The Effect of 3-(Glycidoloxy Propyl) Trimethoxy Silane Concentration on Surface Modification of SiO2 Nanoparticles

SiO2 is widely used in nanocomposites as reinforcement nanoparticle to enhance mechanical properties especially wear resistivity. Prior to use, surface modification with proper and sufficient coupling agent should be performed on it. Coupling agent concentration plays a key role in modification process. In this study, the influence of 3-(glycidoloxy propyl) trimethoxy silane (GPTMS) concentration on surface modification of SiO2 nanoparticles, is experimentally investigated. The surface modification of nano-silica were performed by 30, 50, 80 and 110 wt.% of GPTMS in order to introduce the optimal GPTMS concentration to complete the process. Fourier Transformation Infrared Spectroscopy (FTIR), Field Emission Scanning Electron Microscopy (FESEM), Thermo Gravimetric Analysis (TGA) and X-Ray Diffraction (XRD) characterized the pure and surface modified samples; then, the results were compared to each other to achieve the aim of the research. FTIR results confirmed the silanization proceed due to the silane absorption peak disappearing and shifting of the hydroxyl group bonds in to the amide bonds. This test showed that 30 wt.% GPTMS has not been sufficient for full functionalization of the NPs. According to FESEM images, it seems that the NPs were better modified by 80 wt.% GPTMS due to the least NPs aggregation and lack of coupling agent deposition on the NPs. Also, TGA illustrates that this sample has higher thermal stability because of lower weight loss (11.2%) in coupling agent decomposition temperature range: 130–380 °C. Furthermore, X-Ray Diffraction confirmed the FESEM and TGA results about the mentioned sample due to its highest crystallite size (increase 26.64% in crystallite size in comparison with the pure sample). So, the 80 wt.% of GPTMS introduced as the optimal concentration for surface modification of SiO2 nanoparticles.


Introduction
Silicon dioxide nanoparticles is known as the first produced nanoparticles (NPs) among inorganic materials [1]. This white powder can be found in two major forms; P-type (Porous particles) and S-type (Spherical particles) according to the NPs structure [2]. P-type nano-silica surface containing a number of nano-porous has much larger Specific Surface Area (SSA) comparing to S-type. They have been widely used in nanocomposite preparation as reinforcement nanoparticles due to their excellent properties [3].
Many researchers have been reported that nano-silica could enhance the mechanical and thermal properties of the polymeric matrix, such as toughness, stiffness [4], tensile strength, elongation [5], heat resistance [6], hardness [7],wear resistance [8], adhesion to steel substrate [9] and optical properties [10]. These properties improvements have been mentioned due to hydrogen bonds between nano-silica surface polar groups (-OH) and polar groups of the polymer chains [11]. However, the effects of these NPs on nanocomposites properties, also depends on proper dispersion within the matrix and how to react with the polymer chains [12].
NPs tend to agglomerate in the matrix due to their size and very high surface energy. There are some experiences in prevention of nano-silica agglomeration and increasing of NPs surface reactivity due to surface modification by physical and chemical treatments [13][14][15][16][17]. Functionalized SiO 2 NPs were widely used in nanocomposites preparation as reinforcement particles to achieve higher mechanical properties and more wear resistivity [11,26].
There are some different coupling agents which were used to modify the SiO 2 nanoparticles surfaces, such as acids (H 2 SO 4 , HNO 3 and etc.), carboxyl groups and silane compositions and so on [18][19][20]; However, the most common method to functionalize the nano-silica is organo-silane chemical treatment. Silane modification of SiO 2 NPs in comparison with the other coupling agents causes stronger chemical bonds and more compatibility between the NPs surfaces and polymer chains [4,21].
Silane structure can be introduced as (RO) 3 -Si-R'-X which RO, R' and X refers to hydrolysable group, alkylene group and organo functional group, respectively. "RO" and "X" established strong bonds between the NPs and polymer chains, respectively [22]. Amine groups also created strong bonds between the silane compounds and the polymeric matrix [23].
Recently, silanization of nano-silica processes have been reported in many papers; however, except for few of them, the optimal concentration of the required silane compounds for this issue has not been investigated. For instance, Xu and et al. [20] didn't even mention the exact ratio of used silane compounds to surface modification of nano-silica. In other examples, Chuang and et.al. [24] as like as Ghosh and et al. [18] mentioned this ratio but didn't discuss about different silane concentrations effect on this process. Although, Rong et al. [25] have studied the effect of different silane concentrations (from 10 up to 30 wt.%) on nano-silica functionalization and introduced 30 wt.% silane as optimal amount; the effect of using more than 30 wt.% silane in this process has not been investigated.
So, in this study, nano-SiO 2 functionalization was started by using 30 wt.% silane and continued with higher silane concentrations. The proper methods characterized the samples to determine the effect of silane concentration on SiO 2 NPs surface modification.

Materials
Materials, which were used in the experiments, are introduced in Table 1.

Surface Modification
1 g of raw nano-silica was added to 30 ml solution of distilled water/ethanol 96% (1:1) and it was stirred to obtain uniform solution. Then, 30, 50, 80 and 110 wt.% (with respect to nano-SiO 2 weight) of GPTMS as coupling agent was added to the solution; and in order to set PH of the solution on 4, 0.01 mol of HCl (37%) was added. Mechanical stirrer (1000 rpm) for 30 min stirred the resulted solution and heilscher ultrasonic homogenizer, UIP1000hd, 20 kHz, 1000 W, Germany, with amplitude of 90% for 5 min homogenized it. MSE MISTRAL 3000E, U.K (5000 rpm for 30 min), centrifuged the homogenized solution and the resulted gel was dried in a vacuum oven at 80°C for 36 h to complete the silanization process of nano-SiO 2 .

Characterization
The success of functionalization process was examined by FTIR spectroscopy on GPTMS and pure/functionalized nano-silica using BRUKER-VERTEX 70, Germany, in the frequency range of 4000-400 cm −1 .
Furthermore, Thermo Gravimetric Analysis (TGA) analyzed thermal stability of the functionalized samples (according to ASTM E1131. ISO 11358 standards) in such a way that about 10 mg of each samples were heated from 25 to 800°C in N 2 atmosphere, at rate of 10 K min −1 , using METTLER TOLEDO, U.S., instrument.
Finally, XRD method performed phase identification and crystallite size determination for the functionalized samples using Bruker, D8 Advanced, Germany, with Cu-Ka radiation and 35KV accelerating voltage. Crystallite size was calculated by using well-known Scherrer equation: [29].
Where D is the crystallite size (nm), k is the Scherrer constant (0.91), k is the X-ray wave length (1.5406 A°), b is the full width and half maximum (FWHM), and h is the Bragg angle.
Consequently, proceed in modification process was confirmed by FTIR due to disappearing of silane group absorption peak and also shifting the hydroxyl group bonds in to amide bonds in either modified samples. Moreover, disappearing of the silane group peak in IR spectra of functionalized samples illustrated that all the silane groups were consumed in either sample during functionalization process. Furthermore, intensifying the N-H, C-CH3 and C-CH2 peaks by increasing of silane concentration (from 30 wt.% up to 50 wt.%) at 3425 cm −1 , 2935 cm −1 and 2877 cm −1 , respectively, showed that 30 wt.% silane was not sufficient to complete the surface modification process.
So, the study was followed by performing other tests on the samples which modified with 50 wt.% and more silane concentrations. Figure 2 shows the morphology of the pure and functionalized nano-silica samples with FESEM micrographs. The histogram of the samples also illustrate in Fig. 3. The measured average diameters of the NPs in pure sample ( Fig. 2-a)

*Specific Surface Area
Scheme 1 Nano-SiO 2 silanization mechanism and 26.26 nm, respectively. Decreasing in NPs average diameter of the surface modified samples compared to the pure sample was due to the use of hydrochloric acid in functionalization process [38]. According to Figs. 2 and 3, as the silane concentration increases in the functionalization process, the NPs average diameter increases. It might be due to increase in bonds between the NPs and coupling agent or silane deposition on the NPs [25].
According to Fig. 2-a, nano-SiO 2 particles agglomerated in some areas due to very high surface energy. As shown in Fig.  2-b, NPs functionalization using 50 wt.% GPTMS, appeared to be insufficient because some aggregations were visible. However, NPs functionalization with 80 wt.% GPTMS (Fig.  2-c) should be more effective because of the least visible NPs aggregations. Next, Fig. 2-d shows that functionalization of the NPs with 110 wt.% GPTMS led to some deposition on the NPs instead of the surface modification.

Thermal Analysis
Thermo Gravimetric analysis was performed on the samples and the resulted curves illustrates in Fig. 4. The curves were described in three separated stages in terms of temperature ranges and indicated the stability of the bonds between silane and the NPs for each sample in second stage.
In the first stage, in range 25 and 95°C, there were 0.63, 0.8, 0.7 and 1.16% weight losses in pure, 50, 80 and 110 wt.% GPTMS-SiO 2 nanoparticles, due to evaporation of absorbed water during the modification process. The obtained results showed that the amount of water absorption in the last sample was distinctly different from other samples and this might be due to differences in how silane reacted with nano-silica.
The second stage related to thermal stability of the samples against decomposition of the NPs surfaces' silane groups, in range 130 and 380°C. In this stage, it can be seen 0.3, 11.37, 11.2 and 14.65 weight losses percentage in pure, 50, 80 and 110 wt.% GPTMS-nanosilica, respectively. Lack of functional groups on pure sample surfaces caused negligible weight loss during this stage. Thermal stability in this analysis is directly proportional to bonding strength between coupling agent and NPs. Therefore, in TGA, the higher weight losses percentage, the less bonding strength between silane and samples' surfaces. The third stage corresponded to decomposition of nano-silica structures in range 483.5 and 560°C. C o n s e q u e n t l y , T G A c o n f i r m e d n a n o -s i l i c a functionalization, because there were at least 11.2% weight loss in the specimens, which related to decomposition of silane compounds from nano-silica surfaces in range 130, and 380°C. As a result, 80 wt.% GPTMS-SiO 2 NPs was known as better modified sample due to the least weight loss and more stability of grafted coupling agent on the NPs at related temperature range. Figure 5, shows the XRD patterns of the pure and modified nano-silica samples with different coupling agent concentrations. Tetrahedral arrangement of O atoms around the Si atoms caused two dimensional structure with a local short range ordering in amorphous nano-silica [39]. Therefore, there was not any sharp peak in XRD patterns' and as shown in Fig. 5, it can be seen only a broad peak related to quartz in range 20 and 22 degree [40][41][42].

XRD Analysis
In case of silanization of nano-silica, the characteristic peak in the XRD patterns tends to crystalline mode due to some changes in the ordering of amorphous structure.
The XRD analysis results summarize in Table 2. According to the analysis results, nano-silica functionalization with different amounts of silane groups caused a little change in the broad peaks position in XRD pattern. Therefore, it would be possible to survey the effectiveness of silane concentration on functionalization process with comparing the XRD patterns.
As shown in Fig. 5-a, the broad peak was appeared at 20.792°which was related to pure amorphous nanosilica.
According to Fig. 5-b and Table 2, nano-silica modification with 50 wt.% silane didn't make much differences in the broad   Finally, with respect to Fig. 5-d, increase in silane concentration up to 110 wt.% in functionalization process of sample D, increased the width of the peak again; and also, according to Table 2, the least crystallite size was reported for this sample (decreasing the crystallite size at about 19.3% in comparison with the pure sample). It might be due to precipitate the silane on nano-silica instead of modification and increase in disordering of atomic arrangement in microstructure of sample D.

Conclusion
This paper investigated the effect of 3-(glycidoloxy propyl) trimethoxy silane concentration on surface modification of SiO 2 nanoparticles in details. So, the pure and surface modified samples with 30, 50, 80 and 110 wt.% GPTMS, were analyzed by FTIR, FE-SEM, TGA and XRD. FTIR analysis confirmed progress of nano-silica functionalization process with GPTMS due to disappearing of the silane absorption peak in functionalized samples and shifting of the hydroxyl group peak in pure sample spectra in to the amide in modified specimens. According to FE-SEM images, increasing the GPTMS concentration in the samples modification caused increase in mean diameter size of the NPs due to grafted silane on the NPs surface; and it seemed functionalization of nano-silica with 80 wt.% GPTMS led to minimal NPs agglomeration and also lack of silane deposition on the NPs surfaces. Furthermore, TGA indicated that 80 wt.% GPTMS grafted sample has higher thermal stability in coupling agent decomposition range at about 130-380°C among the samples. Moreover, XRD analysis showed that this sample was better surface modified in comparison with the other samples due to its higher crystallite size. Finally, 80 wt.% GPTMS is introduced as optimal coupling agent concentration to modify the SiO 2 nanoparticles surfaces.
Acknowledgments The authors would like acknowledge from Ministry of Science, Research and Technology of Islamic Republic of Iran for supporting this work.
Author Contributions Bahramnia H. designed and performed experiments, investigated the results and wrote manuscript draft.
Mohammadian Semnani H. supervised the entire research and supported financially the experiments.
Habibolahzadeh A. conceptualized the research. Abdoos H. performed data curation, reviewed and edited the manuscript.
Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Data Availability Electronic supplementary materials contains FTIR, TGA and XRD original data