Selective Electroless Deposition of Copper on Organic Thin Films with

Sep 21, 2011 - However, the deposited films were rough and contained irregular crystallites. Further, the copper penetrated through the film. In this ...
0 downloads 0 Views 5MB Size
Download PDF
ARTICLE pubs.acs.org/Langmuir

Selective Electroless Deposition of Copper on Organic Thin Films with Improved Morphology Peng Lu, Zhiwei Shi, and Amy V. Walker* Department of Materials Science and Engineering, University of Texas at Dallas, 800 West Campbell Road, RL 10, Richardson, Texas 75080, United States

bS Supporting Information ABSTRACT: We have investigated the selective electroless deposition (ELD) of Cu on functionalized self-assembled monolayers (SAMs). Previous studies have demonstrated that Cu deposits on COOH and CH3 terminated SAMs using ELD. However, the deposited films were rough and contained irregular crystallites. Further, the copper penetrated through the film. In this Article, we demonstrate that copper can be selectively deposited on COOH terminated SAMs with improved morphology and without penetration of copper through the organic layer. The method employs a Cu(II) seed layer and an additive, adenine or guanine. We demonstrate the efficacy of the technique on photopatterned CH3/ COOH SAMs. Copper is observed to deposit only atop the COOH terminated SAM area and not on the CH3 terminated SAM. The use of a Cu(II) seed layer increased the Cu ELD rate on both COOH and CH3 terminated SAMs. The deposited copper layer strongly adheres to the COOH terminated SAMs because the copper layer nucleates at Cu2+ carboxylate complexes. In contrast, the deposited copper layer can easily be removed from the CH3 terminated SAM surface because there is no specific copper surface interaction. The additives adenine and guanine mediate the interaction of Cu2+ and the deprotonated COOH terminated SAMs via the formation of additive carboxylate complexes. These complexes lead to significantly reduced copper penetration through the SAM. In the case of adenine, the diffusion of copper through the organic film was eliminated. This new technique for copper deposition will facilitate the development of inexpensive molecular electronics, sensors, and other nanotechological devices.

1. INTRODUCTION Metallized organic thin films have applications ranging from polymer light-emitting diodes (PLEDs)1,2 to memory elements3 to orthopedic implants.4 Organic films are low-cost, can achieve atomic-scale uniformity via self-assembly, and are synthetically flexible. However, they can only be incorporated in practical devices if reliable, robust, and selective metallic contacts can be made, which do not degrade the organic layer. Many studies of the interactions of metals with self-assembled monolayers (SAMs) and polymers have employed vapor deposition5 12 or lift-off methods.5,6,13 Lift-off methods yield homogeneous contacts but have poor reproducibility.5,13 Physical vapor deposition (PVD) of metals on organic thin films results in a range of behaviors from formation of stable contacts7,8 to degradation of the film10,12 to penetration of the metal through the film.8,9,11 Further, the outcome of PVD on organic surfaces is difficult to predict.10 Chemical vapor deposition (CVD)14 17 and atomic layer deposition (ALD)18 have been demonstrated to produce stable contacts, but these methods are generally very slow and require expensive vacuum equipment. Selective deposition of Cu is of particular interest because it is often employed as an interconnect material.19,20 Unfortunately, PVD of Cu on organic surfaces9,21 23 leads to poor interfacial film adhesion, diffusion of copper into the film, and generally r 2011 American Chemical Society

unacceptable device performance. An alternative method by which to deposit a metal overlayer is electroless deposition (ELD). ELD is a soft technique in which deposition occurs via the chemically promoted reduction of metal ions, without an externally applied potential. It can be performed at low temperatures (e50 C) and therefore has the potential to eliminate or greatly reduce copper penetration through SAMs and polymers.24 There are several conditions that must be satisfied to selectively deposit a metal contact on an organic layer using ELD: (a) the reaction must only occur on areas with the target surface functionality; (b) the film must be strongly adherent and electrically conductive; (c) there is no diffusion of metal into the organic layer; (d) the deposited film morphology is regular and homogeneous; and (e) there are no deleterious side reactions between the ELD bath components and the organic layer. ELD often requires a catalyst such as Pd or Sn colloid to be adsorbed on the surface prior to deposition, but these can contaminate the contact.24 These “seeding” steps are not normally spatially selective, although patterning methods such as chemical lithography, scanning probe lithography, and microcontact printing allow Received: July 22, 2011 Revised: September 16, 2011 Published: September 21, 2011 13022

dx.doi.org/10.1021/la202839z | Langmuir 2011, 27, 13022–13028

Langmuir selective deposition.25,26 There are also unseeded ELD conditions, which do not require the preadsorption of a catalyst and have the additional advantage that they are chemically selective.25 Several studies have demonstrated that ELD can be used to deposit Cu on CH3 and COOH terminated SAMs without seeding or other surface pretreatment.25,27,28 Zangmeister and van Zee25 observed that copper could be deposited on 4-mercaptobenzoic acid SAMs via the reduction of Cu2+ by formaldehyde at high pH (pH ≈ 12.8). The deposited layer contained large crystallites and was relatively rough; for a layer thickness of ∼200 nm, the root-mean-square (rms) roughness was 56 nm. These authors proposed that Cu2+ ions formed complexes with deprotonated carboxylic acid terminal groups, which acted as the nucleation sites for subsequent copper deposition.25 However, no copper deposition was observed on 3-mercaptobenzoic acid terminated SAMs, which was attributed to Cu2+ ions being sterically hindered from complex formation. Garno et al.28 also observed copper deposition on COOH terminated SAMs. In addition, they noted that small amounts of copper could be deposited on CH3 terminated SAMs. No copper was deposited on OH terminated SAMs, so that copper could be selectively deposited in the COOH areas of a patterned OH/ COOH terminated SAM.28 In later experiments, Lu and Walker27 investigated the reaction pathways involved in seedless ELD of Cu on functionalized SAMs, using time-of-flight secondary ion mass spectrometry (TOF SIMS), optical microscopy, and scanning electron microscopy. At room temperature (22 C), Cu deposited on CH3 and COOH terminated SAMs but not on OH terminated SAMs. The deposited layers were very rough and contained large crystallites. No copper deposition was observed on OH terminated SAMs because the hydroxyl groups reacted with formaldehyde (the reducing agent) to form an acetal, which prevented copper deposition. Thus, the selective Cu deposition observed by Garno et al.28 was most likely due to the degradation of the OH terminated SAM. After Cu ELD on COOH terminated SAMs in the SIMS spectra, Lu and Walker27 observed Cu(COO(CH2)x+ ions, which are characteristic of the formation of Cu2+ carboxylate complexes, in agreement with the mechanism postulated by Zangmeister and van Zee.25 No specific interaction between Cu2+ ions and CH3 terminal groups was observed. Once Cu2+ ions adsorbed on the CH3 and COOH SAMs, they are reduced by formaldehyde via hydride transfer to deposit Cu(0). At 45 C, no copper deposited on CH3 terminated SAMs, but Cu continued to deposit on COOH terminated SAMs at this temperature. This is because Cu2+ ions were not stabilized on CH3 terminated SAMs at this temperature (due to the lower energy of interaction). However, Cu2+ carboxylate complexes were still able to form on the deprotonated COOH acid terminated SAM at 45 C, and so copper continued to deposit. Finally, copper penetrated through both CH3 and COOH terminated SAMs to the Au/S interface and continued to do so for 48 h after deposition. Taken together, these data suggest that a strongly adherent copper layer can be selectively deposited on the COOH terminated areas of a patterned COOH/ CH3 SAM surface at 45 C without deleterious reactions of the bath components with the surrounding matrix. However, for this to be useful, the penetration of copper through the layer must be prevented, and the morphology of the deposit must be improved so that a layer of uniform crystallites is formed.

ARTICLE

Figure 1. Schematic structures of the additives (a) adenine and (b) guanine.

In this Article, we demonstrate the selective deposition of copper on COOH terminated SAMs with improved morphology and without penetration of copper. Czanderna and coworkers23,29 demonstrated using PVD that the amount of Cu penetration through a monolayer was reduced when the deposition rate was increased. Because the Cu Cu interaction is very strong, at high deposition rates, overlayer formation is favored over Cu penetration. We therefore first increased the rate of Cu ELD on functionalized SAMs by using a Cu(II) seed layer. Cu2+ is catalytically active30 and so is an ideal seed layer. This increases the Cu ELD rate on both COOH and CH3 terminated SAMs. The deposited layer strongly adhered to the COOH terminated SAMs but could easily be removed from CH3 terminated SAMs. However, Cu was still observed to penetrate through the SAMs to the Au/S interface. We then investigated the use of the organic additives adenine and guanine to increase the deposition rate (Figure 1). Organic additives typically lead to a decrease in ELD rates.19,31 33 However, adenine and guanine have been demonstrated to accelerate Cu deposition rates on inorganic substrates, which was attributed, in part, to their role in suppressing oxidation and passivation of the deposited metallic copper layer.19,31 Organic additives can also alter the grain structure of the metallic deposit.19,33 In contrast to previous studies,19,31 the use of these additives did not increase the amount of Cu deposited under seeded conditions. Rather, adenine and guanine mediated the interaction of Cu2+ and the deprotonated carboxylic acid terminated SAMs via the formation of additive carboxylate complexes. The formation of these complexes led to a significant reduction in the amount of copper penetration through the SAM. Indeed, for adenine, the diffusion of copper through the organic film was eliminated. The combination of self-seeding with adenine as an additive led to selective deposition of copper on COOH terminated SAM areas of COOH/ CH3 terminated patterned SAMs. The deposited Cu film is adherent and has small, uniform grain size.

2. EXPERIMENTAL SECTION 2.1. Materials. Copper sulfate (g99%), ethylene-1-diamine tetraacetic acid (EDTA), disodium salt dehydrate (g99.0%), formaldehyde (37 wt % in water), sodium hydroxide (g98%, pellets), adenine (g99%), and guanine (g99%) were purchased from Sigma Aldrich (Saint Louis, MO). Gold and chromium were obtained from Alfa Aesar Inc. and were of 99.995% purity. Hexadecanethiol (HDT) (99%) and 16-mercaptohexadecanoic acid (MHA) (99%) were purchased from Asemblon, Inc. (Redmond, WA). Anhydrous ethanol (ACS grade) was obtained from Aaper Alcohol (Shelbyville, KY). All chemicals were used as-received and without further purification. Native silicon oxide wafers (Æ111æ orientation) were purchased from Addison Technologies, Inc. 13023

dx.doi.org/10.1021/la202839z |Langmuir 2011, 27, 13022–13028

Langmuir (Pottstown, PA) and were cleaned with piranha etch (H2SO4:H2O2 = 3:1) before use. The preparation of alkanethiolate SAMs used in this study has been described in detail previously.7,34 37 Briefly, first Cr (∼5 nm) and then Au (∼100 nm) were thermally deposited onto clean Si native oxide wafers. The prepared Au substrates were then immersed into a 1 mM ethanolic solution of the relevant alkanethiolate molecule for 24 h at ambient temperature (21 ( 2 C) to prepare well-organized SAMs. The size of the samples employed was (∼1  1) cm2. To ensure that the SAMs were well-ordered and free from significant chemical contamination, one sample from each batch was taken and characterized using single-wavelength ellipsometry (Gaertner Scientific Corp., Skokie, IL) and time-of-flight secondary ion mass spectrometry (TOF SIMS) prior to electroless deposition. 2.2. Copper Electroless Deposition. First, a seeding solution was prepared using 4 mM CuSO4 and 10 mM EDTA disodium salt. The pH of the seeding solution was then adjusted to 12.8 with sodium hydroxide. To initiate Cu ELD, 10 mM formaldehyde was added to the seeding solution. The solution containing CuSO4, EDTA, and formaldehyde will be referred to as the plating solution. The seeding and plating solutions were used immediately after preparation. Samples were normally immersed in the seeding solution for 15 min prior to the addition of formaldehyde. After 15 min (“seeding time”), formaldehyde was added to initiate the deposition of copper. This type of deposition is termed “seeded deposition”. In some experiments, 0.04 M adenine or guanine was added to the seeding solution. The samples were then immersed in the seeding solution containing the additive, adenine or guanine, for 15 min (seeding time) prior to addition of formaldehyde. ELD was performed for 60 min at 45 C. After deposition, samples were sonicated for 3 min in deionized water. They were then rinsed thoroughly with copious amounts of deionized water and absolute ethanol, and dried with nitrogen gas. Samples were then immediately transferred to the TOF SIMS or SEM for analysis. 2.3. UV Photopatterning of SAMs. UV photopatterned CH3/ COOH SAM surfaces were prepared in the following way. First, a copper mask (Electron Microscopy, Inc., Hatfield, PA) was placed on top of the COOH terminated SAM (MHA) to be patterned. The sample was placed approximately 50 mm away from a 500 W Hg arc lamp equipped with a dichroic mirror and a narrow-band-pass filter (280 400 nm) (Thermal Oriel, Spectra Physics Inc., Stratford, CT). The COOH terminated SAM was then exposed to the UV light for 2 h to ensure that photooxidation was complete. After UV exposure, the photopatterned COOH terminated SAM was immersed into a freshly made 1 mM ethanolic solution of the second alkanethiol, HDT, for 24 h. After immersion, HDT was adsorbed in the areas exposed to UV light resulting in a CH3/ COOH SAM patterned surface. The patterned surfaces were rinsed thoroughly with degassed ethanol and dried with N2 gas.

2.4. Time-of-Flight Secondary Ion Mass Spectrometry (TOF SIMS). TOF SIMS analyses were performed using an ION TOF IV spectrometer (ION TOF Inc., Chestnut Hill, NY) equipped with a Bi liquid metal ion gun. Briefly, this instrument consists of an air lock for sample introduction, a preparation chamber, and an analysis chamber, each separated by gate valves. The pressure of the preparation and analysis chambers was maintained at e3.8  10 9 mbar. The primary Bi+ ions were accelerated to 25 keV and contained within a ∼100 nm diameter probe beam. The beam was rastered over (100  100) μm2 during spectra acquisition and (500  500) μm2 during image acquisition. All spectra were acquired in the static regime with a total accumulated primary ion dose less than 1011 ions cm 2.38 Thus, the spectra obtained are characteristic of the surface region of the sample. The ejected secondary ions were extracted into a time-of-flight mass spectrometer with a 2 kV potential and were reaccelerated to 10 keV before reaching the detector.

ARTICLE

Figure 2. (a,c) Optical and (b,d) SEM images after Cu ELD on COOH terminated SAMs under (a,b) unseeded conditions and (c,d) seeded conditions. In the optical images, the size of the box is (550  490) μm2. Peak intensities were reproducible to within (10% from scan to scan and from sample to sample. For each electroless deposition experiment, at least two samples were prepared, and three areas on each sample were examined. The spectra and images shown are representative of the data acquired. Optical images were obtained using a video camera (ExwaveHAD, Sony) mounted in the TOF SIMS analysis chamber. 2.5. Scanning Electron Microscopy (SEM). SEM measurements were conducted on a field emission scanning electron microscope (Hitachi s-4500) equipped with a NORAN Instruments energy dispersive X-ray (EDX) microanalysis system, a back scattering detector, and a mechanical straining stage.

3. RESULTS AND DISCUSSION 3.1. “Self” Seeding of Cu ELD on Functionalized SAMs. Figure 2a and b displays optical and SEM images, respectively, after Cu ELD on a COOH terminated SAM using “unseeded” conditions, simply immersion in the plating solution. Both the optical and the SEM images indicated that large scattered Cu crystallites formed. In contrast, after seeded deposition on a COOH terminated SAM, the optical image suggests that a copper layer formed, while the SEM image indicates that there is a densely packed layer of submicrometer crystallites present on the surface (Figure 1c,d). Seeded deposition is performed by immersing the SAM in the seeding solution for 15 min prior to the addition of formaldehyde to initiate Cu deposition. The deposited film is relatively rough, with a background of ∼150 nm crystallites with a few larger scattered crystallites (average size ∼270 nm). The deposited layer was adherent and could not be removed by sonication for 3 min in deionized water or by using adhesive tape. SIMS spectra confirm that the copper film formed under seeded conditions is adherent: the intensity of the 63Cu+ ion is unchanged by sonication in deionized water for 3 min (see the Supporting Information). 13024

dx.doi.org/10.1021/la202839z |Langmuir 2011, 27, 13022–13028

Langmuir

ARTICLE

Figure 3. High-resolution positive ion TOF SIMS spectra centered at m/z 163, of COOH terminated SAMs after Cu ELD using seeded and unseeded deposition conditions. Also shown is the TOF SIMS spectrum of the COOH terminated SAM prior to Cu ELD (“bare”) for reference.

Under seeded deposition conditions in the SIMS spectra, we also observe CuCOO(CH2)x+ ion intensities, for example, CuCOO(CH2)4+ (Figure 3), that are higher than under unseeded conditions. These ions are characteristic of Cu2+ carboxylate complex formation, which acts as the nucleation sites for Cu film formation.27 This observation confirms that, as expected, under seeded deposition conditions there are more Cu2+ carboxylate complexes on the surface, which facilitates an increase in the Cu ELD rate. Further, the increase in the number of nucleation sites for Cu film deposition also improves the film morphology. Seeding also increases the amount of Cu deposited on CH3 terminated SAMs (see the Supporting Information); after seeded deposition in the SIMS spectra, a high intensity of Cu+ ions is observed. The deposited copper only weakly adhered to the CH3 terminated SAM surface and was easily removed by sonication in deionized water for 3 min (see the Supporting Information). This confirms that there is no specific interaction between the Cu2+ ions trapped on the surface (most likely at defects) and the CH3 terminal groups, in agreement with Lu and Walker.27 While seeded deposition increased the amount of Cu deposited, it did not prevent copper penetration. In the TOF SIMS spectra, Cu-, S-, and Au-containing ions such as CuSH2+ were observed, indicating that copper has penetrated through both COOH and CH3 terminated SAMs to the Au/S interface (Figure 4a, “seeded” and the Supporting Information). Further, under seeded and unseeded deposition conditions for COOH terminated SAMs, the intensity of CuSH2+ is approximately the same, indicating that the total amounts of Cu that has penetrated through the monolayer are comparable (Figure 4a). Because more copper is deposited under seeded ELD conditions, this suggests that the fraction of Cu penetrating to the substrate is less, in agreement with previous studies using PVD.23 3.2. Cu ELD Using Additives on Functionalized SAMs. The second method considered by which to increase deposition rates is the use of the organic additives, adenine and guanine.31 Here, the samples were immersed in the seeding solution and the additive, either adenine or guanine, for 15 min prior to the addition of formaldehyde to initiate Cu ELD. After Cu ELD with adenine on COOH terminated SAMs, the intensity of ions characteristic of copper penetration, such as CuSH2+, was zero (Figure 4a, “adenine”). Using guanine as an additive, the CuSH2+ ion intensity

Figure 4. High-resolution positive ion TOF SIMS spectra centered at (a) m/z 97 and (b) m/z 63 of COOH terminated SAMs after Cu ELD under three different protocols: unseeded, seeded, and seeded with adenine added to the bath (“adenine”). The intensities of 63CuSH2+ are indicative of the amount of copper penetration through the SAM, while the intensities of 63Cu+ are indicative of the amount of copper deposited. Shown for reference is the TOF SIMS spectrum of the COOH terminated SAM prior to Cu ELD (“bare”). To make clear the changes in the mass spectra upon Cu ELD, the ion intensities in (a) are normalized to the intensity of C7H13+.

was approximately one-third of that observed under seeded conditions (see the Supporting Information). Taken together, these data indicate that these additives significantly reduce or eliminate copper penetration through COOH terminated SAMs. However, for both organic additives, in the SIMS spectra, the Cux+ ion intensities are significantly reduced, indicating that the amount of copper deposited is less than under seeded conditions (Figure 4b, “adenine” and the Supporting Information). Further, SEM images indicate that the Cu layer is composed of a dense layer of small crystallites with sizes similar to those observed under seeded conditions without the additive (data not shown). Both adenine and guanine are known to form complexes with Cu2+39 43 and/or carboxylates44 46 under a wide range of pH values. Thus, there are two plausible mechanisms by which the bath additives adenine and guanine mitigate copper penetration through the SAM layer. First, the additives form complexes with Cu2+ in solution, which do not penetrate through the organic layer. Second, these additives form weak complexes with the surface carboxylate groups, which mediate the interaction of Cu2+ with the organic layer and the subsequent reduction of Cu2+ ions to Cu(0). To test whether the formation of Cu2+ additive complexes was responsible for the decrease in copper penetration, the following experiment was performed. To form Cu2+ additive 13025

dx.doi.org/10.1021/la202839z |Langmuir 2011, 27, 13022–13028

Langmuir

ARTICLE

Figure 5. High-resolution positive ion TOF SIMS spectra centered at m/z 97 of COOH terminated SAMs after Cu ELD. Two different deposition conditions were employed. “Adenine” indicates that the deposition was performed under standard conditions as in Figure 4. “Cu2+/adenine” indicates that adenine complexed with Cu2+ ions in the seeding solution for 15 min prior to simultaneous addition of the sample and the reducing agent, formaldehyde. Ion intensities are normalized to the intensity of C7H13+.

Figure 7. High-resolution positive ion mass spectra centered at (a) m/z 163 and (b) m/z 97 for COOH terminated SAMs after 1 h ELD at 45 C. Two different deposition conditions were employed. The labels “adenine” indicate that the deposition was performed under standard conditions as in Figure 4. “Adenine/H2CO” indicates that the adenine was added to the plating solution. The ion intensity of 63CuCOO(CH2)4+ is indicative of the number of Cu2+ carboxylate complexes formed, while the ion intensity of 63CuSH2+ is indicative of the amount of copper penetration through the SAM. Ion intensities in (b) are normalized to the intensity of C7H13+. Figure 6. High-resolution positive ion TOF SIMS spectra centered at m/z 568.5 of COOH terminated SAMs after immersion in the seeding solution containing adenine for 15 min at 45 C. The label “adenine” indicates that the spectra were obtained immediately after immersion in the seeding solution with adenine, while the label “sonication” indicates that the sample was sonicated for 3 min in deionized water. The relative ion intensities of adenine (COO)x(CH2)y(CH)z+ indicate the amount of adenine carboxylate complexes formed. Also shown is the TOF SIMS spectrum of the COOH terminated SAM (“bare”) for reference.

complexes, either adenine or guanine was added to the seeding solution and mixed for 15 min. The SAM and formaldehyde were then simultaneously added to the bath, and Cu ELD was performed for 1 h. In the SIMS spectra of the resulting sample surface (Figure 5, “Cu2+/adenine”), the ion intensity of CuSH2+, characteristic of Cu penetration to the Au/S interface, is larger than under standard deposition conditions in which the sample is immersed for 15 min in the seeding solution containing adenine prior to the addition of the reducing agent, formaldehyde, to initiate Cu depostion (Figure 5, “adenine”). Similar results are obtained with guanine (see the Supporting Information). We conclude that the formation of Cu2+ additive complexes in solution prior to ELD actually increases the amount of Cu penetration through the monolayer! The reason for this behavior is unclear at present and is being investigated further.

It appears instead that the formation of adenine or guanine carboxylate complexes at the SAM/solution interface prevents Cu penetration. After 15 min immersion in the seeding solution containing adenine, Cu(COO)x(CH2)y(CH)z+ and adenine (COO)x(CH2)y(CH)z+ ions are observed in the SIMS spectra, indicating that both Cu carboxylate (data not shown) and adenine carboxylate complexes are present on the COOH terminated SAM surface (Figure 6, “adenine”). No ions characteristic of additive penetration to the Au/S interface, such as (Au adenine)(, are observed, indicating that adenine only interacts with the carboxylate terminal group. Similar ions are observed using guanine but at a much lower intensity, suggesting that guanine carboxylate complexes do not form as readily (see the Supporting Information). These complexes are fragile, and the additive can be removed by sonication in deionized water for 3 min (Figure 6, “sonication” and the Supporting Information). We do not observe Cu2+ additive carboxylate complexes in the SIMS spectra. However, metallo nucleobase complexes are known to form in solution,45 48 and so we cannot rule out their presence on the surface. After Cu ELD, there is no evidence in the SIMS spectra that adenine or guanine are present in the deposited layer: no additive carboxylate complexes are observed (data not shown). To further test whether additive carboxylate complexes help prevent Cu penetration through the organic layer, the additive was added to the bath simultaneously with formaldehyde, to 13026

dx.doi.org/10.1021/la202839z |Langmuir 2011, 27, 13022–13028

Langmuir

ARTICLE

diameter). The SIMS data (Figure 8d and e) confirm these observations: 63Cu+ (m/z 63) and OH (m/z 17; a characteristic ion of COOH terminated SAMs) ions are only observed in the “bar” areas. The SIMS data also show that the CH3 terminated SAM (HDT) is unchanged by the ELD process. High intensities of molecular, cluster, and fragment ions characteristic of HDT, such as AuHDT2 (m/z 651), are observed in the “square” areas (Figure 8f).

Figure 8. Optical (a), SEM (b and c), and SIMS (d f) images after Cu ELD at 45 C for 30 min on a photopatterned CH3/ COOH terminated SAM. The ion OH (m/z 17) is characteristic of COOH terminated SAMs, while AuHDT2 (m/z 651, HDT = S(CH2)15CH3) is characteristic of CH3 terminated SAMs. Copper is only found on the COOH terminated SAM (“bar” areas). TOF SIMS analysis: area of analysis = (100  100) μm2 (128 pixels  128 pixels). Ion intensities are shown using a heat scale, and the scale bars display the maximum and minimum number of counts per pixel.

minimize the time for additive surface complex formation. In the resulting SIMS spectra, the ion intensities characteristic of additive surface complexes significantly decrease, while the intensities of CuSH2+ and other ions characteristic of copper penetration increase (Figure 7 and the Supporting Information). The presence of these surface complexes is clearly linked to the formation of the Cu film without penetration through the organic film. It is not clear why adenine forms more complexes at the COOH SAM surface than guanine. Under the reaction conditions employed, pH 12.8, both adenine and guanine are deprotonated (pKa2 = 9.8 and 9.2 for adenine and guanine, respectively49) as well as the COOH terminal group.27 Adenine has a charge of 1 (pKa2 = 9.849), while guanine has a charge of 2 (pKa2 = 9.8 and pKa3 = 12.349). Both adenine and guanine are negatively charged and should be repelled by the negatively charged carboxylate surface. However, guanine has a charge of 2 and so is repelled more than adenine from the surface, leading to fewer guanine carboxylate complexes present on the surface. 3.3. Selective Deposition of ELD Cu on Patterned SAMs. The above data suggest that copper can be selectively deposited atop the COOH areas of patterned CH3/ COOH terminated SAMs. Using a bath temperature of 45 C, seeding, and the additive adenine, copper is deposited on top on the COOH terminated SAM to form a densely packed layer of submicrometer crystallites. Copper is also deposited in the CH3 terminated SAM areas but in a different form that can be easily removed using sonication. Figure 8 displays optical and scanning electron microscopy (SEM) and SIMS images after 30 min ELD under these conditions. The sample was sonicated in deionized water for 3 min after deposition. Clearly copper has only deposited in the COOH terminated SAM areas (“bar” areas). The SEM images (Figure 8b and c) show that the Cu layer is composed of a dense layer of small crystallites (∼180 nm in

4. CONCLUSIONS These findings underscore the important role that surface complexes play in ELD. Previous studies have shown that upon immersion in the deposition bath, Cu2+ ions form weak complexes with the surface carboxylate groups, which serve as the nucleation sites for the formation of stable metallic Cu overlayers.27 Here, we have demonstrated that by increasing the number of these Cu2+ carboxylate complexes by self-seeding, the amount of copper deposition is significantly increased, and the copper forms an even adherent coating of submicrometer crystallites. In contrast, on CH3 terminated SAMs, there is no specific interaction between the terminal group and Cu2+ ions, and so there are few nucleation sites for Cu overlayer formation, most likely surface defect sites.27 Under seeded conditions at 45 C, copper is deposited on CH3 terminated SAMs, but this layer is only weakly bound to the surface and can be easily removed. It is also clear that the formation of additive carboxylate complexes mediates Cu ELD. The addition of guanine to the bath significantly reduces copper penetration through the layer, while the addition of adenine stops copper penetration altogether. These complexes also appear to reduce the number of Cu2+ carboxylate complexes formed and the rate of Cu deposition. Thus, it appears that the concentration of additive carboxylate complexes present on the surface is a critical parameter in the deposition process. If there are too few complexes, as is the case with guanine, copper penetration is reduced but not stopped. If too many complexes are formed, the copper deposition process may be significantly slowed because the formation of Cu2+ carboxylate complexes, which serve as nuclei for the copper overlayer, is impeded. The addition of 0.04 M adenine to the bath appears to give an appropriate balance of adenine and Cu2+ carboxylate complexes at the surface, which leads to the formation of a strongly adherent copper overlayer without metal penetration. In summary, these studies indicate that ELD can be employed to selectively deposit copper overlayers on organic thin films such as polymers and self-assembled monolayers (SAMs) and that the reaction conditions and surface chemistry can be employed to control the morphology of the deposit so as to result in a technologically viable construct. The formation of Cu2+ surface complexes, such as Cu2+ carboxylate, leads to increases in deposition rates, improved film adhesion, and more uniform film morphology. The formation of additive surface complexes mediates copper deposition reducing the deposition rate and prevents penetration of the deposited metal through the organic layer. Because this Cu ELD method is selective, patterned SAMs can be employed as templates for the deposition of large area films for applications in organic/molecular electronics, optoelectronics, and biotechnology. 13027

dx.doi.org/10.1021/la202839z |Langmuir 2011, 27, 13022–13028

Langmuir

’ ASSOCIATED CONTENT

bS

Supporting Information. High-resolution positive ion SIMS spectra centered at m/z 63 of COOH and CH3 terminated SAMs after Cu ELD before and after sonication for 3 min in deionized water; high-resolution positive ion SIMS spectra centered at m/z 97 of CH3 terminated SAMs after Cu ELD under seeded conditions; structures of adenine and guanine; positive ion SIMS spectra of COOH terminated SAMs centered at m/z 63 and m/z 97 under unseeded, seeded, and seeded conditions with guanine added; positive ion SIMS spectra characteristic ions of Cu2+ carboxylate complex formation under seeded conditions with adenine and guanine; high-resolution SIMS spectra centered at m/z 97 of COOH terminated SAMs after Cu ELD when guanine first complexes with Cu2+ ions in solution; SIMS spectra of ions characteristic of guanine carboxylate complex formation; high-resolution SIMS spectra centered at m/z 97 and m/z 163 of COOH terminated SAMs after Cu ELD when guanine was added to the plating solution; and positive ion mass spectra (m/z 2 500) after Cu ELD for 1 h on CH3 and COOH terminated SAMs at 22 and 45 C, and under the following conditions: unseeded, seeded, and seeded with the additives adenine and guanine. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*Phone: (972) 883-5780. Fax: (972) 883-5725. E-mail: amy.walker@ utdallas.edu.

’ ACKNOWLEDGMENT This work was supported by a National Science Foundation Grant (CHE-0847937). ’ REFERENCES (1) Friend, R.; Burroughes, J.; Shimoda, T. Phys. World 1999, 35–40. (2) Friend, R. H.; Gymer, R. W.; Holmes, A. B.; Burroughes, J. H.; Marks, R. N.; Taliani, C.; Bradley, D. D. C.; Dos Santos, D. A.; Bredas, J. L.; L€ogdlund, M.; Salaneck, W. R. Nature 1999, 397, 121–128. (3) Molecular Nanoelectronics; Reed, M. A., Lee, T., Eds.; American Scientific Publishers: Stevenson Ranch, CA, 2003. (4) Geetha, M.; Singh, A. K.; Asokamani, R.; Gogia, A. K. Prog. Surf. Sci. 2009, 54, 397–425. (5) Haick, H.; Cahen, D. Prog. Surf. Sci. 2008, 83, 217–261. (6) Akkerman, H. B.; de Boer, B. J. Phys.: Condens. Matter 2008, 20, 013001. (7) Fisher, G. L.; Hooper, A. E.; Opila, R. L.; Allara, D. L.; Winograd, N. J. Phys. Chem. B 2000, 104, 3267–3273. (8) Hooper, A.; Fisher, G. L.; Konstadinidis, K.; Jung, D.; Nguyen, H.; Opila, R.; Collins, R. W.; Winograd, N.; Allara, D. L. J. Am. Chem. Soc. 1999, 121, 8052–8064. (9) Nagy, G.; Walker, A. V. J. Phys. Chem. B 2006, 110, 12543– 12554. (10) Nagy, G.; Walker, A. V. J. Phys. Chem. C 2007, 111, 8543–8556. (11) Walker, A. V.; Tighe, T. B.; Cabarcos, O. M.; Reinard, M. D.; Haynie, B. C.; Uppili, S.; Winograd, N.; Allara, D. L. J. Am. Chem. Soc. 2004, 126, 3954–3963. (12) Walker, A. V.; Tighe, T. B.; Haynie, B. C.; Uppili, S.; Allara, D. L.; Winograd, N. J. Phys. Chem. B 2005, 109, 11263–11272. (13) Vilan, A.; Cahen, D. Adv. Funct. Mater. 2002, 12, 795–807. (14) Weckenmann, U.; Mittler, S.; Kr€amer, S.; Aliganga, A. K. A.; Fischer, R. A. Chem. Mater. 2004, 16, 621–628.

ARTICLE

(15) Winter, C.; Weckenmann, U.; Fischer, R. A.; K€ashammer, J.; Scheumann, V.; Mittler, S. Chem. Vap. Deposition 2000, 6, 199–205. (16) Wohlfart, P.; Weiβ, J.; K€ashammer, J.; Winter, C.; Scheumann, V.; Fischer, R. A.; Mittler-Neher, S. Thin Solid Films 1999, 340, 274–279. (17) Lu, P.; Demirkan, K.; Opila, R. L.; Walker, A. V. J. Phys. Chem. C 2007, 112, 2091–2098. (18) Seitz, O.; Dai, M.; Aguirre-Tostado, F. S.; Wallace, R. M.; Chabal, Y. J. J. Am. Chem. Soc. 2009, 131, 18159–18167. (19) Paunovic, M. In Modern Electroplating, 4th ed.; Schlesinger, M., Paunovic, M., Eds.; John Wiley & Sons: New York, 2000; pp 645 665. (20) Adriacacos, P. C. Interface 1999, 8, 32–37. (21) Zaporojtchenko, V.; Strunksus, T.; Behnke, K.; von Bechtolsheim, C.; Kiene, M.; Faupel, F. J. Adhes. Sci. Technol. 2000, 14, 467–490. (22) Opila, R. L.; Eng, J., Jr. Prog. Surf. Sci. 2002, 69, 125–163. (23) Dake, L. S.; King, D. E.; Czanderna, A. W. Solid State Sci. 2002, 2, 781–789. (24) Modern Electroplating, 4th ed.; Schlesinger, M., Paunovic, M., Eds.; John Wiley & Sons: New York, 2000. (25) Zangmeister, C. D.; van Zee, R. D. Langmuir 2003, 19, 8065– 8068. (26) Hozumi, A.; Asakura, S.; Fuwa, A.; Shirahata, N.; Kameyama, T. Langmuir 2005, 21, 8234–8242. (27) Lu, P.; Walker, A. V. Langmuir 2007, 23, 12577–12582. (28) Garno, J. C.; Zangmeister, C. D.; Batteas, J. D. Langmuir 2007, 23, 7874–7879. (29) Czanderna, A. W.; King, D. E.; Spaulding, D. J. Vac. Sci. Technol., A 1991, 9, 2607–2613. (30) Shriver, D. F.; Atkins, P. W.; Langford, C. H. Inorganic Chemistry; Oxford University Press: Oxford, 1992. (31) Paunovic, M.; Arndt, R. J. Electrochem. Soc. 1983, 130, 794–799. (32) Schoenberg, L. N. J. Electrochem. Soc. 1972, 119, 1491–1493. (33) Lin, Y.-M.; Yen, S.-C. Appl. Surf. Sci. 2001, 178, 116–126. (34) Allara, D. L.; Nuzzo, R. G. Langmuir 1985, 1, 45–52. (35) Fisher, G. L.; Hooper, A. E.; Opila, R. L.; Jung, D. R.; Allara, D. L.; Winograd, N. J. Electron Spectrosc. Relat. Phenom. 1998, 98 99, 139–148. (36) Nuzzo, R. G.; Allara, D. L. J. Am. Chem. Soc. 1983, 105, 4481– 4483. (37) Nuzzo, R. G.; Dubios, L. H.; L., A. D. J. Am. Chem. Soc. 1990, 112, 558–569. (38) ToF-SIMS: Surface Analysis by Mass Spectrometry; Vickerman, J. C., Briggs, D., Eds.; IM Publications and Surface Spectra Ltd.: Chichester and Manchester, UK, 2001. (39) Masoud, M. S.; Soayed, A. A.; Ali, A. E. Spectrochim. Acta, Part A 2004, 60, 1907–1915. (40) Harkins, T. R.; Freiser, H. J. Am. Chem. Soc. 1958, 80, 1132–135. (41) Pavelka, M.; Shukla, M. K.; Leszczynski, J.; Burda, J. V. J. Phys. Chem. A 2007, 112, 256–267. (42) Farias, P. A. M.; de Luca Rebello Wagener, A.; Castro, A. A. Talanta 2001, 55, 281–290. (43) Sletten, E. Chem. Commun. 1967, 1119–1120. (44) Lancelot, G.; Helene, C. Proc. Natl. Acad. Sci. U.S.A. 1977, 74, 4872–4875. (45) Ilavarasi, R.; Rao, M. N. S.; Udupa, M. R. Proc.-Indian Acad. Sci., Chem. Sci. 1997, 109, 79–87. (46) Samijlenko, S. P.; Krechkivska, O. M.; Kosach, D. A.; Hovorun, D. M. J. Mol. Struct. 2004, 708, 97–104. (47) A~ norbe, M. A.; L€uth, M. S.; Roitzsch, M.; Cerda, M. M.; Lax, P.; Kampf, G.; Sigel, H.; Lippert, B. Chem.-Eur. J. 2004, 10, 1046–1057. (48) Mukherjea, K. K.; Bhaduri, I. Transition Met. Chem. 2002, 27, 22–26. (49) CRC Handbook of Chemistry and Physics, 84th ed.; Lide, D. R., Ed.; CRC Press: Boca Raton, FL, 2003.

13028

dx.doi.org/10.1021/la202839z |Langmuir 2011, 27, 13022–13028