Self-Assembled Monolayer-Assisted Negative Lithography - Langmuir

Feb 23, 2015 - Self-assembled monolayers (SAMs) have been widely employed as etching resists in wet lithography systems to form patterns in which the ...
0 downloads 9 Views 7MB Size
Download PDF
Article pubs.acs.org/Langmuir

Self-Assembled Monolayer-Assisted Negative Lithography Xiaoyan Mu, Aiting Gao, Dehui Wang, and Peng Yang* Key Laboratory of Applied Surface and Colloid Chemistry, Ministry of Education, College of Chemistry and Chemical Engineering, Shaanxi Normal University, Xi’an 710062, China S Supporting Information *

ABSTRACT: Self-assembled monolayers (SAMs) have been widely employed as etching resists in wet lithography systems to form patterns in which the ordered molecular packing of the SAM regions significantly delays the etchant attack. A generally accepted recognition is that the SAMs ability to resist etching is positively correlated to the quality of the surface-assembled structures, and a more ordered molecular packing would correspond to a better etching resistance. Such a classical belief is debated in the present work by providing an alternative SAM-assisted negative lithography where ordered SAM regions are etched more quickly than their disordered counterparts. This method features a unique photoirradiation-imprinted patterning process that simply consists of two steps: (1) UV irradiation on an OH-terminated SAM-modified gold surface through a photomask and (2) the subsequent immersion of the exposed substrate in an aqueous etching solution of N-bromosuccinimide/pyridine to develop a wet lithographic pattern. The entire experimental process reveals a finding from previous work that the etching rate on the UV-exposed regions with disordered molecular packing could be modulated to be slower than that in the unexposed well-defined SAM regions. Longer irradiation times would also revert the patterns from negative to positive. Thus, by merely using one kind of SAM-modified surface to provide both positive and negative micropatterns on gold layers, one could obtain flexible opportunities for high-resolution micro/nanofabrication resembling photolithography.

1. INTRODUCTION Self-assembled monolayers (SAMs) of alkanethiolates are widely used in the surface modification of metals.1 Specifically, the use of thiols with long-chain hydrocarbons provides a densely packed molecular layer on gold surfaces and could therefore slow down wet/dry etching.2−13 This resist has been utilized to create micro/nanostructures on metal substrates through the transcription of SAM patterns to underlying substrates. In principle, a surface without a SAM covering or with disordered SAMs would be more preferentially etched than an area with well-defined SAMs. As a result, the etching proceeds in a positive “copying” process, and a discrete lithographic pattern that is an exact copy of the SAM pattern is created (Scheme 1A). With the process described above, it is difficult to obtain other kinds of structures, particularly negative-tone patterns complementary to the SAM pattern (Scheme 1B), because the etching on SAM-covered areas is always greatly retarded. Therefore, motivated by the great design flexibility in photolithography systems where positive- and negative-tone patterns can be conveniently switched by using corresponding types of photoresists,14 the development of a flexible SAMbased strategy that could be fitted into negative-tone patterns would be interesting for both academic research and industrial applications.15−23 With this aim, Zharnikov and Grunze have developed SAMs that serve as either a positive resist (alkanethiolates with aliphatic backbone) or a negative resist (phenylthiolates with © 2015 American Chemical Society

Scheme 1. Schematic of the Proposed SAM-Templated Positive and Negative Lithography

aromatic backbone).9,19−23 Delamarche et al. have initiated research on positive- and negative-tone microcontact printing (μCP) through the combinational use of two kinds of SAMs possessing different etch-resist abilities.15 Nonetheless, the chemical design in previous approaches15−23 still shares the classical recognition that certain SAMs would produce physical Received: November 19, 2014 Revised: January 16, 2015 Published: February 23, 2015 2922

DOI: 10.1021/la504516e Langmuir 2015, 31, 2922−2930

Article

Langmuir

Figure 1. XPS (a) and TOF−SIMS (b) spectra on the photoexposed SAM−OH surfaces with varying irradiation time. The peaks at low m/z labeled with “*” in panel b were assigned to molecular fragments and recombinations from the pristine SAM−OH and photooxidized SAM−OH, which included typical structures such as CH, O, C2H, C2H2, CH3OH, O2H3, CH3CH2C, CH3CH2C, CH3CH2CH, and CH3COO.

2. RESULTS AND DISCUSSION The present study is based on the use of a special redox microenvironment recently developed by our group.27 Previous research has shown that combining the use of N-bromosuccinmide (NBS) and pyridine (py) in aqueous solution led to an efficient etching of bare gold substrates, while the etching behavior on SAM-protected gold surfaces remains unclear.27 In the present work, the alkanethiolate SAMs with a hydroxyl group as the terminating functionality (HS−C11H22−OH, SAM−OH) were first grown on gold surfaces according to well-established protocols, and their quality was further verified by X-ray photoelectron spectroscopy (XPS, Figure S1a) and water contact angle measurements (WCA, Figure S2). According to XPS, the experimental elemental ratio of C/O on the SAM−OH surface was close to the reported value;28 moreover, the WCA on the SAM−OH surface was in agreement with values reported in the literature.29,30 This evidence indicated that the resultant SAM−OH on gold had an experimental quality with ordered surface constructions. To introduce lithographic patterns, such SAM−OH surfaces were further primed by UV photoxidation to induce disorder. Specifically, a freshly prepared gold surface covered with SAM− OH was exposed to UV irradiation (the UV intensity at 254 nm was 10 mw/cm2) for a determined time. The exposure to UV mainly led to the oxidation of Au−S bonds in the SAMs,31 accompanied by the formation of fragmented molecules, vinyl bonds, oxidized groups, and the cross-linking of adjacent molecules.1,32,33 The resultant weakly bound sulfonate species would be easily detached from the surface during washing, and a well-defined SAM before UV exposure would change to a disordered SAM after UV exposure.31−33 By simply controlling the irradiation time, this extent of disorder could be conveniently tailored,32,33 showing enhanced oxidation when the irradiation duration was prolonged. The chemical changes on the Au−S bonds of SAM−OH were first monitored by XPS and time of flight−secondary ion mass spectroscopy (TOF−SIMS) (Figure 1). With XPS (Figure 1a, Figure S1b−e), it was observed that in addition to the original peak at 162.0 eV for S−II (Au−S bond) and

barriers against etchants and the ability of the resultant resist would be positively correlated to the structure of its selfassembled structures. In practice, a higher etching rate is conventionally observed on the disordered regions rather than on the well-defined SAM surface.2,3,6−8 Very recently, studies on SAM-mediated wet lithography of metal substrates have moved toward a novel method to understand and control the etching process,24,25 as well as ultrafine photocatalytic shaving of SAMs on the nanometer scale.26 Herein we propose a distinctive chemical concept to reveal a novel phenomenon that, under a certain chemical wet etching bath, the UV-exposed disordered SAM area (which is supposed to demonstrate a poor etching resistance) exceptionally induces a slower etching rate for gold film than its unexposed ordered counterpart. We thus successfully obtained negative-tone gold patterning by the alkanethiolate SAM-based strategy without the assistance of μCP and an aromatic backbone as described in previous reports15−23 (Scheme 1B). Interestingly, it was found that the pattern tone was related to the extent of disorder of the SAMs. As a low extent of disorder induced by short UV irradiation led to the negative pattern, a high extent of disorder induced by a long UV irradiation time reverted the pattern tone from negative to positive. A benefit of this negative-tone pattern is the possible exclusion of SAM defect-induced etch errors on protected regions observed in conventional positive-tone patterns. This is due to the SAM area in the present method not being intentionally used as a resist layer but fitted to an etching mediation layer so that the etchant penetration in dedicated SAMs is even. The etching resolution and surface topography could be theoretically maintained without the disturbance from the etching error in this negative-tone patterning. The process without the assistance of μCP also enabled us to avoid the conventional limitations of μCP,15 such as resolution blurring and feature size variation caused by stamp deformation, lateral spreading of thiol inks, and unpolymerized residues remaining on the contact surface. 2923

DOI: 10.1021/la504516e Langmuir 2015, 31, 2922−2930

Article

Langmuir Table 1. O/C, SVI/S−II Atomic Ratio, and WCA on SAM−OH-Modified Gold Surfaces with Varying Exposure Timea entry 1 2 3 4 5

substrate UV UV UV UV UV

(0 min) (10 min) (20 min) (25 min) (50 min)

C(1s)%

O(1s)%

O/C

SII(area) (Au−S)

SVI(area) (Au-SO3−R)

SVI/ SII

WCA (deg)

84.07 73.06 65.11 56.79 68.39

15.93 26.94 34.89 43.03 31.61

0.19 0.37 0.54 0.76 0.46

1891.2 1705.1 1673.5 1487.1 1450.7

0 1431.0 1586.2 1689.8 2406.3

0 0.84 0.95 1.1 1.7

30.0 57.2 56.1 54.0 46.5

a

The O/C ratio increased with the exposure time, with the exception that the O/C ratio in entry 5 decreased somewhat. This might be due to enhanced carbon pollution on the surface during long UV irradiation.

all of the samples, it was found that the spectrum for SAM−OH with short irradiation (e.g., 10 min) was close to that for pristine SAM−OH without photooxidation, indicating the formation of a less-ablated surface after short irradiation. In contrast, long irradiation (e.g., 50 min) afforded the spectrum that deviated the most from the pristine SAM−OH without photooxidation, implying the existence of a heavily ablated surface after long irradiation. Besides the change in peak intensity, the gradual shift of the peak positions to higher wave numbers after the photooxidation further implied that the disorder was introduced onto the UV-exposed surfaces, as the disorganized SAMs were supposed to have more liquidlike peaks at higher wavenumbers than the organized SAMs with more crystalline-like peaks at lower wavenumbers. In addition to the IR information, the thickness of SAM−OH as a function of the irradiation time was measured (Figure S4) to study the disordering effect. It was clearly found that the irradiation time positively produced a gradually decreasing SAM thickness, which was in accordance with the photoablation effect revealed by the grazing-incidence IR spectra (Figure S3). The continuous decrease in the SAM thickness with irradiation time thus reflected a gradual disorder occurring with prolonged exposure. The etching rate on the different types of surfaces was determined on the basis of the ellipsometry thickness measurements (Figure 2). This first attempt at verifying the etching rates on bare gold and SAM−OH-modified gold films

286.1 eV for the C−OH bond, the oxidized SAMs further presented characteristic new oxidation peaks at 168.0 eV assigned to SVI from Au−SO3R,31 and 288.8 eV assigned to COOH. The increasing elemental ratio of SVI/S−II and O/C with exposure time indicated that the extent of oxidation increased when prolonging the irradiation time (Table 1). The XPS results were further confirmed by TOF−SIMS characterizations (Figure 1b). After UV exposure, the peaks at m/z 394, 426, and 427 assigned to the unoxidized Au−S species were markedly mixed with the peaks at m/z 80, 97, 391, 413, 429, and 461 assigned to the oxidized Au-thiol species.34 The peaks at low m/z such as 80 and 97 assigned to SO3− and HSO4− also appeared on the sample before irradiation and were ascribed to the recombination of Au−S fragments with oxygen atoms in air during the measurement. The peaks at the lowerlimit m/z from 13 to 59 were assigned to the recombinational species of SAM fragments before and after the photooxidation.35,36 The newly formed Au−S oxidation peaks including 80 and 97 gradually predominated the spectra when increasing the exposure time, which supported the deduction that UV exposure induced gradual photooxidation on the SAM−OH surface. The O/C ratio on the photooxidized SAM−OH surfaces was also followed by XPS (Table 1), and an increasing elemental ratio of O/C with irradiation time was observed on the photooxidized surfaces due to oxidation.37,38 It was further found that the WCA measurements on the photoexposed surfaces (Table 1) could be directly correlated to the disordering of SAM−OH after UV exposure because all of the exposed surfaces presented a higher WCA than the original unexposed SAM−OH surface, which could be attributed to photooxidation-induced partial peeling or orientational disordering of oxidized alkanethiolates. With a prolonged irradiation time, the WCA on the oxidized surfaces showed a tendency to decrease, which was ascribed to the hydrophilic contribution from polar oxygen-containing functionalities newly formed on the oxidized surface. The extent of disorder was further supported by grazingincidence IR spectroscopy (Figure S3) and ellipsometry measurements (Figure S4). Grazing-incidence IR spectroscopy is considered to be a powerful tool for characterizing the integrity of SAMs formed on a substrate by focusing on the changes in peak intensities and positions for alkyl chains in the SAMs.39 As shown in Figure S3, in a comparison with the pristine SAM−OH without photooxidation, a general tendency for the photooxidized samples was a gradual decrease in the CH2 symmetric stretch of solid alkanes (2849 cm−1), CH2 asymmetric stretch (2918 cm−1), and CH3 symmetric stretch (2960 cm−1), which reflected that the SAMs were gradually photoablated as the irradiation time increased. This photoablation was also in accordance with the thickness measurement of the photoexposed SAM as shown in Figure S4. Among

Figure 2. Comparison of the etching rates on bare gold and on a gold film covered with SAM−OH at different irradiation times. Along the x axis, “0*” symbolizes the bare gold without the SAM−OH covering with UV exposure for 0 min; 0, 10, and 50 represent the gold layer with the SAM−OH covering with UV exposure for 0, 10, and 50 min, respectively. Compared to the etching rate on the gold film with the SAM−OH covering without UV irradiation, a retardation of the etching rate was clearly observed on the gold film with the SAM−OH covering after UV exposure for 10 min. Moreover, as indicated by the arrow, it was shown that the minimum value in the etching rate shifted toward the short irradiation duration, indicating that a short irradiation facilitated a slow etching rate. 2924

DOI: 10.1021/la504516e Langmuir 2015, 31, 2922−2930

Article

Langmuir

result was in contrast to the pattern formed by μCP (Scheme S2),15 where the transferred SAM−OH on the gold surface protected the gold layer from the etching (Figure 3c,d). We found that the resultant gold structure by NBS/py either in negative or positive tone had a similar roughness (12−15 nm in RMS) to that obtained by μCP (15 nm in RMS). Furthermore, a smaller size variation ratio was observed on the samples prepared by the NBS/py method when comparing it with that obtained by μCP (Table S1). The large size variation ratio from μCP might be ascribed to the unsatisfied ability of the resist and inherent lateral ink diffusion of SAM−OH formed by the μCP process. The formation of a negative-tone pattern could be attributed to the use of the unique NBS/py. This was based on the finding from another control experiment by using the commercially well developed cyanide etchant to replace the NBS/py solution while keeping the photomask-controlled UV exposure unchanged. In this case, only the classical positive-tone pattern was found (Figure 3e,f). Also, as indicated in Figure 3f, the resultant positive pattern by the commercial cyanide etchant presented some defects that were attributed to the limitation that small molecular etchants may diffuse through the defects in the SAMs to etch the underlying substrate.4,5 Such a reagent dependence for the formation of the negative pattern also implied that the mechanism was mainly from the etchant system itself and not very correlated to the surface crosslinking/stabilization. If the cross-linking/stabilization of the SAM predominantly accounted for the formation of the negative pattern as just found in the aromatic SAMs,19−23 a similar result would be expected when replacing NBS/py with a conventional cyanide etchant. However, the actual positive pattern shown in Figure 3e,f implied the invalidity of the above mechanism in the present case. By monitoring the pattern evolution with the irradiation time, we further discovered an unexpected transition between negative- and positive-tone patterns on a SAM−OH-modified gold surface, which has never been reported previously. As shown in Figure 4, the negative-tone pattern was first observed after the irradiation for 20 min or less. After that, the positivetone pattern that would be expected in conventional opinions appeared and was stably maintained during prolonged UV exposure. The negative-to-positive transition was observed at 25 min, at which time the pattern tone was ambiguous. As mentioned above, the apparent reason for the visual observation of either negative or positive patterns was ascribed to the difference in height of the gold layer between the exposed and unexposed areas. Specifically, in a typical negative pattern, the layer was thicker on the UV-exposed area than on the unexposed background, while conversely, in a typical positive pattern, the unexposed background was thicker. Such judgments based on the optical inspections were further confirmed by AFM on the cross-sectional profile of the etched gold substrate (Figure 5). The tone feature was clearly consistent with the results from the optical microscopic images (Figures 3 and 4), precluding the disturbance of illusions from microscopic optical verification. The negative-tone pattern reflected that the photoexposed area surprisingly supported a retarded etching rate as compared to the unexposed background, and this phenomenon changed back to normal after the positive-tone pattern appeared. This deduction was directly supported by the earlier evidence (Figure 2) that quantitatively described the etching rate differences among unexposed, shortirradiated, and long-irradiated SAM−OH surfaces. Contrary to

before and after UV exposure clearly showed an unexpected tendency. As can be seen in Figure 2, although both of the etching rates on the SAM−OH-modified gold surface before/ after UV exposure were slower than that on the bare gold surface, an interesting finding was that the etching rate on the SAM−OH surface after the short UV exposure (10 min) was surprisingly slower than that on the SAM−OH surface before UV irradiation. In contrast, the etching rate on the SAM−OH surface after the long UV irradiation (50 min) was naturally faster than that on the unexposed SAM−OH surface. Moreover, from Figure 2, it was told that the minimum value in the etching rate shifted toward the short irradiation duration, which clearly indicated that a short irradiation facilitated a slow etching rate. This distinct result encouraged us to further develop a novel negative pattern based on a UV-primed SAM−OH surface. For this aim, the UV exposure on the SAM−OH surface was deliberately controlled by a photomask so that the UV-exposed and unexposed areas could be exactly defined by the features of the photomask (Scheme S1). After obtaining such a photoprimed SAM−OH surface, we immersed this substrate in NBS/ py solution for a determined time. As shown in Figure 3a,b, it was clear that under the short UV irradiation (∼20 min), the UV-exposed area had a higher elevation than the unexposed background so that a negative-tone pattern was formed. This

Figure 3. Comparison of our method (a, b) with conventional μCP (c, d) and a cyanide etchant system (e, f). (a, c, e) Optical photographs. (b, d, f) SEM micrographs. (a, b) The concentrations of NBS and py were 70 and 240 mM, respectively. (c, d) Labels “SAM area” and “NO SAM” denoted the printed SAM-modified region by μCP and the pristine bare gold region, respectively. In each image, one inset was used to show clearly the cross-sectional profile on the surface after lithography. 2925

DOI: 10.1021/la504516e Langmuir 2015, 31, 2922−2930

Article

Langmuir

Figure 4. Etching process after micropatterned photopriming on a SAM−OH-coated gold layer as a function of the UV irradiation time. Left panel (a−d): optical photographs. Right panel (a−d): SEM micrographs. The concentrations of NBS and py were 70 and 240 mM, respectively. (a, d) One inset was used to show clearly the cross-sectional profile on the surface after lithography.

the imperfect feature size variation observed in conventional μCP, either negative- or positive-tone patterns created by the present method could faithfully maintain the feature size in the employed photomask, and lower size variations were observed for the obtained patterns (Figure S5, Table S1). The preferential etching retardation on the disordered SAM−OH surface urged us to look for an important factor capable of reversing the etching rate order on the disordered and ordered SAM surfaces. In a previous report, we proposed the reaction mechanism for bare gold surface lithography by an NBS/py etchant.27 In the suggested reaction process (Figure S6), the neutral gold atom on the gold surface was first oxidized by a bromine molecule released from the NBS reagent (1), and the resultant AuBr4− then coordinated with a pyridine ligand to form the pyAuBr3 product (2) that could be separated from the solution due to poor solubility in water. As a result, a fresh neutral gold atom was re-exposed for further redox reaction.27 The key intermediate product in the reaction was pyAuBr3, and the formation of such a poorly soluble compound has been fully proven in a previous study,27 where the separation, purification, and characterization of this reaction product from the etched solution were successfully resolved. The present work involved further elaborating the mechanism from our previous report,27 with an emphasis on different etching-resistant behaviors on both unexposed and irradiated SAM surfaces. For this aim, we analyzed the chemical change on the etched SAM-protected surfaces by TOF−SIMS (Figure 6), expecting that the evaluation of the surface reaction product could shed important light on the reaction pathway. Obviously, the collected TOF−SIMS spectra shown in Figure 6 could be classified into two groups. In the first group, the spectra on the unexposed (0 min) and short-irradiated (10 min) SAM surfaces presented a dominant peak at m/z 357, which could be indexed to the reference structure as AuBr2−.40 In contrast, the second group, comprising the spectra of the long-irradiated (20 and 50 min) SAM surfaces, not only presented the peak at m/z 357 but also showed a new peak at m/z 517 that could be indexed to the reference structure as AuBr4−.40 A general tendency was that the AuBr4− signal became enhanced when the irradiation time was prolonged. Besides the main peaks for AuBr2− and AuBr4−, the other peaks for the side oxidation products, labeled with “*” in Figure 6, were observed and attributed to molecular fragments and

Figure 5. AFM scanning on the cross-sectional profile of negative (a, c) and positive (b, d) micropatterns of the etched gold substrate, respectively. (a) Typical negative circular pattern to show the preferential etching on the unexposed SAM area (outside the circular pattern). (b) Typical positive circular pattern to show the preferential etching on the exposed SAM area (inside the circular pattern). (c) Corresponding negative pattern and profile with stripelike features. (d) Corresponding positive pattern and profile with stripelike features. The RMS values on the circular area in (a) and (b) were 12 and 15 nm, respectively.

2926

DOI: 10.1021/la504516e Langmuir 2015, 31, 2922−2930

Article

Langmuir

Figure 6. TOF−SIMS spectra on the photoexposed SAM−OH surfaces with varying irradiation time after etching in an NBS/py solution. The concentrations of NBS and py were 70 and 240 mM, respectively. The peaks around 505, 545, 555, and 586 m/z labeled with “*” were assigned to molecular fragments and recombinations from the etched gold species, which were specifically indentified in Figure S7.

The above-mentioned TOF−SIMS results fit well into the established reaction mechanism between bromine and the gold surface, as developed by Pesic,40 Crooks,41 and Mirkin.42 In our case (Scheme 2), NBS was first introduced to provide a stable

recombinations from the etched gold species. When the irradiation time was increased, the etched samples showed the attenuated signals from the side reaction products at 503/ 505/507 and 586 m/z, while the signal from the main reaction product at 517 m/z for AuBr4− gradually dominated all of the spectra. This result clearly reflected that long UV exposure induced an elevated oxidation extent of SAM (as reflected by the emergence of carboxylate species on the long-irradiated sample), which consequently supported a facilitated etching by NBS/py to produce main reaction product AuBr4−. To study the role of SAM during the etching process, a control experiment was further conducted to investigate by TOF−SIMS the effect of etching on a bare gold surface without a SAM covering (Figure S7). It was found that the reaction products for the etching on bare gold were similar to those on the etched SAM-modified surfaces without irradiation or with short irradiation (e.g., 10 min) (Figure 6). This was understandable because the existence of a SAM would not significantly change the reaction route but would affect the reaction rate. Along this line, one could find that the product signals in Figure S7 were higher than those on the etched SAMmodified surfaces without irradiation or with short irradiation (e.g., 10 min), indicating that the etching rate on the bare gold surface was faster than those on the SAM-covered surfaces. This result was consistent with the etching rate comparisons obtained by the ellipsometry measurements (Figure 2). As the present work claims that differentiated etching rates were obtained on the SAM-modified surfaces with different irradiation durations, the result shown in Figure S7 also pointed out that both the unexposed SAM and the shortirradiated SAM surfaces had a suppressed etching rate as compared to the bare gold surface without a SAM covering. This was reasonable because the existence of SAM served as a physical barrier against etching.

Scheme 2. Schematic Process to Show the Etching Extents on SAM−OH Surfaces with Varying Exposure Time

low-concentration bromine source through a classical radical mechanism.43 The formed bromine (Br−) then attacked and bound to the gold surface40,41,44 by converting neutral Au to AuBr2− (eqs 1 and 2), as reflected in the m/z 357 peak in all of the sample spectra. The resultant AuBr2− subsequently disproportionated to metallic gold and AuBr4− (eq 3), as 2927

DOI: 10.1021/la504516e Langmuir 2015, 31, 2922−2930

Article

Langmuir reflected by the m/z 517 peak in the spectra of the longirradiated samples. AuBr4− could diffuse into the solution40 and be captured by pyridine to form pyAuBr3 (eq 4) that was separated from the system.27 The effective consumption of AuBr4− also diminished the influence of the side reaction to recover AuBr2− from AuBr4− according to eq 5.40,41 The entire process constantly exhausted AuBr4− and consequently facilitated the right-shift of the reaction equilibrium in eqs 1−4 to continuously etch the gold. The mere appearance of an AuBr2− peak without an AuBr4− peak accompanying the spectra of the unexposed and shortirradiated SAM surfaces (Figure 6) proved that, on these surfaces, the lithography was mainly terminated by eq 2. The main consequence of this reaction was surface corrosion passivation (Scheme 2). This was because main product AuBr2− had poor solubility in water and thus bound to the gold surface,40−42 inhibiting the subsequent etch by Br− through steric hindrance and electrostatic repulsion. The lithography retardation further relied on the contribution of SAM−OH during the etching.41 Crooks et al. pointed out that, on a hydrophilic SAM-modified gold surface, e.g., SAMs with OH or COOH as end groups, etching by Br− was first initiated from the SAM defects, after which the resultant pits on the defect regions expanded laterally along the surface plane.41 Once an entire Au monolayer was etched in this way, the surface was fully occupied by alkanethiols that reoriented and adsorbed parallel to the surface to inhibit subsequent deep pitting.41 We further supplied the lithography results on SAMs with different end groups (e.g., COOH, CH3). On the basis of the layer-bylayer lithography,41 a similar lithography transition from negative to positive could be expected on the hydrophilic SAMs with COOH as terminal groups (SAM−COOH) and was proven in Figure 7A. Unlike the layer-by-layer corrosion on these hydrophilic SAMs, localized corrosion (pitting) was suggested on the hydrophobic SAMs with CH3 as end groups (SAM−CH3).41 This behavior resulted in a persistent negative lithography observed in our case under either short (∼20 min)

or long (∼50 min) irradiation (Figure 7B). In the pitting model for SAM−CH3, the etching pits were quickly formed without the alkanethiols laterally covered. As a result, the lithography quickly enabled surface passivation due to the formation of a large amount of insoluble AuBr2−. Accordingly, negative lithography was always obtained under either short (∼20 min) or long (∼50 min) irradiation (Figure 7B). In our case, the etching on the short-irradiated SAM surface would be more inhibited than that on the unexposed SAM surface because the SAM defects formed on the short-irradiated SAM surface accelerated eq 2 more intensively to exert a stronger etching resistance than the unexposed SAM background, resulting in the formation of a negative pattern (Scheme 2). According to this surface passivation model, the etching rate difference between the SAM area and bare gold shown in Figure 2 might be further enlarged by adding other compounds to facilitate or enhance such surface passivation. Some defect healing additives that are generally used in SAMassisted lithography could be selected to fulfill this aim, such as n-octanol or 1-decanesulfonamide.7,8,45 When the irradiation time increased to 20 min, the etching reaction came to a transition point as this time was close to that required for the transition from negative to positive patterns observed in Figure 4. At this stage, besides the appearance of a dominant AuBr2− peak, the TOF−SIMS spectrum (20 min) in Figure 6 also showed a new small peak for AuBr4−. These results reflected that the lithography reaction did not stop at eq 2 but started to enter into eq 3. As the appearance of AuBr4− was an indicative signal of continuous and effective etching,44 the emergence of the AuBr4− peak implied the occurrence of a remarkable etching process. In the TOF−SIMS spectrum of the longirradiated sample (50 min, Figure 6), the AuBr4− peak quickly increased, indicating that eq 3 dominated the entire lithography. At this stage, the more disordered SAM surface provided many defects that initiated the continuous dissolution of gold through eqs 1−4 to form normal positive patterns (Scheme 2). The enhanced AuBr4− peak on the long-irradiated sample surface also suggested that the diffusion of AuBr4− from the surface to the solution was not very fast and was confined by diffusion and binding (with py) kinetics.

3. CONCLUSION AND REMARKS This is a new report of an unknown effect on SAM surfaces that refutes the classical theory describing the function of SAM layers as lithography resists. The finding indicates that disordered SAM−OH would, in an NBS/py system, induce a retarded etching rate that is slower than that on a well-defined SAM−OH surface, thereby promoting the formation of a negative-tone pattern that is complementary to the pattern in the employed photomask. This work provides a new reaction model for wet chemical lithography on a metal substrate. The more important implication of this finding is that, similar to both positivetone and negative-tone photolithography systems, SAM resists commonly used in wet chemical lithography are unconventionally demonstrated to be able to be exploited in both positive-tone and negative-tone patterns, thereby greatly enhancing the design flexibility for complex molecular circuits and submicrometer ultrafine lithography. The simple photopatterning process without the assistance of μCP also avoids the limitations associated with the latter. In light of the fact that UV-induced photopatterning of SAMs has been studied extensively to afford fundamental surface chemistry tailor-

Figure 7. Lithography results on SAM−COOH (A) and SAM−CH3 (B) surfaces by the NBS/py method. (A1) Negative lithography obtained on a SAM−COOH surface under short irradiation. (A2) Positive lithography obtained on a SAM−COOH surface under long irradiation. (B1) Negative lithography obtained on a SAM−CH3 surface under short irradiation. (B2) Negative lithography obtained on a SAM−CH3 surface under long irradiation. 2928

DOI: 10.1021/la504516e Langmuir 2015, 31, 2922−2930

Article

Langmuir ing46−50 and engineering applications such as biointerfaces51−54 and the micro/nanofabrication of metal substrates,55−57 we expect the present approach to impart distinctive and flexible design capabilities to conventional UV-induced photopatterning of SAMs.

■

(12) Ballav, N.; Shaporenko, A.; Terfort, A.; Zharnikov, M. A Flexible Approach to the Fabrication of Chemical Gradients. Adv. Mater. 2007, 19, 998−1000. (13) Ballav, N.; Weidner, T.; Zharnikov, M. UV-Promoted Exchange Reaction as a Tool for Gradual Tuning the Composition of Binary Self-Assembled Monolayers and Chemical Lithography. J. Phys. Chem. C 2007, 111, 12002−12010. (14) Geissler, M.; Xia, Y. Patterning: Principles and Some New Developments. Adv. Mater. 2004, 16, 1249−1269. (15) Delamarche, E.; Geissler, M.; Wolf, H.; Michel, B. Positive Microcontact Printing. J. Am. Chem. Soc. 2002, 124, 3834−3835. (16) Perl, A.; Péter, M.; Ravoo, B. J.; Reinhoudt, D. N.; Huskens, J. Heavyweight Dendritic Inks for positive Microcontact Printing. Langmuir 2006, 22, 7568−7573. (17) Saalmink, M.; van der Marel, C.; Stapert, H. R.; Burdinski, D. Positive Microcontact Printing with Mercaptoalkyloligo(ethylene glycol)s. Langmuir 2006, 22, 1016−1026. (18) Perl, A.; Reinhoudt, D. N.; Huskens, J. Microcontact Printing: Limitations and Achievements. Adv. Mater. 2009, 21, 2257−2268. (19) Weng, C.-C.; Wu, Y.-T.; Tsai, S.-C.; Chen, C.-H.; Zharnikov, M. Patterning of Alkanethiolate Self-Assembled Monolayers by Downstream Microwave Nitrogen Plasma: Negative and Positive Resist Behavior. J. Vac. Sci. Technol., B 2009, 27, 1949−1957. (20) Klauser, R.; Hong, I.-H.; Wang, S.-C.; Zharnikov, M.; Paul, A.; Gölzhäuser, A.; Terfort, A.; Chuang, T. J. Imaging and Patterning of Monomolecular Resists by Zone-Plate-Focused X-ray Microprobe. J. Phys. Chem. B 2003, 107, 13133−13142. (21) Klauser, R.; Huang, M.-L.; Wang, S.-C.; Chen, C.-H.; Chuang, T. J.; Terfort, A.; Zharnikov, M. Lithography with a Focused Soft X-ray Beam and a Monomolecular Resist. Langmuir 2004, 20, 2050−2053. (22) Klauser, R.; Chen, C.-H.; Huang, M.-L.; Wang, S.-C.; Chuang, T. J.; Zharnikov, M. Patterning and Imaging of Self-Assembled Monolayers with a Focused Soft X-ray Beam. J. Electron Spectrosc. Relat. Phenom. 2005, 144−147, 393−396. (23) Wu, Y.-T.; Liao, J.-D.; Weng, C.-C.; Hesieh, Y.-T.; Chen, C.-H.; Wang, M.-C.; Zharnikov, M. Alkanethiolate Self-Assembled Monolayers As a Negative or Positive Resist for Electron Lithography. J. Phys. Chem. C 2009, 113, 4543−4548. (24) Schwartz, J. J.; Hohman, J. N.; Morin, E. I.; Weiss, P. S. Molecular Flux Dependence of Chemical Patterning by Microcontact Printing. ACS Appl. Mater. Interfaces 2013, 5, 10310−10316. (25) Shao, J.; Josephs, E. A.; Lee, C.; Lopez, A.; Ye, T. Electrochemical Etching of Gold within Nanoshaved Self-Assembled Monolayers. ACS Nano 2013, 7, 5421−5429. (26) Ul-Haq, E.; Patole, S.; Moxey, M.; Amstad, E.; Vasilev, C.; Hunter, C. N.; Leggett, G. J.; Spencer, N. D.; Williams, N. H. Photocatalytic Nanolithography of Self-Assembled Monolayers and Proteins. ACS Nano 2013, 7, 7610−7618. (27) Yang, P.; Zhang, X. Nucleic Acid-Mediated Gold Oxidation: Novel Biolithography for Surface Microfabrication and New Insight on Gold-Based Biomaterials. Chem. Commun. 2013, 48, 8787−8789. (28) Gourianova, S.; Willenbacher, N.; Kutschera, M. Chemical Force Microscopy Study of Adhesive Properties of Polypropylene Films: Influence of Surface Polarity and Medium. Langmuir 2005, 21, 5429−5438. (29) Bain, C. D.; Biebuyck, H. A.; Whitesides, G. M. Comparison of Self-Assembled Monolayers on Gold: Coadsorption of Thiols and Disulfides. Langmuir 1989, 5, 723−727. (30) Ista, L. K.; Callow, M. E.; Finlay, J. A.; Coleman, S. E.; Nolasco, A. C.; Simons, R. H.; Callow, J. A.; Lopez, G. P. Effect of Substratum Surface Chemistry and Surface Energy on Attachment of Marine Bacteria and Algal Spores. Appl. Envrion. Microbiol. 2004, 70, 4151− 4157. (31) Rieley, H.; Price, N. J.; Smith, T. L.; Yang, S. Photo-Oxidation and Photo-Reduction in Alkylthiol Monolayers Self-Assembled on Gold. J. Chem. Soc., Faraday Trans. 1996, 92, 3629−3634. (32) Huang, J.; Hemminger, J. C. Photooxidation of Thiols in SelfAssembled Monolayers on Gold. J. Am. Chem. Soc. 1993, 115, 3342− 3343.

ASSOCIATED CONTENT

S Supporting Information *

Experimental details, characterization methods, and additional supporting evidence. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS Grants 21374057 and 51303100 from the National Natural Science Foundation of China (NSFC), the Fundamental Research Funds for the Central Universities (GK201301006), and the 111 Project (B14041) are gratefully acknowledged. This research was also supported by the Program for Changjiang Scholars and Innovative Research Team in University (IRT_14R33).

■

REFERENCES

(1) Love, J. C.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.; Whitesides, G. M. Self-Assembled Monolayers of Thiolates on Metals as a Form of Nanotechnology. Chem. Rev. 2005, 105, 1103−1169. (2) Aizenberg, J.; Black, A. J.; Whitesides, G. M. Controlling Local Disorder in Self-Assembled Monolayers by Patterning the Topography of Their Metallic Supports. Nature 1998, 394, 868−871. (3) Black, A. J.; Paul, K. E.; Aizenberg, J.; Whitesides, G. Patterning Disorder in Monolayer Resists for the Fabrication of Sub-100-nm Structures in Silver, Gold, Silicon, and Aluminum. J. Am. Chem. Soc. 1999, 121, 8356−8365. (4) Zhao, X.-M.; Wilbur, J. L.; Whitesides, G. M. Using Two-Stage Chemical Amplification To Determine the Density of Defects in SelfAssembled Monolayers of Alkanethiolates on Gold. Langmuir 1996, 12, 3257−3264. (5) Xia, Y.; Zhao, X. M.; Kim, E.; Whitesides, G. M. A Selective Etching Solution for Use with Patterned Self-Assembled Monolayers of Alkanethiolates on Gold. Chem. Mater. 1995, 7, 2332−2337. (6) Sundar, V. C.; Aizenberg, J. Narrow Features in Metals at the Interfaces Between Different Etch Resists. Appl. Phys. Lett. 2003, 83, 2259−2261. (7) Geissler, M.; Schmid, H.; Bietsch, A.; Michel, B.; Delamarche, E. Defect-Tolerant and Directional Wet-Etch Systems for Using Monolayers as Resists. Langmuir 2002, 18, 2374−2377. (8) Burdinski, D.; Blees, M. H. Thiosulfate- and Thiosulfonate-Based Etchants for the Patterning of Gold Using Microcontact Printing. Chem. Mater. 2007, 19, 3933−3944. (9) Zharnikov, M.; Grunze, M. Modification of Thiol-Derived SelfAssembling Monolayers by Electron and X-ray Irradiation: Scientific and Lthographic Aspects. J. Vac. Sci. Technol., B 2002, 20, 1793−1807. (10) Geyer, W.; Stadler, V.; Eck, W.; Zharnikov, M.; Gölzhäuser, A.; Grunze, M. Electron-Induced Crosslinking of Aromatic SelfAssembled Monolayers: Negative Resists for Nanolithography. Appl. Phys. Lett. 1999, 75, 2401−2403. (11) Eck, W.; Stadler, V.; Geyer, W.; Zharnikov, M.; Gölzhäuser, A.; Grunze, M. Generation of Surface Amino Groups on Aromatic SelfAssembled Monolayers by Low Energy Electron BeamsA First Step Towards Chemical Lithography. Adv. Mater. 2000, 12, 805−808. 2929

DOI: 10.1021/la504516e Langmuir 2015, 31, 2922−2930

Article

Langmuir

Photopatterned Self-Assembled Monolayers. Chem.Eur. J. 2007, 13, 7957−7964. (53) Hazarika, P.; Behrendt, J. M.; Petersson, L.; Wingren, C.; Turner, M. L. Photopatterning of Self Assembled Monolayers on Oxide Surfaces for the Selective Attachment of Biomolecules. Biosens. Bioelectron. 2014, 53, 82−89. (54) Tarlov, M. J.; Burgess, D. R. F., Jr.; Gillen, G. UV Photopatterning of Alkanethiolate Monolayers Self-Assembled on Gold and Silver. J. Am. Chem. Soc. 1993, 115, 5305−5306. (55) Zhou, C.; Nagy, G.; Walker, A. V. Toward Molecular Electronic Circuitry: Selective Deposition of Metals on Patterned Self-Assembled Monolayer Surfaces. J. Am. Chem. Soc. 2005, 127, 12160−12161. (56) Huang, J.; Dahlgren, D. A.; Hemminger, J. C. Photopatterning of Self-Assembled Alkanethiolate Monolayers on Gold: A Simple Monolayer Photoresist Utilizing Aqueous Chemistry. Langmuir 1994, 10, 626−628. (57) Zhou, C.; Trionfi, A.; Jones, J. C.; Hsu, J. W. P.; Walker, A. V. Comparison of Chemical Lithography Using Alkanethiolate SelfAssembled Monolayers on GaAs (001) and Au. Langmuir 2010, 26, 4523−4528.

(33) Cooper, E.; Leggett, G. J. Static Secondary Ion Mass Spectrometry Studies of Self-Assembled Monolayers: Influence of Adsorbate Chain Length and Terminal Functional Group on Rates of Photooxidation of Alkanethiols on Gold. Langmuir 1998, 14, 4795− 4801. (34) The proposed Au(CH2)12SO2− for 429 and Au(CH2)12SO4− for 461 were deduced on the basis of the comprehensive considerations from the photooxidation of Au−S (XPS in Figure 1a) and the complementary TOF−SIMS data in the low m/z range (80 and 97 for SO3− and HSO4−). Such a molecular cluster ion was from the adduct of the oxidized C11 thiol with CH2 ion, which possibly appeared during TOF−SIMS measurement on the alkanethiolate SAM surface, as reported in ref 35. (35) Graham, D. J.; Ratner, B. D. Multivariate Analysis of TOF− SIMS Spectra from Dodecanethiol SAM Assembly on Gold: Spectral Interpretation and TOF−SIMS Fragmentation Processes. Langmuir 2002, 18, 5861−5868. (36) Wolf, K. V.; Cole, D. A.; Bernasek, S. L. High-Resolution TOF− SIMS Study of Varying Chain Length Self-Assembled Monolayer Surfaces. Anal. Chem. 2002, 74, 5009−5016. (37) King, D. E. Oxidation of Gold by Ultraviolet Light and Ozone at 25 °C. J. Vac. Sci. Technol., A 1995, 13, 1247−1253. (38) Mirsaleh-Kohan, N.; Bass, A. D.; Sanche, L. X-ray Photoelectron Spectroscopy Analysis of Gold Surfaces after Removal of Thiolated DNA Oligomers by Ultraviolet/Ozone Treatment. Langmuir 2010, 26, 6508−6514. (39) Liu, Y.; Kalambur, A. T.; Rabolt, J. F. Time-Resolved IR Studies of Monolayer Self-Assembly on a Gold Substrate Using Planar Array Infrared (PA-IR) Spectroscopy. Langmuir 2004, 20, 816−822. (40) Pesic, B.; Sergent, R. H. Reaction Mechanism of Gold Dissolution with Bromine. Metal. Trans. 1993, 24B, 419−431. (41) Zamborini, F. P.; Crooks, R. M. Corrosion Passivation of Gold by n-Alkanethiol Self-Assembled Monolayers: Effect of Chain Length and End Group. Langmuir 1998, 14, 3279−3286. (42) Langille, M. R.; Personick, M. L.; Zhang, J.; Mirkin, C. A. Defining Rules for the Shape Evolution of Gold Nanoparticles. J. Am. Chem. Soc. 2012, 134, 14542−14554. (43) Huang, R. L.; Goh, S. H.; Ong, S. H. The Chemistry of Free Radicals; Edward Arnold: London, 1974. (44) Singh, S.; Prasad, B. L. V. Nearly Complete Oxidation of Au0 in Hydrophobized Nanoparticles to Au3+ ions by Bromosuccinimide. J. Phys. Chem. C 2007, 111, 14348−14352. (45) Burdinski, D.; Brans, H. J. A.; Decré, M. M. J. Single Etch Patterning of Stacked Silver and Molybdenum Alloy Layers on Glass Using Microcontact Wave Printing. J. Am. Chem. Soc. 2005, 127, 10786−10787. (46) Zhou, C.; Walker, A. V. Dependence of Patterned Binary Alkanethiolate Self-Assembled Monolayers on “UV-Photopatterning” Conditions and Evolution with Time, Terminal Group, and Methylene Chain Length. Langmuir 2006, 22, 11420−11425. (47) Zhou, C.; Walker, A. V. Formation of Multilayer Ultrathin Assemblies Using Chemical Lithography. Langmuir 2010, 26, 8441− 8449. (48) Adams, J.; Tizazu, G.; Janusz, S.; Brueck, S. R. J.; Lopez, G. P.; Leggett, G. J. Large-Area Nanopatterning of Self-Assembled Monolayers of Alkanethiolates by Interferometric Lithography. Langmuir 2010, 26, 13600−13606. (49) Leggett, G. J. Scanning Near-Field Photolithography–Surface Photochemistry with Nanoscale Spatial Resolution. Chem. Soc. Rev. 2006, 35, 1150−1161. (50) Frutos, A. G.; Brockman, J. M.; Corn, R. M. Reversible Protection and Reactive Patterning of Amineand Hydroxyl-Terminated Self-Assembled Monolayers on Gold Surfaces for the Fabrication of Biopolymer Arrays. Langmuir 2000, 16, 2192−2197. (51) Ryan, D.; Parviz, B. A.; Linder, V.; Semetey, V.; Sia, S. K.; Su, J.; Mrksich, M.; Whitesides, G. M. Patterning Multiple Aligned SelfAssembled Monolayers Using Light. Langmuir 2004, 20, 9080−9088. (52) Han, X.; Pradeep, S. N. D.; Critchley, K.; Sheikh, K.; Bushby, R. J.; Evans, S. D. Supported Bilayer Lipid Membrane Arrays on 2930

DOI: 10.1021/la504516e Langmuir 2015, 31, 2922−2930