Tunable Interfacial Chemistry of Spherical Silica by Single- and Dual-Silane Modification
with γ-Glycidoxypropyltrimethoxysilane and Octadecyltriethoxysilane
(Won Seon Seo)
1
(Sangwoo Kim)
2
(Beom Zoo Lee)
1
(Ji Hye Jeon)
3
(Minzhen Li)
3
(Min Young Kim)
3
(Jeong Yun Hwang)
3
(Young-Min Byoun)
4
(Gi Hyeon Han)
3,*
(Changhyun Jin)
3,*
-
(CHEMLAND Co., Ltd., Anyang-si 14087, Republic of Korea)
-
(Materials · Supply Chain R&D Department, Korea Institute of Industrial Technology,
Incheon 21999, Republic of Korea)
-
(Department of Materials Science and Engineering, Yonsei University, Seoul 03722, Republic
of Korea)
-
(ESTECH Co., Ltd., Siheung-si, 15086, Republic of Korea)
Copyright © The Korean Institute of Metals and Materials
Keywords
Silica, γ-Glycidoxypropyltrimethoxysilane, Octadecyltriethoxysilane, Interfaces, Surface modification
1. INTRODUCTION
Silica has been extensively utilized as an inorganic filler in polymer composites
owing to its high hardness, excellent dielectric insulation, low coefficient of thermal
expansion, and outstanding thermal stability[1-
5]. In particular, amorphous silica with an irregular powder morphology is primarily
used as a reinforcing material due to its large surface area and high adsorption capacity,
whereas spherical silica is predominantly employed in precision semiconductor packaging
because of its high fluidity and packing density[6-
9]. However, the surface of pristine silica is rich in hydroxyl groups that form Si–OH
silanol functionalities, making the particles highly susceptible to moisture adsorption
and, consequently, prone to self-aggregation. Furthermore, when the particle size
of spherical silica is reduced to the sub-micron scale, achieving uniform dispersion
becomes challenging due to the increased surface reactivity resulting from the elevated
specific surface area. Owing to these inherent characteristics, the interfacial compatibility
of silica with polymer matrices in composite materials is significantly compromised,
and dispersibility degrades due to particle agglomeration. These factors often limit
the full realization of the performance benefits offered by silica-filled inorganic–organic
composite materials.
To overcome these limitations, surface modification using organosilane coupling agents
is considered a highly effective method for tailoring the surface properties of spherical
silica for specific applications[10-
14]. In general, organosilanes chemically bond to silanol groups on the surface of spherical
silica through hydrolysis–condensation reactions while simultaneously introducing
organic functional groups—such as vinyl, epoxy, styryl, methacryloxy, acryloxy, amino,
and ureide groups—which enhance interfacial interactions with the surrounding polymer
matrix. Among various silanes, γ-glycidoxypropyltrimethoxysilane (GPTMS) possesses
a reactive epoxy group and three methoxy groups; therefore, it can form robust chemical
bonds at the interface between spherical silica and organic polymers[15-
17]. On the other hand, octadecyltriethoxysilane (OTES) provides long hydrophobic C18
alkyl chains upon bonding to the surface of spherical silica[18-
20]. This surface shielding can significantly reduce the surface energy and inhibit moisture
absorption, thereby playing a critical role in preventing particle aggregation and
ensuring long-term dispersibility. In other words, the enhanced resistance to particle
aggregation provided by OTES can significantly improve processability and moisture
resistance. However, owing to the absence of reactive functional groups capable of
interacting with polymer chains, OTES-modified silica may exhibit limited interfacial
adhesion to the organic phase in inorganic–organic composites.
From an interfacial engineering perspective, these two silane coupling agents demonstrate
complementary functionalities, offering the potential to synergistically integrate
their respective advantages while compensating for their inherent shortcomings. More
specifically, GPTMS exhibits strengths in chemical reactivity and adhesion, whereas
OTES excels in hydrophobic performance and anti-aggregation dispersibility. Although
combinations of two silanes tailored to a specific purpose are feasible, most existing
studies have focused exclusively on single-silane processing systems. For this reason,
the effects of combining two silane coupling agents—including changes in surface polarity,
structure-dependent interactions, and interfacial reactivity—have received relatively
limited attention in previous literature[21-
23]. Nevertheless, dual-functional surfaces have the potential to exhibit synergistic
effects that can overcome the limitations of single-silane coupling agents.
In this study, we systematically investigate the surface modification of sub-micrometer
monodisperse spherical silica particles using GPTMS and OTES, which possess distinct
functional groups, focusing on both single-silane systems and mixed dual-functional
systems with varying mixing ratios (3:1, 1:1, and 1:3). This investigation includes
the chemical interactions occurring between the reactive epoxy functional groups of
GPTMS and the long alkyl chains of OTES within the co-grafted dual-functional layer.
We hypothesized that as the relative content of GPTMS and OTES changes, the intermediate
compositions would exhibit intermediate surface and physical properties compared to
those of pure GPTMS and pure OTES. To validate this, we investigated the composition–property
relationships of spherical silica surfaces using a dual-silane coupling agent rather
than a pure single-silane coupling agent. Our findings provide fundamental theoretical
insights into the applicability of these dual-functionalized particles as fillers
for next-generation epoxy composite materials.
2. EXPERIMENTAL PROCEDURES
Sub-micrometer spherical silica particles supplied by Chemland (Korea) were surface-functionalized
using two surface modifiers with contrasting characteristics, namely, GPTMS and OTES,
as well as GPTMS/OTES mixtures with various composition ratios. A 1 L amber container
was used for the treatment process of the spherical silica (240 g). Methyl ethyl ketone
(MEK, 150 g) served as the dispersion solvent, and 1,360 g of zirconia balls (0.3–0.4
mm in diameter) were utilized to ensure uniform mixing and mechanical agitation. A
total of five surface-modification formulations were prepared and introduced: pure
GPTMS (2.4 g), pure OTES (2.4 g), and GPTMS:OTES mixtures at ratios of 3:1 (1.8 g
: 0.6 g), 1:1 (1.2 g : 1.2 g), and 1:3 (0.6 g : 1.8 g). To promote monolayer surface
treatment rather than multilayer deposition, the total silane concentration was fixed
at 1 wt.% relative to the solvent. To ensure sufficient mixing and reaction, the mixtures
were evenly agitated for 3 h in a shaker operated at 600 RPM. No deliberate hydrolysis/condensation
pre-treatment was performed. Instead, silanization proceeded through trace-moisture-assisted
hydrolysis followed by subsequent condensation with surface silanol groups. After
the silanization treatments, the samples were washed three times with deionized water
to remove physically adsorbed or unreacted molecules. The samples were then dried
at room temperature for 24 h. The dried powders were used directly for scanning electron
microscopy (SEM, NNS-450, FEI Hong Kong Company) and Fourier-transform infrared (FT-IR,
IRTracer-100, Shimadzu, Japan) spectroscopy. Additionally, dynamic light scattering
(DLS, Zetasizer Advance Series Pro Blue, Malvern Panalytical, UK) analysis was performed
by redispersing the dried powders in a measurement solvent.
3. RESULTS AND DISCUSSION
Reaction mechanism between spherical silica and surface modifiers
The surface of pure spherical silica is densely covered with silanol groups (Si–OH),
which serve as reactive anchoring sites for organosilane coupling agents. GPTMS and
OTES undergo hydrolysis of their alkoxy groups in the presence of trace amounts of
water or controlled aqueous/alcohol mixed solvents, followed by condensation with
the surface silanol groups[24-
27]. Consequently, a stable siloxane bond (Si–O–Si) is established between the spherical
silica surface and the organosilane molecules.
In the case of GPTMS (Fig. 1), the three methoxy groups are first hydrolyzed to produce a silanol-containing intermediate,
while the terminal epoxy ring remains intact under mild hydrolysis conditions. These
newly formed silanol groups subsequently react with the hydroxyl groups on the silica
surface. Thus, GPTMS introduces epoxy-functionalized organic spacers onto the silica
surface, enabling additional chemical cross-linking with the epoxy matrix or curing
agents during composite processing. OTES follows a similar pathway (Fig. 2).
The triethoxy group is hydrolyzed and converted to silanol, after which it undergoes
a condensation reaction with the hydroxyl groups on the silica surface. This reaction
forms a hydrophobic alkyl chain shell around the silica particles, significantly lowering
their surface energy and inhibiting aggregation via hydrogen bonding.
Fig. 1. Surface reaction between spherical silica and GPTMS.
Fig. 2. Surface reaction between spherical silica and OTES.
Competitive and cooperative reactions between GPTMS and OTES in mixed systems
When GPTMS and OTES are introduced simultaneously at mixing ratios of 3:1, 1:1, or
1:3, the different surface modifiers compete for the available silanol adsorption
and reaction sites on the spherical silica surface. When the concentrations of GPTMS
and OTES are equal (1:1 ratio), the surface adsorption characteristics exhibit intermediate
features, resulting in an even distribution of both GPTMS and OTES (Fig. 3).
However, when one silane component predominates, as in the 3:1 or 1:3 GPTMS/OTES systems,
variations in the hydrolysis rates of the methoxy and ethoxy groups, three-dimensional
steric hindrance, and intermolecular condensation kinetics can increase the likelihood
of developing a non-uniform surface layer. For example, the high chemical reactivity
of the methoxy groups in GPTMS combined with the bulky alkyl chains of OTES may further
amplify surface asymmetry and compositional heterogeneity [28,
29].
Fig. 3. Surface reaction of silica with a 1:1 mixture of GPTMS and OTES.
Morphological characterization of pristine spherical silica
Fig. 4 shows the SEM micrograph of the pristine spherical silica powder (Fig. 4a) and its corresponding particle size distribution (Fig. 4b). The silica particles exhibit a highly uniform morphology characterized by smooth
surfaces and a narrow size distribution, with minimal variation in particle diameter.
No severe aggregation or structural collapse is observed. The average particle diameter
is below approximately 1 µm for all spherical silica samples (Fig. 4a). The particle size histogram confirms that the majority of the silica particles
are concentrated within a narrow diameter range of 0.6–0.7 µm. Although a small fraction
of larger or smaller particles is detected, the overall distribution is highly uniform.
This structural uniformity implies high reproducibility during subsequent silanization
with GPTMS, OTES, and their hybrid systems due to consistent surface interaction areas
(Fig. 4b). In particular, the smooth spherical morphology and uniform size distribution are
expected to prevent particle agglomeration on the silica surface and stabilize the
dispersion when polymer matrices are introduced.
Fig. 4. (a) SEM image and (b) size distribution of pristine spherical silica.
Morphological characterization of silane-modified spherical silica
Fig. 5 shows SEM images of spherical silica particles surface-functionalized using GPTMS
(Fig. 5a), OTES (Fig. 5b), and dual-silane systems with mixing ratios of 3:1 (Fig. 5c), 1:1 (Fig. 5d), and 1:3 (Fig. 5e). Compared to the pristine silica, no physical disintegration of the core structure
or severe particle aggregation is observed in any sample after the silane treatment.
This indicates that mechanical shaking during silanization did not impair the structural
integrity of the spherical silica, regardless of the silane formulation. However,
surface functionalization with GPTMS or OTES does not automatically guarantee complete
dispersion. Depending on the processing parameters, secondary aggregation can occur
due to inappropriate solvent selection, excessive silane concentrations, incomplete
hydrolysis, or over-condensation. For example, Mori et al.[30] reported that GPTMS-modified silica distributes well without significant aggregation
in aqueous media, whereas severe aggregation and sedimentation occur in toluene. This
behavior arises because hydrophilic spherical silica tends to agglomerate more strongly
in hydrophobic toluene, where attractive interparticle forces dominate over repulsive
forces. According to Greenwood[31], the reaction of GPTMS with water generates silanols that undergo siloxane condensation
and silica surface coupling; however, excessive condensation can produce cluster-like
silica aggregates, leaving the colloidal silica unstable. Casagrande et al.[32] suggested that the hydrolysis of GPTMS proceeds initially to increase silanol concentration,
but hydrolysis and condensation reactions occur competitively. Consequently, condensation
may initiate prior to complete hydrolysis, accelerating the formation of dense siloxane
networks and leading to the aggregation of GPTMS layers.
Similarly, functionalization with OTES does not inherently ensure flawless dispersion.
Secondary aggregation may occur depending on the treatment conditions due to the formation
of island-like domains, incomplete monolayers, multilayer deposition, or lateral alkyl-chain
interactions. Poda et al.[33] reported that C18 alkyl chains can form localized hydrophobic domains by strongly
inducing van der Waals interactions, which can trigger particle clustering. Brambilla
et al.[34] further explained that locally excessive silane concentrations can cause incomplete
surface coverage, resulting in silica–silica bridging or heterogeneous shell thicknesses.
However, as shown in Fig. 5, it is difficult to distinguish between aggregated and dispersed states using only
SEM micrographs for the GPTMS, OTES, and hybrid compositions. Contrary to previous
reports[30-
34], the silane concentration used in this study was well-suited for monolayer formation,
and the 3 h mechanical shaking promoted uniform silane bonding across the spherical
silica surface, thereby avoiding the common aggregation issues associated with GPTMS
and the dispersion limitations of OTES.
Fig. 5. SEM images of spherical silicas modified with various silanes (a) GPTMS, (b)
OTES, (c) GPTMS (3) : OTES (1), (d) GPTMS (1) : OTES (1), (e) GPTMS (1) : OTES (3).
FT-IR analysis of surface-modified spherical silica
To investigate the chemical bonding states of the pristine and silane-treated spherical
silica particles, FT-IR spectroscopy was employed to confirm surface functionalization
(Fig. 6).
Because silica consists primarily of siloxane (Si–O–Si) networks and surface silanol
(Si–OH) groups, tracking changes in these bands is highly effective for monitoring
the formation and extent of new surface bonds. As shown in Fig. 6, pristine silica exhibits three characteristic absorption bands: a strong, broad
peak corresponding to the asymmetric stretching vibration of the Si–O–Si bond between
1050 and 1100 cm-1
[35]; a symmetric stretching vibration of the Si–O–Si bond centered around 800 cm-1
[36]; and a Si–O bending vibration centered at 460 cm-1
[37]. Following modification with GPTMS, OTES, or their hybrid combinations, OH-based
bands—such as the O–H stretching vibrations in the 3200–3600 cm-1 range and the Si–OH stretching vibration centered around 950 cm-1—decreased to almost negligible intensities[38,
39]. Instead, changes were predominantly observed within the existing Si–O–Si and Si–O
frameworks.
This indicates that successful surface modification was achieved for all silane functional
groups. Notably, the finding that the single-silane type or dual-silane fraction exerted
only a marginal influence on the infrared spectra implies that the absolute silane
concentration is the governing factor determining the surface modification regime
of spherical silica under these conditions.
Fig. 6. FT-IR spectra of spherical silicas modified with various silanes (a) GPTMS,
(b) OTES, (c) G/O (3:1), (d) G/O (1:1), (e) G/O (1:3).
DLS analysis of surface-modified spherical silica
Subtle differences among the silane-modified spherical silica samples that were not
readily distinguishable via SEM and FT-IR were successfully evaluated using DLS, as
shown in Fig. 7.
While pristine spherical silica particles exhibit an average physical size of 0.6–0.7
µm, DLS measurements reflect an apparent hydrodynamic diameter that accounts for both
the inorganic silica core and the surface-bound organic silane shell. According to
Fig. 7, the pristine spherical silica particles have a hydrodynamic size of approximately
632 nm. Surface modification with GPTMS resulted in only a marginal increase to 634
nm, whereas OTES modification expanded the apparent particle diameter to 821 nm, indicating
that the OTES layer occupies a substantially larger interfacial volume. This difference
is mainly attributed to the lower molecular weight and shorter spatial extension of
GPTMS compared to OTES, alongside the more streamlined, smooth bonding configuration
of GPTMS on the silica surface. For the hybrid-modified samples, the 1:1 GPTMS/OTES
ratio exhibited an intermediate particle size (792 nm) falling precisely between those
of the pure GPTMS- and OTES-modified samples. Conversely, the significantly larger
hydrodynamic sizes observed for the 3:1 and 1:3 mixtures indicate increased interparticle
association, suggesting less uniform dispersion behavior under these specific composition
ratios. This qualitative interpretation, based on the DLS profiles, implies that while
the 1:1 ratio successfully balances the advantages of both GPTMS and OTES to establish
a stable interface, the 3:1 and 1:3 ratios promote the simultaneous formation of isolated
GPTMS or OTES domains or irregular mixed siloxane networks. This structural heterogeneity
increases the likelihood of multilayer formation or bridging flocculation[40,
41]. This hybrid approach is expected to provide a versatile platform for engineering
novel functionalities across a wide range of organic–inorganic composite systems.
Fig. 7. Comparison of DLS in pristine silica and silica with added GPTMS, OTES, G/O
(3:1), G/O (1:1), and G/O (1:3) silanes.
4. CONCLUSIONS
The surface modification of sub-micrometer spherical silica particles was successfully
carried out via wet-chemical functionalization using GPTMS (offering high interfacial
affinity) and OTES (providing excellent anti-aggregation stability) at a fixed concentration
of 1 wt.% each. To integrate the distinct advantages of both silane coupling agents
while mitigating their individual limitations, hybrid GPTMS/OTES dual-silane systems
with composition ratios of 3:1, 1:1, and 1:3 were systematically investigated. SEM,
FT-IR, and DLS analyses revealed that the 1:1 composition successfully achieved well-balanced,
intermediate interfacial properties between those of pure GPTMS and pure OTES, whereas
the 3:1 and 1:3 ratios resulted in broader and less uniform size distributions. These
findings demonstrate that dual-silane hybrid functionalization serves as an effective
strategy for balancing competing interfacial requirements that are otherwise difficult
to satisfy using a single surface modifier.
ACKNOWLEDGEMENTS
Won Seon Seo and Sangwoo Kim contributed equally to this work and are considered co-first
authors. This work was supported by the Industrial Strategic Technology Development
Program (RS-2024-00432810) funded by the Ministry of Trade, Industry & Energy(MOTIE,
Korea). This research was supported by the National Research Council of Science &
Technology(NST) grant by the Korea government (MSIT) (No. CPS25021-120).
REFERENCES
J. Hwang, H.-S. Kim, S.-I. Kim, J.W. Roh,
Korean J. Met. Mater. 62, 464 (2024).
J.H. Lee, S.K. Jeon, M.C. Choi, S.Y. Kim,
Korean J. Met. Mater. 63, 291 (2025).
Mitsuru Tanahashi, Kazuma Hirota,
Polym. Polym. Compos. 31, 1 (2023).
S. Kiba, N. Suzuki, Y Okawauchi,
Chem. – Asian J. 5, 2100 (2010).
N. T. Vo, A. Patra, D. Kim,
Phys. Chem. Chem. Phys. 19, 1937 (2017).
B. Coasne, A. Galarneau, R. J. M. Pellenq, F. Di Renzo,
Chem. Soc. Rev. 42, 4141 (2013).
J. G. Croissant, K. S. Butler, J. I. Zink, C. J. Brinker,
Nat. Rev. Mater. 5, 886 (2020).
X. Du, J. He,
Nanoscale 3, 3984 (2011).
J. Sharma, G. Polizos,
Nanomaterials 10, 1599 (2020).
S. Shokoohi, A. Arefazar, R. Khosrokhavar,
J. Reinf. Plast. Compos. 27, 473 (2008).
J. G. Croissant, K. S. Butler, J. I. Zink, C. J. Brinker,
Nat. Rev. Mater. 5, 886 (2020).
A. Issa, A. Luyt,
Polymers 11, 537 (2019).
D. Wang,
J. Appl. Polym. Sci. 141, e55507 (2024).
J. P. Matinlinna, C. Y. K. Lung, J. K. H. Tsoi,
Dent. Mater. 34, 13 (2018).
S. Shen, P. Sun, W. Li, A. N. Parikh, D. Hu,
Langmuir 26, 7708 (2010).
M. Mousavi, E. Fini,
ACS Sustain. Chem. Eng. 8, 3231 (2020).
L. Gabrielli,
RSC Adv 4, 1841 (2014).
V. G. Parale,
Ceram. Int. 39, 835 (2013).
C. E. Pyo, J. H. Chang,
ACS Omega 6, 16100 (2021).
R. Wang, G. Baran, S. L. Wunder,
Langmuir 16, 6298 (2000).
P. Silberzan, L. Lëger, D. Ausserré, J.J. Benattar,
Langmuir 7, 1647 (1991).
Y. Tong, E. Tyrode, M. Osawa, N. Yoshida, T. Watanabe, A. Nakajima, S. Ye,
Langmuir 27, 5420 (2011).
A. Markov, N. Wolf, X. Yuan, D. Mayer, V. Maybeck, A. Offenhäusser, R. Wördenweber,
ACS Appl. Mater. Interfaces 9, 29265 (2017).
F.D. Osterholtz, E.R. Pohl,
J. Adhes. Sci. Technol. 6, 127 (1992).
F.D. Blum, W. Meesiri, H.J. Kang, J.E. Gambogi,
J. Adhes. Sci. Technol. 5, 479 (1991).
T. Nakamura, R. Tsutsumi, C. Hashiguchi, T. Terao, K. Manabe, T. Hirai, S. Fujii,
Y. Nakamura,
J. Appl. Polym. Sci. 138, e51297 (2021).
M.-C. B. Salon, M. N. Belgacem,
Phosphorus Sulfur Silicon Relat. Elem. 186, 240 (2011).
Y. Nakamura, R. Yamazaki, T. Fukuda, K. Shitajima, S. Fujii, M. Sasaki,
J. Adhes. Sci. Technol. 28, 1895 (2014).
F. Vilmin, I. Bottero, A. Travert, N. Malicki, F. Gaboriaud, A. Trivella, F. Thibault-Starzyk,
J. Phys. Chem. C 118, 4056 (2014).
T. Mori, Y. Okada, H. Kamiya,
Adv. Powder Technol. 27, 830 (2016).
P. Greenwood,
Surface Modifications and Applications of Aqueous Silica Sols, Chalmers Univ. Technol. (2010).
C. A. Casagrande, L. F. Jochem, W. L Repette,
Matéria (Rio de Janeiro) 25, e12811 (2020).
J. Dong, A. Wang, K. Y. Simon Ng, G. Mao,
Thin Solid Films 515, 2116 (2006).
R. P. Bagwe, L. R. Hilliard, W. Tan,
Langmuir 22, 4357 (2006).
J.-W. Park, Y. J. Park, C.-H. Jun,
Chem. Commun. 47, 4860 (2011).
J. Hu, W. Wang, R. Yu, M. Guo, C. He, X. Xie, H. Peng, Z. Xue,
RSC Adv. 7, 54986 (2017).
J.-Y. Lin, B.-X. Wang,
Adv. Mater. Sci. Eng. 2014, 594516 (2014).
B. A. Morrow, A. J. McFarlan,
J. Phys. Chem. 96, 1395 (1992).
S. L. Warring, D. A. Beattie, A. J. McQuillan,
Langmuir 32, 1568 (2016).
N. García, E. Benito, J. Guzmán, P. Tiemblo,
J. Am. Chem. Soc. 129, 5052 (2007).
W. He, D. Wu, J. Li, K. Zhang, Y. Xiang, L. Long, S. Qin, J. Yu, Q. Zhang,
Bull. Korean Chem. Soc. 34, 2747 (2013).