Citation

Hang Bian, Xiaozhe Dong, Shanshan Chen, Dewen Dong, Ning Zhang. Polymer brushes on hydrogen-terminated silicon substrates via stable Si-C bond[J]. Chin. Chem. Lett., 2018, 29(1): 171-174  

Hang Bian^a,b, Xiaozhe Dong^b, Shanshan Chen^b, Dewen Dong^b, Ning Zhang^b    

Surface functionalization of silicon and other semiconductors endowing materials with diverse functionalities is of great significance in the development of new semiconductor-based devices, e.g., sensors, smart switches, recognition, optoelectronic and bioactive device [1-17]. Chemical functionalization, which enables the formation of dense and well defined organic monolayers or tethered polymers on silicon substrate, has gained great attention in the past decades [18-25]. Through modern polymerization methods, it is possible to create well-defined grafted polymer chains, i.e., forming polymer brushes by surface-initiated cationic polymerization, surface-initiated anionic polymerization, surface-initiated radical polymerization and surface-initiated coordination polymerizations [26-35]. Surface-initiated polymerizations on self-assembled monolayers (SAMs) has been the most popular approach for the preparation of well-defined polymer brushes on silicon substrates and allows the control of functionality, density and thickness of the polymer brush layer with almost molecular precision. Silane-based molecules are widely utilized to form this SAMs layer by forming Si—O—Si bonds which bridges the monolayer and the silicon substrate [36]. However, Si—O—Si bond is suffered from hydrolysis under acidic or slightly basic conditions [37]. Therefore, more stable bonding manner which is tolerant with complex conditions is highly demanded in view of the wide application fields for silicon-based devices.

Recently, self-initiated photografting and photopolymerization (SIPGP) has become an attractive strategy to form well-defined, homogeneous and even patterned polymer brushes on various surfaces materials without the necessity to anchor any initiator molecules. Until now, a variety of polymer grafts have been successfully created on carbonaceous, oxides, metal and polymer surfaces [38-42].

Herein, we demonstrate a facile and general method to produce homogeneous polymer brushes on hydrogen-terminated Si(100) and Si(100) substrates under UV irradiation. Stable Si—C bonds were formed between silicon surface and polyethylene chain through the UV-induced photopolymerization of a series of vinyl monomers. Different types of monomers such as N-isopropyl acrylamide, 2-isopropenyl-oxazoline, styrene and fluorinated styrene were successfully applied to achieve stable polymer coatings.

The preparation of the polymer grafts are outlined in Scheme 1. First, HF solution was used to remove the native oxidized layer on silicon substrate to give hydrogen-terminated surface (Si—H). Then, polymer brushes were prepared by SIPGP of monomers using UV light with a spectral distribution between 250 nm and 350 nm (λ_max = 300 nm). After 24 h of the photopolymerization using NIPAM as the monomer, the modified Si(100) substrates were rigorously cleaned by ultrasonication in solvents with different polarities to remove physisorbed polymers and monomers.

Scheme 1

	Download: JPG larger image
Scheme 1. Schematic illustration for the preparation of polymer brushes on hydrogen-terminated silicon surfaces.

The successful modification of the hydrogen-terminated Si(100) substrate by PNIPAM brushes was verified by infrared (IR) spectroscopy and X-ray photoelectron spectroscopy (XPS). As shown in Fig. 1, the typical strong bands at 1655 and 1556 cm^-1 originated from the stretching mode amide Ⅰ and amide Ⅱ were observed, indicating the successful formation of PNIPAM. Moreover, the XPS survey scan and high-resolution scan (Fig. S1 in Supporting information) further corroborate the formation of PNIPAM brushes via Si—C bond. AFM measurements show that the silicon substrate was rendered by homogeneous polymer layer with an average thickness of 49 nm. It was found that the surface roughness slightly increased from 1.5 nm (R_a) for bare silicon surface to 4.2 nm (R_a) for PNIPAM modified substrate. In order to examine the influence of the substrate crystalline phase on the photografting polymerization, SIPGP of NIPAM using Si(111) as substrate was performed. After identical period of UV irradiation (24 h), a 54 nm of PNIPAM layer was formed on Si(111) as evidenced by AFM measurements. The slightly thickness difference might be caused by the minor difference of roughness and reactivity of Si(100) and Si(111).

Figure 1

	Download: JPG larger image
Fig. 1. IR spectra for polymer coatings of PNIPAM, PIPOx, PS and PPHMS brushes on silicon substrates.

In order to demonstrate that the universal approach is applicable to other monomers with different chemical structure and functionalities, besides the photopolymerization of NIPAM, the photografting polymerizations employing other vinyl monomers, i.e., styrene (St), 2-isopropenyl-2-oxazoline (IPOx), and 4-(1H, 1H, 2H, 2H-perfluorohexyl)oxymethyl-styrene (PHMS) were performed, resulting in homogeneous and stable poly(styrene) (PS), poly(2-isopropenyl-2-oxazoline) (PIPOx), and poly(4-(1H, 1H, 2H, 2H-perfluorohexyl)oxymethyl-styrene) (PPHMS) brushes. IR analysis (Fig. 1) revealed that these polymer brushes were successfully created on Si(100) surfaces. Ex-situ kinetic studies of the SIPGP of St were conducted on individual hydrogen-terminated silicon substrate at different UV irradiation times (4–30 h) (Fig. 2). In order to determine the thickness of the PS layer, the sample was scratched with a sharp metallic needle to remove the polymer layers locally. The borders of the scratches were investigated by AFM to determine the height difference, i.e., the brush layer thickness. As indicated in Fig. 2, the thickness of the polymer brushes layer as measured by AFM under ambient conditions is plotted as a function of the irradiation time. An almost constant growth rate of 11 nm/h is observed. Actually, the bulk monomer phase became viscous for longer irradiation time due to photopolymerization of monomer in the bulk phase. Here, the reduced monomer mobility caused by the viscosity increase did not result in significant layer thickness growth rate decrease.

Figure 2

	Download: JPG larger image
Fig. 2. Ex-situ kinetics study for SIPGP of styrene after different UV irradiation time. AFM scans of PS brushes after indicated photopolymerization time (a–d). Polymer layer thickness as a function of photopolymerization time (e).

It should be noted that the increase of the thickness of the polymer layers with the irradiation time was specific for each monomer. For instance, after 48 h of irradiation, photografting polymerization of IPOx gave a polymer layer thickness of 50 nm (thickness growth rate of 1.0 nm/h). However, for SIPGP of PPHMS, a layer thickness of 61 nm was reached in 20 h with a polymer layer thickness growth rate of 3.0 nm/h. The different growth rate is caused by the different polymerization velocity of monomers through a free radical polymerization mechanism.

The influence of the polymer composition on the hydrophilic/hydrophobic character of the polymer layer was investigated by contact angle (CA) measurements (Fig. 3). The bare Si substrate gave a static water CA of 96°, while the hydrophilic PIPOx and amphiphilic PNIPAM brushes gave CAs of 34° and 67°. Hydrophobic PS brushes has a CA of 90°, fluorinated PPHMS brushes give a much hydrophobic surface with a CA of 130°.

Figure 3

	Download: JPG larger image
Fig. 3. Surface wettabilities of hydrogen-terminated silicon substrate and the prepared polymer brushes as determined by static water contact angle (a) and the corresponding surface chemical structures (b).

One of the major advantages of the current approach is the chemical stability of polymer brushes grafted via the Si—C bond during photografting. This chemical bond is more chemically stable than siloxane bond-based polymer brushes against acidic or basic conditions [22]. In order to demonstrate the superiority of polymer brushes via our approach, we also prepared self-assembled monolayer (SAM) of 3-(trimethoxysilyl)propyl amine (TMSPA) via C—Si—O bond on oxidized silicon substrate (Fig. 4), which was employed to prepared PNIPAM brushes. The PNIPAM brushes modified substrates prepared through the two approaches were submerged in NaOH (1 mol/L) aqueous solution. After the treatment in NaOH for 1 h, polymer layer thickness slightly decreased from 40 nm to 36 nm for polymer brushes directly created on bare silicon substrate. However, polymer coating with a thickness of 39 nm on SAM disappeared completely after identical treatment. It is obvious that polymer brushes prepared through our approach exhibits superior stability against basic conditions. The corrosion of O—Si—C bond and SiO_x are both possible when subjected PNIPAM brushes on SAMs of TMSPA to NaOH solution. So far, we could not distinguish which manner is predominant. It is very likely that the decomposition of SiO_x and C—Si—O occurs simultaneously. Nevertheless, our approach not only enables the direct grafting via Si—C bond, but also further avoids the presence of less stable SiO_x layer. Therefore, we have provided an efficient strategy in direct creaction of stable polymer grafts on silicon substrate via Si—C bond. The formed stable polymer coatings show promising potential in applications which require high chemical stability in complex conditions.

Figure 4

	Download: JPG larger image
Fig. 4. Chemical stability investigation of PNIPAM brushes on silicon (a) and on silicon with native oxidized layer (b) in NaOH aqueous solution (1 mol/L).

In summary, we have demonstrated a straightforward method to prepare polymer brushes on hydrogen-terminated silicon substrates through the UV-induced photopolymerization via formation of stable Si—C bond. This approach is applicable to a series of monomers, allowing the tailoring of silicon surface with variable hydrophilic/hydrophobic properties. The photografting polymerization show similar grafting efficiency toward Si(100) and Si(111) substrates. Moreover, the as-prepared polymer brushes show superior stability in comparison to polymer brushes obtained via the C—Si—O bond formation. The provided method is promising in the efficient functionalization of silicon-based surfaces.

H. Bian acknowledges the support of Jilin Provincial Department of Education (No. 2014511) and Department of Changchun Science and Technology (No. 14KP023). Department of Science and Technology of Jiangsu Province (No. BK20151189) is also greatly acknowledged.

Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.cclet.2017.05.011.