Highly moisture-proof polysilsesquioxane coating prepared via facile sol-gel process.

A highly moisture-proof polysilsesquioxane coating was obtained from a new bis-silylated precursor, which was synthesized from 3-aminopropyltriethoxysilane (APTES) and m-xylylene diisocyanate (m-XDI) in tetrahydrofuran (THF) and verified by [.sup.1]H MAS NMR. For direct comparison purposes, an Si[O.sub.2] coating was also prepared by the Stober method using tetraethoxysilane (TEOS) as the reactant. Interestingly, the coating obtained from the polysilsesquioxane sol exhibited a much higher moisture resistance capability than its counterpart, which was attributed to its more compact feature between nanoparticles as characterized by [N.sub.2] absorption experiment and transmission electron microscopy (TEM). Furthermore, its high transparency of about 92% showed potential for application in the protection of optical crystals.

Keywords: Atomic force microscopy, FTIR, ATR, NMR, UV spectroscopy, organosilane, moisture resistance, sol-gel, polysilsesquioxane, silica, bis-silylated precursor

**********

Since Stober developed the sol-gel method more than 30 years ago, (1) remarkable new opportunities have been created to prepare hybrid materials for employment in numerous fields, ranging from microelectronics to healthcare. (2-4) The hybrid materials obtained by the solgel reaction can provide high thermal stability (5) and good mechanical strength, (6,7) such as the large values of the ultimate strength and the rupture energy to the organic polymers, and show high transparency for optical materials (8,9) as well as contact lenses. (10) Recently, there has been a great deal of interest by chemists, physicists, material scientists and integration engineers in preparing polysilsesquioxane through the sol-gel process with the goal of developing low dielectric insulators in microelectronic devices, (11,12) and novel reinforcing elements for hybrid composites. (13-17) Because polysilsesquioxane with an organic group has a low dielectric constant and low moisture absorption compared to silica, it has already been proposed as a replacement for silicon dioxide. (15,18,19) To date, although several kinds of functional films derived from polysilsesquioxane sols have been reported, such as excellent hydrophobic films (15) and photoelectric films, (20,21) few reports on high quality polysilsesquioxane moisture-proof coating have been published.

In this work, we prepared a moisture-proof coating derived from m-xylylene diureidopropyltriethoxysilane (m-XDUPTES). The highly moisture resistance of the coating was clearly evidenced by putting several color-variable silica gel particles coated with polysilsesquioxane sol into water. It was found that these color-variable silica gel particles retained their original color unchanged for more than 100 hr.

EXPERIMENTAL

Synthesis of the New Precursor

The following ingredients were used: 3-aminopropyltriethoxysilane (APTES, 99%, Acros), m-xylylene diisocyanate (m-XDI 99%, TOKYO KASEI), tetraethoxysilane (TEOS, 99%, Acros), N[H.sub.4]OH (Reagent Grade 26-30% N[H.sub.3]), tetrahydrofuran (THF, dehydrated), EtOH (dehydrated), and deionized water.

[GRAPHIC OMITTED]

Two milliliters (0.013 mol) of m-XDI mixed in 13 ml (0.16 mol) of THF was added drop by drop into a typical flask containing 6.04 ml (0.026 mol) of APTES and 15 ml (0.18 mol) of THF under stirring. To ensure the reaction finished smoothly, [N.sub.2] was used as the protective gas to prevent m-XDI from reacting with carbon dioxide (C[O.sub.2]) and water vapor in the air. After m-XDI was added, the obtained liquid was stirred for an additional three hours in an airtight state to react completely. The whole reaction process was carried out at room temperature.

To carry out the NMR analysis, the newly obtained precursor was dissolved in [C.sub.3][D.sub.6]O and placed in a standard 5-mm NMR tube and was conducted with a Bruker DRX 300 NMR using a two-second delay.

Preparation of Polysilsesquioxane Sol and Coating Deposition

A polysilsesquioxane sol was prepared from an m-XDUPTES: EtOH: THF: N[H.sub.3]: [H.sub.2]O molar ratio of 1:26:13:1:3. For direct comparison, Si[O.sub.2] sol was also obtained using TEOS as reactant precursor with the TEOS: EtOH: THF: N[H.sub.3]: [H.sub.2]O molar composition of 1:26:13:0.5:2.

We placed some color-variable silica gel particles into the two different sols, then we removed them and let them dry at room temperature for one hour. In succession, the color-variable silica gel particles were placed in water to study the moisture resistance of the sol coatings, respectively. It was postulated that, if they had excellent moisture resistance, the coated color-variable silica gel particles would keep their color unchanged for a long time. To further characterize the moisture resistance, we dip-coated the polysilsesquioxane sol on nickel sulfate hexahydrated (NSH) crystal, whose transmittance declines rapidly due to its high moisture adsorption, and placed it in a container with fixed relative humidity (60%) for a month to study the change of the crystal's transmittance.

[GRAPHIC OMITTED]

The aged polysilsesquioxane and Si[O.sub.2] sols were used to prepare films on well-cleaned optical glass substrates by spin-coating at a speed of 3000 rpm. The films were dried at room temperature, and their transmittance, hydrophobic ability, and morphology were measured by UV-visible spectrometer (UV 2501PC, Shimadzu, Japan), contact-angle meter (CA-A, Kyowa, Japan), and AFM (Nanoscope III[alpha] Bioscope, Digital Instruments, U.S.), respectively. Furthermore, the polysilsesquioxane sol coated substrate was treated under a 12 W ultraviolet lamp for 72 hr to examine the coating's anti-irradiative ability by observing any change in its color and optical transparency.

Characterization of Xerogels

After being heated in an oven at 333 K for 48 hr, the two sols completely turned into xerogels. They were thoroughly characterized by [N.sub.2] absorption experiment (Tristar 3000, Micromeritics, U.S.) to determine their porosity and surface area. FTIR spectra were recorded on a Nicolet 560 FTIR (U.S.) using a KBr tablet. Their morphology was studied by TEM (Hitachi-600-2, Japan) at a beam voltage of 100 kV.

RESULTS AND DISCUSSION

The new bis-silylated precursor m-XDUPTES was synthesized according to Scheme 1.

In order to avoid the side reactions that can come from combining the extremely active group -NCO with water vapor and C[O.sub.2] in the air and the hydrogen in the group -NHCONH-, two steps had to be taken. In the first step, [N.sub.2] was used as the protective gas to discharge the water vapor and C[O.sub.2] in the container. In the second step, m-XDI/THF was added slowly to ensure that -NCO was fully reacted with the terminal hydrogen at the group -N[H.sub.2] of APTES.

Finally, the obtained system was dried in a vacuum oven at room temperature, and 8.01 g (yield = 99%) of white power was obtained; its [.sup.1]H MAS NMR spectrum is shown in Figure 1. The detailed [.sub.1]H NMR ([C.sub.3][D.sub.6]O) peaks of the sample can be assigned as follows: [delta], 0.6 (4H, 2SiC[H.sub.2]R); 1.3 (18H, 6SiORC[H.sub.3]); 1.6 (4H, 2RC[H.sub.2]R); 3.1 (4H, 2RNHCONHR); 3.8 (12H, 6SiOC[H.sub.2]R); 4.3 (4H, 2RNC[H.sub.2]R); 5.6 (2H, ph-C[H.sub.2]R, above benzene); 5.8 (2H, ph-C[H.sub.2]R, below benzene); 7.2 (4H, R[C.sub.6][H.sub.4]R).

To ensure the full hydrolysis and dissolution of m-XDUPTES, a determined amount of EtOH had to be introduced into the m-XDUPTES/THF solution to increase the system's polarity. THF/EtOH was at the molar ratio of 1/2. The hydrolysis and condensation procedure of m-XDUPTES is illustrated in Scheme 2.

[FIGURE 1 OMITTED]

[FIGURE 2 OMITTED]

[FIGURE 3 OMITTED]

[FIGURE 4 OMITTED]

[FIGURE 5 OMITTED]

[FIGURE 6 OMITTED]

[FIGURE 7 OMITTED]

[FIGURE 8 OMITTED]

Considering that a pure color-variable silica gel particle changes its color rapidly when in contact with water, those color-variable silica gel particles coated, respectively, with polysilsesquioxane sol and Si[O.sub.2] sol were placed in water to study the moisture resistance of the sol coatings. The color-variable silica gel particles coated with Si[O.sub.2] sol only retained their color for several minutes in water. However, the polysilsesquioxane sol showed much better moisture resistance. Under similar conditions, the color-variable silica gel particles coated with polysilsesquioxane sol retained their original color for more than 100 hr.

Figure 2 shows the color change of color-variable silica gel particles coated by polysilsesquioxane sol in water. From the time that these silica gel particles were placed in water, the color did not vary for 100 hr and retained the blue as shown in Figure 2A. One hundred hours later, the color of the silica gel particles gradually changed from blue to purple (Figure 2B), indicating that water molecules had penetrated the polysilsesquioxane sol coating and its moisture resistance was lost.

Figure 3 shows the transmittance spectra of coated and bare NSH crystal treated in a moisture atmosphere. Obviously, the transmittance of the pure water (Figure 3B) surrounded by moisture declined rapidly. Conversely, the transmittance of the coated NSH (Figure 3A) had no obvious change, indicating the present polysilsesquioxane film possessed excellent moisture resistance.

The transmittance spectra of the glass substrates coated with the two different sols is shown in Figure 4; the spectrum of the bare slide was also plotted (blank test). Compared with the Si[O.sub.2] sol film and blank test, although the transmittance for the polysilsesquioxane film was slightly decreased, its value was still beyond 92%, thus indicating that the polysilsesquioxane film had excellent optical property.

Furthermore, the contact angles for water of the polysilsesquioxane film and pure Si[O.sub.2] film were 68[+ or -]2[degrees] and 38 [+ or -] 2, respectively. In comparison with the organically modified Si[O.sub.2] sol film with trimethysilyl coupling agent, (22-25) the present polysilsesquioxane film's hydrophobicity was somewhat limited, which is attributed to the hydrophilic characteristic of -NH- and C=O in the polysilsesquioxane.

The AFM image for the polysilsesquioxane film is displayed in Figure 5; the AFM image of Si[O.sub.2] sol film is also shown for comparison. It is clearly seen from this figure that the surface of the polysilsesquioxane film was much more level than that of the Si[O.sub.2] film. The average roughness of the polysilsesquioxane film was just 0.5 nm, but that of the Si[O.sub.2] sol film was around 1.5 nm. The very slight undulation of polysilsesquioxane film led to the flat surface topography. Conversely, the huge undulation of Si[O.sub.2] sol film contributed to many pits on the surface. Furthermore, it was clearly observed that the particles at the surface of the Si[O.sub.2] sol film were much larger than those of the polysilsesquioxane film.

Figure 6 displays the FTIR spectrum for the polysilsesquioxane xerogel. Clearly, the band at 3330 [cm.sup.-1] was mainly due to the -OH stretching vibration and the -NH- stretching vibration. In the region 2850-2990 [cm.sup.-1], there were three sharp peaks at 2875 [cm.sup.-1], 2930 [cm.sup.-1], and 2985 [cm.sup.-1], which could be ascribed to the symmetric and antisymmetric C-H stretching vibration of methylene (C[H.sub.2]) groups. Moreover, the sharp peak at 1635 [cm.sup.-1] was due to the stretching vibration of carbonyl (C=O) groups. The characteristic absorption of m-xylylene was located at 1580 [cm.sup.-1] and 775 [cm.sup.-1]. The bands at 1085 [cm.sup.-1] and 954 [cm.sup.-1] were due, respectively, to the stretching vibration of Si-O-Si and Si-OH. The peak at 462 [cm.sup.-1] was associated with the bending vibration of Si-O-Si. The occurrence of these peaks was consistent with previously published data. (26,27) The data mentioned above showed that the precursor had hydrolyzed completely and had polycondensed.

Table 1 lists the BET surface area, Langmuir surface area, and pore volume of the different xerogels. Obviously, polysilsesquioxane xerogel was nonporous because of its very low surface area, which is consistent with the research of Loy et al. (28); conversely, the Si[O.sub.2] xerogel had high surface area and high porosity. Taking the data from contact angle for water into consideration, it was therefore deduced that the moisture-proof property of the polysilsesquioxane coating could mainly be attributed to its nonporous character. It was estimated that the BET surface area of Si[O.sub.2] xerogel was 1100 times larger than the pure polysilsesquioxane xerogel, and the Langmuir surface area and pore volume were 900 and 70 times higher, respectively. Clearly the considerable difference in the three aspects could be attributed to the loose cluster of Si[O.sub.2] particles in the hydrolysis and condensation process, which therefore led to the difference in moisture resistance between both sol coatings.

The TEM micrographs of Si[O.sub.2] and polysilsesquioxane xerogels are shown in Figure 7. It is clearly seen from this figure that both xerogels were nanometer-sized. From Figure 7A, we can observe that the Si[O.sub.2] xerogel nanoparticles were homogeneous and approximately 20 nm in size. In Figure 7B, it can be observed that the polysilsesquioxane xerogel nanoparticles clustered intensely. Though we could not exactly differentiate the size of a single polysilsesquioxane xerogel nanoparticle, we could observe that the polysilsesquioxane nanoparticle clusters were approximately 50 nm. The higher clustered degree led to the compact cluster between polysilsesquioxane xerogel nanoparticles, which directly explained the porous structure difference of the two xerogels.

The transmittance of the coating after irradiation is displayed in Figure 8. Clearly, after irradiation, the transmittance of polysilsesquioxane coating declined to some extent, but the transparency was still beyond 91%. Moreover, the color of the coating also did not change. Thus it could be concluded that the unstable aromatic isocyanate in the coating did not lead to the decline of the coating's stability.

CONCLUSION

In summary, a new polysilsesquioxane has been prepared via sol-gel procedure at room temperature. Characterization results based on moisture-proof measurement revealed that the coating obtained from the new precursor had much greater moisture resistance than that of an Si[O.sub.2] coating. The excellent optical property and moisture resistance of the new polysilsesquioxane coating make it possible to be applied widely in the crystal protective field.

ACKNOWLEDGMENTS

Financial support from the National Key Native Science Foundation (20133040) is gratefully acknowledged.

References

(1) Stober, W., Fink, A., and Bohn, E., J. Colloid Interface Sci., 26, 62 (1968).

(2) Orgaz, F. and Rawson, H., J. Non-Cryst. Solids, 82, 57 (1986).

(3) Brinker, C.J. and Scherer, G.W., J. Non-Cryst. Solids, 70, 301 (1985).

(4) Wer, J. and Wilkes, G.L., Chem. Mater., 8, 1667 (1996).

(5) Coltrain, B.K, Ferrar, W.T., Landry, C.J.T., Molaire, T.R., and Zumbulyadis, N., Chem. Mater., 4, 358 (1992).

(6) Mark, J.E. and Sur, G.S., Polym. Bull., 14, 325 (1985).

(7) Sun, C.C. and Mark, J.E., Polymer, 30, 104 (1989).

(8) Davies, B.L., Samoc, M., and Woodruff, M., Chem. Mater., 8, 2586 (1996).

(9) Wung, C.J., Pang, Y., Prasad, P.N., and Karasz, F.E., Polymer, 32, 605 (1991).

(10) Philipp, G. and Schmidt, H., J. Non-Cryst. Solids, 63, 283 (1984).

(11) Lichtenhan, J.D., Vu, N.Q., Carter, J.A., Gilma, J.W., and Feher, F.J., Macromolecules, 26, 2141 (1993).

(12) Zhnag, C. and Laine, R.M., J. Am. Chem. Soc., 122, 6979 (2000).

(13) Kim, K.M., Adiachi, K., and Chujo, Y., Polymer, 43, 1171 (2002).

(14) Wahab, M.A., Kim, I., and Ha, C.S., Polymer, 44, 4705 (2003).

(15) Ameduri, B., Boutevin, B., Moreau, J.J.E., Moutaabbid, H., and Man, M.W.C., J. Fluo. Chem., 104, 185 (2000).

(16) Tsai, M.H. and Whang, W.T., Polymer, 42, 4197 (2001).

(17) Pomogailo, A.D., Russ. Chem. Rev., 69 (1), 53 (2000).

(18) Miller, R.D., Science, 286, 421 (1999).

(19) Nguyen, C.V., Carter, K.R., Hawker, C.I., Hedrick, J.L., Jaffer, R.L., Miller, R.D., Remunar, J.F., Rhee, H.W., Rice, P.M., Toney, M.F., Tollsas, M., and Yoon, D.Y., Chem. Mater., 11, 3080 (1999).

(20) Azinobic, D., Cai, J., Eggs, C., Konig, H., Marsmann, H.C., and Veprek, S., J. Lumin., 97, 40 (2002).

(21) Ossadnik, C., Veprek, S., Marsmann, H.C., and Rikowski, E., Monatsh. Chem., 130, 55 (1999).

(22) Xu, Y., Fan, W.H., Li, Z.H., Wu, D., and Sun, Y.H., Appl. Opt., 42, 108 (2003).

(23) Kytokivi, A. and Haukka, S., J. Phys. Chem., 101, 10365 (1997).

(24) Thomas, I.M., Burnham, A.K., Ertel, J.R., and Frieders, S.C., SPIE, 3492, 220 (1998).

(25) Suratwala, T.I., Hanna, M.L., Miller, E.L., Whitman, P.K., Thomas, I.M., Ehrmann, P.R., Maxwell, R.S., and Burnham, A.K., J. Non-Cryst. Solids, 316, 349 (2003).

(26) Xu, H., Xie, P., and Zhang, R.B., Euro. Polym. J., 37, 2397 (2001).

(27) Ji, X.L., Jiang, S.C., Qiu, X.P., Dong, D.W., Yu, D.H., and Jiang, B.Z., J. Appl. Polym. Sci., 88, 3168 (2003).

(28) Shea, K.J. and Loy, D.A., Chem. Mater., 13, 3306 (2001).

Dongjiang Yang, ([dagger]) Yao Xu, Lei Zhang, Shangru Zhai, Dong Wu, Yuhan Sun -- Chinese Academy of Sciences*

* State Key Laboratory of Coal Conversion, Institute of Coal Chemistry, Chinese Academy of Sciences, Taiyuan 030001, P. R. China.

([dagger]) Graduate School of the Chinese Academy of Sciences. Email address: [e-mail].

Table 1 -- Measured BET Surface Area, Langmuir Surface Area, and Pore
Volume of Polysilsesquioxane Xerogel and Si[O.sub.2] Xerogel

                    BET Surface    Langmuir Surface
                    Area           Area              Pore Volume
Sample              ([m.sup.2]/g)  ([m.sup.2]/g)     ([cm.sup.3]/g)

Polysilsesquioxane    0.2084         0.3315          0.001496
Si[O.sub.2]         231.3319       301.7026          0.101271