Subscriber access provided by UNIV OF CALIFORNIA SAN DIEGO LIBRARIES
Article
Thermochemical Analysis of Molybdenum Thin Films on Porous Alumina Kyoungjin Lee, Charles-François de Lannoy, Simona Liguori, and Jennifer Wilcox Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.6b04149 • Publication Date (Web): 27 Dec 2016 Downloaded from http://pubs.acs.org on January 9, 2017
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Langmuir is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 28
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
Thermochemical Analysis of Molybdenum Thin Films on Porous Alumina Kyoungjin Lee,‡a, b Charles-François de Lannoy,‡a,c Simona Liguoria, d and Jennifer Wilcox*a, d a
Department of Energy Resources Engineering, Stanford University, 367 Panama Street, Stanford, CA 94305, United States b Applied Materials, 974 E Arques Ave, Sunnyvale, CA 94085, United States c Department of Chemical Engineering, McMaster University, 1280 Main Street West, Hamilton, ON, L8S 4L7, Canada d Department of Chemical and Biological Engineering, Colorado School of Mines, 1613 Illinois Street, Golden, CO,80401 ‡
Kyoungjin Lee and Charles-François de Lannoy contributed equally to the work
*
Corresponding author, email:
[email protected] A Manuscript Submitted to Langmuir
1
ACS Paragon Plus Environment
Langmuir
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 2 of 28
Abstract Molybdenum (Mo) thin films (thickness < 100 nm) were physically deposited by e-beam evaporation on a porous alumina substrate and were analyzed for their stability and reactivity under various thermal and gas conditions. The Mo thin-film composites were stable below 300 ˚C but had no reactivity towards gases. Mo thin films showed nitrogen incorporation on the surface as well as in the subsurface at 450 ˚C, confirmed by X-ray photoelectron spectroscopy. The reactivity toward nitrogen was diminished in the presence of CO2, although no carbon species were detected either on the surface or in the subsurface. The Mo thin films have a very stable native oxide layer, which may further oxidize to higher oxidation states above 500 ˚C due to the reaction with the porous anodized alumina substrate. The oxidation of Mo thin films was accelerated in the presence of oxidizing gases. At 600 ˚C in N2, the Mo thin film on anodized alumina was completely oxidized and may also have been volatilized. The results imply that choosing thermally stable and inactive porous supports and operating in non-oxidizing conditions below 500 ˚C will likely maintain the stability of the Mo composite. This study provides key information about the chemical and structural stability of a Mo thin film on a porous substrate for future membrane applications, and offers further insights into the integrity of thin-film composites when exposed to harsh conditions.
2
ACS Paragon Plus Environment
Page 3 of 28
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
1. Introduction Nitrogen-selective membranes represent one of the most unexplored areas of gas permeable membranes. Currently, no nitrogen-selective membranes are commercially available even though it is often desirable to separate nitrogen gas (N2) from other gases such as methane (CH4), carbon dioxide (CO2), and oxygen (O2). The most common ways to remove N2 from gas mixtures is through cryogenic distillation or pressure swing adsorption, both of which are less cost-effective than membranes for small-scale applications.1-2 A membrane technology that selectively removes N2 from gas mixtures of gases would represent an energetically efficient alternative to conventional processes, with the practical advantages of a smaller footprint and operational simplicity. However, N2 selectivity in membranes is challenging because the molecular size of N2 differs little (by less than 0.5 Å) from the size of other gases of interest such as CH4, O2 and CO2.3-4 Polymeric membranes do not have selectivity for N2 since the solubility/sorptivity of many gases in most polymers is higher than that for N2. For example, methane purification can be performed with polymeric membranes that have been shown to preferentially permeate CH4 over N2, albeit with a CH4/N2 selectivity as low as 4–6.5-6 In addition to this low selectivity, polymeric membranes have low thermal resistance and high susceptibility to a reducing environment.7 Some zeolite-based membranes have shown slightly better selectivity (7–11)8-9 than the polymeric N2-selective membranes, but the cost associated with defect-free synthesis tends to remain high. An alternative to these technologies can be the use of dense metallic membranes. Metallic membranes require high temperature conditions to activate gas dissociation and permeation, as well as to avoid undesirable phase transformation in membrane structures at low temperatures. Although metallic membranes are expensive, they may offer distinct advantages over conventional technologies if they can produce high-value, high-purity products, under harsh conditions that are unsuitable for polymeric or zeolite membranes. Metallic membranes are known primarily for high hydrogen permeability.10 While no attempts have been made to create a nitrogen-permeable dense membrane, nitrogen permeation through dense body-centered cubic (bcc) metals has been reported in the literature.11-12 These studies show that nitrogen permeation through V, Nb, and Mo involves molecular dissociation followed by solution and diffusion of atomic species. In other studies,12-14 the solubility of nitrogen in bulk Mo has been reported. The experimental evidence indicates that nitrogen solubility in Mo is low but not negligible, with a solubility limit of 0.004 at.% at temperature over 800 ˚C. Recent
3
ACS Paragon Plus Environment
Langmuir
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 4 of 28
theoretical studies about the nitrogen dissolution and permeation in metals based on first-principles calculations15-17 have supported those experimental observations. The nitrogen absorption in Mo was revealed to be endothermic, and, thus, requires high thermal energy in the dissolution process, corroborating the low nitrogen solubility values in Mo.17 These theoretical studies showed that initially nitrogen dissociation on Vanadium metal surfaces readily happens, but further atomic nitrogen diffusion through the subsurface and bulk is regarded as the limiting step, resembling the mechanism of H2 permeation through dense Pd-based membranes.18 Molybdenum was chosen in this study as a metal of interest due to its bcc structure and its affinity toward nitrogen. In a previous study,15 vanadium was chosen as a representative bcc Group 5 metal, and was shown to have a very high reactivity for surface dissociation and strong surface atomic binding. However, very strong adsorption on the vanadium surface may impede the subsurface and bulk diffusion of nitrogen. Based on electronic structure theory,19-20 Mo exhibits weaker binding toward nitrogen than vanadium in its bulk form, potentially enhancing nitrogen diffusion into the bulk. In addition, Mo nitrides are known to be used as ammonia synthesis catalysts,21-22 showing favorable surface interaction with nitrogen. Therefore, Mo was chosen as a candidate for the first thin-composite membranes for nitrogen separation. The goal of this study is to investigate the initial steps of nitrogen permeation through Mo-based metallic membranes: namely, N2 adsorption on, catalytic dissociation on, and absorption into thin-film molybdenum (Mo) membranes. Generally, in membrane studies, gas permeability and selectivity are the main criteria for evaluating membrane performance, and an accurate measurement of those properties is critical in the development of new membranes. A defect-free but thin membrane is desirable since the thinner the membranes, the higher the flux through the membranes will be. Since nitrogen permeability through metallic membranes is expected to be much lower than that of hydrogen, making a very thin membrane, i.e., at least 10 times thinner than conventional Pd-based hydrogen-selective membranes, is important for the nitrogen-selective membrane to become feasible. However, making a thin membrane (less than a few micrometers) without defects is extremely challenging. To maintain sufficient strength under pressure, thin membranes usually have asymmetric supported structure, composed of a thin coating on top of a porous support. As a membrane becomes thinner, the pores of the support layer are more likely to be exposed, and, therefore, the membrane becomes more defective. Moreover, even though thin membranes are successfully fabricated, they may undergo severe changes and damages in high-temperature operating conditions. Thus, to enhance the long-term stability of membranes, a better
4
ACS Paragon Plus Environment
Page 5 of 28
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
understanding of degrading mechanisms of membranes in gas permeating conditions is crucial. Before overcoming all those existing challenges, in this manuscript we focused on characterizing the physical and chemical properties of Mo thin films to be used as membranes. Specific gas transport properties will be characterized in future studies with the next generation of dense thin-film membranes. In this study, Mo thin films were deposited by e-beam evaporation on a porous alumina substrate. The fabricated Mo thin-film composites were exposed under various thermal and gas conditions that are relevant to the potential operating conditions of nitrogen-selective membranes. The morphological and chemical changes on the Mo thin-film surfaces were analyzed using scanning electron microscopy (SEM) and X-ray photoelectron spectroscopy (XPS) to identify the properties of Mo in dissociating and dissolving nitrogen. Further, the impact of oxygen on the stability of the Mo thin film was investigated. This study provides key information on the chemical and structural stability of a Mo thin film on a porous substrate for membrane applications, and may offer further insights into the integrity of thin-film composites exposed to harsh conditions.
2. Experimental Methods 2.1 Molybdenum thin-film composite fabrication A commercial porous alumina membrane (AnoporeTM, GE) was used as a substrate for Mo thin-film deposition. Prior to deposition, the porous alumina was cleaned by sonicating the surface gently (100 W, 45kHz) for 30 minutes in a sonicator bath filled with DI water. This was followed by soaking the membrane in acetone for 20 minutes, then rinsing thoroughly in methanol, and finally air drying. Molybdenum was deposited by e-beam evaporation in a custom-made system, using an Airco Temescal e-gun and power supply. The Mo target was 99.999% pure and was placed directly in the bare copper hearth with a base pressure of 7 × 10ି Torr. The voltage of the e-beam was approximately 8600 V. To minimize the spot size of the e-beam, no sweep was applied during the deposition, thereby restricting the radiated heat from the Mo slug. The substrate was fixed on to a thick aluminum plate, helping to draw heat away from the substrate during the deposition. The source-to-substrate distance was reduced to 20 cm during deposition, which allowed for a high deposition rate without causing peeling problems on the walls of the station. A shutter was in place during the initial part of the
5
ACS Paragon Plus Environment
Langmuir
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 6 of 28
sputtering to clean the Mo target and to establish the rate of deposition prior to the actual deposition on the alumina substrate. The thickness was monitored and controlled by an Inficon quartz crystal monitor. The deposition rate started at approximately 2 Å/sec, and was gradually ramped up to approximately 10 Å/sec by manual adjustment. The average rate of deposition was 5 Å/sec. Two different thicknesses of Mo deposition were achieved and compared in this study: 20 nm and 100 nm.
2.2 Thermochemical Exposure Gas and heat exposure tests were performed on the thin films of small pieces (approximately 30 mm2). Each piece was supported by quartz wool in the middle of a glass tube reactor, which was connected to continuous gas flow via Teflon fittings. Gas flows in every experiment were precisely controlled at 300 mL/min by mass flow controllers. The entire system was flushed with a desired gas for 20 minutes before heating. The reactor tube was placed in a tube furnace with a heating zone of approximately 30 cm. The reactor was ramped at a rate of 3 ˚C/min to the target temperature, and the target temperature was maintained at a constant level. After four hours of exposure, the reactor was quenched to room temperature (15–20 ˚C). The samples were not exposed to air until it was cooled down to
