Filipfgsftysdw 2016

Subscriber access provided by UNIV OF UTAH
Letter
Band Gaps of the Lead-Free Halide Double Perovskites
CsBiAgCl and CsBiAgBr from Theory and Experiment
2 6 2 6
Marina R Filip, Samuel Hillman, Amir-Abbas Haghighirad, Henry J. Snaith, and Feliciano Giustino
J. Phys. Chem. Lett., Just Accepted Manuscript • DOI: 10.1021/acs.jpclett.6b01041 • Publication Date (Web): 20 Jun 2016
Downloaded from http://pubs.acs.org on June 22, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
The Journal of Physical Chemistry Letters is published by the American Chemical

Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.
Page 1 of 22 The Journal of Physical Chemistry Letters
1
2
3
4
5
6
7
8
Band Gaps of the Lead-Free Halide Double
9
10
11 Perovskites Cs2BiAgCl6 and Cs2BiAgBr6
12
13
14
15
from Theory and Experiment
16
17
18
19 Marina R. Filip,† Samuel Hillman,‡ Amir Abbas Haghighirad,‡ Henry J. Snaith,‡
20
21 and Feliciano Giustino∗,†
22
23
24 Department of Materials, University of Oxford, Parks Road OX1 3PH, Oxford, UK, and
25
26 Department of Physics, University of Oxford, Clarendon Laboratory, Parks Road, Oxford
27
28 OX1 3PU, UK
29
30
31 E-mail: feliciano.giustino@materials.ox.ac.uk
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
∗
56 To whom correspondence should be addressed
57 †
Department of Materials, Unversity of Oxford
58 ‡
Department of Physics, University of Oxford
59
60
1
ACS Paragon Plus Environment
The Journal of Physical Chemistry Letters Page 2 of 22
1
2
3
4
Abstract
5
6 The recent discovery of lead-free halide double perovskites with band gaps in the
7
8 visible represents an important step forward in the design of environmentally-friendly
9
10 perovskite solar cells. Within this new family of semiconductors, Cs2 BiAgCl6 and
11
12 Cs2 BiAgBr6 are stable compounds crystallizing in the elpasolite structure. Following
13
14 the recent computational discovery and experimental synthesis of these compounds, a
15
16 detailed investigation of their electronic properties is warranted in order to establish
17
18 their potential as optoelectronic materials. In this work we perform many-body per-
19
20 turbation theory calculations and obtain high accuracy band gaps for both compounds.
21
22 In addition, we report on the synthesis of Cs2 BiAgBr6 single crystals, which are stable
23
24 in ambient conditions. From our complementary theoretical and experimental analysis
25
26
we are able to assign the indirect character of the band gaps and obtain both exper-
27
imental and theoretical band gaps of these novel semiconductors which are in close
28
29 agreement.
30
31 TOC Graphic
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
Keywords: Double halide perovskites, Lead-free perovskites, Electronic structure,
54
Quasiparticle Band Gap
55
56
57
58
59
60
2
1
2
3
4
Hybrid organic-inorganic halide perovskites are currently the fastest developing photovoltaic
5
6 materials. 1–5 Lead-halide perovskites are solution-processable at low-temperatures, 6 exhibit
7
8 efficient light absorption in the visible, 7 long diffusion lengths, and high mobilities of the
9
10 photogenerated carriers, 8 as well as the ability to recycle the photons emitted upon radia-
11
12 tive recombination. 9 In the last five years, perovskite solar cells based on methylammonium
13
14 lead-iodide and related materials showed an unparalleled growth in power conversion ef-
15
16 ficiency, 10–12 up to a certified record exceeding 22%. 13 However, despite this astounding
17
18 success, perovskite solar cells are also facing important challenges, due to the lack of stabil-
19
20 ity upon prolonged exposure to light, humidity and increased temperatures. 14–20 In addition,
21
22 the toxicity of lead calls into question the long-term environmental impact of perovskite solar
23
24 cells. 21–23
25
26
27 A promising avenue for addressing both the stability and the toxicity bottlenecks is to explore
28
29 closely-related compounds via rational design. For example, it has been shown that by
30
31 employing mixed-halide and mixed cation perovskites not only can the electronic properties
32
33 be tuned, 24–27 but also the composition of the mixes can be explored in order to locate
34
35 a ‘stability sweet-spot’. 28 Similar design principles have been employed in the search for
36
37 potential Pb-replacements, but materials which can rival the unique optoelectronic properties
38
39 of lead-halide perovskites have not been found yet. The strategy of completely replacing Pb
40
41 by Sn has so far been unsuccessful, due to the tendency of Sn to oxidise in the +4 state
42
43 rather than the desired +2 state. 29,30 Furthermore, recent studies focused on the ecotoxity of
44
45 Sn perovskites, and suggested that Sn is not likely to mitigate the health hazards of halide
46
47 perovskite solar cells. 21 In addition, high-throughput computational studies pointed out that
48
49 the substitution of Pb by other metals in the periodic table is likely to compromise the ideal
50
51 optoelectronic properties of lead-halide perovskites, 31,32 and demonstrated that lead is key
52
53 to the electronic properties of these materials.
54
55
56 Very recently, three independent studies reported simultaneously the successful replacement
57
58
59
60
3
1
2
3
4
of Pb via the heterovalent substitution with Bi and Ag, 33–35 so as to form a halide double
5
6 perovskites. In Ref. 35 we reported the computational design of a new family of halide dou-
7
8 ble perovskites of the type Cs2 BB′ X6 (with B = Sb, Bi; B′ = Cu, Ag, Au and X = Cl, Br, I).
9
10 These compounds exhibit band gaps which are highly tunable in the visible range, as well as
11
12 low effective masses. The experimental synthesis of two members of this family, Cs2 BiAgCl6
13
14 and Cs2 BiAgBr6 , was successfully demonstrated, 33–35 both via solid-state reactions and via
15
16 solution processing. Both compounds crystallize in the elpasolite structure, a highly symmet-
17
18 ric face-centred cubic double-perovskite structure, they absorb light in the visible range of the
19
20 solar spectrum, and exhibit an impressive stability under ambient conditions. 33,35 Shortly af-
21
22 ter the discovery of Cs2 BiAgCl6 and Cs2 BiAgBr6 , the synthesis of hybrid organic-inorganic
23
24 halide double perovskite CH3 NH3 BiKCl6 was reported, 36 further expanding the potential
25
26 for materials design in this family of compounds. While photovoltaic devices employing
27
28 halide double perovskites have not yet been reported, these materials promise to be the long
29
30 sought-after stable, non-toxic halide perovskites, and warrant a thorough investigation of
31
32 their optoelectronic properties.
33
34
35 Optical measurements and electronic structure calculations suggest that both Cs2 BiAgCl6
36
37 and Cs2 BiAgBr6 are indirect band gap semiconductors. 33–35 However, there are sizeable
38
39 discrepancies in the measured optical band gaps. The gaps reported for Cs2 BiAgCl6 range
40
41 from 2.2 eV 35 to 2.77 eV, 34 while those of Cs2 BiAgBr6 range from 1.83 eV 33 to 2.19 eV. 34
42
43 These differences are possibly due to the different methods of sample preparation (solid
44
45 state reaction vs. solution processing), and measurement techniques (optical absorption and
46
47 photoluminescence 33,35 vs diffuse reflectance 34 ), but also to the fitting procedures used to
48
49 determine the indirect band gap from Tauc plots. Since the band gap is a crucial property
50
51 in the operation of solar cells, these discrepancies call for further investigations. Refs. 34,35
52
53 also report on the calculation of the band gaps using density functional theory and the
54
55 hybrid functionals PBE0 37,38 and HSE. 38–40 The calculated band gaps also exhibit significant
56
57
discrepancies: the values reported for Cs2 BiAgCl6 range from 2.62 eV 34 to 3.0 eV 35 and
58
59
60
4
1
2
3
4
those for Cs2 BiAgBr6 range from 2.06 eV 34 to 2.3 eV. 35 This may be due to the different
5
6 structural models used in the calculations, but also to the known tendency of PBE0 and
7
8 HSE hybrid functionals to yield band gaps which can differ significantly. 41 Moreover, hybrid
9
10 functional approaches are strongly dependent on the fraction of exact exchange and hence
11
12 not fully predictive. In this work we resolve the ambiguity on the determination of the band
13
14 gap of Cs2 BiAgCl6 and Cs2 BiAgBr6 by performing state-or-the-art many-body perturbation
15
16 theory calculations in the GW approximation. 42–46 In addition, we report on the synthesis
17
18 of Cs2 BiAgBr6 and compare the optical properties of Cs2 BiAgCl6 and Cs2 BiAgBr6 . We
19
20 combine our theoretical and experimental studies to assign the indirect character of the
21
22 band gaps for these two compounds.
23
24
25 Cs2 BiAgX6 (X = Cl, Br) crystallizes in the elpasolite K2 NaAlF6 structure, which corresponds
26
27 to an ideal double perovskite. This structure is a network of corner-sharing octahedra, with
28
29 Cs in the middle of the cuboctahedral cavity. The centres of the octahedra are either occupied
30
31 by Bi or Ag, alternating in a rock-salt configuration, as shown in Figure 1a.
32
33
We synthesize crystalline samples of Cs2 BiAgCl6 and Cs2 BiAgBr6 through both an acidic
34
35
36
solution process and a solid state reaction route, as described in the Supporting Information.
37
38 Through these two methods we obtain high-purity pollycrystalline samples, as well as single
39
40 crystals of up to 1 mm3 in size (shown in Figure 1c-d), respectively. Our single crystals
41
42 exhibit very good stability under environmental conditions. We fully characterize their crys-
43
44 tal structure through powder and single crystal X-Ray diffraction (XRD) measured at room
45
46 temperature (Figure S1 of the Supporting Information). For both compounds we find F m3̄m
47
48 face-centered cubic structures, and resolve the atomic positions and lattice parameters. Our
49
50 refinements are in agreement with Refs. 33–35. The primitive unit cells comprise of only
51
52 one formula unit, as shown in Figure 1b, while the conventional unit cell comprises of four
53
54 formula units (Figure 1a).
55
56
57 We investigate the electronic structure of both compounds by first calculating the band struc-
58
59
60
5
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28 Figure 1: Polyhedral model of (a) the conventional unit cell and (b) the primitive unit
29 cell of Cs2 BiAgX6 (X = Cl, Br). In (b) the red dashed lines mark the primitive lattice
30 vectors. The large dark blue spheres represent the Cs atoms, the small grey, blue and green
31
32 spheres are the Ag, Bi and halogen atoms, respectively. Single crystal of Cs2 BiAgCl6 (c) and
33 Cs2 BiAgBr6 (d), synthesized as described in the Supporting Information.
34
35
36 tures within the local density approximation to density functional theory 47,48 (DFT-LDA,
37
38 as described in the Supporting Information), as implemented in the Quantum Espresso
39
40 suite. 49 In both cases the calculated band gap is indirect, with the valence band top at the
41
42 X (2π/a, 0, 0) point, and the conduction band bottom at the L (π/a, π/a, π/a) point of the
43
44 Brillouin zone. In Figure 2a and 2b we show a comparison between the band structure of
45
46 Cs2 BiAgCl6 and Cs2 BiAgBr6 calculated with or without spin-orbit coupling. We observe a
47
48 large spin-orbit splitting of 1.5-1.6 eV for the first conduction band. The spin-orbit coupling
49
50 leads to the formation of an isolated conduction band in the middle of the scalar relativis-
51
52 tic band gap. Such a large qualitative difference between the scalar-relativistic and the
53
54 fully-relativistic calculations is reminiscent of lead-halide perovskites, where the spin-orbit
55
56
coupling is essential for the correct description of the band edges and effective masses. 50,51
57
58
59
60
6
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26 Figure 2: DFT-LDA band structure of Cs2 BiAgCl6 (a) and Cs2 BiAgBr6 (b). The calculations
27
28 are performed with (blue lines) and without (red lines) spin-orbit coupling. The light blue
29 shading highlights the width of the lowest conduction band as calculated from DFT+SOC:
30 0.6 eV for Cs2 BiAgCl6 and 0.9 eV for Cs2 BiAgBr6 .
31
32
33 In all the following calculations we explicitly take into account spin-orbit coupling.
34
35
36 As shown in Figure 2, the conduction band bottom in both cases has two minima, with
37
38 energies within less than 0.2 eV. This small difference is sensitive to the structure of the
39
40 compound. In Figure S2 of the Supporting Information we show a comparison between the
41
42 DFT band structures calculated for the experimental and for the optimized structures. In
43
44 the case of Cs2 BiAgCl6 , we can see that the conduction band bottom is either at the Γ or L
45
46 point, depending on whether calculations are performed on the optimized or the experimental
47
48 structures. Similar results are obtained in the case of Cs2 BiAgBr6 . In order to avoid any
49
50 ambiguity arising from the sensitivity of the conduction band edge to the crystal structure,
51
52 in all following calculations we use the experimental crystal structures, with atomic positions
53
54 and lattice parameters listed in Table S1 and S2 of the Supporting Information.
55
56
57 In order to clarify some of the peculiar features of the band structures of these compounds,
58
59
60
7
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30 Figure 3: (a) Comparison of the molecular orbital diagrams Cs2 BiAgCl6 (left) and
31 Cs2 BiAgBr6 (right). The atomic one-electron energies are marked by black thick lines,
32 dark blue rectangles correspond to the Bi-halide hybrid bands and the grey rectangles corre-
33
spond to the Ag-halide hybrid bands. The light blue rectangles represent the bands formed
34
35 in Cs2 BiAgCl6 and Cs2 BiAgBr6 respectively. Filled rectangles represent occupied (valence)
36 bands, while the empty rectangles represent the unoccupied (conduction) bands. All bands
37 are aligned with respect to the all-electron energy level of the Bi-5d1/2 and then red-shifted
38
by 1.5 eV for visualization purposes. (b) Square modulus of the wave functions for states
39
40 marked from 1 to 3 on the molecular orbital diagram of Cs2 BiAgCl6 . 1 represents the Bi-
41 6p1/2 /halide-p antibonding orbitals at the bottom of the conduction band (at Γ point). 2 is
42 the antibonding Ag-5s/halide-p at the L-point of the conduction band), while 3 corresponds
43 to the Ag-4d/halide-p antibonding orbitals found at the top of the valence band (at X-point).
44
45 The color codes for the atoms are the same as in Figure 1. In the wave-function plots the
46 Cs atoms are not shown for clarity.
47
48
49 we calculate the atom-projected density of states (Figure S3 of the Supporting Information)
50
51 and use this to construct the molecular orbital diagrams of both halide double perovskites.
52
53 In Figure 3 we show a schematic representation of the bonding and antibonding states in the
54
55 valence and conduction bands. In both compounds the isolated conduction band originates
56
57 primarily from spin-orbit coupling, due to a splitting of the Bi-p1/2 and Bi-p3/2 states in
58
59
60
8
1
2
3
4
the conduction band (Figure 2 and Figure S3). This band is predominantly antibonding
5
6 Bi-p/halogen-p states, and its corresponding bonding orbitals can be found approximately
7
8 5 eV below the valence band top in both cases (Figure S3). In addition, a small contribution
9
10 of the Bi-s/halogen-p antibonding state can be seen at the top of the valence band (diagram
11
12 3 in Figure 3b). Hybridized Ag-d/halogen-p states are predominant throughout the highest
13
14 valence band in both compounds, while the Ag-s states are pushed to the conduction band,
15
16 consistent with the +1 character of the noble metal in this chemical environment.
17
18
19
From Figure 3 we can rationalize the changes in the electronic structure of the two double
20
21
perovskites with the halogen atom. As we move from Cl to Br we observe that the valence
22
23 band width is reduced and the valence band top is slightly higher in energy. The band
24
25 narrowing is due to the smaller difference in energy between the Ag-d and Br-p levels as
26
27 compared to Ag-d and Cl-p; the shift to higher energies is due to the fact that the Br-p
28
29 states are higher than the Cl-p states. The change in the band gap when moving from Cl
30
31 to Br is dominated by the increase in the width of the bottom conduction band from 0.6 eV
32
33 to 0.9 eV (Figure 2). This feature is due to the more delocalized nature of Br-4p orbitals
34
35 as compared to the Cl-3p states, which increases the overlap with the Bi-6p wave functions,
36
37 and thereby the dispersion of the conduction bands. In addition, Br-4p levels are shallower
38
39 and closer in energy to the Bi-6p states than in the case of the Cl-p, leading to a shift of the
40
41 conduction band to lower energies in the case of Cs2 BiAgBr6 .
42
43
44 The electronic structure of the Bi-Ag halide double perovskites exhibits similar features to
45
46 ternary chalcopyrite semiconductors, CuInS2 (CIS), CuGaS2 (CGS) and also the quaternary
47
48 Cu2 ZnSnS4 (CZTS). 52,53 In this family of compounds the noble metal d states are predomi-
49
50 nant in the valence band and exhibit strong bonds with the chalcogen anions, in close analogy
51
52 to Cs2 BiAgCl6 and Cs2 BiAgBr6 . The anion-p orbitals contribute to both the conduction and
53
54 valence bands both in the double perovskites (see Figure S2) and in CZTS. 52 In addition,
55
56 CZTS also exhibits an isolated conduction band, although it corresponds to a Sn-s/S-p char-
57
58
59
60
9
1
2
3
4
acter and therefore is not arising from spin-orbit coupling. In fact, the isolated conduction
5
6 band is a common feature in quaternary compounds, and in particular in elpasolites. 54
7
8 As expected, the band gaps calculated within DFT-LDA including spin-orbit coupling un-
9
10 derestimate the optical band gaps reported in Refs 33–35, the discrepancy being as large as
11
12 1.5 eV. To correct this severe underestimation, we calculate the GW quasiparticle band gaps
13
14 of Cs2 BiAgCl6 and Cs2 BiAgBr6 using the Yambo code. 55 A detailed report of all convergence
15
16
tests with respect to the number of empty states, plane-wave cutoffs as well as the Brillouin
17
18
19
zone sampling can be found in the Supporting Information (see Figure S4).
20
21 The semicore states 4s2 4p6 of Ag are particularly important for the calculation of the quasi-
22
23 particle energies. This is shown in Figure S5 of the Supporting Information, where we
24
25 compare the quasiparticle energies and the separate contributions from the exchange and
26
27 correlation self energies calculated using the 4d10 5s1 and 4s2 4p6 4d10 5s0 valence configura-
28
29 tions of Ag. Here we see that the quasiparticle band gap increases by 1 eV when the semicore
30
31
4s2 4p6 states of Ag are included in the calculation (Figure S5a). As shown in Figure S5b
32
33
and S5c, the largest contribution of the semicore states to the quasiparticle correction comes
34
35
36
from the exchange term. The correlation self-energy, on the other hand, is virtually unaf-
37
38 fected. This effect originates in the large overlap between the 4d10 orbitals and the 4s2 and
39
40 4p6 orbitals of Ag (see Figure S6 of the Supporting Information), and is more pronounced in
41
42 the exchange self-energy, which is energy-independent. The contribution of semicore states
43
44 to the quasiparticle energies is most pronounced at the top of the valence band, where the
45
46 Ag-4d is predominant (Figure S5). The impact of semicore electrons on the calculation of
47
48 the exchange self energy is well documented in the literature, 44,56–59 and is known to have
49
50 a large contribution when d electrons overlap significantly with the s and p electrons of the
51
52 same shell. 56
53
54
55 In Figure 4 we plot the direct and indirect band gaps calculated from G0 W0 along with the
56
57 measured optical band gaps for Cs2 BiAgCl6 and Cs2 BiAgBr6 . For completeness, in Figure 4
58
59
60
10
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Figure 4: Comparison between the quasiparticle indirect (red disks) and direct band gaps
36
37 (green disks) and experimental band gaps (blue lines) for Cs2 BiAgCl6 (a) and Cs2 BiAgBr6
38 (b). The area shaded in light blue is given by the positions of the two shoulders each side
39 of the peaks in the photoluminescence spectra shown in (c) and (d). Room temperature
40 photoluminescence spectra measured for Cs2 BiAgCl6 (c) and Cs2 BiAgBr6 (d). The blue
41
42 continuous lines mark the positions if the main peaks in each case, while the light blue
43 shaded area makes the region between the two shoulders located each side of the main
44 peak. The photoluminescence spectrum of Cs2 BiAgCl6 is the same as the one we reported
45 in Figure 2 of Ref. 35.
46
47
48
we report the two lowest computed indirect band gaps between L-X and Γ-X, as well as the
49
50
51
computed direct band gap at the X-point. For Cs2 BiAgCl6 we obtain indirect quasiparticle
52
53
band gaps of 2.42 eV (L-X) and 2.48 eV (Γ-X) and a direct band gap at X of 2.96 eV, while for
54
55 Cs2 BiAgBr6 we obtain indirect band gaps of 1.83 eV (L-X) and 1.97 eV (Γ-X) and a direct
56
57 band gap of 2.51 eV. These values are smaller than in the hybrid functional calculations
58
59
60
11
1
2
3
4
reported in Ref 34,35 by up to 0.6 eV. We also calculate the high-frequency (electronic)
5
6 dielectric constants of the two compounds within the random-phase approximation (RPA)
7
8 and obtain 4.7 and 5.8 for Cs2 BiAgCl6 and Cs2 BiAgBr6 respectively. These values are similar
9
10 to the high-frequency dielectric constant of 5.8 obtained for CH3 NH3 PbI3 . 50,60
11
12 In order to assess the optical band gap of Cs2 BiAgCl6 and Cs2 BiAgBr6 we perform optical
13
14 absorption (Figure S7a and b of the Supporting Information) and photoluminescence (PL)
15
16
experiments (Figure 4c and d). We observe that the leading edge of the optical absorption
17
18
19
onset is blue-shifted with respect to the position of the main PL peak, by up to 0.5 eV for
20
21
both compounds. Our current measurements are performed on powder samples, rather than
22
23 smooth uniform films, and therefore the optical absorption spectra may be complicated by
24
25 multiple absorptions, reflections and scattering. For this reason, a precise lineshape analysis
26
27 of the optical absorption spectra using Tauc plots is not possible. Instead, we rely on PL to
28
29 probe the band edges.
30
31
The main peaks of the PL spectra are at 2.15 eV for Cs2 BiAgCl6 and 1.89 eV for Cs2 BiAgBr6 .
32
33
In the case of Cs2 BiAgBr6 , the position of the main PL peak is in very good agreement with
34
35
36
Ref. 33. In both PL spectra we observe two shoulders on each side of the main peaks,
37
38 at 2.4 eV and 1.9 eV for Cs2 BiAgCl6 and at 2.1 and 1.7 eV for Cs2 BiAgBr6 . In both
39
40 cases the difference between the position of the main peak and the two shoulders is too
41
42 large to be explained by the absorption or emission of phonons, which have energies in the
43
44 range of 10 meV. 60 On the other hand, these features are similar to the two indirect optical
45
46 transitions observed in the temperature-dependent PL spectra of WSe2 . 61 Therefore, it is
47
48 possible that the peaks and shoulders in the PL spectra may correspond to multiple indirect
49
50 transitions, although we cannot exclude that defects or excitonic effects play a role in these
51
52 features. Recently some of us have shown that electron-phonon interactions can explain
53
54 the PL linewidths of lead-iodide perovskites. 62 In that case the full-width at half-maximum
55
56 (FWHL) of the PL peak is smaller than 100 meV at room temperature. In this case we
57
58
59
60
12
1
2
3
4
could expect a similar effect, enhanced by the fact that the Cl and Br masses are smaller
5
6 than the mass of I. This reasoning would be compatible with the fact that we see a larger
7
8 PL width in the spectrum of Cs2 BiAgCl6 as compared to Cs2 BiAgBr6 . To account for the
9
10 the broad PL lineshape in our analysis, we plot the indirect band gaps obtained from the
11
12 energies of the main peaks of the PL spectra (continuous lines), as well as a shading with a
13
14 width given by the energies of the shoulders on each side of the main peaks (Figure 4c-d),
15
16 and compare these results with our GW band gaps (Figure 4a-b).
17
18
19
In general a direct comparison between optical band gaps measured at room temperature and
20
21
quasiparticle band gaps is not straightforward, due to the absence of excitonic and electron-
22
23 phonon coupling effects in the GW formalism. In the context of halide perovskite, this
24
25 comparison is reasonable given their optoelectronic properties. In the case of CH3 NH3 PbI3 ,
26
27 the excitonic and electron-phonon coupling effects are very small (of the order of 50 meV)
28
29 and tend to cancel each other out. 62–64 Moreover, despite being a hybrid organic-inorganic
30
31 compound, CH3 NH3 PbI3 has a sharp optical absorption edge, with a small Urbach energy
32
33 of 15 meV which resembles closely that of inorganic semiconductors such as GaAs. 7 Given
34
35 these consideration, as well as their common structural features, it is appropriate to draw
36
37 the analogy between CH3 NH3 PbI3 and our inorganic halide double perovskites.
38
39
40 Figure 4 shows that our calculated quasiparticle band gaps are close to the positions of the
41
42 photoluminescence peaks for the two double perovskites. The indirect quasiparticle band
43
44 gaps of Cs2 BiAgCl6 is slightly higher than the optical band gap obtained from photolumi-
45
46 nescence by approximately 0.2 eV, while in the case of Cs2 BiAgBr6 , the two indirect band
47
48 gaps are within less than 0.1 eV from the optical band gap. Moreover, the difference between
49
50 the direct and indirect quasiparticle band gaps is of 0.6-0.7 eV, which is compatible with
51
52 the difference between the leading edge of the optical absorption spectrum and the photolu-
53
54 minescence peak. Based on these observations, we propose to assign the photoluminescence
55
56 peak to a phonon-assisted recombination across the indirect band gap and the leading edge
57
58
59
60
13
1
2
3
4
of the optical absorption onset to the direct band gap.
5
6 In conclusion, we have presented a detailed study of the electronic properties of the newly
7
8 discovered Cs2 BiAgCl6 and Cs2 BiAgBr6 double perovskites within DFT+SOC and state-of-
9
10 the-art GW quasiparticle calculations. We have synthesized Cs2 BiAgCl6 and Cs2 BiAgBr6
11
12 single crystals and performed optical absorption and photoluminescence measurements. Our
13
14 analysis of the electronic structure revealed that the two halide double perovskites are indirect
15
16
band gap semiconductors with electronic properties which resemble closely the well known
17
18
19
lead-halide perovskites as well as the ternary and quaternary chalcopyrite semiconductors
20
21
of the CIGS and CZTS family. We have calculated the quasiparticle band gaps of both
22
23 compounds with high accuracy and obtained indirect gaps of 2.4 eV for Cs2 BiAgCl6 and
24
25 1.8 eV Cs2 BiAgBr6 . These values are very close to our experimental band gaps of 2.2 eV
26
27 and 1.9 eV for Cs2 BiAgCl6 and Cs2 BiAgBr6 respectively and this agreement allowed us to
28
29 assign the indirect nature of the photoluminescence peaks. Future studies should address the
30
31 optical, excitonic and transport properties properties, as well as the temperature dependence
32
33 of the structural and electronic properties of these new compounds, in order to fully assess
34
35 their potential as photovoltaic materials. By following the same materials design strategies
36
37 currently used for lead-halide perovskites, we believe that it will be possible to optimize the
38
39 optoelectronic properties this new class of stable and environmentally-friendly perovskites
40
41 for optoelectronic applications.
42
43
44
45 Supporting Information
46
47
48
49 The Supporting Information includes a detailed description of the computational and ex-
50
51 perimental methods, crystallographic data and additional plots on the GW calculations,
52
53 convergence and optical absorption spectra.
54
55
56
57
58
59
60
14
1
2
3
4
Acknowledgement
5
6
7 The research leading to these results has received funding from the the Graphene Flagship
8
9 (EU FP7 grant no. 604391), the Leverhulme Trust (Grant RL-2012-001), the UK Engineer-
10
11 ing and Physical Sciences Research Council (Grant No. EP/J009857/1), and the European
12
13 Union Seventh Framework Programme (FP7/2007-2013) under grant agreements n◦ 239578
14
15 (ALIGN) and n◦ 604032 (MESO). The authors acknowledge the use of the University of Ox-
16
17 ford Advanced Research Computing (ARC) facility (http://dx.doi.org/10.5281/zenodo.22558)
18
19 and the ARCHER UK National Supercomputing Service under the ‘AMSEC’ Leadership
20
21 project. Figures involving atomic structures were rendered using VESTA. 65
22
23
24
25
26 References
27
28
29
(1) Green, M.; Ho-Baillie, A.; Snaith, H. J. The Emergence of Perovskite Solar Cells. Nature
30
31 Photonics 2014, 8, 506.
32
33
(2) Green, M. A.; Emery, K.; Hishikawa, Y.; Warta, W.; Dunlop, E. D. Solar Cell Efficiency
34
35
36
Tables (Version 45). Progress in Photovoltaics 2015, 23, 1.
37
38
(3) Stranks, S.; Snaith, H. J. Metal-Halide Perovskites for Photovoltaic and Light-Emitting
39
40
41
Devices. Nature Nanotechology 2015, 10, 391.
42
43 (4) Seo, J.; Noh, J. H.; Seok, S. I. Rational Strategies for Efficient Perovskite Solar Cells.
44
45
Acc. Chem. Res. 2016, 49, 562–572.
46
47
48 (5) Choi, J. J.; Billinge, S. J. L. Perovskites at the Nanoscale: From Fundamentals to
49
50 Applications. Nanoscale 2016, 8, 6206–6208.
51
52
53 (6) You, J.; Hong, Z.; Yang, Y. M.; Chen, Q.; Cai, M.; Song, T. B.; Chen, C. C.; Lu, S.;
54
55 Liu, Y.; Zhou, H. et al. Low-Temperature Solution Processed Perovskite Solar Cells
56
57 with High Efficiency and Flexibility. ACS Nano 2014, 8, 1674–1680.
58
59
60
15
1
2
3
4
(7) De Wolf, S.; Holovsky, J.; Moon, S.-J.; Löpert, P.; Niesen, B.; Ledinsky, M.; Haug, F.-
5
6 J.; Yum, J.-H.; Ballif, C. Organometallic Halide Perovskites: Sharp Optical Absorption
7
8 Edge and Its Relation to Photovoltaic Performance. J. Phys. Chem. Lett. 2014, 5,
9
10 1035–1039.
11
12
13 (8) Herz, L. M. Charge-Carrier Dynamics in Organic-Inorganic Metal Halide Perovskites.
14
15 Annu. Rev. Phys. Chem. 2016,
16
17
18
(9) Pazos-Outón, L. M.; Szumilo, M.; Limboll, R.; Richter, J. M.; Crespo-Quesada, M.;
19
20 Adbi-Jalebi, M.; Beeson, H. J.; Vrućnić, M.; Alsari, M.; Snaith, H. J. et al. Photon
21
22 Recycling in Lead-Iodide Perovskite Solar Cells. Science 2016, 351, 1430–1433.
23
24
25
(10) Lee, M. M.; Teuscher, J.; Miyasaka, T.; Myrakami, T. N.; Snaith, H. J. Efficient Hybrid
26
27 Solar Cells Based on Meso-Superstructured Organometal Halide Perovskites. Science
28
29 2012, 338, 643.
30
31
32
(11) Kim, H.-S.; Lee, C. R.; Im, J.-H.; Lee, K.-B.; Moehl, T.; Marchioro, A.; Moon, S.-J.;
33
34 Humphry-Baker, R.; Yum, J.-H.; Moser, J. E. et al. Lead Iodide Perovskite Sensitized
35
36 All-Solid-State Submicron Thin Film Mesoscopic Solar Cell with Efficiency Exceeding
37
38 9%. Sci. Rep. 2012, 2, 591.
39
40
41 (12) Burschka, J.; Pellet, N.; Moon, S.-J.; Humphry-Baker, R.; Gao, P.; Nazeerud-
42
43 din, M. K.; Grätzel, M. Sequential Deposition as a Route to High-Performance
44
45 Perovskite-Sensitized Solar Cells. Nature 2013, 499, 316.
46
47
48 (13) Best Research-Cell Efficiencies. http://www.nrel.gov/ncpv/images/efficiency_
49
50 chart.jpg.
51
52
53
(14) Niu, G.; Guo, X.; Wang, L. Review of Recent Progress in Chemical Stability of Per-
54
55 ovskite Solar Cells. J. Mater. Chem. A 2015, 3, 8970–8980.
56
57
58
59
60
16
1
2
3
4
(15) Divitini, G.; Cacovich, S.; Matteocci, F.; Di Carlo, A.; Ducati, C. In Situ Observation
5
6 of Heat-Induced Degradation of Perovskite Solar Cells. Nature Energy 2016, 1, 15012.
7
8
9
(16) Li, X.; Dar, M. I.; Yi, C.; Luo, J.; Tschumi, M.; Zakeeruddin, S. M.; Nazeeruddin, M. K.;
10
11
Han, H.; Grätzel, M. Improved Performance and Stability of Perovskite Solar Cells
12
13 by Crystal Crosslinking with Alkylphosphonic Acid ω-Ammonium Chlorides. Nature
14
15 Chemistry 2015, 7, 703–711.
16
17
18
(17) Sutton, R. J.; Eperon, G. E.; Miranda, L.; Parrott, E. S.; Kamino, B. A.; Patel, J. B.;
19
20 Hörantner, M. T.; Johnston, M. B.; Haghighirad, A. A.; Moore, D. T. et al. Bandgap-
21
22 Tunable Cesium Lead Halide Perovskites with High Thermal Stability for Efficient Solar
23
24 Cells. Adv. Energ. Mater. 2016,
25
26
27 (18) Manser, J. S.; Saidaminov, M. I.; Christians, J. A.; Bakr, O. M.; Kamat, P. V. Making
28
29 and Breaking of Lead-Halide Perovskites. Acc. Chem. Res. 2016,
30
31
32
(19) Hoke, E. T.; Slotcavage, D. J.; Dohner, E. R.; Bowring, A. R.; Karunadasa, H. I.;
33
34 McGehee, M. D. Reversible Photo-Induced Trap Formation in Mixed-Halide Hybrid
35
36 Perovskites for Photovoltaics. Chem. Sci. 2015, 6, 613–617.
37
38
39
(20) Eames, C.; Frost, J. M.; Barnes, P. R. F.; O’Regan, A., B. C.and Walsh; Saiful Islam, M.
40
41 Ionic Transport in Hybrid Lead Iodide Perovskite Solar Cells. Nature Commun. 2015,
42
43 6, 7497.
44
45
46
(21) Babagayigit, A.; Thanh, D. D.; Ethirajan, A.; Manca, J.; Muller, M.; Boyen, H.-G.;
47
48 Conings, B. Assessing the Toxicity of Pb- and Sn-Based Perovskite Solar Cells in Model
49
50 Organism Danio Rerio. Sci. Rep. 2016, 6, 18721.
51
52
53
(22) Babagayigit, A.; Ethirajan, A.; Muller, M.; Conings, B. Toxicity of Organometal Halide
54
55 Perovskite Solar Cells. Nature Materials 2016, 15, 247.
56
57
58
59
60
17
1
2
3
4
(23) Benmessaoud, I. R.; Mahul-Mellier, A.-L.; Horváth, E.; Maco, B.; Spina, M.;
5
6 Lashuel, L., H. A. Forró Health Hazards of Methylammonium Lead Iodide Based Per-
7
8 ovskites: Cytotoxicity Studies. Toxicol. Res. 2016, 5, 407–419.
9
10
11
(24) Filip, M. R.; Eperon, G.; Snaith, H. J.; Giustino, F. Steric Engineering of Metal-Halide
12
13 Perovskites with Tunable Optical Band Gaps. Nature Commun. 2014, 5, 5757.
14
15
16
(25) Pellet, N.; Gao, P.; Gregori, G.; Yang, T.-Y.; Nazeerrudin, M. K.; Maier, J.; Grätzel, M.
17
18
Mixed-Organic-Cationic Perovskite Photovoltaics for Enhanced Solar-Light Harvesting.
19
20 Angew. Chem. 2014, 53, 3151.
21
22
23
(26) Eperon, G. E.; Stranks, S. D.; Menelaou, C.; Johnston, M.; Hertz, L. M.; Snaith, H. J.
24
25
Formamidinium Lead Trihalide: A Broadly Tunable Perovskite Heterojunction Solar
26
27 Cells. Energ. Environ. Sci. 2014, 7, 982.
28
29
30
(27) Noh, J. H.; Im, S. H.; Heo, T. N., J. H. end Mandal; Seok, S. I. Chemical Management
31
32
for Colorful, Efficient and Stable Inorganic-Organic Hybrid Nanostructured Solar Cells.
33
34 Nano Lett. 2013, 13, 1764.
35
36
37
(28) McMeekin, D. P.; Sadoughi, G.; Rehman, W.; Eperon, G.; Saliba, M.; Hörantner, M. T.;
38
39
Haghighirad, A.; Sakai, N.; Korte, L.; Rech, B. et al. A Mixed-Cation Lead Mixed-
40
41 Halide Perovskite Absorber for Tandem Solar Cells. Science 2016, 351, 151–155.
42
43
44
(29) Hao, F.; Stoumpos, C. C.; Cao, D. H.; Chang, R. P. H.; Kanatzidis, M. G. Lead-free
45
46
Solid-State Organic-Inorganic Halide Perovskite Solar Cells. Nature Photonics 2014,
47
48 8, 489.
49
50
51
(30) Noel, N.; Stranks, S. D.; Abate, A.; Wehrenfennig, C.; Guarnera, S.; Haghighirad, A.-
52
53
A.; Sadhanala, A.; Eperon, G. E.; Pathak, S. K.; Johnston, A., M. B. andPetrozza
54
55 et al. Lead-Free Organic-Inorganic Tin Halide Perovskite for Photovoltaic Applications.
56
57 Energ. Environ. Sci 2014, 7, 3061.
58
59
60
18
1
2
3
4
(31) Filip, M. R.; Giustino, F. Computational Screening of Homovalent Lead Substitution
5
6 in OrganicInorganic Halide Perovskites. J. Phys. Chem. C 2016, 120, 166–173.
7
8
9
(32) Körbel, S.; Marques, M. A. L.; Botti, S. Stability and Electronic Properties of New
10
11
Inorganic Perovskites from High-Throughput Ab Initio Calculations. J. Mater. Chem.
12
13 C 2016,
14
15
16
(33) Slavney, A. H.; Hu, T.; Lindenberg, A. M.; Karunadasa, H. I. A Bismuth-Halde Double
17
18
Perovskite with Long Carrier Recombination Lifetime for Photovoltaic Applications. J.
19
20 Am. Chem. Soc. 2016, 138, 2138–2141.
21
22
23
(34) McClure, E. T.; Ball, M. R.; Windl, W.; Woodward, P. M. Cs2 AgBiX6 (X = Br,
24
25
Cl): New Visible Light Absorbing, Lead-Free Halide Perovskite Semiconductors. Chem.
26
27 Mater. 2016, 28, 1348–1354.
28
29
30
(35) Volonakis, G.; Filip, M. R.; Haghighirad, A. A.; Sakai, N.; Wenger, B.; Snaith, H. J.;
31
32
Giustino, F. Lead-Free Halide Double Perovskites via Heterovalent Substitution of No-
33
34 ble Metals. J. Phys. Chem. Lett. 2016, 7, 1254–1259.
35
36
37
(36) Wei, F.; Deng, Z.; Sun, S.; Xie, F.; Evans, D. M.; Carpenter, M. A.; Bristowe, P. D.;
38
39
Cheetham, A. K. The Synthesis, Structure and Electronic Properties of a Lead-Free
40
41 Hybrid Inorganic-Organic Double Perovskite (MA)2 KBiCl6 (MA = methylammonium).
42
43 Materials Horizons 2016,
44
45
46
(37) Perdew, J. P.; Burke, K.; Ernzerhof, M. Generalized gradient approximation made
47
48 simple. Phys. Rev. Lett. 1996, 77, 3865.
49
50
51
(38) Paier, J.; Hirschl, R.; Marsman, M.; Kresse, G. The Perdew-Burke-Ernzerhof exchange-
52
53
correlation functional applied to the G2-1 test set using a plane-wave basis set. J. Chem.
54
55 Phys. 2005, 122, 234102.
56
57
58
59
60
19
1
2
3
4
(39) Heyd, J.; Scuseria, G. E.; Ernzerhof, M. Hybrid Functionals Based on a Screened
5
6 Coulomb Potential. J. Chem. Phys. 2003, 118, 8207–8215.
7
8
9
(40) Heyd, J.; Scuseria, G. E.; Erzerhof, M. Hybrid Functionals Based on a Screened
10
11
Coulomb Potential. J. Chem. Phys. 2006, 124, 219906.
12
13
(41) Friedrich, C.; Betzinger, M.; Schlipf, M.; Blügel, S.; Schindlmayr, A. Hybrid functionals
14
15
16
and GW approximation in the FLAPW method. J. Phys. Condens. Matter 2012, 24,
17
18
293201.
19
20
(42) Hedin, L. New Method for Calculating the One-Particle Green’s Function with Appli-
21
22
23
cation to the Electron-Gas Problem. Phys. Rev. 1965, 139, A796.
24
25 (43) Hybertsen, M. S.; Louie, S. G. Electron Correlation in Semiconductors and Insulators:
26
27
Band Gaps and Quasiparticle Energies. Phys. Rev. B 1986, 34, 5390.
28
29
30 (44) Aryasetiawan, F.; Gunnarsson, O. The GW Method. Rep. Prog. Phys. 1998, 61, 237.
31
32
33 (45) Giustino, F.; Cohen, M. L.; Louie, S. G. GW Method with the Self-Consistent Stern-
34
35 heimer Equation. Phys. Rev. B 2010, 81, 115105.
36
37
38 (46) Onida, G.; Reining, L.; Rubio, A. Electronic Excitations: Density-Functional Versus
39
40 Many-Body Green’s Function Approaches. Rev. Mod. Phys. 2002, 74, 601–659.
41
42
43 (47) Hohenberg, P.; Kohn, W. Inhomogeneous Electron Gas. Phys. Rev. 1964, 136, B864.
44
45
46 (48) Perdew, J. P.; Zunger, A. Self-Interaction Correction to Density-Functional Approxi-
47
48 mations for Many-Electrons Systems. Phys. Rev. B 1981, 23, 5048.
49
50
51 (49) Gianozzi, P.; Baroni, S.; Bonini, N.; Calandra, M.; Car, R.; Cavazzoni, C.; Ceresoli, D.;
52
53 Chiarotti, G. L.; Cococcioni, M.; Dabo, I. et al. QUANTUM ESPRESSO: A Modular
54
55 and Open-Source Software Project for Quantum Simulations of Materials. J. Phys.:
56
57 Condens. Matter. 2009, 21 .
58
59
60
20
1
2
3
4
(50) Filip, M. R.; Verdi, C.; Giustino, F. GW Band Structures and Carrier Effective Masses
5
6 of CH3NH3PbI3 and Hypothetical Perovskites of the Type APbI3: A = NH4, PH4,
7
8 AsH4 and SbH4. J. Phys. Chem. C 2015, 119, 25209–25219.
9
10
11
(51) Menéndez-Proupin, E.; Palacios, P.; Wahnón, P.; Conesa, J. C. Self-Consistent Rela-
12
13 tivistic Band Structure of the CH3 NH3 PbI3 Perovskite. Phys. Rev. B 2014, 90, 045207.
14
15
16
(52) Paier, J.; Asahi, R.; Nagoya, A.; Kresse, G. Cu2 ZnSnS4 as a potential photovoltaic
17
18
material: A hybrid Hartree-Fock density functional study. Phys. Rev. B 2009, 79,
19
20 115126.
21
22
23
(53) Jaffe, J. E.; Zunger, A. Theory of the band gap anomaly in ABC2 chalcopyrite semi-
24
25
conductors. Phys. Rev. B 1984, 29, 1882.
26
27
(54) Shi, H.; Du, M.-H. Discrete Electronic Bands in Semiconductors and Insulators: Po-
28
29
30
tential High-Light-Yield Scintillators. Phys. Rev. Appl. 2015, 3, 054005.
31
32 (55) Marini, A.; Hogan, C.; Grüning, M.; Varsano, D. Yambo: An Ab Initio Tool for Excited
33
34
State Calculations. Comp. Phys. Commun. 2009, 180, 1392.
35
36
37 (56) Rohlfing, M.; Krüger, P.; Pollmann, J. Quasiparticle band structure of CdS. Phys. Rev.
38
39 Lett. 1995, 75, 3489.
40
41
42 (57) Marini, A.; Onida, G.; Del Sole, R. Quasiparticle Electronic Structure of Copper in the
43
44 GW approximation. Phys. Rev. Lett. 2002, 88, 016403–1.
45
46
47 (58) Filip, M. R.; Patrick, C. E.; Giustino, F. GW Quasiparticle Band Structures of Stibnite,
48
49 Antimonselite, Bismuthinite, and Guanajuatite. Phys. Rev. B 2013, 87, 205125.
50
51
52 (59) Filip, M. R.; Giustino, F. GW Quasiparticle Band Gap of the Hybrid Organic-Inorganic
53
54 Perovskite CH3NH3PbI3: Effect of Spin-Orbit Interaction, Semicore Electrons, and
55
56 Self-Consistency. Phys. Rev. B 2014, 90, 245145.
57
58
59
60
21
1
2
3
4
(60) Perez-Osório, M. A.; Millot, R. L.; Filip, M. R.; Patel, J. B.; Herz, L. M.; John-
5
6 ston, M. B.; Giustino, F. Vibrational Properties of Organic-Inorganic Halide Perovskite
7
8 CH3 NH3 PbI3 from Theory and Experiment: Factor Group Analysis, First Principles
9
10 Calculations and Low-Temperature Infrared Spectra. J. Phys. Chem. C 2015, 119,
11
12 25703–25718.
13
14
15 (61) Zhao, W.; Ribeiro, R. M.; Toh, M.; Carvalho, A.; Kloc, C.; Castron Neto, A. H.;
16
17 Eda, G. Origin of the Indirect Optical Transitions in Few-Layer MoS2 , WS2 and WSe2 .
18
19 Nano Letters 2013, 13, 5267–5634.
20
21
22 (62) Wright, A. D.; Verdi, C.; Millot, R. L.; Eperon, G. E.; Perez-Osorio, M. A.; J., S. H.;
23
24 Giustino, F.; Herz, L. M. Electron-phonon coupling in hybrid lead-halide perovskites.
25
26 Nature Commun. 2016, 7, 11755.
27
28
29 (63) D’Innocenzo, V.; Grancini, G.; Alcocer, M. J. P.; Kandada, A. R. S.; Stranks, S. D.;
30
31 Lee, M. M.; Lanzani, G. L.; Snaith, H. J.; Petrozza, A. Excitons versus Free Charges
32
33 in Organo-Lead Tri-Halide Perovskites. Nature Commun. 2014, 3586.
34
35
36 (64) Millot, R. L.; Eperon, G. E.; Snaith, H. J.; Johnston, M. B.; Herz, L. M. Temperature-
37
38 Dependent Charge-Carrier Dynamics in CH3 NH3 PbI3 Perovskite Thin Films. Adv.
39
40 Func. Mater. 2015, 25, 6218–6227.
41
42
43 (65) Momma, K.; Izumi, F. VESTA: A Three-Dimensional Visualization System for Elec-
44
45 tronic and Structural Analysis. J. Appl. Cryst. 2008, 41, 653.
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
22

Filipfgsftysdw 2016

Uploaded by

Copyright:

Available Formats

You might also like

Filipfgsftysdw 2016

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Filipfgsftysdw 2016

Uploaded by

Copyright:

Available Formats

Subscriber access provided by UNIV OF UTAH

The Journal of Physical Chemistry Letters is published by the American Chemical

You might also like