1 Supporting Information Surficial siloxane-to-silanol interconversion during room temperature hydration/dehydration of amorphous silica films observed by ATR-IR and TIR-Raman spectroscopies Suzanne L. Warring a , David A. Beattie b and A. James McQuillan a * a Department of Chemistry, University of Otago, P. O. Box 56, Dunedin 9054, New Zealand. b Future Industries Institute, University of South Australia, Mawson Lakes, Adelaide SA 5095, Australia. Contents Central network force model Energy dispersive spectroscopy (EDS), dynamic light scattering (DLS) and zeta-potential of colloidal silica Tables of silica IR spectral assignments IR spectra of ~300 nm silica film at different RH relative to bare prism background SEM images and IR spectrum of ~500 nm silica film Variation of effective film thickness with refractive index, wavelength and with humidity Band fitting of TIR Raman spectra References
13
Embed
Supporting Information Surficial siloxane-to-silanol interconversion ...
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
1
Supporting Information
Surficial siloxane-to-silanol interconversion during room
temperature hydration/dehydration of amorphous silica films
observed by ATR-IR and TIR-Raman spectroscopies
Suzanne L. Warringa, David A. Beattie
b and A. James McQuillan
a*
a Department of Chemistry, University of Otago, P. O. Box 56, Dunedin 9054, New Zealand.
b Future Industries Institute, University of South Australia, Mawson Lakes, Adelaide SA
5095, Australia.
Contents
Central network force model
Energy dispersive spectroscopy (EDS), dynamic light scattering (DLS) and zeta-potential of colloidal
silica
Tables of silica IR spectral assignments
IR spectra of ~300 nm silica film at different RH relative to bare prism background
SEM images and IR spectrum of ~500 nm silica film
Variation of effective film thickness with refractive index, wavelength and with humidity
Band fitting of TIR Raman spectra
References
2
Central network force model
The central network force model was first derived by Galeener et al.4 This model assumes
local order consisting of two neighbouring SiO4 tetrahedra sharing a bridged oxygen within
the amorphous structure and correlates the vibrational frequency of the TO mode of the silica
lattice (ω) to the Si-O force constant (k) and Si-O-Si inter-tetrahedral bond angle (θ) through
Equation S1 where mO and mSi are the masses of oxygen and silicon atoms, respectively.
Previous studies have found that higher wavenumber TO modes indicate smaller
intertetrahedral bond angles and less porous structures.5,11
�� =�
��
�1 − �� � +4
3
�
���
(S1)
Energy dispersive spectroscopy (EDS), dynamic light scattering (DLS) and
zeta-potential of colloidal silica
Particle size distribution and zeta-potential measurements were performed on the Zetasizer
Nano ZS90, Malvern, UK with 173° back scatter. Samples were measured at 298 K and
allowed to equilibrate for 120 s. Zeta-potential and size measurements were performed
concurrently using a folded capillary cell. A refractive index of 1.48 from Khlebtsov et al8
was used in the Smoluchowski fitting of size data. For a 4.5 mg mL-1
Ludox suspension pH
~9 the average hydrodynamic diameter (3 measurements) of the particles was 105 nm and the
zetapotential was ~-51 mV.
3
Figure S1 – DLS size data for 4.5 mg mL-1
Ludox suspension pH ~9.
EDS analysis was performed concurrent to recording SEM images of the silica films (Figure
S6). This shows the presence of silicon, oxygen and sodium. The Ludox Syton HT-50 silica
nanoparticles are formed by cationic exchange between sodium silicate and silicic acid,
performed at high temperature in basic aqueous suspension.10
4
Figure S2 - EDS spectrum of silica particle films
Table S1 - IR band assignments for Figure 5 aqueous films and D2O films.
Wavenumber / cm-1 IR band assignment
1–7
3736 Isolated ν(SiO-H)
3647 H-bonded ν(SiO-H)
3400
2756
2688
ν(H2O)
Isolated ν(SiO-D)
H-bonded ν(SiO-D)
2480
1880
ν(D2O)
Overtone νas(Si-O-Si)TO
1870
1188 νas(Si-O)LO
1073
1064
νas(Si-O-Si)TO
966 ν(Si-OH)
5
800 νs(Si-O)
Table S2 - IR band assignments for Figure 6 aqueous film and Figure 7 D2O
film hydration spectra.
Wavenumber / cm-1 IR band assignment
3736 Isolated ν(SiO-H)
3624 H-bonded ν(SiO-H)
3400
3200
2754
2688
2500
ν(H2O)
Isolated ν(SiO-D)
H-bonded ν(SiO-D)
ν(D2O)
1630
1192
δ(H2O)
δ(D2O)
1112
1107
1060
νas(Si-O)TO
1033
1032
980
νas(Si-O)
ν(Si-OD)
956 ν(Si-OH)
888
876
νs(Si-O)
6
Figure S3 - Infrared spectra of silica film exposed initially to RH of 0% (dry argon) then
subsequently to RH of 8, 15, 30 and 40%. Spectral background was bare diamond prism.
Influence of humidity on refractive index and effective sampling thickness
Ellipsometric studies have shown that as humidity increases from ~3 to 40 % the refractive
index of sol-derived silica films shows an increase from ~1.3 to 1.34, dependent on the sol
type.9 Due to the nature of total internal reflection this change causes an increase in
penetration depth (dp) and the effective sampling thickness (de) of the evanescent wave.15
Based on refractive index measurements performed by Rouse et al9 on various sol-derived
silica films the variation in dp and de for silica Sol A has been calculated as shown in Figure
S4.
7
Figure S4 – Variation of effective thickness of evanescent wave with refractive index of the
rarer medium.
At ~1100 cm-1
dp is ~900 nm for RH 3 % and ~925 nm for RH 40 %, giving a thickness
increase of ~3 %. Effective sampling thickness of the evanescent wave (de) at 1100 cm-1
increases from ~1.19 µm to ~1.28 µm, for RH change from 3 to 40 %, an increase of ~91 nm
(~8 %).
The strong absorption in the spectral range of 1100 – 1000 cm-1
could cause anomalous
dispersion9 in this spectral region. This effect causes absorption to be increased on the longer
wavelength side of the absorption envelope and decreased on the shorter wavelength side,
akin to spectral results shown in Figure 6. Films with thickness << dp give ATR-IR spectra
comparable to transmission sampling and the effect of anomalous dispersion is less evident.
In the present results dp is ~1 µm at 1100 cm-1
and the film thickness is ~300 nm, so
anomalous dispersion will have some effect on the 1100 cm-1
spectral envelope.
8
Figure S5 - Variation of effective field thickness with wavelength and with relative humidity.
SEM images and IR spectrum of ~500 nm silica film
Figure S6 - SEMs and IR spectrum of ~500 nm thick silica film formed from 8.4 mg mL-1
Ludox pH 2.5 suspension (a) film morphology, (b) film thickness and (c) IR spectrum.
9
Band Fitting of TIR Raman spectra12-14 in 1250-700 cm
-1 region and second
derivatives f′′(x)
Figure S7 – Deconvolution of TIR-Raman spectrum of a silica film formed from ~80 mg mL
-
1 pH 2.5 silica suspension at RH of ~40 %.
10
Figure S8 – Deconvolution of TIR-Raman spectrum of film formed from ~80 mg mL-1
pH
~9.5 silica suspension at RH of ~40 %.
Figure S9 – Deconvolution of TIR-Raman spectrum of a silica film formed from ~80 mg mL-
1 pH ~9.5 silica suspension at RH of ~0 % (dry argon).
11
Peak deconvolution was performed for the TIR-Raman, spectra were first smoothed using a
27-point Sav-Golvay function. Fitting utilized a combined Gaussian-Lorentzian function,
using the second derivative of the spectrum to determine peak positions. The deconvoluted
spectra show evidence of a minor absorption between the peaks at ~1060 and ~960 cm-1
.
From the deconvolution analysis for pH 10 films there is a small peak at ~1020 cm-1
whereas
for the pH 2.5 film this band has shifted down to 1009 cm-1