Article pubs.acs.org/Langmuir
Thiol Adsorption on and Reduction of Copper Oxide Particles and Surfaces Yiwen Wang,† Jisun Im,† Jason W. Soares,‡ Diane M. Steeves,‡ and James E. Whitten*,† †
Department of Chemistry, The University of Massachusetts Lowell, Lowell, Massachusetts 01854, United States U.S. Army Natick Soldier Research, Development, and Engineering Center, Natick, Massachusetts 01760, United States
‡
ABSTRACT: The adsorption of 1-dodecanethiol at room temperature and at 75 °C on submicron cuprous and cupric oxide particles suspended in ethanol has been investigated by X-ray photoelectron spectroscopy (XPS), Raman spectroscopy, and transmission electron microscopy. Thiol adsorption occurs in all cases via Cu−S bond formation, with partial dissolution of CuO at 75 °C and formation of a copper−thiolate complex replacement layer. Regardless of temperature, the surface of the CuO particles is essentially completely reduced to either Cu2O or metallic copper, as evidenced by loss of the characteristic Cu2+ XPS features of dried powder samples. Companion ultrahighvacuum studies have been performed by dosing clean, oxygen-dosed, and ozone-treated single crystal Cu(111) with methanethiol (MT) gas at room temperature. In the latter case, the surface corresponds to CuO/Cu(111). XPS confirms MT adsorption in all cases, with an S 2p peak binding energy of 162.9 ± 0.1 eV, consistent with methanethiolate adsorption. Heating of MT-covered Cu(111) and oxygen-dosed Cu(111) leads to decomposition/desorption of the MT by 100 °C and formation of copper sulfide with an S 2p binding energy of 161.8 eV. Dosing CuO/Cu(111) with 50−200 L of MT leads to only partial reduction/removal of the CuO surface layers prior to methanethiolate adsorption. This is confirmed by ultraviolet photoelectron spectroscopy (UPS), which measures the occupied states near the Fermi level. For both the colloidal CuO and single crystal CuO/Cu(111) studies, the reduction of the Cu2+ surface is believed to occur by formation and desorption of the corresponding dithiol prior to thiolate adsorption.
■
INTRODUCTION Surface reactions of metal oxide nanoparticles have applications that include dye-sensitized solar cells,1−3 gas sensors,4−6 heterogeneous catalysts,7,8 and photocatalysts.9 Copper oxide is significant due to the ubiquitous use of copper for electrical wiring and interconnects and the deleterious effects of corrosion. Certain nanomorphologies have been shown to exhibit room temperature ferromagnetism.10 It is also of interest for chemical sensing because the electrical resistance of CuO nanowires changes upon exposure to reducing gases (e.g., hydrogen sulfide).11 Xia et al.12 have shown that CuO nanowires may be synthesized by heating copper wire at 500 °C in the presence of oxygen. The first step is oxidation of copper to form the cuprous oxide (Cu2O) intermediate, from which cupric oxide (CuO) vapor forms, condenses, and crystallizes. The detailed surface chemistry between thiols and oxidized copper has not been extensively studied. However, the adsorption of sulfur-containing molecules such as S2, H2S, CH3SH, and thiophene on various metal oxides has been investigated, and bonding may occur (depending on the particular oxide) either to metal or oxygen sites.13,14 In the case of Cu2O, adsorption is mainly via Cu−S bonding.13 While thiol selfassembled monolayers (SAMs) have most commonly been studied on gold surfaces, they also form on clean copper surfaces and have been demonstrated to impede corrosion.15−19 However, it is known that some transition metal oxides (including cupric oxide) may oxidize thiols to disulfide.20 © XXXX American Chemical Society
In one of the few existing detailed investigations, Carbonell and colleagues21 compared the thermal stability and decomposition mechanisms of alkane and aromatic thiol monolayers on clean and oxidized surfaces. They found that SAM formation on the oxidized copper surfaces removed/reduced the oxide layer, and this led to higher surface roughness and higher SAM coverage compared to clean copper surfaces. In the present study, we have investigated the adsorption of 1-dodecanethiol (DDT) on cupric and cuprous oxide particles. The original motivation for this work was to determine if metal oxide nanoparticles could be encapsulated in a shell of a thiol− metal complex, as we had previously observed in the case of ZnO nanoparticles.22 In that work, it was found that heating an ethanolic mixture of ZnO nanorods and a thiol at 75 °C resulted in partial dissolution of the nanorods and formation of a relatively thick (e.g., 100 Å), encapsulating layer of a 1:2 Zn:thiol complex that surrounded the ZnO. As shown in the present study, the encapsulation phenomenon does not occur in the case of Cu2O or CuO particles at room temperature; however, partial dissolution of CuO particles does occur at 75 °C, as evidenced by electron microscopy and thermal gravimetric analysis. In both cases, DDT adsorption at room temperature and 75 °C is confirmed, along with reduction of Received: February 19, 2016 Revised: April 1, 2016
A
DOI: 10.1021/acs.langmuir.6b00651 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir
Figure 1. Mg Kα XPS of the (a) Cu 2p, (b) O 1s, (c) S 2p, and (d) C 1s regions of as-received and room temperature and 75 °C DDT-treated Cu2O particles. Spectra of controls are also included in which the same treatment was performed but without the DDT present in the solution. The surface of the Cu2O particles contains some CuO phase, and peak fitting of the Cu 2p region has been performed to estimate the percentage of each for the different conditions. suspension of 1 mg/mL of the powder in methanol for 5 min. This suspension was then drop-cast onto a copper TEM grid. Nonmonochromatized Mg Kα X-rays, in an ESCALAB photoelectron spectrometer (see below), were used for XPS analysis. For those experiments, a sample stub with double-sided conductive copper tape on its surface was pressed into the powder such that the entire surface of the tape was covered with a uniform layer of the powder, with excess removed by gently shaking prior to introduction of the stub to the vacuum chamber. In the case of the copper oxide and functionalized copper oxide particles, XPS surface charging (due to uncompensated electron ejection) was accounted for by aligning the C 1s peak for either adventitious or aliphatic carbon to 284.6 eV. The amount of correction needed to align this peak was then used for the other XPS peaks for that sample. Single Crystal Experiments. A polished Cu(111) single crystal (1 cm × 1 cm × 0.5 mm, purchased from MTI, Inc.) was ultrasonicated in methanol and then acetone and mounted on a copper sample stub using vacuum compatible silver paint which can withstand high temperature (Aremco, No. 597-A). It was then transferred into a Vacuum Generators MKII ESCALAB photoelectron spectrometer that consists of an introduction chamber, a sample preparation chamber, and an analysis chamber, all separated by gate valves. The latter two chambers are pumped by liquid-nitrogen-trapped oil diffusion pumps and have base pressures in the 10−10 mbar range. Electron energy
cupric oxide to cuprous oxide. The colloidal particle studies have been complemented by single crystal Cu(111) experiments that demonstrate that methanethiol adsorbs via Cu−S bond formation on clean copper, oxygen-dosed copper, and CuO/Cu(111), with reduction of the top Cu2+ layer in the latter case.
■
EXPERIMENTAL SECTION
Nanoparticle Reactions and Characterization. Cuprous oxide (Cu2O, nominal average particle size
