Journal of Colloid and Interface Science 504 (2017) 111–114

Contents lists available at ScienceDirect

Journal of Colloid and Interface Science

journal homepage: www.elsevier.com/locate/jcis

Regular Article

Cationic surfactant blocks radical-inhibiting sites on silica

Tan Zhang a,b,1, Frank D. Blum a,⇑
a
Department of Chemistry, Oklahoma State University, Stillwater, OK 74078, USA
b
Department of Materials Science and Engineering, Texas A&M University, College Station, TX 77843, USA1

h i g h l i g h t s g r a p h i c a l a b s t r a c t

Fumed silica acts as a mild radical

inhibitor.

Proper choice of surfactant can reduce
inhibition of fumed silica.

Decomposition of initiator is
independent of surfactant type.

Radical polymerization of emulsion
gels depends on surfactant type.

a r t i c l e i n f o a b s t r a c t

Article history: Surfactant-catalyzed room-temperature radical polymerization of methyl methacrylate (MMA) was con-
Received 2 February 2017 ducted in the presence of fumed silica nanoparticles and water. Three types of surfactants, cationic
Revised 9 May 2017 (CTAB), nonionic (Triton X-100) or anionic (SDS), were used to catalyze the decomposition of the initiator,
Accepted 11 May 2017
2,20 -azobisisobutyronitrile (AIBN). The surfactant-catalyzed decomposition rate constant for AIBN at
Available online 12 May 2017
room temperature was found to be independent of the surfactant type. However, the rates of polymer-
ization of the MMA emulsion gels at room temperature were found to depend on the types of surfactant
Keywords:
with: cationic > nonionic > anionic. An inhibition period was observed for the polymerizations with non-
Surfactant
Emulsion polymerization
ionic and anionic surfactants. The radical-inhibition was likely due to the reactions between the radicals
Radical inhibition and the silanol groups on fumed silica. This inhibition can be reduced by using cationic surfactants to
Fumed silica block these surface silanols.
Adsorption Ó 2017 Elsevier Inc. All rights reserved.

1. Introduction monomers, such as styrene [2], methyl methacrylate (MMA) [3],

Fumed silica is a hydrophilic and amorphous form of silica with
a primary particle diameter of 5–50 nm and a branched aggregate
structure [1]. Due to the hydrogen-bonding between fumed silica
particles, it can form three-dimensional networks in many organic
⇑ Corresponding author.
E-mail address: fblum@okstate.edu (F.D. Blum).
1
Current address.
and 2-hydroxyethyl methacrylate [4], and consequently induce
gelation in these organic monomers. In these systems, the high vis-
cosity environment increases the rate of free radical polymeriza-
tion by reducing the termination rates [5], thus free radical
polymerization can also be enhanced by the gelation from the
fumed silica. Fumed silica, from this perspective, can be considered
to be a ‘‘catalyst”’ for free radical polymerization. However, we
would like to know if this is the only role that fumed silica can play
in free radical polymerization in these systems.

http://dx.doi.org/10.1016/j.jcis.2017.05.030
0021-9797/Ó 2017 Elsevier Inc. All rights reserved.
112 T. Zhang, F.D. Blum / Journal of Colloid and Interface Science 504 (2017) 111–114
With respect to polymer degradation, fumed silica can be con-
sidered as a ‘‘radical killer” because radical-initiated decomposi-
tion can be significantly suppressed with a small amount of
fumed silica [6]. The active sites on fumed silica are likely the sur-
face silanols, which likely react with free radicals and form rela-
tively stable surface radicals through radical transfer or radical
trapping [7–9]. These silanols should also be able to react with rad-
icals during polymerization, possibly inhibiting free radical poly-
merization to some extent.
Given that fumed silica possibly affects radical reactions, it may
change the kinetics of radical polymerization through aggregation
(gelation) and surface properties (surface silanols). Gelation may
be ‘‘catalytic” in radical polymerization, while the surface silanols
might inhibit polymerization. It is the latter that is the focus of this
work. Although a possible reaction mechanism for radicals and
surface silanols has been proposed [7], little experimental evidence
has been reported, until recently. We observed that compared to
the surfactant-covered fumed silica, bare fumed silica surface sig-
nificantly reduced the rates of polymerization in emulsion gels
and the molecular masses for the resulting polymers [3]. Similar
results were also observed for precipitated silica [4] and meso-
porous silica [10].
Herein, we report the measurements of the kinetics for the free
radical polymerization of MMA in fumed silica emulsion gels at
room temperature. The initiator, 2,20 -azobisisobutyronitrile (AIBN),
decomposed at room temperature in the presence of surfactants
[2,3,11,12]. Three types of surfactants, cationic, nonionic and anio-
nic surfactants, were used in this study and each had different
effectiveness in blocking surface silanols, as demonstrated in
Fig. 1. By comparing the polymerization kinetics in the presence
of different surfactants, the nature of the inhibition of the fumed
silica in these polymerization systems became exposed.
2. Experimental
MMA was purchased from Aldrich, and purified by passing it
through a base activated alumina column before use. AIBN (East-
man Kodak) was purified by recrystallization from methanol, and
dried in a vacuum oven for 48 h to eliminate any methanol residue.
Surfactants, cetyltrimethylammonium bromide (CTAB, Sigma),
octyl phenoxy polyethoxyethanol (Triton X-100, Sigma) and
sodium dodecyl sulfate (SDS, Polysciences) were used as received.
The CTAB used was high purity so as to eliminate any iodide impu-
rities which could terminate the polymerization [2]. Cab-O-Sil M-5
fumed silica obtained from Cabot Corp., with a specific surface area
of 200 m2/g, was used. The fumed silica did not contain chloride
impurities as no precipitate was observed after adding aqueous sil-
ver nitrate after it was boiled in deionized water for 1 h.
Emulsion gels were prepared by mixing the chemicals using the
following procedure: 2.0 ml MMA, 0.10 g fumed silica and 0.04 g
AIBN were mixed first, and then 0.5 ml aqueous solution of surfac-
tant (prepared using pH = 7 reverse osmosis water) was added. The
pH of the surfactant solution was found to also be pH = 7. The mix-
tures were shaken for 1 min with a vortex mixer to form stable
emulsion gels. The emulsion gels were placed in capped glass vials,
and thread sealed with Teflon tape. The appearance and physical
properties of these emulsion gels can be found in literature [2,3].
The vials containing the emulsions were placed in a dark envi-
ronment and the polymerizations were conducted at laboratory
temperature, 21 °C. At different times, aliquots of the polymerizing
samples were taken out and used to determine the conversion of
MMA. Those samples were dried with air flow at ambient condi-
tions and then under vacuum for 24 h to remove water and unre-
acted monomers. The masses of the dried samples were measured,
and the conversions of MMA were calculated by subtracting the
masses of fumed silica and surfactants. At small conversions, the
rate of polymerization Rp was calculated from Eq. (1),
d½M D½M
Rp ¼  ¼ ð1Þ
dt Dt
where D[M] is the concentration of monomer that was converted to
polymer in the period Dt. Since the monomer conversion was fairly
linear at the early stage of free radical polymerization [12], the rate
of polymerization (Rp) was determined from the slope of the mono-
mer conversion verses time line. The rates of polymerization may
vary with time, therefore, the rates of polymerization represent
averages over the time periods studied. The largest conversion
was 20%, with most being less than 10% by mass.
The decomposition of AIBN was conducted at room tempera-
ture in the emulsion gels whose compositions were the same as
the polymerizing samples, except that the oil phase, MMA, was
replaced by toluene which was non-polymerizable under these
conditions. At different times, portions of the emulsion gel samples
were converted into microemulsions for UV spectra measurement
[11]. For the emulsion gels formed using Triton X-100, the
microemulsions were made through the addition of Triton X-100
aqueous solution, to bring their compositions into the microemul-
sion range, toluene/Triton X-100/water 45/40/15 by mass [13]. For
the emulsion gels formed with CTAB or SDS, a co-surfactant,
1-butanol, was added, and the microemulsion compositions (tolu
ene/surfactant/water/1-butanol) were adjusted to be 55/15/9/21
by mass for CTAB or SDS samples [11]. All microemulsion samples
were centrifuged for 20 min before the UV spectra measurement.
The UV spectra of the microemulsion samples were taken with
a Cary 5000 UV/visible spectrophotometer at room temperature.
The absorption spectra at 344–346 nm were used for the azo
absorption [11], and the decomposition rate constant for AIBN,
kd, was determined from a first-order decomposition plot, or
At ¼ A0 expðkd tÞ ð2Þ
where At and A0 are the absorbance for the azo compounds in the
microemulsions at time, t and t = 0, respectively.
3. Results and discussion
Gelation in the fumed silica-MMA emulsions was sensitive to
the type and the concentration of the surfactant. For the cationic
surfactant CTAB, the MMA emulsions gels became viscous emul-
sions when the concentration of CTAB in the aqueous solution
reached around 0.7 M [3]. This critical concentration decreased
slightly to 0.5 M for the nonionic surfactant, Triton X-100, and fur-
ther decreased to 0.2 M for the anionic surfactant, SDS. In order to
compare the polymerization kinetics for different surfactants with
similar compositions, a constant concentration of surfactant, 0.1 M
was used. At this surfactant concentration, the polymerizing sam-
ples with all three surfactants were in the gel forms.
Free radical polymerizations in silica-containing emulsion gels
occurred through enhanced AIBN decomposition at room temper-
ature [11]. The logarithm of the At/A0 for AIBN as a function of time
in the emulsion gels formed with different surfactants is plotted in
Fig. 3. The decomposition rate constants for AIBN at room temper-
ature, kd, determined from Eq. (2) for all three surfactants (CTAB,
Triton-X-100 or SDS) were found to be very close to each other,
around 2.0  107 (±0.2  107) s1, for the emulsion gels. In the
other words, the kd for AIBN was independent of the surfactant
type, which agreed with an earlier observation that the emulsion
polymerization initiated by AIBN at room temperature behaved
similarly in the presence of CTAB, SDS or a nonionic surfactant
when fumed silica was absent [14].
 T. Zhang, F.D. Blum / Journal of Colloid and Interface Science 504 (2017) 111–114 113

Fig. 1. Interactions between surfactant (CTAB, Triton X-100 or SDS) and fumed silica surface when pH > 2. For Triton X-100, n = 9 or 10.

The conversions of MMA to PMMA in silica-containing emulsion 0.00

gels at room temperature are shown in Fig. 3. Without surfactant, CTAB
virtually no polymerization occurred for MMA because the decom- Triton X-100
-0.04
position of AIBN at room temperature occurs very slowly in the SDS
absence of surfactants. Polymerization of MMA was observed for
-0.08
all of the surfactant-containing samples. Based on Eq. (1), the Rp
Log (A t / A0)

values were found to be 4.9 ± 0.2, 2.1 ± 0.4 and 1.1 ± 0.1 mmol L1 -
-0.12
h1 for the emulsion gels containing CTAB, Triton X-100 and SDS,
respectively. These surfactant-catalyzed Rp values at room temper-
ature were very small, compared to that measured for the bulk -0.16
polymerization at 60 °C, around 400 mmol L1 h1, with the same
concentrations of the monomer and AIBN because of the much -0.20
lower concentrations of surfactants.
The differences in Rp for different surfactants are interesting -0.24
because, as mentioned above, the kd for AIBN and the Rp for the 0 50 100 150 200 250 300
polymerizations without fumed silica were independent of the sur- Time (h)
factant type. By extrapolating the polymerization line to the time-
axis (zero conversion) in Fig. 3, it was found that the polymeriza- Fig. 2. Log (At/A0) as a function of time for AIBN in the silica-containing emulsion
gels formed by CTAB, Triton X-100 or SDS at room temperature.
tion with CTAB started almost as soon as the surfactant aqueous
solution was added (that line goes through the origin). For Triton
X-100 and SDS, the extrapolations crossed the time-axis around
48 h. Thus, the free radicals generated in the first 48 h did not ini-
tiate the polymerization. If the radicals were formed as demon- 30
strated in Fig. 2, they must have been consumed by an inhibiting CTAB
component in the silica-containing emulsion gels found in the sys- Triton X-100
25
tems with Triton X-100 or SDS surfactants. SDS
The fumed silica surface played an important role in the radical- No Surfactant
Conversion (%)

inhibiting reactions. The types of surfactants did not affect the 20

kinetics for either the decomposition of the initiator in emulsion
gels, or the polymerization in the absence of fumed silica [14], as 15
noted. The different types of surfactants, however, could modify
the fumed silica surfaces to different extents. Since the fumed silica
surface is negatively charged in an aqueous environment when the 10
pH is greater than 2 [15], cationic surfactants, such as CTAB, will be
readily adsorbed onto the fumed silica surface [16]. The non-ionic 5
surfactant Triton X-100 is able to hydrogen-bond to the fumed sil-
ica surface. The adsorption of Triton X-100 on the fumed silica in
0
the MMA emulsion gels, however, is expected to be more limited,
0 100 200 300 400
because much of the fumed silica surface may form hydrogen-
bonds with MMA, due to the competitive adsorption between Time (h)
MMA and Triton X-100 on fumed silica surfaces. The anionic sur-
Fig. 3. The monomer-polymer conversion (% by mass) of MMA in bulk (no
factant, SDS would not be expected to adsorb on the fumed silica surfactant) and in the presence of CTAB, Triton X-100 or SDS silica-containing
surface due to its electrostatic repulsion. Thus, we expect that in emulsion gels at room temperature for small conversions.
114 T. Zhang, F.D. Blum / Journal of Colloid and Interface Science 504 (2017) 111–114
the different systems, the fumed silica surfaces were covered by
CTAB, but there was probably much less surface coverage by SDS
and Triton X-100 [17,18]. For these systems (SDS and Triton X-
100), the silanols on the fumed silica surfaces would be unblocked
and could act as inhibitors.
A possible mechanism for the inhibition from the silanols on
fumed silica in the emulsion gels, based on a degenerative radical
transfer route [7], is proposed as follows: On the fumed silica sur-
face, H atoms from the silanols are transferred to the free radicals
generated from the decomposition of AIBN, and consequently a
surface radical formed on the fumed silica surface. These surface
radicals are too stable to react with the monomers in the emulsion
gels, leading to an inhibition period for the polymerization. The
inhibition due to the silica was suppressed in the CTAB system
because adsorbed CTAB blocked the surface silanols. In the MMA
emulsion gels with CTAB, the fumed silica surface was not com-
pletely covered by CTAB [3]. The lack of an inhibition period for
fumed silica incompletely covered by CTAB suggested that the sila-
nols that were blocked by CTAB were likely more efficient inhibi-
tion sites than other silanols.
Given the density of the silanols on fumed silica, 2.8 SiOH/nm2
[19], the concentration of the silanols in the emulsion gels was
estimated to be 37 mM. The concentration of the free radicals gen-
erated in the first 48 h (inhibition period) were expected to be
about 6.7 mM, calculated from the initial concentration for the
AIBN in the emulsion gels (97 mM). According to the mechanism
proposed above, in which one silanol reacts with one radical, the
fraction of silanol on fumed silica that were highly efficient at
free-radical inhibition during the inhibition period, was estimated
to be 0.18. If a low-temperature initiator efficiency of AIBN, <0.4
[20], was taken into account, the fraction of the highly efficient
silanols participating in inhibition would be smaller than 0.07.
After the inhibition period, some of the remaining silanols would
decrease the rate of polymerization by transferring H atoms to
the propagating chains, and consequently decreasing the molecu-
lar masses of the resulting polymers [3]. This inhibiting reaction
may also be responsible for the unexpected radical-equilibrium
disturbance and less well-controlled polydispersity from living
radical polymerization in the presence of silanol groups [10].
Inhibition due to the silanols was not observed for polymeriza-
tion initiated at higher temperature (60 °C) [7]. The reason for this
higher temperature behavior was likely due to the faster initiation
rate, which generated more free radicals. For example, the kd for
AIBN at 60 °C is on the order of 105 s1 [11], and is much larger
than that at room temperature, 2.0  107 s1 in this study. If
one assumes that all the silanols on fumed silica can inhibit free
radicals, the presence of fumed silica would only be expected to
reduce the Rp by a few percent at 60 °C, which might not be distin-
guishable in a kinetics measurement.
4. Conclusions
Emulsion gels may play important roles in the preparation of
polymerized emulsions and reinforced composites especially since
these can be made at lower temperatures than conventional free
radical polymerization [2,3,12]. These low temperature polymer-
izations are made possible by the enhanced decomposition of the
free-radical initiator, AIBN, in the presence of surfactant [11]. Poly-
merizations may occur in the presence or absence of silica [11,14].
We have also shown that fumed silica, which is known to be
active in free radical reactions [6,8,9], it acts as a mild inhibitor
in free radical polymerizations. The silanols on fumed silica are
the inhibition sites, and a fraction of them, a fraction of <0.07,
may be reactive enough to inhibit free radicals at room tempera-
ture. The inhibition effect in fumed silica is not strong enough to
dramatically affect radical polymerizations with large initiation
rates, such as those conducted at 60 °C or above. A similar inhibi-
tion effect can also be expected for other metal oxide particles with
surface hydroxyl groups. If a slow initiation occurs for an in-situ
polymerization, these metal oxide surface hydroxyl groups on
metal oxide particles need to be blocked prior to the polymeriza-
tion to avoid the free radical inhibition.
Acknowledgement
The authors acknowledge the financial support of the Army
Research Office under Award No. W911NF-10-1-0476 – United
States and Oklahoma State University – United States.
References
[1] V.M. Gun’ko, I.F. Mironyuk, V.I. Zarko, V.V. Turov, E.F. Voronin, E.M. Pakhlov, E.
V. Goncharuk, R. Leboda, J. Skubiszewska-Zieba, W. Janusz, S. Chibowski, Y.N.
Levchuk, A.V. Klyueva, Fumed silicas possessing different morphology and
hydrophilicity, J. Colloid Interf. Sci. 242 (2001) 90–103.
[2] T. Zhang, G. Xu, Z.F. Li, O. Regev, M. Maddumaarachchi, F.D. Blum, PS/CTAB/
silica composites from room temperature polymerization of high internal
phase emulsion gels, J. Colloid Interf. Sci. 451 (2015) 161–169.
[3] T. Zhang, G. Xu, O. Regev, F.D. Blum, Low-temperature polymerization of
methyl methacrylate emulsion gels through surfactant catalysis, J. Colloid
Interf. Sci. 461 (2016) 128–135.
[4] M. Sadej, E. Andrzejewska, B. Kurc, H. Gojzewski, T. Jesionowski, Surface-
dependent effect of functional silica fillers on photocuring kinetics of hydrogel
materials, J. Polym. Sci. Part A Polym. Chem. 52 (2014) 3472–3487.
[5] E. Trommsdorff, H. Kohle, P. Lagally, Zur polymerisation des
methacrylsauremethylesters, Macromol. Chem. Phys. 1 (1948) 169–198.
[6] N. Garia, T. Corrales, J. Guzman, P. Tiemblo, Understanding the role of
nanosilica particle surfaces in the thermal degradation of nano silica-poly
(methyl methacrylate) solution-blended nanocomposites: From low to high
silica concentration, Polym. Degrad. Stab. 92 (2007) 635–643.
[7] M.M. Demir, M. Memesa, P. Castignolles, G. Wegner, PMMA/zinc oxide
nanocomposites prepared by in-situ bulk polymerization, Macromol. Rapid
Commun. 27 (2006) 763–770.
[8] Y.H. Hu, C.Y. Chen, C.C. Wang, Viscoelastic properties and thermal degradation
kinetics of silica/PMMA nanocomposites, Polym. Degrad. Stab. 84 (2004) 545–
553.
[9] W. Viratyaporn, R.L. Lehman, Effect of nanoparticles on the thermal stability of
PMMA nanocomposites prepared by in situ bulk polymerization, J. Therm.
Anal. Calorim. 103 (2011) 267–273.
[10] K. Khezri, H. Roghani-Mamaqani, Effect of MCM-41 nanoparticles on ARGET
ATRP of styrene: Investigating thermal properties, J. Compos. Mater. 49 (2015)
1525–1535.
[11] G. Xu, R.R. Nambiar, F.D. Blum, Room-temperature decomposition of 2,20 -
azobis(isobutyronitrile) in emulsion gels with and without silica, J. Colloid
Interf. Sci. 302 (2006) 658–661.
[12] G. Xu, F.D. Blum, Surfactant-enhanced free radical polymerization of styrene in
emulsion gels, Polymer 49 (2008) 3233–3238.
[13] M. Almgren, J. Vanstam, S. Swarup, J.E. Lofroth, Structure and transport in the
microemulsion phase of the sysytem triton x-100-toluene-water, Langmuir 2
(1986) 432–438.
[14] K. Tauer, N. Oz, Interfacial energy promotes radical heterophase
polymerization, Macromolecules 37 (2004) 5880–5888.
[15] T. Pradubmook, J.H. O’Haver, P. Malakul, J.H. Harwell, Effect of pH on
adsolubilization of toluene and acetophenone into adsorbed surfactant on
precipitated silica, Colloid Surf. A 224 (2003) 93–98.
[16] T. Zhang, G. Xu, J. Puckette, F.D. Blum, Effect of silica on the structure of
cetyltrimethylammonium bromide, J. Phys. Chem. C 116 (2012) 11626–11634.
[17] X.L. Zhang, D. Taylor, R. Thomas, J. Penfold, I. Tucker, Modifying the adsorption
properties of anionic surfactants onto hydrophilic silica using the pH
dependence of the polyelectrolytes PEI, ethoxylated PEI, and polyamines,
Langmuir 27 (2011) 3569–3577.
[18] S.K. Parida, S. Dash, S. Patel, B.K. Mishra, Adsorption of organic molecules on
silica surface, Adv. Colloid Interf. Sci. 121 (2006) 77–110.
[19] C.H.C. Liu, G.E. Maciel, The fumed silica surface: A study by NMR, J. Am. Chem.
Soc. 118 (1996) 5103–5119.
[20] M. Buback, B. Huckestein, F.D. Kuchta, G.T. Russell, E. Schmid, Initiator
efficiency in 2,2’-azoisobutyronitrile-initiated free-radical polymerizations of
styrene, Macromol. Chem. Phys. 195 (1994) 2117–2140.