Atomic Layer Deposition as Pore Diameter Adjustment Tool for

to load: https://cdn.mathjax.org/mathjax/contrib/a11y/accessibility-menu.js .... Atomic Layer Deposition as Pore Diameter Adjustment Tool for Nano...
0 downloads 0 Views 688KB Size
Langmuir 2008, 24, 4473-4477

4473

Atomic Layer Deposition as Pore Diameter Adjustment Tool for Nanoporous Aluminum Oxide Injection Molding Masks Ville Miikkulainen, Tiina Rasilainen, Esa Puukilainen, Mika Suvanto, and Tapani A. Pakkanen* Department of Chemistry, UniVersity of Joensuu, P.O. Box 111, FI-80101 Joensuu, Finland ReceiVed January 29, 2008. In Final Form: March 10, 2008 The wetting properties of polypropylene (PP) surfaces were modified by adjusting the dimensions of the surface nanostructure. The nanostructures were generated by injection molding with nanoporous anodized aluminum oxide (AAO) as the mold insert. Atomic layer deposition (ALD) of molybdenum nitride film was used to control the pore diameters of the AAO inserts. The original 50-nm pore diameter of AAO was adjusted by depositing films of thickness 5, 10, and 15 nm on AAO. Bis(tert-butylimido)-bis(dimethylamido)molybdenum and ammonia were used as precursors in deposition. The resulting pore diameters in the nitride-coated AAO inserts were 40, 30, and 20 nm, respectively. Injection molding of PP was conducted with the coated inserts, as well as with the non-coated insert. Besides the pore diameter, the injection mold temperature was varied with temperatures of 50, 70, and 90 °C tested. Water contact angles of PP casts were measured and compared with theoretical contact angles calculated from Wenzel and CassieBaxter theories. The highest contact angle, 140°, was observed for PP molded with the AAO mold insert with 30-nm pore diameter. The Cassie-Baxter theory showed better fit than the Wenzel theory to the experimental values. With the optimal AAO mask, the nanofeatures in the molded PP pieces were 100 nm high. In explanation of this finding, it is suggested that some sticking and stretching of the nanofeatures occurs during the molding. Increase in the mold temperature increased the contact angle.

Introduction Polypropylene (PP) is a versatile polymer material, applied, for example, in fibers, packing materials, and car parts.1 The surface wettability is an important parameter in these applications. PP is a slightly hydrophobic polymer, and its contact angle with water is 103°.2 When the wettability of the PP surface was tailored through nanostructuring via a solvent-assisted method, the contact angle increased to 160°.3 A faster and more productive method for thermoplastic replication than the solvent method is microinjection molding (µ-IM).4 In µ-IM, the molten plastic is injected into a cool mold cavity, where the plastic solidifies into the form defined by the mold. The mold temperature has been identified as a key parameter in µ-IM, affecting the filling of the cavities.4 Aspect ratios (feature height divided by its width) over 2 are reported as difficult to produce by µ-IM.4-6 PP, as well as polyethylene and poly(vinylchloride), have been nanostructured by µ-IM with anodized aluminum oxide (AAO) masks as mold inserts.2,7,8 AAO is prepared via electrochemical anodization of aluminum in aqueous solution of polyprotic acid. The pores of the obtained nanoporous aluminum oxide template are aligned perpendicular to the substrate surface and parallel to each other * Corresponding author. Tel.: +35813 2513345; fax: +35813 2513344; e-mail: [email protected]. (1) Oertel, C. G.; Zwygers, T.; Sgarzi, P. In Polypropylene Handbook, 2nd ed.; Pasquini, N., Ed.; Carl Hanser Verlag: Munich, Germany, 2005; Chapter 10.4. (2) Puukilainen, E.; Koponen, H.-K.; Xiao, Z.; Suvanto, S.; Pakkanen, T. A. Colloids Surf., A: Physicochem. Eng. Aspects 2006, 287, 175-181. (3) Erbil, H. Y.; Demirel, A. L.; Avci, Y.; Mert, O. Science 2003, 299, 13771380. (4) Giboz, J.; Copponex, T.; Me´le´, P. J. Micromech. Microeng. 2007, 17, R96-R109. (5) Heckele, H.; Schomburg, W. K. J. Micromech. Microeng. 2004, 14, R1R14. (6) Mo¨nkko¨nen, K.; Hietala, J.; Pa¨a¨kko¨nen, P.; Pa¨a¨kko¨nen, E. J.; Kaikuranta, T.; Pakkanen, T. T.; Ja¨a¨skela¨inen, T. Polym. Eng. Sci. 2002, 42, 1600-1608. (7) Koponen, H.-K.; Saarikoski, I.; Korhonen, T.; Pa¨a¨kko¨, M.; Kuisma, R.; Pakkanen, T. T.; Suvanto, M.; Pakkanen, T. A. Appl. Surf. Sci. 2007, 253, 52085213. (8) Puukilainen, E.; Rasilainen, T.; Suvanto, M.; Pakkanen, T. A. Langmuir 2007, 23, 7263-7268.

in hexagonal symmetry. Pore dimensions can be modified through change in anodization potential and the electrolyte. The pore diameters and interpore distances are dependent on each other, however, and cannot be altered independently.9 It is a challenging task to deposit a conformal coating on an AAO substrate with deep, nanometer-scale pores. One of the methods capable of doing this is atomic layer deposition (ALD). ALD is a chemical gas-phase thin film deposition method, in which the film is deposited by feeding two precursors in separate pulses onto the substrate, and between the pulses the reactor is purged with inert gas.10 During the pulses, the precursors react with the surface in a self-limiting manner. The intermediate purging prevents the gas-phase reactions. The pulses are repeated in cycles, and the film grows layer-by-layer, which means that the film thickness can be controlled at the sub-nanometer level by manipulating the number of deposition cycles. In addition to being of controlled thickness, ALD films are highly conformal and uniform in thickness.11 The above-mentioned features of ALD make it a powerful tool for depositing films on substrates such as AAO with deep nanoscale textures.12 ALD has been applied in chemical modification of AAO pores12-14 as well as in nanotube fabrication of several materials,15-19 including molybdenum nitride,20 by template method. (9) Nielsch, K.; Choi, J.; Schwirn, K.; Wehrspohn, R. B.; Go¨sele, U. Nano Lett. 2002, 2, 677-680. (10) Suntola, T. Mater. Sci. Rep. 1989, 4, 261-312. (11) Ritala, M.; Leskela¨, M. In Handbook of Thin Film Materials; Nalwa, H. S., Ed.; Academic Press: San Diego, CA, 2001; Vol. 1, pp 103-159. (12) Leskela¨, M.; Kemell, M.; Kukli, K.; Pore, V.; Santala, E.; Ritala, M.; Lu, J. Mater. Sci. Eng. C 2007, 27, 1504-1508. (13) Chen, P.; Mitsui, T.; Farmer, D. B.; Golovchenko, J.; Gordon, R. G.; Branton, D. Nano Lett. 2004, 7, 1333-1337. (14) Ott, A. W.; Klaus, J. W.; Johnson, J. M.; George, S. M.; McCarley, K. C.; Way, J. D. Chem. Mater. 1997, 9, 707-714. (15) Pellin, M. J.; Stair, P. C.; Xiong, G.; Elam, J. W.; Birrell, J.; Curtiss, L.; George, S. M.; Han, C. Y.; Iton, L.; Kung, H.; Kung, M.; Wang, H.-H. Catal. Lett. 2005, 102, 127-130. (16) Shin, H.; Jeong, D.-K.; Lee, J.; Sung, M. M.; Kim, J. AdV. Mater. 2004, 16, 1197-1200. (17) Bachmann, J.; Jing, J.; Knez, M.; Barth, S.; Shen, H.; Mathur, S.; Go¨sele, U.; Nielsch, K. J. Am. Chem. Soc. 2007, 129, 9554-9555.

10.1021/la800285s CCC: $40.75 © 2008 American Chemical Society Published on Web 03/21/2008

4474 Langmuir, Vol. 24, No. 9, 2008

Letters

Transition metal nitride films, deposited by chemical vapor deposition (CVD) and physical vapor deposition (PVD), provide effective wear-resistant coatings on injection molds.21-25 Neither method is applicable for AAO substrates, however, because of the low film conformality. Transition metal nitrides exhibit beneficial properties in conventional mold applications, and molybdenum nitride is a good candidate for pore diameter adjustment of AAO masks for µ-IM as well. In our earlier use of ALD in the fabrication of molybdenum nitride nanotubes,20 molybdenum nitride was deposited on AAO substrates from an imido-amido precursor.26 Unlike the more generally used nitride precursors, i.e., transition metal halides,27-31 imido-amido precursors do not emit corrosive hydrogen halides as a side product. They could therefore be considered preferred precursors for depositions on AAO inserts. From the theoretical point of view, the wettability of a solid surface is a function of the surface energy of the material and the surface roughness. The surface energy is a material-dependent factor that can be adjusted by chemical modification of the material. For its part, surface roughness can be formulated by the theories of Wenzel32 or Cassie and Baxter.33 According to Wenzel,32 a rough surface is completely wetted, and the apparent contact angle θ* is

cos θ* ) r cos θ

(1)

where θ is Young’s contact angle for a flat surface, and r is the geometrical area divided by the projected area. Cassie and Baxter,33 in turn, suggest that the liquid lies on top of the surface structures, forming a composite surface where air is trapped between the structures. The apparent contact angle θ* according to Cassie and Baxter is formulated as

cos θ* ) φS(1 + cos θ) - 1

(2)

where φS is the area fraction of the solid surface. Injection molding is a productive method for plastic replication. Injection molding with nanoporous AAO generates nanostructure on the surface of molded PP, and it should be possible, with ALD, to adjust the pore diameter of the AAO insert, and further the dimensions of the nanostructure and the surface wettability of the molded PP. In this study, we tuned the pore diameters of an AAO insert by depositing molybdenum nitride films of different thicknesses by ALD. Injection molding of PP was carried out with the obtained AAO inserts of different pore diameters. As noted above, the mold temperature is an important parameter (18) Yang, C.-J.; Wang, S.-M.; Liang, S.-W.; Chang, Y.-H.; Chen, C.; Shieh, J.-M. Appl. Phys. Lett. 2007, 90, 033104. (19) Kemell, M.; Pore, V.; Tupala, J.; Ritala, M.; Leskela¨, M. Chem. Mater. 2007, 19, 1816-1820. (20) Miikkulainen, V.; Suvanto, M.; Pakkanen, T. A. Thin Solid Films, in press; DOI: 10.1016/j.tsf.2007.10.124. (21) Dearnley, P. A. Wear 1999, 225-229, 1109-1113. (22) Cunha, L.; Andritschky, M.; Pischow, K.; Wang, Z.; Zarychta, A.; Miranda, A. S.; Cunha, A. M. Surf. Coat. Technol. 2002, 153, 160-165. (23) Zhang, W. H.; Hsieh, J. H. Surf. Coat. Technol. 2000, 130, 240-247. (24) Heinze, M. Surf. Coat. Technol. 1998, 105, 38-44. (25) Bienk, E. J.; Mikkelsen, N. J. Wear 1997, 207, 6-9. (26) Miikkulainen, V.; Suvanto, M.; Pakkanen, T. A. Chem. Mater. 2007, 19, 263-269. (27) Ritala, M.; Leskela¨, M.; Rauhala, E.; Haussalo, P. J. Electrochem. Soc. 1995, 142, 2731-2737. (28) Ritala, M.; Leskela¨, M.; Rauhala, E.; Jokinen, J. J. Electrochem. Soc. 1998, 145, 2914-2920. (29) Hiltunen, L.; Leskela¨, M.; Ma¨kela¨, M.; Niinisto¨, L.; Nyka¨nen, E.; Soininen, P. Thin Solid Films 1988, 166, 149-154. (30) Ale´n, P.; Ritala, M.; Arstila, K.; Keinonen, J.; Leskela¨, M. J. Electrochem. Soc. 2005, 152, G361-G366. (31) Juppo, M.; Ritala, M.; Leskela¨, M. J. Electrochem. Soc. 2000, 147, 33773381. (32) Wenzel, R. N. Ind. Eng. Chem. 1936, 28, 988-994. (33) Cassie, A. B. D.; Baxter, S. Trans. Faraday Soc. 1944, 40, 546-551.

in µ-IM, and the effect of mold temperature on the PP surface wettability was studied as well. The contact angles were measured for molded PP pieces and compared with the theoretical contact angles calculated assuming the Wenzel and Cassie-Baxter theories. Experimental Section The AAO masks were fabricated by applying a 40 V anodization voltage in 0.3 M oxalic acid for 24 h. A detailed description of the anodization process can be found elsewhere.2 Use of these parameters gives rise to an 80-µm-thick template with hexagonally arranged pores 50 nm in diameter and 100-120 nm interpore distance. The molybdenum nitride films were deposited by the ALD method with a flow-type reactor modified from the ASM Microchemistry F-120 reactor. Reactor schematics and a detailed description of the reactor setup have been reported previously.26 The precursors were bis(tert-butylimido)-bis(dimethylamido)molybdenum and ammonia, and the deposition temperature was 280 °C. The precursor evaporation temperature was 60 °C, and the pulse and purge times were 7 s. The growth parameters were set at the values found optimal in our previous study on ALD of molybdenum nitride on AAO substrate.20 In the earlier work, we found that the molybdenum nitride film was deposited inside the AAO pores to 20 µm depth, and the film growth per deposition cycle was 0.5 Å. Three different cycle numberss 100, 200, and 300swere applied, resulting in molybdenum nitride thicknesses of 5, 10, and 15 nm, respectively. The corresponding pore diameters for the nitride-coated AAO masks were 40, 30, and 20 nm. In addition, an uncoated AAO mask with pore diameter of 50 nm was used in injection moldings. The mold inserts were studied by scanning electron microscopy (SEM) (Hitachi S-4800) before and after the moldings. The injection molding of PP was done with a DSM Midi 2000 extruder-injection molding machine. On the basis of previous studies2,8 the extruder screw temperature was set at 255 °C, the pressure of the injection piston was set at 4.5 bar, and the screw rotation speed was set at 80 rpm. Mold temperatures of 50 °C, 70 °C, and 90 °C were applied. Mold temperature is a major parameter affecting the filling of micro- and nanofeatures in injection molding4 and, in the case of finer textures, higher mold temperature is required for satisfactory filling of the mold cavities. Puukilainen and coworkers2,8 applied 50 °C mold temperature for PP injection molding of nanofeatures, but higher temperatures are reasonable in the present study where the molded features are narrower. For each mold temperature and AAO pore diameter, nine circular pieces of PP casts, 25 mm in diameter and 1 mm in thickness, were molded. The molded PP casts were analyzed by SEM from the top and at the cross section. Contact angle measurements (KSV Cam 200) were made for all of the molded PP pieces. A 5-µL drop of distilled water was dropped on the sample at each of the three measurement points, and the drop-sample interface was photographed once per second for 30 s. The Young-Laplace function was fitted around the drop for determination of the contact angles.

Results and Discussion Injection mold inserts were studied by SEM before and after the injection molding. SEM micrographs of AAO templates with 15-nm, 10-nm, and 5-nm molybdenum nitride coating, as well as without coating, are presented in Figure 1. As can be seen, the pore diameters a are 20 ( 2, 31 ( 3, 40 ( 2, and 50 ( 4 nm, respectively. The values are averages and standard deviations of 10 measurements. The pore openings in the coated AAO masks are rounded (Figure 2a) because of the nature of the coating method. It can also be seen that the plastic tends to adhere to the mold insert, especially that with 30-nm pores (Figure 2b). Nanostructured PP casts were studied by SEM both in top view and in cross section. Figure 3 shows the micrographs of plastic casts molded at 70 °C mold temperature. The contact angle was higher with 70 °C than with 50 °C mold temperature,

Letters

Langmuir, Vol. 24, No. 9, 2008 4475

Figure 2. SEM micrographs of an AAO insert with 10-nm MoN coating (30-nm pore diameter): (a) cross section before the injection molding, (b) top view after the injection molding.

Figure 1. SEM micrographs of mold inserts with different pore diameters after the injection moldings.

but no notable further increase was found at mold temperature of 90 °C. The contact angles are discussed in more detail below. Measured by SEM, the highest nanofeatures, of about 100 nm, were achieved with the 30-nm pores. The nanopillars were only about 20 nm high with the 20-nm pores and about 40 nm with the 40-nm and 50-nm pores. It is widely reported in the literature that the higher the aspect ratio of a nanofeature, the harder is the replication by injection

molding.4-6 Thus, if the aspect ratio of the molded features is constant with fixed injection molding parameters, a narrower molded feature should be lower in height. This would explain the low filling of the 20-nm pores. Also, other factors must be affecting the feature height, however, since the nanofeatures in the plastic cast were much lower for the 40-nm and 50-nm pores than for the 30-nm pores. One of these other factors may be that the molten plastic adheres slightly to the mold insert, causing stretching of the features when the plastic cast is removed from the mold. As noted, aspect ratios over 2 are difficult to produce by µ-IM.4-6 In the present study, the aspect ratio was over 3 for features 100 nm high and 30 nm wide, which suggests that stretching of the features has occurred. The adhering plastic visible in Figure 2b probably originated from the same cause, since the amount of adhered plastic on the other inserts was lower. For PP casts molded with the other pore diameters, the aspect ratio of the nanopillars was about 1, which lies in the range of the aspect ratios reported for µ-IM. The combined effect of poor replication of high aspect ratio and stretching appears to produce the high nanofeatures with the 30-nm pores. Both the pores in the AAO masks and the nanopillars in the molded PP casts are arranged in hexagonal symmetry, as schematically illustrated in Figure 4 and as can be seen in the SEM micrographs in Figures 1 and 3. In the case of hexagonal symmetry, the area fraction of the solid surface, φS, for the CassieBaxter formula can be formulated as

φS )

πa2 2x3(a + b)2

(3)

4476 Langmuir, Vol. 24, No. 9, 2008

Letters

Figure 5. Experimental and theoretical contact angles, with standard deviations, for PP casts. Experimental values were for casts prepared at mold temperatures of 50 °C, 70 °C, and 90 °C. The horizontal line shows the contact angle of the flat PP surface, 103°.2

Figure 3. SEM micrographs of PP pieces molded with different mold inserts (pore diameters 20, 30, 40, and 50 nm) and 70 °C mold temperature; top views and cross sections.

Figure 4. Hexagonal arrangement of the features, feature height H.

In the Wenzel approach, in turn, the surface roughness term r for hexagonally arranged features of height H can be written out as

r)

x3(a + b)2 + 2πaH x3(a + b)2

(4)

The theoretical contact angles according to Wenzel and CassieBaxter theories can be calculated with eqs 1 and 4 and 2 and 3, respectively. The theoretical and experimental contact angles for the molded PP parts are presented in Figure 5. Experimental values were measured for injection mold temperatures of 50 °C, 70 °C, and 90 °C. The error bars indicate the standard deviations. Feature diameters of 20, 30, 40, and 50 nm were used in the CassieBaxter approach, while the aforementioned diameters and corresponding feature heights of 20, 100, 40, and 40 nm were used in the Wenzel approach. (The height values were measured from the SEM cross sections in Figure 3.) The interpore distance used in calculations was 110 nm.

The highest contact angles, 139-141°, were measured for the PP casts molded with the 30-nm pore diameter AAO inserts (Figure 5). The contact angle of PP molded with a non-coated AAO mask is reported to be 128-136°.2,8 For the PP casts molded with 40-nm pores, and further for the ones molded with 50-nm pores, the contact angle decreases, as does the calculated CassieBaxter value. For the PP pieces molded with 20-nm pores, in turn, the experimental contact angles are much lower than the theoretical Cassie-Baxter angle and close to the Wenzel value. In discussing the cross section micrographs we noted that, as expected, nanofeatures were of lower height for PP molded with the 20-nm pore inserts. The contact angle data support that expectation: the pore filling and resulting nanofeature height is insufficient for the formation of a composite surface. As depicted in Figure 3, the nanofeatures replicated with 20-nm pores are low, and in some places on the surface, the nanopillars are actually missing. However, the nanopillars molded with the 30-nm pore insert are not perpendicular to the surface, as can be seen in Figure 3. Therefore the theoretical approach does not effectively describe the real plastic surface. In addition, the deviations of the experimental contact angles are relatively large, which could question the trends suggested above. Nevertheless, it is evident that the pore diameter of the AAO template, and further the nanofeature dimensions of the molded polymer, can be accurately adjusted with ALD of molybdenum nitride. For the 30-, 40-, and 50-nm features, the experimental contact angle follows the trend expected from the Cassie-Baxter theory, but the values are lower. The lower values probably reflect the nonideality of the molded surfaces (see Figure 3): the molded nanopillars are not cylindrical, and the height of the features tends to vary, contrary to the ideal geometry assumed in the theoretical approach. In general, the Cassie-Baxter theory explains the experimental values better than the Wenzel approach, suggesting that a composite surface is formed between the water droplet and the PP nanostructure. The contact angle increases for 20-, 40-, and 50-nm pores as a function of mold temperature, as would be expected from findings reported in the literature.4 The mold temperature does not seem to influence the contact angle in the case of the 30-nm pores, providing further support for the proposed stretching effect.

Conclusions ALD of molybdenum nitride onto AAO was studied as a method to fine-tune the AAO template pore diameter. Injection

Letters

molding of PP was conducted with templates of four different pore diameters. For the three largest pores, the contact angle of molded PP casts increased as the pore diameter decreased. The contact angle was lowest for PP molded with the narrowest pores, most probably because of insufficient filling of the template structures during the injection molding. The highest contact angle measured for injection molded, nanostructured PP was 140°.

Langmuir, Vol. 24, No. 9, 2008 4477

This was obtained with an AAO mask with 30 nm pores. The deviations of the contact angles were, however, relatively large. The pore diameter in the AAO templates can be accurately adjusted by coating the templates with molybdenum nitride by ALD. Further, fine-tuning of the template pore diameter allows tailoring of the wettability of the polymer cast. LA800285S