Enhanced hydrolytic stability of siliceous surfaces modified with pendant dipodal silanes

Article abstract of DOI:10.1002/chem.201402757

Dipodal silanes possess two silicon atoms that can covalently bond to a surface. They offer a distinct advantage over conventional silanes commonly used for surface modification in terms of maintaining the integrity of surface coatings, adhesive primers, and composites in aqueous environments. New nonfunctional and functional dipodal silanes with structures containing pendant rather than bridged organofunctionality are introduced. The stability of surfaces in aqueous environments prepared from dipodal silanes with hydrophobic alkyl functionality is compared to the stability of similar surfaces prepared from the conventional silanes. In strongly acidic and brine environments, surfaces modified with dipodal silanes demonstrate markedly improved resistance to hydrolysis compared to surfaces prepared from conventional silanes. Pendant dipodal silanes exhibit greater stability than bridged dipodal silanes. The apparent equilibrium constant for the formation of silanol species by the hydrolysis of a disiloxane bond was determined as K_c=[SiOH]²/[Si-O-Si][H₂O]= 6±1×10^-5 and is helpful in understanding the enhanced hydrolytic stability of surfaces modified with dipodal silanes. Two feet are better than one! Nonfunctional and functional dipodal silanes with structures containing pendant rather than bridged organofunctionality were synthesized. Surfaces modified with pendant dipodal silanes were found to be more resistant to hydrolysis than the bridged structure with single-carbon separated (disilapropyl)silanes, demonstrating the greatest resistance to hydrolysis and best stability (see figure).

Full text of DOI:10.1002/chem.201402757

DOI: 10.1002/chem.201402757
Full Paper
&
Surface Chemistry
Enhanced Hydrolytic Stability of Siliceous Surfaces Modified with
Pendant Dipodal Silanes
Barry Arkles,* Youlin Pan, Gerald L. Larson, and Mani Singh[a]
Abstract: Dipodal silanes possess two silicon atoms that can
covalently bond to a surface. They offer a distinct advantage
over conventional silanes commonly used for surface modifi-
cation in terms of maintaining the integrity of surface coat-
ings, adhesive primers, and composites in aqueous environ-
ments. New nonfunctional and functional dipodal silanes
with structures containing “pendant” rather than “bridged”
organofunctionality are introduced. The stability of surfaces
in aqueous environments prepared from dipodal silanes
with hydrophobic alkyl functionality is compared to the sta-
bility of similar surfaces prepared from the conventional si-
lanes. In strongly acidic and brine environments, surfaces
modified with dipodal silanes demonstrate markedly im-
proved resistance to hydrolysis compared to surfaces pre-
pared from conventional silanes. Pendant dipodal silanes ex-
hibit greater stability than bridged dipodal silanes. The ap-
parent equilibrium constant for the formation of silanol spe-
cies by the hydrolysis of a disiloxane bond was determined
as Kc =[SiOH]2/[Si-O-Si][H2O]=6ꢀ1ꢀ10ꢁ5 and is helpful in
understanding the enhanced hydrolytic stability of surfaces
modified with dipodal silanes.
Introduction
the physical region associated with the transition from the
substrate to the matrix. The physical region that extends from
the unmodified bulk phases of both the substrate and matrix
is referred to as the interphase. The interphase region associat-
ed with silane modification is most properly viewed as a dy-
namic region in which a steep gradient in properties is mediat-
ed by the interaction of molecular water with oxane bonds.
More specifically, the dynamics are driven by the hydrolysis
and reformation of oxane bonds. Plueddeman[4] recognized
this dynamic when considering that in composites for which
there were substantial differences between the coefficient of
thermal expansion between inorganic substrates and polymers
physical properties were maintained during high-temperature
processing and room-temperature performance. For example,
mismatched coefficients of thermal expansion bond formation
in a thermoplastic composite, such as nylon 6/6 and glass
fiber, which processes at melt temperatures of 280–2958C
would have sufficient stress when cooled to room temperature
to undergo interfacial failure if the bond between the glass
fiber to the nylon was “fixed”. Separately, he observed that de-
spite the fact that although the oxane bonds between most
metals and silicon were known to be hydrolytically unstable,
the silane surface modification properties of minerals was
maintained. For example, despite the poor hydrolytic stability
of titanium–oxygen–silicon bonds, literally thousands of tons
of octylsilanes are effectively employed in modifying surface
properties for white titanium oxide pigments in paints and
coatings.
Silanes that alter the chemical and physical properties of surfa-
ces have been studied intensively for over sixty years.[1] Silanes
utilized to modify surfaces are generally divided into two cate-
gories. “Functional” silanes or coupling agents are utilized for
polymer composites,[2] biomolecule immobilization, and adhe-
sive technologies. “Nonfunctional” silanes are employed to
modify the surface energy or wettability of substrates and are
not intended to impart chemical reactivity to the substrate. To-
gether, both classes are often referred to as organofunctional
silanes. Both categories rely on oxane bond formation with
a hydroxyl functional substrate to effect the surface modifica-
tion. The role of water is important in this chemistry since the
vast majority of silane surface treatments are carried out by
a hydrolytic deposition process and, after deposition, water ad-
sorption at the interface is an important factor in determining
the durability of the surface modification. The role of water in
silane deposition and modification of surfaces is recognized in
the generally accepted mechanism first proposed for silane
deposition.[3]
The stability of properties associated with surface modifica-
tion depends on maintaining tight equilibrium conditions in
[a] Dr. B. Arkles, Dr. Y. Pan, Prof. Dr. G. L. Larson, Dr. M. Singh
Gelest Inc., 11 East Steel Rd.
Morrisville, PA 19067 (USA)
E-mail: executiveoffice@gelest.com
ꢀ 2014 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA.
This is an open access article under the terms of Creative Commons Attri-
bution NonCommercial-NoDerivs License, which permits use and distribu-
tion in any medium, provided the original work is properly cited, the use is
non-commercial and no modifications or adaptations are made.
The success of silanes in numerous coupling applications is
in large part due to the hydrolysis and reformation of oxane
bonds between silanes and substrates, but at the same time
hydrolysis without oxane bond reformation is a critical factor
Chem. Eur. J. 2014, 20, 9442 – 9450
9442 ꢁ 2014 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
in their failure. It has been observed that overpolymerization
of polymers onto substrates modified with silanes reduces the
mobility of hydrolyzed silanes.[5] Long-term interfacial failure
may be the eventual outcome of scission of the oxane bond
coupled with diffusion of the silane from the interphase and
a swelling or physical displacement of the resin from the origi-
nal interphase region. Both the diffusion of silanols and dis-
placement of the resin away from the inorganic substrate com-
promise system equilibrium and reduce the probability of ref-
ormation of oxane bonds.
soluble organosiloxanes have been limited to surfactants with
poly(ethyleneoxy) linkages and, while these materials degrade
within hours in aqueous systems,[11] the degradation is typically
associated with the oxidatively induced degradation of poly-
(ethylene oxide) segments rather than hydrolysis of the silox-
ane segments.
Several factors can potentially interfere with a straightfor-
ward determination of a silanol–siloxane equilibrium constant.
Ionization of silanol groups can take place at near neutral pH,
the same conditions of interest for studying the silanol–silox-
ane equilibrium. Other considerations that may preclude
simple determination of the silanol–siloxane equilibrium con-
stant is that most compounds of interest, such as silicic acid
and dimethylsilanediol, can form multiple siloxane bonds and
undergo polymerization that would influence the pKa of re-
maining silanol groups. Nevertheless, an understanding of the
apparent equilibrium between siloxane and silanol species
would be helpful in understanding long-term interphase
behavior.
This work explores two issues associated with the long-term
failure of silane-modified inorganic substrates: the dynamism
of siloxane bond hydrolysis and reformation and organofunc-
tional silanes with dipodal structures, which reduce the proba-
bility of diffusion from the inorganic substrate.
While the kinetics of siloxane bond formation in aqueous
systems have been studied extensively, there exists only sparse
data on the equilibrium constant for siloxane bond formation
and hydrolysis in aqueous media (Scheme 1).
The objective of this study was to improve the durability of
organofunctional silane surface modifications by increasing the
number of potentially available sites for an organosilane to
form oxane bonds either to a substrate or to form silsesquiox-
ane polymers. It was of further interest to determine whether
the proximity of the bonding sites within an organofunctional
silane had an effect on the durability of silane surface modifi-
cation under aqueous conditions. To provide a proper context
for the potential of enhancing the hydrolytic stability of orga-
nofunctional silane surface treatments, the apparent equilibri-
Scheme 1. Siloxane–silanol equilibrium.
This is presumably primarily because the condensation of si-
lanol species results in species with siloxane bonds that are
generally insoluble in water and form bulk insoluble phases
that do not readily participate in system equilibria. Silicic acid
and trimethylsilanol are examples of species that form bulk
phases, silica and hexamethyldisiloxane, respectively, that do
not readily participate in equilibrium studies in neutral aque-
ous media.[6–7] Further, the low concentrations of silicic acid
normally found in water, typically 50–100 ppm, leads to the im-
pression that the equilibrium favors the formation of the silox-
ane bond. However, even the determination of equilibrium sili-
cic acid concentration in water is subject to considerable
discussion.[8]
um constant for
determined.
a stable water soluble disiloxane was
Results and Discussion
As probes for the stability of the SiꢁOꢁSi bonds, an organo-
functional disiloxane and an organofunctional silanol were syn-
thesized by hydrosilylation of tetramethyldisiloxane and dime-
thylethoxysilane according to methods previously reported.[12]
Both compounds readily dissolved in water and no tendency
toward phase separation was observed. In the case of the tet-
rahydrofurfuryloxypropyldimethylethoxysilane, ethanol formed
by hydrolysis was removed by azeotropic distillation prior to
the equilibrium studies (Schemes 2 and 3)
On the other hand, phenomena, such as Ostwald ripening of
Stober process silica particles and formation of monolithic
structures by sol–gel processing, suggests that the equilibrium
is much more balanced.[9–10] Studies of the stability of water
Scheme 2. Hydrosilylation of 2-(allyloxymethyl)tetrahydrofurane with tetramethyldisiloxane.
Scheme 3. Hydrolysis of tetrahydrofurfuryloxypropyldimethylethoxysilane.
Chem. Eur. J. 2014, 20, 9442 – 9450 www.chemeurj.org
9443 ꢁ 2014 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Scheme 4. Hydrolysis of 1,3-bis(tetrahydrofurfuryloxypropyl)tetramethyldisiloxane.
In independent experiments, both the disiloxane and the si-
lanol were evaluated in neutral aqueous systems and, after
equilibration, the relative amounts of silanol and disiloxane
were determined (Scheme 4).
Various organofunctional silanes were prepared containing
two silicon atoms with alkoxy substituents. Compounds with
substitution in the alpha and omega position have sometimes
been referred to as “bis(silanes)” and have been the subject of
much investigation, particularly in the field of metal coatings
and primers.[13–15] We have chosen to call this class of silanes
“dipodal silanes” from the Greek in which “two feet” can be
fixed to a substrate. Dipodal silanes can be further differentiat-
ed in a general sense as to whether they are “bridged” dipodal
silanes or as “pendant” dipodal silanes. For the most part, di-
podal silanes utilized both commercially and reported in the
literature may be regarded as “bridged”, that is the silane sub-
stitution is at terminal ends of an organic moiety. Examples of
non-functional bridged dipodal silanes are 1,10-bis(trimethoxy-
silyl)decane and 1,8-bis(triethoxysilyl)octane. Examples of func-
tional bridged dipodal silanes are bis(trimethoxysilylpropyl)a-
mine and bis(triethoxysilylpropyl)tetrasulfide. Pendant dipodal
silanes, which possess substrate reactive sites separated by
only one or two carbon atoms, afford an organic functionality
that extends from the substrate in a manner similar to tradi-
tional or “conventional” organofunctional silanes. Representa-
tive examples of classes of organofunctional silanes are depict-
ed in structures 6–9.
ing from the interphase and, thus, exhibit extended time to
failure. Further, contact-angle studies provide a simple metric
that is readily related to the presence of a silane, whereas me-
chanical properties associated with composites and coatings
are less direct.
Apparent equilibrium constant of the siloxane bond in
water
Initial and equilibrium concentrations of 1,3-bis(tetrahydrofur-
furyloxypropyl)tetramethyldisiloxane and tetrahydrofurfuryloxy-
propyldimethylsilanol in solu-
tions of water and THF were de-
termined by GC analysis (with
adjustments for response fac-
tors) and summarized in the
Scheme 5. Disilylation of decene with trichlorosilane.
Tables 1 and 2. These reagents
were selected because they
Two synthetic methods were utilized to generate the pend-
ant dipodal silanes evaluated in this study. The pendant silanes
with two-carbon separation were synthesized by a disilylation
method as reported earlier (Scheme 5).[16]
The syntheses of the pendant silanes with a one-carbon sep-
aration were achieved in three steps, by starting with a redis-
tribution reaction, followed by hydrosilylation and
then esterification or esterification followed by hydro-
Scheme 6. Redistribution of bis(trichloro)methane with methyldichlorosilane.
silylation (Schemes 6, 7, and 8).[17]
.
Examples of both functional and nonfunctional si-
lanes were prepared. Evaluation of nonfunctional
silane surface treatments in aqueous systems was
considered to be a better probe than functional si-
lanes for hydrolytic stability. A functional silane anch-
ored by reaction with an organic component in
Scheme 7. Hydrosilylation of decene with 1,1,1,3,3-pentachloro-1-3-disilapropane.
a composite coating would be restrained from diffus- Scheme 8. Alkoxylation of 1,1,1,3,3-pentachloro-1-3-disilatridecane.
Chem. Eur. J. 2014, 20, 9442 – 9450
www.chemeurj.org
9444 ꢁ 2014 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
quioxane structures of the polymeric interphase or remain as
unbound or hydrogen bonded silanols. Considering this sim-
plest case, it can be argued that, in contrast to the organo-
functional silanol species that are subject to diffusion, the sila-
nol of the substrate is fixed and unable to leave the equilibri-
um system. Thus, a factor of ꢄ10ꢁ2 is a better metric than
ꢄ10ꢁ4 (rounding 6ꢀ10ꢁ5) of the apparent equilibrium con-
stant. Disregarding the formation of polymeric silsesquioxanes,
a projected improvement in stability for a surface modification
on a silane with two silicon atoms and thus two opportunities
to form siloxane bonds could potentially improve the durabili-
ty of the bond 100-fold. If all of the potential siloxane bonds
of each silicon atom are considered, then the probability of hy-
drolysis followed by diffusion from the substrate would be de-
creased 10ꢁ2 ꢀ10ꢁ2 ꢀ10ꢁ2, resulting in a 1ꢀ106-fold more
stable structure. This discussion is overly simplified, but ap-
pears to be consistent with improvements in the hydrolytic
stability observed with dipodal silanes. Significant factors that
clearly would have an impact on the hydrolytic stability of
a specific system are: 1) the solubility of the monomeric silanol
species, the tendency of these materials to form POSS struc-
tures, and the distance between silicon atoms in a dipodal
silane and 2) water transport to the interphase and adsorption
on the substrate.
Table 1. Initial and final of concentration of siloxane and silanol derivat-
ized from disiloxane.
Components
Disiloxane [mol]
Water [mol]
Initial Final
Silanol [mol]
Initial Final
Initial
Final
1
2
3
0.1000
0.5000
0.1000
0.0999
0.4983
0.9927
0.5000 0.4999 0.0000 0.0018
0.5000 0.4983 0.0000 0.0033
1.0000 0.9927 0.0000 0.0014
Table 2. Initial and final of concentration of siloxane and silanol derivat-
ized from silanol.
Components
Disiloxane [mol]
Water [mol]
Initial Final
Silanol [mol]
Initial Final
Initial
Final
1
2
3
0.0000
0.0000
0.0000
0.0773
0.1293
0.1294
0.5000 0.4991 0.1564 0.0163
0.6518 0.6509 0.2607 0.0019
0.5000 0.4991 0.2607 0.0016
were soluble in water at all concentrations. 1,3-Bis(tetrahydro-
furfuryloxypropyl)tetramethyldisiloxane was used as the start-
ing material in Table 1. (Tetrahydrofurfuryloxypropyl)dimethylsi-
lanol was used as a starting material in Table 2. The mixtures
were stirred in borosilicate glass reactors at room temperature,
and reached equilibrium in no longer than 30 days. Unlike pol-
yethyleneoxy-substituted siloxanes, the formation of low mo-
lecular weight species associated with oxidative scission of Cꢁ
OꢁC bonds was not observed.
Synthesis of dipodal silanes
Both conventional and bridged dipodal silanes were prepared
in high yield by well-established protocols for reacting trichlor-
osilane with terminal olefins in the presence of a Pt0 catalyst.
Pendant dipodal silanes with a two-carbon bridge were pre-
pared from terminal olefins and trichlorosilane in the presence
of a tetraalkylphosphonium chloride catalyst that exhibited
liquid behavior at room temperature (Scheme 9).
The apparent equilibrium constant for the hydrolysis of one
mole of disiloxane to two moles of silanol was calculated for
each data set and then averaged, providing the result:
2
Kc ¼ ½SiOHꢂ =½Si-O-Siꢂ½H2Oꢂ ¼ 6 ꢀ 1 ꢃ 10ꢁ5
The apparent equilibrium constant of 6ꢀ10ꢁ5 is consistent
with the dynamism suggested for the siloxane bond in surface
modification applications. While long-term hydrolytic stability
studies of polymeric siloxane and silica bulk phases are not rel-
evant to the equilibrium solution studies of this investigation,
they provided a perspective that suggested that the siloxane
bond is less susceptible to hydrolysis than reality. The preserva-
tion of the siloxane (SiꢁOꢁSi) bond in aqueous systems is sig-
nificantly more favored than the silicon ester (SiꢁOꢁC) bond
that has been estimated at 2ꢀ10ꢁ2.[18]
The reaction conditions were optimized by independently
varying co-feed ratios of reactants and catalyst into a continu-
ous high pressure reactor, which allowed for greater tempera-
ture control and continuous venting of gaseous byproducts as
depicted in Figure 1.
The data presented in Table 3 indicates that this procedure
proceeds in significantly higher yield than previously reported
for the analogous static reaction utilizing tetrabutylphosphoni-
um chloride as a catalyst,[19] producing >80% of the desired
product in preference to the partially reduced disilylated by-
product and single hydrosilylation byproduct. The structures,
The apparent equilibrium constant of the siloxane bond is
significant when considering the
design of new functional silane
structures. The simplest case for
most conventional silane surface
modifications is that only one of
the three intermediate hydroxyl
groups on each silicon atom
formed during hydrolytic deposi-
tion condenses with a silanol on
a siliceous substrate to form a si-
loxane bond. The remaining sila-
nols condense into the silse- Scheme 9. Disilylation of decene with trichlorosilane.
Chem. Eur. J. 2014, 20, 9442 – 9450
www.chemeurj.org
9445 ꢁ 2014 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
hysteresis exceeded 308, the
data was judged to not be
meaningful and testing was
ended.
Deionized water immersion is
relatively mild in comparison to
the other aqueous media evalu-
ated. The ꢆSi-O-Siꢆ linkage is
very susceptible to hydrolysis at
pH<3 and >8, with basic condi-
tions significantly more deleteri-
ous than acidic conditions. The
static immersion test in 6m HCl
for silane-modified surfaces is an
Figure 1. Schematic of apparatus for continuous high-pressure disilylation. Exemplary conditions for 1,2-bis(tri-
chlorosilyl)decane: Catalyst (n-tetradecyl)tri-n-butylphoshonium chloride; mole ratio decene/trichlorosilane/
R4P+Clꢁ 1:2:0.1; reaction temperature: 1808C; residence time: 2 h; pressure: 23 atm.; yield: 84%; byproducts: 1-
(trichlorosilyl)-2-(dichlorosilyl)decane, trichlorosilyldecane, SiCl4, H2SiCl2.
extreme condition and acts as an accelerated acidic
durability test. 1,2-Bis(trimethoxysilyl)decane, and
1,1,1,3,3-pentamethoxy-1,3-disilatridecane form the
most durable coatings as the static water contact
angle after 60 days in HCl is 102ꢀ2 and 94ꢀ38, re-
spectively. The conventional silane, n-decyltriethoxysi-
lane, and the pendant dipodal, 1,2-bis(trimethoxysi-
lyl)decane contact angles follow each other very
closely up to day 28, after which the conventional
silane coating starts to significantly degrade. The
contact angles drop from 102ꢀ1 on day zero to
88ꢀ18 on day 60 for n-decyltriethoxysilane. Similar
behavior is seen in the static immersion studies in
deionized water for all the silane treatments. The rate
of deterioration of the silane coatings in 3.5%
sodium chloride was greater than anticipated. The bridged di-
podal silane, 1,10-bis(trimethoxysilyl)decane, which appeared
to maintain a relatively stable contact angle over time in the
first part of the experiment, began to exhibit greater hysteresis
Table 3. Nonfunctional pendant silanes (with 2 carbon separation).
Organic
Isolated
B.p.
Density
yield [%] [8C/mmHg] [g/cc, 208C]
87
95
120–28/1
1.250
0.984
130–28/0.4
71
186–98/0.2
1.103
yields, and physical properties of dipodal silanes prepared are
summarized in Tables 3–5.
Behavior of surfaces treated with dipodal silanes
Silane treatment contact angle measurements (ses-
Table 4. Nonfunctional pendant silanes (with one carbon separation).
Organic Isolated B.p.
sile, advancing, and receding) for borosilicate glass
slides treated with representative silanes (6–9) are
listed in Table 6. All the silanes, except 1,10-bis(trime-
thoxysilyl)decane, show sessile water contact angles
(CA) ꢅ908. 1,10-Bis(trimethoxysilyl)decane has two
trialkoxysilyl groups bridged by a long alkyl chain
and, therefore, does not possess terminal methyl
groups that contribute towards low surface energy
(hydrophobic) behavior. All the silane coatings show
a hysteresis (advancing CA–receding CA)>108, which
suggests the possibility of surface defects or inhomo-
geneities, but examination by SEM revealed smooth
homogeneous coatings.
Density
[g/cc,
208C]
yield
[%]
[8C/
mmHg]
59
126/0.1
123/0.1
1.178
0.955
46
91
192–4/0.2 1.059
170–2/0.4 1.262
170–2/0.1 0.994
Static immersion durability tests for silanes (6–9)
were conducted in deionized water, 6m HCl, 3.5%
aqueous sodium chloride and 1m NH4OH at room
temperature. The integrity of the coating is directly
related to changes in the water contact angle and an
increase in contact angle hysteresis measured on the
substrates. Figures 2A–D show the progression of the
sessile water contact angle with time for the sub-
strates coated. In cases for which the contact angle
70
56
Chem. Eur. J. 2014, 20, 9442 – 9450
www.chemeurj.org
9446 ꢁ 2014 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
single-carbon separated (disilapropyl)silanes demon-
strating the greatest resistance to hydrolysis and best
stability.
Table 5. Functional pendant silanes (with one carbon separation).
Organic
Isolated
B.p.
Density
yield [%] [8C/mmHg] [g/cc, 208C]
Experimental Section
15
60
168–78/0.5 1.003
General considerations
All synthetic reactions were run under an atmosphere of
purified nitrogen. Trichlorosilane, dichloromethylsilane
and the various olefins were of a minimum 98% purity,
and were used without further purification. 1H NMR
spectra were recorded on a JEOL 400 MHz instrument.
The INEPT sequence was utilized in collecting 29Si NMR
spectroscopic data. n-Decyltrimethoxysilane was a Gelest
commercial product.
89–91/0.4
110/0.5
1.380
1.020
60
89
Preparation of 1,1,1,3,3-pentachloro-1,3-disilapro-
pane (14)
130–50/0.5 0.990
A 5-gallon autoclave equipped with a stirrer, pot ther-
mometer and addition ports was charged with
1,1,1,3,3,3-hexachloro-1,3-disilapropane (9901.5 g; 35
mol), dichloromethylsilane (4428.7 g; 38.50 mol), and
tri(n-hexyl)(n-tetradecyl)phosphonium chloride (258.0 g, 0.88 mol)
The reaction mixture was heated to between 80 and 908 for 10 h.
The reaction temperature was cooled to room temperature and
the crude reaction product discharged into a 12-liter flask. Vacuum
was applied to remove volatiles and the residue was distilled at
6 mm Hg and a pot temperature of 608C. This served to remove
various impurities and produce a mixture of 1,1,1,3,3-pentachloro-
1,3-disilapropane and 1,1,1,3,3,3-hexachloro-1,3-disilapropane in
a ratio of 46:54. This material was used for the hydrosilylation ex-
periments to prepare the 1,3-disilaalkane derivatives.
Table 6. Initial water contact angles for the various silanes.
Silane
Water contact angle [8]
Sessile
Advancing
Receding
6
8
9
7
94ꢀ1
102ꢀ1
104ꢀ2
78ꢀ2
104ꢀ1
106ꢀ2
110ꢀ1
84ꢀ1
91ꢀ2
95ꢀ1
98ꢀ2
73ꢀ2
and a mottled appearance, with hysteresis exceeding 308 after
120 days. Of all the silanes, 1,1,1,3,3-pentamethoxy-1,3-disilatri-
decane formed the most durable coating in all environments
tested and, notably, was the only silane to form coatings with
even nominal resistance to 1m NH4OH. Resistance to hydrolysis
of organofunctional silanes appears to be optimized when
there is a relatively small number of carbon atoms between
the silicon centers.
Preparation of 1,1,1,3,3-pentaethoxy-1,3-disilapropane
A 5-liter flask equipped with condenser, dropping funnel, and mag-
netic stirrer was charged with of 1,1,1,3,3-pentachloro-1,3-disilapro-
pane (600 g; 1.11 mol active) prepared above. Triethylorthoformate
(8200 g; 55.5 mol) was added dropwise with the reaction tempera-
ture at 908. A gas (ethyl chloride) was evolved. After the evolution
of gas had ceased the temperature was raised to 1208C and tri-
ethylorthoacetate (820 g; 5.55 mol) was added dropwise to finish
the esterification. Chloride analysis indicated that the reaction was
complete. After completion of the reaction, vacuum was applied to
remove the more volatile components. Distillation gave 270 g
(42.9%) of the title product as a 44:56 mixture of 1,1,1,3,3-penta-
ethoxy-1,3-disilapropane and 1,1,1,3,3,3-hexaethoxy-1,3-disilapro-
pane. B.p. 88–1038C/1.5 mmHg. This mixture was used for the hy-
drosilylation experiments.
Conclusion
Dipodal silanes possess two silicon atoms that can covalently
bond to a surface. New nonfunctional and functional dipodal
silanes with structures containing “pendant,” rather than
“bridged,” organofunctionality were synthesized. Both pendant
and bridged silanes offer a distinctive advantage over conven-
tional silanes in terms of maintaining the integrity of surface
coatings, adhesive primers and composites in aqueous and ag-
gressive environments. The improved durability of such dipo-
dal silanes is associated with an increased cross-link density of
the interphase and the inherent resistance to hydrolysis, as
they can potentially form five or six, rather than three, oxane
bonds. Pendant dipodal silane structures are more resistant to
hydrolysis than the bridged dipodal silane structure with
Preparation of 1,2-bis(trichlorosilyl)decane (11)
A 5-gallon autoclave equipped with a stirrer, pot thermometer, and
addition ports was charged with a premix of 1-decene (1616.5 g;
11.5 mol), trichlorosilane (4683.2 g; 34.6 mol), and a 50% toluene
solution of (tetradecyl)tri-n-butylphosphonium chloride (679.7 g;
2.3 mol). The reaction mixture was slowly heated to 2208C and
held at that temperature for 18 h. The reaction temperature was
cooled to room temperature and the crude reaction product dis-
charged into a 12-liter flask. Distillation of the product gave
3684.7 g (87%) of the title dipodal silane. B.p. 120–28C/0.2 mmHg;
Chem. Eur. J. 2014, 20, 9442 – 9450
www.chemeurj.org
9447 ꢁ 2014 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
1
density, 208C: 1.249 g/cc; H NMR (400 MHz): d=1.88–1.70 (m, 3H),
1.66–1.56 (m, 1H), 1.56–1.36 (m, 2H), 1.36–1.18 (m, 11H), 0.88 ppm
(t, 3H).
Preparation of 1,2-bis(trimethoxysilyl)decane (8)—represen-
tative esterification procedure
A 22-liter flask equipped with condenser, dropping funnel, and
overhead stirrer was charged with (6,463 g, 17.25 mol) of 1,2-bis-
(trichlorosilyl)decane and heated to about 1008C. Trimethylortho-
formate (7,322 g; 69.0 mol) was charged to the addition funnel
and added dropwise at such a rate so as to control the evolution
of the chloromethane and methyl formate. The progress of the re-
action was followed by measurement of the pH with a pH of 6–7
indicating full reaction of the chlorosilane. To finish the reaction,
trimethylorthoacetate (5,181 g; 43.13 mol) was added dropwise
and the reaction temperature increased to 1208C. After the reac-
tion was complete the volatiles and excess ortho esters were re-
moved at atmospheric pressure up to a pot temperature of 1608C.
Reduced-pressure distillation gave 5610.0 g (85%) of the title prod-
uct. B.p. 130–1328/0.4 mmHg; density, 208C: 0.984 g/cc; 1H NMR
(400 MHz): d=3.54 (s, 9H), 3.51 (s, 9H), 1.50–1.30 (m, 4H), 1.25
(brs, 10H), 1.05–0.98 (m, 1H), 0.90–0.81 (m, 4H), 0.46 ppm (dd,
1H); 29Si{1H} NMR ( 400 MHz, CDCl3): d =ꢁ44.41, ꢁ41.95 ppm.
Preparation of 1,2-bis(trichlorosilyl)octadecane (19)
A procedure similar to that for the preparation of 1,2-bis(trime-
thoxysilyl)decane employing 1-octadecene (2,200 g; 9.0 mol) and
trichlorosilane (3,600 g; 27 mol) gave 3257.6 g (71%) of 1,2-bis(tri-
chlorosilyl)octadecane as colorless liquid. B.p. 1908C/0.1 mmHg;
1
density, 208C: 0.923 g/cc; H NMR (400 MHz): d=1.88–1.72 (m, 4H),
1.60 (dd, 1H), 1.54–1.36 (m, 2H), 1.36–1.20 (m, 25H), 0.88 ppm (t,
3H).
1,1,1,3,3-Pentachloro-1,3-disilatridecane (16)
Under a nitrogen atmosphere, a 1-liter, 3-necked flask equipped
with a heating mantle, magnetic stirrer, pot thermometer, addition
funnel, and condenser was charged with 276.8 g of a mixture of
approximately 46% (dichlorosilylmethyl)trichlorosilane and 54%
bis(trichlorosilyl)methane [molar ratio 1:1]. The mixture was heated
to 908C and 10 g of 1-decene were added to the mixture by an ad-
dition funnel, followed by 0.5 mL of 5% hexachloroplatinic acid in
tetrahydrofuran. An immediate exotherm was observed and the re-
action mixture changed from clear to dark brown. An additional
173 g of 1-decene was added at an appropriate rate to maintain
the temperature between 90 and 1108C. Upon completion of the
addition, an additional 0.25 mL of 5% hexachloroplatinic acid solu-
tion was added and the mixture was stirred at 1008C for 45 min.
GC analysis of the mixture indicated the presence of some unreact-
ed pentachlorodisilapropane. An additional 10 g of 1-decene were
added and the reaction mixture was stirred an additional 45 min at
1008. Distillation provided 242 g (59%) of the title compound. B.p.
1
1268/0.1 mmHg; density, 208C: 1.178 g/cc; H NMR (400 MHz): d=
1.58–1.48 (m, 4H), 1.46–1.34 (m, 2H), 1.34–1.20 (m, 14H), 0.88 ppm
(t, 3H).
Preparation of 1,1,1,3,3-pentamethoxy-1,3-disilatridecane (9)
Figure 2. Sessile water contact-angle measurements for deionized water (A),
The product from above was reacted according to the representa-
tive esterification procedure to give an 82% yield of 9. B.p. 123–
48C/0.1 mmHg, density, 208C: 0.955 g/cc; 1H NMR (400 MHz): d=
3.55 (s, 9H), 3.50 (s, 6H), 1.40–1.15 (m, 16H), 0.85 (t, 3H), 0.70–0.60
(m, 2H), ꢁ0.05 ppm (s, 2H); 29Si{1H}( 400 MHz, CDCl3): d ꢁ41.85,
ꢁ4.22 ppm.
&
6m HCl (B), 3.5% aqueous NaCl (C), and 1m NH4OH (D) ( : n-decyltriethoxysi-
^
*
lane (6); : 1,2-bis(trimethoxysilyl)decane (7); : 1,1,1,3,3-pentamethoxy-1,3-
disilatridecane (8); : 1,10-bis(trimethoxysilyl)decane (7)).
~
Chem. Eur. J. 2014, 20, 9442 – 9450
www.chemeurj.org
9448 ꢁ 2014 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
tion mixture was subjected to distillation through a short Vigreaux
column with three successive distillations giving 100 g (64.5%) of
the title product in 96% purity. B.p. 125–308C/0.1 mmHg; H NMR
(400 MHz): d=3.82–3.71 (m, 10H), 3.65 (dd, 1H), 3.48–3.31 (m, 3H),
3.12–3.08 (m, 1H), 2.75 (t, 1H), 3.59–3.54 (m, 1H), 1.69–1.61 (m,
2H), 1.20–1.11 (m, 15H), 0.68–0.61 (m, 2H), ꢁ0.07 ppm (s, 2H).
Preparation of 1,1,1,3,3-pentachloro-1,3-disilaheneicosane
(20)
1
A 2-liter flask equipped with heating mantel, magnetic stirrer, pot
thermometer, addition funnel, and
a dry-ice condenser was
charged with the 1,1,1,3,3-pentachloro-1,3-disilapropane/
1,1,1,3,3,3-hexachloro-1,-disilapropane mixture (276.8 g; 0.51 mol
active reagent). The reaction mixture was warmed to 908C and 5%
chloroplatinic acid in THF (0.5 g) was added followed by the addi-
tion of 10 g of 1-octadecene. An immediate exotherm was ob-
served indicating initiation of the hydrosilylation. The remainder of
the 187 g (0.72 mol) of 1-octadecene was added at a rate to main-
tain the reaction temperature between 90 and 1108C. After the ad-
dition was complete, an additional 0.25 mL of the platinum catalyst
solution was added and the reaction mixture was heated at 1008C
for an additional 45 min. Distillation provided 229 g (91%) of the
title product. B.p. 192–48C/0.2 mmHg; density, 208C: 1.059 g/cc;
1H NMR (400 MHz): d=1.57–1.49 (m, 4H), 1.42–1.35 (m, 2H), 1.34–
1.20 (m, 30H), 0.88 ppm (t, 3H).
Preparation of 1,1,1,3,3,6-hexachloro-5-methyl-1,3-disilahex-
ane (24)
A 2-liter flask equipped with magnetic stirrer, condenser, pot ther-
mometer, and addition funnel was charged with the 1,1,1,3,3-pen-
tachloro-1,3-disilapropane/1,1,1,3,3,3-hexachloro-1,3-disilapropane
mixture prepared above (1,455 g; 2.05 mol active). The addition
funnel was charged with methallyl chloride (186 g; 2.05 mol). The
reaction mixture was warmed to 1108C and to initiate the reaction
20 g of the olefin was added followed by the addition of 0.5 mL of
a 5% solution of chloroplatinic acid in THF. The remainder of the
olefin was added a rate such as to maintain a reaction temperature
between 110 and 1208C. After completion of the olefin addition,
the reaction was heated for an additional 3 h at 1308C. Distillation
of the product directly from the reaction vessel provided 412 g
(60%) of the title dipodal silane. B.p. 89–918C/0.4 mmHg; density,
208C: 1.380 g/cc.
Preparation of [2-methoxy(triethyleneoxy)propyl]-1,1,1,3,3-
pentachloro-1,3-disilapropane (21)
Under an atmosphere of nitrogen, a standard-equipped 500 mL
flask was charged with allyloxytriethyleneoxymethyl ether (51.1 g;
0.25 mol). A 46:54 mixture of 1,1,1,3,3-pentachloro-1,3-disilapro-
pane and 1,1,1,3,3,3-hexachloro-1,3-disilapropane (141.8 g; 0.26
mol of reactive silane) was placed in an addition funnel and ca.
25 g of the mixture containing 14 was added to the reaction mix-
ture. The reaction mixture was heated to 858C and a 5% THF solu-
tion of chloroplatinic acid (0.25 mL) was added. An immediate exo-
therm was observed with the solution changing from clear and
colorless to dark brown. The balance of the chlorosilane mixture
was added at a rate so as to maintain a reaction temperature of 85
to 1058C. Upon completion of the silane addition, an additional
0.25 mL of the catalyst solution was added and the reaction mix-
ture heated at 958 for 45 min. Distillation provided 80 g (70%) of
the title organosilane. B.p. 170–28C/0.5 mmHg; density, 208C:
Preparation of 1,1,1,3,3-pentaethoxy-6-chloro-5-methyl-1,3-
disilahexane (25)
The product from above was reacted according to the representa-
tive esterification procedure to give a 60% yield of the title orga-
nosilane. B.p. 115–1178C/0.5 mmHg; density, 208C: 1.020 g/cc;
1H NMR (400 MHz): d=3.85–3.72 (m, 10H), 3.56–3.50 (m, 1H), 3.39–
3.33 (m, 1H), 2.15–2.05 (m, 1H), 1.23–1.15 (m, 15H), 1.07 (d, 3H),
0.9–0.83 (m, 1H), 0.68–0.61 (m, 1H), ꢁ0.03 ppm (s, 2H).
Preparation of 3-[2-(aminoethylamino-5-methyl)]-1,1,1,3,3-
pentaethoxydisilahexane (26)
1
1.262 g/cc; H NMR (400 MHz): d=3.72- 3.56 (m, 8H), 3.65–3.48 (m,
A 500 mL flask suitably equipped was charged with 1,1,1,3,3-pen-
taethoxy-6-hexachloro-5-methyl-1,3-disilahexane (96.8 g; 0.25 mol)
and ethylene diamine (75.1 g; 1.25 mol). The reaction mixture was
heated to 1108C for 16 h and 25 g (0.54 mol) of ethanol added
and the bottom layer separated and wiped-film distilled to yield
81 g (89%) of the title dipodal silane. B.p. (est) 130–1408C/
4H), 3.17 (t, 2H), 3.41 (s, 3H), 1.71–1.59 (m, 4H), 0.73–0.68 ppm (m,
2H).
Preparation of [2-methoxy(triethyleneoxy)propyl]-1,1,1,3,3-
pentaethoxy-1,3-disilapropane (22)
1
0.5 mmHg; density, 208C: 0.990 g/cc; H NMR (400 MHz): d=3.83–
In a one-pot, two-step process based on the above, allyloxytriethy-
leneoxymethyl ether (204 g; 1 mol) was hydrosilylated and the
crude pentachloro derivative converted directly to the ethyl ester
to give after distillation 310 g (62%) of the title organosilane. B.p.
176–88C/0.1 mmHg: density, 208C: 0.994 g/cc; nD20 =1.4359;
1H NMR (400 MHz): d=3.80–3.65 (m, 10H), 3.62- 3.56 (m, 8H),
3.55–3.48 (m, 4H), 3.37 (t, 2H), 3.31 (s, 3H), 1.68–1.59 (m, 2H),
1.19–1.10 (m, 15H), 0.65–0.58 (m, 2H), ꢁ0.09 ppm (s, 2H).
3.69 (m, 10H), 2.76–2.71 (m, 2H), 2.68–2.59 (m, 2H), 2.51–2.45 (m,
1H), 2.40–2.32 (m, 1H), 2.88–2.78 (m, 1H), 1.34 (brs, 3H), 1.22–1.11
(m, 15H), 0.95 (d, 3H), 0.79–0.71 (m, 1H), 0.56–0.48 (m, 1H),
ꢁ0.06 ppm (s, 2H); 29Si{1H} NMR (400 MHz, CDCl3): d=ꢁ45.62,
ꢁ8.50 ppm.
Methodology for deposition of silanes
Borosilicate glass slides (Schott North America) were acid etched
before silane treatment by dipping in 4 wt.% HCl for 45 min
before washing with 1) deionized water (5 mL), 2) ethanol (5 mL),
and 3) acetone (5 mL). The slides were dried under N2 and treated
with the various silanes immediately.
Preparation of 1,1,1,3,3-pentaethoxy-1,3-disilahexylglycidyl
ether (23)
A 1-liter flask equipped with condenser, addition funnel, magnetic
stirrer, and pot thermometer was charged with the 44:56 mixture
of 1,1,1,3,3-pentaethoxy-1,3-disilapropane and 1,1,1,3,3,3-hexa-
ethoxy-1,3-disilapropane (283.75 g; 0.42 mol of active) prepared
above and phenothiazine (0.04 g). The reaction mixture was
heated to 908C and allylglycidyl ether (43.3 g; 0.38 mol) was
added keeping the temperature between 85 and 1158C. The reac-
A 2 wt.% solution of n-decyltriethoxysilane (6) was prepared in cy-
clohexane and a small amount of deionized water (less than
0.5 wt.%; containing enough glacial acetic acid so that the pH 3–5)
was added with stirring. The stirred silane solution was allowed to
hydrolyze for 5 h prior to dipping the cleaned glass slides into the
stirred solution. After 20 h, the glass slides were removed and
Chem. Eur. J. 2014, 20, 9442 – 9450
www.chemeurj.org
9449 ꢁ 2014 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[1] J. Bjorksten, L. L. Yaeger, Modern Plastics 1952, 29, 124.
[2] J. Theberge, B. Arkles, P. Cloud, Machine Design, 1971, 43, 58; 26th Soc.,
Plastics Ind. Composites Inst. Proc., 1971, 3A.
rinsed with cyclohexane, and dried under N2 before being put into
an oven at 1108C for 30 min. The glass slides were cooled to room
temperature in a desiccator and the water contact angles were
measured. The contact angle measurements were carried out on
a Ramꢂ–Hart 100 goniometer. Multiple sessile drop contact angle
(advancing and receding) measurements were carried out for each
silane and the average numbers (minimum of 6 measurements)
are reported. Glass slides were also prepared for silanes 1,2-bis(tri-
methoxysilyl)decane (8), 1,1,1,3,3-pentamethoxy-1,3-disilatridecane
(9), and 1,10-bis(trimethoxysilyl)decane (7), and the respective
water contact angles were measured following the same
procedure.
[3] B. Arkles, Chemtech 1977, 7, 766.
[4] E. P. Plueddemann, “Silane Coupling Agents”, 1982, 111, Plenum.
[5] F. Blum, W. Meesiri, H. Kang, J. Gambogi, J. Adhes. Sci. Technol. 1991, 5,
479.
[6] W. T. Grubb, J. Am. Chem. Soc. 1954, 76, 3408.
[7] J. Sefcik, S. Rankin, S. Kirchner, A. McCormick, J. Non-Cryst. Solids 1999,
258, 187.
[8] R. K. Iler, “The Chemistry of Silica”, 1979, 30, Wiley.
[9] J. Schlomach, M. Kind, J. Colloid Interface Sci. 2004, 277, 316.
[10] J. Baldyga, M. Jasinska, K. Jodko, P. Petelski, Chem. Eng. Sci. 2012, 77,
207.
[11] H. Hillborg, U. Gedde, Polymer 1998, 39, 1991.
[12] B. Arkles, Y. Pan, J. Goff, US Pat. Appl. 20120226002, 2012.
[13] P. Pape, E. Plueddemann, in “ Silanes and Other Coupling Agents” ed. K.
Mittal, 1992 VSP, p105.
[14] J. Song, W. J. van Ooij, J. Adhes. Sci. Technol. 2003, 17, 2191.
[15] A. Seth, W. J. van Ooij, P. Puomi, Z. Yin, A. Ashigrade, S. Bafna, C. Shi-
vane, Prog. Org. Coat. 2007, 58, 136.
[16] C. Y. K. Lung, J. P. Matinlinna, Dent. Mater. 2012, 28, 467.
[17] B. Arkles, US Pat. 7,235,683, 2007.
[18] B. Arkles, Y. Pan, G. Larson, Pendant Dipodal Silanes with a Disilapropyl
Terminus, US Pat Appl., 2014.
Methodology for contact angle durability studies
The durability of the silane coating was assessed by static immer-
sion tests at room temperature in deionized water, 6m HCl, 1m
NH4OH, and 3.5 wt.% aqueous NaCl. Each slide was removed from
the solution, rinsed with water, and dried under nitrogen before
being put into an oven at 1108C for 30 min. On cooling to room
temperature, the water contact angles were measured. After each
contact angle measurement, the treated slides were re-immersed
in their parent baths (deionized water, 6m HCl, 1m NH4OH or
3.5 wt.% aqueous NaCl). The process was repeated for all the sam-
ples at specified intervals for ca. 140 days or until the coatings
were completely delaminated.
[19] B. Arkles, J. Steinmetz, J. Zazyczny, P. Mehta, in “Silanes and Other Cou-
pling Agents”, ed. K. Mittal, 1992, VSP, 91.
[20] I. N. Jung, J. Am. Chem. Soc. 2001, 123, 5584; US Pat. 6,251,057, 2001.
Keywords: dipodal silanes · hydrolytic stability · silicon ·
siloxane–silanol equilibria · surface chemistry
Received: March 24, 2014
Chem. Eur. J. 2014, 20, 9442 – 9450
www.chemeurj.org
9450 ꢁ 2014 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim