Article pubs.acs.org/Langmuir
Selective Adsorption of Thiols Using Gold Nanoparticles Supported on Metal Oxides Ruohong Sui,† Kevin L. Lesage,† Sarah K. Carefoot,† Tobias Fürstenhaupt,‡ Chelsea J. Rose,† and Robert A. Marriott*,† †
Department of Chemistry, University of Calgary, 2500 University Drive, Northwest, Calgary, Alberta, Canada T2L 1N4 Health Science Center, University of Calgary, 3330 Hospital Drive, Northwest, Calgary, Alberta, Canada T2N 4N1
‡
S Supporting Information *
ABSTRACT: Selective capture of thiols from a synthetic hydrogen sulfide containing mixture using supported nanogold materials has been explored for the potential removal of thiols from sour gas production fluids. In this research, TiO2-, Al2O3-, SiO2-, and ZnO-supported gold nanoparticles have been studied for their usage as regeneratable adsorbents to capture CH3SH, C2H5SH, and i-C3H7SH. Au/TiO2 and Au/Al2O3 showed promising properties for removing the thiols efficiently from a gas-phase mixture; however, Au/Al2O3 did catalyze some undesirable side reactions, e.g., carbonyl sulfide formation. It was found that a mild temperature of T = 200 °C was sufficient for regeneration of either Au/TiO2 or Au/ Al2O3 adsorbent. The metal oxide mesopores played an important role for accommodating gold particles and chemisorption of the thiols, where smaller pore sizes were found to inhibit the agglomeration/growth of gold particles. The nature of thiol adsorption and the impact of multiple adsorption−desorption cycles on the adsorbents have been studied using electron microscopy, XPS, XRD, GC, and physi/chemiadsorption analyses.
■
INTRODUCTION The interaction of thiols with gold has been extensively studied in relation to self-assembled monolayer (SAM) techniques1,2 and the stabilization of gold colloidal particles.3,4 The current targeted applications for these materials include molecular electronic devices,5 drug delivery,6 sensors,7 and catalysts.8 Despite the tremendous number of experimental and theoretical reports related to the SAM of thiol on gold in the past two decades, relatively little research and development has been documented for using supported gold for selective adsorption of thiols for separation purposes. As most SAM studies are conducted in the liquid phase,1 mainly due to the low vapor pressures of the adsorbates, knowledge regarding gas-phase adsorption/desorption of thiols is limited.9−12 Separation of thiols is of interest for many applications in the fields of biological and analytical chemistry,16 power generation, petroleum industry,17 and natural gas production.18 Our group is specifically interested in improving the processing of sour gas, where H2S and CO2 are separated from methane and other valuable low carbon natural gas components before being transported for sales. Traditional sour fluid treating involves contact with aqueous amines of various types, where H2S (and/ or CO2) is removed from a high-pressure gas stream by preferential dissolution. When thiols are present in these sour fluids, gas processors can choose to (i) increase amine circulation rates (higher energy), (ii) select from a limited number of hybrid amines (e.g., HySWEET or UCARSOL), or © XXXX American Chemical Society
(iii) allow thiol to slip through the amine contactor and then remove them with a downstream adsorbent. In the first two cases there remains some slipping of thiols and/or other species that are removed beyond design specifications. An alternative process could be implemented upstream of the amine if one could selectively remove thiol but not H2S and CO2. Removal of thiol upstream of the amine contactor allows the processor to protect the amine from other thiol-related complications, such as foaming or degradation caused by the increased circulation. Thus, our goal was to find materials which could selectively adsorb thiols and allow higher concentrations of H2S to flow through the adsorbent. These high-pressure fluids are also saturated with water; therefore, highly hydroscopic adsorbents are less desirable. Aside from this application example, an adsorbent which is selective for thiol could potentially be used to clean up other thiol-contaminated streams, such as biogases, oil upgrading off-gases, or liquid natural gas feed streams. It is currently known that the interactions of thiols with gold can be through either strong covalent bonds or weaker coordination bonds.2 In the case of covalent bond formation, the sulfhydryl group of thiols is deprotonated during the adsorption process and the formed Au−thiolate bond is Received: July 6, 2016 Revised: August 19, 2016
A
DOI: 10.1021/acs.langmuir.6b02497 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir
Figure 1. Schematic for the measurement of selective thiol adsorption and thermal swing regeneration (closed cycle). (A) Sour gas mixture; (B) gas regulator; (C) mass flow controller; (D) 10-way GC valve; (E) adsorbent bed; (F) heating block with a controlled temperature 333 K; (G) bypass valve; (H) GC sample loop; (I) thiol perm tubes; (J) KOH trap. and i-C3H7SH were added to flowing streams using permeation tubes, purchased from Kin Tek Corp. (USA). Breakthrough Experiments. In a typical breakthrough experiment, the gold/metal oxide adsorbent was loaded in a stainless steel column, where the 5 μm stainless steel frits (Swagelok) were used to contain the adsorbent powder. As illustrated in Figure 1, a gas mixture of 2.0% H2S, 4.5% CO2 balanced with CH4 was delivered by means of a mass flow controller with the flow rate of 5 Scm3 min−1. When the 10-way valve was in the adsorption mode, the gas mixture flows through the thiol permeation tube chamber, delivering 112 ppm of CH3SH, 98 ppm of C2H5SH, and 49 ppm i-C3H7SH. The thiolcontaining fluid then passed through the adsorbent bed at room temperature (295 K) before flowing through the GC injection loop, which was set to inject onto the column every 20 min. The GC effluent then passed through a KOH trap to remove H2S and thiols before being vented to the exhaust plenum. With the 10-way valve in the regeneration mode, the H2S/CO2-containing gas mixture first passed through the adsorbent vessel (without picking up thiol) that was heated to the regeneration conditions in order to remove the adsorbed thiols. Note that in the regeneration mode, effluent from the GC injection valve passed through the permeation tube chamber in order to keep thiols from building up. A bypass (controlled by valve G) of the adsorption bed was used to circumvent the adsorbent bed in order to measure the feed gas compositions. Material Characterization. The BET surface area and mesopore size distribution of the adsorbents were determined by N2 adsorption at 77 K on a 3Flex Surface Characterization Analyzer (Micromeritics). The hydrogen adsorption isotherms also were collected using the Micrometeritics equipment at 308 and 373 K. The samples were pretreated with H2 purging at 473 K for 2 h before the first H2 isotherm measurement. The second H2 isotherms were measured after 2 h of evacuation at the measurement temperature to remove the weakly adsorbed hydrogen. The total adsorbed amount of hydrogen was obtained from the isotherms at 100 KPa. The amount of strongly adsorbed hydrogen was determined by the difference between the first and the second isotherms. The high-resolution transmission electron microscopy (HRTEM) and scanning transmission electron microscopy (STEM) images were obtained using an FEI Tecnai F20 operated at 200 kV. The samples were ground to a fine powder before
believed to be strong enough to prevent low-temperature thermal decomposition. As a result, exotic methods have been used for removing thiols from the gold surface, e.g., electrochemical desorption,13 exchange displacement,14 and photooxidation.15 For gas-phase adsorption, temperature swing regeneration would be the preferable option. On the other hand, coordination bond formation does not involve deprotonation of thiols, and the coordination bonds may make it feasible for mild thermal desorption of thiols for the purpose of regeneration of gold surface. In this research, selective adsorption/desorption is conducted using metal oxide-supported gold nanoparticles, because the latter provides a high surface area per gold mass unit. It is recognized that the metal oxide support plays an important role for the performance of these materials.19,20 For comparison, three widely studied metal oxides were tested as the gold support: TiO2, Al2O3, and ZnO. A fourth SiO2 support was investigated but not reported here due to poor results. A synthetic natural gas mixture of CH3SH, C2H5SH, i-C3H7SH, H2S, CO2, and CH4 was used to examine the selectivity of the adsorbents by means of a custom breakthrough apparatus. The breakthrough gas composition was analyzed using gas chromatography (GC) equipped with a thermal conductivity detector (TCD) and pulsed flame photometric detector (PFPD). The adsorbents were characterized using electron microscopy, X-ray photoelectron spectroscopy (XPS), X-ray diffraction (XRD), N2 physisorption, and H2 chemisorption.
■
EXPERIMENTAL METHODS
Safety Notes. Because the breakthrough experiments described here involved toxic H2S gas, an H2S detector-equipped cabinet and a KOH caustic scrubbing system were used for safety purposes. Materials. All gases were obtained from Praxair (Canada) with a minimum purity of 99.999%; 1% Au/TiO2, 1% Au/Al2O3, and 1% Au/ ZnO were purchased from Strem Chemicals (USA). CH3SH, C2H5SH, B
DOI: 10.1021/acs.langmuir.6b02497 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir
Figure 2. Gas-phase concentrations for three thiols in a synthetic sour gas stream for three adsorption and desorption cycles. Left axis shows the concentrations for CH3SH (blue), C2H5SH (red), and i-C3H7SH (green) analyzed by PFPD below 50 ppm and TCD above 50 ppm. Right axis shows the concentrations for H2S (purple) and CO2 (orange) (TCD). being placed on a nickel grid covered with Formvar film. XRD was analyzed using a Rigaku Ultima IV diffractometer at a speed of 2° min−1 with a step size of 0.02°. XPS analysis was carried out using a Kratos Axis Ultra spectrometer. The binding energy was calibrated with C 1s at 285.0 eV. For H2 detection, an SRI 8610 GC equipped with a Molsieve 5A column and a TCD detector was used (argon carrier gas at 20 mL min−1).
After the breakthrough of i-C3H7SH for each adsorption stage shown in Figure 2, the adsorbent bed was regenerated at 573 K. While we expected some chemisorption of H2S and CO2 on TiO222 the concentrations for these species did not change large appreciably, i.e., the support surface remained saturated. The small apparent concentration changes of H2S (ca. 5.0%) and CO2 (ca. 2.5%) were due to a small system pressure change. Here there was a slight back-pressure change on the GC injection loop between each regeneration and adsorption mode, because we needed to reroute the effluent through the permeation chamber. After each regeneration process, the feed gas mixture was introduced first through the bypass for analyzing the feed concentrations and then passed through the adsorption bed to start the next cycle. Different regeneration temperatures were tested for Au/TiO2 in order to determine the acceptable regeneration conditions. In these early studies the adsorbent was not replaced between regeneration tests. As expected and shown in Table 1, regeneration at the highest temperature (673 K) gave rise to a longer subsequent breakthrough time (higher adsorption capacity due to less residual adsorption), but the low regeneration temperature at 373 K was still able to provide a reasonable adsorption capacity. Even though a higher temperature was more effective at regenerating active sites for the next cycle of adsorption, it can also increase the rate of sintering for the adsorbent and shorten adsorbent lifetime. It has been reported that gold was dissolved in H2S gas at the parts per billion level when temperature was raised above 673 K, suggesting that some interaction with H2S at high temperature could aid gold surface mobility.23 In addition, with consideration of close-loop adsorption, a higher regeneration temperature increases energy consumption and the capital cost for the high-temperature facilities. When Au/Al2O3 and Au/ZnO were tested, 473 K was arbitrarily selected for a mild regeneration temperature (this also corresponds closely to the recoverable steam heat within a processing plant). The latter adsorbents were less promising than Au/TiO2, hence the extra experiments for Au/TiO2. From Table 1 the conclusion can be made that Au/Al2O3 had a
■
RESULTS Breakthrough Experiments. The purpose of the breakthrough tests was 2-fold: (1) to determine which adsorbent has a higher capacity for thiol adsorption and (2) to investigate conditions adequate for thiol desorption. These experiments are discussed by referring to the online GC analysis results with the 10-way valve in either adsorption or desorption mode as illustrated in Figure 1. The flow rate for the sour gas was kept at 5 Scm3 min−1 during all experiments. Figure 2 shows the gas concentration profiles during three adsorption−desorption cycles when Au/TiO2 was loaded as the adsorbent. During the adsorption stage, the thiols were removed from the gas stream by the adsorbent bed, where C 2 H 5 SH > CH 3 SH (corresponding to the thiol vapor pressure). The competitive adsorption of CO2 and H2S on metal oxides and H2 on gold was found to be insignificant and did not affect the thiol adsorptions. In fact, regeneration using an H2S-containing sour fluid helped to regenerate the adsorbents further. Third, electron microscope and XRD analysis results showed that H
DOI: 10.1021/acs.langmuir.6b02497 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir (9) Pensa, E.; Carro, P.; Rubert, A. A.; Benítez, G.; Vericat, C.; Salvarezza, R. C. Thiol with an Unusual Adsorption-Desorption Behavior: 6-Mercaptopurine on Au(111). Langmuir 2010, 26, 17068− 17074. (10) Thomas, R. C.; Sun, L.; Crooks, R. M.; Ricco, A. J. Real-time Measurements of the Gas-phase Adsorption of n-alkylthiol Mono- and Multilayers on Gold. Langmuir 1991, 7, 620−622. (11) Dishner, M. H.; Hemminger, J. C.; Feher, F. J. Direct Observation of Substrate Influence on Chemisorption of Methanethiol Adsorbed from the Gas Phase onto the Reconstructed Au(111) Surface. Langmuir 1997, 13, 2318−2322. (12) Dubois, H. L.; Zegarski, R. B.; Nuzzo, G. R. Molecular Ordering of Organosulfur Compounds on Au(111) and Au(100): Adsorption from Solution and in Ultrahigh Vacuum. J. Chem. Phys. 1993, 98, 678. (13) Munakata, H.; Oyamatsu, D.; Kuwabata, S. Effects of ωFunctional Groups on pH-Dependent Reductive Desorption of Alkanethiol Self-Assembled Monolayers. Langmuir 2004, 20, 10123− 10128. (14) Collard, D. M.; Fox, M. A. Use of Electroactive Thiols to Study the Formation and Exchange of Alkanethiol Monolayers on Gold. Langmuir 1991, 7, 1192−1197. (15) Huang, J.; Hemminger, J. C. Photooxidation of Thiols in Selfassembled Monolayers on Gold. J. Am. Chem. Soc. 1993, 115, 3342− 3343. (16) O’Shea, T. J.; Lunte, S. M. Selective Detection of Free Thiols by Capillary Electrophoresis-Electrochemistry using a Gold/Mercury Amalgam Microelectrode. Anal. Chem. 1993, 65, 247−250. (17) Mochizuki, Y.; Sugawara, K. Selective Separation of Thiols from a Model Fuel by Metal Oxides. Energy Fuels 2008, 22, 3939−3943. (18) Cai, C.; Worden, R. H.; Bottrell, S. H.; Wang, L.; Yang, C. Thermochemical Sulphate Reduction and the Generation of Hydrogen Sulphide and Thiols (Mercaptans) in Triassic Carbonate Reservoirs from the Sichuan Basin, China. Chem. Geol. 2003, 202, 39−57. (19) Haruta, M. Catalysis of Gold Nanoparticles Deposited on Metal Oxides. CATTECH 2002, 6, 102. (20) Remediakis, I. N.; Lopez, N.; Nørskov, J. K. CO Oxidation on Gold Nanoparticles: Theoretical Studies. Appl. Catal., A 2005, 291, 13−20. (21) Shevade, A. V.; Zhou, J.; Zin, M. T.; Jiang, S. Phase Behavior of Mixed Self-Assembled Monolayers of Alkanethiols on Au(111): A Configurational-Bias Monte Carlo Simulation Study. Langmuir 2001, 17, 7566−7572. (22) Linsebigler, A. L.; Lu, G.; Yates, J. T. Photocatalysis on TiO2 Surfaces: Principles, Mechanisms, and Selected Results. Chem. Rev. 1995, 95, 735−758. (23) Zezin, D. Y.; Migdisov, A. A.; Williams-Jones, A. The Solubility of Gold in Hydrogen Sulfide Gas: An experimental study. Geochim. Cosmochim. Acta 2007, 71, 3070−3081. (24) Thommes, M.; Kaneko, K.; Neimark, A. V.; Olivier, J. P.; Rodriguez-Reinoso, F.; Rouquerol, J.; Sing, K. S. W. Physisorption of Gases, with Special Reference to the Evaluation of Surface Area and Pore Size Distribution. Pure Appl. Chem. 2015, 87 (9−10), 1051− 1069. (25) Bus, E.; Miller, J. T.; van Bokhoven, J. A. Hydrogen Chemisorption on Al2O3-Supported Gold Catalysts. J. Phys. Chem. B 2005, 109, 14581−14587. (26) Bus, E.; van Bokhoven, J. A. Hydrogen Chemisorption on Supported Platinum, Gold, and Platinum-Gold-Alloy Catalysts. Phys. Chem. Chem. Phys. 2007, 9, 2894−2902. (27) Patterson, A. L. The Scherrer Formula for X-Ray Particle Size Determination. Phys. Rev. 1939, 56, 978−982. (28) Zhang, H.; Banfield, J. F. Understanding Polymorphic Phase Transformation Behavior during Growth of Nanocrystalline Aggregates: Insights from TiO2. J. Phys. Chem. B 2000, 104, 3481−3487. (29) Pineda, M.; Palacios, J. M.; Alonso, L.; Garcia, E.; Moliner, R. Performance of Zinc Oxide based Sorbents for Hot Coal Gas Desulfurization in Multicycle Tests in a Fixed-bed Reactor. Fuel 2000, 79, 885−895.
(30) Westmoreland, P. R.; Gibson, J. B.; Harrison, D. P. Comparative Kinetics of High-temperature Reaction between Hydrogen Sulfide and Selected Metal Oxides. Environ. Sci. Technol. 1977, 11, 488−491. (31) Sasaoka, E.; Hirano, S.; Kasaoka, S.; Sakata, Y. Characterization of Reaction between Zinc Oxide and Hydrogen Sulfide. Energy Fuels 1994, 8, 1100−1105. (32) Oeter, D.; Ziegler, C.; Gópel, W.; Haarmann, H. Spectroscopic and Electrical Investigations of Undoped and Doped α-oligothiophene Thin Films. Synth. Met. 1994, 67, 267−271. (33) Kutty, T. R. N. A Controlled Copper-coating Method for the Preparation of ZnS: Mn DC Electroluminescent Powder Phosphors. Mater. Res. Bull. 1991, 26, 399−406. (34) Chen, M.; Goodman, D. W. Catalytically Active Gold: From Nanoparticles to Ultrathin Films. Acc. Chem. Res. 2006, 39, 739−746.
I
DOI: 10.1021/acs.langmuir.6b02497 Langmuir XXXX, XXX, XXX−XXX
