Citation
Superhydrophobic Coating Synthesis via Silane Modifiers

Material Information

Title:
Superhydrophobic Coating Synthesis via Silane Modifiers
Creator:
Tang, Eric L.
Publication Date:
Language:
English

Notes

Abstract:
Silica nanoparticles modified with three different silane coupling agents: n-octyltrichlorosilane, n-decyltrichlorosilane, and dodecyltrichlorosilane were dispersed into four different polymer matrices: Lumiflon, styrene-butadiene-styrene, poly (methyl methylacrylate), and poly vinyl alcohol. Coated samples were prepared and contact angles of water droplets on the samples were measured to characterize their wettability. It was observed that modification of silica nanoparticles using n-octyltrichlorosilane yields superhydrophobic surfaces but using n-decyltrichlorosilane or dodecyltrichlorosilane yields surfaces with significantly lower contact angle. Samples that were dispersed in polyvinyl alcohol obtained superhydrophobicity, possibly due to functionalization of silica nanoparticles from hydroxyl groups in polyvinyl alcohol. The potential of this functionalization sparks the idea of possible functionalization of silica nanoparticles using other polymers. ( en )
General Note:
Awarded Bachelor of Science in Chemical Engineering, magna cum laude, on May 8, 2018. Major: Chemical Engineering
General Note:
College or School: College of Engineering
General Note:
Advisor: Wolfgang Sigmund. Advisor Department or School: Materials Science and Engineering

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Eric L. Tang. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.

UFDC Membership

Aggregations:
UF Undergraduate Honors Theses

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

Eric Tang April 2018 Department of Chemical Engineering

PAGE 2

2 Table of Contents Abstract ................................ ................................ ................................ ................................ ......................... 3 Introduction ................................ ................................ ................................ ................................ .................. 3 Methods ................................ ................................ ................................ ................................ ........................ 7 Silica Nanoparticle Modification ................................ ................................ ................................ ............... 7 Preparation of Coating Samples ................................ ................................ ................................ ............... 8 Results and Discussion ................................ ................................ ................................ ................................ 10 Silica Modified with dodecyltrichlorosilane ................................ ................................ ............................ 10 Silica Modified with n decyltrichlorosilane ................................ ................................ ............................. 12 Silica Mod ified with n ocyltrichlorosilane ................................ ................................ ............................... 13 Analysis of Contact Angle Measurements ................................ ................................ .............................. 14 Dispersion of Modified Silica into Lumiflon matrix ................................ ................................ ............. 14 Effect of Different Silane Coupling Agents on Contact Angle ................................ ............................. 14 Dispersion of Modified Silica into PVA matri x ................................ ................................ .................... 15 Conclusion ................................ ................................ ................................ ................................ ................... 15 References ................................ ................................ ................................ ................................ .................. 16

PAGE 3

3 Abstract Silica nanopartic l es modified with three different silane coupling agents: n octyltrichlorosilane, n decyltrichlorosilane, and dodecyltrichlorosilane were dispersed into four different polymer matrices: Lumiflon, styrene butadiene styrene, poly ( methyl methylacrylate), and poly vinyl alcohol. Coated samples were prepared and c ontact angles of water droplets on the samples were measured to characterize their wettability It was observed that modification of silica nanoparticles using n octyltrichlorosilane yields superhydrophobic surfaces but using n decyltrichlorosilane or dodecyltrichlorosilane yields surfaces with significantly lower contact angle. Samples that were dispersed in poly vinyl alcohol obtained superhydrophobicity possibly due to functionaliza tion of silica nanoparticles from hydroxyl groups in polyvinyl alcohol. The potential of this functionalization sparks the idea of possible functionalization of silica nanoparticles using other polymers. Introduction Superhydrophobic surfaces have excellen t water repellency and a re a growing field in surface science due to the ir potential applications in self cleaning materials and in anti fouling and anti corrosion applications. Superhydrophobic surfaces are defined as surfaces that yield a contact angle g reater than 150 and sliding angle less than 10 when a droplet of water is placed on top of it The contact angle is the angle between the bottom of a surface and the point where a water droplet starts to advance from the surface as seen in Figure 1. Contact angle measurements measure the wettability of a solid surface by a liquid If a contact angle is less than 90 then the water droplet has spread over the surface. This is a case of good wetting and is characteristic of hydrophilic surfaces. On th e other hand, if a contact angle is greater than 90 then the water droplet does not spread over the surface and is an example of poor wetting. Surfaces with contact angles greater than 90 are hydrophobic. Figure 1 Examples o f Contact Angles. (a) No wetting. (b) Hydrophobic (c) Neither hydrophobic or hydrophilic (d) Hydrophilic (e) Perfect wetting A contributing factor to the wettability of a liquid on a solid is surface tension. Surface tension of a liquid is what gives it s f ree drop spherical shape and also determines its wettability on a solid surface. For a given solid, a liquid with low surface tension will produce more wetness than a liquid with high surface tension [1] Additionally, surface energy of the solid surface is another important factor in the wettability of a solid liquid interface. Surfaces with high er surface energy show a greater attraction to water, and therefore, will be easy to wet. Surfaces with lower surface energy repel water, and thus, are difficult to wet and are more hydrophobic. Surface tension and surface energy between liquids and solid ultimately contribute to the free energy of a surface. Water droplets will take the form that is most

PAGE 4

4 thermodynamically stable on a surface will therefore exhibi t a stationary or static contact angle. The static contact angle equation C reated by T. Young in 1805, equation creates a relationship between the surface tensions between the surface and the droplet assum ing the simplest case that a solid surface is molecularly flat and chemically homogeneous [2] Where t he represent the surface tensions of solid vapor, solid liquid, and liquid vapor interfaces an d eq represents the contact angle at equilibrium on a smooth surface. However, surfaces are generally rough surfaces, and surface roughness and chemical heterogeneity combine to affect the value of the contact angle. Two additional models account for the se factors, and transition between these two models are largely what determines the superhydrophobic state of a surface Where w represents the resultant contact angle of a rough surface that is directly proportional to r, the surface roughing factor. completely removes the air between the solid vapor interface. However, it has been observed, that this is not always the case, that when a droplet is placed on a rough surface, it can remove hardly any air, and there exists a heterogeneous contact region of solid vapor and solid liquid. This state can be described by the Cassie Baxter model [ 3 ] Where f SL represents the contact fraction between solid and liquid. A comparison of a water droplet in the Wenzel state and Cassie Baxter state is shown in Figure 2. Figure 2 (a) Wenzel state: Water droplet adheres strongly to and fills in all of the gaps of a rough surface. (b) Cassie Baxter state: Water droplet sits on top of the surface, and gaps in surface are filled in with vapor, creating a hetero geneous contact region. As stated, another property of superhydrophobic surfaces is that the sliding angle is less than 10 The sliding angle occurs when a surface is tilted on an incline until the water droplet on it begins to roll. The angle at which the surface was tilted to is known as the sliding angle, as seen in Figure 3

PAGE 5

5 Figure 3 Example of sliding angle, the lowest inclined angle required to cause a water droplet to roll off a surface Just before the water droplet slides off the surface, the angles on both sides of the droplet can be measured. These angles are known as the advancing angle and receding angle, which can also be seen from Figure 3 The difference between these angles is known as contact angle h ysteresis [4] Surfaces with both low contact angle hysteresis and low sliding angle are considered to be in the Cassie Baxter state. Surfaces with high contact angle hysteresis and high sliding angle are considered to be in the Wenzel st ate. It is important to know the properties between both states, because each state is not fixed. In fact, in some cases, there exists coexistence between these states, and the energy barrier between these two states will cause transition from one state to the other [ 4]. It is important to know factors that cause this transitio n when designing superhydrophobic materials Surfaces can be made superhydrophobic by application of superhydrophobic coatings. Superhydrophobic coatings are commercially available and are made from polytetrafluoroethylene, otherwise known as Teflon Current applications of Teflon based superhydrophobic coatings are for anti wetting, corrosion prevention, for use in engines, and boats to reduce drag. However, c urrent limitations of superhydrophobic coatings are that they have limited durability. Therefore, work is being done to find ways to create superhydrophobic coatings that are durable. One method to making superhydrophobic coatings is to use silica nanoparticles Silica nanopar ticles have suitable properties for use in superhydrophobic surfaces due to their high thermal resistance, good mechanical strength, and porous structures that allow for rough surfaces [ 2] Methods to make superhydrophobic surfaces from silica include dip coating, electrospinning, etching and chemical vapor deposition. Superhydrophobic coatings can also be made through modification of silica nanoparticles. Silica SiO 2 readily hydrolyzes in the presence of water, or water vapor from humidity to form silanol groups, which are Si bonded with OH groups Silanes coupling agents are commonly used to modify silanol groups due to their structure, which is shown in Figure 4 Silanes generally are made of a Si atom that is bonded to an organic R group and thr ee hydrolysable groups such as alkoxy groups, halogens, or amines When t hese three groups become hydrolyzed they form a silanol group that is bonded with three hydroxyl groups and the R group [5] When this newly formed silanol group comes into contact with the silanol formed from silica, a condensation reaction occurs to form a linkage. This general reaction scheme is shown below in Figure s 5 and 6

PAGE 6

6 The resulting product are modified silica particles that now instead of being bonded to just hydroxyl groups, are bonded to an organic R group. The R group is a nonhydrolyzable organic radical such as a long aliphatic carbon chain [6] This group is a dangling bond that can be introduced onto the surface o f a substrate to increase the roughness of the surface. Introduction of more dangling bonds to the surface will lower the overall surface energy of the surface and increase the surface tension on the liquid until the dangling bonds are broken. For thi s project, trichloro alkyl silanes of different carbon chain length s were used to modify silica nanoparticles, and their effect on surface roughness was observed through contact angle measurements. Figure 4 General structure of a silane coupling agent. Si atom bonded to three hydrolysable groups and an organic functional group Figure 5 General r eaction scheme for hydrolyzation of a silane Figure 6 Reaction of formation of siloxane groups Silica nanoparticles are commonly dispersed into polymer matrices to increase their mechanical properties. Because polymers vary widely in their properties including intrinsic wettability, polymer selection plays a n important role the ultimate wettability of the coated surface. Because surface

PAGE 7

7 roughness can be modified through the use of silane coupling agents, it is not necessary to use a polymer that is intrinsically hydrophobic. It may be more advantageous to use polymers that have high mechanical properties that will assist the surface in handling wear to preserve the superhydrophobic state of the coating. Thus, o ne of the goals of this project was to investigate the effects of dispersing silica nanoparticles in to polymers with different functional groups, intrinsic wettabilities, and mechanical properties. Methods Silica Nanoparticle Modification Silica nanoparticles were synthesized from Aerosil Ox 50 Silica particles The particles were dispersed in chloroform for 1 h ou r using a magnetic stirrer, at a ratio of 3 g silica per 3 0 mL chloroform. Silane coupling agents from Gelest used were: n octyltrichlorosilane, n decyltrichlorosilane, and dodecyltrichlorosilane shown below in Figures 7 9 For different samples 3 mL of alkyl trichlorosilane coupling agent was added (1 mL of alkyl trichlorosilane per g of silica) and the particles were treated for 1 hour. After, the particles w ere centrifuged and decanted and then cleaned with 25 mL of ethanol, centrifuged and decanted. The resulting modified silica nanoparticles were dried in an oven at 70 C overnight. Figure 7 Structure of n octyltrichlorosilane Figure 8 Structure of n decyltrichlorosilane

PAGE 8

8 Figure 9 Structure of dodecyltrichlorosilane P reparation of Coating Samples Coating sample s were performed using a solid loading ratio of 60 volume percent silica and 40 volume percent polymer. This ratio was chosen because higher solid loa dings of silica will result in higher contact angle s ; however, at higher concentrations of silica, it is likely that the coating will not be durable. Therefore, 60 volume percent of silica was deemed a reasonable ratio to use. Polymer s used were: Lumiflon LF 200F, styrene butadiene styrene block copolymer ( SBS ) p oly(methyl methacrylate) ( PMMA ) and polyvinyl alcohol ( PVA ) and structure of each polymer is shown below in Figures 10 13 Figure 10 Lumiflon is a fluorinated polymer that is commercially available for use in many coatings. It is expected that Lumiflon based coatings will exhibit high degrees of superhydrophobicity, and the hydrophobicity of the other polymers will be compare d to Lu miflon Figure 11 SBS is a block copolymer and an elastomer, with domains within its matrices that can allow for better dispersion of nanoparticles. It will be observed if these properties enhance the surface roughing of the modified silica nanopa rticles.

PAGE 9

9 Figure 12 PMMA is a transparent thermoplastic that is commonly used as an alternative to glass. Figure 13 PVA is a synthetic polymer, with a repeating hydroxyl group. The effects of hydroxyl gr oups in PVA and if it reacts with the hydroxyl groups on a silanol will be explored. Polymers were individually dissolved in 5 mL of methyl ethyl ketone (MEK) Modified silica nanoparticles were added. Particles were stirred for 3 hours to allow for dispersion. After, dispersed nanoparticles were spin coated onto a glass slide using a dynamic dispensing method. The spin coating w as performed using 200 L of solution per sample at 1000 revolutions per mi nute (rpm) for 30 seconds. Samples were left to sit for a day to allow all of the MEK to evaporate. Pictures of water droplets were taken using a camera an d ImageJ

PAGE 10

10 Results and Discussion Before samples with modified silica were prepared, samples of nonmodified silica were prepared at the same constant solids loading ratio of 60 volume percent silica to 40 volume percent polymer. This was done to know the contact angle of si lica only, so that the degree of surface roughness that the silane coupling agents contribute could be quantified Additionally, contact angles of each polymer individually was measured to record the intrinsic wettabilities of the polymers. This was done b y preparing samples using the same method stated in the procedure but using 100 volume percent polymer. Resulting contact angles of coated samples with nonmodified silica and intrinsic contact angles of each polymer are shown below in Figures 14 and 15 Figure 14 Intrinsic contact angles of: (a) Lumiflon ( 85 1 ), (b) SBS (69 1 ), (c) PMMA (74 1 ), (d) PVA (41 1 ) Figure 15 Contact angles of: (a) Lumiflon + nonmodified silica ( 134 2 ), (b) SBS + nonmodifi ed silica ( 70 1 ), (c) PMMA + nonmodified silica ( 68 1 ), (d) PVA + nonmodified silica ( 1 5 1 ) As seen from Figure 14 three of the polymers used were slightly hydrophilic with contact angles within 90 , and PVA is moderately hydrophilic with a contact angle of 41 Upon dispersing silica nanoparticles into the polymers, the contact angles for Lumiflon, SBS, and PMMA increased expectedly, as silica adds surface roughness. Surprisingly, the contact angle of PVA decreased from its intrinsic contact angle w hen it is dispersed with silica. Reasons for this occurrence could be hydroxyl groups from PVA bonding with silica as proposed; however, it was expected that additional bonds and linkages between molecules would lead to higher surface roughness. Another ex planation for the almost perfect wetting of water onto the PVA coated sample is that PVA is water soluble. In the opposite scenario, it could be that PVA does not form any bonds with nonmodified silica, and therefore, due to its high solubility with water, is very easy to wet despite potentially having a rougher surface from silica dispersion S ilica Modified with dodecyltrichlorosilane In order to assess reproducibility of coating samples for each coupling agent, multiple batches of silica nanoparticles were modified and a new batch was spin coated onto samples for each day a trial was performed Two glass sides were coated per polymer per trial. For contact angle measurements, two water droplets were placed on each glass slide totaling 4 contact angle measurements for each sample

PAGE 11

11 per trial Contact angle measurements were uniform on the same sample but varied across different samples. Achieving samples with consistent values of contact angles across different days proved to be the most challenging task of this project. Therefore, many samples were prepared over days, and statistical analysis of the contact angle values were performed. Average values and ranges of contact angles and a picture of a water droplet close to that value for samples containing silica modified with dodecyltrichlorosilane are presented in Table 1 It is important to point out that while most of the ranges for contact angles were about 6 samples modified with dodecyltrichlorosilane and dispersed in PMMA varied significantly Pol ymer Average Contact Angle Lumiflon 14 6 (Ranged from 142 to 150) SBS 12 3 (Ranged from 120 to 126 ) PMMA 11 8 (Ranged from 104 to 130 ) PVA 15 4 (Ranged from 150 to 156) Table 1 Average Contact Angles for Samples Modified with Dodecyltrichlorosilane From the table, most notably, the average contact angle of silica nanoparticles dispersed in PVA was 154 This value is larger than the value of Lumiflon at 146 and classifies the c oated sample as superhydrophobic. After witnessing a sample coated with PVA and nonmodified silica exhibit a contact

PAGE 12

12 angle of 15 it is surprising again, to see a significant change in its wettability. Reason for this change could be that modifying silica introduces more hydroxyl groups and siloxane linkages that cause PVA to bond to it and lose its solubility with water. It will be seen if this observation stayed consistent for silane coupling agents of n decyltrichlorosilane and n octyltrichlorosilane. S ilica Modified with n decyltrichlorosilane Average values and ranges of contact angles for samples containing silica modified with n decyltrichlorosilane are presented in Table 2 Polymer Average Contact Angle Lumiflon 149 (Ranged from 145 to 153) SBS 124 (Ranged from 120 to 127) PMMA 117 (Ranged from 105 to 126) PVA 158 (Ranged from 155 to 160) Table 2 Average Contact Angles for Samples Modified with n decyltrichlorosilane

PAGE 13

13 Silica Modified with n ocyltrichlorosilane Average values and ranges of contact angles for samples containing silica modified with n ocyltrichlorosilane are presented in Table 3 Polymer Average Contact Angle Lumiflon 149 (Ranged from 147 to 155 ) SBS 147 (Ranged from 145 to 150 ) PMMA 143 (Ranged from 133 to 148 ) PVA 152 (Ranged from 148 to 154 ) Table 3 Average Contact Angles for Samples Modified with n o c t yltrichlorosilane

PAGE 14

14 Analysis of Contact Angle Measurements The bar graph in Figure 16 summarizes the average values for the contact angles of all the coated samples. Figure 16 Average Contact Angles for Coated Samples Dispersion of Modified Silica into Lumiflon matrix From the graph, it is observed that regardless of coupling agent, dispersion of modified silica into Lumiflon and PVA can both produce superhydrophobic surfaces. Coupling agent used does not seem to affect the contact angle of a surface coa ted with Lumiflon. This is likely because Lumiflon is fluorinated, and the fluoro groups are also hydrolysable. The fluoro groups may interact with the modified silica and limit the effect of changing the R group. Effect of Different Silane Coupling Agen ts on Contact Angle For all polymers used, modification of silica using a silane coupling agent significantly increased the contact angle from the contact angle values from before they were modified. Varying coupling agent had no observable effect on silica particles dispersed in Lumiflon. For PVA, the average contact angle of samples coated with n decyltrichlorosilane and dodecyltrichlorosilane is significantly greater than that of n octyltrichlorosil ane. I t appears that using a longer carbon chain trichlorosilane increases surface roughness, and contact angle. However, because the ranges of measured contact angles for samples coated with PVA was around 6 and the values of contact angles measured on samples coated with n octyltrichlorosilane overlapped with values of the other coupling agents, while it is possible, it cannot be said for certain that increasing the carbon chain length of an R group of a trichlorosilane increases its contact angle. N o ctyltrichlorosilane was observed to be the most effective silane coupling agent to introduce surface roughness. Coatings that were modified with n octyltrichlorosilane and dispersed in SBS and PMMA showed significant increases in contact angle reaching av erage values of 147 and 143 However, samples dispersed in SBS and PMMA that were modified with n decyltrichlorosilane and dodecyltrichlorosilane surprisingly showed much lower contact angles Reasons that modification of silica nanoparticles using only n octyltrichlorosilane was capable of introducing enough surface roughness to create superhydrophobic surfaces could be because 8 carbon chain length R group could be the limit for which contributions to surface roughness is effective Adding 8 carbon chai n silane groups to the silica 115 120 125 130 135 140 145 150 155 160 Lumiflon SBS PMMA PVA Contact Angle ( ) Polymer dodecyltrichlorosilane n-decyltrichlorosilane n-octyltrichlorosilane

PAGE 15

15 nanoparticles could be the maximum length that leads to the most agglomerations that make linkages within the siloxane groups more packed closer together, increasing surface roughness significantly. Therefore, for greater carbon chain lengths such as 10 and 12 carbons, surpassing the limit could result in less effective agglomerations and cause contributions to surface roughness to decrease. An experiment to test this hypothesis could be to modify si lica nanoparticles using a trichlorosilane with an R group of 6 carbon chain length. It would be expected that this trichlorosilane will also yield high contact angles for coated samples Dispersion of Modified Silica into PVA matrix Dispersion of al l three silane coupling agents into PVA resulted in a coated surface with average contact angle greater than 150 As hypothesized, this is possibly a result of PVA forming additional linkages with the hydroxyl and siloxane groups of the modified silica c reating more dangling bonds, and contributing further to the surface roughness It is important to remember that when a coated sample with nonmodified silica and PVA was tested with water, the water droplet spread almost completely over the surface, possib ly because PVA does not form bonds with nonfunctionalized silica but does with functionalized silica. Due to the high solubility PVA has with water, it is not possible to test whether PVA alone can functionalize silica. However, it might be possible to investigate the us e of non water soluble polymers with similar structure to PVA to add functional groups to silica to form superhydrophobic surfaces in place of silane coupling agents It could possibly be a faster and less expensiv e modification. Another reason that modified nanoparticles dispersed in PVA yield superhydrophobic surfaces is because dispersion of modified silica nanoparticles into the PVA matrix could lead to phase separation between the nanoparticles and PVA. This wo uld mean that on a PVA coated sample, most of the silica nanoparticles are lying on top of the surface, and the polymer is below the nanoparticles. Because modified silica nanoparticles contribute fully to increasing surface roughness, a phase separation t hat leads to maximum modified nanoparticles on the surface will also maximize the measured contact angle. Future work that can be done to verify the contact angle measurements that are above 150 are due to increased surface roughness would to be perform surface roughness tests on the samples coated with PVA. It would also be beneficial to find out exactly wh at causes the contact angle to be very high in the PVA coated samples Obtaining more clarity on this matter could lead to the ability to synthesize s uperhydrophobic coatings using novel approaches of using polymers that can create sufficient roughness to obtain superhydrophobic surfaces. Conclusion Goals of this project were to characterize the effects various silane coupling agents used to modify silica nanoparticles dispersed in polymers with different functional groups have on the wettability and measured contact angle of their coated surface. Coupling agents used to modify silica nanoparticles were n octyltrichlorosilane, n decyltrichlorosilane, and dodecyltrichlorosilane. Modified silica nanoparticles were dispersed into the matrices of Lumiflon, SBS, PMMA, and PVA. It was observed that length of carbon chain on trichlorosilanes have no effect on nan oparticles dispersed in Lumiflon but all of the coupling agents allows Lumiflon coated samples to reach superhydrophobicity. Nanoparticles dispersed in SBS and PMMA reached superhydrophobicity when they were modified with n octyltrichlorosilane, but parti cles modified with longer length carbon chains obtained contact angles that were significantly less. All nanoparticles that were dispersed in PVA achieved superhydrophobicity and obtained contact angles that were higher than particles dispersed In Lumiflon This could possibly be because the repeating hydroxyl

PAGE 16

16 groups in PVA provide additional functionalization of silica in addition to the linkages that are provided by the silane coupling agent. From this observation, it is believed that a s a novel approach to making superhydrophobic coatings from silica, adding functional groups to silica through the simple dispersion of silica into capable polymers could be an alternative to the functionalization of silica nanoparticles via silane coupling agents. Referenc es [1] A. Gatenby, "Initiation to Contact Angle", Csc Scientific 2016. [Online]. Available: http://www.cscscientific.com/csc cientific blog/initiation to contact angle. [Accessed: 17 Apr 2018]. [ 2 based superhyd Applied Surface Science 21 Jul 2017. [Online]. Available: https://www.sciencedirect.com/science/article/pii/S0169433217321268. [Accessed: 1 7 Apr 201 8 ]. Chemistry Letters 28 Jul 2016. [Online]. Available: http://www.journal.csj.jp/doi/fpi/10.1246/cl.160621. [Accessed: 17 Apr 2018]. [4] "Superhydrophobic Polymers", Onlinelibrary Wiley 2013. [Online]. Available: https://onlinelibrary.wiley.com/d oi/pdf/10.1002/0471440264.pst594. [Accessed: 17 Apr 2018]. [5] "Understanding Silane Functionalization", Surface Science and Technology 2018. [Online]. Available: http://www.surface.mat.ethz.ch/research/surface functionalization/understanding silane fu nctionalization.html. [Accessed: 17 Apr 2018]. [6] B. Arkles, "Silane Coupling Agents", Gelest 2014. [Online]. Available: https://www.gelest.com/wp content/uploads/Goods PDF brochures couplingagents.pdf. [Accessed: 17 Apr 2018].