Atmospheric degradation of volatile methyl-silicon compounds

Article abstract of DOI:10.1021/es9910053

The current widespread use of poly(dimethylsiloxane)s (PDMS) in a broad range of applications leads to their release into soil environments where they degrade to monomeric products, primarily dimethylsilanediol, most of which enter the atmosphere by volatilization. The major degradation pathway of volatile organosilicon compounds in the atmosphere is expected to be a reaction with hydroxyl (OH) radicals. In this work, the kinetics of the gas- phase reactions of dimethylsilanediol, trimethylsilanol, and tetramethylsilane with the OH radical were measured using a relative rate method which employed the N₂H₄ + O₃ reaction as a nonphotolytic source of OH radicals, with analysis by Fourier transform infrared (FT-IR) spectroscopy in a 5870 L chamber. The measured values of the OH radical reaction rate constants (cm³ molecule^-1 s^-1) at 298 ± 2 K are as follows: dimethylsilanediol, (8.1 ± 1.0) x 10^-13; trimethylsilanol, (7.2 ± 0.8) x 10^-13; and tetramethylsilane, (8.5 ± 0.9) x 10^-13. These values lead to an estimate of tropospheric lifetimes with respect to reaction with the OH radical of ca. 15 days for these organosilicon compounds. FT-IR spectroscopy and atmospheric pressure ionization mass spectrometry (API-MS) were employed to analyze the products of OH radical- and Cl atom-initiated photooxidations of dimethylsilanediol and trimethylsilanol. Infrared signatures of the probable formate ester intermediate products from both silanols were detected. API-MS analyses indicated the formation of methylsilanetriol from dimethylsilanediol, of both dimethylsilanediol and methylsilanetriol from trimethylsilanol, and of the corresponding intermediate formate esters. Possible reaction mechanisms are discussed. The current widespread use of poly(dimethylsiloxane)s (PDMS) in a broad range of applications leads to their release into soil environments where they degrade to monomeric products, primarily dimethylsilanediol, most of which enter the atmosphere by volatilization. The major degradation pathway of volatile organosilicon compounds in the atmosphere is expected to be a reaction with hydroxyl (OH) radicals. In this work, the kinetics of the gas-phase reactions of dimethylsilanediol, trimethylsilanol, and tetramethylsilane with the OH radical were measured using a relative rate method which employed the N₂H₄ + O₃ reaction as a nonphotolytic source of OH radicals, with analysis by Fourier transform infrared (FT-IR) spectroscopy in a 5870 L chamber. The measured values of the OH radical reaction rate constants (cm³ molecule^-1 s^-1) at 298 ± 2 K are as follows: dimethylsilanediol, (8.1 ± 1.0) × 10^-13; trimethylsilanol, (7.2 ± 0.8) × 10¹³; and tetramethylsilane, (8.5 ± 0.9) × 10^-13. These values lead to an estimate of tropospheric lifetimes with respect to reaction with the OH radical of ca. 15 days for these organosilicon compounds. FT-IR spectroscopy and atmospheric pressure ionization mass spectrometry (API-MS) were employed to analyze the products of OH radical- and Cl atom-initiated photooxidations of dimethylsilanediol and trimethylsilanol. Infrared signatures of the probable formate ester intermediate products from both silanols were detected. API-MS analyses indicated the formation of methylsilanetriol from dimethylsilanediol, of both dimethylsilanediol and methylsilanetriol from trimethylsilanol, and of the corresponding intermediate formate esters. Possible reaction mechanisms are discussed.

Full text of DOI:10.1021/es9910053

Environ. Sci. Technol. 2000, 34, 1970-1976
between environmental compartments (air, soil, and water),
Atmospheric Degradation of Volatile
Methyl-Silicon Compounds
E R N E S T O C . T U A Z O N , * , †
S A R A M . A S C H M A N N , † A N D
R O G E R A T K I N S O N * , † , ‡
the processes which influence their concentrations, and
their ultimate fate. The predicted mass flow of PDMS in the
United States is predominantly to the soil environment as
a component of sludge (3), where they undergo soil-catalyzed
hydrolysis to produce dimethylsilanediol as the major product
(3). The major fraction of the dimethylsilanediol produced
in soil is expected to migrate into the atmosphere, since in
several types of soil its volatilization has been shown to
dominate over biodegradation (4). Hence, a knowledge of
the rate and mechanism by which dimethylsilanediol is
transformed in the atmosphere is required for a complete
assessment of the impact of PDMS release into the environ-
ment.
Air Pollution Research Center and Departments of
Environmental Sciences and Chemistry, University of
California, Riverside, California 92521
The current widespread use of poly(dimethylsiloxane)s
(PDMS) in a broad range of applications leads to their
release into soil environments where they degrade
to monomeric products, primarily dimethylsilanediol, most
of which enter the atmosphere by volatilization. The
major degradation pathway of volatile organosilicon
compounds in the atmosphere is expected to be a reaction
with hydroxyl (OH) radicals. In this work, the kinetics of
the gas-phase reactions of dimethylsilanediol, trimethylsilanol,
and tetramethylsilane with the OH radical were measured
Reaction with the hydroxyl (OH) radical has been shown
to be the main atmospheric degradation pathway for a
number of volatile methyl siloxanes and silanols (5), and
OH radical reaction rate constants for a number of organo-
silicon compounds, including (CH3)3SiOH and (CH3)3SiOSi-
(CH3)2OH, have been reported (5-9). However, no similar
measurement has been carried out for dimethylsilanediol.
Hence, in this work we have measured the rate constant for
reaction of the OH radical with dimethylsilanediol and carried
out similar measurements for tetramethylsilane (to test the
experimental technique) and trimethylsilanol. The products
of the atmospheric reactions of silylalkoxy radicals, formed
after H-atom abstraction by OH radicals and/ or Cl atoms
from dimethylsilanediol and trimethylsilanol, were also
investigated.
using a relative rate method which employed the N2H
4
+ O reaction as a nonphotolytic source of OH radicals,
3
with analysis by Fourier transform infrared (FT-IR)
spectroscopy in a 5870 L chamber. The measured values
of the OH radical reaction rate constants (cm3 molecule-1
s-1) at 298 ( 2 K are as follows: dimethylsilanediol, (8.1
( 1.0) × 10-13; trimethylsilanol, (7.2 ( 0.8) × 10-13; and
tetramethylsilane, (8.5 ( 0.9) × 10-13. These values lead
to an estimate of tropospheric lifetimes with respect to reaction
with the OH radical of ca. 15 days for these organosilicon
compounds. FT-IR spectroscopy and atmospheric
Experimental Section
Experiments were carried out in two reaction chambers: a
5870 L evacuable, Teflon-coated chamber containing an in
situ multiple-reflection optical system interfaced to a Nicolet
7199 Fourier transform infrared (FT-IR) absorption spec-
trometer, with irradiation being provided by a 24 kW xenon
arc lamp filtered through a 6.4 mm thick Pyrex pane to remove
wavelengths <300 nm and a 7500 L Teflon chamber equipped
with two parallel banks of blacklamps and interfaced to a PE
SCIEX API III MS/ MS direct air sampling, atmospheric
pressure ionization tandem mass spectrometer (API-MS).
Both chambers are fitted with Teflon-coated fans to ensure
rapid mixing of reactants during their introduction to the
chambers.
Kinetic Studies. Kinetic experiments were carried out in
the 5870 L chamber at 298 ( 2 K and 740 Torr of dry synthetic
air (80% N2 + 20% O2), with analysis by FT-IR absorption
spectroscopy. The OH radical rate constants of dimethyl-
silanediol, trimethylsilanol, and tetramethylsilane were mea-
sured using a relative rate method which employed the
reaction of hydrazine with ozone as the OH radical source
(10) and dimethyl ether as the reference compound. The use
of the fast N2H4 + O3 reaction (10) to generate OH radicals
enabled short-duration experiments to be carried out which
allowed corrections for the autodecay of dimethylsilanediol
to be reliably made (cf. later). For these experiments, aliquots
of hydrazine were introduced (with mixing) to O3-organo-
silicon-dimethyl ether-air mixtures, with the N2H4 + O3
reaction at each stage being essentially complete during the
2-3 min mixing time. Typical initial concentrations (molecule
cm-3) were as follows: organosilicon compound, (1.2-1.8)
× 1014; dimethyl ether, 1.8 × 1014; and O3, 4.8 × 1014; with
each N2H4 aliquot corresponding to a concentration in the
chamber of (1.0-2.4) × 1014 molecule cm-3 . The decay rates
of the organosilicon compounds in the O3-organosilicon-
dimethyl ether-air mixtures were monitored by FT-IR
pressure ionization mass spectrometry (API-MS) were
employed to analyze the products of OH radical- and Cl atom-
initiated photooxidations of dimethylsilanediol and
trimethylsilanol. Infrared signatures of the probable
formate ester intermediate products from both silanols
were detected. API-MS analyses indicated the formation
of methylsilanetriol from dimethylsilanediol, of both
dimethylsilanediol and methylsilanetriol from trimethylsilanol,
and of the corresponding intermediate formate esters.
Possible reaction mechanisms are discussed.
Introduction
The rapid growth of the silicone industry and resulting
widespread use of poly(dimethylsiloxane)s (PDMS) [(CH3)3-
SiO[Si(CH3)2O]nSi(CH3)3] in a broad range of industrial,
healthcare, and consumer applications has resulted in the
release of these substances into the environment. An over-
view of the scope of usage and relative environmental
loadings of commercial silicon materials, including PDMS,
and the possible pathways of their entry into the environment
is given by the recent articles of Allen et al. (1) and Chandra
et al. (2). Fendinger et al. (3) discuss the partitioning of PDMS
* Corresponding author phone: (714)787-5140; fax: (909)787-5004;
e-mail: ernesto.tuazon@ucr.edu.
† Air Pollution Research Center.
‡ Departments of Environmental Sciences and Chemistry.
9
1 9 7 0 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 34, NO. 10, 2000
10.1021/es9910053 CCC: $19.00
2000 Am erican Chem ical Society
Published on Web 04/05/2000

absorption spectroscopy for 20-60 min prior to the intro-
duction of N2H4. The kinetic experiments consisted of
recording spectra of the initial mixture and of the reaction
mixture after each N2H4 injection. In these experiments, and
those for product studies, infrared spectra were recorded
using a path length of 62.9 m, a full-width at half-maximum
resolution of 0.7 cm-1, and a 2-min averaging time (64 scans/
interferograms).
The applicability of the N2H4 + O3 reaction as a non-
photolytic source of OH radicals for OH rate constant
measurements has previously been validated (10, 11). Where
the only loss processes for the organosilicon and the dimethyl
ether reference compound are due to their reactions with
OH radicals
generator at a precalibrated settings of voltage and input
flow of high purity O2 (Puritan-Bennett Corp., 99.994%).
Dimethylsilanediol was prepared by a procedure similar
to that described by Kantor (14). All vessels employed in the
preparation, handling, and storage of the product were acid-
washed and rinsed thoroughly before use to minimize
condensation reactions known to be catalyzed by traces of
acids and bases (14). The infrared spectrum of the solid
prepared as a Nujol mull agreed with that reported by Kantor
(14), and API-MS analysis of a vapor sample showed the
presence of the protonated [M + H]+ ion at 93 u.
Trimethylsilanol was prepared by heating a mixture of
methoxytrimethylsilane [(CH3)3SiOCH3] (15 g) and distilled
water (50 mL) at ca. 320 K with vigorous stirring for 4 h in
a similar reflux setup used to prepare dimethylsilanediol and
with the same attention given to acid-washing the reaction
and storage vessels. The organic phase was separated, and
a top fraction, presumably containing mostly methanol, was
distilled off and discarded. The remaining fraction was stored
under vacuum at 77 K. The IR spectrum of trimethylsilanol
vapor from this sample is presented below; an API-MS
analysis showed the presence of a protonated [M + H]+ ion
at 91 u. Vapors obtained directly from this sample at room
temperature typically showed a molar composition of 83%
trimethylsilanol, 7% hexamethyldisiloxane, and 10% metha-
nol as deduced from the IR calibrations of the latter two
components and the measured total vapor pressure of the
sample. The trimethylsilanol contents of these vapor samples
were significantly higher than the 33-35% reported by Kantor
(14) for the trimethylsilanol-hexamethyldisiloxane azeotrope
which was obtained by ether extraction and distillation of
the organic layer.
organosilicon + OH f products
(CH3)2O + OH f products
(1)
(2)
then
ln{[organosilicon]t0/ [organosilicon]t} )
(k1/k2) ln{[(CH3)2O]t0/ [(CH3)2O]t} (I)
where [organosilicon]t0 and [(CH3)2O]t0 are the organosilicon
and dimethyl ether concentrations at time t0, respectively;
[organosilicon]t and [(CH3)2O]t are the corresponding con-
centrations at time t; and k1 and k2 are the rate constants for
reactions 1 and 2, respectively.
Product Studies. Chlorine atom-initiated photooxidations
of dimethylsilanediol and trimethysilanol were conducted
in both reaction chambers. The experiments in the 5870 L
chamber with FT-IR analysis were carried out by photolyzing
mixtures of the organosilicon compound and Cl2 in dry
synthetic air at 298 ( 2 K and 740 Torr. The initial
organosilicon and Cl2 concentrations were (1.2-1.8) × 1014
and (2.4-5.1) × 1014 molecule cm-3, respectively. Irradiations
were carried out continuously for 20-40 min, with spectra
being recorded every 5 min. Prior to the above runs, an
exploratory photolysis of an air mixture of dimethylsilanediol,
Except for dimethylsilanediol, the reactants were mea-
sured into calibrated 2 L- and 5 L-Pyrex bulbs with an MKS
Baratron capacitance manometer (100 Torr sensor) and
flushed with a stream of nitrogen with mixing into the
chamber. Dimethylsilanediol was weighed into a stopcocked
Pyrex vial (acid-washed prior to use) which was then
connected to the chamber, and the sample was introduced
with heating and flushing with nitrogen gas.
CH3ONO, and NO with initial concentrations (molecule cm-3
)
of 1.8 × 1014, 2.4 × 1014, and 1.8 × 1014, respectively, was
Results
carried out.
Infrared Spectra and Dark Decay in the 5870 L Cham ber.
Dimethylsilanediol. Vapor samples of (CH3)2Si(OH)2 intro-
duced into the 5870 L chamber showed the initial presence
and growth of another component with time. Figure 1A is
the spectrum of a mixture of this product and (CH3)2Si(OH)2
which was recorded 61 min after introduction of a vapor
The experiments conducted in the 7500 L Teflon chamber
with API-MS analysis consisted of irradiations of each of
dimethylsilanediol and trimethylsilanol with Cl2 at 298 ( 2
K and 740 Torr total pressure of purified air at ca. 5% relative
humidity. The initial organosilicon and Cl2 concentrations
were (2-4) × 1013 and 1.2 × 1014 molecule cm-3, respectively.
Prior to these experiments, an exploratory photolysis of a
mixture of dimethylsilanediol with CH3ONO and NO was
sample (total concentration of 1.3 × 1014 molecule cm-3
)
into the evacuated chamber (<5 Torr N2 diluent) maintained
at room temperature. From the known amount of the sample
and the differences between the spectra with time, the
spectrum of the “decay” product(s) at t ) 61 min and of
dimethylsilanediol at t ) 0 were derived and are shown in
Figure 1B,C, respectively. (The concentration shown for
dimethylsilanediol in Figure 1C was derived assuming that
all components remained in the gas phase.) The OH stretch
frequency of (CH3)2Si(OH)2 is at 3735 cm-1 (not included in
Figure 1). The Q-branch feature at 1265.1 cm-1 allowed
accurate monitoring of the (CH3)2Si(OH)2 concentration
despite its overlap with a broader absorption band by the
“decay” product(s). Figure 1D shows the spectrum of the
solid in Nujol mull (Nujol bands subtracted), and, as noted
above, the band positions agree with those reported by Kantor
(14).
also conducted with initial concentrations (molecule cm-3
)
of 4 × 1013, 2.4 × 1014, and 1.8 × 1014, respectively. The
mixtures were irradiated for periods of 0.5-1.0 min at 100%
of the maximum light intensity. The chamber contents were
sampled through a 25 mm diameter × 75 cm length Pyrex
tube at ∼20 L min-1 directly into the API mass spectrometer
source. MS (scanning) and MS/ MS [with collision activated
dissociation (CAD)] analyses of the mixtures were performed
prior to irradiation and after each illumination period (12).
The instrument was operated in the positive ion mode, and
details of the operation of the API-MS under these conditions
have been described previously (12).
Chem icals. The chemicals used, and their stated purities,
were as follows: tetramethylsilane (99.9+%), dimethoxy-
dimethylsilane (95%), and methoxytrimethylsilane (99%),
Aldrich Chemical Co.; Cl2 (g99.5%) and NO (g99.0%),
Matheson Gas Products. CH3ONO was prepared as described
by Taylor et al. (13) and stored under vacuum at liquid
nitrogen temperature. O3 was produced in a Welsbach T-408
The disappearance of (CH3)2Si(OH)2 was faster in the
evacuated chamber than in the presence of one atmosphere
of air. In both cases, however, the decays could be described
as first order in (CH3)2Si(OH)2, with observed decay rates of
9
VOL. 34, NO. 10, 2000 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 1 9 7 1

FIGURE 3. Least-squares plot of eq I for the gas-phase reaction of
OH radicals with dimethylsilanediol, with dimethyl ether as
FIGURE 1. Vapor-phase IR spectra of (A) (CH ) Si(OH)2 sample after
61 min in evacuated chamber, (B) “decay” product(s), and (C) “pure”
3 2
reference compound. The measured (CH ) Si(OH)2 concentrations
3 2
have been corrected to account for the (CH ) Si(OH)2 “prereaction”
3 2
(CH ) Si(OH)2 derived from the dark decay experiment. (D) Spectrum
3 2
losses using the (CH ) Si(OH)2 decay rates measured prior to addition
3 2
of solid (CH ) Si(OH)2 sample in Nujol mull (Nujol bands subtracted).
3 2
of N2H (see text).
4
Number in parentheses is concentration in molecule cm-3.
air and 298 K showed no measurable dark decay as monitored
by FT-IR spectroscopy over periods of 30-40 min.
No measurable dark decay of Si(CH3)4 in 1 atm of air
diluent at 298 K was observed. The sharp Q-branch feature
of its spectrum at 1255.0 cm-1 (Figure 2C) was employed for
FT-IR analysis.
Kinetic Studies. Initial experiments with (CH3)2Si(OH)2
using the photolysis of CH3ONO as the source of OH radicals
resulted in acutely nonlinear plots of eq I, with the curvature
indicating a marked acceleration of the (CH3)2Si(OH)2 loss
rate as the irradiations progressed. It is likely that the acid
products of this reaction system (i.e., HNO3, HONO) catalyzed
the decay and/ or self-condensation of (CH3)2Si(OH)2, as
discussed later. Hence, the relative rate technique involving
the use of the N2H4 + O3 reaction as the OH radical source
was employed to measure the rate constants for the reactions
of the OH radical with dimethylsilanediol, trimethylsilanol,
and tetramethylsilane.
The observed decay rates of (CH3)2Si(OH)2 in its mixtures
with (CH3)2O and O3 prior to injection of N2H4 were in the
range (0.6-1.3) × 10-2 min-1, and these decay losses (due
to self-reaction and/ or wall losses) were significant fractions
(48-62%) of the total losses after introduction of N2H4. Hence,
only one injection of N2H4 per experiment was made, with
the last spectral record of the “prereaction” decay measure-
ment being used to establish the initial concentrations of
(CH3)2Si(OH)2 and (CH3)2O and the final concentrations being
determined from the next spectrum recorded after N2H4
injection and the mixing time of ca. 3 min duration. The
elapsed time between these two spectra was 6-8 min. No
measurable decays of the reference compound (CH3)2O in
the reaction chamber prior to N2H4 injection were observed.
The data from four separate experiments, corrected for the
“prereaction” decay losses, are plotted according to eq I in
Figure 3, resulting in a slope ) (k1/k2) of 0.285 ( 0.020, with
the indicated error being two least-squares standard devia-
tions.
FIGURE 2. (A) IR spectrum of vapor from a trimethylsilanol sample
showing the presence of methanol and hexamethyldisiloxane
(HMDS) as impurities. (B) (CH ) SiOH spectrum from (A) after
3 3
subtraction of the impurity absorptions. (C) Si(CH ) spectrum.
3 4
Numbers in parentheses are concentrations in 1013 molecule cm-3.
0.015 min-1 at reduced pressure (<5 Torr N2 diluent) and
0.0022 min-1 for the next experiment which was carried out
in 1 atm of air. During the prereaction stages of the kinetic
experiments described below, however, varying (and higher)
dark decay rates of (CH3)2Si(OH)2 in 1 atm of diluent air were
obtained.
Trimethylsilanol and Tetramethylsilane. The spectrum of
a vapor sample from the (CH3)3SiOH preparation with a
molar composition of 83% (CH3)3SiOH, 7% (CH3)3SiOSi(CH3)3,
and 10% CH3OH is shown in Figure 2A. The spectrum of
(CH3)3SiOH with the absorption bands of impurities sub-
tracted is shown in Figure 2B. The OH stretch frequency is
found at 3735 cm-1 (not illustrated). The sharp Q-branch at
1258.4 cm-1 was employed for FT-IR analysis. Trimethyl-
silanol samples in the chamber at atmospheric pressure of
The negligible decay rates of (CH3)3SiOH and Si(CH3)4 in
their respective mixtures in air with (CH3)2O and excess O3
allowed more than one N2H4 injection to be made during an
experiment and, hence, to obtain more data points per
experiment. Plots of eq I for the combined data from three
experiments for (CH3)3SiOH and from two experiments for
9
1 9 7 2 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 34, NO. 10, 2000

TABLE 1. Summary of Rate Constant Ratios, k1/k2, and Rate Constants, k1, for the Gas-Phase Reactions of OH Radicals with
Organosilicon Compounds and Other Selected Organic Compounds at 298 ( 2 K
1012 × k1 (cm3 molecule-1 s-1
)
a
compound
k1/k2
this workb
literature
ref
(CH3)2Si(OH)2
(CH3)3SiOH
Si(CH3)4
0.285 ( 0.020
0.253 ( 0.011
0.301 ( 0.008
0.81 ( 0.10
0.72 ( 0.08
0.85 ( 0.09
3.95 ( 0.65
1.00 ( 0.27
1.28 ( 0.55
1.38 ( 0.36
1.19 ( 0.30
1.89 ( 0.60
0.85 ( 0.10
1.08 ( 0.10
0.99 ( 0.26c
(7)
(6)
(7)
(6)
(7)
(9)
(16)
(17)
(CH3)3SiOSi(CH3)3
(CH3)3SiOSi(CH3)2OH
C(CH3)4
(CH3)3COH
a
b
Reference organic is dim ethyl ether; indicated errors are two least-squares standard deviations. Placed on an absolute basis using
k2(OH + dim ethyl ether) ) 2.84 × 10-12 cm 3 m olecule-1 s-1 (ca. ( 10%) (15). The indicated errors include the ( 10% uncertainty in the rate constant
c
k2. Average of literature values sum m arized in ref 17.
FIGURE 4. Least-squares plot of eq I for the gas-phase reactions
of OH radicals with trimethylsilanol and tetramethylsilane, with
dimethyl ether as reference compound. The ordinates of the
tetramethylsilane plot are offset for clarity.
FIGURE 5. IR spectra of products from the photolysis of a
dimethylsilanediol-CH ONO-NO-air mixture (with absorptions
3
of products from the photooxidation of CH ONO and NO subtracted)
3
Si(CH3)4 are shown in Figure 4, with good straight-line plots
being observed. The k1/k2 values for the three organosilicon
compounds studied are given in Table 1. These rate constant
ratios are placed on an absolute basis with the use of a value
of k2, the rate constant for the (CH3)2O + OH reaction, of 2.84
× 10-12 cm3 molecule-1 s-1 (ca. ( 10%) (15). The resulting
rate constants, k1, are also presented in Table 1.
after irradiation times of (A) 13 min, (B) 31 min, and (C) 71 min. (D)
Vapor-phase spectrum of product obtained from reacting dimethoxy-
dimethylsilane with water (to prepare dimethylsilanediol) without
acid-washing of the reaction vessels (see text).
that (CH3)2Si(OH)2 can undergo during preparation in the
presence of traces of alkali or acid (see ref 14). It is possible
that the significant amounts of HNO3 formed during the
CH3ONO-NO-air photolysis catalyzed the transformation
of (CH3)2Si(OH)2 and its intermediate decay product(s) to
the product(s) represented by Figure 5D.
Product Studies. Dimethylsilanediol. FT-IR Measure-
ments. The reaction of (CH3)2Si(OH)2 with OH radicals was
examined by photolysis of a mixture of (CH3)2Si(OH)2 with
CH3ONO and NO in air. The qualitative results are presented
in Figure 5, where only the infrared spectra of the products
arising from (CH3)2Si(OH)2 itself are depicted, i.e., absorptions
by remaining reactants and by the photooxidation products
from CH3ONO and NO have been subtracted. The sequence
of spectra presented in Figure 5 (parts A and B) shows that
the product formed in the early stage of photolysis is the
same one observed during the dark decay of (CH3)2Si(OH)2
(see Figure 1B), indicating that the reaction with OH radicals
probably did not occur to a significant extent. A different
major product is formed later in the photolysis (Figure 5C).
However, this latter product has a spectrum that is similar
to the product or mixture of products obtained (see Figure
5D) during our first attempts to prepare (CH3)2Si(OH)2. In
those earlier attempts the reaction vessels were not acid-
washed, and the literature warns of the transformation
The results of a Cl-atom initiated photooxidation of (CH3)2-
Si(OH)2, with a mixture containing initial concentrations of
1.3 × 1014 molecule cm-3 of (CH3)2Si(OH)2 and 3.0 × 1014
molecule cm-3 of Cl2 in 1 atm of dry air, are presented in
Figure 6. Figure 6A is the spectrum of the reaction mixture
prior to irradiation. Figure 6 (parts B and D) corresponds to
2 min (41% consumption of (CH3)2Si(OH)2) and 12 min
(∼100% consumption of (CH3)2Si(OH)2) of photolysis, re-
spectively. Figure 6 (parts C and E) is corresponding spectra
where the absorptions of H2O, HC(O)OH, and unreacted
(CH3)2Si(OH)2 have been subtracted. An intermediate product
is indicated in Figure 6 (parts B and C) by absorption bands
at 1721 and 1213 cm-1, most likely the characteristic pair of
CdO stretch and C-C stretch frequencies of an ester group
(18, 19) and attributed to (HO)2Si(CH3)OCHO, which dis-
9
VOL. 34, NO. 10, 2000 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 1 9 7 3

FIGURE 7. API-MS spectra (A) before and (B) after irradiation of
a mixture of dimethylsilanediol and Cl2 in air.
FIGURE 6. IR spectra from a Cl atom-initiated photooxidation of
dimethylsilanediol: (A) initial dimethylsilanediol, (B) reaction
mixture after 2 min of irradiation and 41% consumption of reactant,
(222 + 92 + H]), and 333 ([166 + 166 + H] and/ or [240 + 92
+ H]). Other ion peaks observed (not shown) and their
assignments are 407 ((222 + 92 + 92 + H]), 481 ([240 + 240
+ H] and/ or [240 + 222 + H + H2O]), and 555 u ([240 + 222
+ 92 + H] and/ or [222 + 222 + 92 + H + H2O]).
(C) from (B) after subtraction of absorptions by HC(O)OH, H O, and
2
remaining dimethylsilanediol, (D) reaction mixture after 12 min of
irradiation and 100% loss of reactant, and (E) from (D) after subtraction
of absorptions by HC(O)OH and H O.
2
After 1.5 min irradiation of this Cl2-(CH3)2Si(OH)2-air
mixture, the API-MS spectrum in Figure 7B was obtained.
Most of the ion peaks associated with the condensation
products of (CH3)2Si(OH)2 have disappeared, and the spec-
trum shows the presence of fewer dimer ions than is the case
in Figure 7A. Of special interest in Figure 7B are the presence
of ion peaks at 95, 113, and 131 u, which are attributed to
the [M + H]+, [M + H + H2O]+, and [M + H + 2H2O]+ ions
of a MW 94 product. Additional ion peaks associated with
this MW 94 product are those at 187 ([94 + 92 + H]) and 189
([94 + 94 + H]), and we attribute this MW 94 product to
CH3Si(OH)3. An ion peak at 141 u may be the [M + H + H2O]+
ion of a MW 122 product which could then be the inter-
mediate formate ester CH3Si(OH)2OCHO.
This experiment therefore provides evidence for the Cl-
atom initiated reaction of (CH3)2Si(OH)2 leading to the
formation of CH3Si(OH)3, possibly via the intermediate
formation of the formate ester CH3Si(OH)2OCHO.
The API-MS spectra obtained during the photolysis of a
mixture of dimethysilanediol with CH3ONO and NO showed
no products except the dark decay/ condensation species
enumerated above which were already present prior to
irradiation, in agreement with the results of the experiment
conducted with FT-IR analysis.
Trimethylsilanol. The Cl atom-initiated photooxidation
of trimethylsilanol carried out with FT-IR analysis showed
the formation of an ester-like product, as indicated by a pair
of absorption bands at 1720 cm-1 (CdO stretch) and 1206
cm-1 (C-C stretch) (18, 19). These absorptions overlapped
with the corresponding absorptions of the formate ester
product from the more reactive hexamethyldisiloxane (8),
the latter being an impurity in the trimethylsilanol sample.
However, the 1720 cm-1 band of the trimethylsilanol product
has a sharp Q branch that distinguishes it from the CdO
stretch absorption of the hexamethyldisiloxane product.
Similar to the Cl atom-initiated photooxidation of hexamethyl-
disiloxane (8), formic acid was also observed during the course
of the irradiation. Dimethylsilanediol was not observed by
appeared in the later stage of the photolysis with formation
of an unidentified major product(s) which was accompanied
by an increase in the production of HC(O)OH. Since HCl is
a product of the Cl atom-initiated photooxidation of organic
compounds, it was possible that HCl also catalyzed the self-
condensation of (CH3)2Si(OH)2 in this experiment but to a
much lesser degree than the acid catalysis observed in the
CH3ONO-NO systems because of the much shorter reaction
times involved.
API-MS Measurements. A preirradiation API-MS spectrum
of a mixture of (CH3)2Si(OH)2 and Cl2 in air is presented in
Figure 7A, which shows the presence of (CH3)2Si(OH)2 along
with its dark decay products. Product ion peaks were
identified based on the observation of the protonated homo-
or heterodimer ions (for example, [(MP1)2 + H]+, [(MP2)2 +
H]+, and [MP1 + MP2 + H]+, where P1 and P2 are a product
or a reactant) in the API-MS/ MS “parent ion” spectra and
consistency of the API-MS/ MS “daughter ion” spectrum of
a homo- or heterodimer ion with the “parent ion” spectra
of the various [MP + H]+ peaks (12). Water cluster ion peaks
of the product ions, [M + H + H2O]+, were often observed.
Thus, in Figure 7A, the ion peaks at 93 (weak), 167, 223, and
241 u are assigned to the [M + H]+ ions of (CH3)2Si(OH)2
(molecular weight (MW) ) 92) and its known condensation
products (CH3)2Si(OH)OSi(OH)(CH3)2 (MW ) 166),
(MW ) 222) and (CH3)2Si(OH)OSi(CH3)2OSi(OH)(CH3)2
(MW ) 240) (20). All of the other ion peaks shown in Figure
7A can be assigned to homo- or heterodimers of these MW
92, 166, 222, and 240 species, including 111 ([92 + H + H2O]),
185 ([92 + 92 + H] and/ or [166 + H + H2O]), 203 ([92 + 92
+ H + H2O])]), 241 ([222 + H + H2O] and/ or [240 + H]), 259
([92 + 166 + H], [222 + H + 2H2O] and/ or [240 + H + H2O]),
277 ([92 + 166 + H + H2O] and/ or [240 + H + 2H2O]), 315
9
1 9 7 4 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 34, NO. 10, 2000

formation of ion peaks at 93, 111, and 129 u (attributed to
the [M + H]+, [M + H + H2O]+, and [M + H + 2H2O]+ ions
of a MW 92 product) and at 121 and 139 u (attributed to the
[M + H]+ and [M + H + H2O]+ ions of a MW 120 product).
Additionally, the API-MS spectrum after 1.0 min of irradiation
shows the presence of ion peaks at 95 and 113 u which are
attributed to the [M + H]+ and [M + H + H2O]+ ions of a MW
94 product. After 2.0 min of irradiation (Figure 8D), the ion
peaks associated with (CH3)3SiOH have disappeared, and
the ion peaks present are those at 93, 111, and 129 u (from
the MW 92 product); 121 and 139 u (from the MW 120
product); 95, 113, and 131 u (the [M + H]+, [M + H + H2O]+
and [M + H + 2H2O]+ ions of the MW 94 product); and 123
and 141 u (not shown) (attributed to the [M + H]+ and
[M + H + H2O]+ ions of a MW 122 product). [The 135 and
137 u ion peaks observed in Figure 8 (parts B-D, respectively)
could be the [M + H]+ of M ) (CH3)2Si(OH)OCHO and the
[M + H + H2O]+ of (CH3)3SiOCHO, respectively, formed from
the (CH3)3SiOCH3 impurity.]
We interpret these observations as involving the initial
formation of (CH3)2Si(OH)OCHO (MW 120) and (CH3)2Si-
(OH)2 (MW 92) from the reaction of (CH3)3SiOH, followed by
reaction of the first-generation product (CH3)2Si(OH)2 to form
CH3Si(OH)2OCHO (MW 122) and CH3Si(OH)3 (MW 94).
Discussion
The value of (8.5 ( 0.9) × 10-13 cm3 molecule-1 s-1 for the
OH radical rate constant measured in this work for tetra-
methylsilane is in good agreement with the value reported
by Atkinson (6) and, to a lesser degree, with the measurement
by Sommerlade et al. (7), with overall agreement within the
reported error limits (see Table 1). However, our present OH
radical rate constant of (7.2 ( 0.8) × 10-13 cm3 molecule-1
s-1 for trimethylsilanol is a factor of ∼5 lower than that of
(3.95 ( 0.65) × 10-12 cm3 molecule-1 s-1 reported by
Sommerlade et al. (7). The source of this discrepancy is not
clear, but Sommerlade et al. (7) employed a much smaller
reaction chamber than used in this work, and it is possible
that their study (7) was complicated by wall losses of (CH3)3-
SiOH but not of Si(CH3)4 and (CH3)3SiOSi(CH3)3. Trimethyl-
silanol, like tetramethylsilane, did not decay in our 5870 L
chamber within the time frames of the kinetic experiments,
and both trimethylsilanol and tetramethylsilane possess well-
isolated absorption features that enabled their time-
concentration profiles to be measured with very good
precision by FT-IR spectroscopy. As noted in Table 1, our
previous data for the analogous case of (CH3)3SiOSi(CH3)3
(6) and (CH3)3SiOSi(CH3)2OH (9) also show that their OH
radical rate constants do not differ to a significant degree,
and this is also true for C(CH3)4 versus (CH3)3COH. No data
for dimethylsilanediol have been reported in the literature,
but the OH radical rate constant of (8.1 ( 1.0) × 10-13 cm3
molecule-1 s-1 measured in this work (Table 1) is very similar
to those obtained for tetramethylsilane and trimethylsilanol.
From the above-measured OH radical rate constants and
using a 24-h average OH radical concentrations of 1.0 × 106
molecules cm-3 (21, 22), tropospheric lifetimes of 14 days,
17 days, and 15 days are calculated for tetramethylsilane,
trimethylsilanol, and dimethylsilanediol, respectively.
Product studies for both dimethylsilanediol and trimethyl-
silanol could not be carried out on pure samples due to their
inherent tendencies to form condensation products in the
bulk sample or in the gaseous reaction mixtures. Thus, the
investigation of the fates of their respective alkoxy radicals,
formed after the initial OH radical and/ or Cl atom abstraction
of an H atom from a methyl group, was complicated by
formation and subsequent reaction of these impurities. In
the case of dimethylsilanediol, the possible interferences
from a number of these products limited the analysis by
FT-IR spectroscopy, although the product spectra provided
FIGURE 8. API-MS spectra (A) before and after (B) 0.5 min, (C) 1.0
min, and (D) 2.0 min irradiation of a mixture of trimethylsilanol and
Cl2 in air.
FT-IR spectroscopy as a product from trimethylsilanol in
this experiment. However, during the kinetic experiments
described above which employed N2H4 + O3 as a source of
OH radicals, a weak but distinct Q-branch absorption at
1265.1 cm-1 was observed in the spectrum of the reaction
mixture, indicating the formation of (CH3)2Si(OH)2 from
(CH3)3SiOH via OH radical-initiated reaction in this non-
photolytic system. It should be noted that, due to H2O
production in the N2H4 + O3 reaction systems, the prevailing
H2O concentrations in these experiments were significantly
higher than in the photolysis of (CH3)3SiOH-Cl2-air mixtures,
which favors hydrolysis of an organosilicon ester to a silanol
(8). Addition of significant amounts of H2O to a Cl2 photolysis
mixture in the 5870 L chamber was not attempted, because
a previous, similar experiment resulted in the destruction of
the gold coatings of the multiple-reflection mirrors.
Photolysis of a Cl2-(CH3)3SiOH-air mixture was carried
out in the 7500 L Teflon chamber with analyses by API-MS
and API-MS/ MS. Figure 8A shows the API-MS spectrum
(80-140 u region) of the reactants before irradiation, and
Figure 8B-D shows the corresponding API-MS spectra after
0.5, 1.0, and 2.0 min of irradiation, respectively. Before
irradiation, ion peaks are observed at 91 ([90 + H]), 109
([90 + H + H2O]), and 127 u ([90 + H + 2H2O]) with the MW
90 species being (CH3)3SiOH and at 105 ([104 + H]) and 123
u ([104 + H + H2O]), with the MW 104 species being an
(CH3)3SiOCH3 impurity in the (CH3)3SiOH sample (presum-
ably from reaction of methanol with (CH3)3SiOH during the
preparation of the trimethylsilanol).
The API-MS spectra after irradiation show the gradual
disappearance of the 91 u ion peak (and its associated water
cluster peaks at 109 and 127 u) [and of the 105 u ion peak
of the (CH3)3SiOCH3 impurity] and the appearance of a
number of product ion peaks. The API-MS spectra after 0.5
and 1.0 min irradiation (Figure 8 (parts B and C) show the
9
VOL. 34, NO. 10, 2000 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 1 9 7 5

evidence for the formation of formate ester intermediates
from both dimethylsilanediol and trimethylsilanol. The most
significant results were provided by API-MS analysis which
indicated formation of methylsilanetriol from dimethyl-
silanediol and of both dimethylsilanediol and methylsilane-
triol from trimethylsilanol. Our present findings combined
with previous observations that (CH3)3SiOSi(CH3)2OH is
formed from (CH3)3SiOSi(CH3)3 via the hydrolysis of the
formate ester (CH3)3SiOSi(CH3)2OCHO (8) provide further
support to the following tropospheric degradation scheme
for tetramethylsilane and product silanols via OH radical
initiation:
Atkinson/ 97. We thank Dr. Grish Chandra for valuable advice
during this project and Professor Janet Arey for providing
expertise on the API-MS analyses.
Literature Cited
(1) Allen, R. B.; Kochs, P.; Chandra, G. Industrial Organosilicon
Materials, their Environmental Entry and Predicted Fate. In
Organosilicon Materials, The Handbook of Environmental
Chemistry; Chandra, G., Ed.; Springer-Verlag: Berlin, 1997; Vol.
3, Part H, p 1.
(2) Chandra, G.; Maxim, L. D.; Sawano, T. The Silicone Industry
and its Environmental Impact. In Organosilicon Materials, The
Handbook of Environmental Chemistry; Chandra, G., Ed.;
Springer-Verlag: Berlin, 1997; Vol. 3, Part H, p 295.
(3) Fendinger, N. J.; Lehmann, R. G.; Mihaich, E. M. Polydimeth-
ylsiloxane. In Organosilicon Materials, The Handbook of En-
vironmental Chemistry; Chandra, G., Ed.; Springer-Verlag:
Berlin, 1997; Vol. 3, Part H, p 181.
(4) Lehmann, R. G.; Miller, J. R. Environ. Toxicol. Chem. 1996, 15,
1455.
(5) Hobson, J.; Atkinson, R.; Carter, W. P. L. Volatile Methylsiloxanes.
In Organosilicon Materials, The Handbook of Environmental
Chemistry; Chandra, G., Ed.; Springer-Verlag: Berlin, 1997; Vol.
3, Part H, p 137.
Si(CH3)4 + OH f (CH3)3SiC˙ H2 + H2O
(3)
(CH3)3SiC˙ H2 + O2 f [(CH3)3SiCH2OO˙ ] f
(CH3)3SiOCH2O˙ (4)
(CH3)3SiOCH2O˙ + O2 f (CH3)3SiOCHO + HO2 (5)
(CH3)3SiOCHO + H2O f (CH3)3SiOH + HC(O)OH (6)
(6) Atkinson, R. Environ. Sci. Technol. 1991, 25, 863.
(7) Sommerlade, R.; Parlar, H.; Wrobel, D.; Kochs, P. Environ. Sci.
Technol. 1993, 27, 2435.
and similarly for (CH3)3SiOH:
(8) Atkinson, R.; Tuazon, E. C.; Kwok, E. S. C.; Arey, J.; Aschmann,
S. M.; Bridier, I. J. Chem. Soc., Faraday Trans. 1995, 91, 3033.
(9) Atkinson, R.; Aschmann, S. M.; Arey, J. 1993, Unpublished data
cited in refs 5 and 16.
(10) Tuazon, E. C.; Carter, W. P. L.; Atkinson, R.; Pitts, J. N., Jr. Int.
J. Chem. Kinet. 1983, 15, 619.
(11) Tuazon, E. C.; Carter, W. P. L.; Atkinson, R.; Winer, A. M.; Pitts,
J. N., Jr. Environ. Sci. Technol. 1984, 18, 49.
(12) Aschmann, S. M.; Chew, A. A.; Arey, J.; Atkinson, R. J. Phys.
Chem. A 1997, 101, 8042.
(13) Taylor, W. D.; Allston, T. D.; Moscato, M. J.; Fazekas, G. B.;
Kozlowski, R.; Takacs, G. A. Int. J. Chem. Kinet. 1980, 12, 231.
(14) Kantor, S. W. J. Am. Chem. Soc. 1953, 75, 2712.
(15) Atkinson, R.; Baulch, D. L.; Cox, R. A.; Hampson, R. F., Jr.; Kerr,
J. A.; Rossi, M. J.; Troe, J. J. Phys. Chem. Ref. Data, 1999, 28, 191.
(16) Atkinson, R. J. Phys. Chem. Ref. Data 1994, Monograph 2, p
1-216.
(17) Teton, S.; Mellouki, A.; Le Bras, G.; Sidebottom, H. Int. J. Chem.
Kinet. 1996, 28, 291.
(18) Colthup, N. B.; Daly, L. H.; Wiberley, S. E. Introduction to Infrared
and Raman Spectroscopy; Academic Press: New York, 1964; p
248.
O2
98 f
(CH3)3SiOH OH8 (CH3)2Si(OH)C˙ H2
O2
9
8 (CH3)2Si(OH)OCHO H O8
2
(CH3)2Si(OH)OCH2O˙
(CH3)2Si(OH)2
and similarly for (CH3)2Si(OH)2:
O2
98 f
(CH3)2Si(OH)2 OH8 CH3Si(OH)2C˙ H2
CH3Si(OH)2OCH2O· 98 CH3Si(OH)2OCHO H O8
O2
2
CH3Si(OH)3
The above reaction schemes can probably be extended
to include the transformation of CH3Si(OH)3 to Si(OH)4 [SiO2‚
xH2O], even though the latter was not detected in extended
irradiations of dimethylsilanediol-Cl2-air and trimethyl-
silanol-Cl2-air mixtures described above. The presence of
two hydroxyl groups in dimethylsilanediol and its observed
decay behavior in the chamber suggest that surfaces will
play a significant role in its atmospheric removal, and this
is also expected to be the case for methylsilanetriol. Since
dimethylsilanediol will be released into the atmosphere at
concentrations where autocondensation will be negligible,
incorporation into aerosols (and subsequent washout) could
be significant removal pathways for dimethylsilanediol in
the atmosphere in addition to reaction with OH radicals and
wet deposition from the gas phase.
(19) Pouchert, C. J. The Aldrich Library of FTIR Spectra, 1st ed.; Aldrich
Chemical Co., Inc.: 1989; Vol. 3, pp 605-633.
(20) Spivack, J. L.; Pohl, E. R.; Kochs, P. Organoalkoxysilanes,
Organosilanols and Organosiloxanols. In Organosilicon Materi-
als, The Handbook of Environmental Chemistry; Chandra, G.,
Ed.; Springer-Verlag: Berlin, 1997; Vol. 3, Part H, p 105.
(21) Prinn, R.; Weiss, R. F.; Miller, B. R.; Huang, J.; Alyea, F. N.;
Cunnold, D. M.; Fraser, P. J.; Hartley, D. E.; Simmonds, P. G.
Science 1995, 269, 187.
(22) Hein, R.; Crutzen, P. J.; Heimann, M. Global Biogeochem. Cycles
1997, 11, 43.
Received for review August 30, 1999. Revised manuscript
received January 28, 2000. Accepted February 22, 2000.
Acknowledgments
The authors gratefully acknowledge the financial support
from Dow Corning Corporation through Contract No. Dow/
ES9910053
9
1 9 7 6 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 34, NO. 10, 2000