Professional Documents
Culture Documents
All Papers Merged
All Papers Merged
A R T I C L E I N F O A B S T R A C T
Keywords: Solvatochromic studies of small IR probes are beneficial to study the structural dynamics of biomolecules as well
Solvatochromism as measuring the strength of surrounding electric fields in condensed phases. Herein, we have investigated the
Hydrogen Bond Population C––O and S––O vibrational probes of structurally similar acetone and dimethyl sulfoxide respectively to compare
Alpha C-H/D interaction
the solvatochromism. Interestingly in water C– –O formed both one hydrogen bond and two hydrogen bond
Disparity of acetone and DMSO
complexes wherein S– –O formed only two hydrogen bond complexes. Not only this, but we have also investi
gated how different solvents alter the IR spectra of the C– –O and S– –O vibrating modes during the interaction
with α C–H of acetone and DMSO respectively. The IR absorption spectra of C– –O of acetone altered after
α deuteration (acetoned6) in aprotic organic solvents as well as aqueous solvents. But, surprisingly for DMSO and
DMSOd6, the IR spectra of S– –O mode remain intact during α isotopic exchange (CH3 to CD3) in organic aprotic
solvents but in strong coordinating polar protic solvents (D2O) it is altered remarkably. The equivalent frequency
gap of C––O of acetone and acetoned6 in organic solvents and water represent the no interaction of α C–H/D
with solvent molecules. Parallelly unaltered IR spectral signature of S– –O of both DMSO and DMSOd6 in aprotic
organic solvent represents the stationary state of the ‘S’ atom and altered S– –O IR absorption in water signifies
the different α C–H and α C-D interaction with the surrounding water molecules. These results are the
cautionary of perturbation on IR probe during α C–H interaction with solvents as well as guest molecules.
* Corresponding authors at: Department of Condensed Matter Physics and Materials Sciences, S. N. Bose National Centre for Basic Sciences, Kolkata 700106, India.
E-mail addresses: anupg86@gmail.com, anup.ghosh@bose.res.in (A. Ghosh).
https://doi.org/10.1016/j.molliq.2023.122005
Received 25 November 2022; Received in revised form 25 April 2023; Accepted 1 May 2023
Available online 3 May 2023
0167-7322/© 2023 Elsevier B.V. All rights reserved.
S. Chakrabarty et al. Journal of Molecular Liquids 382 (2023) 122005
quantification of hydrogen bond interaction in water (D2O) and the spectra, condensed phase structure, and dynamics of acetone and DMSO.
disparities in physicochemical properties of acetone/acetoned6 and
DMSO/DMSOd6 in various solvents such as DCM, ACN and D2O, we 2. Materials and methods
have employed IR absorption, DFT calculations and MD simulations.
However, collective experimental, and theoretical studies offer us to Sample preparation. Dimethyl sulfoxide (DMSO 99.9 %), dimethyl
understand the hydrogen bond population disparity of acetone and sulfoxide d6 (DMSOd6, 99.9 %), acetone (99.9 %), acetoned6 (99.9 %)
DMSO as well as the miscellaneous interaction of CH3 of acetone and and all solvents (99.9 %) namely dichloromethane (DCM), acetonitrile
DMSO with different solvents. In this regard, our present report is ad (ACN), and water (D2O) are bought from Sigma Aldrich and used
ditive studies wherein we have explored inconsistent behavior of without further purifications. All the recorded linear IR absorption
DMSO/DMSOd6 and acetone/acetoned6 depending on the α H/D in spectra were obtained by using an FTIR spectrometer (Bruker, Tensor
teractions with different types of organic and protic solvents. Moreover, 70). 0.1 M solutions of DMSO/DMSOd6 and acetone/acetoned6 mole
the diversity in the interactions of DMSO and acetone with organic cules in different solvents were used during the study. Between the 2 mm
solvents and water have been explored to understand how α C–H/D thick CaF2 windows separated by a Teflon spacer, all the mentioned
interaction can affect the IR absorptions of S–
– O/C–– O stretching modes
sample solutions were placed.
differently. Interestingly we observed that IR spectra of S–– O of DMSO
and DMSOd6 remain unaltered in organic solvents but altered in pres
3. Theoretical Section
ence of water wherein in the case of acetone and acetoned6 C– – O fre
quency is altered in every solvent (DCM, ACN) including D2O. Overall,
DFT Calculations. All the molecular geometries of acetone/ace
this type of finding plays a prominent role to explain the details about
toned6, as well as DMSO/DMSOd6 in organic solvents and their H-bond
the implications of micro solvation environment on different vibrational
complexes in water, were optimized by density functional theory with
2
S. Chakrabarty et al. Journal of Molecular Liquids 382 (2023) 122005
the employment of the B3LYP (6-311G +(d,p)) level of theory in the case of DMSO, S– – O is mostly 2hb (~96 %) populated wherein the
Gaussian 09. [49] Herein we have used atomic Mulliken charges from negligible amount of 0hb (~0%) and 1hb (~3%) population is observed
the optimized structure of acetone and DMSO in DFT calculation. (Fig. 2). So from the MD simulation, we have observed though acetone
MD Simulation. MD simulations were carried out with the and DMSO are similar in structure, the H-bond interaction patterns are
CHARMM general forcefield using the GROMACS version 4.6.5 package completely different. The mentioned H-bond population calculations
for acetone and DMSO in water. [66–68] All the forcefield parameters of command close proximity of C– – O IR absorption of 1hb and 2hb com
acetone and DMSO were taken from virtualchemistry.org, which has plexes which makes them very difficult to separate. But, certainly, for
been systematically tested earlier. [62–63] We have used the TIP4P DMSO, the S– – O IR absorption is mostly coming from 2hb populated
water model for the MD simulation of acetone and DMSO. [62–63] A 5 complexes.
nm cubic box, filled with water molecules was used to run an MD To investigate the reason led behind this disparity in H-bond pop
simulation of acetone and DMSO. The MD simulations were performed ulations, we have done multiple DFT calculations to compute the for
for each minimization, using the steepest-descent algorithm followed by mation energy of 1hb and 2hb complexes of acetone and DMSO
the equilibrium in the NVT ensemble (1 ns) and equilibrium in NPT (1 simultaneously. Due to the similar effect of water on acetone and DMSO,
ns) at 1 bar pressure at 300 K temperature. [62–63] Final coordinates we have ignored the effect of solvent dynamics in DFT calculations. The
were recorded for 1 ns MD simulation run and used for the hydrogen calculated formation energy of the 2hb complex is − 22kj/mol and the
bond population of acetone and DMSO in neat water. Hydrogen bond 1hb complex is − 11.5 kJ/mol of acetone. Wherein for DMSO, they are
populations were computed using standard 3.5 A donor (D)-to-acceptor about − 42 kJ/mol and − 21 kJ/mol respectively. The graphical repre
(A) distance cutoff, and a 30◦ angle cutoff. The distance parameter was sentation in Fig. 3, clearly shows the corresponding formation energies
selected to match the first minimum of the D − A radial distribution and the energy gaps between 1hb and 2hb for both acetone and DMSO.
function. Due to a small stabilization energy gap (Δ9.5 kJ/mol), 1hb and 2hb
conformation of acetone can be converted on either side very easily. On
4. Results and discussion contrary, due to the high stabilization energy of 2hb and high energy gap
(Δ21.5 kJ/mol) between 1hb and 2hb of DMSO, they can’t convert
To understand the solvation of acetone and DMSO in water, IR ab either side of the equilibrium easily (Fig. 3). The equilibrium is mostly
sorption spectra of 0.1 M of each have been collected. All the collected shifted towards the 2hb complex side. Now to establish the reason of the
IR absorption spectra of acetone and DMSO in D2O are depicted in huge discrepancy in stabilization energies of 2hb and 1hb complexes, we
Fig. 1a and 1b respectively. Second derivative IR spectra have also been have looked on the atomic charge distribution of acetone and DMSO
calculated for a better understanding of the spectral signature (Fig. 1c, (Fig. 4). The δ- charge on the ‘O’ atom (-0.871) of DMSO is much higher
1d). The gaussian line shape of C– – O IR absorption at 1696 cm− 1 dic than the ‘O’ atom (-0.517) of acetone. The disparity in charge distri
tates either 1hb and 2hb complexes are equally populated with almost bution over the ‘O’ atom is due to the higher polarizability of the ‘S’
equal extinction coefficients or they are very close to separate out. atom than the ‘C’ atom of DMSO and acetone respectively. So the
Another possibility also may come either of 1hb or 2hb population is chances of hydrogen bond formation with S– – O are higher for DMSO
negligibly small to separate. Correspondingly, for DMSO in Fig. 1b, a than C– – O of acetone and the coulombic attraction probability of H
gaussian shaped S– – O IR absorption spectrum at 1012 cm− 1, one CH3 atoms of water is also higher towards S– – O than C–– O mode. As the
rocking mode, and two other modes were obtained and the probable solvent is the same for both cases, the dynamical effect on acetone and
preliminary explanation is similar to acetone. To solve the experimental DMSO are presumably similar. But surprisingly we have different
difficulty for hydrogen bond population calculations by IR spectra, MD hydrogen bond populations. So even though solvent dynamics may have
simulations have been performed in water to obtain a better represen some effect on hydrogen bond population, the intrinsic properties of the
tative picture of comparative H-bond populations of acetone and DMSO solute play a key role here.
and correspondingly probable explanation of IR line shapes. The MD We have also done a comparative study of α C–H interactions of
simulation predicts a discrepancy between C– – O and S– – O H-bond acetone and DMSO with water. For such investigation, we have
populations of acetone and DMSO shown in Fig. 2. The C– – O mode of considered acetone and isotope substituted acetoned6 as well as DMSO
acetone forms approximately a significant amount of 1hb (38 %) and and DMSOd6. To resolve the detailed molecular interaction strategy of
2hb (60 %) with a negligible amount of 0hb (~2%) in water. While in acetone/acetoned6 and DMSO/DMSOd6 in different solvents a series of
infrared absorption spectra of 0.1 M acetone and acetoned6 in different
solvents such as DCM, ACN and D2O have been collected (Fig. 5).
Similarly, for DMSO and DMSOd6, we have also taken a series of IR
spectra in the same solvents (Fig. 6). We have deconvoluted the IR ab
sorption spectra of DMSO and DMSOd6 after second order derivative
spectra calculations (See Figure S1 and S2). Based on the second-order
derivative spectra in the region from 950 to 1100 cm− 1 we have
clearly assigned the individual fitted peak (S– – O symmetric stretching
modes-red, other modes- green, blue- and CH3/CD3 rocking modes-
magenta) of DMSO/DMSOd6 (Section S1). All the Experimental fre
quencies of acetone and acetoned6 as well as DMSO/DMSOd6 are sum
marized in Table 1. The constant frequency gap (10 cm− 1) of C– – O IR
absorption spectra of acetone and acetoned6 are observed in each sol
vent, shown in Fig. 5. Wherein, the S– – O IR absorption of DMSO and
DMSOd6 remain unaltered in each organic solvent (DCM, ACN,) but
altered in D2O, depicted in Fig. 6 and Table 1.
Parallelly, we have further evaluated the IR absorption frequency of
acetone/acetoned6 and DMSO/DMSOd6 in different organic solvents as
well as in D2O by DFT calculations. All the calculated frequencies and
Fig. 2. Bar diagram of H-bonded complexes. The H-bonded complex of acetone the respective frequency gaps of C– – O of acetone/acetoned6 and S–– O of
through C– –O (Orange) and the H-bonded complex of DMSO through DMSO/DMSOd6 in different solvents are shortened in Table 2. Where
S–
– O (Green). the C–– O IR absorption frequency gap is almost constant for acetone and
3
S. Chakrabarty et al. Journal of Molecular Liquids 382 (2023) 122005
Fig. 3. Stabilisation Energy of water hydrogen-bonded complexes of acetone (blue) and DMSO (red) in water.
acetoned6 in organic as well as protic solvents (D2O). Again S– – O IR It is well known that the amplitude of the C – H vibration is larger
absorption of DMSO and DMSOd6 are comparable in organic solvents than C − D, which leads to the strengthening of the H-bond bridge of C
but significantly altered by 23 cm− 1 in D2O (Table 2). The calculated − D—OD2 than C – H—OD2. The higher mass of the − CD3 group ro
results exactly matched with experimental observations which drive us tates slower over the C–S bond in DMSOd6, wherein the − CH3 group
to find out the reason of the disparity in IR absorption of C–
– O and S–
–O rotates faster over the C–S bond, which makes the C – H—OD2 H-bond
of acetone/acetoned6 and DMSO /DMSOd6 respectively. We have seen weaker than the C − D—OD2 H-bond. We have also done DFT calcula
from DFT calculations that weak vibrational coupling exists between the tions to study the H bond formation energy of α-H/D of DMSO/DMSOd6.
modes in DMSO. However, as weak coupling cannot explain the large For DMSOd6 it is about − 7.6310 kJ/mol; for DMSO it is − 7.0798 kJ/mol
frequency shift, we have not considered the coupling. which is significantly smaller than DMSOd6. So definitely the α-D H-
In the case of acetone and acetoned6, both the atomic mass of the ‘C’ bond is stronger in DMSOd6 than the α-H H-bond of DMSO. So the
and ‘O atoms are comparable. So, C and O mutually vibrate simulta delocalization of the bond over the S–– O mode of DMSOd6 is more than
neously during C– – O stretching vibration. As the relative mass of the DMSO. Consequently, bond order decreases more in S– – O of DMSOd6
central ‘C’ atom (attached with the heavier CD3 group) is more in ace than DMSO. Subsequently, we have more red-shifted S– – O IR absorption
toned6 than acetone, C– – O IR absorption shifted towards the lower peaks in DMSOd6 than in DMSO.
frequency region. Whereas for DMSO and DMSOd6, the heavy ‘S’ atom On the other hand, the constant frequency gap of IR absorption
of S–– O mode remains in the stationary state and mostly the ‘O’ atom spectra of the C–– O mode of acetone and acetoned6 in organic solvents
vibrates significantly. So, during S– – O stretching vibration, the addi and water also suggests that α C–H/D negligibly interacts with water
tional perturbation from the CH3/CD3 mass is almost negligible in (Table 1 and Table 2). Also, the carbonyl ‘C’ atom of acetone and ace
different organic solvents. But on the contrary, despite the structurally toned6, is not polarizable, and the acidity of α H/D is almost negligible.
and chemically similar DMSO and DMSOd6, S– – O IR absorption is Hence, D2O rarely interacts with α C–H/D of acetone/acetoned6. The
altered remarkably in D2O (Fig. 6). These could be the possibility for calculated CH3 rocking mode of acetone suggests the non-interaction of
different kinds of H-bond interaction of α C–H/D of DMSO and DMSOd6 α H wherein alteration of the mode confirms the α H interaction of
with water. Herein, the high polarizability of the S–– O bond makes α H/ DMSO with D2O (Section S1). These could be the feasible reason for
D acidic and suitable to form H-bond with D2O. dissimilar vibrational C–– O and S– – O IR absorption peaks for both
4
S. Chakrabarty et al. Journal of Molecular Liquids 382 (2023) 122005
Fig. 6. The black spectra denote the experimental IR absorption spectra (highest intensity point normalized) of DMSO ((a) DCM, (b) ACN and (c) D2O) and DMSOd6
((d) DCM, (e) ACN and (f) D2O). All the spectra are fitted with Voigt line shape functions. S–
– O stretching spectra (red), magenta peaks signify CH3 and CD3 rocking
modes and the green, blue represent other modes of vibrations. The cyan color denotes cumulative spectra.
Table 1 Table 2
Experimental IR absorption peak positions of acetone/acetoned6 and DMSO/ Calculated IR absorption frequencies of acetone/acetoned6 and DMSO/DMSOd6
DMSOd6 in different solvents. in different solvents by employing the MP2/cc-pVDZ level of theory.
Solvents νC¼O of acetone (cm¡1) νC¼O of acetoned6 (cm¡1) Δν (cm¡1) Solvents νC¼O of acetone νC¼O of acetoned6 ΔνC¼O
(cm¡1) (cm¡1) (cm¡1)
DCM 1713.3 1702.8 ~10
ACN 1712.5 1703.0 ~10 DCM 1753.8 1744.8 9.0
D2O 1692.0 1682.5 ~10 ACN 1747.0 1738.0 9.0
Solvents νS¼O of DMSO (cm¡1) νS¼O of DMSOd6 (cm¡1) Δν (cm¡1) D2O 1746.0 1736.8 9.2
DCM 1059.0 1058.0 ~1 Solvents νS¼O of DMSO νS¼O of DMSOd6 (cm¡1) ΔνS¼O
ACN 1061.0 1060.0 ~1 (cm¡1) (cm¡1)
D2O 1012.0 989.0 ~23 DCM 1059.5 1061.3 1.8
ACN 1055.1 1058.1 3.0
D2O 1021.0 1001.0 20.0
5
S. Chakrabarty et al. Journal of Molecular Liquids 382 (2023) 122005
acetone/acetoned6 and DMSO/DMSOd6 in a distinct organic solvent and [8] D.L. Akins, Theory of Raman scattering by aggregated molecules, J. Phys. Chem. 90
(8) (1986) 1530–1534.
water.
[9] J.L. McHale, Intermolecular vibrational resonance coupling: Intensity borrowing in
polarized Raman spectroscopy, Int. J. Quantum Chem. Symp. 40 (S25) (1991)
5. Conclusions 593–602.
[10] A. Mortensen, O.F. Nielsen, J. Yarwood, V. Shelley, Anomalies in the Isotropic
Raman Spectra of Liquid Mixtures of Isotopomers of Formamide. Intermolecular
In this work, we inspected the disparity of solvatochromism of Interactions in the Carbonyl Stretching Region, J. Phys. Chem. 99 (13) (1995)
acetone and DMSO through IR absorption spectroscopy, DFT calcula 4435–4440.
[11] A.M. Benson Jr., H.G. Drickamer, Stretching Vibrations in Condensed Systems:
tions and MD simulation. Even though DMSO and acetone have equiv
Especially Bonds Containing Hydrogen, J. Chem. Phys. 27 (5) (1957) 1164–1174.
alent chemical structures except for central sulphur and carbon atom, [12] A.D.E. Pullin, The variation of infra-red vibration frequencies with solvent,
they showed significant discrepancies in hydrogen bond populations Spectrochim. Acta 13 (1–2) (1958) 125–138.
and IR absorptions. During hydrogen bond formation dissimilar per [13] A.D. Buckingham, Solvent effects in vibrational spectroscopy, Trans. Faraday Soc.
56 (1960) 753–760.
centages of the population have been obtained for acetone (1hb-40 % [14] F.G. Dijkman, J.H. Van der Maas, Inhomogeneous broadening of Morse oscillators
and 2hb-60 %) and DMSO (1hb-3 % and 2hb-96 %). On the other aspect, in liquids, J. Chem. Phys. 66 (9) (1977) 3871–3878.
dissimilarity was also observed in the IR absorption spectra of S–
– O and [15] H. Wang, L. Wang, S. Shen, W. Zhang, M. Li, L. Du, X. Zheng, D.L. Phillips, Effects
of hydrogen bond and solvent polarity on the C=O stretching of bis(2-thienyl)
C– O in different organic and protic solvents. We showed that the iso
–
ketone in solution, J. Chem. Phys. 136 (12) (2012), 124509.
topic exchange of α C–H to α C-D, the IR absorption of the S–– O mode of [16] J.L. Finney, D.T. Bowron, A.K. Soper, The structure of aqueous solutions of tertiary
DMSO and DMSOd6 remains unaltered in organic solvents but altered in butanol, J. Phys. Condens. Matter. 12 (8A) (2000) A123.
[17] D.T. Bowron, J.L. Finney, A.K. Soper, Structural Investigation of Solute− Solute
aqueous solvent. Wherein the C– – O frequency of acetone and acetoned6
Interactions in Aqueous Solutions of Tertiary Butanol, J. Phys. Chem. B 102 (18)
significantly altered in every solvent including D2O. Herein we have (1998) 3551–3563.
exposed that α C–H/D of DMSO/DMSOd6 and acetone/acetoned6 [18] J. Turner, A.K. Soper, J.L. Finney, Water structure in aqueous solutions of
tetramethylammonium chloride, Mol. Phys. 77 (3) (1992) 411–429.
interact differently in presence of different organic aprotic and protic [19] S.E. McLain, A.K. Soper, A. Luzar, Orientational correlations in liquid acetone and
solvents and their perturbation on S– – O and C–– O modes IR absorption dimethyl sulfoxide: A comparative study, J. Chem. Phys. 124 (7) (2006), 074502.
frequency is significantly different. The α C H/D of acetone and ace
– [20] A. Idrissi, F. Sokolic, A. Perera, A molecular dynamics study of the urea/water
mixture, J. Chem. Phys. 112 (21) (2000) 9479–9488.
toned6 couldn’t form hydrogen bonds with D2O whereas in the case of
[21] A. Idrissi, S. Longelin, F. Sokolic, Study of Aqueous Acetone Solution at Various
DMSO/DMSOd6, they formed hydrogen bonds which induce a signifi Concentrations: Low-Frequency Raman and Molecular Dynamics Simulations,
cant shift in the S–
– O IR absorption frequency. In conclusion, our results J. Phys. Chem. B 105 (25) (2001) 6004–6009.
delivered an outline of the contradiction between DMSO and acetone [22] J.J. Max, C. Chapados, Infrared spectroscopy of acetone–water liquid mixtures. II.
Molecular model, J. Chem. Phys. 120 (14) (2004) 6625–6641.
which are structurally comparable compounds. We hope that in the [23] Y. Takebayashi, S. Yoda, T. Sugeta, K. Otake, T. Sako, M. Nakahara, Acetone
future, our findings will help in solvatochromic studies of biomolecules hydration in supercritical water: 13 C-NMR spectroscopy and Monte Carlo
such as in DMSO and acetone in the arena of IR spectroscopy. simulation, J. Chem. Phys. 120 (13) (2004) 6100–6110.
[24] D.S. Venables, C.A. Schmuttenmaer, Spectroscopy and dynamics of mixtures of
water with acetone, acetonitrile, and methanol, J. Chem. Phys. 113 (24) (2000)
11222–11236.
Declaration of Competing Interest [25] P.E. Mason, G.W. Neilson, J.E. Enderby, M.L. Saboungi, E.D. Christopher, A.
D. MacKerell, J.W. Brady, The Structure of Aqueous Guanidinium Chloride
The authors declare that they have no known competing financial Solutions, J. Am. Chem. Soc. 126 (37) (2004) 11462–11470.
[26] A.K. Soper, A. Luzar, neutron diffraction study of dimethyl sulphoxide–water
interests or personal relationships that could have appeared to influence
mixtures, J. Chem. Phys. 97 (2) (1992) 1320–1331.
the work reported in this paper. [27] W.K. Orland, Effect of Hydrogen Bonding Solvents on the Infrared Absorption Band
for the Fundamental Vibration of the Carbonyl Group in 1,1,3,3-Tetramethylurea,
Trans. Kans. Acad. Sci. 102 (1999) 53–56.
Data availability
[28] L.C.J. Almeida, P.S. Santos, The anomalous solvent effect in the vibrational
spectrum of 2,3-diphenyl-cycloprop-2-enone: an experimental and theoretical
Data will be made available on request. investigation, Spectrochim. Acta: A 58 (14) (2002) 3139–3148.
[29] R.A. Nyquist, H.A. Fouchea, G.A. Hoffman, D.L. Hasha, Infrared Study of β
-Propiolactone in Various Solvent Systems and Other Lactones, Appl. Spectrosc. 45
Acknowledgments (1991) 860–867.
[30] R.A. Nyquist, C.L. Putzig, D.L. Hasha, Solvent Effect Correlations for Acetone: IR
versus NMR Data for the Carbonyl Group, Appl. Spectrosc. 43 (6) (1989)
A.G thanks the DST- India for financial support and SNBNCBS, Kol
1049–1053.
kata for instrumental facilities. [31] R.A. Nyquist, C.L. Putzig, L. Yurga, Solvent and Concentration Effects on Carbonyl
Stretching Frequencies: Ketones, Appl. Spectrosc. 43 (6) (1989) 983–991.
[32] R.A. Nyquist, T.D. Clark, R. Streck, Infrared study of alkyl carboxylic acids in CCl4
Appendix A. Supplementary data and/or CHCl3 solutions, Vib. Spectrosc. 7 (3) (1994) 275.
[33] P. Bruni, E. Giorginy, E. Maurelli, G. Tosi, Fourier transform infrared spectrometry
Supplementary data to this article can be found online at https://doi. investigation of solvent effect on NH and CO stretching modes in N-
acylaminopyridines, Vib. Spectrosc. 12 (2) (1996) 249–255.
org/10.1016/j.molliq.2023.122005.
[34] R.A. Nyquist, C.L. Putzig, T.D. Clark, Infrared study of 1,3-dimethyl-2-imidazoli
dinone in various solvents, Vib. Spectrosc. 12 (1) (1996) 81–91.
References [35] J.B.F.N. Engberts, G.R. Famini, A. Perjessy, L.Y.J. Wilson, Solvent effects on C=O
stretching frequencies of some 1-substituted 2-pyrrolidinones, Phys. Org. Chem. 11
(4) (1998) 261–272.
[1] W. Liptay, Electrochromism and Solvatochromism, Angew. Chem. internat. Edit. 8
[36] A. Ghosh, B. Cohn, A.K. Prasad, L. Chuntonov, Quantifying conformations of ester
(3) (1969) 177–188.
vibrational probes with hydrogen-bond-induced Fermi resonances, J. Chem. Phys.
[2] M. Cho, Vibrational solvatochromism and electrochromism: Coarse-grained models
149 (18) (2018) 184501–184511.
and their relationships, J. Chem. Phys. 130 (9) (2009), 094505.
[37] A. Ghosh, Vibrational Coupling on Stepwise Hydrogen Bond Formation of Amide I,
[3] B.A. Lindquist, K.E. Furse, S.A. Corcelli, Nitrile groups as vibrational probes of
J. Phys. Chem. B 123 (37) (2019) 7771–7776.
biomolecular structure and dynamics: an overview, Phys. Chem. Chem. Phys. 11
[38] K.I. Oh, K. Rajesh, J.F. Stanton, C.R. Baiz, Quantifying hydrogen-bond populations
(32) (2009) 8119–8132.
in DMSO/water mixtures, Angew. Chem. Int. Ed. 56 (38) (2016) 11375–11379.
[4] K.S. Schwiezer, D. Chandler, Vibrational dephasing and frequency shifts of
[39] K.I. Oh, C.R. Baiz, Crowding Stabilizes DMSO− Water Hydrogen-Bonding
polyatomic molecules in solution, J. Chem. Phys. 76 (5) (1982) 2296–2314.
Interactions, J. Phys. Chem. B 122 (22) (2018) 5984–5990.
[5] C.M. Huggins, G.C. Pimentel, Infrared Intensity of the C — D Stretch of
[40] S.M. Kashid, G.Y. Jin, S. Chakrabarty, S. Bagchi, Y.S. Kim, Two-Dimensional
Chloroform-d in Various Solvents, J. Chem. Phys. 23 (5) (1955) 895–898.
Infrared Spectroscopy Reveals Cosolvent Composition-Dependent Crossover in
[6] C.H. Veas, J.L. McHale, Intermolecular resonance coupling of solute and solvent
Intermolecular Hydrogen Bond, J. Phys. Chem. Lett. 8 (7) (2017) 1604–1609.
vibrational modes, J. Am. Chem. Soc. 111 (18) (1989) 7042–7046.
[41] S.M. Kashid, G.Y. Jin, S. Bagchi, Y.S. Kim, Cosolvent Effects on Solute− Solvent
[7] R.C. Lord, B. Nolin, H.D. Stidham, Quantitative Study of the Bonding of
Hydrogen-Bond Dynamics: Ultrafast 2D IR Investigations, J. Phys. Chem. B 119
Chloroform-d in Various Solvents by Infrared Spectrometry, J. Am. Chem. Soc. 77
(49) (2015) 15334–15343.
(5) (1955) 1365–1368.
6
S. Chakrabarty et al. Journal of Molecular Liquids 382 (2023) 122005
[42] I.A. Borin, M.S. Skaf, Molecular Association between Water and Dimethyl Sulfoxide [55] N.C. Santos, J.F. Coelho, J.M. Silva, C. Saldanha, Multidisciplinary utilization of
in Solution: The Librational Dynamics of Water, Chem. Phys. Lett. 296 (1–2) dimethyl sulfoxide: pharmacological, cellular, and molecular aspects, Biochem
(1998) 125–130. Pharmacol. 65 (7) (2003) 1035–1041.
[43] A. Luzar, D. Chandler, Structure and Hydrogen-Bond Dynamics of Water-Dimethyl [56] C.D. Wood, J. Wood, Pharmacologic and Biochemical considerations of Dimethyl
Sulfoxide Mixtures by ComputerSimulations, J. Chem. Phys. 98 (10) (1993) Sulfoxide, Ann N Y Acad Sci. 243 (1) (1975) 7–19.
8160–8173. [57] W.R. Fawcett, A.A. Kloss, Attenuated total reflection fourier-transform IR
[44] S.M. Kashid, R.K. Singh, H. Kwon, Y.S. Kim, A. Mukherjee, S. Bagchi, Arresting an spectroscopic study of dimethyl sulfoxide self-association in acetonitrile solutions,
Unusual Amide Tautomer Using Divalent Cations, J. Phys. Chem. B 123 (40) J. Chem. Soc. Faraday Trans. 92 (18) (1996) 3333–3337.
(2019) 8419–8424. [58] W.J. Stanley, H. Robert, Pharmacology of DMSO, Cryobiology 23 (1) (1986)
[45] A. Wulf, R. Ludwig, Structure and Dynamics of Water Confined in Dimethyl 14–27.
Sulfoxide, Chem Phys Chem. 7 (1) (2006) 266–272. [59] R.H. Figueroa, E. Roig, H.H. Szmant, Infrared study on the self-association of
[46] R. Ludwig, T.C. Farrar, M.D. Zeidler, Temperature Dependence of the Deuteron and dimethyl sulfoxide, Spectrochem. Acta. 22 (4) (1966) 587–592.
Oxygen Quadrupole Coupling Constants of Water in the System Water/Dimethyl [60] T. Clark, J.S. Murray, P. Lane, P. Politzer, Why are dimethyl sulfoxide and dimethyl
Sulfoxide, J. Phys. Chem. D 98 (1994) 6684–6687. sulfone such good solvents? J Mol Model. 14 (8) (2008) 689–697.
[47] S.D. Fried, S. Bagchi, S.G. Boxer, Measuring Electrostatic Fields in Both Hydrogen- [61] S. Singh, S.K. Srivastava, D.K. Singh, Raman scattering and DFT calculations used
Bonding and Non-Hydrogen-Bonding Environments Using Carbonyl Vibrational for analyzing the structural features of DMSO in water and methanol, RSC Adv. 3
Probes, J. Am. Chem. Soc. 135 (30) (2013) 11181–11192. (13) (2013) 4381–4390.
[48] S.G. Boxer, S. Bagchi, S.D. Fried, N. Levinson, M. Saggu, L. Xu, Vibrational Stark [62] M.I.S. Sastry, S.J. Singh, Self-association of dimethyl sulphoxide and its dipolar
Spectroscopy Directly Probes Electric Fields in Proteins, Biophys. J . 104 (2) (2013) interactions with water: Raman spectral studies, Raman Spectrosc. 15 (2) (1984)
355A. 80–85.
[49] M.S. Kashid, S. Bagchi, Experimental Determination of the Electrostatic Nature of [63] S.A. Kirillov, M.I. Gorobets, M.M. Gafurov, K.S. Rabadanov, M.B. Ataev,
Carbonyl Hydrogen-Bonding Interactions Using IR-NMR Correlations, J. Phys. Temperature dependence of associative equilibria of DMSO according to raman
Chem. Lett. 5 (18) (2014) 3211–3215. scattering spectra, Russ. J. Phys. Chem A 88 (1) (2013) 175–177.
[50] S. Chakrabarty, A. Ghosh, Inconsistent hydrogen bond-mediated vibrational [64] S. Chakrabarty, H.S. Deshmukh, A. Barman, S. Bagchi, A. Ghosh, On− Off Infrared
coupling of amide I, RSC Adv. 13 (2023) 1295. Absorption of the S=O Vibrational Probe of Dimethyl Sulfoxide, J. Phys. Chem. B
[51] A. Greenberg, C.M. Breneman, J.F. Liebman, The Amide Linkage: Structural 126 (24) (2022) 4501–4508.
Significance, Chemistry, Biochemistry and Material Science, Wiley, New York, NY, [65] S. Chakrabarty, A. Barman, A. Ghosh, Anomalous Infrared Absorbance of S=O: A
USA, 2000. Perturbation Study of α-C–H/D, J. Phys. Chem. B 126 (29) (2022) 5490–5496.
[52] L. Brunton, B. Chabner, B. Knollman, Goodman and Gilman’s the Pharmacological [66] P.J. Stephens, F.J. Devlin, C.F. Chabalowski, M.J. Frisch, Ab Initio Calculation of
Basis of Therapeutics, MacGraw-Hill, New York, NY, USA, 2010. Vibrational Absorption and Circular Dichroism Spectra Using Density Functional
[53] V.R. Pattabiraman, J.W. Bode, Rethinking amide bond synthesis, Nature 480 Force Fields, J. Phys. Chem. A 98 (45) (1994) 11623–11627.
(2011) 471–479. [67] A.D. Becke, Density-functional thermochemistry. III. The role of exact exchange,
[54] A.B. Hughes, Amino Acids, Peptides and Proteins in Organic Chemistry, Wiley- J. Chem. Phys. 98 (7) (1993) 5648–5652.
VCH, Weinheim, Germany, 2009. [68] S. Chakrabarty, S. Maity, D. Yazhini, A. Ghosh, Surface-Directed Disparity in Self-
Assembled Structures of Small-Peptide L-Glutathione on Gold and Silver
Nanoparticles, Langmuir 36 (38) (2020) 11255–11261.
7
RSC Advances
This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.
Using infrared spectroscopy and density functional theory (DFT) calculations, we scrutinized an amide
(dimethylformamide) as a “model” compound to interpret the interactions of amide 1 with different
phenol derivatives (para-chlorophenol (PCP) and para-cresol (CP)) as “model guest molecules”. We
established the involvement of amide I in vibrational coupling with symmetric and asymmetric C]C
modes of different phenolic derivatives and how their coupling was dependent upon different guest
aromatic phenolic compounds. Interestingly, substitution of phenol perturbed the pattern of vibrational
coupling with amide I. The symmetric and asymmetric C]C modes of PC were coupled significantly
with amide 1. For PCP, the symmetric C]C mode coupled significantly, but the asymmetric mode
Received 12th November 2022
Accepted 16th December 2022
coupled negligibly, with amide I. Here, we reveal the nature of vibrational coupling based on the
structure of a guest molecule hydrogen-bonded with amide I. Our conclusions could be valuable for
DOI: 10.1039/d2ra07177k
depiction of the unusual dynamics of coupled amide-I modes as well as the dependency of vibrational
rsc.li/rsc-advances coupling on altered factors.
© 2023 The Author(s). Published by the Royal Society of Chemistry RSC Adv., 2023, 13, 1295–1300 | 1295
View Article Online
Fig. 1 Linear IR spectra of amide I (dimethylformamide) in the presence of different concentrations of (a) para-cresol and (b) para-chlorophenol.
The colour of the spectra is described in the inset with different concentrations of para-cresol and para-chlorophenol. The green (up) and red
arrow (down) represent enhancement and suppression of absorbance, respectively.
1296 | RSC Adv., 2023, 13, 1295–1300 © 2023 The Author(s). Published by the Royal Society of Chemistry
View Article Online
Fig. 2 Linear IR absorption of the C]C mode of (a) para-cresol (0.1 M) and (b) para-chlorophenol (0.1 M) in the presence of dimethylformamide
at 0.00 M (black), 0.02 M (red), 0.04 M (blue), 0.06 M (green), 0.08 M (violet), and 0.10 M (dark-yellow). The green arrow represents enhancement
of absorbance. The red cross represents no enhancement of absorbance.
absorption peaks for symmetric and asymmetric C]C modes We did not know whether the enhancement was due to the
were at 1597.5 cm−1 and 1618.5 cm−1 (Fig. 2a) whereas, for PCP altered electron density of the phenolic ring (C]C mode) aer
they were at 1594.0 cm−1 and 1606.5 cm−1, respectively (Fig. 2b). hydrogen bonding with DMF, so we undertook IR spectroscopy
For PC and PCP, with an increasing concentration of DMF of a mixture of acetonitrile (ACN) and phenol at a ratio of 1 : 1
(0.000–0.10 M), a signicant difference in IR absorbance was (Fig. 3b). The shiing to a higher IR absorbance frequency of CN
observed between symmetric and asymmetric C]C modes signied formation of a hydrogen bond between ACN with
(Fig. 2). For PC, with a gradual increase in the DMF concen- phenolic OH (Fig. 3b). An absence of enrichment of IR absor-
tration from 0.000 M to 0.10 M, IR absorbance for symmetric bance of C]C explained the non-involvement of hydrogen
and asymmetric C]C modes was enhanced signicantly. bonds. Hence, these results suggested that the enhancement in
However, in contrast with PCP, though IR absorbance for IR absorbance of the C]C mode was not due to n–p* bonding
symmetric C]C stretching was enhanced, IR absorbance for or hydrogen bonding. Therefore, the enhancement was prob-
asymmetric C]C stretching was altered negligibly throughout ably due to vibrational coupling between the C]C and C]O
the experiment (Fig. 2). The IR-absorbance enhancement ratio modes of amide (DMF) and phenol derivatives (PC and PCP).
for the symmetric and asymmetric C]C modes of PC was 1.35 To check that our hypotheses on vibrational coupling were
and 1.40 whereas, for PCP, it was 5.97 and 1.00, respectively robust, we carried out DFT calculations (b3lyp, 6311 G (d, p)) for
(Table 1), as calculated from Fig. 2. The transition dipole PC and PCP hydrogen-bonded with DMF, and the videos are
moment of the symmetric and asymmetric modes of PC and provided in ESI† (AV1, AV2, AV3, AV4). Vibrational couplings
PCP changed accordingly (Table 1). These unusual phenomena were visualized between the C]C of phenol derivatives and
focused our attention on the intermolecular interactions C]O of DMF. Interestingly, our experimental observations
between amide (C]O) and phenolic compounds (PC and PCP). aligned with the videos for DFT calculations. In AV1 and AV2,
The increase in IR absorbance of the C]C mode could have the symmetric and asymmetric C]C modes of PC were coupled
been due to the altered hydrogen bonding with vibrational signicantly with amide 1 of DMF; for PCP, the symmetric C]C
modes or n–p* bonding of PC and the PCP ring with the C]O mode was coupled signicantly (AV3) but the asymmetric mode
mode of DMF (Fig. 2a and b). To elucidate the precise reason was not (AV4). PC and PCP are phenol derivatives and form
underpinning the enhancement, we undertook IR spectroscopy hydrogen bonds with amide 1 of DMF, but they showed
of a 1 : 1 DMF : anisole mixture. The unaltered enhancement of different vibrational coupling. Hence, phenolic substitution
IR absorbance for the C]C mode of the DMF : anisole (1 : 1) altered the pattern of vibrational coupling.
mixture signied no interaction of the non-bonded electron of Vibrational coupling is dependent upon the frequency gap of
the oxygen atom of DMF with the p electron cloud of anisole. IR absorption, the distance between two vibration modes, and
Table 1 Vibrational frequencies of IR absorption for the C]C mode and the ratio of IR absorption area for the hydrogen-bonded C]C mode
and free C]C mode of phenol derivatives. The enhancement factor and transition dipole moment ratio were calculated from the experimental
data shown in Fig. 2
© 2023 The Author(s). Published by the Royal Society of Chemistry RSC Adv., 2023, 13, 1295–1300 | 1297
View Article Online
Fig. 3 (a) Linear IR absorbance spectra of the C]C mode of anisole (0.1 M) in the absence (black) and presence (red) of dimethylformamide (0.1
M). (b) IR spectra of the CN mode of ACN (0.1 M) in the absence (black) and presence (green) of phenol (0.1 M) in CCl4. (c) IR spectra of the C]C
mode of phenol (0.1 M) in the absence (black) and presence (red) of ACN (0.1 M).
orientation. The frequency gap between the asymmetric C]C observed for the symmetric C]C mode, but negligible
mode of PC and C]O stretching mode of DMF was smaller enhancement was observed for the asymmetric C]C mode
(56 cm−1) than that of the symmetric C]C and C]O (76 cm−1) (Fig. 4b). The distance between the C]C and C]O modes of
modes of PC (Fig. 4a). However, the intensity of asymmetric and phenol derivatives and DMF according to DFT calculations are
symmetric C]C modes was enhanced simultaneously. The shown in Fig. 4c and d. An identical distance (5.2 Å) between
frequency gap was smaller for the asymmetric C]C mode C]C and C]O could not explain the disparity in vibrational
(67 cm−1) than symmetric C]C mode (79.5 cm−1) of PCP with coupling. Hence, we assumed that a different transition dipole
amide 1 of DMF. Signicant enhancement in IR absorbance was angle between the C]C mode and amide-I mode was the cause
Fig. 4 Linear IR spectra of amide I (dimethylformamide) and the C]C mode of (a) para-cresol and (b) para-chlorophenol. The IR spectra of
amide I (violet), free amide I (red), hydrogen-bonded amide I (blue), cumulative spectra (green), and C]C (gold) are shown. (c) Distance between
C]O and C]C of dimethylformamide and para-cresol. (d) Distance between C]O and C]C of dimethylformamide and para-chlorophenol. A
distance of 5.2 Å was calculated (by DFT) for para-cresol and para-chlorophenol, respectively, with amide I of dimethylformamide.
1298 | RSC Adv., 2023, 13, 1295–1300 © 2023 The Author(s). Published by the Royal Society of Chemistry
View Article Online
Fig. 5 Red and green denote the linear vibration modes of C]O and C]C, respectively. bij is the coupling constant, and mi, and mj are the
transition dipole moments of the C]C mode and C]O mode, respectively. Orientation of C]O and C]C modes- (a) parallel (b) antiparallel (c)
linearly parallel (d) oppositely parallel and (e) perpendicular.
of this difference in vibrational coupling. To ascertain the AV10). The symmetric C]C mode of ortho-chlorophenol was
reason for this anomalous coupling behaviour, we exposed coupled with amide I, but the asymmetric C]C mode was not
different orientations between vibration modes (C]C and C] (AV11, AV12). Hence, the C]C symmetric and asymmetric
O) (Fig. 5). The sign of the coupling constant (bij) was dependent modes coupled with amide 1 for para-electron-promoting
upon the geometry of the hydrogen-bonded DMF and phenol phenolic compounds. Only the C]C symmetric mode
derivatives (eqn (1)).55 If C]C and C]O modes are parallel coupled with amide 1 for para- and ortho-electron-withdrawing-
(Fig. 5a) and bij is positive, the intensity of the C]O vibrational substituted phenolic compounds. The symmetric and asym-
mode will be enhanced, with a transition dipole moment mi + mj. metric C]C modes of electron-withdrawing meta-substituted
In contrast, the intensity of the C]C vibrational mode will be phenolic compounds were coupled like para-electron-
weaker, with a transition dipole moment mi − mj. In the case of promoting phenolic compounds.
antiparallel C]C and C]O modes (Fig. 5b) and negative
coupling constant bij, the C]C mode will be weaker and
correspondingly the C]O mode will be stronger, with a transi- Conclusions
tion dipole moment of mi + mj and mi − mj, respectively. In the
Employment of FTIR spectroscopy and DFT calculations
head-to-tail (Fig. 5c) or head-to-head orientation (Fig. 5d) of
revealed the anomalous vibrational coupling between amide 1
C]C and C]O, the lower frequency mode of C]C will carry
and the C]C mode of phenol derivatives. For PC, IR absor-
more oscillator strength compared with the C]O mode. An
bance was enhanced markedly for symmetric and asymmetric
unperturbed intensity will result if the C]C mode and C]O
C]C modes upon gradual addition of DMF. For PCP, the
mode are perpendicular to each other (Fig. 5e). The geometries
symmetric C]C mode was enhanced signicantly, whereas the
of the DMF–PC complex and DMF–PCP complex are in-between
asymmetric C]C mode was not. Even though the frequency gap
the limits we modelled in Fig. 5. The symmetric and asymmetric
was less for the asymmetric C]C transition compared with that
C]C modes of PC and symmetric C]C mode of PCP are
for the C]O transition, and the distance between C]C and
coupled in the manner shown in Fig. 5c and d. Probably, the
C]O was constant for symmetric and asymmetric C]C modes
asymmetric C]C mode of PCP was coupled with the C]O
for PC and PCP, distinctive behaviour was observed for different
mode of DMF as shown in Fig. 5e or they were coupled very
phenol derivatives in the presence of DMF. Theoretical and
weakly according to the model shown in Fig. 5a–d.
3 experimental observations revealed that, with alteration of
2
! ! mi !
rij ! rij !
mi ! mj phenol substituents, the coupling pattern changed. The area of
mj
bij ¼ 1=4p30 4 3 5 (1) IR absorbance was enhanced z5.97 times for the C]C
r3ij r5ij
symmetric mode, whereas it was altered negligibly for the
asymmetric mode, in the case of PCP. For PC, the area of IR
where bij is the coupling constant, 30 is permittivity, mi and mj are
absorbance for symmetric and asymmetric modes was
transition dipole moments, and rij is the distance between two
enhanced signicantly (z1.4 times). Thus, we revealed the
vibration modes.
nature of vibrational coupling based on the structure of a guest
We wished to validate our hypotheses of coupling of the
molecule hydrogen-bonded with amide I. Our conclusions
symmetric and asymmetric C]C modes of PC and PCP with
could be valuable for depiction of the unusual dynamics of
amide 1. Hence, we carried out DFT calculations of para-ethyl-
coupled amide-I modes as well as the dependency of vibrational
phenol and para-nitrophenol. Surprisingly, we observed the
coupling on altered factors.
same results as PC for para-ethylphenol (AV5, AV6) and as PCP
for para-nitrophenol (AV7, AV8). We calculated the vibrational
coupling for meta-chlorophenol and ortho-chlorophenol
complexes with DMF. The symmetric and asymmetric C]C
Conflicts of interest
modes of meta-chlorophenol were coupled with amide I (AV9, The authors declare no competing nancial interest.
© 2023 The Author(s). Published by the Royal Society of Chemistry RSC Adv., 2023, 13, 1295–1300 | 1299
View Article Online
for a fellowship. 29 D. G. Brown and J. Bostrom, J. Med. Chem., 2016, 59, 4443–
4458.
References 30 V. R. Pattabiraman and J. W. Bode, Nature, 2011, 480, 471–479.
31 A. B. Hughes, Origins and Synthesis of Amino Acids, Amino
1 R. E. Dickerson and I. Geis, The Structure and Action of Acids, Peptides and Proteins in Organic Chemistry, Wiley-
Proteins, Harper and Row, New York, Evanston, London, VCH, Weinheim, Germany, 2009, vol. 1.
Open Access Article. Published on 05 January 2023. Downloaded on 1/6/2023 5:20:04 AM.
1300 | RSC Adv., 2023, 13, 1295–1300 © 2023 The Author(s). Published by the Royal Society of Chemistry
pubs.acs.org/JPCB Article
the strength of the surrounding electric fields and the hydrogen bonding
status. Herein, we demonstrated how the solvents alter the infrared (IR)
spectra of the S�O vibrating mode. The experimental measurement of the
involvement of α-H/D isotopic interactions with different solvents and
their effects on the IR absorbance spectra of the vibrational probe provides
detailed knowledge of the microsolvation environment despite the
complexity of overlapping bands in the spectra. Herein, we discover how
the solvents interact differently with DMSO and DMSO-d6, while being
electronically and structurally the same. Interestingly, the IR spectrum of
the S�O mode remains unaltered during α-isotopic replacement in the
presence of aprotic solvents (acetone, acetonitrile, and dichloromethane),
but in strongly coordinating polar solvents (D2O), it is altered remarkably.
There is a lack of quantitative information about the influence of the α-H atom or α-isotopic substitution on the vibrational probe in
the literature. Our experiments provide a detailed molecular understanding of the structure of DMSO in DMSO−solvent binary
mixtures. As DMSO plays an important role in virtually all subdisciplines of chemistry and biology, we believe that our work will be
of interest to a large diversity of studies in these fields.
■ INTRODUCTION
Interactions between molecules in the condensed phase
good solvent/cosolvent and a pharmaceutically important
compound, recently, similar S�O bond-containing DMSO
assessments become a challenging area of research. A proven molecules have gained interest to be explored.27−31 Although
commanding tool is vibrational spectroscopy to unravel the numerous vibrational spectroscopic reports on DMSO have
secret of molecular liquids. The infrared spectrum of a been published, a consensus regarding the interaction with the
molecule not only depends on the strength of a certain bond solvent is yet to be reached.32−44 Several vibrational studies
and mass of the vibrating atoms but also can be markedly have been reported based on the IR absorbance peak
influenced by medium factors. It has been reported that assignment of S�O and the corresponding mode of
solute−solvent interactions can have a reckonable influence on DMSO.33−44 Unaltered S�O IR absorption in organic
the frequencies and intensities of vibrational bands in infrared solvents and altered S�O IR absorption in aqueous solvents
(IR) spectroscopy.1−7 For many years, the impact of the warn us to explore the α-H/D interaction of DMSO and
solvent on the spectral properties of molecules, usually DMSO-d6 with the surrounding solvent molecules, as they are
mentioned as solvatochromism, has been investigated.8,9 structurally and electronically the same.33,43,44 Such an
Particularly, wavenumber values, intensities, and line shapes observation cannot be explained based on the S�O bond
of IR absorption spectra are influenced by the nature of the interaction of DMSO/DMSO-d6 with solvents; however,
solvent via dielectric effects and specific interactions. There are mostly in the literature, only the interaction of the S�O
lots of models that have already been projected to treat the bond of DMSO with the solvent has been investigated
shifts in vibrational frequencies over varied ranges of liquid widely.31−37,40,41 In the binary mixture of DMSO−water, the
densities.1,10−13 A net solvent-induced force invoked by these S�O bond exists in four structural forms, namely, self-
models can disturb the normal-mode frequencies of an isolated associated (aggregated), free monomer (0HB), singly hydro-
solute molecule. Thus, the obtained frequency shifts in the gen-bonded (1HB), and doubly hydrogen-bonded (2HB)
stretching vibrational frequency are directly proportional to the species.42,43 In this regard, our present report is an additive
changes in bond lengths, which is induced by the surrounding study wherein we have explored the H-bond interaction not
medium of the solutes.10,14 In this context, molecules bearing
the C�O moiety have been extensively investigated, as their Received: February 25, 2022
polarity and hydrogen bond-accepting nature make them very Revised: June 21, 2022
sensitive to the solvent influence.15−26 However, being a very
only with the S�O bond but also with α-H/D of DMSO/ functional and a 6-311G +(d,p) basis set as implemented in
DMSO-d6 in an aqueous solvent. Gaussian 09.44−46
To identify structural and physicochemical behaviors, we MD Simulation. MD simulations of DMSO in water and
have performed IR absorption, density functional theory organic solvents were carried out with the CHARMM general
(DFT) calculations, and molecular dynamics (MD) simu- forcefield using the GROMACS version 4.6.5 package.47−50
lations of DMSO dissolved in various solvents, namely acetone, The forcefield parameters of DMSO and organic solvents were
ACN, DCM, D2O, and hexafluoroisopropanol (HFIP). obtained from virtualchemistry.org, which has been thoroughly
However, a combined experimental and theoretical approach tested earlier.51 We have used the TIP4P water model for MD
offers the potential to understand the diverse interaction of the simulation in water.52 A single DMSO molecule was placed in
DMSO mode with solvents, which motivates the present study. the center of a 5 nm cubic box of the respective solvent
Moreover, the nature of the interaction of DMSO in water has molecules. Then, energy minimizations of each MD simulation
been explored to understand α-H-bond effects on the IR were performed using the steepest-descent algorithm followed
absorptions of S�O stretching modes. To distinguish between by the equilibrium in the NVT ensemble using a velocity-
different electrostatic environments and H-bond interactions rescaling thermostat for 100 ps and equilibrium in NPT at 1
of DMSO, we turned to link experimental IR absorption with bar pressure using a Parrinnello−Rahman barostat for 1 ns at
MD simulation and DFT calculations. The important role of 300 K temperature.53,54 Finally, the coordinates were recorded
this type of interaction is causing the condensed-phase for a 1 ns MD simulation run and used for the calculation of
structure and dynamics and its significances on the different the electric field and hydrogen bond population of the S�O
vibrational spectra (Scheme 1). bond of DMSO. Hydrogen bond populations were computed
using a 3.5 Å donor (D)-to-acceptor (A) distance cutoff,
Scheme 1. Chemical Structure of DMSO and DMSO-d6 between the oxygen atoms of the donor and acceptor, and a
30° D−H−A angle cutoff. The distance parameter was selected
to match the first minimum of the D−A radial distribution
function. The electric fields, which the surrounding solvent
molecules exerted on the S�O bond of DMSO and DMSO-
d6, were calculated by following similar protocol used in ref 55.
Figure 1. Black spectra denote the experimental IR absorption spectra of DMSO ((a) acetone, (b) ACN, (c) DCM, and (d) D2O) and DMSO-d6
((e) acetone, (f) ACN, (g) DCM, and (h) D2O). The experimental IR spectra are fitted to four peaks using Voigt lineshape functions. The red
peaks correspond to the S�O stretch, and the green, blue, and magenta peaks correspond to the CH3 (a−d) and CD3 (e−h) rocking modes. All
red dotted lines signify the S�O IR absorption peak position.
B https://doi.org/10.1021/acs.jpcb.2c01374
J. Phys. Chem. B XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry B pubs.acs.org/JPCB Article
C https://doi.org/10.1021/acs.jpcb.2c01374
J. Phys. Chem. B XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry B pubs.acs.org/JPCB Article
Figure 5. Illustration of (a) CH3 and (b) CD3 group rotation over the C−S axis from DMSO and DMSO-d6.
having a lower mass, −CH3 rotates faster than the −CD3 above-mentioned IR experiments. The lower bond order of
group, which makes the C−H---O H-bond weaker than the DMSO-d6 illustrated the lower IR absorbance frequency
C−D---O H-bond (Figure 5). The combination of the compared to DMSO (Table 1). A close analysis indicated
mentioned effects leads to the situation where a deuterated that an increase of the H-bond strength correlated with a
H-bond bridge is more stable. decrease of the C−H/D bond strength. Due to the H-bond
formation of α-H/D, delocalization of the σ C−H/D bond
i 1y h K
E v = jjjv + zzz occurred. It caused an amplified overlap with the antibonding
k 2{2 (1) orbital of the C−H/D bond, resulting in the reduction of the
m1m2 α-C−H/D bond order, and the α-C−H/D bond became weak.
= Parallelly, the S�O bond becomes weaker, as shown in
m1 + m2 (2) Scheme S1. As the C−D forms a stronger H-bond,
Therefore, the plausible explanation of the S�O IR absorption delocalization of the C−D bond is more, and thus, the S�
peak difference is the delocalization of the bond after the H- O bond of DMSO-d6 becomes weaker than the S�O bond of
bond formation of the α-H/D atom with D2O. As the α-D DMSO. Consequently, the S�O IR absorption shifted more
forms a stronger H-bond than α-H with D2O, bond for DMSO-d6 than for DMSO in D2O, wherein as there is no
delocalization is higher in DMSO-d6 than in DMSO (Scheme chance to form H-bonds in aprotic organic solvents, the IR
S1). Therefore, unsurprisingly, the bond order decreases and absorption peak position remains unaltered. One “C−D---O”
the bond distance increases in DMSO-d6. Subsequently, S�O H-bond with D2O of DMSO-d6 is 0.551 kJ/mol stronger than
IR absorption of DMSO-d6 shifted toward a lower wave- the “C−H---O” H-bond formation energy of DMSO (Table
number compared to the DMSO one. The maximum no of 1). Naturally, six C−D H-bonds reduced the S�O bond order
water molecules that can be accommodated by the α-H atom more effectively compared to six C−H H-bonds. Based on this
of a single DMSO molecule is shown in Scheme S1. MD apparent inconsistency of S�O IR absorption, we have
simulation results of the α-C−H H-bond population validate explained how DMSO and DMSO-d6 interact differently with
the arguments (Figure S5). D2O instituted on α-H/D.
To support this viewpoint, DMSO and DMSO-d6 were Furthermore, the dimerization of DMSO validates our
optimized by DFT calculations. Most importantly, the S�O argument of the acidity of α-H, which we have proved by
bond orders of DMSO and DMSO-d6 of the optimized quantum calculations (Figure S4).31−37 It is well established
geometry are calculated, and their values are shown in Table 1. that DMSO forms an antiparallel dimer by involving the α-C−
H H-bond and Coulombic interactions of the opposite S�O
Table 1. Bond Order (BO) of S�O in Free and after α-H- dipole in neat and even in lower concentrations in organic
Bond Formation of DMSO and DMSO-d6 with Water solvents.56 Herein, we have also designed the dimerization that
(D2O)a occurred due to the α-C−H hydrogen bonding with the “O”
atom of DMSO (Figure S4). We have used the MP2 (cc-
BO of BO of S�O (After α-H/D α-H-bond formation
S�O H-bond formation) energy (kJ/mol) pVDZ) level of theory to optimize the antiparallel DMSO
dimer structure.56
DMSO 1.1518 1.1116 −7.0798
The unusual influence of the fluorinated solvent (HFIP) is
DMSO- 1.1517 1.1061 −7.6310
d6 observed on IR absorption of the S�O mode of DMSO
a
Hydrogen bond formation energy of α-C−H/D with D2O (column (Figure 6). Surprisingly, we have observed a lower frequency of
four). S�O IR absorbance of DMSO (0.1 M) in HFIP (black)
compared to that in D2O (red). The relative polarity of HFIP
(0.96) is lower than that of D2O (1), and the acidity (pKa =
Parallelly, we have calculated the α-H/D H-bond energy by 7.97) is also lower than that of D2O (pKa = 7). As the acidity
optimizing the geometry of DMSO and DMSO-d6 with one of HFIP is lower, it is obvious that it will form a weak H-bond
D2O molecule in neat D2O (Figure 4). The higher α-H-bond compared to that of D2O with the S�O mode. However, the
formation energy of “C−D---O” (−7.631 kJ/mol) illustrated H-bond energy of the α-H atom of DMSO with D2O (7.07 kJ/
the higher stability than that of “C−H---O” (−7.0798 kJ/mol). mol) is lower compared to that of the F atom of HFIP (8.75
Naturally, the α-D atom of DMSO-d6 forms a stronger H-bond kJ/mol), which commands the S�O IR absorption peak in
than the α-H atom of DMSO, which we have claimed in the lower frequency due to more delocalization of the C−H bond
D https://doi.org/10.1021/acs.jpcb.2c01374
J. Phys. Chem. B XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry B pubs.acs.org/JPCB Article
■
*
ASSOCIATED CONTENT
sı Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.jpcb.2c01374.
Fitted IR absorption spectra and second-order derivative
spectra of DMSO and DMSO-d6 in different solvents;
solvated structure of DMSO and DMSO-d6; exper-
Figure 6. IR absorption of DMSO in (a) D2O (blue) and (b) HFIP imental IR spectra of acetone and acetone d6 in different
(black). solvents; schematic structure of N methyl acetamide and
deuterated N methyl acetamide; dimeric structure of
over the S�O bond in HFIP. Subsequently, we have DMSO; fitting parameters of SO IR absorptions;
calculated the electric field of the S�O mode in HFIP and calculated SO IR absorption frequency of DMSO and
D2O. Surprisingly the higher electric field of S�O in HFIP DMSO-d6; and histogram of α-C−H hydrogen bond
(125 MV/cm) than in D2O (80 MV/cm) prescribed a lower populations (PDF)
frequency of S�O IR absorbance of DMSO in HFIP
compared to D2O. Therefore, combining these two effects
described the unusual S�O IR absorbances of DMSO in a
■ AUTHOR INFORMATION
Corresponding Authors
fluorinated solvent. Anjan Barman − Department of Condensed Matter Physics
To check whether the α-C−H/D effect is true only for and Materials Sciences, S. N. Bose National Centre for Basic
DMSO-DMSO-d6, we have performed a similar experiment for Sciences, Kolkata 700106, India; orcid.org/0000-0002-
structurally same acetone and acetone d6 in acetonitrile and in 4106-5658; Email: abarman@bose.res.in
D2O, shown in Figure S3. Anup Ghosh − Department of Condensed Matter Physics and
The evolution of these intrinsic α-H properties can reveal Materials Sciences, S. N. Bose National Centre for Basic
intricate details about the nature of the interactions within the Sciences, Kolkata 700106, India; orcid.org/0000-0002-
system. These could be utilized to study a diverse range of 1442-8740; Email: anupg86@gmail.com, anup.ghosh@
problems, including sulfoxide, carbonyl, ester, and amide I with bose.res.in
the α-H atom in liquid water or protic solvents.
The evaluation of the α-C−H H-bond effect on IR Author
absorption of the S�O mode is the indication of awareness Suranjana Chakrabarty − Department of Condensed Matter
of the α-(N−H/D) H-bond effect on the IR absorption of Physics and Materials Sciences, S. N. Bose National Centre
amide I in D2O. During the IR absorption experiment of for Basic Sciences, Kolkata 700106, India
biological molecules at different pH values, mostly, D2O is
Complete contact information is available at:
used as a solvent. Therefore, there are a lot of chances to
https://pubs.acs.org/10.1021/acs.jpcb.2c01374
exchange α-N−H to α-N−D (Scheme S2).57 Thus, misinter-
pretation may occur on the observed amide I IR absorption, as
Notes
it is shifted toward a lower frequency not only because of
deuteration but also due to hydrogen bond formation of N−D The authors declare no competing financial interest.
with water (Scheme S2). It is very clear from Scheme S2 that
when N−D forms a hydrogen bond with D2O, the amide I IR
absorption frequency shifted significantly more than in organic
■ ACKNOWLEDGMENTS
A.G. thanks SNBNCBS, Kolkata, for instrumental facilities and
solvents. financial support from the DST, India. A.G. also thanks Dr.
■ CONCLUSIONS
Suman Chakrabarty (SNBNCBS) for his valuable suggestions
during MD simulations.
In this work, we have inspected the isotopic influence on the
structural properties of DMSO and DMSO-d6 in different
solvent binary mixtures. Due to the various biological and
■ REFERENCES
(1) Schweizer, K. S.; Chandler, D. Vibrational dephasing and
chemical implications of DMSO and its binary mixture frequency shifts of polyatomic molecules in solution. J. Chem. Phys.
solutions, the S�O probe has gained our attention for 1982, 76, 2296−2314.
research. Isotopic exchange of α-C−H to α-C−D altered the (2) Huggins, C. M.; Pimentel, G. C. Infrared Intensity of the C - D
IR absorption of the S�O mode of DMSO in protic solvents Stretch of Chloroform-d in Various Solvents. J. Chem. Phys. 1955, 23,
895−898.
but unaltered in an organic aprotic solvent. Herein, we have (3) Akins, D. L. Theory of Raman scattering by aggregated
exposed that α-C−D forms a stronger H bond than α-C−H in molecules. J. Phys. Chem. A 1986, 90, 1530−1534.
D2O, which is the reason behind the IR absorption spectral (4) Mortensen, A.; Nielsen, O. F.; Yarwood, J.; Shelley, V.
shift of the S�O mode toward a lower frequency. Parallelly, Anomalies in the Isotropic Raman Spectra of Liquid Mixtures of
we have clarified an unusual IR absorption of the S�O mode Isotopomers of Formamide. Intermolecular Interactions in the
in fluorinated solvents. This study increases our knowledge, Carbonyl Stretching Region. J. Phys. Chem. B 1995, 99, 4435−4440.
E https://doi.org/10.1021/acs.jpcb.2c01374
J. Phys. Chem. B XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry B pubs.acs.org/JPCB Article
(5) Lord, R. C.; Nolin, B.; Stidham, H. D. Quantitative Study of the (28) Wood, C. D.; Wood, J. Pharmacologic and Biochemical
Bonding of Chloroform-d in Various Solvents by Infrared considerations of Dimethyl Sulfoxide. Ann. N. Y. Acad. Sci. 1975, 243,
Spectrometry. J. Am. Chem. Soc. 1955, 77, 1365−1368. 7−19.
(6) McHale, J. L. Intermolecular vibrational resonance coupling: (29) David, N. A. The Pharmacology of Dimethyl Sulfoxide. Annu.
Intensity borrowing in polarized Raman spectroscopy. Int. J. Quantum Rev. Pharmacol. 1972, 12, 353−374.
Chem. 1991, 40, 593−602. (30) Santos, N. C.; Coelho, J. F.; Silva, J. M.; Saldanha, C.
(7) Veas, C. H.; McHale, J. L. Intermolecular resonance coupling of Multidisciplinary utilization of dimethyl sulfoxide: pharmacological,
solute and solvent vibrational modes. J. Am. Chem. Soc. 1989, 111, cellular, and molecular aspects. Biochem. Pharmacol. 2003, 65, 1035−
7042−7046. 1041.
(8) Reichardt, C. Solvents and Solvent Effects in Organic Chemistry, (31) Fawcett, W. R.; Kloss, A. A. Attenuated total reflection fourier-
3rd ed.; Wiley-VCH, 2003. transform IR spectroscopic study of dimethyl sulfoxide self-association
(9) Buckingham, A. D. Solvent Effects in Infra-Red Spectroscopy. in acetonitrile solutions. J. Chem. Soc., Faraday Trans. 1996, 92,
Proc. R. Soc. London, Ser. A 1958, 248, 169−182. 3333−3337.
(10) Benson, A. M., Jr.; Drickalner, H. G. Stretching Vibrations in (32) Fawcett, W. R.; Kloss, A. A. Solvent-Induced Frequency Shifts
Condensed Systems: Especially Bonds Containing Hydrogen. J. Chem. in the Infrared Spectrum of Dimethyl Sulfoxide in Organic Solvents. J.
Phys. 1957, 27, 1164−1174. Phys. Chem. C 1996, 100, 2019−2024.
(11) Pullin, A. D. E. The variation of infra-red vibration frequencies (33) Sastry, M. I. S.; Singh, S. J. Self-association of dimethyl
with solvent. Spectrochim. Acta 1958, 13, 125−138. sulphoxide and its dipolar interactions with water: Raman spectral
(12) Buckingham, A. D. Solvent effects in vibrational spectroscopy. studies. J. Raman Spectrosc. 1984, 15, 80−85.
Trans. Faraday Soc. 1960, 56, 753−760. (34) Skripkin, M. Y.; Lindqvist-Reis, P.; Abbasi, A.; Mink, J.;
(13) Dijkman, F. G.; van der Maas, J. H. Inhomogeneous Persson, I.; Sandstrom, M. Basic materials physics of transparent
broadening of Morse oscillators in liquids. J. Chem. Phys. 1977, 66, conducting oxides. Dalton Trans. 2004, 4038−4049.
3871−3878. (35) Perelygin, I. S.; Itkulov, I. G.; Krauze, A. S. Temperature
(14) Zakin, M. R.; Herschbach, D. R. Relation of vibrational dependence of associative equilibria of DMSO according to raman
frequency shifts to molecular compression in liquid benzene. J. Chem. scattering spectra. Russ. J. Phys. Chem. 1991, 65, 410−414.
Phys. 1985, 83, 6540−6541. (36) Shikata, T.; Sugimoto, N. Dimeric Molecular Association of
(15) Wang, H.; Wang, L.; Shen, S.; Zhang, W.; Li, M.; Du, L.; Dimethyl Sulfoxide in Solutions of Nonpolar Liquids. J. Phys. Chem. A
Zheng, X.; Phillips, D. L. Effects of hydrogen bond and solvent 2012, 116, 990−999.
polarity on the C�O stretching of bis(2-thienyl) ketone in solution. (37) Figueroa, R. H.; Roig, E.; Szmant, H. H. Infrared study on the
J. Chem. Phys. 2012, 136, No. 124509. self-association of dimethyl sulfoxide. Spectrochem. Acta 1966, 22,
(16) Kolling, O. W. Effect of Hydrogen Bonding Solvents on the 587−592.
Infrared Absorption Band for the Fundamental Vibration of the (38) Singh, S.; Srivastava, S. K.; Singh, D. K. Raman scattering and
Carbonyl Group in 1,1,3,3-Tetramethylurea. Trans. Kans. Acad. Sci. DFT calculations used for analyzing the structural features of DMSO
1999, 102, 53−56. in water and methanol. RSC Adv. 2013, 3, 4381−4390.
(17) Almeida, L. C. J.; Santos, P. S. The anomalous solvent effect in (39) Gajda, R.; Katrusiak, A. Electrostatic matching versus close-
the vibrational spectrum of 2,3-diphenyl-cycloprop-2-enone: an packing molecular arrangement in compressed dimethyl sulfoxide
experimental and theoretical investigation. Spectrochim. Acta, Part A (DMSO) polymorphs. J. Phys. Chem. B 2009, 113, 2436−2442.
2002, 58, 3139−3148. (40) Clark, T.; Murray, J. S.; Lane, P.; Politzer, P. Why are dimethyl
(18) Nyquist, R. A.; Fouchea, H. A.; Hoffman, G. A.; Hasha, D. L. sulfoxide and dimethyl sulfone such good solvents? J. Mol. Model.
Infrared Study of β -Propiolactone in Various Solvent Systems and 2008, 14, 689−697.
Other Lactones. Appl. Spectrosc. 1991, 45, 860−867. (41) Venkataramanan, N. S.; Suvithaa, A.; Kawazoea, Y. Density
(19) Nyquist, R. A.; Putzig, C. L.; Hasha, D. L. Solvent Effect functional theory study on the dihydrogen bond cooperativity in the
Correlations for Acetone: IR versus NMR Data for the Carbonyl growth behavior of dimethyl sulfoxide clusters. J. Mol. Liq. 2018,
Group. Appl. Spectrosc. 1989, 43, 1049−1053. 454−462.
(20) Nyquist, R. A.; Putzig, C. L.; Yurga, L. Solvent and (42) Oh, K. I.; Rajesh, K.; Stanton, J. F.; Baiz, C. R. Quantifying
Concentration Effects on Carbonyl Stretching Frequencies: Ketones. hydrogen-bond populations in DMSO/water mixtures. Angew. Chem.,
Appl. Spectrosc. 1989, 43, 983−991. Int. Ed. 2017, 56, 11375−11379.
(21) Nyquist, R. A.; Clark, T. D.; Streck, R. Infrared study of alkyl (43) Oh, K. I.; Baiz, C. R. Crowding Stabilizes DMSO−Water
carboxylic acids in CCl4 and/or CHCl3 solutions. Vib. Spectrosc. 1994, Hydrogen-Bonding Interactions. J. Phys. Chem. B 2018, 122, 5984−
7, 275−286. 5990.
(22) Bruni, P.; Giorginy, E.; Maurelli, E.; Tosi, G. Fourier transform (44) Stephens, P. J.; Devlin, F. J.; Chabalowski, C. F.; Frisch, M. J.
infrared spectrometry investigation of solvent effect on NH and CO Ab Initio Calculation of Vibrational Absorption and Circular
stretching modes in N-acylaminopyridines. Vib. Spectrosc. 1996, 12, Dichroism Spectra Using Density Functional Force Fields. J. Phys.
249−255. Chem. A 1994, 98, 11623−11627.
(23) Nyquist, R. A.; Putzig, C. L.; Clark, T. D. Infrared study of 1,3- (45) Becke, A. D. Density-functional thermochemistry. III. The role
dimethyl-2-imidazolidinone in various solvents. Vib. Spectrosc. 1996, of exact exchange. J. Chem. Phys. 1993, 98, 5648−5652.
12, 81−91. (46) Chakrabarty, S.; Maity, S.; Yazhini, D.; Ghosh, A. Surface-
(24) Engberts, J. B. F. N.; Famini, G. R.; Perjessy, A.; Wilson, L. Y. J. Directed Disparity in Self-Assembled Structures of Small-Peptide L-
Solvent effects on stretching frequencies of some 1-substituted 2- Glutathione on Gold and Silver Nanoparticles. Langmuir 2020, 36,
pyrrolidinones. J. Phys. Org. Chem. 1998, 11, 261−272. 11255−11261.
(25) Ghosh, A.; Cohn, B.; Prasad, A. K.; Chuntonov, L. Quantifying (47) Van Der Spoel, D.; Lindahl, E.; Hess, B.; Groenhof, G.; Mark,
conformations of ester vibrational probes with hydrogen-bond- A. E.; Berendsen, H. J. C. GROMACS: Fast, Flexible, and Free. J.
induced Fermi resonances. J. Chem. Phys. 2018, 149, No. 184501. Comput. Chem. 2005, 26, 1701−1718.
(26) Ghosh, A. Vibrational Coupling on Stepwise Hydrogen Bond (48) Hess, B.; Kutzner, C.; van der Spoel, D.; Lindahl, E.
Formation of Amide I. J. Phys. Chem. B 2019, 123, 7771−7776. GROMACS 4: Algorithms for Highly Efficient, Load-Balanced, and
(27) Jacob, S. W.; Herschler, R. Pharmacology of DMSO. Scalable Molecular Simulation. J. Chem. Theory Comput. 2008, 4,
Cryobiology 1986, 23, 14−27. 435−447.
F https://doi.org/10.1021/acs.jpcb.2c01374
J. Phys. Chem. B XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry B pubs.acs.org/JPCB Article
(49) Wang, J.; Wolf, R. M.; Caldwell, J. W.; Kollman, P. A.; Case, D.
A. Development and Testing of a General Amber Force Field. J.
Comput. Chem. 2004, 25, 1157−1174.
(50) Caleman, C.; van Maaren, P. J.; Hong, M.; Hub, J. S.; Costa, L.
T.; van der Spoel, D. Force Field Benchmark of Organic Liquids:
Density, Enthalpy of Vaporization, Heat Capacities, Surface Tension,
Isothermal Compressibility, Volumetric Expansion Coefficient, and
Dielectric Constant. J. Chem. Theory Comput. 2012, 8, 61−74.
(51) van der Spoel, D.; van Maaren, P. J.; Caleman, C. GROMACS
Molecule & Liquid Database. Bioinformatics 2012, 28, 752−753.
(52) Horn, H. W.; Swope, W. C.; Pitera, J. W.; Madura, J. D.; Dick,
T. J.; Hura, G. L.; Head-Gordon, T. Development of an Improved
Four-Site Water Model for Biomolecular Simulations: TIP4P-Ew. J.
Chem. Phys. 2004, 120, 9665−9678.
(53) Bussi, G.; Donadio, D.; Parrinello, M. Canonical Sampling
through Velocity Rescaling. J. Chem. Phys. 2007, 126, No. 014101.
(54) Parrinello, M.; Rahman, A. Polymorphic Transitions in Single
Crystals: A New Molecular Dynamics Method. J. Appl. Phys. 1981, 52,
7182−7190.
(55) Fried, S. D.; Bagchi, S.; Boxer, S. G. Measuring Electrostatic
Fields in Both Hydrogen-Bonding and Non-Hydrogen-Bonding
Environments Using Carbonyl Vibrational Probes. J. Am. Chem. Soc.
2013, 135, 11181−11192.
(56) Chakrabarty, S.; Deshmukh, S. H.; Barman, A.; Bagchi, S.;
Ghosh, A. On−Off Infrared Absorption of the S�O Vibrational
Probe of Dimethyl Sulfoxide. J. Phys. Chem. B 2022, 126, 4501−4508.
(57) Nielsen, S. O. Hydrogen-deuterium exchange in N-methyl-
acetamide. Biochim. Biophysics. Acta 1960, 37, 146−147.
Recommended by ACS
Cationic Effects on the Structural Dynamics of the
Metal Ion–Crown Ether Complexes Investigated by
Ultrafast Infrared Spectroscopy
Dexia Zhou, Hongtao Bian, et al.
NOVEMBER 11, 2021
THE JOURNAL OF PHYSICAL CHEMISTRY B READ
G https://doi.org/10.1021/acs.jpcb.2c01374
J. Phys. Chem. B XXXX, XXX, XXX−XXX
* Unknown * | ACSJCA | JCA11.2.5208/W Library-x64 | manuscript.3f (R5.1.i4:5009 | 2.1) 2021/10/27 08:51:00 | PROD-WS-121 | rq_3200684 | 6/03/2022 22:22:29 | 8 | JCA-DEFAULT
pubs.acs.org/JPCB Article
3 Suranjana Chakrabarty,∥ Samadhan H. Deshmukh,∥ Anjan Barman, Sayan Bagchi,* and Anup Ghosh*
Cite This: https://doi.org/10.1021/acs.jpcb.1c10558 Read Online
21
22
■ INTRODUCTION
Dimethyl sulfoxide (DMSO) is one of the polar aprotic
dictate molecular behaviors, enhancing polarizability in the
molecules.15−17 Having the propensity toward intermolecular
48
49
23 molecules with a large dipole moment.1 In the field of associations, either with itself or with other molecules (such as 50
24 pharmacology, toxicology, biochemistry, and other analytical water), DMSO occasionally exhibits abnormal physicochemical 51
25 applications, DMSO is extensively used as a prospective properties and can form DMSO−water aggregates with various 52
26 cosolvent in various chemical reactions.1,2 Being a cryopro- stoichiometries.18−21 At a higher DMSO concentration, the 53
27 tective agent, DMSO averts the development of intracellular hydrogen-bond network of water is ruptured and stable water− 54
28 and extracellular crystals during the cell and tissue freezing DMSO complexes are formed.22−24 For a detailed exploration 55
29 processes.3−7 In addition, DMSO has various therapeutic and of the structures of DMSO, infrared (IR) absorption and 56
30 pharmaceutical uses, such as anti-inflammatory, analgesic, Raman scattering spectroscopies were employed.25−30 These 57
31 antibacterial, antifungal, antiviral, and membrane penetration experiments revealed that DMSO can form antiparallel dimers 58
32 enhancement properties that remain underexplored by and chain-like associations both in the pure liquid state and in 59
33 pharmaceutical developers and ophthalmologists.8 In bio- solutions.25−31 60
34 chemical and biophysical assays, it is reported that DMSO Although several vibrational spectroscopic reports on 61
35 concentrations influence protein aggregation and stability.9,10 DMSO have been published, a consensus regarding the peak 62
36 Experimental and computational studies revealed that DMSO assignments has yet to be reached.26,30 The band at 1027 cm−1 63
37 induces structural and conformational changes in pro- is assigned to the SO stretch arising from parallel chain-like 64
38 teins.11−13 These wide ranges of applications have generated molecular associations. For antiparallel cyclic associations, the 65
39 a profound interest in the structural behavior of DMSO, either bands at 1058 and 1044 cm−1 are attributed to the out-of- 66
40 neat or in the solution state, in recent years.14−25
41 The structural properties and the dimerization strategies of
42 DMSO have been previously reported.25−30 The intermolec- Received: December 14, 2021
43 ular geometry of DMSO in the liquid state, determined from Revised: May 24, 2022
44 X-ray and neutron diffraction studies, showed that DMSO
45 mostly prefers to arrange in the antiparallel ordering of SO
46 dipoles in the liquid state.14 The formation of hydrogen bonds
47 between DMSO and water molecules has been shown to
67 phase and in-phase SO stretch modes, respectively.31 The spectroscopic approach to elucidate the details of the 130
68 out-of-phase SO stretch of the cyclic dimer is detected at microsolvation environment of DMSO despite the complexity 131
69 1063 cm−1 in the IR spectra.30 Toshiyuki Shikata et al. of overlapping bands in the spectra. 132
■
70 reported that associations of DMSO dimers are formed in the
71 pure liquid and also in the presence of nonpolar solvents.32 MATERIALS AND METHODS 133
72 When dissolved in CCl4, antiparallel cyclic dimers of DMSO
73 are formed due to limited self-interaction,33 with peaks at 1060 Sample Preparation. Dimethyl sulfoxide (DMSO 99.9%), 134
74 and 1000 cm−1 arising from the SO stretch of monomers thiophene oxide (TPSO, 99.9%), and all solvents (99.9%) were 135
75 and dimers, respectively.33 Recently, it has been reported that bought from Sigma Aldrich. Karl Fischer titration (coulo- 136
76 DMSO dimerization happens even at low DMSO concen- metric) was performed to quantify the moisture present in 137
77 trations.34 Gajda and Katrusiak used single-crystal XRD at high used solvents. These titrations indicated negligible water 138
78 pressure to study the detailed dimeric structure of DMSO.35 content. Solutions of DMSO and TPSO at a 0.1M 139
79 The calculated electrostatic potentials revealed a large positive concentration in all solvents were prepared for IR absorption 140
80 partial charge on the “S” atom and a negative partial charge on measurements. 141
81 the “O” atom, through which a variety of intermolecular IR Spectroscopy. All IR absorption spectra were collected 142
82 electrostatic interactions can occur.36 Natarajan et al. on a Bruker 70S FTIR spectrometer equipped with a DTGS 143
83 investigated the structure and stability of DMSO molecules detector at 2 cm−1 resolutions. A 50 μL volume of each 144
84 using molecular electrostatic potential and atoms-in-molecules solution was placed between two CaF2 windows separated by a 145
85 (AIM) analysis with the employment of density functional 64 μm Teflon spacer, housed in a homemade sample cell. 146
86 theory (DFT) studies.37 Along with the SO stretch region, the OH stretch region was 147
87 Another interesting aspect of DMSO is the structural and also monitored to identify any small amount of water 148
physicochemical behavior difference in aprotic and protic contamination. A flat baseline in the OH stretch region 149
88
indicated that the samples were not contaminated by water. 150
89 solvents. To this end, similar behavioral changes can be found
DFT Calculations. All initial geometries of DMSO, TPSO, 151
90 in molecules containing a carbonyl (CO) moiety where the
and their water complexes were optimized preliminarily by 152
91 less polarizable C atom replaces the S atom. Furthermore,
density functional theory and B3LYP/6-311G(d) calculations 153
92 studies on hydrogen-bond occupancy have shown that SO-
using the Gaussian 09 suite of programs to estimate the dipole 154
93 and CO-containing molecules can form one or more
moments of the DMSO monomer and the dimer.50,51 Further, 155
94 hydrogen bonds in a protic solvent.38−49 Recently, a
to expose the dissociation energies of one hydrogen bond and 156
95 temperature- and concentration-dependent study on DMSO/
two hydrogen bond complexes of DMSO in water, the “scan” 157
96 water mixtures indicated the presence of four distinct forms of
analyses were performed on the B3LYP/6-311G(d) optimized 158
97 DMSO, namely self-associated (aggregated), free-monomer
structures.52 159
98 (0HB), singly hydrogen-bonded (1HB), and doubly hydrogen-
MD Simulations. Classical molecular dynamics simulation 160
99 bonded (2HB) species.38,39 On the contrary, a recent study has
was conducted on the GROMACS 2016.5 package.53 161
100 shown that CO-containing esters are exclusively doubly
Parameters for DMSO were derived from quantum chemical 162
101 hydrogen-bonded in water.40 The reported differences in
calculation followed by RESP charge fitting by ANTECHAM- 163
102 structures’ hydrogen-bond occupancies upon changing the C
BER software to produce the general AMBER force field.54 164
103 atom in CO to the S atom in SO warrants a detailed study
Simulation box was prepared by adding 7 DMSO molecules 165
104 of DMSO in the solution state. Moreover, a significant SO
with 4000 water molecules. Here, water molecules were 166
105 IR absorbance signature for DMSO dimers, either neat or in
modeled using the TIP4P-EW model. For short-range 167
106 the solution state, has been surmised in the previous
electrostatic interactions, a periodic boundary condition with 168
107 studies.25−38 Such an observation cannot be explained based
a cutoff distance of 16 Å was used. Before the simulation, an 169
108 on the symmetric structure of the DMSO dimer.
energy minimization using a steepest-descent algorithm was 170
109 To identify the structural conformers underlying the performed, followed by equilibrium in the NVT ensemble at 171
110 overlapping DMSO vibrational spectra and to determine 300 K for 1 ns using a velocity rescale thermostat and 172
111 their structural details and hydrogen-bonding nature in water, equilibrium for the NPT ensemble at 300 K and 1 bar using a 173
112 we performed a detailed investigation using IR absorption Parrinello−Rahman barostat for 5 ns. LINCS was used to 174
113 measurements, DFT calculations, and MD simulations on neat constrain all covalently bonded hydrogen atoms. MD run was 175
114 DMSO and DMSO dissolved in various solvents, namely carried out for 10 ns. A hydrogen-bond analysis was performed 176
115 carbon tetrachloride (CCl4), dichloromethane (DCM), and on GROMACS with the last 5 ns of the trajectory where a 177
116 water (D2O). We combined IR spectroscopy with ab initio cutoff distance of 3.5 Å and an angle cutoff of 30° were used. 178
■
117 quantum mechanical calculations to analyze the monomer−
118 dimer equilibrium in different solvents to show that the relative
119 intensity of the SO stretch and the CH3 rocking modes RESULTS AND DISCUSSION 179
120 provide an estimate of the DMSO dimer present in the The delicate balance in the noncovalent (solute−solute and 180
121 solution. Furthermore, in the presence of a protic solvent like solute−solvent) interactions drives the formation and dis- 181
122 water, we showed that the extent of dimerization is drastically ruption of intermolecular interactions of DMSO dissolved in 182
123 reduced due to the formation of DMSO−water hydrogen different solvents. To resolve the detailed molecular structure 183
124 bonds. We subsequently analyzed the hydrogen-bonded of DMSO, in both neat and solution states, we performed a 184
125 conformations of DMSO in water. Interestingly, we observed series of infrared absorption experiments on 0.1 M DMSO in 185
126 that DMSO interacts predominantly with two water molecules different solvents (Figures 1 and 2). The second derivatives of 186 f1f2
127 to form a double hydrogen-bonded conformer. Such the broad overlapping IR absorption spectra (Figures 1b, and 187
128 conformations are well known for molecules containing the 2b,2d,2f) in the 950−1100 cm−1 region indicate four 188
129 CO moiety.40 Overall, our findings suggest an efficient underlying peaks. Each spectrum is fitted with four peaks 189
B https://doi.org/10.1021/acs.jpcb.1c10558
J. Phys. Chem. B XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry B pubs.acs.org/JPCB Article
(Table S1). The fitted peaks in Figures 1 and 2 are assigned to 193
the SO symmetric stretching mode and three CH3 rocking 194
modes based on previous reports.31−34,38,39 In neat DMSO 195
(Figure 1), three rocking modes and the symmetric SO 196
mode are observed at ∼1019, 1030, 1045, and 1059 cm−1, 197
respectively. Interestingly, the intensity of the SO mode is 198
comparable to that of the CH3 rocking modes even though the 199
transition dipole moment of SO is larger than those of the 200
rocking modes. One plausible explanation for the comparable 201
intensities in Figure 1 can be that neat DMSO predominantly 202
exists as symmetric dimers even at low concentrations, as 203
reported previously.34 Theoretical calculations (discussed 204
later) suggest that the SO mode of antiparallel dimers is 205
not IR active. However, when dissolved in another solvent, the 206
dimers should not be formed at a low DMSO concentration. 207
Indeed, the IR absorption spectra of a dilute solution of 208
DMSO in different solvents show that the intensity of the S 209
O stretch band is distinctly larger than those of the CH3 210
rocking modes (Figure 2). 211
We performed DFT calculations at the B3LYP/6-311G(d) 212
level of theory50,51 to estimate the dipole moments of the 213
DMSO monomer and the dimer. Our calculations suggest that 214
the dipole moment is zero for an antiparallel DMSO dimer, 215
whereas a considerable dipole moment (5.6D) is estimated for 216
the monomer. We surmise that the SO peak in the IR 217
Figure 1. (a) IR absorption spectrum of neat DMSO (black). The spectrum (Figure 1) arises from the monomeric species, which 218
experimental spectrum was fitted to four peaks corresponding to the are in equilibrium with DMSO dimers in neat DMSO. The 219
SO stretch (orange) and the CH3 rocking modes (green, blue, and self-association of the monomers to form antiparallel dimers is 220
magenta) using Voigt line shape functions. During the fitting protocol, quite likely due to the high polarity of the SO bond. In 221
the peak positions are assigned based on the (b) second derivative addition, the dynamic monomer−dimer equilibrium happens 222
spectrum.
due to the intrinsic molecular fluctuations. 223
We further evaluated the ratios of SO peak areas to the 224
190 using Voigt line shape functions. The fits are shown in Figures integrated peak areas of the rest of the bands in the 950−1100 225
191 S1 and S2. The peak positions are fixed to the frequencies cm−1 region (Figure 3a). The ratio has a minimum value for 226 f3
192 corresponding to the minima in the second derivative spectra neat DMSO and subsequently increases with the polarity of 227
Figure 2. Black curves denote the IR absorption spectra of DMSO in (a) CCl4, (c) DCM, and (e) D2O. The experimental IR spectra are fitted to
four peaks using Voigt line shape functions. The orange peaks correspond to the SO stretch, and the green, blue, and magenta peaks correspond
to the CH3 rocking modes. The peak positions during the fitting protocol are obtained from the second derivative spectra of DMSO in (b) CCl4,
(d) DCM, and (f) D2O.
C https://doi.org/10.1021/acs.jpcb.1c10558
J. Phys. Chem. B XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry B pubs.acs.org/JPCB Article
Figure 3. (a) Ratio (R) of the IR absorption area of the SO stretching mode and the integrated area of the rocking mode at around 1045 cm−1.
(b) Ratio (R) of the integrated IR absorption area of the SO stretching mode and the integrated area of all overlapping rocking modes.
228 the solvent. In water, the ratio is remarkably enhanced by a To confirm our arguments, we performed control IR 254
229 factor of ∼6 from that in neat DMSO. The polarity and the absorption experiments and DFT calculations on TPSO. The 255
230 hydrogen-bonding capacity of the solvent play a key role in the DFT calculations suggest that the bulky five-membered ring of 256
231 monomer−dimer equilibrium of DMSO. A nonpolar aprotic TPSO provided steric hindrance when two TPSO monomers 257
232 solvent like CCl4 or DCM weakly interacts with the solute approached one another, making the dimerization unfeasible 258
233 (DMSO), but a polar protic solvent like water can form even in neat TPSO. We performed a series of IR absorption 259
234 hydrogen bonds with the SO moiety of DMSO (see Figure experiments of TPSO and analyzed the 950−1100 cm−1 260
S3 in the Supporting Information). Therefore, the propensity spectral region (same frequency range as DMSO). The IR 261
235
spectra of neat TPSO show three peaks (Figure 5a), one 262 f5
236 of the monomeric species is higher in water than in other
arising from the SO stretch and the other two arising from 263
237 organic solvents. In neat DMSO, two polar DMSO molecules
CH2 rocking modes. Interestingly the absorption of the SO 264
238 can form the antiparallel dimer, giving rise to an increase in the mode was stronger than that of the rocking modes. Similar 265
239 dimeric species. This effect is corroborated by the ratios shown results are obtained when TPSO was dissolved at a low 266
240 in Figure 3. The trend remains the almost same if the area of concentration in other solvents (Figure 5b−d). Herein, we 267
241 one of the rocking modes is considered (Figure 3a) instead of fitted TPSO IR absorption spectra based on the quality (Figure 268
242 the integrated area (Figure 3b). The small discrepancy in the S4) and second-order derivative spectra (Figure S5). Figure 6 269 f6
243 ratios for CCl4 and DCM solvents (Figure 3b) most plausibly shows the ratios of the SO peak area to that of the rocking 270
244 arises due to the effect of the coupling between the SO mode (∼990 cm−1). The similar ratios in TPSO, CCl4, and 271
245 stretch and the CH3 rocking modes (see the Supporting DCM corroborate the DFT prediction of the dimerization 272
246 Information). However, the overall trend remains the same. infeasibility in neat TPSO. The lower polarity of TPSO when 273
247 To elucidate this phenomenon, we calculated the dimeriza- compared with DMSO, along with the steric hindrance in 274
248 tion energy of DMSO and compared it with solvation energy TPSO to the formation of hydrogen bonds might have resulted 275
f4 249 in organic and aqueous solvents (Figure 4). When compared in a lower ratio in water. However, the control experiments in 276
250 to aprotic solvents, Figure 4 clearly shows that the stability TPSO provide additional support for the formation of an 277
251 obtained from the dimerization of DMSO is much smaller than antiparallel DMSO dimer in a neat solution. 278
252 that obtained from the solvation of DMSO by water molecules, To resolve the hydrogen-bonded populations of DMSO, we 279
253 thereby increasing the ratio in water (Figure 3). further performed MD simulation with GROMACS (version 280
4.6.5) in neat water. The hydrogen-bond occupancy of DMSO 281
has been explored by keeping the angle of H-donor−acceptor 282
(H−O−O) between 0 and 30° and the O (donor)−O 283
(acceptor) distance was below 3.5Å.40 The force-field 284
parameters for DMSO were obtained from virtualchemis- 285
try.org.41 The TIP4P-Ew model was used for water.42 The 286
simulations reveal the hydrogen-bond formation and breaking 287
mechanisms of DMSO in neat water. The hydrogen-bond 288
analysis from the MD simulation trajectories demonstrates that 289
SO can form zero, one, or two HB complexes with the 290
surrounding water molecules.44−48 Interestingly, the free 291
DMSO (0.35%) and 1HB complexes (3.5%) are found to be 292
negligibly small in comparison to 2HB (96.15%) populations. 293
Similarly, TPSO predominantly forms 2HB populations (92%) 294
compared to 1HB (8%) and 0HB (0%) in neat water. These 295
results support our experimental results wherein we mentioned 296
Figure 4. Solvation energy (orange) and dimerization energy (green) only one SO stretching IR absorption peak for both DMSO 297
of DMSO in different solvents. (Figure 2e) and TPSO (Figure 5d) in neat water. 298
D https://doi.org/10.1021/acs.jpcb.1c10558
J. Phys. Chem. B XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry B pubs.acs.org/JPCB Article
Figure 5. IR absorption spectra (a) of neat TPSO (black) and TPSO in (b) CCl4 (c) DCM, and (d) D2O. The experimental spectra were fitted to
three peaks corresponding to the SO stretch (red) and the CH2 rocking modes (green, blue) using Voigt line shape functions. During the fitting
protocol, the peak positions are assigned based on the second derivative spectrum shown in Figure S5.
E https://doi.org/10.1021/acs.jpcb.1c10558
J. Phys. Chem. B XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry B pubs.acs.org/JPCB Article
Table 1. Thermodynamical Data Obtained from DFT spectra of TPSO, details of coupling, and mathematical 373
Calculations calculations of H-bond populations (PDF) 374
complexes
1HB
formation
energy
(kJ/mol)
−24.33
change of Gibbs
free energy
(kJ/mol)
−22.25
equilibrium
constant
(mol)
7942.63
populations
(%)
0.0025
■ AUTHOR INFORMATION
Corresponding Authors
375
376
2HB −22.72 −16.29 718.38 99.9975 Sayan Bagchi − Physical and Materials Chemistry Division, 377
CSIR-National Chemical Laboratory, Pune 411008, India; 378
330 clear parity of H-bond populations is undoubtedly recognized. Academy of Scientific and Innovative Research (AcSIR), 379
331 Subsequently, we confirmed conclusive 2HB populations of Ghaziabad 201002, India; orcid.org/0000-0001-6932- 380
332 DMSO in neat water, which is also logically explained in 3113; Email: s.bagchi@ncl.res.in 381
333 Scheme S1. Especially, in the mixing process, hydrogen Anup Ghosh − Department of Condensed Matter Physics and 382
334 bonding is simply transferred from water−water interactions Materials Sciences, S. N. Bose National Centre for Basic 383
335 (18.5 kJ/mol) to water−DMSO interactions as the H-bond Sciences, Kolkata 700106, India; Physical and Materials 384
336 interaction of water−water is lower than the H-bond Chemistry Division, CSIR-National Chemical Laboratory, 385
interaction of water−DMSO in neat water (22.72 kJ/mol). Pune 411008, India; Academy of Scientific and Innovative 386
■
337
Research (AcSIR), Ghaziabad 201002, India; orcid.org/ 387
338 CONCLUSIONS 0000-0002-1442-8740; Email: anupg86@gmail.com 388
■
358
359 results provided an overview of the influence of solvent polarity
360 in the dissociation of DMSO dimers. We envisage that our ACKNOWLEDGMENTS 407
361 findings will help in future studies based on the solvent effects A.G. thanks SNBNCBS, Kolkata, for instrumental facilities and 408
362 on dynamical configurations of biomolecules in DMSO−- financial support from the DST, India. S.B. thanks CSIR-NCL 409
s1 363 water mixtures (Scheme 1). and SERB, India, (SR/S2/RJN-142/2012 and EMR/2016/ 410
000576) for financial support. 411
■
Scheme 1. Chemical Structure of the Sulfoxide Compound
REFERENCES 412
(1) Vaisman, I. I.; Berkowitz, M. L. J. Local structural order and 413
molecular associations in water-DMSO mixtures. Molecular dynamics 414
study. J. Am. Chem. Soc. 1992, 114, 7889−7896. 415
(2) Chalaris, M.; Marinakis, S.; Dellis, D. Temperature effects on the 416
structure and dynamics of liquid dimethyl sulfoxide: A molecular 417
dynamics study. Fluid Phase Equilib. 2008, 267, 47−60. 418
■
(3) Lovelock, J. E.; Bishop, M. W. H. Prevention of Freezing 419
Damage to Living Cells by Dimethyl Sulphoxide. Nature 1959, 183, 420
364 ASSOCIATED CONTENT 1394−1395. 421
365 *
sı Supporting Information (4) Rall, W. F.; Fahy, G. M. Ice-free cryopreservation of mouse 422
366 The Supporting Information is available free of charge at embryos at −196 °C by vitrification. Nature 1985, 313, 573−575. 423
367 https://pubs.acs.org/doi/10.1021/acs.jpcb.1c10558. (5) Fahy, G. M. Cryoprotectant toxicity neutralization. Cryobiology 424
2010, 60, S45−53. 425
368 Fitted linear IR spectra of neat DMSO, fitted linear IR (6) Rammler, D. H.; Zaffaroni, A. Biological implications of DMSO 426
369 spectra of DMSO in different solvents, table of linear IR based on a review of its chemical properties. Ann. N. Y. Acad. Sci. 427
370 absorption peak positions of DMSO and TPSO, fitted 1967, 141, 13−23. 428
371 linear IR spectra of DMSO in DCM and D2O mixtures, (7) Jacob, S. W.; Herschler, R. Pharmacology of DMSO. Cryobiology 429
372 fitted linear IR spectra of TPSO, second-order derivative 1986, 23, 14−27. 430
F https://doi.org/10.1021/acs.jpcb.1c10558
J. Phys. Chem. B XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry B pubs.acs.org/JPCB Article
431 (8) Hoang, C.; Nguyen, A. K.; Nguyen, T. Q.; Fang, W.; Han, B.; (28) Fine, G.; Mirone, P. Short-range orientation effects in dipolar 500
432 Hoang, B. X.; Tran, H. D. Application of Dimethyl Sulfoxide as a aprotic liquids–HI. Intermolecular coupling of vibrations in sulfoxides, 501
433 Therapeutic Agent and Drug Vehicle for Eye Diseases. J. Ocul. sulfones, nitriles and other compounds. Spectrochim. Acta, Part A 502
434 Pharmacol. Ther. 2021, 37, 441−451. 1976, 32A, 625−629. 503
435 (9) Chan, D.-H.; Madeline, E. K.; Kirsty, J. M.; Andrew, W. M.; (29) Sastry, M. I. S.; Singh, S. J. Self-association of dimethyl 504
436 Dijana, M. V.; Anthony, G. C.; Chris, A. Effect of DMSO on Protein sulphoxide and its dipolar interactions with water: Raman spectral 505
437 Structure and Interactions Assessed By Collision-induced Dissocia- studies. J. Raman Spectrosc. 1984, 15, 80−85. 506
438 tion and Unfolding. Anal. Chem. 2017, 89, 9976−9983. (30) Skripkin, M. Y.; Lindqvist-Reis, P.; Abbasi, A.; Mink, J.; 507
439 (10) Tjernberg, A.; Markova, N.; Griffiths, W. J.; Hallén, D. DMSO- Persson, I.; Sandstrom, M. Basic materials physics of transparent 508
440 Related Effects in Protein Characterization. SLAS Discovery 2006, 11, conducting oxides. Dalton Trans. 2004, 4038−4049. 509
441 131−137. (31) Perelygin, I. S.; Itkulov, I. G.; Krauze, A. S. Temperature 510
442 (11) Ghosh, S.; Chattoraj, S.; Chowdhury, R.; Bhattacharyya, K. dependence of associative equilibria of DMSO according to Raman 511
443 Structure and dynamics of lysozyme in DMSO−water binary mixture: scattering spectra. Russ. J. Phys. Chem. 1991, 65, 410−414. 512
444 fluorescence correlation spectroscopy. RSC Adv. 2014, 4, 14378− (32) Shikata, T.; Sugimoto, N. Dimeric Molecular Association of 513
445 14384. Dimethyl Sulfoxide in Solutions of Nonpolar Liquids. J. Phys. Chem. A 514
446 (12) Roy, S.; Jana, B.; Bagchi, B. Dimethyl sulfoxide induced 2012, 116, 990−999. 515
447 structural transformations and non-monotonic concentration depend- (33) Figueroa, R. H.; Roig, E.; Szmant, H. H. Infrared study on the 516
448 ence of conformational fluctuation around active site of lysozyme. J. self-association of dimethyl sulfoxide. Spectrochim. Acta, Part A 1966, 517
449 Chem. Phys. 2012, 136, No. 115103. 22, 587−592. 518
450 (13) Sterling, H. J.; Prell, J. S.; Cassou, C. A.; Williams, E. R. Protein (34) Singh, S.; Srivastava, S. K.; Singh, D. K. Raman scattering and 519
451 conformation and supercharging with DMSO from aqueous solution. DFT calculations used for analyzing the structural features of DMSO 520
452 J. Am. Soc. Mass Spectrom. 2011, 22, 1178−1186. in water and methanol. RSC Adv. 2013, 3, 4381−4390. 521
453 (14) Onthong, U.; Megyes, T.; Bakó, I.; Radnai, T.; Grósz, T.; (35) Gajda, R.; Katrusiak, A. Electrostatic matching versus close- 522
454 Hermansson, K.; Probst, M. X-ray and neutron diffraction studies and packing molecular arrangement in compressed dimethyl sulfoxide 523
455 molecular dynamics simulation of liquid DMSO. Phys. Chem. Chem. (DMSO) polymorphs. J. Phys. Chem. B 2009, 113, 2436−2442. 524
456 Phys. 2004, 6, 2136−2144. (36) Clark, T.; Murray, J. S.; Lane, P.; Politzer, P. Why are dimethyl 525
457 (15) Soper, A. K.; Luzar, A. Orientation of Water Molecules around sulfoxide and dimethyl sulfone such good solvents? J. Mol. Model. 526
458 Small Polar and Nonpolar Groups in Solution: A Neutron Diffraction 2008, 14, 689−697. 527
459 and Computer Simulation Study. J. Phys. Chem. A 1996, 100, 1357− (37) Venkataramanan, N. S.; Suvithaa, A.; Kawazoea, Y. Density 528
460 1367. functional theory study on the dihydrogen bond cooperativity in the 529
461 (16) Mancera, R. L.; Chalaris, M.; Samios, J. The Concentration growth behavior of dimethyl sulfoxide clusters. J. Mol. Liq. 2018, 249, 530
462 Effect on the ‘Hydrophobic’ and ‘Hydrophilic’ Behaviour around 454−462. 531
463 DMSO in Dilute Aqueous DMSO Solutions. A Computer Simulation (38) Oh, K. I.; Rajesh, K.; Stanton, J. F.; Baiz, C. R. Quantifying 532
464 Study. J. Mol. Liq. 2004, 110, 147−153. hydrogen-bond populations in DMSO/water mixtures. Angew. Chem. 533
465 (17) Kirchner, B.; Hutter, J. Solvent Effects on Electronic Properties Int. Ed. 2017, 56, 11375−11379. 534
466 from Wannier Functions in a Dimethyl Sulfoxide/Water Mixture. J. (39) Oh, K. I.; Baiz, C. R. Crowding Stabilizes DMSO−Water 535
467 Chem. Phys. 2004, 121, 5133−5142. Hydrogen-Bonding Interactions. J. Phys. Chem. B 2018, 122, 5984− 536
468 (18) Borin, I. A.; Skaf, M. S. Molecular association between water 5990. 537
469 and dimethyl sulfoxide in solution: A molecular dynamics simulation (40) Ghosh, A.; Cohn, B.; Prasad, A. K.; Chuntonov, L. Quantifying 538
470 study. J. Chem. Phys. 1999, 110, 6412−6420. conformations of ester vibrational probes with hydrogen-bond- 539
471 (19) Luzar, A.; Chandler, D. Structure and hydrogen bond dynamics induced Fermi resonances. J. Chem. Phys. 2018, 149, 184501−184511. 540
472 of water dimethyl sulfoxide mixtures by computer simulations. J. (41) Ghosh, A. Vibrational Coupling on Stepwise Hydrogen Bond 541
473 Chem. Phys. 1993, 98, 8160−8173. Formation of Amide I. J. Phys. Chem. B 2019, 123, 7771−7776. 542
474 (20) Vishnyakov, A.; Lyubartsev, A. P.; Laaksonen, A. Molecular (42) Kashid, S. M.; Jin, G. Y.; Chakrabarty, S.; Bagchi, S.; Kim, Y. S. 543
475 Dynamics Simulations of Dimethyl Sulfoxide and Dimethyl Two-Dimensional Infrared Spectroscopy Reveals Cosolvent Compo- 544
476 Sulfoxide−Water Mixture. J. Phys. Chem. A 2001, 105, 1702−1710. sition-Dependent Crossover in Intermolecular HydrogenBond. J. 545
477 (21) Kirchner, B.; Reiher, M. The Secret of Dimethyl Sulfoxide− Phys. Chem. Lett. 2017, 8, 1604−1609. 546
478 Water Mixtures. A Quantum Chemical Study of 1DMSO−nWater (43) Kashid, S. M.; Jin, G. Y.; Bagchi, S.; Kim, Y. S. Cosolvent 547
479 Clusters. J. Am. Chem. Soc. 2002, 124, 6206−6215. Effects on Solute−Solvent Hydrogen-Bond Dynamics: Ultrafast 2D 548
480 (22) Baker, E. S.; Jonas, J. Transport and relaxation properties of IR Investigations. J. Phys. Chem. B 2015, 119, 15334−15343. 549
481 dimethyl sulfoxide-water mixtures at high pressure. J. Phys. Chem. B (44) Borin, I. A.; Skaf, M. S. Molecular Association between Water 550
482 1985, 89, 1730−1735. and Dimethyl Sulfoxide in Solution: The Librational Dynamics of 551
483 (23) Chowdhuri, S.; Pattanayak, S. K. Pressure dependence on the Water. Chem. Phys. Lett. 1998, 296, 125−130. 552
484 single-particle dynamics and hydrogen bond structural relaxation of (45) Luzar, A.; Chandler, D. Structure and Hydrogen-Bond 553
485 water-DMSO mixtures under ambient and cold conditions. Mol. Phys. Dynamics of Water-Dimethyl Sulfoxide Mixtures by Computer- 554
486 2013, 111, 135−146. Simulations. J. Chem. Phys. 1993, 98, 8160−8173. 555
487 (24) Kirchner, B.; Hutter, J. The Structure of a DMSO-Water (46) Kashid, S. M.; Singh, R. K.; Kwon, H.; Kim, Y. S.; Mukherjee, 556
488 Mixture from Car-Parrinello Simulations. Chem. Phys. Lett. 2002, 364, A.; Bagchi, S. Arresting an Unusual Amide Tautomer Using Divalent 557
489 497−502. Cations. J. Phys. Chem. B 2019, 123, 8419−8424. 558
490 (25) Forel, M. -T.; Tranquille, M. Spectres de vibration du (47) Kashid, S. M.; Singh, R. K.; Kwon, H.; Seol, J. G.; Kim, Y. S.; 559
491 diméthylsulfoxyde et dn diméthylsulfoxyde-d6. Spectrochim. Acta, Mukherjee, A.; Bagchi, S. Reply to “Comment on ‘Arresting an 560
492 Part A 1970, 26, 1023−1034. Unusual Amide Tautomer Using Divalent Cations’”. J. Phys. Chem. B 561
493 (26) Fawcett, W. R.; Kloss, A. A. Attenuated total reflection fourier- 2021, 125, 479−483. 562
494 transform IR spectroscopic study of dimethyl sulfoxide self-association (48) Wulf, A.; Ludwig, R. Structure and Dynamics of Water 563
495 in acetonitrile solutions. J. Chem. Soc., Faraday Trans. 1996, 92, Confined in Dimethyl Sulfoxide. ChemPhysChem. 2006, 7, 266−272. 564
496 3333−3337. (49) Ludwig, R.; Farrar, T. C. Temperature Dependence of the 565
497 (27) Fawcett, W. R.; Kloss, A. A. Solvent-Induced Frequency Shifts Deuteron and Oxygen Quadrupole Coupling Constants of Water in 566
498 in the Infrared Spectrum of Dimethyl Sulfoxide in Organic Solvents. J. the System Wated Dimethyl Sulfoxide. J. Phys. Chem. D 1994, 98, 567
499 Phys. Chem. C 1996, 100, 2019−2024. 6684−6687. 568
G https://doi.org/10.1021/acs.jpcb.1c10558
J. Phys. Chem. B XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry B pubs.acs.org/JPCB Article
H https://doi.org/10.1021/acs.jpcb.1c10558
J. Phys. Chem. B XXXX, XXX, XXX−XXX
nmt00 | ACSJCA | JCA11.2.5208/W Library-x64 | manuscript.3f (R5.1.i4:5009 | 2.1) 2021/10/27 08:51:00 | PROD-WS-121 | rq_2820701 | 2/08/2022 22:22:35 | 7 | JCA-DEFAULT
pubs.acs.org/JPCB Article
3 Published as part of The Journal of Physical Chemistry virtual special issue “Kankan Bhattacharyya Festschrift”.
4 Chayan K. De, Anup Ghosh, and Prasun K. Mandal*
Cite This: https://doi.org/10.1021/acs.jpcb.1c08598 Read Online
5 ABSTRACT: The hydrophobicity of room-temperature ionic liquids (RTILs) has been shown to have a very significant effect on
6 the optical and structural properties of and in RTILs. The average excited state lifetime of neat RTILs has been shown to be
7 increasing with increasing hydrophobicity of the RTILs. By employing pico-nanosecond-based fluorescence anisotropy decay, the
8 volume of the nanoaggregates in neat RTILs have been calculated. The volume of these nanoaggregates have been shown to be
9 decreasing with increase in hydrophobicity of the RTILs. Thus, hydrophobicity has been shown to have an important role, i.e.,
10 hydrophobicity can be used as a handle to tune the properties of RTILs as designer solvents. Moreover, the excited-state lifetime of
11 red-emitting fluorophores, i.e., whose fluorescence emission is not perturbed by the inherent emission of RTILs, has been shown to
12 increase with the increasing hydrophobicity of the RTILs. Highly hydrophobic RTILs have been shown to exhibit positive deviation
13 and highly hydrophilic RTIL has been shown to exhibit negative deviation from the linear correlation between average solvation time
14 (τs) versus viscosity/temperature (η/T).
15 Hydrophobicity is a unique property in the case of ionic and normal solvents.5−12 Varying the physical and chemical 35
16 polar room-temperature ionic liquids (RTILs). There have properties of RTILs have been obtained by the iterative 36
17 been reports about the synthesis and applications of hydro- combinations of various cations and anions. Various 37
18 phobic and polar ionic liquids,1−4 however, effect of hydro- experimental techniques like small-angle X-ray scattering, 38
19 phobicity toward inherent structural properties and fluores- neutron scattering, optical spectroscopy, as well as simulation 39
20 cence dynamics have not been investigated in detail. In this analyses have shown the existence of different types of 40
21 context, we have chosen a few well-known hydrophobic ionic mesoscopic local heterostructures in liquid state.12−24 Thus, 41
22 liquids containing (hexafluorophosphate) PF6 and FAP RTILs are known as ‘“nanostructured fluids”’.19 Although 42
23 (tris(pentafluoroethyl)trifluorophosphate) anions. Generally, visibly transparent, however, imidazolium-based RTILs possess 43
24 the hydrophobicity follows the following for the anions: BF4 significant absorption in the UV region, and the tail of 44
25 (tetrafluoroborate) < PF6 < FAP. We have investigated and absorption extends up to a part of the visible region.25,26 45
26 compared the optical and structural properties of the
27 hydrophobic RTILs with hydrophilic ionic liquids containing
28 BF4 anion keeping the same cation (either bmim (1-butyl-3- Received: September 30, 2021
29 methylimidazolium) or hmim (1-hexyl-3-methylimidazolium), Revised: January 30, 2022
c1 30 Chart 1). A few related physicochemical properties of these
31 RTILs have been summarized in Table S1.
32 The existence of a multitude of interactions, such as
33 Coulombic, dipolar, van der Waals, and H-bonding, among
Chart 1. Chemical Structures of RTILs and Fluorophores goodness of the fit. Viscosity of the RTILs at different 91
Explored in This Study temperatures were recorded on Brookfield DV3T rheometer. 92
2.3. Method. For fluorescence anisotropy decay, parallel 93
(∥) and perpendicular (⊥) polarized fluorescence decays were 94
measured until their peak difference reaches 5000 count. Then 95
the G factor was calculated. Then the time-resolved anisotropy 96
decay (r(t)) was calculated with the equation 97
IVV(t ) − GIVH(t )
r (t ) =
IVV(t ) + 2GIVH(t ) (1) 98
46 Imidazolium RTILs exhibit excitation wavelength-dependent keeping the excitation polarizer in horizontal position, 104
47 emission maximum and fluorescence decay.25−29 alternating the emission polarizer in vertical and horizontal 105
48 Hydrophobicity is an important physical parameter which positions, respectively. Suitable exponential decay functions 106
49 controls several physical processes such as solubility, were used to fit the fluorescence decay/anisotropy decay 107
50 association of suitable functional groups, interaction and curves. Fitting of these decay curves were done using iterative 108
51 binding kinetics of ligands to the protein, protein folding, reconvolution program (IBH). 109
B https://doi.org/10.1021/acs.jpcb.1c08598
J. Phys. Chem. B XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry B pubs.acs.org/JPCB Article
C https://doi.org/10.1021/acs.jpcb.1c08598
J. Phys. Chem. B XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry B pubs.acs.org/JPCB Article
199 time constants would be less than those of IRF, i.e., less than
200 100 ps. The rotational correlation times we obtained are in the
201 range of several nanoseconds, and these values are at least 10−
202 60 times higher than the RET time, if RET is present at all.
203 Thus, further analyses based on rotational correlation time are
204 free from any significant perturbation from RET. RTILs are
205 known as “designer solvents” as the property of the RTILs
206 could be tuned by changing the nature of cation and anion.43
207 Quite interesting are the observations that the (i) average
208 excited state lifetime increases and (ii) the size of the
209 nanoaggregates decreases with increasing hydrophobicity of
210 the RTILs. Thus, hydrophobicity can be considered as another
211 handle toward tuning properties of RTILs as designer solvents.
212 3.3. Hydrophobicity-Dependent Excited State-Life-
213 time of Red-Emitting Dyes. As hydrophobicity has been
214 shown to be a useful handle to control structural as well as
215 optical property of neat RTILs, it was necessary to probe the
216 optical behavior of red-emitting fluorophores in RTILs of
217 varying hydrophobicity. We have chosen red-emitting
218 fluorophores so that inherent fluorescence of RTILs do not
219 cause any perturbation in the investigation. We have explored
220 two red-emitting fluorophores, namely, LD700 and OX725 in
221 six different RTILs, viz., bmim[BF4], bmim[PF6], [bmim-
222 [FAP], hmim[BF4], hmim[PF6], and hmim[FAP] (Chart 1).
223 These dyes are positively charged hence varying the anion
224 (with the same cation) thus varying the hydrophobicity would
225 provide important insight about the nature of nanoscopic
226 location these dyes reside into as well as nature of fluorescence
227 decay pathway these dyes follow in these RTILs. Both the red
228 emitting fluorophores exhibit single exponential fluorescence
229 decay behavior in all six RTILs for all three different excitation
230 wavelengths (Figures S9 and S10).
231 Please recall the hydrophobicity increases in the following
232 order: BF4 < PF6 < FAP. Interestingly, as the hydrophobicity
233 increases from bmim[BF4] to bmim[PF6] to bmim[FAP] the
234 excited state lifetime of LD700 has been observed to increase
f3 235 from 2.7 to 2.9 to 4.6 ns (Figure 3a). Similarly as the
236 hydrophobicity increases from hmim[BF4] to hmim[PF6] to
237 hmim[FAP], the excited state lifetime of LD700 has been
238 observed to increase (Figure 3b). These observations have Figure 3. Variation of excited state lifetime of LD700 in (a) bmim-
239 been noted for all three excitation wavelengths, i.e., 375, 402, based RTILs (1) bmim[BF4], (2) bmim[PF6], and (3) bmim[FAP],
t2 240 and 635 nm (Figure 3 and Table 2). Such observations have and in (b) hmim-based RTILs, (4) hmim[BF4], (5) hmim[PF6], and
241 been noted for OX725 in six different RTILs and for all three (6) hmim[FAP]. λem = 680 nm. The error in the excited-state lifetime
242 excitation wavelengths as well. is less than ±5%.
243 However, as the alkyl chain length increases from bmim to
244 hmim keeping the anion same, the fluorescence lifetime of
245 LD700 or OX725 does not change at all (Table 2). Thus, we S12). The time constants of the solvation dynamics have been 262
246 can conclude that both LD700 and OX725 could sense the depicted in Table S5. As can be seen from this table, the 263
247 hydrophobicity of the anion. This means that being positively average solvation time has been found to be dependent on the 264
248 charged these dyes occupy a region much closer to the anion viscosity of the RTILs. 265
249 and quite far from the alkyl chain length of the cation. Such Maroncelli and co-workers have shown that the average 266
250 hitherto unreported consistent hydrophobicity-dependent solvation time (τs) depends on the size of the cation and 267
optical behavior of red-emitting fluorophores in RTILs is follows the eq 3.47
iηy
268
kT {
252
253 3.4. Hydrophobicity-Dependent Solvation in RTILs.
(3) 269
254 Solvation has been one of the most studied dynamical
255 phenomenon in RTILs.27,29,33−40,58−60 In this work, we have where τs is average solvation time, η is the viscosity of the 270
256 investigated solvation dynamics of C153 in two mostly RTIL at temperature T, R+ is the radius of the cation, p ∼ 1, 271
257 hydrophobic RTILs (bmim[FAP] and hmim[FAP]). Nature and q ∼ 4. To check whether hydrophobicity of the RTILs has 272
258 of the fluorescence decay of C153 in both bmim[FAP] and any role in solvation, we have plotted the average solvation 273
259 hmim[FAP] are dependent on monitoring wavelength (Figure time (τs) versus viscosity/temperature (η/T) for bmim cation 274
260 S11), and the time-resolved emission spectra showed time- based RTILs with varying anions (Figure 4). The solvation 275 f4
261 dependent shift of the fluorescence emission maximum (Figure time of RTILs other than FAP is taken from literature.47 It is 276
D https://doi.org/10.1021/acs.jpcb.1c08598
J. Phys. Chem. B XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry B pubs.acs.org/JPCB Article
■
277 reported that the hydrophobicity in RTILs follows the order
278 FAP− > Tf2N− > PF6−> BF4−> Cl−.61 ASSOCIATED CONTENT 314
*
sı Supporting Information 315
The Supporting Information is available free of charge at 316
https://pubs.acs.org/doi/10.1021/acs.jpcb.1c08598. 317
■ AUTHOR INFORMATION
Corresponding Author
320
321
Prasun K. Mandal − Department of Chemical Sciences and 322
Centre for Advanced Functional Materials, Indian Institute of 323
Science Education and Research (IISER) Kolkata, 324
Mohanpur, West Bengal 741246, India; orcid.org/0000- 325
0002-5543-5090; Email: prasunchem@iiserkol.ac.in 326
Authors 327
Chayan K. De − Department of Chemical Sciences, Indian 328
Institute of Science Education and Research (IISER) Kolkata, 329
Mohanpur, West Bengal 741246, India 330
Figure 4. Correlation between average solvation time τs, with η/T in Anup Ghosh − Department of Chemical Sciences, Indian 331
bmim-containing RTILs displaying varying hydrophobicities. Institute of Science Education and Research (IISER) Kolkata, 332
Mohanpur, West Bengal 741246, India; orcid.org/0000- 333
0002-1442-8740 334
340
286 hydrophilic anion [Cl]. Thus, hydrophobicity (or hydro- mental facilities. Support from the Fast-Track Project (SR/ 341
287 philicity) of RTIL has been shown to exhibit a significant role FT/CS-52/2011) of DST-India is gratefully acknowledged. 342
288 over the solvation phenomenon as well. Financial support from the SERB-DST India, Project No. 343
E https://doi.org/10.1021/acs.jpcb.1c08598
J. Phys. Chem. B XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry B pubs.acs.org/JPCB Article
344 CRG/2019/003605 is gratefully acknowledged. C.K.D. thanks (20) Urahata, S. M.; Ribeiro, M. C. C. Structure of ionic liquids of 1- 411
345 DST- INSPIRE, AG thanks UGC for Fellowship. alkyl-3-methylimidazolium cations: A systematic computer simulation 412
■
study. J. Chem. Phys. 2004, 120, 1855−1863. 413
(21) Wang, Y.; Voth, G. A. Unique Spatial Heterogeneity in Ionic 414
346 REFERENCES Liquids. J. Am. Chem. Soc. 2005, 127, 12192−12193. 415
347 (1) Fukaya, Y.; Ohno, H. Hydrophobic and polar ionic liquids. Phys. (22) Canongia Lopes, J. N.; Costa Gomes, M. F.; Pádua, A. A. H. 416
348 Chem. Chem. Phys. 2013, 15, 4066−4072. Nonpolar, Polar, and Associating Solutes in Ionic Liquids. J. Phys. 417
349 (2) Yao, W.; Wang, H.; Cui, G.; Li, Z.; Zhu, A.; Zhang, S.; Wang, J. Chem. B 2006, 110, 16816−16818. 418
350 Tuning the Hydrophilicity and Hydrophobicity of the Respective (23) Raabe, G.; Köhler, J. Thermodynamical and structural 419
351 Cation and Anion: Reversible Phase Transfer of Ionic Liquids. Angew. properties of imidazolium-based ionic liquids from molecular 420
352 Chem., Int. Ed. 2016, 55, 7934−7938. simulations. J. Chem. Phys. 2008, 128, 154509. 421
353 (3) Zhao, X.; Zhou, K.; Zhong, Y.; Liu, P.; Li, Z.; Pan, J.; Long, Y.; (24) de Andrade, J.; Böes, E. S.; Stassen, H. Liquid-Phase Structure 422
354 Huang, M.; Brakat, A.; Zhu, H. Hydrophobic ionic liquid-in-polymer of Dialkylimidazolium Ionic Liquids from Computer Simulations. J. 423
355 composites for ultrafast, linear response and highly sensitive humidity Phys. Chem. B 2008, 112, 8966−8974. 424
356 sensing. Nano Res. 2021, 14, 1202−1209. (25) Paul, A.; Mandal, P. K.; Samanta, A. How transparent are the 425
357 (4) Wang, Y.; Leng, W.; Gao, Y.; Guo, J. Thermo-Sensitive Polymer- imidazolium ionic liquids? A case study with 1-methyl-3-butylimida- 426
358 Grafted Carbon Nanotubes with Temperature-Controlled Phase zolium hexafluorophosphate, [bmim][PF6]. Chem. Phys. Lett. 2005, 427
359 Transfer Behavior between Water and a Hydrophobic Ionic Liquid. 402, 375−379. 428
360 ACS Appl. Mater. Interfaces. 2014, 6, 4143−4148. (26) Paul, A.; Mandal, P. K.; Samanta, A. On the Optical Properties 429
361 (5) Welton, T. Room-Temperature Ionic Liquids. Solvents for of the Imidazolium Ionic Liquids. J. Phys. Chem. B 2005, 109, 9148− 430
362 Synthesis and catalysis. Chem. Rev. 1999, 99, 2071−2083. 9153. 431
363 (6) Wasserscheid, P.; Keim, W. Ionic liquids - new solutions for (27) Hu, Z.; Margulis, C. J. Room-Temperature Ionic Liquids: Slow 432
364 transitions metal catalysts. Angew. Chem., Int. Ed. 2000, 39, 3772− Dynamics, Viscosity, and the Red Edge Effect. Acc. Chem. Res. 2007, 433
365 3789. 40, 1097−1105. 434
366 (7) Dupont, J.; de Souza, R. F.; Suarez, P. A. Z. Liquid (Molten Salt) (28) Ghosh, A.; Chatterjee, T.; Mandal, P. K. On the heterogeneity 435
367 Phase Organometallic Catalysis. Chem. Rev. 2002, 102, 3667−3692. of fluorescence lifetime of room temperature ionic liquids: onset of a 436
368 (8) Rogers, R. D.; Seddon, K. R. Ionic liquids - solvents of the future. journey for exploring red emitting dyes. Chem. Commun. 2012, 48, 437
369 Science 2003, 302, 792−793. 6250−6252. 438
370 (9) Seddon, K. R. Ionic liquids: A taste of the future. Nat. Mater. (29) Mandal, P. K.; Samanta, A. Fluorescence Studies in a 439
371 2003, 2, 363−365. Pyrrolidinium Ionic Liquid: Polarity of the Medium and Solvation 440
372 (10) WeinGaertner, H. Understanding ionic liquids at the molecular Dynamics. J. Phys. Chem. B 2005, 109, 15172−15177. 441
373 level: facts, problems and controversies. Angew. Chem., Int. Ed. 2008, (30) Lakowicz, J. R. Principles of Fluorescence Spectroscopy, 3rd ed.; 442
374 47, 654−670. Springer-Verlag: New York, 2006. 443
375 (11) Hallett, J. P.; Welton, T. Room-Temperature Ionic Liquids: (31) Reichardt, C.; Welton, T. Solvents and Solvent Effects in Organic 444
376 Solvents for Synthesis and catalysis. Chem. Rev. 2011, 111, 3508−
Chemistry, 4th ed.; Wiley-VCH Verlag GmbH & Co. KgaA: 445
377 3576.
Weinheim, 2010. 446
378 (12) Holbrey, J. D.; Reichert, W. M.; Rogers, R. D. Crystal structures
(32) Egorova, K. S.; Gordeev, E. G.; Ananikov, V. P. Biological 447
379 of imidazolium bis (trifluoromethanesulfonyl)imide ionic liquid’ salts:
Activity of Ionic Liquids and Their Application in Pharmaceutics and 448
380 the first organic salt with a cis-TFSI anion conformation. Dalton
Medicine. Chem. Rev. 2017, 117, 7132−7189. 449
381 Trans. 2004, 2267−2271.
(33) Karmakar, R.; Samanta, A. Solvation Dynamics of Coumarin- 450
382 (13) Chen, S.-H.; Yang, F.-R.; Wang, M.-T.; Wang, N.-n. Synthesis,
153 in a Room-Temperature Ionic Liquid. J. Phys. Chem. A 2002, 106, 451
383 characterization, and crystal structure of several novel acidic ionic
384 liquids based on the corresponding 1-alkylbenzimidazole with 4447−4452. 452
385 tetrafluoroboric acid. C. R. Chimie. 2010, 13, 1391−1396. (34) Karmakar, R.; Samanta, A. Steady-State and Time-Resolved 453
386 (14) Henderson, W. A.; Fylstra, P.; De Long, H. C.; Trulove, P. C.; Fluorescence Behavior of C153 and PRODAN in Room-Temperature 454
387 Parsons, S. Crystal structure of the ionic liquid EtNH3NO3− insights Ionic Liquid. J. Phys. Chem. A 2002, 106, 6670−6675. 455
388 into the thermal phase behavior of protic ionic liquids. Phys. Chem. (35) Mandal, P. K.; Sarkar, M.; Samanta, A. Excitation wavelength- 456
389 Chem. Phys. 2012, 14, 16041−16046. dependent fluorescence behavior of the room temperature ionic 457
390 (15) Triolo, A.; Russina, O.; Bleif, H.-J.; Di Cola, E. Nanoscale liquids and dissolved dipolar solutes. J. Phys. Chem. A 2004, 108, 458
391 Segregation in Room Temperature Ionic Liquids. J. Phys. Chem. B 9048−9053. 459
392 2007, 111, 4641−4644. (36) Mandal, P. K.; Paul, A.; Samanta, A. Excitation-Wavelength- 460
393 (16) Hardacre, C.; Holbrey, J. D.; Mullan, C. L.; Nieuwenhuyzen, Dependent Fluorescence Behavior of Some Dipolar Molecules in 461
394 M.; Youngs, T. G. A.; Bowron, D. T. Liquid Structure of the Ionic Room-Temperature Ionic Liquids. J. Photochem. Photobiol. A: Chem. 462
395 Liquid, 1-Methyl-4-cyanopyridinium Bis{(trifluoromethyl)sulfonyl}- 2006, 182, 113−120. 463
396 imide Determined from Neutron Scattering and Molecular Dynamics (37) Samanta, A. Dynamic Stokes Shift and Excitation Wavelength 464
397 Simulations. J. Phys. Chem. B 2008, 112, 8049−8056. Dependent Fluorescence of Dipolar Molecules in Room Temperature 465
398 (17) Hardacre, C.; Holbrey, J. D.; Mullan, C. L.; Youngs, T. G. A.; Ionic Liquids. J. Phys. Chem. B 2006, 110, 13704−13716. 466
399 Bowron, D. T. Small-angle neutron scattering from 1-alkyl-3- (38) Mandal, P. K.; Saha, S.; Karmakar, R.; Samanta, A. Solvation 467
400 methylimidazolium hexafluorophosphate ionic liquids ([Cnmim]- dynamics in room temperature ionic liquids Dynamic stokes shift 468
401 [PF6], n = 4, 6, and 8). J. Chem. Phys. 2010, 133, 074510. studies of fluorescence of dipolar molecules. Curr. Sci. 2006, 90, 301− 469
402 (18) Mele, A.; Tran, C. D.; De Paoli Lacerda, S. H. D. P. The 310. 470
403 structure of a room-temperature ionic liquid with and without trace (39) Samanta, A. Solvation Dynamics in Ionic Liquids: What We 471
404 amounts of water: The role of C-H···O and C-H···F interactions in 1- Have Learned from the Dynamic Fluorescence Stokes Shift Studies. J. 472
405 n-butyl-3-methylimidazolium tetrafluoroborate. Angew. Chem., Int. Ed. Phys. Chem. Lett. 2010, 1, 1557−1562. 473
406 2003, 42, 4364−4366. (40) Zhang, X.-X.; Liang, M.; Ernsting, N. P.; Maroncelli, M. 474
407 (19) Mele, A.; Romano, G.; Giannone, M.; Ragg, E.; Fronza, G.; Complete Solvation Response of Coumarin 153 in Ionic Liquids. J. 475
408 Raos, G.; Marcon, V. The local structure of ionic liquids: cation-cation Phys. Chem. B 2013, 117, 4291−4304. 476
409 NOE interactions and internuclear distances in neat [BMIM][BF4] (41) Ghosh, A.; Chatterjee, T.; Roy, D.; Das, A.; Mandal, P. K. On 477
410 and [BDMIM][BF4]. Angew. Chem., Int. Ed. 2006, 45, 1123−1126. the nanoscopic environment a neutral fluorophore experiences in 478
F https://doi.org/10.1021/acs.jpcb.1c08598
J. Phys. Chem. B XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry B pubs.acs.org/JPCB Article
479 room temperature ionic liquids. J. Phys. Chem. C 2014, 118, 5051− (59) Roy, D.; Maroncelli, M. Simulations of Solvation and Solvation 547
480 5057. Dynamics in an Idealized Ionic Liquid Model. J. Phys. Chem. B 2012, 548
481 (42) Ghosh, A.; De, C. K.; Chatterjee, T.; Mandal, P. K. What type 116, 5951−5970. 549
482 of nanoscopic environment does a cationic fluorophore experience in (60) Maroncelli, M.; Zhang, X.; Liang, M.; Roy, D.; Ernsting, N. P. 550
483 room temperature ionic liquids? Phys. Chem. Chem. Phys. 2015, 17, Measurements of the complete solvation response of coumarin 153 in 551
484 16587−16593. ionic liquids and the accuracy of simple dielectric continuum 552
485 (43) Majhi, D.; Sarkar, M. Probing the microscopic structural predictions. Faraday Discuss. 2012, 154, 409−424. 553
486 organization of neat ionic liquids (ILs) and ionic liquid-based gels (61) Huddleston, J. G.; Visser, A. E.; Reichert, W. M.; Willauer, H. 554
487 through resonance energy transfer (RET) studies. Phys. Chem. Chem. D.; Broker, G. A.; Rogers, R. D. Characterization and Comparison of 555
488 Phys. 2017, 19, 23194−23203. Hydrophilic and Hydrophobic Room Temperature Ionic Liquids 556
489 (44) Arzhantsev, S.; Ito, N.; Heitz, M.; Maroncelli, M. Solvation Incorporating the Imidazolium Cation. Green Chem. 2001, 3, 156− 557
490 Dynamics of Coumarin 153 in Several Classes of Ionic Liquids: 164. 558
491 Cation Dependence of the Ultrafast Component. Chem. Phys. Lett.
492 2003, 381, 278−286.
493 (45) Ito, N.; Arzhantsev, S.; Maroncelli, M. The probe dependence
494 of solvation dynamics and rotation in the ionic liquid 1-butyl-3-
495 methyl-imidazolium hexafluorophosphate. Chem. Phys. Lett. 2004,
496 396, 83−91.
497 (46) Yan, T.; Wang, Y.; Knox, C. Solvation and Rotational Dynamics
498 of Coumarin 153 in Ionic Liquids: Comparisons to Conventional
499 Solvents. J. Phys. Chem. B 2010, 114, 6905−6921.
500 (47) Jin, H.; Baker, G. A.; Arzhantsev, S.; Dong, J.; Maroncelli, M.
501 Solvation and Rotational Dynamics of Coumarin 153 in Ionic
502 Liquids: Comparisons to Conventional Solvents. J. Phys. Chem. B
503 2007, 111, 7291−7302.
504 (48) Wang, Y.; Voth, G. A. Tail Aggregation and Domain Diffusion
505 in Ionic Liquids. J. Phys. Chem. B 2006, 110, 18601−18608.
506 (49) Das, A.; Roy, D.; De, C. K.; Mandal, P. K. Where does the
507 fluorescing moiety reside in a carbon dot?” - Investigations based on
508 fluorescence anisotropy decay and resonance energy transfer
509 dynamics. Phys. Chem. Chem. Phys. 2018, 20, 2251−2259.
510 (50) Das, S. K.; Sarkar, M. Rotational dynamics of coumarin-153
511 and 4-aminophthalimide in 1-ethyl-3-methylimidazolium alkylsulfate
512 ionic liquids: effect of alkyl chain length on the rotational dynamics. J.
513 Phys. Chem. B 2012, 116, 194−202.
514 (51) Das, S. K.; Majhi, D.; Sahu, P. K.; Sarkar, M. Investigation of
515 the influence of alkyl side chain length on the fluorescence response of
516 C153 in a series of room temperature ionic liquids. RSC Adv. 2015, 5,
517 41585−41594.
518 (52) Prabhu, S. R.; Dutt, G. B. Effect of Low Viscous Nondipolar
519 Solvent on the Rotational Diffusion of Structurally Similar Nondipolar
520 Solutes in an Ionic Liquid. J. Phys. Chem. B 2015, 119 (5), 2019−
521 2025.
522 (53) Ghosh, A.; De, C. K.; Chatterjee, T.; Das, A.; Roy, D.; Routh,
523 T.; Mandal, P. K. Correlation between size of nano-aggregates and
524 excitation wavelength dependent fluorescence emission in room
525 temperature ionic liquids: A case study with emim[FAP]. Chem. Phys.
526 Impact 2021, 3, 100054.
527 (54) Ghiggino, K. P.; Reek, J. N. H.; Crossley, M. J.; Bosman, A. W.;
528 Schenning, A. P. H. J.; Meijer, E. W. The Dynamics of Electronic
529 Energy Transfer in Novel Multiporphyrin Functionalized Dendrimers:
530 A Time-Resolved Fluorescence Anisotropy Study. J. Phys. Chem. B
531 2000, 104, 2596−2606.
532 (55) Silori, Y.; De, A. K. Tuning effect of local environment to
533 control mechanism of fluorescence depolarization: Rotational
534 diffusion and resonance energy transfer within homo-aggregates of
535 xanthenes. J. Photochem. Photobiol. A: Chem. 2019, 377, 198−206.
536 (56) Xu, Q.-H.; Wang, S.; Korystov, D.; Mikhailovsky, A.; Bazan, G.
537 C.; Moses, D.; Heeger, A. J. The fluorescence resonance energy
538 transfer (FRET) gate: A time-resolved study. Proc. Nat. Acad. Sci. USA
539 2005, 102, 530.
540 (57) Warren, S. C.; Margineanu, A.; Katan, M.; Dunsby, C.; French,
541 P. M. W. Homo-FRET Based Biosensors and Their Application to
542 Multiplexed Imaging of Signalling Events in Live Cells. Int. J. Mol. Sci.
543 2015, 16, 14695−14716.
544 (58) Roy, D.; Patel, N.; Conte, S.; Maroncelli, M. Dynamics in an
545 Idealized Ionic Liquid Model. J. Phys. Chem. B 2010, 114, 8410−
546 8424.
G https://doi.org/10.1021/acs.jpcb.1c08598
J. Phys. Chem. B XXXX, XXX, XXX−XXX
Chemical Physics Impact 3 (2021) 100054
A R T I C L E I N F O A B S T R A C T
Keywords: Excitation wavelength dependent emission maximum (400 - 560 nm) of a room temperature ionic liquid has been
Ionic liquid investigated employing time-resolved fluorescence anisotropy based rotational dynamical analyses. Employing
Nano-aggregate Stokes-Einstein-Debye relationship we could show that nano-aggregate having aggregation volume of ~250 Å3,
Rotational dynamics
~350 Å3, ~400 Å3 and ~500 Å3 emit at ~400 nm, ~450 nm, ~500 nm, and ~560 nm respectively.
SED equation
Size dependent emission
Research on Room Temperature Ionic Liquids (RTILs) has been Neat RTILs exhibit excitation wavelength dependent emission
increasing rapidly because of their interesting properties like low behavior which is known as red edge effect [33,34]. It has been invoked
melting point, high thermal stability, negligible vapor pressure, recy that different nano-aggregates behave as different species and their
clability, high ionic conductivity etc. [1–7]. Thus RTILs have been ground state and excited state (electronic) energy levels are different
employed in separation, catalysis and energy related applications like [33–34] Lack of energy transfer between these nanostructures causes
batteries, fuel cell photovoltaics, super-capacitors etc. [1–7]. Single excitation wavelength dependent emission maximum in RTILs. It has
crystal XRD, neutron scattering, Raman spectroscopic study, small-wide been shown that fluorescence decay behavior of neat RTILs are multi
angle X-ray scattering, simulation studies etc. reveal microscopic het exponential in nature and dependent on both excitation and emission
erogeneity in the structure of RTIL [8–28]. Structural heterogeneities wavelength [56.57] Average excited state lifetimes have been shown to
have been shown to be ranging from Angstrom (Å) to a few nanometre vary by a factor of three or more [57]. This observed behavior has been
(nm) in the spatial scale as a consequence of the segregation of the alkyl ascribed to the absence of energy transfer among the different
tails into mesoscopic domains [29–31]. Such an organization might be nano-aggregating emitting unit inherently present in RTILs [51]. It has
responsible for observed unique properties of RTILs like their ability to been speculated that perhaps different (sized) nano-aggregates are
dissolve both polar and not so polar compounds, anomalous trend for responsible for the said behavior. However, detailed effort in order to
the physical properties like viscosity upon increasing the alkyl chain quantify the size of the nano-aggregated species present in the liquid
length of the imidazolium cation, [1–7] excitation wavelength depen state and it’s correlation with excitation wavelength dependent fluo
dent emission behavior of neat RTILs [32–34] and complex solvation rescence emission maximum is lacking. Although several rotational
dynamics [35–42] of dissolved dipolar solutes. anisotropic studies of the dissolved fluorophores have been carried out
Out of all these properties, excitation wavelength dependent fluo in RTILs, however, the same measurements in neat RTILs are very few
rescence emission is still very intriguing and bear interesting analogy [58–69].
with size dependent emission maximum of semiconductor quantum dots Using fast rotational dynamics we have probed the existence of
(QDs) (say CdSe, InP etc.) [43–56]. Thus, using same chemical compo different nano-aggregates in the liquid structure of RTILs and measured
sition but by varying the size of the QDs it is possible to cover the entire the size of the nano-aggregating units. Moreover, whether there is any
range of the UV-–Vis-NIR region of the electromagnetic spectrum. correlation between the size of the nano-aggregating species with the
* Corresponding author.
E-mail address: prasunchem@iiserkol.ac.in (P.K. Mandal).
https://doi.org/10.1016/j.chphi.2021.100054
Received 15 September 2021; Received in revised form 31 October 2021; Accepted 9 November 2021
Available online 22 November 2021
2667-0224/© 2021 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
A. Ghosh et al. Chemical Physics Impact 3 (2021) 100054
emission maximum was also probed into. In order to shed light in that
direction we have performed a case study using 1-Ethyl-3-methylimida
zolium tris(pentafluoroethyl)trifluorophosphate (emim[FAP]) (see
Chart 1 and Table S0 for physico-chemical properties) as a model RTIL
due to its interesting photo-physical and physiochemical property [69,
70]. The choice of using this particular RTIL stems from the fact that this
RTIL is highly hydrophobic and less viscous in nature [28,56,57]. emim
[FAP] showed a wider electrochemical window, despite lower viscosity,
larger self-diffusion coefficients and higher ionic conductivity. emim
[FAP] has weaker inter-ionic interactions compared to the emim+ with
smaller anion like [PF6], [BF4], halides [70]. We have investigated
rotational dynamics of the neat RTIL, emim[FAP] as a function of tem
perature. We have explored how the hydrodynamic volume of the
different nano-aggregating emitting unit in neat RTIL changes with
excitation and monitoring wavelengths.
As we have planned to perform temperature dependent fluorescence
dynamical experiments so we had to first do a steady state control
measurement in order to check that the RTIL structure remains intact in Fig. 1. Temperature independence of variation of emission maximum against
the entire temperature range (298 K to 338 K) of the experiment. As a excitation wavelength.
first step we investigated whether the absorption spectrum gets modified
with change in temperature. As can be seen (see Fig. S1) the absorption correlation time, and viscosity of the medium respectively.
spectrum remains same in the entire temperature range. A more sensi
tive experiment is to check whether the excitation wavelength depen ηVh
τr = (1)
dent emission maximum of RTIL change with change in temperature. As kB T
can be seen from Fig. 1, the excitation wavelength dependent emission With increasing temperature rotational correlation time has been
maximum does not change as we vary the temperature from 298 K to observed to decrease gradually. These are not due to the breakdown of
338 K. We would like to recall a report based on MD computer simu the nano-aggregation because emission maximum does not change with
lation of temperature dependent behavior of RTIL [23]. It has been change temperature (Fig. 1& Supp Info Figure S1). Rather this obser
shown that static dielectric constant of bmim[PF6] does not change vation is due to the decreasing the viscosity of the medium.
significantly with change in temperature [17]. The ultrafast inertial According to Eq. (1), rotational diffusion of a solute not only depends
component of solvation also remains unchanged. The cation-anion on the viscosity but also on the size of the solute. In order to investigate
structure and the potential of mean force relevant to the relative mo whether rotational anisotropy of neat emim[FAP] depends on moni
tion of cations and anions in the radial direction, i.e. vibration as an ion toring wavelength or not we have probed different monitoring wave
pair do not get altered substantially with change in temperature [23] lengths. Correlation between τr and η/T has been depicted in Fig. 2 and
Thus, our experimental result as well as reported MD simulation result the corresponding rotational correlation time constants have been
show that the nature of the nano-aggregates does not get altered in the depicted in Table 1. As can be seen from Fig. 2 as well as from Table 1,
temperature range of 298 K to 338 K. rotational anisotropy decay becomes more slower with increasing the
We have measured rotational anisotropy of neat emim[FAP] varying monitoring wavelength from 400 nm to 560 nm for a fixed excitation
both excitation and emission wavelengths over the temperature range wavelength (say 377 nm). This means that rotational correlation time
298 K to 338 K (see S2 to S8). Experimentally measured reorientation increases with increasing the monitoring wavelength. As can be seen
times of neat emim[FAP] have been analyzed using Stokes-Einstein- from linear correlation plot of τr and η/T depicted in Fig. 2, SED equa
Debye (SED) hydrodynamic theory. According to the SED theory, the tion can depict the situation quite well for emim[FAP]. From Table 1 we
solute reorientation time is related to the bulk viscosity of the solvent could notice that at lower temperature (298 K) rotational correlation
according to eqn. (1) (shown below), where kB, T, Vh, τr, η are the time (θ) increases by a factor of ~1.8 (with change of monitoring
Boltzmann constant, absolute temperature, the hydrodynamic volume of wavelength from 400 nm to 560 nm), whereas, at higher temperature
the fluorescing unit, (nano-aggregates in the present case), rotational
2
A. Ghosh et al. Chemical Physics Impact 3 (2021) 100054
Table 1
Rotational correlational time (τr) at different temperatures, volume or radius of nano-aggregates.
λex (nm) λem (nm) τr 298K (ns) τr 308K (ns) τr 318K (ns) τr 328K (ns) τr 338K (ns) Volume of nano-aggregate (Å3) Radius of the nano-aggregate (Å)
3
A. Ghosh et al. Chemical Physics Impact 3 (2021) 100054
impurity rather the excitation wavelength dependent emission is help. Support from the Fast-Track Project (SR/FT/CS-52/2011) of DST-
because of inherently present nano-aggregates. The fact that different India is gratefully acknowledged. AG thanks UGC, CKD and TR thank
reorientation times are because of different sized nano-aggregates pre IISER Kolkata, and AD, TC, DR thank CSIR for their respective
sent in neat RTIL can be supported by another control experiment. If a Fellowship.
polar solvent such as acetonitrile is added to the neat RTIL then it is
expected that different nanoaggregates will break and form an average Supplementary materials
sized moiety. It can be seen (see S9 and S10) that fluorescence anisot
ropy decays of emim[FAP] in acetonitrile are much faster in comparison Supplementary material associated with this article can be found, in
to neat RTIL. These decay profiles are identical to each other i.e. inde the online version, at doi:10.1016/j.chphi.2021.100054.
pendent of monitoring wavelengths (for a fixed λex=377 nm). Moreover,
rotational correlation time is same for all monitoring wavelengths and References
the magnitude (200 ps) is much less than what has been observed for
neat RTIL (~1 to ~7 ns). Thus, we can conclude that in acetonitrile [1] T. Welton, Room-temperature ionic liquids. Solvents for synthesis and catalysis,
Chem. Rev. 99 (1999) 2071–2083.
different nano-aggregating units have been broken down. This control [2] P. Wasserscheid, W. Keim, Ionic liquids: new solutions for transitions metal
experiment further confirms that the emission from RTIL is not from any catalysts, Angew. Chem. Int. Ed. 39 (2000) 3772–3789.
impurity rather from the nano-aggregating units. Thus, using fast rota [3] J. Dupont, R.F.D. Souza, P.A.Z. Suarez, Liquid (molten salt) phase organometallic
catalysis, Chem. Rev. 102 (2002) 3667–3692.
tion dynamical measurements we were able to measure the size of [4] R.D. Rogers, K.R. Seddon, Ionic liquids: solvents of the future, Science 302 (2003)
different nano-aggregates inherently present in RTIL and correlate the 792–793.
emission maximum with the size of the nano-aggregates. With increase [5] K.R. Seddon, Ionic liquids: a taste of the future, Nature 2 (2003) 363–365.
[6] H. WeinGaertner, Understanding ionic liquids at the molecular level: facts,
in size of the nano-aggregates the emission maximum has been shown to problems and controversies, Angew. Chem., Int. Ed. 47 (2008) 655–670.
be increased. In this regard RTIL behave in a similar way to that of the [7] J.P. Hallet, T. Welton, Room-temperature ionic liquids: solvents for synthesis and
QDs. Moreover, in case of QDs photolumiscent quantum yield depends catalysis, Chem. Rev. 111 (2011) 3508–3576.
[8] A. Mele, C.D. Tran, S.H. De Paoli Lacerda, The structure of a room-temperature
on excitation wavelength. [48,49,53] However, in order to understand
ionic liquid with and without trace amounts of water: the role of C− H•••O and
the underlying phenomenon in a much detailed way, theoret C− H•••F interactions in 1-n-butyl- 3-methylimidazolium tetrafluoroborate,
ical/computational analyses need to be performed, and these in Angew. Chem., Int. Ed. 42 (2003) 4364–4366.
vestigations need to be repeated for many such RTILs. These [9] S.M. Urahata, M.C.C. Ribeiro, Structure of ionic liquids of 1-alkyl-3-methylimida
zolium cations: a systematic computer simulation study, J. Chem. Phys. 120 (2004)
experiments are currently underway. 1855–1863.
In conclusion, we could show that emission maxima of RTILs do not [10] J. Dupont, Liquid (molten salt) phase organometallic catalysis, J. Braz. Chem. Soc.
change with change in temperature (298 K to 338 K). This means that 15 (2004) 341–350.
[11] M.G.D. Pópolo, G.A. Voth, On the structure and dynamics of ionic liquids, J. Phys.
the nature of nanoaggregates do not change significantly in the said Chem. B 108 (2004) 1744–1752.
temperature range. For a particular temperature rotational correlation [12] Y. Wang, G.A. Voth, Unique spatial heterogeneity in ionic liquids, J. Am. Chem.
time has been observed to increase with increase in monitoring wave Soc. 127 (2005) 12192–12193.
[13] A. Mele, G. Romano, M. Giannone, E. Ragg, G. Fronza, G. Raos, V. Marcon, The
length. For a particular monitoring wavelength the rotational correla local structure of ionic liquids: cation-cation NOE interactions and internuclear
tion time has been observed to decrease by a factor of 3 to 4 with distances in neat [BMIM][BF4] and [BDMIM][BF4], Angew. Chem. Int. Ed. 45
increase in temperature from 298 K to 338 K. Stokes-Einstein-Debye (2006) 1123–1126.
[14] A. Triolo, O. Russina, H.-.J. Bleif, E.J. Di Cola, Segregation in room temperature
model has been shown to be depict nano-spatial-information for emim ionic liquids, Phys. Chem. B 111 (2007) 4641–4644.
[FAP] quite well and thus size of the nano-aggregates could be calcu [15] W. Jiang, Y. Wang, G.A. Voth, Molecular dynamics simulation of nanostructural
lated to vary from ~~500 Å3. Finally we could correlate the emission organization in ionic liquid/water, J. Phys. Chem. B 111 (2007) 4812–4818.
[16] H. Weingaertner, Understanding ionic liquids at the molecular level: facts,
maximum with the size of the nano-aggregates inherently present in an
problems and controversies, Angew. Chem. Int. Ed. 47 (2008) 654–670.
RTIL namely emim[FAP]. We could show that nano-aggregate having [17] E.W. Castner, C.J. Margulis, M. Maroncelli, J.F. Wishart, Ionic liquids: strctures and
aggregation volume of ~250 Å3, ~350 Å3, ~400 Å3 and ~500 Å3 emit at photochemical reactions, Annu. Rev. Phys. Chem. 62 (2011) 85–105.
~400 nm, ~450 nm, ~500 nm, and ~560 nm respectively. [18] O. Russina, A. Triolo, L. Gontrani, R. Caminiti, Mesoscopic structural
heterogeneities in room-temperature ionic liquids, J. Phys. Chem. Lett. 3 (2012)
27–33.
Author contributions [19] M. Macchiagodena, F. Ramondo, A. Triolo, L. Gontrani, R. Caminiti, Liquid
structure of 1-ethyl-3-methylimidazolium alkyl sulfates by X-ray scattering and
molecular dynamics, J. Phys. Chem. B 116 (2012) 13448–13458.
PKM envisaged the project. TC, AD, DR and TR purified the RTIL, [20] A. Triolo, O. Russina, R. Caminiti, H. Shirota, H.Y. Lee, C.S. Santos, N.S. Murthy, E.
performed temperature dependent steady state optical spectroscopic W. Castner Jr, Comparing intermediate range order for alkyl- vs.ether-substituted
measurements, and measured the viscosity of the RTIL at different cations in ionic liquids, Chem. Commun. 48 (2012) 4959–4961.
[21] H.K. Kashyap, J.J. Hettige, H.V.R. Annapureddy, C.J. Margulis, SAXS anti-peaks
temperatures. AG and CKD performed rotational anisotropy based time reveal the length-scales of dual positive–negative and polar–apolar ordering in
resolved rotational dynamical measurements and analyzed the data. AG room-temperature ionic liquids, Chem. Commun 48 (2012) 5103–5105.
and CKD made the figures and the tables. AG, CKD and PKM wrote the [22] S. Li, G. Feng, J.L. Banuelos, G. Rother, P.F. Fulvio, S. Dai, P.T. Cummings,
Distinctive nanoscale organization of dicationic versus monocationic ionic liquids,
manuscript. J. Phys. Chem. C 117 (2013) 18251–18257.
[23] Y. Shim, H.J. Kim, Dielectric relaxation and solvation dynamics in a room-
Research data temperature ionic liquid: temperature dependence, J. Phys. Chem. B 117 (2013)
11743–11752.
[24] T. Ishida, H. Shirota, Dicationic versus monocationic ionic liquids: distinctive ionic
Research Data can be provided if necessary. dynamics and dynamical heterogeneity, J. Phys. Chem. B 117 (2013) 1136–1150.
[25] H.K. Kashyap, C.S. Santos, R.P. Daly, J.J. Hettige, N.J. Murthy, H. Shirota, Jr.E.
W. Castner, C.J. Margulis, How does the ionic liquid organizational landscape
Declaration of Competing Interest
change when nonpolar cationic alkyl groups are replaced by polar isoelectronic
diethers? J. Phys. Chem. B 117 (2013) 1130–1135.
The authors declare that they have no known competing financial [26] J.J. Hettige, H.K. Kashyap, H.V.R. Annapureddy, C.J. Margulis, Anions, the
interests or personal relationships that could have appeared to influence reporters of structure in ionic liquids, J. Phys. Chem. Lett. 4 (2013) 105–110.
[27] K. Wei, L. Deng, Y. Wang, Z.-.C. Ou-Yang, G. Wang, Effect of side-chain length on
the work reported in this paper. structural and dynamic properties of ionic liquids with hydroxyl cationic tails,
J. Phys. Chem. B 118 (2014) 3642–3649.
Acknowledgement [28] A. Ghosh, T. Chatterjee, D. Roy, A. Das, P.K. Mandal, On the nanoscopic
environment a neutral fluorophore experiences in room temperature ionic liquids,
J. Phys. Chem. C. 118 (2014) 5051–5057.
PKM thanks IISER Kolkata for instrumental facilities and financial
4
A. Ghosh et al. Chemical Physics Impact 3 (2021) 100054
[29] Y. Wang, G.A. Voth, Tail aggregation and domain diffusion in ionic liquids, J. Phys. dependent photoluminescence quantum yield, photoluminescence decay
Chem. B 110 (2006) 18601–18608. dynamics, and single particle blinking dynamic, J. Phys. Chem. C 122 (2018)
[30] R. Hayes, G.G. Warr, Rob atkin, structure and nanostructure in ionic liquids, Chem. 964–973.
Rev. 115 (13) (2015) 6357–6426. [51] C.K. De, D. Roy, S. Mandal, P.K. Mandal, Suppressed blinking under normal air
[31] W. Yong-Lei, L. Bin, S. Sten, F. Mocci, Z.-.Y. Lu, J. Yuan, A. Laaksonen, M.D. Fayer, atmosphere in toxic-metal-free, small sized, InP-based core/alloy-shell/shell
Microstructural and dynamical heterogeneities in ionic liquids, Chem. Rev. 120 quantum dots, J. Phys. Chem. Lett. 10 (2019) 4330–4338.
(13) (2020) 5798–5877. [52] D. Roy, S. Mandal, C.K. De, K. Kumar, P.K. Mandal, Nearly suppressed
[32] P.K. Mandal, M. Sarkar, A. Samanta, Excitation wavelength dependent photoluminescence blinking of small-sized, blue–green–orange–red emitting single
fluorescence behavior of the room temperature ionic liquids and dissolved dipolar CdSe-based core/gradient alloy shell/shell quantum dots: correlation between
solutes, J. Phys. Chem. A 108 (2004) 9048–9053. truncation time and photoluminescence quantum yield, Phys. Chem. Chem. Phys.
[33] A. Paul, P.K. Mandal, A. Samanta, How transparent are the imidazolium ionic 20 (2018) 10332–10344.
liquids? A case study with 1-methyl-3- butylimidazolium hexafluorophosphate, [53] S. Mandal, S. Mukherjee, C.K. De, D. Roy, S. Ghosh, P.K. Mandal, Extent of
[bmim][PF6], Chem. Phys. Lett. 402 (2005) 375–379. shallow/deep trap states beyond the conduction band minimum in defect-tolerant
[34] A. Paul, P.K. Mandal, A. Samanta, On the optical properties of the imidazolium CsPbBr3 perovskite quantum dot: control over the degree of charge carrier
ionic liquids, J. Phys. Chem. B 109 (2005) 9148–9153. recombination, J. Phys. Chem. Lett. 11 (2020) 1702–1707.
[35] A. Samanta, Dynamic stokes shift and excitation wavelength dependent [54] S. Mandal, S. Ghosh, S. Mukherjee, C.K. De, D. Roy, T. Samanta, P.K. Mandal,
fluorescence of dipolar molecules in room temperature ionic liquids, J. Phys. Unravelling halide-dependent charge carrier dynamics in CsPb(Br/Cl)3 perovskite
Chem. B 110 (2006) 13704–13716. nanocrystals, Nanoscale 13 (2021) 3654–3661.
[36] A. Samanta, Solvation dynamics in ionic liquids: what we have learned from the [55] S. Ghosh, S. Mandal, S. Mukherjee, C.K. De, T. Samanta, M. Mandal, D. Roy, P.
dynamic fluorescence stokes shift studies, J. Phys. Chem. Lett. 1 (2010) K. Mandal, Near-unity photoluminescence quantum yield and highly suppressed
1557–1562. blinking in a toxic-metal-free quantum dot, J. Phys. Chem. Lett. 12 (2020)
[37] S. Daschakraborty, R. Biswas, Composition dependent stokes shift dynamics in 1426–1431.
binary mixtures of 1-butyl-3-methylimidazolium tetrafluoroborate with water and [56] A. Ghosh, C.K. De, T. Chatterjee, P.K. Mandal, What type of nanoscopic
acetonitrile: quantitative comparison between theory and complete measurements, environment does a cationic fluorophore experience in room temperature ionic
J. Phys. Chem. B 118 (2014) 1327–1339. liquids? Phys. Chem. chem. phys. 17 (2015) 16587–16593.
[38] H.K. Kashyap, R. Biswas, Dipolar solvation dynamics in room temperature ionic [57] A. Ghosh, T. Chatterjee, P.K. Mandal, On the heterogeneity of fluorescence lifetime
liquids: an effective medium calculation using dielectric relaxation data, J. Phys. of room temperature ionic liquids: onset of a journey for exploring red emitting
Chem. B 112 (2008) 12431–12438. dyes, Chem. Commun. 4 (2012) 6250–6252.
[39] H.K. Kashyap, R. Biswas, Solvation dynamics of dipolar probes in dipolar room [58] V. Gangamallaiah, G.B. Dutt, Fluorescence anisotropy of a nonpolar solute in 1-
temperature ionic liquids: separation of ion− dipole and dipole− dipole interaction alkyl-3-methylimidazolium-based ionic liquids: does the organized structure of the
contributions, J. Phys. Chem. B 114 (2010) 254–268. ionic liquid influence solute rotation? J. Phys. Chem. B 117 (2013) 5050–5057.
[40] H.K. Kashyap, R. Biswas, Stokes shift dynamics in ionic liquids: temperature [59] S.R. Prabhu, G.B. Dutt, Rotational diffusion of nondipolar and charged solutes in
dependence, J. Phys. Chem. B 114 (2010) 16811–16823. alkyl-substituted imidazolium triflimides: effect of C2 methylation on solute
[41] S. Daschakraborty, T. Pal, R. Biswas, Stokes shift dynamics of ionic liquids: solute rotation, J. Phys. Chem. B 118 (2014) 9420–9426.
probe dependence, and effects of self-motion, dielectric relaxation window, and [60] D.C. Khara, J.P. Kumar, N. Mondal, A. Samanta, Effect of the alkyl chain length on
collective intermolecular solvent modes, J. Chem. Phys. 139 (2013) the rotational dynamics of nonpolar and dipolar solutes in a series of N-alkyl-N-
164503–164515. methylmorpholinium ionic liquids, J. Phys. Chem. B 117 (2013) 5156–5164.
[42] S. Daschakraborty, R. Biswas, Ultrafast solvation response in room temperature [61] S.K. Das, M. Sarkar, Rotational dynamics of coumarin-153 and 4-aminophthali
ionic liquids: possible origin and importance of the collective and the nearest mide in 1-ethyl-3-methylimidazolium alkylsulfate ionic liquids: effect of alkyl
neighbour solvent modes, J. Chem. Phys. 137 (2012) 114501–114513. chain length on the rotational dynamics, J. Phys. Chem. B 116 (2012) 194–202.
[43] C.B. Murray, D.J. Noms, M.G. Bawendi, Synthesis and characterization of nearly [62] S.K. Das, P.K. Sahu, M. Sarkar, Diffusion–viscosity decoupling in solute rotation
monodisperse CdE (E = sulfur, selenium, tellurium) semiconductor and solvent relaxation of coumarin153 in ionic liquids containing
nanocrystallites, J. Am. Chem. Soc. 115 (1993) 8706–8715. fluoroalkylphosphate (FAP) anion: a thermophysical and photophysical study,
[44] Y. Masumoto, K. Sonobe, Size-dependent energy levels of CdTe quantum dots, J. Phys. Chem. B 117 (2013) 636–647.
Phys. Rev. B 56 (1997) 9734–9737. [63] P.K. Sahu, S.K. Das, M. Sarkar, Toward understanding solute− solvent interaction in
[45] A.A. Guzelian, J.E.B. Katari, A.V. Kadavanich, U. Banin, K. Hamad, E. Juban, A. roomtemperature mono- and dicationic ionic liquids: a combined fluorescence
P. Alivisatos, R.H. Wolters, C.C. Arnold, J.R. Heath, Synthesis of size-selected, spectroscopy and mass spectrometry analysis, J. Phys. Chem. B 118 (2014)
surface-passivated InP nanocrystals, J. Phys. Chem. 100 (1996) 7212–7219. 1907–1915.
[46] D. Roy, T. Routh, A.V. Asaithambi, S. Mandal, P.K. Mandal, Spectral and temporal [64] S.R. Prabhu, G.B. Dutt, Effect of low viscous nondipolar solvent on the rotational
optical behavior of blue-, green-, orange-, and red-emitting cdse-based core/ diffusion of structurally similar nondipolar solutes in an ionic liquid, J. Phys.
gradient alloy shell/shell quantum dots: ensemble and single-particle investigation Chem. B 119 (2015) 2019–2025.
results, J. Phys. Chem. C 120 (2016) 3483–3491. [65] S.K. Das, D. Majhi, P.K. Sahu, M. Sarkar, Investigation of the influence of alkyl side
[47] S. Mandal, D. Roy, C.K. De, S. Ghosh, M. Mandal, A. Das, P.K. Mandal, chain length on the fluorescence response of C153 in a series of room temperature
Instantaneous, room-temperature, open-air atmosphere, solution-phase synthesis ionic liquids, RSC Adv 5 (2015) 41585–41594.
of perovskite quantum dots through halide exchange employing non-metal based [66] C. Lawler, M.D. Fayer, The influence of lithium cations on dynamics and structure
inexpensive HCl/HI: ensemble and single particle spectroscopy, Nanoscale Adv 1 of room temperature ionic liquids, J. Phys. Chem. B 117 (2013) 9768–9774.
(2019) 3506–3513. [67] S.R. Prabhu, G.B. Dutt, Rotational dynamics of imidazolium-based ionic liquids: do
[48] D. Roy, A. Das, C.K. De, S. Mandal, P.R. Bangal, P.K. Mandal, Why does the the nature of the anion and the length of the alkyl chain influence the dynamics?
photoluminescence efficiency depend on excitation energy in case of a quantum J. Phys. Chem. B 118 (2014) 13244–13251.
dot? A case study of CdSe-based core/alloy shell/shell quantum dots employing [68] S. Cha, T. Shim, Y. Ouchi, D. Kim, Characteristics of visible fluorescence from ionic
ultrafast pump–probe spectroscopy and single particle spectroscopy, J. Phys. liquids, J. Phys. Chem. B 117 (2013) 10818–10825.
Chem. C 123 (2019) 6922–6933. [69] S.K. Das, P.K. Sahu, M. Sarkar, Studies on electronic energy transfer (EET) on a
[49] C.K. De, S. Mandal, D. Roy, S. Ghosh, A. Konar, P.K. Mandal, Ultrafast dynamics series of room temperature ionic liquids (RTILs): can the EET studies on RTILs be
and ultrasensitive single-particle intermittency in small-sized toxic metal free InP- exploited to predict their structural organization? RSC Adv. 4 (2014), 39184-3919.
based core/alloy-shell/shell quantum dots: excitation wavelength dependency [70] A. Nazet, S. Sokolov, T. Sonnleitner, T. Makino, M. Kanakubo, R. Buchner,
toward variation of PLQY, J. Phys. Chem. C 123 (2019) 28502–28510. Densities, viscosities, and conductivities of the imidazolium ionic liquids [Emim]
[50] C.K. De, T. Routh, D. Roy, S. Mandal, P.K. Mandal, Highly photoluminescent inp [Ac], [Emim][FAP], [Bmim][BETI], [Bmim][FSI], [Hmim][TFSI], and [Omim]
based core alloy shell QDs from air-stable precursors: excitation wavelength [TFSI], J. Chem. Eng. Data 60 (2015) 2400–2411.
5
pubs.acs.org/Langmuir Article
■ INTRODUCTION
Recently, nanoparticle-oriented research has become an area of
an anticancer drug, doxorubicin, delivery.15 Despite a large
number of investigations mainly on ligand-capped nanoparticle
interest due to its various potential scientific and industrial synthesis with different shapes and sizes, structural details of
applications, such as in chemical, medical, and biological the capping layer are not explored in detail.16−19
fields.1 The versatile properties of nanoparticles, such as The subject of molecular ordering and self-assembly on the
biocompatibility, drug delivery, and plasmonic activity, are nanoparticle surface has been actively discussed during the last
typically controlled by the molecular functionalization of their decade. However, the efficacy of nanomedicines remains
surfaces.2−5 Thus, nanoparticle surface chemistry is one of the inadequate, mainly because of a lack of information about
most thriving modern research fields. Over the past few years, the surface structural properties of nano−biointeractions.17
metal nanoparticles have played a key role in modern The self-assembled monolayer structure of larger peptides on
medicine, being a contrast carrier for drug delivery to target the nanoparticle surfaces fully depends on the curvature of the
cells.6,7 Constancy under appropriate physicochemical environ- nanoparticles.18,19 In the presence of mercaptobenzoic acid,
ments is a challenging requirement for such in vivo peptides form the β-sheet structure on AuNPs.20 A detailed
applications.8−10 Nowadays, various coating agents, such as investigation of ligand surface conformation is obligatory to
organic molecules, peptides, proteins, DNA, etc., are fully discover the broad prospects of the capping ligands in
extensively used for stability and to increase the solubility of
nanomaterials in ionic solutions.11,12 Among these capping
agents, peptides provide invaluable support in material sciences Received: May 22, 2020
by interacting with the surface of noble metal nanoparticles.8,13 Revised: September 2, 2020
The inorganic nano−biointeractions are comprised of Published: September 3, 2020
exchanges of kinetic, thermodynamic, and physicochemical
interactions between the nanoparticle surfaces and peptides.14
It is reported that DNA-capped gold nanoparticles are used for
■
GSH. Using CD measurement, we compared and confirmed
the structural conformation of GSH on AgNP and AuNP with
RESULTS AND DISCUSSION our preliminary findings estimated from linear IR studies.
IR absorption spectroscopy is one of the widely used Figure 2a (red curve) shows a negative peak at 225 nm and a
techniques for characterizing the secondary structure of positive peak at around 205 nm, which indicates the formation
proteins and peptides. The most intense IR absorbance band of the β-sheet of GSH on AgNPs. From Figure 2b, it is clear
(1600−1700 cm−1) of amide I is mostly used to investigate the that the negative band at around 190 nm and a positive band at
structures of proteins and peptides. The conformation- around 213 nm revealed the random structure of GSH on
sensitive amide I vibrational band directly obtained from the AuNPs. No spectral signature is observed for the ordered
backbone of proteins and peptides is associated with the CO structure of GSH on AuNPs. On the other hand, Figure 2a
stretching vibration. All of the linear IR absorption spectra of (black curve) shows two intensely negative bands at 204 and
GSH are taken in D2O. The representative spectra are shown 218 nm of free GSH. Hence, from the above studies, we have
in Figure 1. We have compared here the IR absorption demonstrated that after the coating of GSH on nanoparticles, it
spectrum of amide I of free GSH with those of GSH capped on adopts different molecular conformations on the surface of the
AuNP and AgNP (comparable size ∼37 nm) to reveal their same-sized and -shaped AgNP and AuNP. Thus, the above
11256 https://dx.doi.org/10.1021/acs.langmuir.0c01527
Langmuir 2020, 36, 11255−11261
Langmuir pubs.acs.org/Langmuir Article
linear IR and CD spectra display the nanoparticle-induced Figure 4. Orientation of amide I stretching mode of GSH on (a)
structural changes and the secondary structure formation of AgNP and (b) AuNP.
the tripeptide GSH after capping on 37 nm nanoparticles. It is
commonly known that to facilitate the arrangement of peptides
(Figure 4b). Thus, we showed that the GSH molecules on a
into β-sheets, sequences with at least three to six amide units
AgNP form β-sheet-like structures, in which amide I is almost
are needed.43−45 Remarkably, GSH has only two amide units
parallel to the surface of the AgNP, while on the same-sized
participating in the β-sheet formation, and the driving force for
AuNP surface, they are randomly oriented where amide I is
its self-assembly is provided by the interaction with the
perpendicular to the AuNP surface. This disparity of surface
nanoparticle’s surface.
■
structure of GSH on AgNP and AuNP undoubtedly describes
SURFACE-ENHANCED INFRARED ABSORPTION the suppression and enhancement of amide I.
(SEIRA) In this work, we have demonstrated that GSH forms β-sheet-
like structures with an orientation nearly parallel to the AgNP
IR absorption is enhanced when molecules are adsorbed on surface, thus “wrapping” the nanoparticle, and a random
metal surfaces (such as gold and silver) and the phenomenon structure on AuNP in an aqueous solution.
is known as surface-enhanced infrared absorption Now, the question as to why GSH forms β-sheet structures
(SEIRA).46−51 The strength of the SEIRA depends on the on AgNP but not on AuNP arises. To achieve surface
metal type, surface morphology, as well as the orientation of structural disparity of GSH, we monitored the carboxylic
the molecular vibrations with the metal surfaces.52 The amide I asymmetric stretching frequency. The asymmetric stretching
symmetric stretch is the strongest peak visible in the SEIRA frequency of the carboxylic group of GSH on AgNP shifted
spectra. Figure 3a,b presents a comparison of the SEIRA toward the blue region (1735 cm−1) compared to that of free
GSH (1725 cm−1), which is probably due to the bond
formation of the carboxylic group with the AgNP surface
(Figure S3). Therefore, GSH was probably attached to the
AgNP surface by three ends, one thiol and two other carboxylic
ends. Consequently, GSHs formed a β-sheet structure by
positioning parallelly to each other. In the case of GSH on the
AuNP, the IR peak position of carboxylic asymmetric
stretching modes is almost constant compared to that of free
GSH (Figure S3). These could be due to the attachment of the
thiol group to the surface of AuNPs, and the carboxylic groups
Figure 3. (a) Concentration-normalized IR absorption spectra of are freely present in the bulk with a greater distance from the
amide I of free GSH (black curve) and GSH on AgNP (red curve). surface. To validate our assessment, we performed the
following theoretical calculation.
■
(b) Concentration-normalized IR absorption spectra of amide I of
free GSH (black curve) and GSH on AuNP (red curve).
ESTIMATION OF BINDING ENERGY: DFT
spectra of GSH on AgNPs and AuNPs. The spectra, such as CALCULATIONS
those in Figure 3a,b, allow the estimation of relative As there is a competition between the carboxylic group and the
enhancements. The IR absorption of amide I of GSH on thiol group of GSH to attach to the surface of the
AgNPs (red color band) is suppressed to half of that of free nanoparticles, there is a need to define the exact interaction.
GSH (Figure 3a), while in contrast, enhancement of 10 times To gain exact knowledge about the interaction of GSH on the
is observed for the amide I IR absorption spectrum at 1653 surface of AuNP and AgNP in neat water, we performed
cm−1 of GSH anchored on AuNPs (Figure 3b). Compared to density functional theory (DFT) calculations by employing
the free GSH, GSH on AgNP and AuNP display distinctive Gaussian 09 software.53,54 In these calculations, we have used
properties that make it more attractive as a capping ligand. The two different basis sets, 631G for lower atomic weights (C, H,
optical properties of GSH are radically changed after N, O, S) and LanL2DZ is for higher atomic weights (Au and
adsorption on metal nanoparticles. To explain this anomalous Ag), for preliminary geometry optimization. To obtain the
behavior of GSH on plasmonic nanomaterials of the same size, binding energy of GSH on nanoparticles through thiol and
shape, and group of the periodic table, a “surface selection carboxylic oxygen bonds, the “scan” analyses are performed on
rule” was introduced. This rule states that the IR absorption of the B3LYP−LANL2DZ optimized structures. To calculate the
11257 https://dx.doi.org/10.1021/acs.langmuir.0c01527
Langmuir 2020, 36, 11255−11261
Langmuir pubs.acs.org/Langmuir Article
11258 https://dx.doi.org/10.1021/acs.langmuir.0c01527
Langmuir 2020, 36, 11255−11261
Langmuir pubs.acs.org/Langmuir Article
To investigate the stability of the β-sheet of GSH on AgNPs, dependency on different physiological conditions (such as salt
linear IR absorption of amide I in the presence of different salt and pH) is important for nano-biotechnological applications.
concentrations was considered (Figure 8a). The secondary Our work is also important as GSH is an abundant molecule in
the cytoplasm and can be potentially exchanged with various
nanoparticle ligands during nanoparticle-based drug delivery.
Our findings of the surface conformational structure of GSH
would also be useful for other inorganic colloidal nanoparticle
studies.
■
*
ASSOCIATED CONTENT
sı Supporting Information
■ CONCLUSIONS
The present work provides a fundamental understanding of the
Sciences, Kolkata 700106, India
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.langmuir.0c01527
surface conformational structure of small tripeptide GSH on
silver and gold nanoparticles. The sulfur−silver bonds and the Notes
interaction of the carboxylic anions with the silver surface The authors declare no competing financial interest.
■
facilitate the formation of the parallel β-sheet and fibril-like
structures. On the contrary, no secondary structure is formed ACKNOWLEDGMENTS
on the AuNP surface even though they belong to the same
family of the periodic table and have almost the same size and This work was supported in part by the DST, Government of
shape. Despite their similarities, they exhibit different behavior India [DST/INSPIRE/04/2018/002031].
with GSH to form different surface conformational structures.
In this study, we have reported the suppression of the amide I
IR absorption band of GSH on the AgNP surface because of
■ REFERENCES
(1) Khan, I.; Saeed, K.; Khan, I. Nanoparticles: Properties,
the nearly parallel orientation of the stretching mode with the applications and toxicities. Arabian J. Chem. 2019, 12, 908−931.
surface. However, IR absorption of amide I was enhanced (2) Jain, P. K.; Huang, X.; El-Sayed, I. H.; El-Sayed, M. A. Review of
significantly for its nonparallel stretching mode with the AuNP Some Interesting Surface Plasmon Resonance-enhanced Properties of
surface. In comparison with spherical AgNPs, superior Noble Metal Nanoparticles and Their Applications to Biosystems.
enhancement due to the GSH conformational structure was Plasmonics 2007, 2, 107−118.
(3) Hu, C.; Rong, J.; Cui, J.; Yang, Y.; Yang, L.; Wang, Y.; Liu, Y.
confirmed for AuNPs. The obtained results could demonstrate Fabrication of a graphene oxide-gold nanorod hybrid material by
the utility of Au and Ag nanostructures in SEIRA applications electrostatic self-assembly for surface-enhanced Raman scattering.
of amide I, which is considered to be relevant in biosensing Carbon 2013, 51, 255−264.
and biomarkers. GSH-capped AgNPs are more stable (4) Castellana, E. T.; Gamez, R. C.; Russell, D. H. Label-free
compared to AuNP in the presence of calcium chloride. The Biosensing with Lipid-Functionalized Gold Nanorods. J. Am. Chem.
structural conformation study of GSH on nanoparticles and its Soc. 2011, 133, 4182−4185.
11259 https://dx.doi.org/10.1021/acs.langmuir.0c01527
Langmuir 2020, 36, 11255−11261
Langmuir pubs.acs.org/Langmuir Article
(5) Yin, P. T.; Kim, T.-H.; Choi, J.-W.; Lee, K.-B. Prospects for Targeted Intracellular Drug and Gene Delivery. J. Controlled Release
graphene-nanoparticle-based hybrid sensors. Phys. Chem. Chem. Phys. 2011, 152, 2−12.
2013, 15, 12785−12799. (26) Lund, T.; Callaghan, M. F.; Williams, P.; Turmaine, M.;
(6) Murthy, S. K. Nanoparticles in modern medicine: State of the art Bachmann, C.; Rademacher, T.; Roitt, I. M.; Bayford, R. The
and future challenges. Int. J. Nanomed. 2007, 2, 129−141. influence of ligand organization on the rate of uptake of gold
(7) Zou, L.; Wang, H.; He, B.; Zeng, L.; Tan, T.; Cao, H.; He, X.; nanoparticles by colorectal cancer cells. Biomaterials 2011, 32, 9776−
Zhang, Z.; Guo, S.; Li, Y. Current Approaches of Photothermal 9784.
Therapy in Treating Cancer Metastasis with Nanotherapeutics. (27) Berg, J. M.; Tymoczko, J. L.; Stryer, L. Biochemistry, 5th ed.;
Theranostics 2016, 6, 762−772. Freeman & Company: New York, 2002.
(8) Cobley, C. M.; Chen, J.; Cho, E. C.; Wang, L. V.; Xia, Y. Gold (28) Hepel, M.; Tewksbury, E. Ion-Gating Phenomena of Self-
nanostructures: a class of multifunctional materials for biomedical Assembling Glutathione Films on Gold Piezoelectrodes. J. Electro-
applications. Chem. Soc. Rev. 2011, 40, 44−56. anal.Chem. 2003, 552, 291−305.
(9) Han, G.; Martin, C. T.; Rotello, V. M. Stability of Gold (29) Bieri, M.; Bürgi, T. L-Glutathione Chemisorption on Gold and
Nanoparticle-Bound DNA toward Biological, Physical, and Chemical Acid/Base Induced Structural Changes: A PM-IRRAS and Time-
Agents. Chem. Biol. Drug. Des. 2006, 67, 78−82. Resolved in Situ ATR-IR Spectroscopic Study. Langmuir 2005, 21,
(10) Giljohann, D. A.; Seferos, D. S.; Daniel, W. L.; Massich, M. D.; 1354−1363.
Patel, P. C.; Mirkin, C. A. Gold Nanoparticles for Biology and (30) Bieri, M.; Bürgi, T. Adsorption Kinetics of L-Glutathione on
Medicine. Angew. Chem., Int. Ed. 2010, 49, 3280−3294. Gold and Structural Changes During Self-Assembly: An in Situ
(11) Duchesne, L.; Gentili, D.; Comes-Franchini, M.; Fernig, D. G. ATRIR and QCM Study. Phys. Chem. Chem. Phys. 2006, 8, 513−520.
Robust Ligand Shells for Biological Applications of Gold Nano- (31) Maza, F. L.; Leo, L. M. D.; Rubert, A. A.; Carro, P.; Salvarezza,
particles. Langmuir 2008, 24, 13572−13580. R. C.; Vericat, C. New Insight into the Interface Chemistry and
(12) Han, G.; Ghosh, P.; Rotello, V. M. Functionalized Gold Stability of Glutathione Self-Assembled Monolayers on Au(III). J.
Nanoparticles for Drug Delivery. Nanomedicine 2007, 2, 113−123. Phys. Chem. C 2016, 120, 14597−14607.
(13) Slocik, J. M.; Naik, R. R. Probing peptide−nanomaterial (32) Vallée, A.; Humblot, V.; Méthivier, C.; Pradier, C.-M.
interactions. Chem. Soc. Rev. 2010, 39, 3454−3463. Glutathione Adsorption from UHV to the Liquid Phase at VariouspH
(14) Nel, A. E.; Madler, L.; Velegol, D.; Xia, T.; Hoek; on Gold and Subsequent Modification of Protein Interaction. Surf.
Somasundaran, P.; Klaessig, F.; Castranova, V.; Thompson, M. Interface Anal. 2008, 40, 395−399.
understanding biophysicochemical interactions at the nano-bio (33) Vallée, A.; Humblot, V.; Méthivier, C.; Pradier, C.-M.
interface. Nat. Mater. 2009, 8, 543−557. Adsorption of a Tripeptide, GSH, on Au (111) under UHVCondi-
(15) Alexander, C. M.; Maye, M. M.; Dabrowiak, J. C. DNA-capped tions; Pm-Rairs and Low T-XPS Characterisation. Surf. Sci. 2008, 602,
nanoparticles designed for doxorubicin drug delivery. Chem. Commun. 2256−2263.
2011, 47, 3418−3420. (34) Moaseri, E.; Bollinger, J. A.; Changalvaie, B.; Johnson, L.;
(16) Park, J.-W.; Shumaker-Perry, J. S. Structural Study of Citrate Schroer, J.; Johnston, K. P.; Truskett, T. M. Reversible Self-Assembly
Layers on Gold Nanoparticles: Role of Intermolecular Interactions in of Glutathione-Coated Gold Nanoparticle Clusters via pH-Tuna-
Stabilizing Nanoparticles. J. Am. Chem. Soc. 2014, 136, 1907−1921. bleInteractions. Langmuir 2017, 33, 12244−12253.
(17) Wang, Y.; Cai, R.; Chen, C. The Nano-Bio Interactions of (35) Ghosh, A.; Prasad, A. K.; Chuntonov, L. Two-Dimensional
Nanomedicines: Understanding the Biochemical Driving Forces and Infrared Spectroscopy Reveals Molecular Self-Assembly on the
Redox Reactions. Acc. Chem. Res. 2019, 52, 1507−1518. Surface of Silver Nanoparticles. J. Phys. Chem. Lett. 2019, 10,
(18) Shaw, C. P.; Middleton, D. A.; Volk, M.; Levy, R. Amyloid- 2481−2486.
Derived Peptide Forms Self-Assembled Monolayers on Gold (36) Qian, W.; Krimm, S. Vibrational Analysis of Glutathione.
Nanoparticle with a Curvature-Dependent β-Sheet Structure. ACS Biopolymers 1994, 34, 1377−1394.
Nano 2012, 6, 1416−1426. (37) Wang, L.; Middleton, C. T.; Singh, S.; Reddy, A. S.; Woys, A.
(19) Mandal, H. S.; Kraatz, H.-B. Effect of the Surface Curvature on M.; Strasfeld, D. B.; Marek, P.; Raleigh, D. P.; de Pablo, J. J.; Zanni,
the Secondary Structure of Peptides Adsorbed on Nanoparticles. J. M. T.; Skinner, J. L. 2DIR Spectroscopy of Human Amylin Fibrils
Am. Chem. Soc. 2007, 129, 6356−6357. Reflects Stable B-Sheet Structure. J. Am. Chem. Soc. 2011, 133,
(20) Ho, J.-J.; Ghosh, A.; Zhang, T. O.; Zanni, M. T. Heterogeneous 16062−16071.
Amyloid B-Sheet Polymorphs Identified on Hydrogen Bond (38) Hahn, S.; Kim, S.-S.; Lee, C.; Cho, M. Characteristic Two-
Promoting Surfaces Using 2D SFG Spectroscopy. J. Phys. Chem. A Dimensional IR Spectroscopic Features of Antiparallel and Parallel B-
2018, 122, 1270−1282. Sheet Polypeptides: Simulation Studies. J. Chem. Phys. 2005, 123,
(21) Mosquera, J.; Henriksen-Lacey, M.; García, I.; Martínez-Calvo, 084905−084912.
M.; Rodríguez, J.; Mascareñas, J. L.; Liz-Marzán, L. M. Cellular (39) Moran, S. D.; Zanni, M. T. How to Get Insight into Amyloid
Uptake of Gold Nanoparticles Triggered by Host−Guest Interactions. Structure and Formation from Infrared Spectroscopy. J. Phys.
J. Am. Chem. Soc. 2018, 140, 4469−4472. Chem.Lett. 2014, 5, 1984−1993.
(22) Eibling, M. J.; MacDermaid, C. M.; Qian, Z.; Lanci, C. J.; Park, (40) Lomont, J. P.; Rich, K. L.; Maj, M.; Ho, J.-J.; Ostrander, J. S.;
S.-J.; Saven, J. G. Controlling Association and Separation of Zanni, M. T. Spectroscopic Signature for Stable B-Amyloid Fibrils
GoldNanoparticles with Computationally Designed Zinc-Coordinat- Versus β-Sheet-Rich Oligomers. J. Phys. Chem. B 2018, 122, 144−153.
ing Proteins. J. Am. Chem. Soc. 2017, 139, 17811−17823. (41) Marsh, D. Dichroic Ratios in Polarized Fourier Transform
(23) Pengo, P.; Ş ologan, M.; Pasquato, L.; Guida, F.; Pacor, S.; Infrared for Nonaxial Symmetry of Beta-Sheet Structures. Biophys. J.
Tossi, A.; Stellacci, F.; Marson, D.; Boccardo, S.; Pricl, S.; Posocco, P. 1997, 72, 2710−2718.
Gold Nanoparticles with Patterned Surface Monolayers for Nano- (42) Greenfield, N. J. Using Circular Dichroism Spectra to Estimate
medicine: Current Perspectives. Eur. Biophys. J. 2017, 46, 749−771. Protein Secondary Structure. Nat. Protoc. 2007, 1, 2876−2890.
(24) Ling, X.; Tu, J.; Wang, J.; Shajii, A.; Kong, Na.; Feng, C.; (43) Panda, J. J.; Chauhan, S. V. Short peptide based self-assembled
Zhang, Y.; Yu, M.; Xie, T.; Bharwani, Z.; Aljaeid, B. M.; Shi, B.; Tao, nanostructures: implications in drug delivery and tissue engineering.
W.; Farokhzad, O. C. Glutathione-Responsive Prodrug Nanoparticles Polym. Chem. 2014, 5, 4418−4436.
for Effective Drug Delivery and Cancer Therapy. ACS Nano 2019, 13, (44) Nowick, S. J. Exploring B-Sheet Structure and Interactions with
357−370. Chemical Model Systems. Acc. Chem. Res. 2008, 41, 1319−1330.
(25) Cheng, R.; Feng, F.; Meng, F.; Deng, C.; Feijen, J.; Zhong, Z. (45) Graf, P.; Masic, A.; Thunemann, F. A.; Taubert, A.; et al.
Glutathione-Responsive Nano-Vehicles as a Promising Platform for Peptide-Coated Silver Nanoparticles: Synthesis, Surface Chemistry,
11260 https://dx.doi.org/10.1021/acs.langmuir.0c01527
Langmuir 2020, 36, 11255−11261
Langmuir pubs.acs.org/Langmuir Article
11261 https://dx.doi.org/10.1021/acs.langmuir.0c01527
Langmuir 2020, 36, 11255−11261
Article
■ INTRODUCTION
Hydrogen bond (hb) interactions are ubiquitous in protein
coupling.17−19 Recently, vibrational couplings between carbon-
yl bonds located on DNA nucleobases have been revealed by
molecules and play key roles in numerous biological processes 2D IR spectroscopy.20−22 Coupling between two isotope-based
like molecular association, catalysis, and signal transmission.1−7 nitrile groups (IR reporter) has been studied on DNA
Although peptide (−CONH−) units in proteins provide templates.23 Parallel β-sheet vibrational couplings were
several infrared (IR) signatures (amide I, amide II, amide III, revealed by 2D IR spectroscopy.24 Vibrational dynamics and
and amide A), the amide I band (1600−1700 cm−1) is mostly vibrational modes of coupling of amide A and amide I/II
studied in conjunctions to the protein structure and dynamics modes within the same amide unit and those connected by a
because of its sensitivity to different secondary structures. hb for several model dipeptides have been studied by two-color
Moreover, the amide I mode is not strongly influenced by the 2D IR spectroscopy.25 As hb’s of amide I make a very crucial
side-chain absorptions.5 Further, amide I is widely utilized as role in the structural conformation of protein and peptide,
IR probes because of their sensitivity to the local electrostatic vibrational dynamic studies of amide I is a recent utmost topic
environment, solvation, and large molar extinction coeffi- of interest.26−32 To understand hydrogen bonding ability of
cient.8−12 In particular, IR absorption measurements of amide amide I with hb donor phenol derivatives, dimethyl formamide
I band is sensitive in a variety of biological conformations and and dimethyl acetamide have been studied.33,34 Recently, a
solvation in a way that it reflects vibrational couplings and single peptide unit N-methylacetamide has been studied to
hydrogen bonding.4−7 A lot of experiments have been done to explore the nature of solute−solvent interactions on amide I
investigate the relationships between amide I vibrational vibrational dynamics by using 2D IR experiments and
couplings and peptide or protein conformations, and most
recently, peptide conformational dynamics has been stud- Received: May 29, 2019
ied.8−16 Theoretical two-dimensional IR (2D IR) spectro- Revised: August 24, 2019
scopic approach has been taken to model the vibrational Published: August 24, 2019
■ EXPERIMENTAL SECTION
Phenol, para-chlorophenol (p-CP), para-cresol (PC), anisole,
components (6.81, 0.872, and 0.024), such that the third
component can be neglected undoubtedly. I have performed
multivariate curve resolution (MCR) analysis on the linear IR
DMF, and methyl acetate (MA) are purchased from Sigma- data of amide I transition of DMF for the lower concentrations
Aldrich and used without any further purification. A solution of range of phenol.40,41 During the MCR analysis on lower
0.1 M DMF in CCl4 was used for Fourier transform infrared concentrations data, two distinct components with different
(FTIR) experiments. The sample was placed in a homemade line shapes of amide I were observed. For a higher
FTIR sample cell with CaF2 windows and 60 μm Teflon concentrations range, phenol concentration was increased by
spacer. The linear IR absorption spectrum was measured using a step of 0.05 M up to 0.5 M. The SVD analysis of the higher
an FTIR spectrometer (JASCO-FTIR-6300). The solvent concentrations of phenol indicates the presence of three
background of CCl4 was also measured and subtracted from all significant components, with the first four singular values of
the linear spectra. The same procedure was followed for MA− 7.37, 3.80, 0.807, and 0.111. Thus, the fourth component could
phenol hb interaction in CCl 4 and purine−DMF in be easily neglected. The obtained line shapes for both 0 hb and
acetonitrile. 1 hb complexes at the lower concentrations of phenol (up to
■
0.050 M) was used as the initial guess in the MCR analysis of
the high phenol concentration spectral series (up to 0.5 M).
RESULT AND DISCUSSION The final MCR results were checked for consistency such
A series of the IR absorption spectra of 0.1 M DMF solutions that the line shapes obtained from the low and high
in carbon tetrachloride with phenol concentrations ranging concentration series become identical. The obtained spectral
from 0 to 0.5 M were collected. The representative spectra are profiles of the individual components, bonding to either a
shown in Figure 1a. In the absence of phenol molecules, one single phenol (1 hb) or to a pair of phenol molecules (2 hb)
black single peak spectrum with frequency 1686 cm−1 of amide with transition frequency 1673 and 1664 cm−1, respectively,
I transition was obtained in Figure 1a. In the lower range of are presented in Figure 1b. Amide I absorption frequency shifts
concentrations, the phenol concentrations were increased in to a different extent in each case because hb formation reduces
steps of 0.005 M up to a concentration of 0.050 M (to avoid electron density from amide I (CO bond) and consequently
self-aggregation of phenol). Absorption of amide I transition red-shifts of its stretching frequency. Ensemble populations of
was gradually decreased and new hydrogen-bonded amide I all components were extracted from the individual peak area,
transition was generated in the red-shifted region. The singular obtained from MCR analysis in the presence of different
7772 DOI: 10.1021/acs.jpcb.9b05118
J. Phys. Chem. B 2019, 123, 7771−7776
The Journal of Physical Chemistry B Article
Figure 2. IR absorbance spectra of 0.1 M of (a) phenol, (b) PC, and (c) p-CP with the increasing concentration of DMF (0.00 Mblack, 0.02
Mred, 0.04 Mgreen, 0.06 Mblue, 0.08 Mcyan, 0.10 Mmagenta).
concentrations of phenol. Figure 1c shows the population tional coupling (see Videos S1 and S2). Tucker and co-workers
ensembles as a function of phenol concentrations. Three have shown the vibrational coupling of azide (2124 cm−1) and
distinct concentration regimes are observed for 0 hb, 1 hb, and nitrile (2242−1) reporters in RNA nucleoside even if they have
2 hb conformations. The molar integrated extinction a 118 cm−1 frequency gap.45 Thus, their work also validates our
coefficient of 1 hb conformation of amide I was calculated observation. The amide I and CC transition couples their
from lower concentrations data of phenol and that value was local modes to create delocalized normal modes. Hence, Figure
used to calculate the concentrations of 2 hb conformation in 2a−c shows the clear agreement of vibrational coupling in
higher concentration series. As in lower concentrations of between amide I transition of DMF and CC transition of
phenol, only the 1 hb conformation formed with DMF, the phenol, PC, and p-CP, respectively. Thus, people can predict
corresponding data were fitted to a 1:1 binding isotherm Y11 = coupled modes and therefore can identify the molecular
(K1X)/(1 + K1X), where Y is the fraction of the H-bonded structures by either the frequency shift or the IR absorbance
DMF, X is the concentration of the free phenol molecules in change caused by delocalization of the vibrational modes. The
solution, and K1 is the hb formation constant of the 1 hb obtained results from Figure 4b,c clearly indicate that MA
complex with value 26.41. At higher phenol concentrations, the (1748 cm−1) and phenol (1597, 1605 cm−1) have enough
corresponding data were fitted to a 1:2 binding isotherm frequency gap (∼140 cm−1) for not coupling or negligible
(Figure 1d) given by Y12 = (K1X + K1K2X2)/(1 + K1X + coupling. Now the question is the dependency of relative
K1K2X2) where K2 is the stepwise formation constant of the 2 coupling strength on individual hb conformation. As CC
hb complex with the value of 6.98.42−44 transition for 0 hb, 1hb, and 2 hb conformations are very close
Two CC antisymmetric and symmetric stretching modes to each other, it is quite difficult to isolate the individual
of the phenol ring have a Lorentzian line shape with transition transition peak and peak area contribution (Figure 2a). To do
frequencies 1597 and 1605 cm−1 (Figure 2a). The data so, 0.1 M of DMF has been taken, and the same experiment
presented in Figure 2a suggest that upon the formation of the like Figure 1a was repeated by monitoring the CC transition
hb complexes, IR absorbance of CC transition of phenol (Figure 3a) from 1580 to 1620 cm−1. To calculate the
enhances gradually and frequency shifts slightly toward lower individual peak area of CC transition for free, 1 hb, and 2 hb
frequency with the increasing concentrations of DMF. Now conformations, the integrated extinction coefficient of
the question arises, why the CC transition absorption has respective conformations have been calculated. The integrated
enhanced. This enhancement could be the reason of n−π* extinction coefficient of CC transition of free phenol was
bonding of DMF and phenol ring or polarity change or altered calculated from the standard plot, the concentration of free
electron density of the vibrating bond or vibrational coupling. phenol versus CC transition IR absorbance (Figure 3b). To
To rule out the effect of n−π* bonding of lone pair of “O” calculate the integrated extinction coefficient of 1 hb
atom in DMF and “π” cloud of phenol ring, 1:1 concentration conformation, lower concentrations (0−0.05 M) of phenol
of DMF (0.1 M) and anisole (0.1 M) were mixed and the IR series has been taken to avoid the formation of 2 hb
absorption spectrum was collected. The spectra in Figure 4a conformation. As the concentrations of free phenol and 1 hb
clearly suggest that there was no interaction of nonbonded conformations were known in lower concentration series, the
electron of “O” atom in DMF with the “π” electron cloud of area contribution of free phenol was calculated by multiplying
anisole. This result also confirms that the CC IR transition the integrated extinction coefficient with the concentration of
enhancements are not due to the polarity change of the free phenol. The area contribution of CC transition of 1 hb
solvent, as the volume of additional DMF (9.4 μL) in anisole conformation was calculated by subtracting the area of CC
solution (990.6 μL) and other phenol derivatives solution transition of free phenol from total ensemble area of lower
(990.6 μL) are same. To check the altered electron density of concentration series (0−0.05 M). The subtracted area versus
the vibrating bond, IR absorption spectra were collected from concentration of 1 hb conformation was plotted in Figure 3c to
MA and phenol in 1:1 concentration. The spectra shown in calculate the integrated extinction coefficient CC transition
Figure 4b speak the formation of hb conformation of MA (0.1 of 1 hb conformation. Similarly, to calculate the area
M) with phenol (0.1 M), but CC IR absorption enhance- contribution of CC transition for 2 hb conformation, the
ment, position, and line shape were negligibly changed (Figure area contribution of free phenol and 1 hb conformation was
4c). These results invalidate the arguments of enhancement subtracted from the total CC transition area of ensembles in
because of altered electron density of the vibrating bond. Thus, higher concentration series. Area of CC transition versus 2
the observed enhancement (Figure 2a) in IR absorption clearly × concentration of 2 hb conformation was plotted in Figure 3d
indicates the presence of vibrational coupling even though to calculate the integrated extinction coefficient. By comparing
amide I and CC transitions have ∼59 cm−1 (1 hb) and ∼68 the slope of 0 hb, 1 hb, and 2 hb, it is clear that the slope
cm−1 (2 hb) frequency gap. Density functional theory (DFT) of Figure 3c is higher than that of Figure 3d, and consequently,
calculation videos have strengthened my argument of vibra- the slope of Figure 3d is higher than that of 3a, which
7773 DOI: 10.1021/acs.jpcb.9b05118
J. Phys. Chem. B 2019, 123, 7771−7776
The Journal of Physical Chemistry B Article
Figure 4. (a) IR absorbance of CC transition of anisole (0.1 M) in the absence (black) and presence (red) of DMF (0.1 M). (b) IR spectra of
MA (0.1 M) in the absence (black) and presence (red) of phenol (0.1 M) in CCl4. (c) IR spectra of CC transition of phenol (0.1 M) in the
absence (black) and presence (red) of MA (0.1 M).
Table 1. Analysis of H-Bond Formation Constant, Relative Integrated Area, and Average Relative Transition Dipole Momenta
H-bond donor pKa K1 (M−1) K2 (M−1) ≈
ΔCC−CO
1hb (cm−1) ≈
ΔCC−CO
2hb (cm−1) ICC CC
1hb/I0hb ICC CC
2hb/I0hb μCC CC
1hb/μ0hb μCC CC
2hb/μ0hb
■
*
ASSOCIATED CONTENT
S Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.jpcb.9b05118.
Linear IR data of PC and p-CP (PDF)
Vibrational video of 1 hb conformation (from DFT
Figure 5. (a) Purine ring vibrational transition spectra in the absence calculation) (AVI)
(black) and presence (red) of DMF in acetonitrile with 1:1 Vibrational video of 2 hb conformation (from DFT
concentration ratio. (b) One hydrogen-bonded conformation of calculation) (AVI)
■
purine and DMF.
AUTHOR INFORMATION
pair of DNA, amide I and CC transition vibrationally may Corresponding Author
be coupled. As still there is no report of vibrational coupling *E-mail: anupg86@gmail.com, anup.ghosh@bose.res.in.
between carbonyl and ring transition of nucleobases, this study ORCID
could be a point of reference for future studies. Anup Ghosh: 0000-0002-1442-8740
■ CONCLUSIONS
Notes
The author declares no competing financial interest.
In conclusion, I have established experimentally the hydrogen-
bonded structural information of amide I in the presence of hb
donor phenol derivatives using linear IR spectroscopy. The
■ ACKNOWLEDGMENTS
This work was supported in part by DST Government of India
main spectroscopic features from all phenol derivatives [DST/INSPIRE/04/2018/002031].
■
approaches agree very well, enabling interpretation in terms
of hydrogen bonding population and stepwise hb formation- REFERENCES
based vibrational coupling. I found that amide I forms two (1) Dickerson, R. E.; Geis, I. The Structure and Action of Proteins; W.
main hydrogen-bonded conformations, of which the vibra- A. Benjamin, Inc.: Menlo Park, CA, 1969.
tional coupling of amide I and CC transition are different for (2) Nelson, D. L.; Cox, M. M. Lehninger Principles of Biochemistry;
different hb conformations. Relative area and average transition W. H. Freeman and Company: New York, 2005.
dipole moment enhancement of CC IR absorbance were (3) Berg, J. M.; Tymoczko, J. L.; Stryer, L. Biochemistry; W. H.
higher for 1 hb conformation than 2 hb conformation for all Freeman and Company: New York, 2007.
the phenol derivatives, whereas the energy gap of the two (4) Krimm, S.; Bandekar, J. Vibrational spectroscopy and
dipoles is less in the case of 2 hb conformation than 1 hb conformation of peptides, polypeptides, and proteins. Adv. Protein
conformation. For the first time, hb formation dependent Chem. 1986, 38, 181−364.
vibrational coupling of amide I where mechanical effects and (5) Barth, A.; Zscherp, C. What vibrations tell us about proteins. Q.
Rev. Biophys. 2002, 35, 369−430.
distance between the two coupled modes might play significant (6) Woutersen, S.; Hamm, P. Nonlinear two-dimensional vibrational
roles has been reported. This observation definitely demon- spectroscopy of peptides. J. Phys.: Condens. Matter 2002, 14, R1035−
strates a delicate interplay between coupling effects and hb R1062.
formation constant of amide I. Thus, by linear IR spectroscopy, (7) Hamm, P.; Lim, M.; Hochstrasser, R. M. Structure of the Amide
I have unraveled the interplay between hydrogen bonding and I Band of Peptides Measured by Femtosecond Nonlinear-Infrared
vibrational coupling. My study could be an important point of Spectroscopy. J. Phys. Chem. B 1998, 102, 6123−6138.
reference for the protein and nucleic acid structure and (8) Dyer, R. B.; Gai, F.; Woodruff, W. H.; Gilmanshin, R.; Callender,
dynamics experiment to unravel the IR spectral signatures. I R. H. Infrared Studies of Fast Events in Protein Folding. Acc. Chem.
believe that this report provides the most inclusive results into Res. 1998, 31, 709−716.
the origins of the couplings in the host−guest hydrogen (9) Woutersen, S.; Pfister, R.; Hamm, P.; Mu, Y.; Kosov, D. S.;
Stock, G. Peptide conformational heterogeneity revealed from
bonding study in protein and peptides. The results presented
nonlinear vibrational spectroscopy and molecular-dynamics simu-
in this paper provide an insight into the quantification of hb lations. J. Chem. Phys. 2002, 117, 6833−6840.
conformation of amide and quantification of the relative (10) Demirdöven, N.; Cheatum, C. M.; Chung, H. S.; Khalil, M.;
coupling modes utilized to interpret the IR spectra of proteins, Knoester, J.; Tokmakoff, A. Two-Dimensional Infrared Spectroscopy
peptides, and guest molecules. The detailed comparison among of Antiparallel β-Sheet Secondary Structure. J. Am. Chem. Soc. 2004,
all phenol derivatives strengthens our confidence in the ability 126, 7981−7990.
(11) Waegele, M. M.; Culik, R. M.; Gai, F. Site-Specific (30) Kubelka, J.; Keiderling, T. A. Ab Initio Calculation of Amide
Spectroscopic Reporters of the Local Electric Field, Hydration, Carbonyl Stretch Vibrational Frequencies in Solution with Modified
Structure, and Dynamics of Biomolecules. J. Phys. Chem. Lett. 2011, 2, Basis Sets. 1.N-Methyl Acetamide. J. Phys. Chem. A 2001, 105,
2598−2609. 10922−10928.
(12) Serrano, A. L.; Waegele, M. M.; Gai, F. Spectroscopic studies of (31) Ham, S.; Kim, J.-H.; Lee, H.; Cho, M. Correlation between
protein folding: Linear and nonlinear methods. Protein Sci. 2012, 21, electronic and molecular structure distortions and vibrational
157−170. properties. II. Amide I modes of NMA-nD2O complexes. J. Chem.
(13) Hamm, P.; Hochstrasser, R. M. Structure and dynamics of Phys. 2003, 118, 3491−3498.
proteins and peptides: femtosecond two-dimensional infrared spec- (32) Ham, S.; Cho, M. Amide I modes in the N-methylacetamide
troscopy. In Ultrafast Infrared and Raman Spectroscopy; Fayer, M. D., dimer and glycine dipeptide analog: Diagonal force constants. J. Chem.
Ed.; Marcel Dekker, Inc.: New York, 2001; Vol. 26, p 273. Phys. 2003, 118, 6915−6922.
(14) Gnanakaran, S.; Hochstrasser, R. M. Conformational (33) Hatton, J. V.; Richards, R. E. Solvent effects in the proton
Preferences and Vibrational Frequency Distributions of Short resonance spectra of dimethyl-formamide and dimethyl-acetamide.
Peptides in Relation to Multidimensional Infrared Spectroscopy. J. Mol. Phys. 1960, 3, 253−263.
Am. Chem. Soc. 2001, 123, 12886−12898. (34) Malathi, M.; Sabesan, R.; Krishnan, S. IR carbonyl band
(15) Woutersen, S.; Hamm, P. Isotope-edited two-dimensional intensity studies in N, N-dimethyl formamide and N, N- dimethyl
vibrational spectroscopy of trialanine in aqueous solution. J. Chem. acetamide on formation with phenols. Curr. Sci. 2002, 86, 838−842.
Phys. 2001, 114, 2727−2737. (35) Woutersen, S.; Mu, Y.; Stock, G.; Hamm, P. Hydrogen-bond
(16) Woutersen, S.; Mu, Y.; Stock, G.; Hamm, P. Subpicosecond lifetime measured by time-resolved 2D-IR spectroscopy: N-methyl-
conformational dynamics of small peptides probed by two-dimen- acetamide in methanol. Chem. Phys. 2001, 266, 137−147.
sional vibrational spectroscopy. Proc. Natl. Acad. Sci. U.S.A. 2001, 98, (36) Zanni, M. T.; Asplund, M. C.; Hochstrasser, R. M. Two-
11254−11258. dimensional heterodyned and stimulated infrared photon echoes of
(17) Zhang, W. M.; Chernyak, V.; Mukamel, S. Multidimensional N-methylacetamide-D. J. Chem. Phys. 2001, 114, 4579−4590.
femtosecond correlation spectroscopies of electronic and vibrational (37) Kwac, K.; Cho, M. Molecular dynamics simulation study of N-
excitons. J. Chem. Phys. 1999, 110, 5011−5028. methylacetamide in water. I. Amide I mode frequency fluctuation. J.
(18) Scheurer, C.; Piryatinski, A.; Mukamel, S. Signatures of β- Chem. Phys. 2003, 119, 2247−2255.
Peptide Unfolding in Two-Dimensional Vibrational Echo Spectros- (38) Kwac, K.; Lee, H.; Cho, M. Non-Gaussian statistics of amide I
copy: A Simulation Study. J. Am. Chem. Soc. 2001, 123, 3114−3124. mode frequency fluctuation of N-methylacetamide in methanol
(19) Moran, A.; Mukamel, S. The origin of vibrational mode solution: linear and nonlinear vibrational spectra. J. Chem. Phys.
couplings in various secondary structural motifs of polypeptides. Proc. 2004, 120, 1477−1490.
Natl. Acad. Sci. U.S.A. 2004, 101, 506−510. (39) Lawton, W. H.; Sylvestre, E. A. Self Modeling Curve
(20) Krummel, A. T.; Zanni, M. T. DNA vibrational coupling Resolution. Technometrics 1971, 13, 617−633.
revealed with two-dimensional infrared spectroscopy: Insight into why (40) Borgen, O. S.; Kowalski, B. R. An extension of the multivariate
vibrational spectroscopy is sensitive to DNA structure. J. Phys. Chem. component-resolution method to three components. Anal. Chim. Acta
B 2006, 110, 13991−14000. 1985, 174, 1−26.
(21) Krummel, A. T.; Mukherjee, P.; Zanni, M. T. Inter and (41) Jaumot, J.; de Juan, A.; Tauler, R. MCR-ALS GUI 2.0: new
Intrastrand Vibrational Coupling in DNA Studied with Heterodyned features and applications. Chemom. Intell. Lab. Syst. 2015, 140, 1−12.
2D-IR Spectroscopy. J. Phys. Chem. B 2003, 107, 9165−9169. (42) Connors, K. A. Binding Constants: The Measurement of
(22) Krummel, A. T.; Zanni, M. T. Interpreting DNA Vibrational Molecular Complex Stability; Wiley: New York, 1987; Vol. 33.
Circular Dichroism Spectra Using a Coupling Model from Two- (43) Thordarson, P. Determining association constants from
Dimensional Infrared Spectroscopy. J. Phys. Chem. B 2006, 110, titration experiments in supramolecular chemistry. Chem. Soc. Rev.
24720−24727. 2011, 40, 1305−1323.
(23) Krummel, A. T.; Zanni, M. T. Evidence for Coupling between (44) Hibbert, D. B.; Thordarson, P. The death of the Job plot,
Nitrile Groups Using DNA Templates: A Promising New Method for transparency, open science and online tools, uncertainty estimation
Monitoring Structures with Infrared Spectroscopy. J. Phys. Chem. B methods and other developments in supramolecular chemistry data
2008, 112, 1336−1338. analysis. Chem. Commun. 2016, 52, 12792−12805.
(24) Woys, A. M.; Almeida, A. M.; Wang, L.; Chiu, C.-C.; (45) Schmitz, A. J.; Hogle, D. G.; Gai, X. S.; Fenlon, E. E.; Brewer, S.
McGovern, M.; de Pablo, J. J.; Skinner, J. L.; Gellman, S. H.; Zanni, H.; Tucker, M. J. Two-Dimensional Infrared Study of Vibrational
M. T. Parallel β-Sheet Vibrational Couplings Revealed by 2D IR Coupling between Azide and Nitrile Reporters in a RNA Nucleoside.
Spectroscopy of an Isotopically Labeled Macrocycle: Quantitative J. Phys. Chem. B 2016, 120, 9387−9394.
Benchmark for the Interpretation of Amyloid and Protein Infrared
Spectra. J. Am. Chem. Soc. 2012, 134, 19118−19128.
(25) Rubtsov, I. V.; Wang, J.; Hochstrasser, R. M. Vibrational
Coupling between Amide-I and Amide-A Modes Revealed by
Femtosecond Two Color Infrared Spectroscopy†. J. Phys. Chem. A
2003, 107, 3384−3396.
(26) Mirkin, N. G.; Krimm, S. Ab initio vibrational analysis of
hydrogen-bonded trans- and cis-N-methylacetamide. J. Am. Chem. Soc.
1991, 113, 9742−9747.
(27) Mirkin, N. G.; Krimm, S. Ab initio vibrational analysis of
isotopic derivatives of aqueous hydrogen-bonded trans-N-methyl-
acetamide. J. Mol. Struct. 1996, 377, 219−234.
(28) Torii, H.; Tatsumi, T.; Kanazawa, T.; Tasumi, M. Effects of
Intermolecular Hydrogen-Bonding Interactions on the Amide I Mode
ofN-Methylacetamide: Matrix-Isolation Infrared Studies and ab Initio
Molecular Orbital Calculations. J. Phys. Chem. B 1998, 102, 309−314.
(29) Herrebout, W. A.; Clou, K.; Desseyn, H. O. Vibrational
Spectroscopy ofN-Methylacetamide Revisited. J. Phys. Chem. A 2001,
105, 4865−4881.
living cells; it is extensively used in NP chemistry and related research. We identified conditions where the interaction of
glutathione with the NP surface facilitates formation of a β-sheet-like structure enclosing the NPs. A spectroscopic signature
associated with the assembly of β-sheets into an amyloid fibril-like structure was also observed. Remarkably, the interaction with
the metal surface promotes formation of a fibril-like structure by a small peptide involving only two amino acids.
Figure 2. Polarization-selective 2DIR spectroscopy of free GSH. The polarization of the excitation pulses and the waiting times corresponding to
the spectra in each column are indicated at the top. The pD values for each row appear on the right.
of the peak areas for both linear and 2DIR spectra represents In the pD 5.5 solutions of free GSH, the signal associated
the ratio of the concentrations of the two conformations. We with the protonated fraction of the carboxylic groups is not
obtained that ca. 15% of the sample adopts fibril-like observed. Regarding SAM GSH, the corresponding transition
conformation, as compared to the 25% reported in ref 47 for appears 6 times weaker than that of the amide-I band in the
amyloid peptides. linear spectrum (Figure S3) and 40 times weaker in 2DIR. The
Reduction in both the inhomogeneous and the homoge- presence of this transition provides interesting information on
neous line width of the SAM GSH, as compared to the free the GSH SAM structure. Because the pKa value of Gly carboxyl
GSH, is reflected in the diagonal and the antidiagonal width of is higher than that of Glu, it is more prone to protonation, for
the 2DIR peak, respectively. The line narrowing is consistent example, by the Glu amine of the neighboring strand, available
with the formation of the ordered intermolecular structure for hydrogen bonding in the β-sheet arrangement. At the same
constrained to the NP surface.23,54 Narrower line shapes time, in contrast to free GSH, no transition associated with the
generally indicate slower molecular dynamics. The correlation anionic carboxyl groups is seen for GSH SAM. This can be
function of the frequency fluctuations (FFCF), reflecting rationalized by the sensitivity of the SAM conformation to the
SAMs’ ultrafast dynamics, decays slower, compared with that silver surface charge. At high pD values, the surface of the silver
of free molecules in solution.55,56 For a single-oscillator NPs61 is negatively charged by the adsorbed hydroxyl anions,
transition, the Tw-dependence of the nodal line slope (NLS) whereas at low pD values the surface charge is positive, and the
of the 2D peak serves as a measure of the spectral diffusion.57 point-of-zero-charge is at pD 7, as estimated by Merga et al.62
In the cases of coupled vibrational modes (free GSH) and Thus, at pD 5.5 the carboxyl anions are prone to interact with
excitonic transitions (GSH SAM), equilibration of the surface cations. This interaction leads to elimination of the
excitation between the coupled modes can scramble the NLS carboxyl stretching transition from the spectral region
signature of the FFCF decay. However, even though the NLSs monitored in our experiment either because of the strong
in our data cannot be strictly associated with the FFCF, it can deformation of the electronic potential, leading to a dramatic
still serve as a parameter qualitatively describing the very shift of the transition frequency, or because of the orientation
different spectral evolution observed in the free GSH and GSH of the transition dipole parallel to the surface and manifestation
SAM.58,59 of the surface selection rules.63
The evolution of the NLS for free and SAM GSH evaluated In order to test the sensitivity of the β-sheet conformation to
at the lower amide-I transition frequency is compared in Figure the solvation environment, we added to the GSH-capped NP
3. The NLS of the free GSH signal decays exponentially (τ = solution either metal ions (BaCl2) or alkali base (KOD).
Barium ions compete with the positively charged NP surface,
because they are chelated by the carboxylic anions,28 leading to
a conformational change in GSH SAM and disruption of β-
sheets.64 The corresponding spectra in Figure 4A,B show
corresponding spectra in Figure 4C,D resemble those observed nanotechnological applications of the GSH-capped NPs but
for solutions of free GSH at high pD values, with emphasized also for NP-based medicine, because GSH is abundant in the
transition at ωex = 1595 cm−1. Analysis of the NLS (Figure 3) cytosol and can potentially exchange with various NP ligands.
indicates that also here static inhomogeneity exceeds that of Internalized NPs with GSH SAM eventually reach the acidic
free GSH. lysosome, where the fibril-like structure on their surface may
Recently, Mandal et al.65 observed that leucine-rich peptide, initiate various processes.72
fully helical in solution, assembles into an antiparallel β-sheet
on the surface of 5 nm NPs (ωex = 1625 cm−1) but not on the
surface of larger 20 nm NPs and on flat surfaces. In contrast,
■
*
ASSOCIATED CONTENT
S Supporting Information
Shaw et al.54 observed that the fraction of the parallel β-sheet The Supporting Information is available free of charge on the
formed by the amyloid-derived peptide (ωex = 1665 and 1640 ACS Publications website at DOI: 10.1021/acs.jp-
cm−1) increased with NP size. Our own results for GSH SAM clett.9b00530.
on the surface of 2 nm silver NPs66 indicate that β-sheets are
not formed and that the corresponding spectra (Figures S3 and Experimental methods and additional experimental
S5) qualitatively resemble those of free GSH at pD 5.5, results (PDF)
■
illustrating the critical role of surface flatness in promoting the
formation of GSH β-sheets. Di Gregorio et al.67 noted that AUTHOR INFORMATION
chiroptical signal of the GSH-capped silver nanocubes vanishes Corresponding Author
upon raising the pH. Indeed, in our samples we observed that *E-mail: chunt@technion.ac.il.
the β-sheet’s circular dichroism, a positive transition at 203 nm,
disappears upon addition of the barium or deuteryl ions ORCID
(Figure S6), confirming disruption of β-sheets. Lev Chuntonov: 0000-0002-2316-4708
In 2DIR spectroscopy, the parallel and antiparallel β-sheets Author Contributions
are distinguished by the cross-peaks appearing in the latter †
A.G. and A.K.P. contributed equally to this work.
case.45,46,52,68 On metal NPs, the electromagnetic boundary Notes
conditions require that direction of the electric field is normal The authors declare no competing financial interest.
■
to the surface69,70 and surface selection rules are implied.63
The low-frequency amide-I exciton mode of the β-sheet has ACKNOWLEDGMENTS
transition dipole moment perpendicular to the strand axis.71
For the parallel β-sheet, the transition dipole is inclined to the This research is supported by the Israel Science Foundation
plane of the sheet, such that if the sheet lies parallel to the (grant 1118/15) and the Russell Berrie Nanotechnology
Institute, Technion.
■
metal surface, its out-of-plane component would survive the
selection rules. On the other hand, for the antiparallel β-sheet,
the corresponding transition dipole moment is in-plane with REFERENCES
the sheet71 and would not be observed. The weaker high- (1) Sapsford, K. E.; Algar, W. R.; Berti, L.; Gemmill, K. B.; Casey, B.
frequency exciton mode of the antiparallel β-sheets has J.; Oh, E.; Stewart, M. H.; Medintz, I. L. Functionalizing
transition dipole moment parallel to the strand, in-plane with Nanoparticles with Biological Molecules: Developing Chemistries
the sheet.71 Thus, any signal from the antiparallel β-sheet That Facilitate Nanotechnology. Chem. Rev. 2013, 113, 1904−2074.
(2) Nel, A. E.; Mädler, L.; Velegol, D.; Xia, T.; Hoek, E. M. V.;
whose plane is parallel to the NP surface would vanish.
Somasundaran, P.; Klaessig, F.; Castranova, V.; Thompson, M.
Interestingly, we have not detected high-frequency transition Understanding Biophysicochemical Interactions at the Nano−Bio
associated with the antiparallel beta-sheet nor the correspond- Interface. Nat. Mater. 2009, 8, 543.
ing cross-peaks in the polarization-selective measurements of (3) Vallee, A.; Humblot, V.; Pradier, C.-M. Peptide Interactions with
SAM GSH (Figure S4), suggesting that if any antiparallel β- Metal and Oxide Surfaces. Acc. Chem. Res. 2010, 43, 1297−1306.
sheet SAM is formed, its plane is parallel to the NP surface and (4) Slocik, J. M.; Naik, R. R. Probing Peptide−Nanomaterial
it is not observed in our measurements. Therefore, the signal Interactions. Chem. Soc. Rev. 2010, 39, 3454−3463.
reported in the present work represents the parallel β-sheet (5) Haran, G.; Chuntonov, L. Artificial Plasmonic Molecules and
SAM. Interestingly, it is in contrast to conclusions of NMR Their Interaction with Real Molecules. Chem. Rev. 2018, 118, 5539−
studies of the oxidized disulfide GSH dimer, self-assembled 5580.
(6) Howes, P. D.; Chandrawati, R.; Stevens, M. M. Colloidal
into gels, where antiparallel β-sheets are stabilized by the Nanoparticles as Advanced Biological Sensors. Science 2014, 346,
restricted rotation about the disulfide bond.17 1247390.
In conclusion, using polarization-selective 2DIR spectrosco- (7) Mikolajczak, D. J.; Koksch, B. Peptide-Gold Nanoparticle
py, we revealed molecular conformations of GSH on the Conjugates as Sequential Cascade Catalysts. ChemCatChem 2018, 10,
surface of silver NPs. The thiol−silver bonds and the 4324−4328.
interaction of the carboxylic anions with silver surface cations (8) Han, G.; Ghosh, P.; Rotello, V. M. Functionalized Gold
facilitate formation of the β-sheet-like and fibril-like structures Nanoparticles for Drug Delivery. Nanomedicine 2007, 2, 113−123.
enclosing the NP. Remarkably, the fibril-like structure is (9) Dykman, L. A.; Khlebtsov, N. G. Uptake of Engineered Gold
formed by a small peptide with only two amino acids. A weak, Nanoparticles into Mammalian Cells. Chem. Rev. 2014, 114, 1258−
1288.
yet visible signal of the protonated carboxyl is also seen, when
(10) Gladytz, A.; Abel, B.; Risselada, H. J. Gold-Induced Fibril
β-sheets are formed. The conformation of the β-sheet appears Growth: The Mechanism of Surface-Facilitated Amyloid Aggregation.
predominantly parallel; however, we cannot rule out the Angew. Chem., Int. Ed. 2016, 55, 11242−11246.
presence of the antiparallel β-sheets, whose plane is parallel to (11) Gao, G.; Zhang, M.; Gong, D.; Chen, R.; Hu, X.; Sun, T. The
the NP surface. The structural response of β-sheets to a change Size-Effect of Gold Nanoparticles and Nanoclusters in the Inhibition
in the environment is important not only for the chemical and of Amyloid-Β Fibrillation. Nanoscale 2017, 9, 4107−4113.
(12) Pengo, P.; Ş ologan, M.; Pasquato, L.; Guida, F.; Pacor, S.; (30) Bieri, M.; Bürgi, T. Adsorption Kinetics of L-Glutathione on
Tossi, A.; Stellacci, F.; Marson, D.; Boccardo, S.; Pricl, S.; Posocco, P. Gold and Structural Changes During Self-Assembly: An in Situ ATR-
Gold Nanoparticles with Patterned Surface Monolayers for Nano- IR and QCM Study. Phys. Chem. Chem. Phys. 2006, 8, 513−520.
medicine: Current Perspectives. Eur. Biophys. J. 2017, 46, 749−771. (31) Lobo Maza, F.; Méndez De Leo, L.; Rubert, A. A.; Carro, P.;
(13) Hong, R.; Han, G.; Fernández, J. M.; Kim, B.-j.; Forbes, N. S.; Salvarezza, R. C.; Vericat, C. New Insight into the Interface Chemistry
Rotello, V. M. Glutathione-Mediated Delivery and Release Using and Stability of Glutathione Self-Assembled Monolayers on Au(111).
Monolayer Protected Nanoparticle Carriers. J. Am. Chem. Soc. 2006, J. Phys. Chem. C 2016, 120, 14597−14607.
128, 1078−1079. (32) Vallée, A.; Humblot, V.; Méthivier, C.; Pradier, C.-M.
(14) Lund, T.; Callaghan, M. F.; Williams, P.; Turmaine, M.; Glutathione Adsorption from UHV to the Liquid Phase at Various
Bachmann, C.; Rademacher, T.; Roitt, I. M.; Bayford, R. The pH on Gold and Subsequent Modification of Protein Interaction. Surf.
Influence of Ligand Organization on the Rate of Uptake of Gold Interface Anal. 2008, 40, 395−399.
Nanoparticles by Colorectal Cancer Cells. Biomaterials 2011, 32, (33) Vallée, A.; Humblot, V.; Méthivier, C.; Pradier, C.-M.
9776−9784. Adsorption of a Tripeptide, GSH, on Au (1 1 1) under UHV
(15) Amarnath, K.; Mathew, N. L.; Nellore, J.; Siddarth, C. R. V.; Conditions; Pm-Rairs and Low T-XPS Characterisation. Surf. Sci.
Kumar, J. Facile Synthesis of Biocompatible Gold Nanoparticles from 2008, 602, 2256−2263.
Vites Vinefera and Its Cellular Internalization against Hbl-100 Cells. (34) Donaldson, P. M.; Hamm, P. Gold Nanoparticle Capping
Cancer Nanotechnol. 2011, 2, 121−132. Layers: Structure, Dynamics, and Surface Enhancement Measured
(16) Sousa, A. A.; Morgan, J. T.; Brown, P. H.; Adams, A.; Using 2D-IR Spectroscopy. Angew. Chem. 2013, 125, 662−666.
Jayasekara, M. S.; Zhang, G.; Ackerson, C. J.; Kruhlak, M. J.; (35) Li, J.; Qian, H.; Chen, H.; Zhao, Z.; Yuan, K.; Chen, G.;
Leapman, R. D. Synthesis, Characterization, and Direct Intracellular Miranda, A.; Guo, X.; Chen, Y.; Zheng, N.; Wong, M. S.; Zheng, J.
Imaging of Ultrasmall and Uniform Glutathione-Coated Gold Two Distinctive Energy Migration Pathways of Monolayer Molecules
Nanoparticles. Small 2012, 8, 2277−2286. on Metal Nanoparticle Surfaces. Nat. Commun. 2016, 7, 10749.
(17) Lyon, R. P.; Atkins, W. M. Self-Assembly and Gelation of (36) Bian, H.; Li, J.; Chen, H.; Yuan, K.; Wen, X.; Li, Y.; Sun, Z.;
Oxidized Glutathione in Organic Solvents. J. Am. Chem. Soc. 2001, Zheng, J. Molecular Conformations and Dynamics on Surfaces of
123, 4408−4413. Gold Nanoparticles Probed with Multiple-Mode Multiple-Dimen-
(18) Matsuura, K.; Matsuyama, H.; Fukuda, T.; Teramoto, T.; sional Infrared Spectroscopy. J. Phys. Chem. C 2012, 116, 7913−7924.
Watanabe, K.; Murasato, K.; Kimizuka, N. Spontaneous Self-Assembly (37) Petti, M. K.; Lomont, J. P.; Maj, M.; Zanni, M. T. Two-
of Nanospheres from Trigonal Conjugate of Glutathione in Water. Dimensional Spectroscopy Is Being Used to Address Core Scientific
Soft Matter 2009, 5, 2463−2470. Questions in Biology and Materials Science. J. Phys. Chem. B 2018,
(19) Lim, I.-I. S.; Mott, D.; Ip, W.; Njoki, P. N.; Pan, Y.; Zhou, S.; 122, 1771−1780.
Zhong, C.-J. Interparticle Interactions in Glutathione Mediated (38) Hamm, P.; Lim, M.; DeGrado, W. F.; Hochstrasser, R. M. The
Assembly of Gold Nanoparticles. Langmuir 2008, 24, 8857−8863. Two-Dimensional IR Nonlinear Spectroscopy of a Cyclic Penta-
(20) Moaseri, E.; Bollinger, J. A.; Changalvaie, B.; Johnson, L.; Peptide in Relation to Its Three-Dimensional Structure. Proc. Natl.
Schroer, J.; Johnston, K. P.; Truskett, T. M. Reversible Self-Assembly Acad. Sci. U. S. A. 1999, 96, 2036−2041.
of Glutathione-Coated Gold Nanoparticle Clusters via pH-Tunable (39) Woutersen, S.; Hamm, P. Structure Determination of
Interactions. Langmuir 2017, 33, 12244−12253. Trialanine in Water Using Polarization Sensitive Two-Dimensional
(21) Mosquera, J.; Henriksen-Lacey, M.; García, I.; Martínez-Calvo, Vibrational Spectroscopy. J. Phys. Chem. B 2000, 104, 11316−11320.
M.; Rodríguez, J.; Mascareñas, J. L.; Liz-Marzán, L. M. Cellular (40) Woutersen, S.; Mu, Y.; Stock, G.; Hamm, P. Subpicosecond
Uptake of Gold Nanoparticles Triggered by Host−Guest Interactions. Conformational Dynamics of Small Peptides Probed by Two-
J. Am. Chem. Soc. 2018, 140, 4469−4472. Dimensional Vibrational Spectroscopy. Proc. Natl. Acad. Sci. U. S. A.
(22) Eibling, M. J.; MacDermaid, C. M.; Qian, Z.; Lanci, C. J.; Park, 2001, 98, 11254−11258.
S.-J.; Saven, J. G. Controlling Association and Separation of Gold (41) Baiz, C. R.; Reppert, M.; Tokmakoff, A. Amide I Two-
Nanoparticles with Computationally Designed Zinc-Coordinating Dimensional Infrared Spectroscopy: Methods for Visualizing the
Proteins. J. Am. Chem. Soc. 2017, 139, 17811−17823. Vibrational Structure of Large Proteins. J. Phys. Chem. A 2013, 117,
(23) Ho, J.-J.; Ghosh, A.; Zhang, T. O.; Zanni, M. T. Heterogeneous 5955−5961.
Amyloid Β-Sheet Polymorphs Identified on Hydrogen Bond (42) Ganim, Z.; Chung, H. S.; Smith, A. W.; DeFlores, L. P.; Jones,
Promoting Surfaces Using 2D SFG Spectroscopy. J. Phys. Chem. A K. C.; Tokmakoff, A. Amide I Two-Dimensional Infrared Spectros-
2018, 122, 1270−1282. copy of Proteins. Acc. Chem. Res. 2008, 41, 432−441.
(24) Kozlowski, R.; Ragupathi, A.; Dyer, R. B. Characterizing the (43) Hamm, P.; Lim, M.; Hochstrasser, R. M. Structure of the
Surface Coverage of Protein−Gold Nanoparticle Bioconjugates. Amide I Band of Peptides Measured by Femtosecond Nonlinear-
Bioconjugate Chem. 2018, 29, 2691−2700. Infrared Spectroscopy. J. Phys. Chem. B 1998, 102, 6123−6138.
(25) Colangelo, E.; Comenge, J.; Paramelle, D.; Volk, M.; Chen, Q.; (44) Wang, L.; Middleton, C. T.; Singh, S.; Reddy, A. S.; Woys, A.
Lévy, R. Characterizing Self-Assembled Monolayers on Gold M.; Strasfeld, D. B.; Marek, P.; Raleigh, D. P.; de Pablo, J. J.; Zanni,
Nanoparticles. Bioconjugate Chem. 2017, 28, 11−22. M. T.; Skinner, J. L. 2DIR Spectroscopy of Human Amylin Fibrils
(26) Ong, Q.; Luo, Z.; Stellacci, F. Characterization of Ligand Shell Reflects Stable Β-Sheet Structure. J. Am. Chem. Soc. 2011, 133,
for Mixed-Ligand Coated Gold Nanoparticles. Acc. Chem. Res. 2017, 16062−16071.
50, 1911−1919. (45) Hahn, S.; Kim, S.-S.; Lee, C.; Cho, M. Characteristic Two-
(27) Smith, A. M.; Johnston, K. A.; Crawford, S. E.; Marbella, L. E.; Dimensional IR Spectroscopic Features of Antiparallel and Parallel Β-
Millstone, J. E. Ligand Density Quantification on Colloidal Inorganic Sheet Polypeptides: Simulation Studies. J. Chem. Phys. 2005, 123,
Nanoparticles. Analyst 2017, 142, 11−29. 084905.
(28) Hepel, M.; Tewksbury, E. Ion-Gating Phenomena of Self- (46) Moran, S. D.; Zanni, M. T. How to Get Insight into Amyloid
Assembling Glutathione Films on Gold Piezoelectrodes. J. Electroanal. Structure and Formation from Infrared Spectroscopy. J. Phys. Chem.
Chem. 2003, 552, 291−305. Lett. 2014, 5, 1984−1993.
(29) Bieri, M.; Bürgi, T. L-Glutathione Chemisorption on Gold and (47) Lomont, J. P.; Rich, K. L.; Maj, M.; Ho, J.-J.; Ostrander, J. S.;
Acid/Base Induced Structural Changes: A PM-IRRAS and Time- Zanni, M. T. Spectroscopic Signature for Stable Β-Amyloid Fibrils
Resolved in Situ ATR-IR Spectroscopic Study. Langmuir 2005, 21, Versus Β-Sheet-Rich Oligomers. J. Phys. Chem. B 2018, 122, 144−
1354−1363. 153.
(48) Evanoff, D. D.; Chumanov, G. Size-Controlled Synthesis of Dimensional Infrared Spectroscopy and Atomic Force Microscopy.
Nanoparticles. 1. “Silver-Only” Aqueous Suspensions Via Hydrogen Sci. Rep. 2017, 7, 41051.
Reduction. J. Phys. Chem. B 2004, 108, 13948−13956. (69) Chuntonov, L.; Haran, G. Maximal Raman Optical Activity in
(49) Qian, W.; Krimm, S. Vibrational Analysis of Glutathione. Hybrid Single Molecule-Plasmonic Nanostructures with Multiple
Biopolymers 1994, 34, 1377−1394. Dipolar Resonances. Nano Lett. 2013, 13, 1285−1290.
(50) Rubtsov, I. V. Relaxation-Assisted Two-Dimensional Infrared (70) Cohn, B.; Engelman, B.; Goldner, A.; Chuntonov, L. Two-
(RA 2DIR) Method: Accessing Distances over 10 Å and Measuring Dimensional Infrared Spectroscopy with Local Plasmonic Fields of a
Bond Connectivity Patterns. Acc. Chem. Res. 2009, 42, 1385−1394. Trimer Gap-Antenna Array. J. Phys. Chem. Lett. 2018, 9, 4596−4601.
(51) Müller-Werkmeister, H. M.; Li, Y.-L.; Lerch, E.-B. W.; Bigourd, (71) Marsh, D. Dichroic Ratios in Polarized Fourier Transform
D.; Bredenbeck, J. Ultrafast Hopping from Band to Band: Assigning Infrared for Nonaxial Symmetry of Beta-Sheet Structures. Biophys. J.
Infrared Spectra Based on Vibrational Energy Transfer. Angew. Chem., 1997, 72, 2710−2718.
Int. Ed. 2013, 52, 6214−6217. (72) Stern, S. T.; Adiseshaiah, P. P.; Crist, R. M. Autophagy and
(52) Cheatum, C. M.; Tokmakoff, A.; Knoester, J. Signatures of Β- Lysosomal Dysfunction as Emerging Mechanisms of Nanomaterial
Sheet Secondary Structures in Linear and Two-Dimensional Infrared Toxicity. Part. Fibre Toxicol. 2012, 9, 20.
Spectroscopy. J. Chem. Phys. 2004, 120, 8201−8215.
(53) Abramavicius, D.; Zhuang, W.; Mukamel, S. Peptide Secondary
Structure Determination by Three-Pulse Coherent Vibrational
Spectroscopies: A Simulation Study. J. Phys. Chem. B 2004, 108,
18034−18045.
(54) Shaw, C. P.; Middleton, D. A.; Volk, M.; Lévy, R. Amyloid-
Derived Peptide Forms Self-Assembled Monolayers on Gold
Nanoparticle with a Curvature-Dependent Β-Sheet Structure. ACS
Nano 2012, 6, 1416−1426.
(55) Kraack, J. P.; Hamm, P. Surface-Sensitive and Surface-Specific
Ultrafast Two-Dimensional Vibrational Spectroscopy. Chem. Rev.
2017, 117, 10623−10664.
(56) Rosenfeld, D. E.; Gengeliczki, Z.; Smith, B. J.; Stack, T.; Fayer,
M. Structural Dynamics of a Catalytic Monolayer Probed by Ultrafast
2D IR Vibrational Echoes. Science 2011, 334, 634−639.
(57) Kwac, K.; Cho, M. Molecular Dynamics Simulation Study of N-
Methylacetamide in Water. II. Two-Dimensional Infrared Pump−
Probe Spectra. J. Chem. Phys. 2003, 119, 2256−2263.
(58) Kuroda, D. G.; Vorobyev, D. Y.; Hochstrasser, R. M. Ultrafast
Relaxation and 2D IR of the Aqueous Trifluorocarboxylate Ion. J.
Chem. Phys. 2010, 132, 044501.
(59) King, J. T.; Baiz, C. R.; Kubarych, K. J. Solvent-Dependent
Spectral Diffusion in a Hydrogen Bonded “Vibrational Aggregate. J.
Phys. Chem. A 2010, 114, 10590−10604.
(60) Kwak, K.; Park, S.; Finkelstein, I. J.; Fayer, M. D. Frequency-
Frequency Correlation Functions and Apodization in Two-Dimen-
sional Infrared Vibrational Echo Spectroscopy: A New Approach. J.
Chem. Phys. 2007, 127, 124503.
(61) Adamczyk, K.; Simpson, N.; Greetham, G. M.; Gumiero, A.;
Walsh, M. A.; Towrie, M.; Parker, A. W.; Hunt, N. T. Ultrafast
Infrared Spectroscopy Reveals Water-Mediated Coherent Dynamics
in an Enzyme Active Site. Chemical Science 2015, 6, 505−516.
(62) Merga, G.; Cass, L. C.; Chipman, D. M.; Meisel, D. Probing
Silver Nanoparticles During Catalytic H2 Evolution. J. Am. Chem. Soc.
2008, 130, 7067−7076.
(63) Greenler, R. G.; Snider, D. R.; Witt, D.; Sorbello, R. S. The
Metal-Surface Selection Rule for Infrared Spectra of Molecules
Adsorbed on Small Metal Particles. Surf. Sci. 1982, 118, 415−428.
(64) Edington, S. C.; Gonzalez, A.; Middendorf, T. R.; Halling, D.
B.; Aldrich, R. W.; Baiz, C. R. Coordination to Lanthanide Ions
Distorts Binding Site Conformation in Calmodulin. Proc. Natl. Acad.
Sci. U. S. A. 2018, 115, E3126−E3134.
(65) Mandal, H. S.; Kraatz, H.-B. Effect of the Surface Curvature on
the Secondary Structure of Peptides Adsorbed on Nanoparticles. J.
Am. Chem. Soc. 2007, 129, 6356−6357.
(66) Farrag, M.; Thämer, M.; Tschurl, M.; Bürgi, T.; Heiz, U.
Preparation and Spectroscopic Properties of Monolayer-Protected
Silver Nanoclusters. J. Phys. Chem. C 2012, 116, 8034−8043.
(67) di Gregorio, M. C.; Ben Moshe, A.; Tirosh, E.; Galantini, L.;
Markovich, G. Chiroptical Study of Plasmon−Molecule Interaction:
The Case of Interaction of Glutathione with Silver Nanocubes. J. Phys.
Chem. C 2015, 119, 17111−17116.
(68) Roeters, S. J.; Iyer, A.; Pletikapić, G.; Kogan, V.; Subramaniam,
V.; Woutersen, S. Evidence for Intramolecular Antiparallel Beta-Sheet
Structure in Alpha-Synuclein Fibrils from a Combination of Two-
Low scaling EOM-CCSD and EOM-MBPT(2) method with natural transition orbitals
The Journal of Chemical Physics 149, 184103 (2018); 10.1063/1.5045340
Nanoscale domains in ionic liquids: A statistical mechanics definition for molecular dynamics studies
The Journal of Chemical Physics 149, 184502 (2018); 10.1063/1.5054999
The activation energy for water reorientation differs between IR pump-probe and NMR measurements
The Journal of Chemical Physics 149, 164504 (2018); 10.1063/1.5050203
THE JOURNAL OF CHEMICAL PHYSICS 149, 184501 (2018)
Solvatochromic shifts of local vibrational probes report on the strength of the surrounding electric
fields and the probe’s hydrogen bonding status. Stretching vibrational mode of the ester carbonyl group
is a popular solvatochromic reporter used in the studies of peptides and proteins. Small molecules,
used to calibrate the response of the vibrational probes, sometimes involve Fermi resonances (FRs)
induced by inter-molecular interactions. In the present work, we focus on the scenario where FR does
not appear in the infrared spectrum of the ester carbonyl stretching mode in aprotic solvents; however,
it is intensified when a hydrogen bond with the reporter is established. When two molecules form
hydrogen bonds to the same carbonyl oxygen atom, FR leads to strong hybridization of the involved
modes and splitting of the absorption peak. Spectral overlap between the Fermi doublets associated
with singly and doubly hydrogen-bonded carbonyl groups significantly complicates quantifying dif-
ferent hydrogen-bonded conformations. We employed a combination of linear and third-order (2DIR)
infrared spectroscopy with chemometrics analysis to reveal the individual line shapes and to estimate
the occupations of the hydrogen-bonded conformations in methyl acetate, a model small molecule.
We identified a hydrogen-bond-induced FR in complexes of methyl acetate with alcohols and water
and found that FR is lifted in larger molecules used for control experiments—cholesteryl stearate
and methyl cyanoacetate. Applying this methodology to analyze acetonitrile-water solutions revealed
that when dissolved in neat water, methyl acetate occupies a single hydrogen-bonding conformation,
which is in contrast to the conclusions of previous studies. Our approach can be generally used when
FRs prevent direct quantification of the hydrogen bonding status of the vibrational probe. Published
by AIP Publishing. https://doi.org/10.1063/1.5055041
solvent corresponding to the 0hb conformation.25 In methanol, context of the spectroscopy of small molecules.34,36 Recent
this transition does not shift appreciably; however, two addi- examples include the work of Kwon et al.,37 who observed
tional peaks appear at 1729.9 cm−1 and 1705.0 cm−1 .16 In that molecule–ion interactions can significantly enhance the
water (D2 O), only two peaks are observed at 1726.0 cm−1 and strength of FR. Rodgers et al.39 noted that the presence of FR
1704.2 cm−1 .20 could limit the ability of the vibrational probe to report on
Based on the extensive experimental studies, comple- the H-bonded conformation. Earlier studies by Wassermann
mented by theoretical analysis, Banno et al.16 assigned these et al.40 and Greve et al.35 on the N−−H stretching modes in
lower-frequency transitions to the 1hb and 2hb conforma- pyrazole and aniline molecules, respectively, have shown that
tions of MA, which can coexist simultaneously in protic the H-bond can dramatically enhance the coupling strength
solvents, especially in alcohols, and can interconvert on the between states participating in FR. In situations where differ-
picosecond time scale. Later, this assignment was supported ent H-bonded conformations possess overlapping FR doublets
by the studies of Righini and co-workers.19,20 However, despite that are not observed in the free (non-bonded) state, the iden-
the apparent agreement between the experimentally observed tification and quantification of the H-bonded configurations
shifts of C==O transitions in different H-bonded conforma- using vibrational probes become especially challenging.
tions and their interpretation, based on the quantum chemistry In the present work, we revealed the presence of FR in
calculations,19,20 these studies suggest that two very differ- the ester C==O transition of MA using infrared absorption
ent solvation states, 1hb and 2hb, can be simultaneously measurements in solutions with increasing concentrations of
populated by a small and highly hydrophilic molecule such the H-bond-donating molecules. The paper is organized as
as MA in the bulk of aqueous solution; this might appear follows: First, the corresponding data for para-chlorophenol
counterintuitive. (p-CP) as the H-bond donor in the carbon tetrachloride sol-
MA can be readily compared with another extensively vent are presented. This substituted phenol is the most acidic
studied small molecule, N-Methylacetamide (NMA),26–28 homologue from the series of homologues that we investi-
which is the smallest molecule involving the amide group, a gated, including para-tert-butylphenol, para-cresol, phenol,
building block of peptides and proteins.29 It is well known that and para-cyanophenol.41 Because its H-bond formation con-
in diluted aqueous solutions, NMA has a single broad transi- stant is the highest among the homologues, p-CP is the most
tion of its amide I vibrational mode that reflects a nearly con- appropriate choice for the scope of the present discussion. A
tinuous distribution of multiple H-bonded configurations.30–32 series of the infrared absorption spectra are analyzed by sin-
The asymmetry of the C==O line shape of NMA in water gular value decomposition (SVD) to identify the number of
was initially attributed to Fermi resonances (FRs).33 Two- possible H-bonded conformations present in solution, as well
dimensional vibrational spectroscopy (2DIR) is well known as by self-similar modeling42 and multivariate curve resolu-
for its ability to reveal FRs because of its high sensitivity tion (MCR) analysis43–45 to reveal their individual line shapes
to the inter-mode coupling.34–37 However, there is no clear and the corresponding contribution to the signal in each sam-
indication of the presence of FRs in the 2DIR spectrum of ple of the series. In order to support the assignment of the
NMA.27,31 The ultrafast chemical exchange between different lower-frequency C==O peaks to the FRs, we conducted 2DIR
H-bonded conformations was first demonstrated using 2DIR spectroscopy, where the coupled vibrational modes feature the
spectroscopy in NMA–methanol complexes.26 Stimulated by corresponding cross-peaks.
the opportunity provided by this experimental advance, in In order to exclude the possibility that the observed FR
recent 2DIR experiments and molecular dynamics simulation arises from the interaction with vibrational modes specific
studies, researchers focused on elucidating the rates of inter- to phenols, we conducted control experiments with a differ-
conversion between different H-bonded conformations of MA ent H-bond-donating molecule, trifluoroethanol (TFE). The
in aqueous and methanol solutions.16–20 individual infrared spectra obtained for the 1hb and 2hb con-
Because of an apparent similarity in the molecular struc- formations allowed us to extract an unperturbed shift of the
tures of NMA and MA and of their relatively high solubility in C==O transition in MA upon H-bond formation, invoking the
water,38 the existence of the 1hb and 2hb conformers of MA model of Refs. 39 and 46. We compared the obtained shifts
in the diluted aqueous solution requires additional experimen- to those of C==O in the cholesteryl stearate (CS), which is a
tal verification. In the present work, we address this aspect convenient choice of molecule for the additional control exper-
via a detailed study of the H-bonded conformations between iments: In CS, which is a larger molecule than MA, the FRs
the ester C==O group of MA and several H-bond-donating are lifted; CS is well soluble in carbon tetrachloride, which is a
molecules, including a series of substituted phenols, trifluo- convenient solvent for studies of the corresponding complexes
roethanol, and water (D2 O). We conducted a series of infrared with alcohols.
absorption and 2DIR experiments and found that two low- Finally, we apply the combination of FTIR, MCR, and
frequency peaks observed in the infrared spectrum of MA in 2DIR methods to analyze hydrogen-bonded conformations of
the region of the C==O transition involve FRs, associated with MA in acetonitrile-water mixtures with gradually changing
both 1hb and 2hb conformations; this effect is particularly water volume fractions ranging from 0 to 1.47–49 Interestingly,
strong in the latter case. we found that in neat water, MA occupies only a single H-
It is well known that accidental degeneracies between the bonding conformation, where carbonyl oxygen of the ester
infrared-active vibrational mode and a dark overtone or a com- group in MA is bound to two hydroxyl groups of water (2hb).
bination band of a different mode can lead to FR. Such occur- Similar results are obtained in a series of the control exper-
rences are extensively discussed in the literature, mainly in the iments with the homologue molecule methyl cyanoacetate
184501-3 Ghosh et al. J. Chem. Phys. 149, 184501 (2018)
Methyl acetate–p-CP
0hb 1748.5
1hb 1727.5 1709.9 3.1 7.6 9.0 1723.2 1714.2
2hb 1723.8 1705.7 1.5 8.9 3.6 1716.6 1712.9
Methyl acetate–TFE
0hb 1748.2
1hb 1730.9 1705.0 5.0 9.7 17.3 1726.6 1709.3
2hb 1725.9 1709.0 1.1 8.4 0.8 1717.9 1717.0
Cholesteryl stearate–p-CP
0hb 1732.5
1hb 1709.8
2hb 1696.1
Methyl acetate–acetonitrile-D2 O
0hb 1741.9
1hb 1729.9 1707.1 5.0 8.5 15.2 1726.1 1710.9
2hb 1726.0 1704.2 0.79 10.8 2.6 1716.4 1713.8
Methyl cyanoacetate–acetonitrile-D2 O
0hb 1755.7
1hb 1751.6
2hb 1742.1
a Transition frequency, as obtained from fitting the individual line shapes of the H-bonded molecules.
b Ratio between the areas of the transitions associated with the FR in H-bonded molecules as obtained from fitting the individual
line shapes.
c The strength of coupling between the modes involved in the FR, as obtained from fitting to the model of Ref. 46.
d Frequency difference between the transitions associated with the unperturbed modes involved in the FR, as obtained from fitting
The contribution of the individual spectral component 160 ± 15 km/mol, which is in a good agreement with the
(i.e., of the line shape of the particular H-bonded conforma- known values.52,53 The relations between the transition dipole
tion) to the experimental spectrum obtained by MCR analysis moments of the H-bonded conformers are in agreement with
involves a product of its concentration in the sample and the that of Candelaresi et al.’s theoretical analysis of the electronic
excitation probability, which scales with µ2 , where µ is the structure of the H-bonded MA.20
transition dipole moment of the C==O stretching mode of the In order to check the results of Fig. 1(b) for consistency,
H-bonded complex. Transition dipole moments are expected to we fitted the data to the binding isotherm. As evident from
differ with the H-bonded conformation.20 The concentrations Fig. 1(c), with up to ca. 0.1M concentration of p-CP, only
of the H-bonded complexes in our samples were obtained by the 1hb complex with MA is formed. The corresponding data
the following method: First, the extinction coefficient of the were fitted to a 1:1 binding isotherm Y11 = 1+K K1 X
, where Y
1X
0hb MA solution in carbon tetrachloride (without p-CP) was is the fraction of the H-bonded MA, X is the concentration
measured using Beer’s law. Second, the concentration of the of the free p-CP molecules in solution, and K 1 is the equilib-
0hb molecules and, consequently, that of the 1hb molecules rium constant of the 1hb complex formation. At higher p-CP
were determined in a series of samples with low concentrations concentrations, the 2hb complex is also formed and the cor-
of p-CP, where only two singular components were present. responding data are fitted to a 1:2 binding isotherm given by
Once the concentrations of the 1hb species are known, the K1 X+K1 K2 X 2
corresponding extinction coefficient is obtained and used to Y12 = 1+K 2 , where K 2 is the stepwise formation equi-
1 X+K1 K2 X
determine the concentrations and the extinction coefficient librium constant of the 2hb complex.54 The values of K 1 = 17.1
of the 2hb species in samples with higher concentrations of l/mol and K 2 = 4.3 l/mol were obtained, which correspond to a
p-CP. Our results suggest the following relations between the non-cooperative binding process with α = 4K 2 /K 1 ≈ 1, where
transition dipole moments: µ0hb /µ1hb = 0.85 and µ0hb /µ2hb α is the interaction parameter,55,56 suggesting that a stacking
= 0.75, where the value of µ0hb = 0.3 D was estimated interaction between the phenol rings does not affect the H-
following Ref. 51. The integrated intensity associated with bonding equilibrium. The corresponding results are shown in
the 0hb transition of MA measured in our experiments is Fig. 1(d).
184501-5 Ghosh et al. J. Chem. Phys. 149, 184501 (2018)
FIG. 8. 2DIR spectroscopy of H-bonded complexes of MA and MCA with water. [(a) and (b)] 2DIR spectra of MA and [(d) and (e)] that of MCA in pure D2 O.
Waiting times are indicated. (c) The waiting time dependence of the 2DIR peak amplitudes of MA. The lines show the bi-exponential and the rise-and-decay fits.
Diagonal peaks: green—{1730; 1743}, τ 1 = 0.3 ps (0.7), τ 2 = 1.7 ps (0.3); black—{1705; 1710}, τ 1 = 0.2 ps (0.6), τ 2 = 1.4 ps (0.4). Cross-peaks: blue—{1707;
1743}, τ 1 = 0.2 ps (0.6), τ 2 = 1.8 ps (0.4); red—{1730; 1709}, τ rise = 2.4 ps, τ decay = 1.8 ps. (f) The waiting time evolution of the nodal line slope in 2DIR
spectra of MCA. The red line shows fitting to an exponential decay, τ = 0.86 ps.
of 2.4 ps and a decay time of 1.8 ps, whereas the ampli- is observed in neat water, as seen in Fig. 7(f). The 2DIR
tude of the second cross-peak {ωτ = 1707 cm−1 ; ωt = 1743 spectrum of MCA in water [see Figs. 8(d) and 8(e)] also
cm−1 } does not rise, but decays bi-exponentially, with a fast has a single peak that undergoes spectral diffusion with a
time constant of 0.2 ps and a similar slow time constant of characteristic time of 0.86 ps, as obtained from the analy-
1.7 ps. In addition, we noted that the amplitudes of both diag- sis of its nodal line slope in Fig. 8(f). The latter reflects the
onal peaks decay with the fast time constants of 0.2-0.3 ps loss of correlation between the fluctuating vibrational frequen-
and slow time constants of 1.4-1.7 ps. Bi-exponential decay cies,62,63 which in aqueous solutions is broadly attributed to
of the C==O stretching in MA was previously observed in the reconfiguration of H-bonds.64–66
both aprotic and protic solvents, with the fast component
attributed to the equilibration of the vibrational excitation
E. Analysis of Fermi resonance in MA
between the coupled modes, whereas the slower one to popula-
tion relaxation.18,25,41 The evolution of the cross-peak ampli- The interaction between the ester C==O modes and vari-
tudes, which is different from what is expected in the case ous combination bands in aromatic acetate molecules is well
of chemical exchange, supports the assignment of the origin known;67,68 however, the aliphatic acetates were generally
of the observed cross-peaks in FR. Note also the qualitative considered not to exhibit such an effect.69,70 Although the
similarity between the 2DIR spectrum in Figs. 8(a) and 8(b) occurrence of FR in the C==O stretching mode of H-bonded
and the 2DIR spectrum of FR in benzoylchloride, discussed in MA molecules has not been discussed before, it is not very
Ref. 34. surprising since small molecules are generally well known to
In order to alter the resonance conditions between the be highly susceptible to this phenomenon. Careful inspection
modes involved in FRs also under conditions of aqueous solu- of the 0hb spectrum of MA reveals the presence of a very weak
tions, we focused on a substituted homologue of MA–methyl peak at 1710 cm−1 , which can be possibly assigned to the over-
cyanoacetate (MCA). The corresponding infrared absorption tone of the skeleton deformation mode at 850 cm−1 , which can
spectra are shown in Fig. 7(d). Analogous to the case of MA, couple to the C==O stretching.21
we found that MCA has three distinct H-bonded conform- In order to obtain more details on the coupling between
ers in acetonitrile-water mixtures, which can be distinguished the modes, we used the modeling of Refs. 39 and 46, where
experimentally and assigned to 0-2hb C==O groups (MCR the coupling to a mode with zero transition dipole moment
R2 = 0.9999). However, as shown in Fig. 7(e), each con- is assumed. The individual line shapes of the 1hb and 2hb
former has a single C==O transition with the corresponding conformers possessing FRs from Figs. 1, 3, and 7 were fit-
frequencies of 1755.7 cm−1 , 1751.6 cm−1 , and 1742.1 cm−1 ted to pairs of Lorentzian profiles with central frequencies
for the 0hb, 1hb, and 2hb complexes, respectively. In MCA, ω1 and ω2 and peak areas I 1 and I 2 . Based on the parame-
the C==O transition frequencies are shifted to the blue, as ters obtained from fits, we calculated the frequency difference
compared to those of MA and CS, and the corresponding between the coupled modes, ∆ = ω1 − ω2 , as well as the ratios
spectra indicate that the FR is removed by the substitution. of the peak intensities R = I 1 /I 2 , as summarized in Table I.
Analysis of the component fractions revealed behavior sim- The strength of the inter-mode coupling W and the energy
ilar to MA, indicating that only a single 2hb conformation difference between the unperturbed modes ∆0 were extracted
184501-9 Ghosh et al. J. Chem. Phys. 149, 184501 (2018)
IV. CONCLUSIONS
An infrared study of H-bonded complexes of MA indi-
cates that the ester C==O stretching vibrational mode is
involved in FR, where the strength of the coupling with the dark
mode strongly depends on the H-bonded configuration. The
formation of H-bonds with the C==O group of MA generally
red shifts the frequency of the C==O transition, can blue shift
the frequency of the dark mode, and enhances the coupling
strength between them; all together, it dramatically increases
the degree of hybridization between the involved modes. This
FIG. 9. Infrared absorption of H-bonded MA and MCA in the low-frequency effect is much stronger for 2hb conformers than for 1hb con-
range. Blue lines—free MA molecules in carbon tetrachloride solution [panels
formers. Following this finding, we have demonstrated that
(a) and (b)], in acetonitrile [panel (c)], and free MCA molecules [panel (d)].
The concentration of MA and MCA is 0.1M in all samples. Red line—MA the two transitions in the region of C==O stretching observed
H-bonded molecules to p-PC (concentration 1.5M) in panel (a), TFE (con- in aqueous solutions of MA are associated with the H-bond-
centration 1M) in panel (b), D2 O (neat) in panel (c), and MCA H-bonded induced FR that appears in a single H-bonded conformer. The
to D2 O (neat) in panel (d). Black lines show absorption in samples with the
H-bond-donating molecules only. The magenta line in panel (b) shows the
emergence of FR induced by intermolecular interactions has
spectrum of the CS molecules (concentration 0.1M) lacking any transitions in been previously reported in cases of N−−H and C−−N stretch-
this spectral region. ing; it appears to be a general phenomenon. This effect is likely
184501-10 Ghosh et al. J. Chem. Phys. 149, 184501 (2018)
to characterize also other small molecules, which are gener- 13 S. D. Fried, S. Bagchi, and S. G. Boxer, Science 346, 1510 (2014).
14 S. M. Kashid and S. Bagchi, J. Phys. Chem. Lett. 5, 3211 (2014).
ally known to be prone to FR, whereas this is not observed, for 15 T. Haldar, S. M. Kashid, P. Deb, S. Kesh, and S. Bagchi, J. Phys. Chem.
example, in the ester C==O stretching in larger CS and MCA Lett. 7, 2456 (2016).
molecules studied in this work. The identity of the H-bond 16 M. Banno, K. Ohta, and K. Tominaga, J. Raman Spectrosc. 39, 1531
donating molecules used in the present study (p-CP, TFE, and (2008).
17 M. Banno, K. Ohta, S. Yamaguchi, S. Hirai, and K. Tominaga, Acc. Chem.
D2 O) was found not to affect the general phenomena of the
Res. 42, 1259 (2009).
emergence of FR in MA. In the view of the present results, the 18 L. Chuntonov, I. M. Pazos, J. Ma, and F. Gai, J. Phys. Chem. B 119, 4512
assignment of the C==O transitions in small ester molecules to (2014).
different H-bonded conformations should be taken with cau- 19 M. Pagliai, F. Muniz-Miranda, G. Cardini, R. Righini, and V. Schettino,
tion. Interestingly, the 2DIR spectra of the MA complex with J. Phys. Chem. Lett. 1, 2951 (2010).
20 M. Candelaresi, M. Pagliai, M. Lima, and R. Righini, J. Phys. Chem. A 113,
cyanophenol reported by us earlier 41 did not show a clear cross- 12783 (2009).
peak pattern characteristic to the FR, and the cross-peaks’ 21 J. Wilmshurst, J. Mol. Spectrosc. 1, 201 (1957).
growth that was attributed to the chemical exchange among 22 J. Dybal and S. Krimm, J. Mol. Struct. 189, 383 (1988).
23 M. L. Senent, R. Domı́nguez-Gómez, M. Carvajal, and I. Kleiner, J. Chem.
the H-bonded conformations was relatively slow. However,
Phys. 138, 044319 (2013).
our present findings suggest that it is necessary to consider the 24 V. V. Volkov, R. Chelli, W. Zhuang, F. Nuti, Y. Takaoka, A. M. Papini,
anharmonic interactions in great detail. Substituted phenols S. Mukamel, and R. Righini, Proc. Natl. Acad. Sci. U. S. A. 104, 15323
provide a convenient platform to study the H-bond-induced (2007).
25 M. Lim and R. M. Hochstrasser, J. Chem. Phys. 115, 7629 (2001).
FR, where the anharmonic interactions induced by H-bonding 26 S. Woutersen, Y. Mu, G. Stock, and P. Hamm, Chem. Phys. 266, 137
can be systematically varied via the alternation of the substi-
(2001).
tution group on the phenol ring. We are currently investigating 27 M. F. DeCamp, L. DeFlores, J. M. McCracken, A. Tokmakoff, K. Kwac,
the role of the strength of the H-bonding interaction on the and M. Cho, J. Phys. Chem. B 109, 11016 (2005).
28 M. T. Zanni, M. C. Asplund, and R. M. Hochstrasser, J. Chem. Phys. 114,
anharmonic coupling in the complexes of substituted phenols
4579 (2001).
with different ester molecules. 29 A. Ghosh and R. M. Hochstrasser, Chem. Phys. 390, 1 (2011).
In the context of solvatochromic vibrational probes of 30 E. Salamatova, A. V. Cunha, R. Bloem, S. J. Roeters, S. Woutersen, T. L.
(bio-)molecular environment, where the probe’s response is C. Jansen, and M. S. Pshenichnikov, J. Phys. Chem. A 122, 2468 (2018).
31 S. Woutersen, R. Pfister, P. Hamm, Y. Mu, D. S. Kosov, and G. Stock,
assessed with the help of small molecules, it is very impor-
J. Chem. Phys. 117, 6833 (2002).
tant to be able to recognize FRs. Because of its accidental 32 S. Ham, J.-H. Kim, H. Lee, and M. Cho, J. Chem. Phys. 118, 3491 (2003).
nature, FR induced by intermolecular interactions may com- 33 P. Hamm, M. Lim, and R. M. Hochstrasser, J. Phys. Chem. B 102, 6123
plicate the development of a general approach for describing (1998).
34 J. Edler and P. Hamm, J. Chem. Phys. 119, 2709 (2003).
molecular vibrational solvatochromism. Here, the unperturbed 35 C. Greve, E. T. J. Nibbering, and H. Fidder, J. Phys. Chem. B 117, 15843
vibrational transitions could be considered in constructing the
(2013).
frequency-field maps for the vibrational probe. The system- 36 M. J. Tucker, Y. S. Kim, and R. M. Hochstrasser, Chem. Phys. Lett. 470,
atic approach presented in this work to identify FRs involves 80 (2009).
37 Y. Kwon, C. Lee, and S. Park, Chem. Phys. 445, 38 (2014).
linear infrared spectroscopy as well as 2DIR and MCR anal- 38 P. Howard and W. Meylan, Handbook of Physical Properties of Organic
ysis. We believe that this approach is general and therefore Chemicals (Lewis Publishers, Boca Raton, FL, 1997).
can be broadly used to quantify the occupations of different 39 J. M. Rodgers, R. M. Abaskharon, B. Ding, J. Chen, W. Zhang, and F. Gai,
dynamics in amyloid fibrils, J. Phys. Chem. B (to be published). 57 P. Hamm and M. T. Zanni, Concepts and Methods of 2D Infrared
12 S. D. Fried, S. Bagchi, and S. G. Boxer, J. Am. Chem. Soc. 135, 11181 Spectroscopy (Cambridge University Press, Cambridge, 2011).
(2013). 58 L. Chuntonov and J. Ma, J. Phys. Chem. B 117, 13631 (2013).
184501-11 Ghosh et al. J. Chem. Phys. 149, 184501 (2018)
59 Y. S. Kim and R. M. Hochstrasser, Proc. Natl. Acad. Sci. U. S. A. 102, 65 R. Kumar, J. R. Schmidt, and J. L. Skinner, J. Chem. Phys. 126, 204107
11185 (2005). (2007).
60 J. Zheng, K. Kwak, J. Asbury, X. Chen, I. R. Piletic, and M. D. Fayer, 66 S. Park and M. D. Fayer, Proc. Natl. Acad. Sci. U. S. A. 104, 16731
DOI: 10.1039/c6cc05947c
www.rsc.org/chemcomm
This journal is © The Royal Society of Chemistry 2016 Chem. Commun., 2016, 52, 11159--11162 | 11159
View Article Online
Communication ChemComm
The disodium salt of 5 0 -GMP was irradiated under micro- with intensification of a broad peak centered at 340 nm,
wave conditions at 160 1C for 5 minutes. The resulting solution suggesting an increase in electron delocalization efficiency
exhibited intense fluorescence (Scheme 1) upon exposure to (Fig. 1c). Like any other CDs, G-dots also exhibited the char-
UV-light thereby confirming the transformation of molecular acteristic excitation wavelength dependent fluorescence emis-
5 0 -GMP to 5 0 -GMP nano dots (G-dots). It should be pointed out sion. Upon increasing the excitation wavelength from 310 nm
that 5 0 -GMP does not exhibit any fluorescence. The condition of to 420 nm, the photoluminescence peaks of G-dots shifted from
Published on 25 August 2016. Downloaded by Technion - Israel Institute of Technology on 27/02/2017 10:08:09.
irradiation was important because exposure to lower tempera- 447 nm to 462 nm (Fig. 1d). At an excitation wavelength of
ture or lesser time led to incomplete conversion, while exposure 360 nm, the photoluminescence quantum yield (QY) of G-dots
to higher temperatures for prolonged time led to precipitation. is determined to be 32.5%. In addition to PL spectra, G-dots
The morphology of the G-dots was confirmed by high resolution- exhibited up-conversion luminescence with their maxima in
transmission electron microscopy (HR-TEM) and atomic force the region of 430 nm to 462 nm, when excited in the higher
microscopy (AFM) (Fig. 1a and b). HR-TEM images showed wavelength region (600 nm to 825 nm) (Fig. 1e). Surprisingly,
spherical particles with an average diameter less than 5 nm time resolved photoluminescence spectra (lex = 375 nm) of the
without any graphitic lattices (Fig. 1a and Fig. S1, ESI†). The 5 0 -GMP dots monitored at different emission decayed differently
absence of the lattice pattern suggests them to be highly for different monitoring wavelengths and the average decay
amorphous. The AFM height profile was in well agreement with constant decreased with an increase of monitoring wavelength
the HR-TEM analysis (Fig. 1b). (Fig. 1f and Table S1, ESI†), which indicate the variation in
UV/Vis spectra and PL studies were carried out to character- shapes, sizes and band gaps of G-dots. We have observed that
ize the optical properties of G-dots. The UV/Vis spectrum of the fluorescence of G-dots was quenched in acidic pH and
5 0 -GMP exhibited two well defined absorption peaks at 219 nm remain unaffected at basic pH (Fig. S2, ESI†).
and 293 nm and a weak shoulder at 345 nm regions, where the The PXRD patterns of the G-dots (Fig. S3, ESI†) displayed a
absorption peak at 219 nm and 293 nm are attributed to the broad peak centered at 27.41 (d = 0.33 nm), which could be
p - p* transition of the guanine ring and the n - p* transition related to the G-quadruplex type stacking between two adjacent
of the carbonyl group (Fig. 1c). The weak shoulder at 345 nm is vertical G–G stacks. The peaks in the lower Bragg’s angle
attributed to the weak propensity of the overall p - p* regime were nearly absent for the G-dot, thus suggesting a
conjugation of the 5 0 -GMP quartet helix formed in the solution. spherical symmetry of the resulting G-dot. For further surface
Upon exposure to high energy microwave irradiation, the structural elaboration, Fourier transform infrared (FTIR) spectros-
absorption features at 219 nm and 293 nm diminished along copy of the G-dots was performed and compared with that of
5 0 -GMP (Fig. S4, ESI†). The peaks at 1693, 1524 and 1484 cm 1
are attributed to the amide I (CQO), amide II (N–H) and amide
III (C–N) stretching frequency. The broad-bands centered at
3423 and 3313 cm 1 are assigned to the O–H (Ws) and N–H
(Ws) stretching frequency. The sharp peak at 1086 cm 1 corre-
sponds to the PQO stretching frequency and that at 1365 cm 1
corresponds to C–H bending vibration. The phosphate group
renders the colloidal stability of the G-dots, with a measured
zeta potential value of 19.4 mV (Fig. S5, ESI†).
Surprisingly, the G-dots did not show the characteristic
graphitic D band and G band in the RAMAN spectra (Fig. S6,
ESI†), suggesting the presence of a highly disordered (amorphous)
state of carbon in the core corroborating our observation from
HR-TEM. We propose that the presence of a graphitic layer is not
a necessary condition to form G-dots. We believe that the inherent
aromatic nature of the guanine nucleotide is sufficient to exhibit
CDs like the photoluminescence property. The circular dichroism
spectrum of G-dots showed intense positive peaks at 218, 270 nm
and a shoulder at B300 nm. It also showed negative peaks
at B230 and B260 nm, suggesting the presence of continuous
self-assemblies of a 50 -GMP like structure (Fig. S7, ESI†). Moreover,
the XPS spectra of carbon (1s), oxygen (2p), nitrogen (2p), and
Fig. 1 HR-TEM (a) and AFM (b) image of G-dots. (c) Absorption spectra phosphorous (p,s) (Fig. S8, ESI†) also showed peaks corres-
of G-dots, (d) fluorescence spectra of G-dots at different excitation ponding to 5 0 -GMP.
wavelengths (lex = 310 to 420 nm with increment of 10 nm) (inset-
To understand the structure of G-dots, the 1H NMR spectra
normalised one). (e) UP-conversion spectra of G-dots (lex = 600 to
850 nm with increment of 25 nm) (inset-normalised one). (f) Fluorescence
of the G-dots were recorded in both D2O (Fig. 2a) and DMSO-d6
decay at different monitoring wavelengths for a particular excitation (Fig. S9, ESI†), and the 31P NMR spectrum was recorded in D2O
wavelength (lex = 375 nm). (Fig. 2b). Interestingly, all the protons were intact in the G-dots.
11160 | Chem. Commun., 2016, 52, 11159--11162 This journal is © The Royal Society of Chemistry 2016
View Article Online
ChemComm Communication
Published on 25 August 2016. Downloaded by Technion - Israel Institute of Technology on 27/02/2017 10:08:09.
Fig. 2 (a) 1H NMR spectra of 5 0 -GMP and (b) G-dots. (c) MALDI-TOF Fig. 3 (a) Solution after microwave treatment of 5 0 -GMP. (b) Inverted
spectra of G-dots. image showing the formation of the gel. (c) Fluorescence image of the
hydrogel, when excited at 365 nm. (d) HR-TEM image of the G-dot
hydrogel. (e) G 0 and G00 versus angular frequency at a constant strain
(0.01%) for the G-dot gel and G-dot gel + K+ hydrogel.
However, in comparison to 5 0 -GMP,16 all the proton peaks for
the G-dots gave sharp lines and up-fielded to some extent,
suggesting that G-dots contain highly symmetrical species in an self-assemble to form a G-quartet based short range columnar
electron-rich environment. The NMR spectra thus indicate a self-assembled structure, where sodium stabilizes the stacking
complete conversion of 5 0 -GMP to G-dots. In the spectra of quartets to some extent. It is water soluble and forms
recorded in D2O, the methylene protons attached to the 5 0 hydrogels at acidic pH or in the presence of selective cations
carbon on the sugar ring appeared at 3.65 and 3.73 as two such as K+, Ag+, etc. It also forms binary hydrogels with addition
doublets of doublets suggesting that the methylene protons are of guanosine derived insoluble components.17 It is interesting
in a structurally rigid environment (Fig. 2). to observe that the G-dots are sparingly soluble in water. We
In the phosphorous (31P) NMR spectra of 5 0 -GMP, a single envisaged that a high concentration of G-dots might form
peak was observed at 4.39 corresponding to the ortho- hydrogels, where the presence of a small number of guanine
phosphate monoester, whereas 31P NMR spectra of G-dots showed motifs of 5 0 -GMP would facilitate the self-assembly of the
a peak at 3.17, suggesting the formation of orthophosphate G-dots with the help of sodium to form extended structures.
diester bonds (Fig. 2b). A similar comparison for 1H NMR spectra Accordingly, microwave irradiation of a 140 mM 50 -GMP in Milli-Q
recorded in DMSO-d6 suggested that all the protons were intact water (1.5 mL) resulted in a transparent solution, which formed
and were slightly up-fielded. A new peak was observed at 5.06, an opaque hydrogel after cooling (Fig. 3a–c). TEM analysis of the
which was attributed to the –OH group in the phosphate. dispersed hydrogel showed a network of fibrillar structures of
Matrix assisted laser desorption ionization-time of flight lengths extending to several micrometres (Fig. 3d). The fibrils
(MALDI-TOF) confirms the existence of a dimer, a tetramer were possibly formed due to the super-assembly of G-dots, which
and other higher order oligomeric structures of 5 0 -GMP (Fig. 2c induces heterogeneity and renders stability to the gel. The CD
and Fig. S10, S11, ESI†). spectrum of the G-dot derived hydrogel showed features similar
All the above observations can be collectively put-together to to the G-dots (Fig. S7, ESI†) and suggested that the hydrogel
elaborate the structure of G-dots. We believe that the spontaneous retains the chirality.
self-assembly of guanosine drives the formation of G-dots. Moreover, To investigate the visco-elastic behavior of the covalently
stacking of guanosine quartets allows self-assembly and the quartets cross-linked G-dot hydrogels, oscillatory rheological measure-
are further cross-linked by the orthophosphate diester bonds. ments were performed in frequency sweep modes at room
The other oxygen (O ) of the phosphate gets protonated and is temperature. The storage modulus (G) was higher than the loss
neutralized to reduce repulsion between adjacent stacking. modulus (G00 ) and they did not intersect throughout the applied
Continued polymerization of 5 0 -GMP through formation of frequency range, suggesting the elastic nature of the hydrogel
phosphodiester bonds produces guanosine-polymers of variable (Fig. 3e). Importantly, the hydrogels retain the structural
length, which contribute to the rigidity of the G-dots. Moreover, similarity and elasticity even after terminating the shear-stress.
we believe that the presence of a wide range of polymeric length The thermal properties of the G-dot hydrogel examined using
of guanosine forms surface traps to contribute to the excitation differential scanning calorimetry (DSC) gave a gel–sol transition
dependent emission property of the G-dots. temperature of 58 1C (Fig. S12, ESI†). The potassium cation
Guanosine derivatives are known to form supramolecular is known to drive the G-quartet assembly.1–3 We have then
hydrogels, which have found applications in tissue engineering, examined the effect of K+ ions on the G-dot self-assembly to
birefringent materials and sensors.2 5 0 -GMP is known to form stable hydrogels. The G-dot gel prepared in the presence
This journal is © The Royal Society of Chemistry 2016 Chem. Commun., 2016, 52, 11159--11162 | 11161
View Article Online
Communication ChemComm
of K+ was found to exhibit a fibrillar structure with thinner 7 (a) W. Shi, X. Li and H. Ma, Angew. Chem., Int. Ed., 2012, 51, 6432;
diameter (Fig. S1b, ESI†) with a higher thermal stability (gel–sol (b) S. Zhu, Q. Meng, L. Wang, J. Zhang, Y. Song, H. Jin, K. Zhang,
H. Sun, H. Wang and B. Yang, Angew. Chem., Int. Ed., 2013,
transition temperature of 70 1C) and higher elasticity compared 125, 4045; (c) M. Lin, H. Y. Zou, T. Yang, Z. X. Liu, H. Liu and
to the G-dot gel (Fig. 3). All the above observations suggest that C. Z. Huang, Nanoscale, 2016, 8, 2999; (d) M. Lan, Y. Di, X. Zhu,
the self-assembly drives the formation of G-dot hydrogels. T.-W. Ng, J. Xia, W. Liu, X. Meng, P. Wang, C.-S. Lee and W. Zhang,
Chem. Commun., 2015, 51, 15574.
In summary, we have demonstrated the unprecedented 8 (a) J. Kim, J. Park, H. Kim, K. Singha and W. J. Kim, Biomaterials,
Published on 25 August 2016. Downloaded by Technion - Israel Institute of Technology on 27/02/2017 10:08:09.
formation of highly photoluminescent G-dots from the molecular 2013, 34, 7168; (b) M. Zheng, S. Liu, J. Li, D. Qu, H. Zhao, X. Guan,
precursor 5 0 -guanosine monophosphate. The G-dots consist of X. Hu, Z. Xie, X. Jing and Z. Sun, Adv. Mater., 2014, 26, 3554;
(c) J. Wang and J. Qiu, J. Mater. Sci., 2016, 51, 4728; (d) M. Zheng,
5 0 -GMP oligomers that form G-quartet like self-assemblies, which S. Ruan, S. Liu, T. Sun, D. Qu, H. Zhao, Z. Xie, H. Gao, X. Jing and
is in sharp contrast to the conventional graphitic core based Z. Sun, ACS Nano, 2015, 9, 11455.
carbon dots. The spectroscopic evidence collectively suggests that 9 (a) P. Mirtchev, E. J. Henderson, N. Soheilnia, C. M. Yip and
G. A. Ozin, J. Mater. Chem., 2012, 22, 1265; (b) Z. Ma, Y.-L. Zhang,
guanine and sugar component do not undergo rearrangement L. Wang, H. Ming, H. Li, X. Zhang, F. Wang, Y. Liu, Z. Kang and
during the microwave treatment. RNA oligomers are believed to S.-T. Lee, ACS Appl. Mater. Interfaces, 2013, 5, 5080.
be formed in a high energy environment under the prebiotic 10 (a) A. Y.-Y. Tam and V. W.-W. Yam, Chem. Soc. Rev., 2013, 42, 1540;
(b) D. K. Smith, Nat. Chem., 2010, 2, 162.
conditions,18 and such a self-assembled structure formation of 11 (a) F. Wang, Y.-H. Chen, C.-Y. Liu and D.-G. Ma, Chem. Commun.,
G-dots from the 5 0 -GMP monomer may provide evidence for 2011, 47, 3502; (b) X. Zhang, Y. Zhang, Y. Wang, S. Kalytchuk,
that. The G-dots utilize the inherent self-assembly property of S. V. Kershaw, Y. Wang, P. Wang, T. Zhang, Y. Zhao and H. Zhang,
ACS Nano, 2013, 7, 11234; (c) W. Kwon, S. Do, J. Lee, S. Hwang,
50 -GMP to form fluorescent hydrogels without any externally added J. K. Kim and S.-W. Rhee, Chem. Mater., 2013, 25, 1893.
mono-positive cations indicating that these nano-dots can be used 12 (a) J. C. Vinci, I. M. Ferrer, S. J. Seedhouse, A. K. Bourdon,
as a design platform for creating complex nano-structured J. M. Reynard, B. A. Foster, F. V. Bright and L. A. Coloón, J. Phys.
Chem. Lett., 2013, 4, 239; (b) Q. Xu, Y. Liu, C. Gao, J. Wei, H. Zhou,
materials.19 Moreover, bio-compatible and biodegradable hydrogels Y. Chen, C. Dong, T. S. Sreeprasad, N. Li and Z. Xia, J. Mater.
can find potential applications in long term cell tracking, drug Chem. C, 2015, 3, 9885; (c) S. Hu, W. Zhang, Q. Chang, J. Yang and
delivery and biomedical applications.20 K. Lin, Carbon, 2016, 103, 391; (d) F. Arcudi, L. Ðord-ević and
M. Prato, Angew. Chem., Int. Ed., 2016, 55, 2107.
We thank Department of Atomic Energy (BNRS) and the 13 (a) S. K. Das, Y. Liu, S. Yeom, D. Y. Kim and C. I. Richards, Nano
Department of Science and Technology (DST) Nanomission Lett., 2014, 14, 620; (b) Y. Lu, L. Zhang and H. Lin, Chem. – Eur. J.,
for research funding. TB thanks CSIR, India, for a research 2014, 20, 4246; (c) V. Strauss, J. T. Margraf, C. Dolle, B. Butz,
T. J. Nacken, J. Walter, W. Bauer, W. Peukert, E. Spiecker, T. Clark
fellowship. We thank Mr Pavan Kumar and Rabindra Nath Das and D. M. Guldi, J. Am. Chem. Soc., 2014, 136, 17308; (d) Y. H. Yuan,
for useful suggestions. Z. X. Liu, R. S. Li, H. Y. Zou, M. Lin, H. Liu and C. Z. Huang,
Nanoscale, 2016, 8, 6770.
References 14 (a) L. Wang, S. J. Zhu, H. Y. Wang, S. N. Qu, Y. L. Zhang, J. H. Zhang,
Q. D. Chen, H. L. Xu, W. Han, B. Yang and H. B. Sun, ACS Nano,
1 (a) F. A. Aldaye, A. L. Palmer and H. F. Sleiman, Science, 2008, 2014, 8, 2541; (b) H. P. Liu, T. Ye and C. D. Mao, Angew. Chem.,
321, 1795; (b) F. Wang, C.-H. Lu and I. Willner, Chem. Rev., 2014, Int. Ed., 2007, 46, 6473; (c) M. J. Krysmann, A. Kelarakis, P. Dallas
114, 2881; (c) F. Wang, X. Liu and I. Willner, Angew. Chem., Int. Ed., and E. P. Giannelis, J. Am. Chem. Soc., 2012, 134, 747.
2015, 54, 1098. 15 (a) J. Dash, A. J. Patil, R. N. Das, F. L. Dowdall and S. Mann,
2 (a) L. E. Buerkle and S. J. Rowan, Chem. Soc. Rev., 2012, 41, 6089; Soft Matter, 2011, 7, 8120; (b) R. N. Das, Y. P. Kumar, S. Pagoti,
(b) J. Dash and P. Saha, Org. Biomol. Chem., 2016, 14, 2157; A. J. Patil and J. Dash, Chem. – Eur. J., 2012, 19, 6008; (c) Y. P. Kumar,
(c) G. M. Peters and J. T. Davis, Chem. Soc. Rev., 2016, 45, 3188. R. N. Das, S. Kumar, O. M. Schutte, C. Steinem and J. Dash,
3 (a) I. Bang, Biochem. Z., 1910, 26, 293; (b) M. Gellert, M. Lipsett and Chem. – Eur. J., 2014, 20, 3023; (d) R. N. Das, Y. P. Kumar, O. M.
D. Davies, Proc. Natl. Acad. Sci. U. S. A., 1962, 48, 2013; (c) J. T. Davis, Schutte, C. Steinem and J. Dash, J. Am. Chem. Soc., 2015, 137, 34;
Angew. Chem., Int. Ed., 2004, 43, 668; (d) J. T. Davis and G. P. Spada, (e) Y. P. Kumar, R. Das, O. M. Schütte, C. Steinem and J. Dash,
Chem. Soc. Rev., 2007, 36, 296. Nat. Protoc., 2016, 11, 1039.
4 (a) Y. Qin and L. H. Hurley, Biochimie, 2008, 90, 1149; (b) A. De Cian, 16 (a) A. Wong, R. Ida, l. Spindler and G. Wu, J. Am. Chem. Soc., 2005,
L. Lacroix, C. Douarre, N. Temime-Smaali, C. Trentesaux, J.-F. Riou 127, 6990; (b) G. Wu and I. C. M. Kwan, J. Am. Chem. Soc., 2009,
and J.-L. Mergny, Biochimie, 2008, 90, 131; (c) A. Bugaut and 131, 3180.
S. Balasubramanian, Nucleic Acids Res., 2012, 40, 4727; (d) P. Murat 17 (a) Y. Yu, D. Nakamura, K. DeBoyace, A. W. Neisius and L. B. McGown,
and S. Balasubramanian, Curr. Opin. Genet. Dev., 2014, 25, 22. J. Phys. Chem. B, 2008, 112, 1130; (b) L. E. Buerkle, Z. Li, A. M. Jamieson
5 (a) G. Hong, S. Diao, A. L. Antaris and H. Dai, Chem. Rev., 2015, and S. J. Rowan, Langmuir, 2009, 25, 8833; (c) Z. Li, L. E. Buerkle,
115, 10816; (b) S. Y. Lim, W. Shen and Z. Gao, Chem. Soc. Rev., 2015, M. R. Orseno, K. A. Streletzky, S. Seifert, A. M. Jamieson and S. J. Rowan,
44, 362; (c) C. Ding, A. Zhu and Y. Tian, Acc. Chem. Res., 2013, 47, 20; Langmuir, 2010, 26, 10093; (d) L. E. Buerkle, H. A. von Recum and
(d) L. Cao, M. J. Meziani, S. Sahu and Y. P. Sun, Acc. Chem. Res., 2013, S. J. Rowan, Chem. Sci., 2012, 3, 564.
46, 171; (e) C. T. Nguyen, Y. Zhu, X. Chen, G. A. Sotzing, S. G. Focil 18 A. C. Rios, Proc. Natl. Acad. Sci. U. S. A., 2015, 112, 643.
and R. M. Kasi, J. Mater. Chem. C, 2015, 3, 399. 19 (a) J.-M. Lehn, Science, 2002, 295, 2400; (b) G. M. Whitesides and
6 (a) A. B. Bourlinos, A. Bakandritsos, A. Kouloumpis, D. Gournis, B. Grzybowski, Science, 2002, 295, 2418.
M. Krysmann, E. P. Giannelis, K. Polakova, K. Safarova, K. Hola and 20 (a) R. Xing, K. Liu, T. Jiao, N. Zhang, K. Ma, R. Zhang, Q. Zou, G. Ma
R. Zboril, J. Mater. Chem., 2012, 22, 23327; (b) P. G. Luo, S. Sahu, and X. Yan, Adv. Mater., 2016, 28, 3669; (b) K. Liu, R. Xing, Q. Zou,
S.-T. Yang, S. K. Sonkar, J. Wang, H. Wang, G. E. LeCroy, L. Cao and G. Ma, H. Möhwald and X. Yan, Angew. Chem., Int. Ed., 2016,
Y.-P. Sun, J. Mater. Chem. B, 2013, 1, 2116. 55, 3036.
11162 | Chem. Commun., 2016, 52, 11159--11162 This journal is © The Royal Society of Chemistry 2016
PCCP
This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.
In the presence of a cationic fluorophore (rhodamine 6G) whose absorption has a significant spectral
overlap with the emission of a room temperature ionic liquid (RTIL), the emission of the latter gets
quenched, and the quenching has been shown to be dynamic in nature. It has been shown that
resonance energy transfer (RET) indeed happens between the RTIL (donor) and rhodamine 6G (cationic
acceptor), and RET is the reason for the quenching of the RTIL emission. The spectral and temporal
aspects of the RET (between neat RTILs as the donors and rhodamine 6G as the acceptor) were closely
studied by steady-state and picosecond time-resolved fluorescence spectroscopy. The influence of the
alkyl chain length of the cation, size of the anion, excitation wavelength and concentration of the
acceptor on the RET dynamics were also investigated. The energy transfer time (obtained from the rise
time of the acceptor) was noted to vary from 2.5 ns to 4.1 ns. By employing the Förster formulation, the
donor–acceptor distance was obtained, and its magnitude was found to vary between 31.8 and 37.1 Å.
The magnitude of the donor–acceptor distance was shown to be independent of the alkyl chain length of
the cation but dependent on the size of the anion of the RTIL. Moreover, the donor–acceptor distance
was observed to be independent of the excitation wavelength or concentration of the acceptor. It was
Received 8th April 2015, shown that the Förster formulation can possibly account for the mechanism and hence can explain the
Accepted 26th May 2015 experimental observables in the RET phenomenon. Following the detailed experiments and rigorous
DOI: 10.1039/c5cp02036k analysis, a model has been put forward, which can successfully explain the nanoscopic environment that
a cationic fluorophore experiences in an RTIL. Moreover, the nanoscopic environment experienced by
www.rsc.org/pccp the cationic probe has been noted to be different from that experienced by a neutral fluorophore.
This journal is © the Owner Societies 2015 Phys. Chem. Chem. Phys., 2015, 17, 16587--16593 | 16587
View Article Online
Paper PCCP
the existence of specific interactions (like H-bonding) between Chart 1 Chemical structure of 5 RTILs and rhodamine 6G.
the dipolar probe and RTIL.33–35 Fluorescence dynamical studies
have shown that probe-dependent solvation in RTILs is biphasic
the concentration of the acceptor on the RET dynamics. Moreover,
in nature, where one component is typically of a few hundred
since different fluorescence dynamics in RTILs have been shown
picoseconds and the other component is in the range of a few
to be excitation and monitoring wavelength dependent, we
nanoseconds.35–38 However, because of the complex structural
probed different excitation and monitoring wavelengths to probe
aspect of RTILs, the nature of the solvation structure is still not
their effect on RET dynamics. All these rigorous experiments, are
fully understood. Hence, it is very much necessary to understand
helpful in order to understand whether the Förster method44 can
the nanoscopic environment a probe experiences while being
account for the experimentally observable RET phenomenon.
dissolved in an RTIL.
The chemical structure of the RTILs and rhodamine 6G are
Resonance energy transfer studies in RTILs have been reported
depicted in Chart 1.
earlier.39–42 Very recently a modest but significant effort has been
made towards understanding the nanoscopic environment a
fluorophore experiences while being dissolved in an RTIL by Experimental
exploring this with the resonance energy transfer (RET) techni-
que.43 The probe chosen was a neutral one.43 Förster formulation44 Steady-state absorption and fluorescence spectra were recorded
has been shown to explain the fluorescence dynamical parameters with a UV-vis spectrophotometer (Cary100, Varian) and a spectro-
related to the RET phenomenon with reasonable accuracy. The fluorimeter (Fluoromax-3), respectively. The experimental tempera-
rise time (varying from 3.00 to 4.00 ns) or rate constant of energy ture was maintained at 298 K. Time resolved fluorescence decay
transfer has been shown to be independent of the excitation experiments were carried out using a time-correlated single-photon
wavelength and the concentration of the acceptor. However, the counting (TCSPC) spectrometer (5000, IBH). Two diode lasers
magnitude of the rise time has been shown to vary with (377 nm and 402 nm) were used as the excitation sources, and an
the nature of the cation (alkyl chain length) and with the size MCP photomultiplier was used as the detector. The instrument
of the anion. By employing Förster formulation, the donor– response function (IRF) was detected by placing a scatterer, and
acceptor distance inside nanostructural RTIL media could be the FWHM of the IRF was B70 ps. A nonlinear least-squares
obtained, and were seen to vary from 30.35 to 36.48 Å.43 The size iteration procedure using IBH DAS 6 (Version 2.2) decay analysis
of the nanostructural cage has been shown to be dependent on software was used for the fitting of all the decay curves. w2 values
the size of the alkyl chain length of the cation, as well as on the and the plot of residuals were used as parameters to check the
size of the anion of the RTILs.43 It was shown that the size of the goodness of the fit. The fluorescence quantum yields were
nanoscopic cages does not depend on the excitation wavelength calculated using coumarin 153 as a reference.
or concentration of the acceptor. Following detailed experiments
and rigorous analysis, a model has been put forward that can Results and discussion
successfully explain the nanoscopic environment that a neutral
fluorophore experiences in an RTIL.43 The RTILs show (see the ESI†) typical broad absorption spectra
RTILs are composed of cations and anions. Hence the kind of in UV region, with a long tail extending in to the visible region.
environment an ionic probe will experience should be different The emission behavior of these five RTILs is quite broad in
from the environment experienced by a neutral probe. In order the UV-vis region (see ESI†). A representative plot emission of
to gain an insight in this direction, we used rhodamine 6G as a hmim[BF4] is shown in Fig. 1. Also shown in the same figure is
reference cationic probe to help us explore the RET phenomenon the absorption spectrum of rhodamine 6G. The significant
between RTILs (as a donor) and rhodamine 6G (as an acceptor). spectral overlap between the emission spectrum of the donor
We employed RTILs with the same cation but different anion (RTIL) and the absorption spectrum of the acceptor (rhodamine
and vice versa in order to understand the effect of the cation 6G) fulfils the prerequisite condition to observe RET between
and anion on the nanocages that are formed surrounding the the donor and the acceptor.45,46
cationic probe. We used different concentrations of the reference With increasing the concentration of the acceptor (rhodamine
cationic probe (acceptor) in order to understand the effect of 6G), the emission intensity of the donor (hmim[BF4]) decreases
16588 | Phys. Chem. Chem. Phys., 2015, 17, 16587--16593 This journal is © the Owner Societies 2015
View Article Online
PCCP Paper
This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.
Open Access Article. Published on 26 May 2015. Downloaded on 1/31/2023 9:54:00 AM.
Fig. 1 Emission spectrum of the donor (hmim[BF4]) [lex = 377 nm] and the Fig. 3 Fluorescence decay curve of RTIL (hmim[BF4]) in the absence and
absorption spectrum of the acceptor (rhodamine 6G). in the presence of the acceptor (lex = 377 nm and lem = 450 nm)
(The decay behavior has been fitted with a tri-exponential decay function
P P
y = y0 + Bi* exp(t/ti). (Note that Bi = 100)).
gradually (Fig. 2a). This observation indicates that more and more
quenching of the donor emission is happening with increase in
the concentration of the acceptor. From the steady-state emission decay becomes faster with increasing the concentration of the
of the donor, a Stern–Volmer plot was obtained and is depicted acceptor (rhodamine 6G). This observation clearly indicates
in Fig. 2b. A linear Stern–Volmer plot (Fig. 2b) indicates that the that the nature of the quenching is dynamic.
nature of the quenching is either static or dynamic.45,46 The excited state time constants of the donor decay are
In order to know whether static or dynamic quenching is depicted in Table 1. As can be seen from Table 1, for excitation
happening, the time resolved fluorescence decay of the donor was at 377 nm, the excitation average lifetime of the donor (RTIL)
measured. As can be seen from Fig. 3, the donor (hmim[BF4]) decreases from 4.87 ns to 4.08 ns on changing the concentration of
the acceptor (rhodamine 6G) from 0 to 236 mM. Thus, a reduction
of B16% of the average lifetime of the donor was noted. Also,
for excitation at 402 nm, the reduction of the average lifetime of
the donor was noted to be B16% (from 6.14 ns to 5.16 ns).
Thus, the reduction of the average lifetime of the donor was
noted to be independent of the excitation wavelength.
However, that still does not prove that the quenching occurs
because of the resonance energy transfer between the RTIL
as a donor and rhodamine 6G as an acceptor. It has been
documented in literature that it could only be the quenching of
the donor fluorescence, without resonance energy transfer to
acceptor.45–48 We would like to emphasise that observation of
the Stern–Volmer plot or the faster fluorescence decay of the
donor in the presence of the acceptor does not necessarily
402 470 0 0.88 13.25 3.31 43.84 10.67 42.91 6.14 1.06
88 0.34 7.82 2.42 39.25 9.10 52.93 5.79 1.22
Fig. 2 (a) Variation of the RTIL (hmim[BF4]) emission intensity (lex = 377 nm) 110 0.34 8.63 2.36 39.86 8.78 51.51 5.49 1.21
w.r.t. the increasing concentration of rhodamine 6G; (b) Stern–Volmer 142 0.26 8.37 2.20 38.73 8.45 52.90 5.34 1.21
plot showing the concentration-dependent change of fluorescence of 176 0.27 7.80 2.21 38.28 8.42 53.91 5.26 1.18
the donor. 236 0.26 6.92 1.96 36.61 7.85 56.47 5.16 1.28
This journal is © the Owner Societies 2015 Phys. Chem. Chem. Phys., 2015, 17, 16587--16593 | 16589
View Article Online
Paper PCCP
16590 | Phys. Chem. Chem. Phys., 2015, 17, 16587--16593 This journal is © the Owner Societies 2015
View Article Online
PCCP Paper
402 emim[FAP] 6.37 0.13 1.98 4.14 39.81 37.17 2.41 Scheme 1 Nanocage model structure of RTILs in the presence of a
bmim[FAP] 7.30 0.19 2.03 3.24 42.63 37.10 3.08 positively charged fluorophore.
Open Access Article. Published on 26 May 2015. Downloaded on 1/31/2023 9:54:00 AM.
This journal is © the Owner Societies 2015 Phys. Chem. Chem. Phys., 2015, 17, 16587--16593 | 16591
View Article Online
Paper PCCP
concentration of the donor or acceptor should be B100 mM or 9 S.-H. Chen, F.-R. Yang, M.-T. Wang and N.-N. Wang,
higher. Moreover, different RET dynamics should be observed C. R. Chim., 2010, 13, 1391–1396.
for different types of acceptor molecules (neutral and cationic 10 W. A. Henderson, P. Fylstra, H. C. D. Long, P. C. Trulove and
probe) using the same RTILs. Thus, it could be shown beyond S. Parsons, Phys. Chem. Chem. Phys., 2012, 14, 16041–16046.
This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.
any doubt that it is not the impurity but rather the inherent 11 A. Triolo, O. Russina, H.-J. Bleif and E. Di Cola, J. Phys.
emission of imidazolium that is involved in the RET process. Chem. B, 2007, 111, 4641–4644.
12 C. Hardacre, J. D. Holbrey, C. L. Mullan, M. Nieuwenhuyzen,
T. G. A. Youngs and D. T. Bowron, J. Phys. Chem. B, 2008,
Conclusions 112, 8049–8056.
In conclusion, we have shown that in the presence of a cationic 13 C. Hardacre, J. D. Holbrey, C. L. Mullan, T. G. A. Youngs and
fluorophore (rhodamine 6G) as an acceptor whose absorption D. T. Bowron, J. Chem. Phys., 2010, 133, 074510.
Open Access Article. Published on 26 May 2015. Downloaded on 1/31/2023 9:54:00 AM.
has significant spectral overlap with the emission of the RTIL, 14 A. Mele, C. D. Tran and S. H. D. P. Lacerda, Angew. Chem.,
the emission of the latter can be quenched and the nature of Int. Ed., 2003, 42, 4364–4366.
the quenching has been shown to be dynamic. It has been 15 A. Mele, G. Romano, M. Giannone, E. Ragg, G. Fronza,
shown that RET is indeed happening between the RTIL (donor) G. Raos and V. Marcon, Angew. Chem., Int. Ed., 2006, 45,
and the rhodamine 6G (cationic acceptor) and RET is the 1123–1126.
reason for quenching of the RTIL emission. By employing a 16 S. M. Urahata and M. C. C. Ribeiro, J. Chem. Phys., 2004, 120,
Förster formulation, we obtained the donor acceptor distance 1855–1863.
(31.8 Å to 37.1 Å). The magnitude of the donor–acceptor 17 Y. Wang and G. A. Voth, J. Am. Chem. Soc., 2005, 127,
distance was shown to be independent of the alkyl chain length 12192–12193.
of the cation but dependent on the size of the anion of the 18 J. N. C. Lopes, M. F. C. Gomes and A. A. H Padua, J. Phys.
RTILs. Moreover, the donor–acceptor distance was noted to be Chem. B, 2006, 110, 16816–16818.
independent of the excitation wavelength or concentration of 19 O. Borodin and G. D. Smith, J. Phys. Chem. B, 2006, 110,
the acceptor. It has been shown that the Förster formulation 11481–11490.
can perhaps account for the mechanism and hence can explain 20 W. Jiang, Y. Wang and G. A. Voth, J. Phys. Chem. B, 2007,
the experimental observables in the RET phenomenon. Finally, 111, 4812–4818.
we put forward a model that can explain the experimental 21 Z. Hu and C. J. Margulis, Acc. Chem. Res., 2007, 40, 1097–1105.
observables with reasonable accuracy. Also, the nanoscopic 22 G. Raabea and J. Köhler, J. Chem. Phys., 2008, 128, 154509.
environment experienced by the cationic probe was noted to 23 J. D. Andrade, E. S. Böes and H. Stassen, J. Phys. Chem. B,
be different from that of the neutral probe. 2008, 112, 8966–8974.
24 T. Yan, Y. Wang and C. Knox, J. Phys. Chem. B, 2010, 114,
6905–6921.
Notes 25 S. N. V. K. Aki, J. F. Brennecke and A. Samanta, Chem.
Commun., 2001, 413–414.
The authors declare no competing financial interests.
26 P. K. Mandal and A. Samanta, J. Phys. Chem. B, 2005, 109,
15172–15177.
Acknowledgements 27 C. Wakai, A. Oleinikova and H. Weingaertner, J. Phys. Chem.
Lett., 2005, 109, 17028–17030.
PKM thanks IISER – Kolkata for financial help and instrumental 28 F. V. Bright and G. A. Baker, J. Phys. Chem. B, 2006, 110,
facilities. Support from the Fast-Track Project (SR/FT/CS-52/2011) 5822–5823.
of DST-India is gratefully acknowledged. AG thanks UGC, CKD 29 C. Wakai, A. Oleinikova and H. Weingaertner, J. Phys. Chem. B,
thanks IISER – Kolkata, TC thanks CSIR – India for Fellowship. 2006, 110, 5824.
30 A. Paul, P. K. Mandal and A. Samanta, Chem. Phys. Lett.,
2005, 402, 375–379.
References
31 A. Paul, P. K. Mandal and A. Samanta, J. Phys. Chem. B, 2005,
1 T. Welton, Chem. Rev., 1999, 99, 2071–2083. 109, 9148–9153.
2 P. Wasserscheid and W. Keim, Angew. Chem., Int. Ed., 2000, 32 P. K. Mandal, M. Sarkar and A. Samanta, J. Phys. Chem. A,
39, 3772–3789. 2004, 108, 9048–9053.
3 J. Dupont, R. F. D. Souza and P. A. Z. Suarez, Chem. Rev., 33 P. K. Mandal, A. Paul and A. Samanta, J. Photochem.
2002, 102, 3667–3692. Photobiol., A, 2006, 182, 113–120.
4 R. D. Rogers and K. R. Seddon, Science, 2003, 302, 792–793. 34 A. Samanta, J. Phys. Chem. B, 2006, 110, 13704–13716.
5 K. R. Seddon, Nat. Mater., 2003, 2, 363–365. 35 P. K. Mandal, S. Saha, R. Karmakar and A. Samanta,
6 H. Weingaertner, Angew. Chem., Int. Ed., 2008, 47, 654–670. Curr. Sci., 2006, 90, 301–310.
7 J. P. Hallet and T. Welton, Chem. Rev., 2011, 111, 3508–3576. 36 A. Samanta, J. Phys. Chem. Lett., 2010, 1, 1557–1562.
8 J. D. Holbrey, W. M. Reichert and R. D. Rogers, Dalton 37 X.-X. Zhang, M. Liang and N. P. Ernsting, J. Phys. Chem. B,
Trans., 2004, 2267–2271. 2013, 117, 4291–4304.
16592 | Phys. Chem. Chem. Phys., 2015, 17, 16587--16593 This journal is © the Owner Societies 2015
View Article Online
PCCP Paper
38 A. Ghosh, T. Chatterjee and P. K. Mandal, Chem. Commun., 43 A. Ghosh, T. Chatterjee, D. Roy and A. Das, J. Phys. Chem. C,
2012, 48, 6250–6252. 2014, 118, 5051–5057.
39 A. Adhikari, D. K. Das, D. K. Sasmal and K. Bhattacharyya, 44 Th. Förster, Ann. Phys., 1948, 2, 55–75.
J. Phys. Chem. B, 2009, 113, 3737–3743. 45 B. Valeur, Molecular Fluorescence Principles and Applications,
This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.
40 D. K. Das, A. K. Das, T. Mondal, A. K. Mandal and Wiley-VCH Verlag GmbH, Weinheim, 2002.
K. Bhattacharyya, J. Phys. Chem. B, 2010, 114, 13159–13166. 46 J. R. Lakowicz, Principles of Fluorescence Spectroscopy,
41 H. Izawa, S. Wakizono and J. Kadokawa, Chem. Commun., Plenum Press, New York, 3rd edn, 1999.
2010, 46, 6359–6361. 47 S. Saha and A. Samanta, J. Phys. Chem. A, 1998, 102, 7903–7912.
42 V. G. Rao, S. Mandal, S. Ghosh, C. Banerjee and N. Sarkar, 48 K. Santhosh, S. Patra, S. Soumya, D. C. Khara and
J. Phys. Chem. B, 2012, 121, 12021–12029. A. Samanta, ChemPhysChem, 2011, 12, 2735–2741.
Open Access Article. Published on 26 May 2015. Downloaded on 1/31/2023 9:54:00 AM.
This journal is © the Owner Societies 2015 Phys. Chem. Chem. Phys., 2015, 17, 16587--16593 | 16593
Research Article
pubs.acs.org/acscatalysis
■ INTRODUCTION
The phenalenyl (PLY) unit has played an intriguing role in
describe the present status of phenalenyl based radical
materials.6
different fields of research spanning from chemistry, to material For some time, we started working on another possible
chemistry, to device physics, acting as a key electronic reservoir. electronic state of phenalenyl moiety, where it stays in its
Phenalenyl is a well-known odd alternant hydrocarbon with cationic state possessing an empty NBMO. Recently, we have
high symmetry (D3h) which has the ability to form three redox demonstrated that the cationic state can be generated by metal
species: cation, neutral radical, and anion.1 Formation of this ion coordination, and organometallic phenalenyl complexes can
redox triad involves the use of the formally nonbonding act as catalysts, in which the phenalenyl unit played a significant
molecular orbital (NBMO) of the phenalenyl molecule and role in modulating the catalytic outcome.7 Furthermore, we
hence does not greatly affect the stability of the resulting recently demonstrated the potential of cationic phenalenyl
species.2,3 The transformation from cation to radical to anion of moiety in molecular spintronics.8 A zinc ion coordinated
phenalenyl unit progresses through successive electron accept- cationic phenalenyl unit in an organometallic compound acted
ance in an accessible NBMO as predicted from an early Hückel as a spin trap, leading to the development of the phenalenyl
molecular orbital (HMO) calculations in 1960s.2 Later, the based spin memory device. 8 When the nonmagnetic
concept of using this NBMO of phenalenyl was proposed for phenalenylorganozinc compound is deposited over a ferro-
designing neutral free radical based molecular conductors by magnetic substrate, the cationic phenalenyl accepts electron
Haddon.3 Subsequently, this idea led to the discovery of the from the substrate and becomes magnetic, resulting in an
best neutral organic conductor at room temperature.4 The unexpectedly large magnetoresistance of 20% at near room
radical state of phenalenyl molecules has been used as a temperature which can be manipulated by external stimuli.8
building block to prepare intriguing materials for exploring new This result could provide a platform for developing molecular
conjugated electronic systems, such as multifunctional elec- memory devices of next generation with superior strength,
tronic and magnetic materials exhibiting simultaneous bist-
ability in multiple physical channels.4a,5 Until recently, most Received: July 24, 2014
studies focused on the radical state of phenalenyl based Revised: October 7, 2014
molecules, and recent articles by Morita, Takui, and Hicks Published: October 22, 2014
© 2014 American Chemical Society 4307 dx.doi.org/10.1021/cs5010695 | ACS Catal. 2014, 4, 4307−4319
ACS Catalysis Research Article
which is based on the phenalenyl’s ability to accept an electron which the NBMO has been used as the major pathway of
utilizing its empty NBMO. interaction, until the contribution of our study.
In the present study, we used metal-free cationic phenalenyl
compounds to design an organic Lewis acid catalyst which
utilizes the empty NBMO of phenalenyl for electron accept-
■ RESULTS AND DISCUSSION
The phenalenyl (PLY)-based cations (Chart 1a) 9-methoxy-1-
ance. In this regard, we posed a question whether the presence ethoxyphenalenium tetrafluoroborate (1, O,O-system) and 9-
of this readily available NBMO in the phenalenyl system in its methylamino-1-ethoxyphenalenium tetrafluoroborate (2, N,O-
cationic state can be utilized to design well-defined molecular system) were synthesized following the reported procedure.12
catalysts. It was postulated that the vacant NBMO of For the present study, we have chosen the epoxide ring-
phenalenyl unit may have a great influence in catalytic opening reaction with amine as a model organic reaction. The
reactions. The initial interaction of the electron density of the epoxide ring-opening reaction is an important one as it can lead
approaching nucleophile could be supported by the presence of to the formation of β-hydroxy amine, which has been key
an energetically accessible orbital of the ligand system (Chart intermediate for the synthesis of several drugs, pharmaceuticals,
1). It is now well documented that dramatic enhancement of and natural products.13 The classical approach for the synthesis
of β-hydroxy amine involves Lewis acid based14 (preferably a
Chart 1. (a) Cationic Phenalenyl Compounds 1 and 2; (b) metal salt) activation of epoxides followed by nucleophilic
Schematic Description of Molecular Orbitals of Cationic attack with amines. As a result, various Lewis acid catalysts15,16
PLY1a have been developed for activating the epoxides. As proof of
concept, we began the study by testing the catalytic activity of 1
(2.5 mol %) for the model reaction of 2-(phenoxymethyl)-
oxirane (3a) with N-methylaniline (4a) under neat condition at
40 °C. To our expectation, the desired β-hydroxy amine, 1-
(methyl(phenyl)amino)-3-phenoxypropan-2-ol (5) was ob-
tained in 84% yield (Table 1, entry 1). Subsequently, the
a
The schema above show that NBMO of PLY cation can support
interaction with nucleophilic substrate (Nu).
spectrum, we did not observe any significant change in exclusively on the less-hindered side of the terminal epoxide
chemical shift from either of these two individual moieties, (Table 2, entries 1−2, 4−5, etc.). When the cyclohexene oxide
which ruled out the possibility of HBF4 catalyzed reaction. The (3c) was treated with 4a (Table 2, entries 3), it gave the
parent 9-hydroxyphenalenone, when used as a catalyst, resulted diastereoselective trans-β-hydroxy amine exclusively. The
in poor conversion (11%, Table 1, entry 4). In another control reaction of epichlorohydrin 3e (Table 2, entry 5) provided
experiment, we used NaBF4 as catalyst which also led to very excellent example of chemo and regioselectivity and afforded
poor catalytic activity (19%, Table 1, entry 5) revealing that the 85% yield of the β-hydroxy amine corresponding to
presence of the PLY-based cationic moiety as catalyst has nucleophilic attack at the terminal carbon of the epoxide
prominent effect on the catalysis. In another control experi- moiety. Different electronically and sterically diverse amines
ment, when this reaction was performed in the absence of were reacted in the presence of catalysts 1 and 2 at 40 °C under
catalyst 1, only 8% (Table 1, entry 6) conversion was observed. solvent-free condition. The reaction rate appeared to be
This observation confirms the indispensable role of the PLY- influenced by the nature of the amine (Table 2, entries 1 and
based cation during catalysis. 6) as evident from the reaction performed with 4a (an aromatic
In order to check the versatility of this catalytic process, we amine), resulting in less conversion than with an aliphatic
investigated the scope of catalyst 1 (2.5 mol %) for the amine 4b. Treatment of terminal epoxides 3a and 3b with
aminolysis reaction of several functionalized epoxides. We also morpholine 4c (Table 2, entries 9−11) and piperidine 4d
used another PLY-based cation, 9-methyiamino-1-ethoxy- (Table 2, entries 12−14) as a congener of cyclic aliphatic
phenalenium tetrafluoroborate (2, N,O-system), as catalyst to amine, gives the corresponding β-hydroxy amine in good to
check the change in efficacy of the catalyst depending on the excellent yield. A sterically hindered amine, such as di-
nature of PLY cation used. Under the optimized procedure, isopropylamine 4e (Table 2, entry 15), also works under the
reactions proceed smoothly and gave the corresponding β- optimized condition. We tried to synthesize the β-hydroxy
hydroxy amine in good to excellent yield. The regioselective amine with long chain acyclic amine. The regioselective
nucleophilic attack by different amines took place almost nucleophilic attack took place almost exclusively on the less
4309 dx.doi.org/10.1021/cs5010695 | ACS Catal. 2014, 4, 4307−4319
ACS Catalysis Research Article
hindered side of the terminal epoxide (Table 2, entry 16−18) At the same time, catalyst 2 has a comparatively smaller LUMO
upon the treatment of dibutylamine (4f) with the terminal − LUMO+1 gap of 2.34 eV and LUMO−1 − LUMO gap of
epoxides (4a, 4b, and 4d). This methodology is equally 5.22 eV. Thus, the DFT calculation supports the presence of
effective for aromatic as well as aliphatic amines. The catalytic energetically accessible empty molecular orbital for electron
results reveal that irrespective of amines, reaction proceeds with acceptance in these PLY-based catalysts. Interestingly, the
excellent conversion in case of catalyst 1. Though there are only calculation reveals that the relative energy of the LUMO largely
very few reports for the epoxide ring-opening reaction in the depends on the nature of the substitutions in the PLY. In other
presence of metal-free catalysts,16a,c the present catalysts show words, the LUMO energy levels in these PLY units can be
comparable or better catalytic activity than the reported metal- tuned by changing the substitution from O,O (in catalyst 1) to
free catalysts. Moreover, this catalyst system can stay live for a N,O (in catalyst 2) in the present study (Figure 2c).
number of consecutive catalytic cycles without losing the To understand the electron-acceptance property of this
catalytic activity significantly (see Supporting Information for organic Lewis acid and to further support our DFT-calculation-
details). As a general observation, it was noticed from Table 2, based findings, we carried out a cyclic voltammetric study.
in all the reactions tested in the present study, the catalyst 1 These experiments indeed revealed the presence of two one-
outperforms catalyst 2 in terms of activity under the identical electron acceptance processes in both catalyst 1 and 2 (Figure
reaction condition. 3a and b, respectively). The E1/2 values of 1 (E11/2 = −0.408 V
We have performed a comparative kinetic study to and E21/2 = −1.486 V) and 2 (E11/2 = −0.931 V and E21/2 =
understand the relative efficacy of catalysts 1 and 2 for the −1.820 V) indicate that electron acceptance from the first to
model reaction involving 3a and 4a via 1H NMR spectroscopy the second one becomes progressively difficult as a result of
over the course of the first 6 h (Figure 1). From the kinetic increasing Coulombic repulsion in the system. A similar
analysis, it is evident that catalyst 1 is superior than that of observation was also noted by Haddon and co-workers with
catalyst 2, as also observed consistently from the data presented the 9-ethoxy-1-ethoxyphenalenium tetrafluoroborate system.17
in Table 2. Earlier EPR studies on spirobiphenalenyl boron salts4,18 had
shown that the first reduction leads to the formation of an EPR
active radical species and the second reduction results in the
EPR silent anionic species, supporting our result that the
LUMO of a PLY system can accommodate two electrons. The
cyclic voltammetry thus in the present case can be explained by
considering the generation of a neutral radical and an anionic
species (Figure 3c). Comparing first reduction potential data of
1 and 2 [1 (E11/2 = −0.408 V) and 2 (E11/2 = −0.931 V)], it is
clear that the reduction of 1 is much more facile than 2. This
result points out that the LUMO of 1 is more accessible for
electron acceptance as compared to that of 2, which is in
consonance with our DFT calculation (Figure 2c).
Next, we attempted to envisage the molecular level
interaction of the catalyst with the substrates to understand
the mechanistic insight of the epoxide ring opening with amine.
A general perception for the Lewis acid catalyzed aminolysis
reaction of epoxide ring-opening reaction is that a Lewis acid
(preferably a metal salt) can activate the epoxide substrate by
forming a strong coordinate bond with the oxygen atom of the
epoxide ring. This type of interaction increases the electro-
philicity of the carbon atoms of the epoxide ring which assists
in the opening of epoxide ring.14−16 However, when catalyst 1
Figure 1. Kinetic studies revealing that catalyst 1 has better efficacy was mixed with amine 4a in a ratio of 1:1, there was a sharp
than catalyst 2. visual change in color from orange to dark brown. On the other
hand, no such sharp color change was observed when we
The difference in reactivity between the catalysts 1 and 2 physically mix catalyst 1 and epoxide 3a. This observation
intrigues us to find its relationship in the molecular level, and clearly highlights that NBMO-based organic Lewis acid works
therefore, we have carried out density functional theory (DFT) in a different pathway than the traditional Lewis acid catalyzed
study at the M06-2X/6-31+G(d) level of theory. An analysis of aminolysis reaction of epoxide and further prompted us to
the charge density distributions is shown in Figure 2a. The record a detailed absorption and emission spectra of these
calculated orbital energy diagrams of 1 and 2 show that the mixtures.
LUMO (see Figure 2b) of these catalysts are largely We followed the interaction between amine/epoxide (see
nonbonding in nature and is highly separated from the Supporting Information for epoxide) with the catalyst
LUMO−1 and LUMO+1(Figure 2c). The presence of the spectroscopically. After a careful stepwise addition of amine
NBMO of PLY cation as the LUMO has earlier been predicted 4a to a solution of 1 (25 μM), the absorbance of 1 was found to
for the cationic state of parent phenalenyl molecule.1,3 The be gradually decreased with increasing the concentration of 4a.
LUMO of catalyst 1 (Figure 2c) is well separated from its Presence of an isosbestic point (Figure 4a) at 468 nm for 4a
nearest molecular orbitals with a calculated LUMO−(LUMO proves beyond any doubt that a ground state interaction has
+1) gap of 2.52 eV and a LUMO−1 − LUMO gap of 5.37 eV. indeed happened with catalyst 1 and 4a.
4310 dx.doi.org/10.1021/cs5010695 | ACS Catal. 2014, 4, 4307−4319
ACS Catalysis Research Article
Figure 2. (a) Mulliken charge densities on selected carbon atoms in catalyst 1. (b) Computed LUMO of the catalyst 1. (c) Molecular orbital energy
profile diagrams of catalysts 1 and 2. Energy of LUMO level (in eV) is shown.
Figure 3. (a,b) Cyclic voltammogram of catalysts 1 and 2 in acetonitrile, referenced to SCE via internal ferrocene revealing two successive one-
electron reduction processes. (c) Successive electron acceptance by catalysts 1 and 2 generates a neutral radical and an anionic species.
In order to obtain the stoichoimetry of the amine−catalyst We have also estimated the association constant of this process
adduct we have carried out Job’s plot (ΔOD vs mole fraction of between the 1 and 4a from emission spectroscopic measure-
4a; where ΔOD = I0 − I; I0 = absorbance of 1 in absence of 4a, ments by using Benesi−Hildebrand equation. From the
and I = absorbance of 1 in the presence of quencher 4a) which Benesi−Hildebrand plot (Figure 4e), the cumulative associa-
clearly demonstrates a 1:1 association of 1 and 4a (Figure 4b). tion constant has been calculated to be 13.3 × 103 LM−1 for the
Further, we have monitored the emission spectroscopy of 1 at catalyst−amine pair and that for the catalyst−epoxy pair is 4.8
different concentration of amine to understand the exact × 103 LM−1 (see Supporting Information for details). These
interaction of 1 with 4a. Sequential addition of 4a to a solution values suggest that the association between catalyst and amine
of 1 (25 μM) resulted in a gradual decrease in the fluorescence
is 2−3 times much stronger than that between catalyst epoxide.
intensity of the catalyst 1 (Figure 4c), further confirms the
Furthermore, to understand the nature of association between
interaction between amine and catalyst. The extent of the
interaction has been followed by the Stern−Volmer plot (plot catalyst and amine, we mixed the catalyst 1 and amine 4a (in
of F0/F against the concentration of 4a; where F0 = 1:1 ratio) and recorded the 1H NMR spectrum. We did not
fluorescence intensity in absence of any quencher, F = observe any significant change in the chemical shift from either
fluorescence intensity in the presence of quencher) (Figure of these two individual starting moieties (see Supporting
4d) as well as by Benesi−Hildebrand plot (plot of 1/[F − F0] Information for NMR spectra). This NMR experiment clearly
versus 1/[4a]) (Figure 4e). As it can be seen from Figure 4d, a suggests that there is no strong association (covalent bond
strict linear behavior of the quenching (obtained from Stern− formation) between catalyst 1 and amine 4a. However, from
Volmer plot) confirms that the quenching is static in nature.19a the absorption and emission spectroscopic studies, we can
4311 dx.doi.org/10.1021/cs5010695 | ACS Catal. 2014, 4, 4307−4319
ACS Catalysis Research Article
Figure 4. Spectroscopic signature of the binding of catalyst 1 with amine 4a. (a) Absorption spectra in MeCN. (b) Job’s Plot (plot of ΔOD vs mole
fraction of amine 4a). (c) Steady state emission spectra in MeCN. (d) Stern-Volmer plot (plot of F0/F vs 4a concentration). (e) Benesi Hildebrand
plot (plot of 1/[F − F0] vs 1/[4a]). For details see text.
Figure 5. Optimized geometries of catalyst 1 with (a) a model epoxide 3a and (b) amine 4a. Selected bond distances (in Å) are shown. (c)
Molecular orbital energy profile diagrams of catalyst 1, amine 4a, and epoxide 3a. Energy gaps between the LUMO level of catalyst 1 and HOMO
levels of reactants (in eV) are shown.
Figure 6. Kinetic studies of epoxide ring-opening reaction for the formation of 5 monitored by 1HNMR spectroscopy in CDCl3. (a,b) Plots showing
the change of epoxide concentration with time for the reaction of 3a (2 mmol), 4a (2 mol) with varying catalyst 1 concentration. (c,d) Plots of rate
(mmol/h) versus epoxide concentration (mmol).
conclude that there is a loose association (noncovalent) were then computed using N-methylaniline as the model amine
between catalyst 1 and amine 4a. (Figure 5b). It is important to note that catalyst 1 can bind with
The spectroscopic results were rationalized on the basis of N-methylaniline through a large number of stacked and
computations on the interaction of catalyst 1 with both the nonstacked orientations in addition to a lone pair−π interaction
amine and epoxide reactants. Ethylene oxide was used as the model. We have optimized all important geometries that could
simplified model to mimic the epoxide reactants. Calculations be formed from the binding of 4a with different binding sites
show that epoxide could form a complex with the planar surface on catalyst 1 to identify the most stable complex (details of the
of catalyst 1 through the lone pair−π interaction resulting from optimized structures and their energetics are tabulated in the
an interaction of 1:1 mixture of 1 and ethylene oxide as shown Supporting Information). Among them, the most suitable
in Figure 5a. The distance of 2.83 Å between the aromatic reactive intermediate (Figure 5b) is stabilized by a binding
surface of catalyst 1 and epoxide oxygen is within the range of energy of ΔE (BSSE corrected) = −11.1 kcal/mol due to
conventional lone pair−π interactions.19b,c However, the favorable π−π stacking interactions.
binding energy of this complex was found to be small. Binding Furthermore, it was observed from the calculations that the
energies obtained for adduct of 1 with epoxide was of ΔE LUMO of catalyst 1 remains 0.72 eV above the HOMO of
(BSSE corrected) = −5.5 kcal/mol. Also, the Gibb’s free energy amine 4a while for the HOMO of epoxide 3a, the LUMO of 1
changes were found to be clearly positive; +2.5 kcal/mol. This is placed 1.62 eV above (Figure 5c). Thus, the comparatively
suggests a poor association of 1 with epoxide and possibly shorter HOMO−LUMO gap in the case of adduct formed
eliminates the well-documented electrophilic activation of between 1 and 4a results in stronger donor−acceptor
epoxide in the present case.15,16 interaction and initiate the nucleophilic activation. These
Alternatively, it may be considered that amine might form a computational studies are also in agreement with our
more stable adduct with 1 as the nucleophilicity of the nitrogen spectroscopic results.
of the amine is more than that of the oxygen atom of the Detailed kinetic studies were performed following the
epoxide. Interactions of catalyst 1 with the amine reactants literature method reported earlier20 to gain further insight
4313 dx.doi.org/10.1021/cs5010695 | ACS Catal. 2014, 4, 4307−4319
ACS Catalysis Research Article
Figure 7. Computed potential energy surface (PES) for the reaction between 4a and ethylene oxide in the presence of catalyst 1. Selected bond
distances (in Å) and relative energies with respect to the reactants (in kcal/mol) are shown.
into the aminolysis process for the model reaction between 3a Efforts were also given to determine the order of the reaction
and 4a under neat condition using 1 as catalyst. Kinetic studies with respect to the amine concentration as well as with respect
of all the reactions were performed by following the standard to epoxide concentration, respectively, in a similar way (see
optimized reaction condition. The evolution of the specific Supporting Information for details). It shows a second-order
resonances of the product β-hydroxy amine 5 was monitored by dependency of the reaction rate with respect to the amine
1
H NMR spectroscopy relative to the concentration of either concentration and a first-order dependency with epoxide
starting amine or epoxide. From this conversion, the effective concentration.
concentration (mmol) of the 4a/3a over the courses of the first Thus, from the present study, the overall rate law for the
9 h was determined, which in turn was used for the calculation aminolysis reaction of 3a with 4a catalyzed by 1 can be
of rate (mmol/h) of the reaction. summarized as shown in eq 1.
To determine the order of the reaction with respect to
Rate = k[Catalyst]1 × [Amine]2 × [Epoxide]1 ............ (1)
catalyst concentration, the aminolysis reaction was carried out
in two different concentrations of catalyst 1 (Run 1: with 5 mol Subsequently, in order to understand the complete
% catalyst loading and Run 2: with 7.5 mol% catalyst loading mechanistic cycle and the role of catalyst, we have modeled
separately) and keeping the concentration of epoxide 3a (2 the reaction between ethylene oxide and 4a in the presence (or
mmol) and amine 4a (2 mmol) fixed. The progress of the absence) of catalysts 1 or 2. Calculations at M06-2X/6-
reaction was monitored during the first 9 h (Figure 6a,b) which 31+G(d) level of theory predicted that, in the absence of any
shows gradual decrease of epoxide concentration with time. catalyst, a very high activation energy (ΔΔG‡ = 35.5 kcal/mol)
The amine and epoxide concentration at a given time can be is required for this chemical transformation (see Supporting
quantified with the help of 1H NMR spectroscopy, and the rate Information for detail).
of the reaction can be calculated by determining the conversion The computed potential energy surface for this reaction in
at a given time. A plot of rate (mmol/h) versus different the presence of catalyst 1 is shown in Figure 7. The π−π
epoxide concentrations was determined for two sets of stacked reactive intermediate 6 can readily undergo two
reactions (using two different catalyst concentration), which intramolecular rearrangements, first to 7 with strengthening
reveals a linear increase of the reaction rate with epoxide of catalyst−nitrogen bond and then to an unfolded
concentration (Figure 6c,d). Comparing the rate of reaction for conformation in 8. Small activation barriers of ΔG ‡ = 13.0
two sets of catalyst concentration, one can calculate (see kcal/mol (TS1) and 5.1 kcal/mol (TS2) were found for these
Supporting Information for detail) and confirm the first-order rearrangements. Both of these rearrangements occur with
dependency of the reaction rate with respect to the catalyst strengthening of newly formed C−N bond between the catalyst
concentration. and reactant as well as with the activation of N−H bond. The
4314 dx.doi.org/10.1021/cs5010695 | ACS Catal. 2014, 4, 4307−4319
ACS Catalysis Research Article
Scheme 1. Schematic Presentation of Plausible Mechanism for the Phenalenyl Cation Catalyzed Reaction between Amine with
Epoxide Forming β-Hydroxy Aminea
a
A generalized presentation has been adopted for the sake of simplicity where A, B, C, and D represent 6, 9, 10, and 11 of Figure 7, respectively.
C−N bond distance decreases from 3.05 Å in 6 to 1.63 Å in 7 that the reaction is second order with respect to the amine and
and subsequently shortens to 1.61 Å in the nonstacked, first order with respect to the epoxide and the catalyst (see eq
unfolded conformer 8. At the same time, the N−H bond 1).
weakens from 1.01 Å in free 4a to 1.03 Å in 8. In addition, the From the rate law of the reaction between 3a and 4a in the
DFT calculations show that the formation of 7 and 8 with presence of catalyst 1, it suggests the involvement of two
shorter C··· N bond distances (1.61−1.63 Å) are endothermic molecules of amine with one molecule of each of catalyst and
processes, and the reverse reaction to return to the noncovalent epoxide during the rate-determining step of the aminolysis
adduct 6 should be a rapid processes. This suggests that the reaction. This is also reflected in the DFT calculated reaction
amine catalyst adduct stays in a dynamic equilibrium between 6, profile which revealed that the rate-determining step involves
7, and 8 with low energy barrier. Both the calculations and one epoxide molecule, two amine molecules, and one catalyst
spectroscopic experiments show that the stable catalyst···amine molecule (during the transformation of 10 to 11, Figure 7).
complex is noncovalent in nature. Combining all these results, we propose a schematic
In the following step, the epoxide molecule forms a N−H··· presentation of plausible mechanism for the phenalenyl cation
O hydrogen bond by interacting with 8 resulting in the based organic Lewis acid catalyzed reaction of amine with
intermediate 9. A second molecule of amine (4a) would further epoxide (Scheme 1).
activate the reaction by binding with the epoxide carbon atom The greater nucleophilicity of the nitrogen atom of the amine
to form the weakly bound complex 10. In the succeeding step, moiety in comparison to the oxygen atom of the epoxide
the proton abstraction by the epoxide and the formation of a prefers stronger interaction between the catalyst and amine-
carbon−nitrogen bond between the epoxide and second generating catalytically active species A. The primary
molecule of amine (4a) occurs through the concerted transition interaction between amine and catalyst in A turned out to be
state TS3 to form the complex 11. In the presence of epoxide, a noncovalent type. This is evident from spectroscopic studies
the catalyst amine interaction switches form loose association and the DFT calculation which have shown that the minimum
to a stronger association, and in 11, the C−N bond distance energy geometry of the initial complex between 1 and 4a (6 in
approaches to 1.5 Å, indicating a covalent type bonding Figure 7) is purely noncovalent in nature,22 and it can be
interaction as shown in Figure 7. 11 can undergo a proton ascertained that there is a loose association between catalyst 1
transfer process forming the desired product and regenerating and amine 4a. Interestingly, this 1:1 amine−catalyst association
6.21 Similar to the uncatalyzed reaction, this step was identified in turn activates the hydrogen atom of N−H functionality of
as the rate-determining step. Importantly, the activation energy the amine and prompts the formation of hydrogen bond with
of 25.7 kcal/mol (ΔΔG‡) required to cross the barrier TS3, oxygen of the epoxide molecule forming B. This hydrogen
which is significantly lower compared to that of 35.5 kcal/mol bonding with epoxide in B activates the epoxide molecule by
required for the uncatalyzed reaction (see Supporting increasing its electrophilicy. Then, the second molecule of
Information for detail). amine approaches B and forms the intermediate C by binding
Thus, our calculations justify the role of catalyst in the with the epoxide carbon atom at the sterically and electronically
reaction profile. Further, our computational results have been favorable site. Subsequently the C−O bond of epoxide breaks,
corroborated by the experimental kinetic studies, which show and the new C−N bond between epoxide and second amine
4315 dx.doi.org/10.1021/cs5010695 | ACS Catal. 2014, 4, 4307−4319
ACS Catalysis Research Article
Figure 8. Reaction coordinate for the rate-determining step for the reaction between N-methylaniline and ethylene oxide in the presence of catalyst
2. Selected bond distances (in Å) and relative energies in kcal/mol are shown.
molecule is formed in intermediate D. Finally, a proton transfer attracted to utilize these cationic PLY molecules as versatile
from D via E regenerates the catalytically active species A. DFT organic Lewis acids for different aldehyde based reactions, and
calculation reveals that the Gibb’s free energy change is we planned to use them as catalysts for the synthesis of
thermodynamically feasible (−23.9 kcal/mol, see Supporting bis(indolyl)methanes (see Supporting Information for plausible
Information for details) for the proton transfer process mechanism). The development of new methodologies23 for the
transforming D to A forming desired β-hydroxy amine via synthesis of bis(indolyl)methanes is a subject of continuous
epoxide ring opening and regenerating the catalyst−amine interest to synthetic organic/medicinal chemists as indole and
adduct A. their derivatives have versatile biological activities24 and are
Furthermore, to understand the superior activity of catalyst 1 widely present in various biologically active natural products.25
over catalyst 2, it was found by DFT calculation that the free To examine the effect of the cationic PLY-based molecule, the
energy of activation required for the rate-determining step
reaction of indole with 4-methoxybenzaldehyde (Scheme 2)
(shown in Figure 8) in the presence of catalyst 2 is higher in
under neat conditions and at room temperature was considered
energy than the respective barrier TS3 in the presence of
for a model study. An initial screening was done to obtain
catalyst 1 by 1.7 kcal/mol.
Thus, although catalyst 2 shows significant catalytic activity maximum conversion to the product in short period. The
compared to the uncatalyzed reaction, it is less efficient than
the catalyst 1. Hence, our computational study fully Scheme 2. Synthesis of Bisindolylmethanes in the Presence
substantiates the experimental findings. The enhanced activity of PLY Cations (1 and 2) as Organic Lewis Acid Catalysts
of catalyst 1 could be explained from a comparison of HOMO
energy of N-methylaniline with LUMO energies of both Lewis
acid catalysts 1 and 2. The LUMO of catalyst 1 lays 0.72 eV
above the HOMO of N-methylaniline, whereas for catalyst 2, it
is placed 1.07 eV above the HOMO of N-methylamine,
resulting in a larger HOMO−LUMO gap for catalyst 2 as
compared to catalyst 1. Such significant changes in the catalytic
activity were understood on the basis of the fact that the
substitution on PLY system can tune its LUMO energy. In
general, the substitution of an electron-withdrawing group
stabilizes the LUMO, whereas the substitution of an electron-
donating group can lead to the reverse effect.
To establish that our concept of using the NBMO of the PLY
cation is not limited to only epoxide ring-opening reaction and
that the same concept can be extended and generalized for
more organic transformations, we tried to check the efficacy of
catalysts 1 and 2 in different organic transformations. We can
anticipate that NBMO of the PLY cation can also be able to
activate the aldehyde. Keeping this goal in mind, we were
4316 dx.doi.org/10.1021/cs5010695 | ACS Catal. 2014, 4, 4307−4319
ACS Catalysis Research Article
progress of the reactions was monitored by TLC taking into In these MCR reactions, we also observed the consistent
consideration the complete consumption of indole. The catalytic activity ordering between 1 and 2 when the reactions
reaction was completed in 40 min using catalyst 1. In a control were carried out under identical condition, which shows
experiment when the model reaction was performed in the catalyst 1 is more active in comparison to catalyst 2.
■
absence of catalyst, only 10% desired product was formed. This
observation clearly highlights the role of NBMO based catalyst CONCLUSIONS
for this chemical transformation.
To establish the generality of the organic Lewis acid In summary, we have demonstrated that the NBMOs of the
catalyzed bisindolylmethane formation, various functionalized phenalenyl cation can be used as highly active Lewis acid
aldehydes were treated with indole (Scheme 2). The reaction is catalysts for organic transformations such as the epoxide ring-
compatible with a variety of functional groups as well as with opening reaction with amine in the present case. The chemistry
heterocyclic aldehyde. In all cases, the reaction proceeded of phenalenyl system started more than six decades back, and a
smoothly at room temperature to afford the corresponding number of phenalenyl-based neutral free radicals have already
bisindolylmethanes in excellent yields (85−95%). To compare been realized in the solid state; their promise as fascinating
the catalytic activity between the catalysts 1 and 2, few parallel materials has been very well documented. However, with this
experiments were performed. Here we also found that catalyst 1 study, we show that the empty NBMO of organic phenalenyl
is more effective than catalyst 2. cation can be utilized as Lewis acid catalyst. This work
Furthermore, we extend this concept of using NBMO based represents a major step in phenalenyl chemistry, dealing with
organic Lewis acid concept in multicomponent reactions cationic phenalenyl states to design molecular catalysts using its
(MCRs) as well. In recent years, several new criteria have easily accessible electron-acceptance capability into the empty
NBMO. This particular concept was also further successfully
been brought forward in the synthetic process to solve the
extended in different types of organic transformations, showing
challenging synthetic problem. Among them, MCRs26 have
the versatility of cataionic PLY as organic Lewis acid.
been recognized as a new synthetic support. The perpetual
Furthermore, we have shown that by chemically tuning the
interest of synthetic chemists toward the development of newer
energy of NBMO of PLY cation, one can alter the Lewis acidity,
methodologies for the synthesis of naphthylbenzamide27
and eventually, it can change the catalytic outcome of the
moiety relies on its broad range of biological activities.28 To
reaction. This phenomenon clearly describes the importance of
confirm the efficiency of the organic Lewis acid catalysts 1 and
the NBMO mediated organocatalysis.
■
2 for MCR transformation, we used a model reaction that
comprises a three-component reaction (3-MCR) involving an
benzaldehyde, β-napthol, and benzamide in the presence of EXPERIMENTAL SECTION
catalyst 1 (2.5 mol %) at 100 °C under neat condition (Scheme General Considerations. All solvents were distilled from
3). To our delight, the desired naphthylbenzamide was formed Na/benzophenone prior to use. All chemicals were purchased
in 95% yield after 1 h. However, in the absence of catalyst, very from Aldrich and Fluka Chemicals and used as received. The
1
poor conversion was observed. H and 13C NMR spectra were recorded on 400 and 500 MHz
Encouraged with this preliminary result, we applied this spectrometer in CD3CN/CDCl3/DMSO-d6 with residual
present methodology with the various functionalized aldehydes. undeuterated solvent (CD3CN: 1.94/1.3, 118.2, CDCl3: 7.26/
Irrespective of functionality, the final product was formed in 77.0, DMSO-d6: 2.5/39.5) as an internal standard. Chemical
excellent yield (Scheme 3). shifts (δ) are given in ppm and J values are given in Hz. The
HR-MS data were obtained using a Q-TOF Micromass. Open-
Scheme 3. Synthesis of Naphthylbenzamide in the Presence column chromatography and thin-layer chromatography (TLC)
of PLY-Based Organic Lewis Acid Catalysts 1 and 2 were performed on Silica gel [60−120 mesh]. Evaporation of
solvents was performed at reduced pressure using a rotary
evaporator.
Experimental Procedure for the Cyclic Voltammetry.
Cyclic Voltammetry was performed using a PAR potentiostat at
room temperature in dry acetonitrile under argon atmosphere
with n-Bu4NClO4 (0.1 M) as supporting electrolyte. Potentials
were scanned with respect to the quasi-reference electrode in a
single compartment cell fitted with Pt electrodes (working,
auxiliary and reference) and referenced to the Fc/Fc+ couple of
ferrocene at 0.38 V versus SCE. The Epa-Epc separation of the
reversible couples was within 10% of that of the Fc/Fc+ couple.
Typical Procedure for the Formation of β-Hydroxy
Amines (Table 2, Entry 1). A mixture of 2-(phenoxymethyl)-
oxirane 3a (0.30g, 2 mmol) and N-methylaniline 4a (0.214g, 2
mmol, 1 equiv) were stirred magnetically at 40 °C in the
presence of catalyst 1 (18 mg, 2.5 mol %). After 11 h, the
reaction mixture was diluted with EtOAc (5 mL) followed by
addition of water (2 mL). The supernatant EtOAc solution was
decanted. The aqueous layer/portion was extracted with EtOAc
(2 × 5 mL). The combined EtOAc extracts were dried
(Na2SO4) and concentrated under vacuum. The conversion of
the reaction was checked by 1H NMR spectroscopy. When the
4317 dx.doi.org/10.1021/cs5010695 | ACS Catal. 2014, 4, 4307−4319
ACS Catalysis
■
Research Article
crude product was estimated to be greater than 80% pure, flash AUTHOR INFORMATION
chromatography (1:0 to 7:3 hexane/EtOAc) was then Corresponding Authors
performed to get the isolated yield of the corresponding pure *E-mail: swadhin.mandal@iiserkol.ac.in.
β-hydroxy amines. *E-mail: spad@iacs.res.in.
Procedure for the Synthesis of Bisindolylmethane *E-mail: prasunchem@iiserkol.ac.in.
(Scheme 2). A mixture of aldehyde (2 mmol, 1 equiv) and
Notes
indole (0.464 g, 4 mmol, 2 equiv) was stirred at 40 °C in the
The authors declare no competing financial interest.
■
presence of catalyst 1 (18 mg, 2.5 mol %) for 40 min. After
completion of the reaction (TLC), the mixture was diluted with ACKNOWLEDGMENTS
water (2 mL) and EtOAc (5 mL). The supernatant EtOAc
solution was decanted off. The aqueous layer/portion was S.K.M. thanks SERB grant No. SR/S1/IC-25/2012 (DST),
extracted with EtOAc (2 × 5 mL). The combined EtOAc New Delhi for financial support. S.R.R. and A.P. thank CSIR
extracts were dried over Na2SO4. The organic phase was and SERB (DST) for RA fellowships, respectively. P.K.V.
concentrated in vacuo, and the purity of the residue was thanks UGC for a research fellowship. A.D. is grateful to the
checked by 1H NMR spectroscopy. Flash chromatography was DST and INSA for partial support. The authors also
acknowledge the NMR facility of IISER-Kolkata.
■
performed with eluant EtOAc/petroleum ether (1:9) unless
crude product was estimated to be greater than 95% pure.
Procedure for the Synthesis of Naphthylbenzamide
REFERENCES
(Scheme 3). A mixture of aldehyde (1 mmol, 1 equiv), 2- (1) Reid, D. H. Quart. Rev. 1965, 19, 274−302.
napthol (0.144 g, 1 mmol, 1 equiv), and benzamide (0.121 g, 1 (2) Reid, D. H. Tetrahedron 1958, 3, 339−352.
(3) (a) Haddon, R. C. Nature 1975, 256, 394−396. (b) Haddon, R.
mmol, 1 equiv) was stirred at 100 °C in the presence of catalyst C. Aust. J. Chem. 1975, 28, 2343−2351.
1 (9 mg, 2.5 mol %) for 1 h. After completion of the reaction (4) (a) Itkis, M. E.; Chi, X.; Cordes, A. W.; Haddon, R. C. Science
(monitored by TLC), the mixture was diluted with water (2 2002, 296, 1443−1445. (b) Pal, S. K.; Itkis, M. E.; Tham, F. S.; Reed,
mL) and EtOAc (5 mL). The supernatant EtOAc solution was R. W.; Oakley, R. T.; Haddon, R. C. Science 2005, 309, 281−284.
decanted. The aqueous layer/portion was extracted with EtOAc (c) Mandal, S. K.; Itkis, M. E.; Chi, X.; Samanta, S.; Lidsky, D.; Reed,
(2 × 5 mL). The combined EtOAc extracts were dried R. W.; Oakley, R. T.; Tham, F. S.; Haddon, R. C. J. Am. Chem. Soc.
(Na2SO4) and concentrated under vacuum. The crude reaction 2005, 127, 8185−8196. (d) Mandal, S. K.; Samanta, S.; Itkis, M. E.;
mixture was recrystallized from EtOH to obtain the pure pale Jensen, D. W.; Reed, R. W.; Oakley, R. T.; Tham, F. S.; Donnadieu, B.;
yellow solid product. Haddon, R. C. J. Am. Chem. Soc. 2006, 128, 1982−1994. (e) Bag, P.;
Computational Methods. The ground-state and tran- Pal, S. K.; Itkis, M. E.; Sarkar, S.; Tham, F. S.; Donnadieu, B.; Haddon,
R. C. J. Am. Chem. Soc. 2013, 135, 12936−12939.
sition-state structures were optimized using the hybrid meta (5) Miller, J. S. Angew. Chem., Int. Ed. 2003, 115, 27−29.
exchange correlation functional M06-2X with 6-31+G(d) basis (6) (a) Morita, Y.; Suzuki, S.; Sato, K.; Takui, T. Nat. Chem. 2011, 3,
set.29 Truhlar’s M06-2X functional has been successful for the 197−204. (b) Hicks, R. G. Nat. Chem. 2011, 3, 189−191.
calculation of accurate thermochemistry, reaction barriers, and (7) (a) Mukherjee, A.; Sen, T. K.; Ghorai, P. K.; Mandal, S. K.
noncovalent interactions present in the reactions involving Organometallics 2013, 32, 7213−7224. (b) Sen, T. K.; Mukherjee, A.;
main group elements.30 Vibrational frequencies were calculated Modak, A.; Mandal, S. K.; Koley, D. Dalton Trans. 2013, 42, 1893−
to ensure that the located minima and transition states were 1904. (c) Mukherjee, A.; Sen, T. K.; Ghorai, P. K.; Mandal, S. K. Sci.
characterized by zero and one imaginary frequencies Rep. 2013, 3, 2821. (d) Mukherjee, A.; Sen, T. K.; Ghorai, P. K.;
respectively and to obtain the zero point energy (ZPE) Samuel, P. P.; Schulzke, C.; Mandal, S. K. Chem.Eur. J. 2012, 18,
corrections to the energies. Relaxed intrinsic reaction 10530−10545. (e) Sen, T. K.; Mukherjee, A.; Modak, A.; Ghorai, P.
K.; Kratzert, D.; Granitzka, M.; Stalke, D.; Mandal, S. K. Chem.Eur.
coordinate scans were performed to confirm that the located J. 2012, 18, 54−58.
saddle points do connect the reactants and products (or (8) Raman, K. V.; Kamerbeek, A. M.; Mukherjee, A.; Atodiresei, N.;
intermediates).31 Basis set super position errors (BSSE) were Sen, T. K.; Lazic, P.; Caciuc, V.; Michel, R.; Stalke, D.; Mandal, S. K.;
calculated using a counterpoise correction scheme, whenever Blugel, S.; Munzenberg, M.; Moodera, J. S. Nature 2013, 493, 509−
necessary.32 All structural calculations were done using standard 513.
quantum chemistry programs.33 (9) (a) Ishihara, K. Lewis acids in organic synthesis; Yamamoto, H.,
Experimental Details on Spectroscopy. The steady-state Ed.; Wiley-VCH: Weinheim, Germany, 2000; Vol 1, p 135. (b) Carey,
absorption and emission spectra were recorded in a UV F. A.; Sundberg, R. J. Advanced Organic Chemistry: Part A: Structure
spectrophotometer (Cary100, Varian) and FluoroMax-3 and Mechanisms, 5th ed.; Springer: Berlin, 2007.
spectrofluorimeter, respectively. All experiments were carried (10) Sereda, O. Tabassum, S.Wilhelm, R. Lewis Acid Organocatalysts,
in Asymmetric Organocatalysis, Topic In Current Organic Chemistry; List,
out at room temperature (298 K). We have taken 25 μM B., Ed.; Springer: Berlin, 2009; pp 86−117.
concentration of catalyst 1 or 2 in each measurement.
■
(11) Bah, J.; Franzen, J. Chem.Eur. J. 2014, 20, 1066−1072.
(12) (a) Franz, K. D.; Martin, R. L. Tetrahedron 1978, 34, 2147−
ASSOCIATED CONTENT 2152. (b) Haddon, R. C.; Rayford, R.; Hirani, A. M. J. Org. Chem.
1981, 46, 4587−4588.
* Supporting Information
S
(13) (a) Corey, E. J.; Zhang, F. Angew. Chem., Int. Ed. 1999, 38,
The following files are available free of charge on the ACS 1931−1934. (b) O’Brien, P. Angew. Chem., Int. Ed. 1999, 38, 326−329.
Publications website at DOI: 10.1021/cs5010695. (c) Johannes, C. W.; Visser, M. S.; Weatherhead, G. S.; Hoveyda, A. H.
Experimental procedure, spectroscopic data, and scanned J. Am. Chem. Soc. 1998, 120, 8340−8347. (d) Ager, D. J.; Prakash, I.;
Schaad, D. R. Chem. Rev. 1996, 96, 835−876. (e) Li, G.; Chang, H. T.;
spectra; Cartesian coordinates, vibrational frequencies, Sharpless, K. B. Angew. Chem., Int. Ed. 1996, 35, 451−454.
and energies for all of the optimized geometries; and (14) (a) Posner, G. H.; Rogers, D. Z. J. Am. Chem. Soc. 1977, 99,
additional results from the spectroscopy and calculations 8208−8214. (b) Carre, M. C.; Houmounou, J. P.; Caubere, P.
(PDF). Tetrahedron Lett. 1985, 26, 3107−3110. (c) Cossy, J.; Bellosta, V.;
Hamoir, C.; Desmurs, J. R. Tetrahedron Lett. 2002, 43, 7083−7086. (25) Sundberg, R. J. The Chemistry of Indoles; Academic Press: New
(d) Papini, A.; Ricci, I.; Taddei, M.; Seconi, G.; Dembach, P. J. Chem. York, 1996.
Soc., Perkin Trans. 1984, 2261−2265. (e) Fujiwara, M.; Imada, M.; (26) (a) Corey, E. J.; Clark, D. A.; Goto, G.; Marfat, A.; Mioskowski,
Baba, A.; Matsuda, H. Tetrahedron Lett. 1989, 30, 739−740. (f) Fan, R. C.; Sameulson, B.; Hammerson, S. J. Am. Chem. Soc. 1980, 102, 1436−
H.; Hou, X. L. J. Org. Chem. 2003, 68, 726−730. (g) Chakraborti, A. 1439. (b) Luly, J. R.; Yi, N.; Soderquist, J.; Stein, H.; Cohen, J.; Perun,
K.; Rudrawar, S.; Kondaskar, A. Org. Biomol. Chem. 2004, 2, 1277− T. J.; Plattner, J. J. J. Med. Chem. 1987, 30, 1609−1616. (c) Hashiyama,
1280. (h) Bonollo, S.; Fringuelli, F.; Pizzo, F.; Vaccaro, L. Synlett 2008, T. Med. Res. Rev. 2000, 20, 485−501. For review: (d) Domling, A.
10, 1574−1578 DOI: 10.1055/s-2008-1078410. Chem. Rev. 2006, 106, 17−89. (e) Zhu, J.Multicomponent Reactions;
(15) (a) Yamada, J. I.; Yumoto, M.; Yamamoto, Y. Tetrahedron Lett. Bienayme, H., Ed.; Wiley-VCH: Weinheim, Germany, 2005. (f) Wipf,
1989, 30, 4255−4258. (b) Yamamoto, Y.; Asao, N.; Meguro, M.; P.; Stephenson, C. R. J.; Okumura, K. J. Am. Chem. Soc. 2003, 125,
Tsukade, N.; Nemoto, H.; Adayari, N.; Wilson, J. G.; Nakamura, H. J. 14694−14695.
Chem. Soc. Chem. Commun. 1993, 1201−1203. (c) Chini, M.; Crotti, (27) (a) Das, B.; Laxminarayana, K.; Ravikanth, B.; Rao, B. R. J. Mol.
P.; Favero, L.; Machhia, F.; Pineschi, M. Tetrahedron Lett. 1994, 35, Catal. A: Chem. 2007, 261, 180−183. (b) Nagarapu, L.; Baseeruddin,
433−436. (d) Meguro, M.; Asao, N.; Yamamoto, Y. J. Chem. Soc., M.; Apuri, S.; Kantevari, S. Catal. Commun. 2007, 8, 1729−1734.
Perkin Trans. 1994, 1, 2597−2601. (e) Auge, J.; Leroy, F. Tetrahedron (c) Khodaei, M. M.; Khosropour, A. R.; Moghanian, H. Synlett 2006,
Lett. 1996, 37, 7715−7716. (f) Sekar, G.; Singh, V. K. J. Org. Chem. 6, 916−920. (d) Patil, S. B.; Singh, P. R.; Surpur, M. P.; Samant, S. D.
1999, 64, 287−289. (g) Sagava, S.; Abe, H.; Hase, Y.; Inaba, T. J. Org. Ultrason. Sonochem. 2007, 14, 515−518. (e) Lei, M.; Ma, L.; Hu, L.
Chem. 1999, 64, 4962−4965. (h) Pujala, B.; Rana, S.; Chakraborti, A. Tetrahedron Lett. 2009, 50, 6393−6397.
K. J. Org. Chem. 2011, 76, 8768−8780. (28) (a) Juaristi, E. Enantioselective Synthesis of β-Amino Acids; Wiley:
(16) (a) Bonnet-Delpon, D.; Begue, J.-P. J. Org. Chem. 2000, 65, New York, 1997. (b) Knapp, S. Chem. Rev. 1995, 95, 1859−1876.
6749−6751. (b) Yadav, J. S.; Reddy, B. V. S.; Basak, A. K.; Narsaiah, A. (c) Wang, Y.-F.; Izawa, T.; Kobayashi, S.; Ohno, M. J. Am. Chem. Soc.
V. Tetrahedron Lett. 2003, 44, 1047−1050. (c) Surendra, K.; 1982, 104, 6465−6466. (d) Dingermann, T.; Steinhilber, D.; Folkers,
Krishnaveni, N. S.; Rao, K. R. Synlett 2005, 506−510 G.Front Matter in Molecular Biology in Medicinal Chemistry; Wiley-
DOI: 10.1055/s-2005-862359. (d) Azizi, N.; Saidi, M. R. Org. Lett. VCH: Weinheim, Germany, 2004. (e) Shen, A. Y.; Tsai, C. T.; Chen,
2005, 7, 3649−3651. C. L. Eur. J. Med. Chem. 1999, 34, 877−882. (f) Shen, A. Y.; Chen, C.
(17) Sarkar, A.; Tham, F. S.; Haddon, R. C. J. Mater. Chem. 2011, 21, L.; Lin, C. I. Chin. J. Physiol. 1992, 35, 45−52.
1574−1581. (29) (a) Zhao, Y.; Schultz, N. E.; Truhlar, D. G. J. Chem. Theory
(18) Chi, X.; Itkis, M. E.; Patrick, B. O.; Barclay, T. M.; Reed, R. W.; Comput. 2006, 2, 364−382. (b) Zhao, Y.; Truhlar, D. G. Theor. Chem.
Oakley, R. T.; Cordes, A. W.; Haddon, R. C. J. Am. Chem. Soc. 1999, Acc. 2008, 120, 215−241. (c) Hariharan, P. C.; Pople, J. A. Theor.
121, 10395−10402. Chim. Acta 1973, 28, 213−222.
(19) (a) Lakowicz, J. R. Principles of Fluorescence Spectroscopy, 3rd ed.; (30) (a) Jissy, A. K.; Datta, A. J. Phys. Chem. Lett. 2013, 4, 1018−
Plenum Press: New York, 1999. (b) Egli, M.; Sarkhel, S. Acc. Chem. 1022. (b) Nijamudheen, A.; Jose, D.; Shine, A.; Datta, A. J. Phys. Chem.
Res. 2007, 40, 197−205. (c) Mooibroek, T. J.; Gamez, P.; Reedijk, J. Lett. 2012, 3, 1493−1496. (c) Pieniazek, S. N.; Clemente, F. R.; Houk,
CrystEngComm. 2008, 10, 1501−1515. K. N. Angew. Chem., Int. Ed. 2008, 47, 7746−7749. (d) Wheeler, S. E.;
(20) Sharma, U.; Togati, N.; Maji, A.; Manna, S.; Maiti, D. Angew. Houk, K. N.; Schleyer, P. V. R.; Allen, W. D. J. Am. Chem. Soc. 2009,
Chem., Int. Ed. 2013, 52, 12669−12673. 131, 2547−2560.
(21) For examples of proton transfer mechanisms assisted by (31) (a) Fukui, K. Acc. Chem. Res. 1981, 14, 363−368. (b) Hratchian,
counterion or other cooperative mechanisms, see (a) Dudnik, A. S.; H. P.; Schlegel, H. B. J. Chem. Phys. 2004, 120, 9918−9924.
Xia, Y.; Li, Y.; Gevorgyan, V. J. Am. Chem. Soc. 2010, 132, 7645−7655. (32) Boys, S. F.; Bernardi, F. Mol. Phys. 1970, 19, 553−566.
(b) Xia, Y.; Dudnik, A. S.; Gevorgyan, V.; Li, Y. J. Am. Chem. Soc. (33) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
2008, 130, 6940−6941. (c) Shao, Z.; Zhang, H. Chem. Soc. Rev. 2009, Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
38, 2745−2755. (d) Masuda, Y.; Mori, Y.; Sakurai, K. J. Phys. Chem. A B.; Petersson, G. A.; et al. Gaussian 09; Gaussian, Inc.: Wallingford,
2013, 117, 10576−10587. (e) Nijamudheen, A.; Jose, D.; Datta, A. J. CT, 2009.
Phys. Chem. C 2011, 115, 2187−2195. (f) Patil, N. T.; Nijamudheen,
A.; Datta, A. J. Org. Chem. 2012, 77, 6179−6185.
(22) The calculations predict that the formation of 7 and 8 (in Figure
7) are endothermic processes and the reverse reactions to go back to
the weakly bound noncovalent adduct 6 should be very fast (ΔG‡ =
0.5 and 4.8 kcal/mol only). Therefore, in a mixture of 1 and 4a, the
much smaller populations of the structures 7 and 8 compared to 6
makes the detection of 7 and 8 by spectroscopy intractable.
(23) (a) Chakraborti, A. K.; Raha Roy, S.; Kumar, D.; Chopra, P.
Green Chem. 2008, 10, 1111−1118. (b) Wang, S. Y.; Ji, S.-J. Synth.
Commun. 2008, 38, 1291−1298. (c) Kamble, V. T.; Kadama, K. R.;
Joshia, N. S.; Muleya, D. B. Catal. Commun. 2007, 8, 498−502.
(d) Firouzabadi, H.; Iranpoor, N.; Jafari, A. A. J. Mol. Catal. A: Chem.
2006, 244, 168−172. (e) Deb, M. L.; Bhuyan, P. J. Tetrahedron Lett.
2006, 47, 1441−1443. (f) Wang, L.-M.; Han, J.-W.; Tian, H.; Sheng,
J.; Fan, Z.-Y.; Tang, X.-P. Synlett 2005, 2, 337−340. (g) Zhang, Z. H.;
Yin, L.; Wang, Y.-M. Synthesis 2005, 12, 1949−1954 DOI: 10.1055/s-
2005-869959. (h) Mo, L.-P.; Chuan, Z.; Zhang, Z.-H. Synth. Commun.
2005, 35, 1997−2004. (i) Singh, P. R.; Singh, D. U.; Samant, S. D.
Synth. Commun. 2005, 35, 2133−2138. (j) Ji, S.-J.; Wang, S.-Y.; Zhang,
Y.; Loh, T.-P. Tetrahedron 2004, 60, 2051−2055.
(24) (a) Casapullo, A.; Bifulco, G.; Bruno, I.; Riccio, R. J. Nat. Prod.
2000, 63, 447−451. (b) Garbe, T. R.; Kobayashi, M.; Shimizu, N.;
Takesue, N.; Ozawa, M.; Yukawa, H. J. Nat. Prod. 2000, 63, 596−598.
(c) Bao, B.; Sun, Q.; Yao, X.; Hong, J.; Lee, C. O.; Sim, C. J.; Im, K. S.;
Jung, J. H. J. Nat. Prod. 2005, 68, 711−715.
pubs.acs.org/JPCC
Room temperature ionic liquids (RTILs) have obtained a bonding.11,12 Neutron scattering as well as simulation studies
significant place as a class of nonvolatile environmentally green have shown that RTILs do possess local mesoscopic
solvents; however, what kind of nanoenvironment a molecular heterogeneities, i.e., nanoscale segregation of local structures
probe experiences while being dissolved in an RTIL is still not is prevailing, and the size of the nonpolar domain increases with
known. RTILs are composed of mostly an organic cation and increase in alkyl chain length of the imidazole ring.13−24
an inorganic anion.1−7 Stupendous interest in RTILs has been RTILs are highly polar in nature, and the polarity is in the
generated because of the interesting properties like vanishing range of short to long chain alcohol.25,26 However, there exists
vapor pressure, low melting point, high thermal stability, report of much lower polarity of RTILs from the dielectric
inflammability, recyclability, high ionic conductivity, etc.1−7 Out constant point of view.27−29 From the optical spectroscopic
of different RTILs, those based on imidazolium cationic moiety studies it has been shown that the absorption of RTILs is due
are maximum explored. Thus, RTILs have made its place in to imidazolium moiety, and by using different control
synthesis, (bio)catalysis, material sciences, and chemical experiments, it has been shown beyond any scientific doubt
engineering, specially in energy related applications, such as that the absorption as well as emission from RTILs are inherent
batteries, fuel cell photovoltaics, supercapacitors, etc.1−7 RTILs to RTILs and not due to any impurity.30,31 Proper rigorous
are different from conventional solvents in many respects. For purification procedure needs to be followed in order to remove
example, unlike conventional solvents, RTILs possess structural
the impurity. The purity of thus produced RTIL is confirmed
nanodomains both in solid and liquid states.8−24 From crystal
by NMR and MS measurements. As can be seen (see
structure studies it has been shown that there exists
nanostructural heterogeneities in neat RTILs. These spatial Supporting Information), the RTILs produced are of very
heterogeneities can range up to a few nanometers, and the high purity grade and do not contain any impurity (see
magnitude depends on the alkyl chain length of the imidazole discussion part for detail). It has been reported that neat RTILs
ring.8−10 Existence of π−π interaction and intermolecular exhibit excitation wavelength-dependent emission behavior, a
hydrogen bonding between imidazole rings have been phenomenon commonly known as red edge effect (REE).30,31
documented.8 It has also been reported that the hydrogen
atom attached to the carbon atom located in between two Received: February 7, 2014
nitrogen atoms of the imidazole ring can take part in hydrogen Published: February 11, 2014
© 2014 American Chemical Society 5051 dx.doi.org/10.1021/jp501342g | J. Phys. Chem. C 2014, 118, 5051−5057
The Journal of Physical Chemistry C Article
This unusual behavior has been assigned to the existence of the Chart 1. Chemical Structures of RTILs and the 4NBD Dye
different energetically associated forms and to the lack of
solvation/energy transfer between those species. It has also
been shown that some dipolar solutes do show excitation
wavelength-dependent emission while being dissolved in an
RTIL.32−34 Incomplete solvation or the existence of specific
interaction (like H-bonding) between the dipolar probe and
RTIL has been speculated to be responsible for the observed
REE for the dipolar solutes.32−34 Fluorescence dynamical
studies have shown that solvation34−37 in RTILs is biphasic in
nature, one component is typically of a few hundred
picoseconds and the other component is in the range of a
few nanoseconds. However, there exists different views
regarding the mechanism (especially regarding the faster
component) of solvation in different RTILs.34−37 Recently, it
has been shown that neat RTILs do exhibit excitation as well as
monitoring wavelength-dependent fluorescence decay behav-
ior.38 However, no rise time in picosecond or longer time
regime could be observed.38 This proves that no energy transfer
among different nanodomains take place in picosecond or
longer time scale. Thus, the reason for the observation of red
edge effect in neat RTILs could be understood. wavelengths, viz. 377 and 402 nm. In addition to those
Although a lot of photophysical measurements have been elaborate and rigorous experiments we have also performed
performed in RTILs, however, experiments based on resonance acceptor concentration-dependent RET dynamics in order to
energy transfer (RET) have not been explored in great verify whether RET dynamics depends on the concentration of
detail.39−42 Specially the dynamics of RET in neat RTILs has the acceptor. Results from all these experiments have been
not been explored at all. RET is a phenomenon in which the elaborated. On the basis of all these rigorous experimental
excitation energy of the donor is nonradiatively transferred to results a model has been put forward that could explain the
the acceptor provided some electrodynamic conditions are nanoscopic environment experienced by a neutral probe while
fulfilled and donor and acceptor are in close proximity (20 to being dissolved in an RTIL.
200 Å).43−45 The acceptor should be chosen in such a way that Steady-state absorption and emission: Absorption and
there exists significant spectral overlap between emission emission spectra of hmim[FAP] and that of neutral dye
spectrum of donor and the absorption spectrum of acceptor 4NBD have been shown in Supporting Information The
,and the extent of direct excitation of the acceptor is minimal. absorption and emission spectra of RTIL (donor) are well
The application of RET phenomenon is that it enables us to separated from that of the acceptor (4NBD). Absorbance at the
measure the distance between donor and acceptor in excitation wavelengths (377 and 402 nm) of the acceptor
heterogeneous matrices and environments like live cell, etc., (4NBD) is very low compared to that of the donor (RTIL)
in a noninvasive way. It has been shown recently41 that RTILs (see Supporting Information). This is a prerequisite condition
can engage in RET. We have explored RET to probe the in order to keep direct excitation of the acceptor at the minimal
distance between the imidazolium moiety (donor) and the level. In order to observe RET between any donor−acceptor
neutral dye (as acceptor), thus having an insight regarding the hybrid, the emission spectrum of the donor should have
nanoscopic environment that a neutral probe experiences in an spectral overlap with the absorption spectrum of the accept-
RTIL. or.43−45 It is evident from Figure 1 that there is significant
Hereby in this article we report the RET phenomenon in spectral overlap between the emission spectrum of RTIL
neat RTILs and have measured the dynamics of RET using
FAP, PF6, and BF4 anion based imidazolium RTIL as donor
and 4NBD (neutral dye) as an acceptor (see Chart 1). The
choice of the dye stems from the fact that the direct excitation
of the dye at 377 and 402 nm is minimal (see Supporting
Information).46 In order to check whether RET dynamics
depends on the alkyl chain length of the cation we have carried
out experiments in FAP anion based RTILs with three different
alkyl chain length of the cation, viz., ethyl, butyl, and hexyl. In
order to check whether the RET dynamics depend on the
nature of anion we have performed experiments in three RTILs
with varying anions, viz. FAP, BF4, and PF6 anion with same
cation (hmim). There exists a few literature reports that have
shown that different spectral and temporal fluorescence
behavior of neat RTILs or dissolved solutes do depend on
excitation wavelength or on the chain length of the cation or
the nature of the anion. In order to check whether the
dynamics of RET depends on the excitation wavelength, we Figure 1. Emission spectrum of hmim[PF6] (λex = 377 nm) (black, ---)
have performed the experiments for two different excitation and absorption spectrum of 4NBD (red, ).
Table 1. Time Constants of Fluorescence Decay of the Donor (hmim[FAP]) in the Presence of Acceptor (4NBD) (Different
Concentrations)
λex (nm) λem (nm) concentration (μM) τ1 (ns) B1 τ2 (ns) B2 τ3 (ns) B3 ⟨τ⟩ (ns) χ2
377 450 0 0.26 11.56 2.63 50.12 8.95 38.33 4.82 1.03
176 0.78 14.59 2.56 45.93 8.26 39.48 4.55 1.18
220 0.26 6.21 1.97 44.57 7.08 49.22 4.37 1.14
264 0.28 7.23 1.97 47.49 7.30 45.27 4.26 1.15
352 0.41 10.07 2.13 47.24 7.45 42.69 4.22 1.10
402 450 0 0.26 9.08 2.25 50.57 7.07 40.35 4.01 1.13
176 0.27 11.06 2.09 52.79 6.89 36.14 3.62 1.15
220 0.23 11.19 1.94 50.87 6.43 37.93 3.45 1.19
264 0.24 11.28 1.85 49.50 6.34 37.44 3.31 1.14
352 0.21 11.05 1.78 49.53 5.95 39.42 3.25 1.15
λex (nm) RTILs τ0D (ns) ϕD J(λ) (1015) (M−1 cm−1 nm4) τrise (ns) R0 (Å) RDA (Å) KFRET (108 s−1) E
377 emim[FAP] 8.48 0.21 1.19 3.00 39.68 33.37 3.33 73.45
bmim[FAP] 8.61 0.24 1.03 3.73 39.54 34.39 2.68 69.79
hmim[FAP] 4.82 0.23 0.68 3.97 37.48 36.30 2.51 54.79
hmim[PF6] 4.33 0.08 1.02 4.00 32.73 31.93 2.50 58.13
hmim[BF4] 4.87 0.09 0.91 3.19 32.68 30.35 3.13 61.34
402 emim[FAP] 6.37 0.13 1.42 3.04 37.64 33.27 3.28 67.71
bmim[FAP] 7.30 0.19 1.15 3.72 38.75 34.63 2.68 66.25
hmim[FAP] 4.02 0.18 0.87 4.00 36.48 36.48 2.50 50.00
hmim[PF6] 5.13 0.08 1.15 4.00 33.33 31.80 2.50 57.14
hmim[BF4] 6.14 0.10 0.96 3.44 33.43 30.35 2.90 64.10
the distance between donor and acceptor have been calculated acceptor is expected to decrease and hence the energy transfer
(using the rise time) following eq 3. time. However, in the present case the energy transfer time has
been found to be dependent on the nature of the cation and the
1 ⎛ R0 ⎞
6
1 anion, but independent of the concentration. This means that
kFRET = A = 0⎜ ⎟ with an increase in acceptor concentration, the distance
τrise τD ⎝ RDA ⎠ (3)
between D−A pair remains same. In order to explain these
where τArise is the rise time of the acceptor. τ0D is the lifetime of experimental observations we have put forward a model shown
the donor in the absence of any acceptor. R0 and RDA are the in Scheme1.
Förster distance and the actual distance between the donor and
the acceptor. The magnitude of J(λ), R0, KFRET, RDA, etc., have Scheme 1. Nanocage Assembly with Varying Anions and the
been depicted in Table 3. Acceptor Inside the Nanocage
As can be seen from Table 3, the magnitude of the spectral
overlap (and Förster radius R0) are dependent on the nature of
the RTIL. For RTILs having the same cation with different
anions, the value of the spectral overlap has varied. The value of
spectral overlap has varied for RTIL having same anion but
varied cation chain length. Moreover, the magnitude of RET
time (τrise) (or the rate constant of RET) has been found to
vary for different RTILs with the same cation and different
anions as well as for RTILs with the same anion but different
cations. The concentration-dependent change in RET dynamics is
Using Förster formulation we have calculated the donor− true for a homogeneous media. However, RTILs are known to
acceptor distance (RDA). The magnitude of RDA has been found form nanocages, and the size of the nanocages has been shown
to be dependent on the nature of the RTIL. With the same to extend up to a few nanometers depending on size of the alkyl
anion (FAP), the magnitude of RDA has been found to increase chain length of the cation.15,17,18,20,22 According to our model
as the chain length of the cation increases. When the chain (see Scheme 1), cation and anion of the RTIL remain side by
length of the cation increases from ethyl to hexyl, the side, and the tail of the cation remains inside the cage.
magnitude of RDA has been found to increase from 33.37 to Formation of different nanocages within RTIL has been
36.30 Å. Moreover, with the same cation hmim, as the size of invoked earlier.15,17,18,20,22 It has been shown earlier that the
the anion increases from BF4 to PF6 to FAP the magnitude of emission of the RTIL is because of the imidazolium moiety.30,31
RDA has been found to increase from 30.54 to 36.30 Å. Thus, As the tail of the cation remains inside the cage, thus the
the magnitude of RDA has been shown to be dependent both on distance between the donor (imidazolium ring) and the
the nature of the cation (alkyl chain length) and size of the acceptor increases with increase of the alkyl chain length.
anion of the RTIL. Thus, for a particular anion, the distance between the donor
The magnitude of the rise time or the rate constant of energy and the acceptor should increase with an increase in the chain
transfer for a particular RTIL has been noted to be same for length. This is exactly what we have observed experimentally.
402 nm excitation wavelength in comparison to 377 nm as Moreover, as the size of the anion increases from BF4 to PF6 to
excitation wavelength. Different photophysical processes in a FAP (for the same cation, hmim) the bulkier the anion more
particular RTIL have been shown to be dependent on the will be the cation pushed outward (Scheme 1), and thus, the
excitation wavelength.32−34 However, a recent communication distance between the donor (imidazolium ring) and the
has shown that dynamical fluorescence behavior of a acceptor will increase with increase in size of the anion. This
fluorophore (dissolved in an RTIL), whose emission is beyond is exactly what we have observed experimentally. We have put
that of the RTIL, is excitation wavelength independent.38 In the forward a model (Scheme 1) to explain the experimentally
present report, we have 4NBD whose emission is beyond that observed phenomenon.
of the RTIL, and the dynamics of RET process has been found According to our model (Scheme 1), 4NBD (acceptor) will
to be excitation wavelength independent. go to these nanocages, and since the concentration of neat
Quite interestingly for different excitation wavelengths for a RTIL (∼M) is a few order higher than the concentration of the
particular RTIL, the magnitude of donor−acceptor distance dye (∼a few hundred μM), with increase in acceptor
(RDA) has been found to be exactly the same. No matter what concentration the dye will go to different nanocages, and as a
methodology we use, the donor−acceptor distance (RDA) result, the distance between donor and acceptor will remain the
should be fixed and independent of the method we employ. For same for a particular RTIL-4NBD pair. Hence, the rise time or
different excitation wavelengths the magnitude of spectral rate constant of the RET phenomenon should remain the same
overlap (J(λ)) as well as Förster distance (R0) are different as even with an increase in concentration of the acceptor. This is
the emission of these RTILs is excitation wavelength what we have observed experimentally. Thus, rigorous
dependent. However, using Förster formulation, even for experiments provide evidence that nanocages do form in
different excitation wavelengths, the magnitude of donor− RTIL surrounding the acceptor. Thus, the optical spectroscopic
acceptor distance (RDA) has been found to be exactly the same results mentioned in the manuscript help us understand the
for a particular RTIL. This observation points to the fact that nanoscopic environment that the probe experiences inside
perhaps Förster formulation can explain the mechanism of RET RTIL. This first report of dynamics of RET in neat RTIL as
dynamics in RTILs. donor and a neutral dye as acceptor will prompt similar other
However, what is still intriguing is that the RET dynamics experiments using cationic and anionic dyes. These experiments
(rise time) is independent of concentration. With an increase in are currently underway, and very soon the results of these
concentration of the acceptor the distance between donor and investigations will be reported.
5055 dx.doi.org/10.1021/jp501342g | J. Phys. Chem. C 2014, 118, 5051−5057
The Journal of Physical Chemistry C
■
Article
■
*
ASSOCIATED CONTENT
S Supporting Information
Simulations. J. Phys. Chem. B 2008, 112, 8049−8056.
(14) Hardacre, C.; Holbrey, J. D.; Mullan, C. L.; Youngs, T. G. A.;
Bowron, D. T. Small Angle Neutron Scattering from 1-Alkyl-3-
methylimidazolium Hexafluorophosphate Ionic Liquids ([Cnmim]-
Experimental details, characterization of synthesized RTILs, [PF6], n = 4, 6, and 8). J. Chem. Phys. 2010, 133, 074510.
steady-state absorption, fluorescence, time-resolved fluores- (15) Triolo, A.; Russina, O.; Bleif, H.-J.; Di Cola, E. Nanoscale
cence decay, time constants of fluorescence decay, Stern− Segregation in Room Temperature Ionic Liquids. J. Phys. Chem. B
Volmer plot, etc. This material is available free of charge via the 2007, 111, 4641−4644.
Internet at http://pubs.acs.org. (16) Urahata, S. M.; Ribeiro, M. C. C. Structure of Ionic Liquids of 1-
■
Alkyl-3-methylimidazolium Cations: A Systematic Computer Simu-
lation Study. J. Chem. Phys. 2004, 120, 1855−1863.
AUTHOR INFORMATION (17) Wang, Y.; Voth, G. A. Unique Spatial Heterogeneity in Ionic
Liquids. J. Am. Chem. Soc. 2005, 127, 12192−12193.
Corresponding Author (18) Lopes, J. N. C.; Gomes, M. F. C.; Padua, A. A. H. Nonpolar,
*(P.K.M.) E-mail: prasunchem@iiserkol.ac.in. Polar, and Associating Solutes in Ionic Liquids. J. Phys. Chem. B 2006,
Notes 110, 16816−16818.
(19) Borodin, O.; Smith, G. D. Structure and Dynamics of N-Methyl-
The authors declare no competing financial interest.
■
N-propylpyrrolidinium Bis(trifluoromethanesulfonyl)imide Ionic
Liquid from Molecular Dynamics Simulations. J. Phys. Chem. B
ACKNOWLEDGMENTS 2006, 110, 11481−11490.
(20) Jiang, W.; Wang, Y.; Voth, G. A. Molecular Dynamics
P.K.M. thanks IISER-Kolkata for financial help and instrumen- Simulation of Nanostructural Organization in Ionic Liquid/Water. J.
tal facilities. Support from the Fast-Track Project (SR/FT/CS- Phys. Chem. B 2007, 111, 4812−4818.
52/2011) of DST-India is gratefully acknowledged. A.G. thanks (21) Hu, Z.; Margulis, C. J. Room-Temperature Ionic Liquids: Slow
UGC, and T.C., D.R., and A.D. thank CSIR for their respective Dynamics, Viscosity, and the Red Edge Effect. Acc. Chem. Res. 2007,
Fellowships. 40, 1097−1105.
(22) Raabea, G.; Kö hler, J. Thermodynamical and Structural Showing Emissions by Excitation at Wide Wavelength Areas. Chem.
Properties of Imidazolium Based Ionic Liquids from Molecular Commun. 2010, 46, 6359−6361.
Simulations. J. Chem. Phys. 2008, 128, 154509. (42) Rao, V. G.; Mandal, S.; Ghosh, S.; Banerjee, C.; Sarkar, N. Study
(23) Andrade, J. D.; Böes, E. S.; Stassen, H. Liquid-Phase Structure of of Fluorescence Resonance Energy Transfer in Zweitter Ionic Micelle:
Dialkylimidazolium Ionic Liquids from Computer Simulations. J. Phys. Ionic Liquids-Induced Changes in FRET Parameters. J. Phys. Chem. B
Chem. B 2008, 112, 8966−8974. 2012, 121, 12021−12029.
(24) Yan, T.; Wang, Y.; Knox, C. On the Structure of Ionic Liquids: (43) Förster, T. Intermolecular Energy Migration and Fluorescence
Comparison between Electronically Polarizable and Nonpolarizable (Trans1 RS Knox). Ann. Phys. 1948, 2, 55−75.
Models. J. Phys. Chem. B 2010, 114, 6905−6921. (44) Valeur, B. Molecular Fluorescence Principles and Applications;
(25) Aki, S. N. V. K.; Brennecke, J. F.; Samanta, A. How Polar Are Wiley-VCH Verlag GmbH: Weinheim, Germany, 2002.
Room-Temperature Ionic Liquids? Chem. Commun. 2001, 413−414. (45) Lakowicz, J. R. Principles of Fluorescence Spectroscopy, 3rd ed.;
(26) Mandal, P. K.; Samanta, A. Fluorescence Studies in a Plenum Press: New York, 1999.
Pyrrolidinium Ionic Liquid: Polarity of the Medium and Solvation (46) Saha, S.; Samanta, A. Photophysical and Dynamic NMR Studies
Dynamics. J. Phys. Chem. B 2005, 109, 15172−15177. on 4-Amino-7-nitro-2-oxa-1,3-diazole Derivatives: Elucidation of the
(27) Wakai, C.; Oleinikova, A.; Weingaertner, H. How Polar Are Nonradiative Deactivation Pathway. J. Phys. Chem. A 1998, 102,
Ionic Liquids? Determination of the Static Dielectric Constant of an 7903−7912.
Imidazolium-Based Ionic Liquid by Microwave Spectroscopy. J. Phys. (47) Santhosh, K.; Patra, S.; Soumya, S.; Khara, D. C.; Samanta, A.
Chem. Lett. 2005, 109, 17028−17030. Fluorescence Quenching of CdS Quantum Dots by 4-Azetidinyl-7-
(28) Bright, F. V.; Baker, G. A. Comment on “How Polar Are Ionic nitrobenz-2-oxa-1,3-diazole: A Mechanistic Study. ChemPhysChem
Liquids? Determination of the Static Dielectric Constant of an 2011, 12, 2735−2741.
Imidazolium-Based Ionic Liquid by Microwave Dielectric Spectrosco-
py. J. Phys. Chem. B 2006, 110, 5822−5823.
(29) Wakai, C.; Oleinikova, A.; Weingaertner, H. How Polar Are
Ionic Liquids? Determination of the Static Dielectric Constant of an
Imidazolium-Based Ionic Liquid by Microwave Spectroscopy. Reply. J.
Phys. Chem. B 2006, 110, 5824−5824.
(30) Paul, A.; Mandal, P. K.; Samanta, A. How Transparent Are the
Imidazolium Ionic Liquids? A Case Study with 1-Methyl-3-
butylimidazolium Hexafluorophosphate, [bmim][PF6]. Chem. Phys.
Lett. 2005, 402, 375−379.
(31) Paul, A.; Mandal, P. K.; Samanta, A. On the Optical Properties
of the Imidazolium Ionic Liquids. J. Phys. Chem. B 2005, 109, 9148−
9153.
(32) Mandal, P. K.; Sarkar, M.; Samanta, A. Excitation Wavelength
Dependent Fluorescence Behavior of the Room Temperature Ionic
Liquids and Dissolved Dipolar Solutes. J. Phys. Chem. A 2004, 108,
9048−9053.
(33) Mandal, P. K.; Paul, A.; Samanta, A. Excitation-Wavelength-
Dependent Fluorescence Behavior of Some Dipolar Molecules in
Room-Temperature Ionic Liquids. J. Photochem. Photobiol. A: Chem.
2006, 182, 113−120.
(34) Samanta, A. Dynamic Stokes Shift and Excitation Wavelength
Dependent Fluorescence of Dipolar Molecules in Room Temperature
Ionic Liquids. J. Phys. Chem. B 2006, 110, 13704−13716.
(35) Mandal, P. K.; Saha, S.; Karmakar, R.; Samanta, A. Solvation
Dynamics in Room Temperature Ionic Liquids: Dynamic Stokes Shift
Studies of Fluorescence of Dipolar Molecules. Curr. Sci. 2006, 90,
301−310.
(36) Samanta, A. Solvation Dynamics in Ionic Liquids: What We
Have Learned from the Dynamic Fluorescence Stokes Shift Studies. J.
Phys. Chem. Lett. 2010, 1, 1557−1562.
(37) Zhang, X.-X.; Liang, M.; Ernsting, N. P. Complete Solvation
Response of Coumarin 153 in Ionic Liquids. J. Phys. Chem. B 2013,
117, 4291−4304.
(38) Ghosh, A.; Chatterjee, T.; Mandal, P. K. On the Heterogeneity
of Fluorescence Lifetime of Room Temperature Ionic Liquids: Onset
of a Journey for Exploring Red Emitting Dyes. Chem. Commun. 2012,
48, 6250−6252.
(39) Adhikari, A.; Das, D. K.; Sasmal, D. K.; Bhattacharyya, K.
Ultrafast FRET in a Room Temperature Ionic Liquid Microemulsion:
A Femtosecond Excitation Wavelength Dependent Study. J. Phys.
Chem. B 2009, 113, 3737−3743.
(40) Das, D. K.; Das, A. K.; Mondal, T.; Mandal, A. K.;
Bhattacharyya, K. Ultrafast FRET in Ionic Liquid-P123 Mixed
Micelles: Region and Counterion Dependence. J. Phys. Chem. B
2010, 114, 13159−13166.
(41) Izawa, H.; Wakizono, S.; Kadokawa, J. Fluorescence Resonance-
Energy-Transfer in Systems of Rhodamine 6g with Ionic Liquid
Three step chemical tweaking of nonfluorescent parent GFP chro- important towards bioimaging.12 There are quite a few number
mophore has yielded a three hundred times as bright coumarinic of photoactivable compounds, however, there is only one report
fluorophore. A fluorogenic compound has been prepared which so far where a photoactivable compound has been made with
after photo-uncaging yields the same highly bright fluorophore that GFP chromophore analogue.13
could be used for live stem cell imaging. In this manuscript we describe how three step chemical
tweaking of non-uorescent parent GFP chromophore (p-HBDI)
could lead to a highly uorescent coumarinic uorophore
The quest to enhance the uorescence quantum yield of (Scheme 1). Three modications that has been incorporated are
nonuorescent GFP (green uorescent protein)1,2 chromo- (i) p-hydroxy has been replaced by o-hydroxy, (ii) N,N-diethyl-
phores (p-HBDI, see Scheme 1) has engaged many researchers amino group has been introduced at the p-position, and (iii) sp3
from biology and chemistry for more than a decade. Strong nitrogen in imidazole ring has been replaced by oxygen. In this
deviation of uorescence quantum yield (f) of p-HBDI (f < direction we have made four compounds: 4-(2-hydroxy-4-N,
0.001) from that of wild type GFP (f 0.8) has been noted in N-diethylamino-benzylidene)-1,2-dimethyl-1H-imidazol-5(4H)-
solution.3,4 Different mechanisms have been put forward as one (OHIM), 4-(2-methoxy-4-N,N-diethylamino-benzylidene)-1,
plausible reasons for the low uorescence quantum yield of GFP
chromophore.5,6 Several successful attempts7,8 to increase f of
GFP chromophore by restricting the torsional motion include
mimicking GFP barrel encapsulation in human serum albumin,
in octaacid cavitand, and using truncated or split GFP tech-
nique, and selective complexation with RNA aptamers,
complexation with metal salts (such as Zn2+) or BF2 moiety.
However, concerns have been raised regarding their usage due
to lability of metal–BF2 complexes in solution.9 Different
chemical modications/substitutions in benzene and imidazole
ring of p-HBDI have been tried but so far a maximum of only
10% uorescence quantum yield could be achieved in GFP
chromophore analogues.10 Thus, the quest for chemically
modied GFP chromophore analogue with enhanced uores-
cence quantum yield is still very much active.
Photoactivation process provides spatial and temporal
control over the release of desired chemical11 and thus photo-
activable compounds and uorescent proteins are quite
This journal is ª The Royal Society of Chemistry 2013 RSC Adv., 2013, 3, 24021–24024 | 24021
View Article Online
Table 1 Photophysical properties of OHIM and cOHBO rearrangement, presence of –OH group at ortho position and
1 1
oxygen in oxazole ring are necessary conditions. For OHIM,
Dye Solvent lmaxa (nm) 3 (M cm ) f s
OMIM or OMBO these conditions are not fullled and thus no
OHIM Toluene 484(448) 29 075 2.2 103 0.76, 32.8 ps intramolecular rearrangement is possible. Thus, uorescence
Methanol 505(451) 30 580 1.4 103 0.73, 13.9 ps properties of OHIM, OMIM, and OMBO are similar to each
cOHBO Toluene 444(377) 19 030 0.90 3.27 ns other but quite different from that of cOHBO. Absorption
Methanol 485(392) 19 000 0.70 3.88 ns maximum of cOHBO matches with that of many other
a
Emission maximum, absorption maximum (nm) in parentheses. coumarinic probes that absorb at around 375–400 nm.14 The
best proof in support of coumarinic structure stems from the
This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.
single crystal X-ray structure of cOHBO (see Fig. 1). (For lattice
parameters etc. see ESI†).
Open Access Article. Published on 14 October 2013. Downloaded on 27/02/2017 10:12:14.
24022 | RSC Adv., 2013, 3, 24021–24024 This journal is ª The Royal Society of Chemistry 2013
View Article Online
This journal is ª The Royal Society of Chemistry 2013 RSC Adv., 2013, 3, 24021–24024 | 24023
View Article Online
Commun., 2010, 46, 5686–5688; (e) K. P. Kent and S. G. Boxer, 12 (a) G. H. Patterson and J. Lippincott-Schwartz, Science, 2002,
J. Am. Chem. Soc., 2011, 133, 4046–4052; (f) K. Do and 297, 1873–1877; (b) R. Ando, H. Hama, M. Yamamoto-Hino,
S. G. Boxer, J. Am. Chem. Soc., 2011, 133, 18078–18081; (g) H. Mizun and A. Miyawaki, Proc. Natl. Acad. Sci. U. S. A.,
A. Baldridge, S. Feng, Y. T. Chang and L. M. Tolbert, ACS 2002, 99, 12651–12656; (c) H. Tsutsui, S. Karasawa,
Comb. Sci., 2011, 13, 214–217; (h) J. S. Paige, K. Y. Wu and H. Shimizu, N. Nukina and A. Miyawaki, EMBO Rep., 2005,
S. R. Jaffrey, Science, 2011, 333, 642–646. 6, 233–238; (d) S. Habuchi, R. Ando, P. Dedecker,
8 M. S. Baranov, K. A. Lukyanov, A. O. Borissova, J. Shamir, W. Verheijen, H. Mizuno, A. Miyawaki and J. Hoens,
D. Kosenkov, L. V. Slipchenko, L. M. Tolbert, Proc. Natl. Acad. Sci. U. S. A., 2005, 102, 9511–9516; (e)
I. V. Yampolsky and K. M. Solntsev, J. Am. Chem. Soc., N. Gagey, M. Emond, P. Neveu, C. Benbrahim, B. Goetz,
This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.
2012, 134, 6025–6032. I. Aujard, J. B. Baudin and L. Jullien, Org. Lett., 2008, 10,
9 J. Kang, G. Zhao, J. Xu and W. Yang, Chem. Commun., 2010, 2341–2344; (f) E. Betzig, G. H. Patterson, R. Sougrat,
Open Access Article. Published on 14 October 2013. Downloaded on 27/02/2017 10:12:14.
24024 | RSC Adv., 2013, 3, 24021–24024 This journal is ª The Royal Society of Chemistry 2013
View Article Online / Journal Homepage / Table of Contents for this issue
www.rsc.org/chemcomm COMMUNICATION
On the heterogeneity of fluorescence lifetime of room temperature ionic
liquids: onset of a journey for exploring red emitting dyesw
Published on 01 May 2012. Downloaded by North Dakota State University on 22/10/2014 02:55:34.
An excitation and emission wavelength dependent non-exponential because of longer-time trapping of solutes in quasistatic local
fluorescence decay behaviour of room temperature ionic liquids solvent cages in comparison to their fluorescence lifetime.22,23
(RTILs) has been noted. Average fluorescence lifetimes have been Fluorescence decay of dissolved dyes (e.g. C153, AP etc.) has
found to vary by a factor of three or more. Red emitting dyes been reported to be non-exponential in nature. It was speculated
dissolved in RTILs are found to follow hitherto unobserved single that these dyes trapped in different solvent cages behave as
exponential fluorescence decay behaviour. different species and thus decay differently.21,24,25
The fluorescence quantum yield of imidazolium RTILs has
Room temperature ionic liquids (RTILs) have been considered been shown to be 0.003.19,20 This means that RTILs are very
to be green alternative solvents for various applications for over weakly fluorescent. However, while performing single molecule
a decade, owing to their better solvent properties in comparison spectroscopic studies with bright dyes it was observed that it is
to the volatile organic solvents.1–7 Coexistence of different types not possible to see single dyes unless we excite beyond 550 nm.
of interactions, such as coulombic, dipolar, van der Waals, and This surprising observation made us believe that although
H-bonding etc., makes these RTILs unique and complex in fluorescence quantum yield of RTILs is very low, significant
comparison to normal solvents.8–18 background fluorescence is generated as the number of fluorescing
Various optical spectroscopic studies as well as simulation species around the single dye is high.
analyses have shown the existence of heterogeneous cation–cation, In this communication, we report hitherto unknown monitoring
cation–anion mesoscopic local ordering in liquid state.8–18 Thus, wavelength dependent heterogeneity of the fluorescence lifetime of
RTILs are considered to be ‘‘nano-structured fluids’’.15 Optical most hydrophobic FAP anion based imidazolium RTILs (Chart 1).
studies on RTILs have shown that imidazolium-based RTILs We show spectroscopic evidence for the reason behind the
possess significant absorption in the UV region, extending into a observation of REE in neat RTILS. In contrast to previously
part of the visible region.19,20 Imidazolium RTILs are known to reported nonexponential decay behaviour of dyes in RTILs, we
exhibit excitation wavelength dependent emission behaviour.19,20 demonstrate and provide a possible reason for the single
It has been shown unambiguously that both absorption and exponential fluorescence decay behaviour of two red emitting
emission are inherent to RTIL itself and not due to any dyes (LD700 and OX725 (Chart 1)).
impurity.19,20 However, although there exists detailed investi- The steady state absorption behaviour of these RTILs resembles
gation on the spectral behaviour of RTILs, there is no that of other imidazolium based RTILs (see S1, ESIw). These
rigorous analysis of the fluorescence decay behaviour of neat RTILs also show an excitation wavelength dependent emission
RTILs. An unusual excitation wavelength dependent fluores- behaviour (see S2, ESIw) similar to what has been observed for
cence behaviour (known as Red Edge Effect or REE) has been other imidazolium based RTILs.19,20
reported for neat RTILs.19,20 REE has also been observed for However, monitoring the fluorescence decay of both RTILs
some dipolar solutes dissolved in RTILs.21 It has been speculated when excited at 375 nm, we have noticed hitherto unobserved
that different associated species are responsible for the observed monitoring wavelength dependent emission decay behaviour.
phenomenon, perhaps either because of inefficiency of the excita- As can be seen from Fig. 1, when the monitoring wavelength is
tion energy-transfer between them or an incomplete solvation changed from 390 nm to 440 nm, the fluorescence lifetime
process.19–21 It has also been mentioned that REE is observed increased (see Table 1) (for complete chart see S3, ESIw). With
a further increase in monitoring wavelength (to 660 nm),
6250 Chem. Commun., 2012, 48, 6250–6252 This journal is c The Royal Society of Chemistry 2012
View Article Online
differently.
Several fluorescent dyes have so far been explored in RTILs.24,25
Fluorescence decay behaviour of these dyes has been characterized
Fig. 1 Fluorescence decay curves of [emim][FAP]; lex = 377 nm. by a non-exponential decay function. It has been proposed that
Monitoring wavelengths are shown as inset. these dyes get trapped in different nanostructural cages and thus
decay differently. In our study, we have measured the fluorescence
Table 1 Monitoring wavelength dependent fluorescence lifetime of decay curve of two red emitting dyes namely LD700 and OX725
[emim][FAP]. lex = 377 nm in [emim][FAP] (see Fig. 2). Interestingly, we have observed
lem (nm) t1 (ns) (B1) t2 (ns) (B2) t3 (ns) (B3) hti w2 a single exponential fluorescence decay behaviour. This obser-
vation is in stark contrast to literature reports to date (e.g.
390 0.12 (5.42) 1.25 (29.81) 10.85 (64.77) 7.41 1.24
non-exponential decay of C153, AP, PRODAN etc.).
430 0.84 (19.31) 3.25 (25.22) 13.30 (55.44) 8.35 1.05
440 1.01 (18.52) 4.25 (27.35) 14.50 (54.13) 9.20 1.15 Decay curves of both red emitting dyes in other RTILs are
620 0.41 (13.36) 2.20 (36.98) 8.90 (49.66) 5.28 1.09 shown in the ESIw (see S6 and S7). From all these decay
660 0.45 (12.62) 1.75 (57.64) 6.10 (29.69) 2.88 1.04 curves, a clear single exponential decay behaviour is evident.
The decay time constants are given in S8 (ESIw). It might be
the average fluorescence lifetime continued to decrease. A argued that the excitation wavelength is 635 nm, which is
triexponential decay function was necessary to fit the decay much beyond the tail of absorption spectra of RTILs, so single
function at all wavelengths. Strikingly, a three-fold change in exponential decay behaviour is expected. However, in order to
the average fluorescence lifetime was noted on changing the prove our claim we have also used 377 nm and 402 nm as the
monitoring wavelength (Table 1). excitation wavelengths while keeping the monitoring wave-
We repeated the fluorescence decay measurements for both length the same. Even under these experimental conditions, we
RTILs by exciting at 402 nm. Even for this particular excita- have observed a clear single exponential decay for both dyes
tion wavelength, we observed a two-fold increase in average in both RTILs (see S9–S12, ESIw). Fluorescence lifetimes
lifetime (see S3, ESIw). A similar observation was noted for the for different excitation wavelengths are found to be almost
[hydemim][FAP] (see S4 and S5, ESIw). Thus, a non-exponential identical (S8, ESIw). This observation is in stark contrast to
fluorescence decay behaviour was quite evident for different what is currently known about the fluorescence decay behaviour
excitation wavelengths and for different monitoring wavelengths. of the dissolved dyes in RTILs. It is believed by the RTIL
Therefore, for both RTILs, a heterogeneity of fluorescence life-
time is clearly demonstrated.
We would like to report here another interesting observation.
While studying REE of RTILs,19,20 it was speculated that the lack
of solvation and energy transfer between the subensembles could
be the reason for the observation of REE. On closer look we
found no rise-time after fitting decay curves of neat RTILs. This
observation has been noted for both excitation wavelengths and
for monitoring wavelengths ranging from the lower wavelength
end through to the extreme red edge of the emission curve. This
phenomenon has been observed for both RTILs. This is a clear
indication of the lack of any excited state phenomenon (like
solvation) in the neat RTILs. Moreover, it also tells that there
exists no resonance energy transfer among various heterogeneous
nano-structured subensembles inherently present in the RTILs.
The time-resolution of our setup is about 70 ps so we could
safely mention that there is no solvation or energy transfer
process slower than 35 ps. However, in principle, an ultrafast
solvation with subpicosecond time constant cannot be over-
ruled. If there has been an ultrafast energy transfer between
subensembles in neat RTILs then we should have observed
similar decay time constants when fluorescence decays were
monitored at the same wavelength. However, that is not what we Fig. 2 Fluorescence decay behaviour of LD700 (top) and OX725
observed. Instead, we have found different decay time constants. (bottom) in [emim][FAP]. lex = 635 nm and lem = 650 nm.
This journal is c The Royal Society of Chemistry 2012 Chem. Commun., 2012, 48, 6250–6252 6251
View Article Online
community that subensembles of dyes associated with a neutral and ionic red emitting dyes in RTILs is currently
particular local nanoenvironment absorb and emit at different underway.
wavelengths. This was known to be the reason behind their Abbreviations used in the text: FAP, tris(pentafluoroethyl)-
nonexponential fluorescence decay behaviour. Similar results trifluorophosphate; RTIL, room temperature ionic liquid;
should have been observed in our case also. However, instead emim, 1-ethyl-3-methyl-imidazolium; hydemim, 1-(2-hydroxyl-
we have observed a clear single exponential decay of these two ethyl)-3-methyl-imidazolium; C153, coumarin 153; AP, 4-amino-
dyes in both RTILs for different excitation wavelengths. Thus, phthalimide; PRODAN, 6-propionyl-2-dimethylaminonaphthalene.
it is evident that it is not the excitation wavelength rather the PKM thanks IISER-Kolkata for financial help and instru-
Published on 01 May 2012. Downloaded by North Dakota State University on 22/10/2014 02:55:34.
monitoring wavelength which dictates the exponentiality of mental facilities. AG and TC thank UGC and CSIR for their
the fluorescence decay behaviour. respective Fellowship.
In order to understand this observation we look back at the
steady state optical behaviour of the dyes in RTILs. A closer Notes and references
look at the optical behaviour of most of these dyes (say, C153, 1 T. Welton, Chem. Rev., 1999, 99, 2071.
AP, PRODAN etc.)21,24,25 would reveal that these dyes have 2 P. Wasserscheid and W. Keim, Angew. Chem., Int. Ed., 2000,
an emission curve which overlaps with that of the RTILs. 39, 3772.
3 J. Dupont, R. F. D. Souza and P. A. Z. Suarez, Chem. Rev., 2002,
However, the absorption and emission spectra for LD700 and 102, 3667.
Oxazine725 are well beyond the tail of absorption and emission 4 R. D. Rogers and K. R. Seddon, Science, 2003, 302, 792.
curves of RTILs (see S13 and S14, ESIw). The sizes as well as 5 K. R. Seddon, Nat. Mater., 2003, 2, 363.
6 H. WeinGaertner, Angew. Chem., Int. Ed., 2008, 47, 655.
the electronic property of LD700 and Oxazine725 are not 7 J. P. Hallet and T. Welton, Chem. Rev., 2011, 111, 3508.
drastically different from those of Coumarin 153 or PRODAN. 8 J. Dupont, P. A. Z. Suarez, R. F. D. Souza, R. A. Burrow and
Thus, if Coumarin 153 or PRODAN can be trapped in J. Kintzinger, Chem.–Eur. J., 2000, 6, 2377.
nanostructural cages then LD700 and Oxazine725 could also 9 C. Chiappe and D. Pieraccini, J. Phys. Org. Chem., 2005, 18, 275.
10 Y. Wang and G. A. Voth, J. Am. Chem. Soc., 2005, 127, 12192.
go to similar cages. 11 J. N. C. Lopes and A. A. H. Padua, J. Phys. Chem. B, 2006,
Recent research on ionic dyes in RTILs has shown the 110, 3330.
existence of ionic probe–RTIL counterion interaction that is 12 J. N. C. Lopes, M. F. C. Gomes and A. A. H. Padua, J. Phys.
Chem. B, 2006, 110, 16816.
different from neutral probe–RTIL interaction.26–28 So it may
13 D. Jeong, Y. Shim, M. Y. Choi and H. J. Kim, J. Phys. Chem. B,
be surmised that the cationic dyes employed in this work 2007, 111, 4920.
would go to specific regions and thus produce only one type 14 L. M. N. B. F. Santos, J. N. C. Lopes, J. A. P. Coutinho, J. M. S. S.
of species and hence would show single exponential decay. Esperanca, L. R. Gomes, I. M. Marrucho and L. P. N. Rebelo, J. Am.
Chem. Soc., 2007, 129, 284.
However, earlier time-resolved fluorescence studies employing 15 K. Iwata, H. Okajima, S. Saha and H. Hamaguchi, Acc. Chem.
ionic dyes have shown a non/stretched exponential fluores- Res., 2007, 40, 1174.
cence decay.29,30 Emission spectra of all those ionic dyes have 16 Y. Wang, W. Jiang, T. Yan and G. A. Voth, Acc. Chem. Res.,
some magnitude of overlap with those of RTIL.29,30 2007, 40, 1193.
17 L. P. N. Rebelo, J. N. C. Lopes, J. M. S. S. Esperanca, H. J. R.
It is thus quite evident that, because of absence of perturbation Guedes, J. Lachwa, V. N. Visak and Z. P. Visak, Acc. Chem. Res.,
of RTIL emission, these two dyes do follow a single mode of 2007, 40, 1114.
fluorescence decay process. Thus, our observation points to the 18 A. A. H. Padua, M. F. C. Gomes and J. N. A. C. Lopes, Acc.
Chem. Res., 2007, 40, 1087.
fact that a reassignment for the nonexponential decay behaviour 19 A. Paul, P. K. Mandal and A. Samanta, Chem. Phys. Lett., 2005,
of dyes, while dissolved in RTILs, is necessary in light of the 402, 375.
present results. 20 A. Paul, P. K. Mandal and A. Samanta, J. Phys. Chem. B, 2005,
In conclusion, we have demonstrated hitherto unknown 109, 9148.
21 P. K. Mandal, M. Sarkar and A. Samanta, J. Phys. Chem. A, 2004,
excitation wavelength dependent fluorescence decay behaviour 108, 9048.
and hence a heterogeneity of fluorescence lifetime of neat RTILs. 22 Z. Hu and C. J. Margulis, Proc. Natl. Acad. Sci. U. S. A., 2006,
The evidence of the speculated reason for the observation of a 103, 831.
23 Z. Hu and C. J. Margulis, Acc. Chem. Res., 2007, 40, 1097.
red edge effect, i.e. the lack of solvation or energy transfer
24 A. Samanta, J. Phys. Chem. B, 2006, 110, 13704.
phenomenon among different nano-structured subensembles 25 A. Samanta, J. Phys. Chem. Lett., 2010, 1, 1557.
inherently present within RTILs, is established. We have 26 M. Muramats, Y. Nagasawa and H. Miyasaka, J. Phys. Chem. A,
shown a clear single exponential fluorescence decay behaviour 2011, 115, 3886.
27 H. Qiu, M. Takafuji, T. Sawada, X. Liu, S. Jiang and H. Ihara,
of red emitting dyes whose fluorescence is not perturbed by Chem. Commun., 2010, 46, 8740.
that of RTILs. We strongly believe that these observations will 28 S. Bruzzone, M. Malvaldi and C. Chiappe, J. Chem. Phys., 2008,
open a new domain of spectroscopic studies where perturbation 129, 074509.
29 N. Ito, S. Arzhantsev and M. Maroncelli, Chem. Phys. Lett., 2004,
of RTIL emission is no longer present on the fluorescence
396, 83.
processes of the probes. Work towards understanding dynamic 30 D. C. Khara and A. Samanta, Indian J. Chem., Sect. A, 2010,
phenomena like solvation, resonance energy transfer employing 49, 714.
6252 Chem. Commun., 2012, 48, 6250–6252 This journal is c The Royal Society of Chemistry 2012