10912
Langmuir 2006, 22, 10912-10914
Relative Humidity Effects in Dip-Pen Nanolithography of Alkanethiol Mixtures Omkar A. Nafday and Brandon L. Weeks* Texas Tech UniVersity, Department of Chemical Engineering, 6th Street and Canton, Lubbock, Texas 79409 ReceiVed July 19, 2006. In Final Form: NoVember 1, 2006 In the dip-pen nanolithography of a binary alkanethiol mixture of mercaptohexadecanoic acid (MHA) and 1-octadecanethiol (ODT) at a relative humidity (RH) of less than 80%, two distinct phases of MHA and ODT were patterned. However, on ramping up the RH to greater than 80%, only MHA was observed to pattern. This effect was reversible, as shown by the fact that two distinct thiol regions were again patterned on lowering the RH. This segregation could be exploited for generating exclusive MHA (hydrophilic) templates for subsequent architectures from a mixture of alkanethiols driven solely by the RH.
Introduction The dip-pen nanolithography (DPN) method1 is widely used today to control surface architectures on the nanoscale. A variety of inks2-4 have been patterned with DPN, but the most common inks used to demonstrate DPN are n-alkanethiols, either mercaptohexadecanoic acid (MHA, HS(CH2)15CO2H) or octadecanethiol (ODT, CH3(CH2)17SH), patterned on gold substrates because of the thiol self-assembly and the formation of strong gold-thiol bonds. When the inked nanoscale atomic force microscope (AFM) tip (∼20 nm radius) is in contact with the substrate, thiol features are formed by the transport of ink from the tip to the substrate. The factors influencing ink transport and their influence on feature size have been well studied. One of these factors is the relative humidity (RH) of the DPN environment, which affects ink transport as a result of the subtle interplay between the water meniscus formed and the ink molecules.1 Ink transport models to predict dot size as a function of dwell time and to explain the role of the meniscus exist in the literature; however, there is a fair bit of controversy about the meniscus contribution to patterning.5-8 It has been proposed that the ink transport occurs through the bulk water meniscus, but the transport of water-insoluble inks such as ODT cannot be easily accounted for with a meniscus model. Varying ink transport rates were reported for an ink-on-ink9 study (MHA on ODT and ODT on MHA); enhanced transport rates were observed in the case of ODT on MHA whereas MHA on ODT led to reduced ink transport rates. This study also reported that the shape of the preexisting thiol pattern was unaltered when subsequent thiol DPN was conducted. A co-dissolved binary thiol DPN study, specifically DPN using an MHA-ODT thiol mixture, first demonstrated that the thiols phase separate into * Corresponding author. E-mail:
[email protected]. (1) Piner, R.; Zhu, J.; Xu, F.; Hong, S.; Mirkin, C. Science 1999, 283, 661. (2) McKendry, R.; Huck, W. T.; Weeks, B.; Fiorini, M.; Abell, C.; Rayment, T. Nano Lett. 2002, 2, 713. (3) Nyamjav, D.; Ivanisevic, A. Chem. Mater. 2004, 16, 5216. (4) Sheehan, P. E.; Whitman, L. J.; King, W. P.; Nelson, B. A. Appl. Phys. Lett. 2004, 85, 1589. (5) Schwartz, P. V. Langmuir 2002, 18, 4041. (6) Weeks, B. L.; Noy, A.; Miller, A. E.; De Yoreo, J. J. Phys. ReV. Lett. 2002, 88, 255505. (7) Sheehan, P. E.; Whitman, L. J. Phys. ReV. Lett. 2002, 88, 156104. (8) Jang, J.; Hong, S.; Schatz, G. C.; Ratner, M .A. J. Chem. Phys. 2001, 115, 2721. (9) Hampton, J. R.; Dameron, A. A.; Weiss, P. S. J. Am. Chem. Soc. 2006, 128, 1648. (10) Salaita, K.; Amarnath, A.; Maspoch, D.; Higgins, T. B.; Mirkin, C. A. J. Am. Chem. Soc. 2005, 127, 11283.
two distinct regions within the patterned feature.10 Phase separation was also observed in the case of a mixture of polymer films spin coated on DPN-generated MHA templates.11 Other phase separation studies relating to alkanethiol self-assembled monolayers (SAM) also exist in literature.12,13 Surprisingly, none of these studies has examined the role of RH on thiol phase separation although RH is proposed to be one of the principal factors influencing feature characteristics in DPN. In this letter, we report the reversible phase separation of MHA-ODT thiol mixtures driven solely by changes in RH. This demonstration of RH as a control tool in single-tip DPN is important because of its potential application to multi-thiol ink systems. Experimental Section MHA (90% pure) and ODT (98% pure) were obtained from SigmaAldrich (St. Louis, MO). A 2.5 mM MHA-ODT inking solution in ethanol (95%) was prepared by mixing equal parts by weight of MHA and ODT. The mixture was sonicated to ensure uniform mixing. Freshly sputtered gold deposited on freshly cleaved 1 cm2 mica sheets was used as the substrate. DPN patterning was performed using a Pacific Nanotechnology (Santa Clara, CA) Nano-R AFM with an RH control chamber built in-house. To increase the RH, an ultrasonic humidifier was used, and compressed air was circulated inside the chamber to decrease the RH. The RH was measured with a hygrothermometer with accuracies of (3% RH and (1 °C. DPN experiments were performed by cycling the RH from 25 to 88% and back down to 25%. Sufficient time was allowed for each RH value to equilibrate before DPN patterning was started. The AFM was operated in contact mode at room temperature (24 ( 3 °C). Standard “A”-type silicon nitride (Si3N4) AFM cantilevers provided by NanoInk Inc. (Skokie, IL), were used for DPN. The cantilever was inked with the 2.5 mM MHA-ODT solution, and the excess solvent was blown off with a light spray of compressed 1,1-difluoroethane gas. The DPN-generated dots were patterned by allowing the inked cantilever tip to be in contact with the gold for dwell times ranging from 1 to 60 s. All images shown are in lateral force microscopy (LFM) mode to reveal friction force contrast information. The thiol dot radii (outer and inner) were measured using a Matlab routine9 (MathWorks Inc., Natick, MA). (11) Coffey, D.; Ginger, D. S. J. Am. Chem. Soc. 2005, 127, 4564. (12) Stranick, S. J.; Atre, S. V.; Parikh, A. N.; Wood, M. C.; Allara, D. L.; Winograd, W.; Weiss, P. S. Nanotechnology 1996, 7, 438. (13) Kakiuchi, T.; Sato, K.; Iida, M.; Hobara, D.; Imabayashi, S.; Niki, K. Langmuir 2000, 16, 7238. (14) Nafday, O. A.; Vaughn, M. W.; Weeks, B. L. J. Chem. Phys. 2006, 125, 144703.
10.1021/la062110a CCC: $33.50 © 2006 American Chemical Society Published on Web 11/17/2006
Letters
Langmuir, Vol. 22, No. 26, 2006 10913
Figure 1. (a) Phase-separated MHA (inner, light contrast) and ODT (outer, dark contrast) regions observed for dwell times of 1-60 s at RH 31%. (b) RH was then increased to 85% when only MHA ink was patterned. (c) RH was then decreased to 69%, when distinct MHA-ODT areas were patterned again.
Results and Discussion Figure 1 shows the two cases of mixed-thiol DPN observed after tip dwell times from 1 to 60 s. The RH was ramped up from 31 to 87% and back down to 31% in steps of ∼20% RH. Figure 1a shows the representative phase-separated MHA (inner, light contrast) and ODT (outer, dark contrast) regions patterned in each of the dots at RH 31%, which is consistent with a previous study.10 Other work has shown that there can be contrast differences, particularly in the patterning of dots, depending on whether the monolayer formed was standing or prone on the surface.7 However, on increasing the RH to 85%, only MHA dots were patterned (Figure 1b), and little evidence of ODT patterning was observed even after multiple repetitions. The existence of minority populations of ODT on the boundary of MHA areas (Figure 1b) at higher RH is possible, although ODT has been shown to exhibit very poor edge resolution as a result of the “surface protection” effect of the meniscus.5 In some instances, a faint discontinuous dark region was observed around the MHA dots patterned at RH >80%, but it was not reproducible. When the RH was decreased, the two distinct MHA-ODT regions were observed again as shown in Figure 1c. This reversibility in the patterning nature of the binary thiol mixture was reproducibly observed and was driven exclusively by the changes in RH because other factors including tip and substrate conditions were the same.
Figure 2. (a) Average MHA and ODT dot areas over all RH ramping up and down cycles plotted against RH after 40 s of tip dwell time. (b) Average MHA dot area relative to the total dot area after 40 s dwell time, plotted for all RH. Both a and b indicate that MHA is almost exclusively patterned at RH 84%. This phenomenon was observed reproducibly for all dwell times (1-60 s); however, only the 40 s dwell time dot data is shown for clarity.
Control experiments were performed to determine whether the contrast changes in the patterned dots could be attributed to thiol phase separation or to an artifact produced by localized humidity. Dots (1-60 s dwell times) were patterned at high RH, and the features were imaged at low RH. Similar experiments were performed at low RH and showed the two distinct regions as expected. However, in both cases no changes in dot size and contrast were observed, which demonstrated that the inner region observed at high/low RH is indeed MHA/MHA-ODT. Figure 2 shows the graphs of MHA, ODT, and total dot areas plotted against RH after 40 s of dwell time, together with the % MHA coverage at various RH values. Experiments at each (160 s) dwell time were performed twice in a single RH ramping cycle, and four such ramping (up and down) cycles were conducted in a typical run. Figure 2a shows the average areas of the ODT and MHA dots. The ODT dot area was determined by subtracting the inner MHA area from the total area. The MHA and ODT dot areas are relatively unchanged as a function of RH up to 80%; however, above 80%, only MHA is patterned to fill the entire dot. Figure 2b shows the average % MHA dot area with respect to the total dot area after a tip dwell time of 40 s. No clear trend was observed in the % MHA coverage with respect to RH below 80%. However, at RH 84% only MHA
10914 Langmuir, Vol. 22, No. 26, 2006
could be observed to pattern in every experiment corresponding to an ∼100% coverage area. Although we cannot verify that the features are exclusively MHA, the patterning was reproducible above 80% RH over all of the contact times. The lowest-energy state for ODT has been postulated to remain at the air-water interface in DPN.5 This helps to explain the observation of outer regions of ODT because the MHA will travel through the bulk meniscus, filling the inner region, and the ODT will transport at the air-water interface, patterning at the outer area of a dot. Other studies have verified that inks can transport at the air-water interface, producing hollow features.5,14 At low humidity, thiol phase separation is postulated to be driven by enthalpic energy and hydrogen bonding interactions.10 However, at high humidity (>80%) we propose that water covers the gold surface and a developed water meniscus exists at the tip-gold junction. Gold is a hydrophilic surface (contact angle ) 70°),15 and a 60-nm-thick water film has been observed to cover the entire surface at RH 80%.16 When MHA-ODT thiols coexist, the thiol-thiol interactions become important.9 However, in the presence of a developed water meniscus observed at high RH,17 the thiol-meniscus interactions become dominant. The transition to patterning primarily MHA at ∼80% RH is also
Letters
consistent with previous studies where a transition in the patterning rate was observed at high humidity, suggesting that a fully developed meniscus was formed.18 Because MHA is more watersoluble than ODT, only the MHA ink is deposited, and ODT deposition has been shown to be inhibited at high humidity.5,19
Conclusions We have demonstrated that a binary MHA-ODT thiol mixture phase separates to pattern predominantly MHA at RH > 80%. Our data also shows patterning reversibility on lowering the RH to
