Research Updates Summer 2021

 

Jump to Section: Center News Research News Education News Diversity News   Industry News Publications    

 

Jump to Research Section: Thrust 1 Thrust 2 Thrust 4 Thrust 6   Testbed

 


Research Feature:

Thrust 3 - Soft Oxidant Coupling of Methane

Allison Arinaga and Tobin J. Marks, Northwestern University

Proposed mechanism of SOCM over FeS2. Although methane is the predominate constituent of natural gas, very little of it is upgraded to value-added products. The oxidative coupling of methane using gaseous disulfur as a soft oxidant (SOCM, 2CH4 + S2 → C2H4 + 2H2S) is an “out of the box” approach for methane utilization that proceeds with promising ethylene selectivity. CISTAR PI Tobin Marks and his student Shanfu Liu recently submitted a manuscript to PNAS reporting detailed experimental and theoretical studies that examine the mechanism for the conversion of CH4 to C2H4 over a cheap, simple Fe3O4-derived catalyst (FeS2).1 Kinetic experiments and density functional theory analysis of “SOCM’ in collaboration with Matthew Neurock and his student Sagar Udyavara of the University of Minnesota show that ethylene is largely produced as a primary product of methane activation via the coupling of CH2 intermediates over Fe-S sites. In contrast, the CS2 overoxidation byproduct forms predominantly via CH4 oxidation, rather than from the C2 products, via a series of C-H activation and S-addition steps at adsorbed sulfur sites on the FeS2 surface. The experimental rate law for methane conversion is 1st order in both CH4 and S2, consistent with the involvement of two S sites in the rate-determining methane C-H activation, with a CD4/CH4 kinetic isotope effect of 1.78±0.18. The experimental apparent activation energy for methane conversion is 66 ± 8 kJ/mol, which is significantly lower than those reported for CH4 oxidative coupling with O2. All evidence argues that SOCM proceeds via a very different pathway than traditional OCM.

 

Graduate students Allison Arinaga (left) and Shanfu Liu (right) in the lab.

Marks, along with graduate student Allison Arinaga and CISTAR REU student Morgan Ziegelski, also compiled a review article published in Angewandte Chemie International Edition covering the past and recent literature on alternative oxidants for the OCM reaction.2 The review, which discusses oxidants such as N2O, CO2, and S2, provides a thorough overview of the field, comparing reaction mechanisms and catalyst designs with the traditional O2-OCM systems. Potential opportunities for future research are also discussed for each oxidant, which is sure to be of interest to the CISTAR community.

 

Additionally, in collaboration with CISTAR Thrust 1, Marks, Shanfu Liu, and Allison Arinaga have now expanded the scope of the S2 oxidant approach to other light alkane oxidation reactions such as oxidative dehydrogenation to olefins (ODH). For ethane ODH with sulfur (SODHE, C2H6 + ½S2 → C2H4 + H2S), was recently published detailing a maximum ethylene yield of 76% over an earthabundant FeSxbased catalyst.3 Yield and selectivity are stable for 50 hours on stream. The SODHE rate law is 1st order in ethane and ½ order in S2, supporting a proposed Mars van Krevelenlike mechanism at temperatures <700°C. Furthermore, conversion and selectivity become insensitive to catalyst identity at temperatures >860°C, suggesting the intrusion of radical pathways. They have also published a manuscript reporting the development of a propane oxidative dehydrogenation system with disulfur (SODHP, C3H8 + ½S2 → C3H6 + H2S).4 They find that different metal sulfide catalysts generate very different  reaction product distributions, and that propylene selectivities as high as 86% can be achieved at 450 - 550°C. For a group of 6 metal sulfide catalysts, the apparent activation energies for propylene formation range from 72-134 kJ/mol and parallel the corresponding catalyst XPS sulfur binding energies, indicating that M-S bond strength plays a key role in SODHP activity. Kinetic data over a sulfided ZrO2 catalyst indicate a rate law which is first-order in propane and zero-order in sulfur, suggesting that SODHP may also occur via a mechanism analogous to the Mars van Krevelen cycle of traditional ODH reactions.

 

1.  Liu, S.; Udyavara, S.; Peter, M.; Lohr, T.L.; Neurock, M.; Marks, T.J. Soft Oxidative Coupling of Methane. An Experimental and Theoretical Mechanistic Investigation, PNAS, 2020, under revision.

2.  Arinaga, A. M.; Ziegelski, M. C.; Marks, T. J. Alternative Oxidants for the Catalytic Oxidative Coupling of Methane. Angew. Chem. Int. Ed. 2020, 6. DOI: 10.1002/anie.202012862

3.  Liu, S.; Arinaga, A.M.; Peter, M.; Lohr, T.L.; Marks, T.J. High ethylene yield oxidative dehydrogenation of ethane using sulfur vapor as a soft oxidant, ChemCatChem. 2020, 12, 4538-4542. DOI: 10.1002/cctc.202000858R1.

4. Arinaga, A.M.; Liu, S.; Marks, T.J. Oxidative Dehydrogenation of Propane to Propene over Transition Metal Sulfides Using Sulfur as an Alternative Oxidant, 2020, 10, 6840-6948. DOI: 10.1039/D0CY01039A

 

   

Thrust 1

Jason Hicks, University of Notre Dame

Thrust 1 researchers continue to identify new, stable catalysts for alkane dehydrogenation reactions.  In a recent study, CISTAR researchers determined that strong metal-support interactions (SMSI) could be exploited to increase the selectivity to alkane dehydrogenation (ZhuChen et al.).  Specifically, SMSI of CeO2 on Pt nanoparticles leads to low rates for ethylene hydrogenation and propane dehydrogenation when reduced at 975 °C, but high olefin selectivity for the latter reaction. By contrast, reduction at the lower temperature of 550 °C for Pt/CeO2, leads to 5 nm Pt nanoparticles with higher ethylene hydrogenation and propane dehydrogenation turnover rates but low olefin selectivity similar to Pt/SiO2. Detailed characterization with X-ray absorption fine structure and scanning transmission electron microscopy was performed on these materials and showed formation of ∼15 nm monometallic Pt nanoparticles partially covered by SMSI CeO2, see Figure 1. The propylene selectivity of Pt nanoparticles covered by SMSI CeO2 was ∼95 % suggesting there were few Pt ensembles capable of catalyzing propane hydrogenolysis, while the remaining exposed Pt remained active for dehydrogenation reactions.

Johnny ZhuChen, Abhijit Talpade, Griffin A. Canning, Paige R. Probus, Fabio H. Ribeiro, Abhaya K. Datye, Jeffrey T. Miller, Strong metal-support interaction (SMSI) of Pt/CeO2 and its effect on propane dehydrogenation, Catalysis Today, 2020,  https://doi.org/10.1016/j.cattod.2020.06.075.

Thrust 1 is also focused on enhancing the stability of catalysts and catalyst supports.  Ceria is known to be a very good catalyst as well as a support for oxygen transfer (oxidation) as well as for hydrogen transfer (hydrogenation and dehydrogenation) reactions. Many of these reactions occur at high temperatures where ceria is known to sinter, leading to loss of surface area. The thermal stability of ceria can be improved by the addition of dopants, but the location of the dopant atoms and the mechanisms by which ceria stabilization occurs are poorly understood.

We show here that dopants located on the surface of ceria are remarkably effective at stabilizing ceria surface area, see Figure 2.  Keeping metal loading constant at 0.88 mol%, we found that surface area of the ceria aged at 800 °C in air for 5 h ranged from 45 m2/g to 2 m2/g. Strongly bound dopants in atomically dispersed form help to pin surface sites and lower the mobility of ceria.

R. Alcala, A. DeLaRiva, E.J. Peterson, A. Benavidez, C.E. Garcia-Vargas, D. Jiang, X.I. Pereira-Hernández, H.H. Brongersma, R.t. Veen, J. Stank, J.T. Miller, Y. Wang, and A. Datye, Atomically Dispersed Dopants for Stabilizing Ceria Surface Area. Applied Catalysis B: Environmental, 2021. 284: p. 119722.
 

Microscope PhotoMeasured BET surface area of surface doped ceria after aging in 50 sccm of flowing air at 800 degrees C for 5

 

 

 

 

 

 

 

 

 

Thrust Two

Linda Broadbelt, Northwestern University

In Thrust 2, combined experimental and modeling efforts continue to provide insight into the complex chemistry underlying Brønsted acid-catalyzed oligomerization of olefins. The microkinetic modelling methodology that Vernuccio, Bickel, Gounder and Broadbelt developed previously to describe propene oligomerization on medium-pore MFI zeolites has been extended successfully to large-pore Beta zeolites. The extension of the model was supported by the identification of the key descriptors that account for the different topologies and acid strengths of the zeolite frameworks (physisorption enthalpies, stabilization enthalpies, and frequency factors). The model was validated with experimental conversion and selectivity data measured in a plug-flow reactor on a commercial Beta zeolite over a range of operating conditions. Analysis of net reaction rates allowed identifying the preferred pathways that increase oligomerization selectivity toward C9 species with increasing propene pressure. The model was additionally used to investigate how the stabilization enthalpies of chemisorbed intermediates, an important catalyst descriptor, influenced the selectivity and surface coverage at iso-conversion. This analysis provides mechanistic insights into the propene oligomerization reaction network and its dependence on zeolite topology and demonstrates how microkinetic models can describe catalyst behavior and aid in catalyst and process optimization.

Experimental (symbols) and calculated (lines) selectivity and propene conversion graph

Sergio Vernuccio, Elizabeth E. Bickel, Rajamani Gounder, Linda J. Broadbelt, “Propene oligomerization on Beta zeolites: Development of a microkinetic model and experimental validation”, Journal of Catalysis, 2021, https://doi.org/10.1016/j.jcat.2021.01.018

 

Figure: Experimental (symbols) and calculated (lines) selectivity (left) and propene conversion (right) as a function of the ratio of propene pressure [kPa] and propene space velocity [molC3(molH+ s)-1] at 503 K. Propene pressure is 75% of the total pressure in the reactor. Experimental errors are 2% for selectivity and 0.1% for conversion. 


 

Thrust 4

Rakesh Agrawal, Purdue University and David Allen, University of Texas at Austin 

Thrust 4 has continued to perform process synthesis, design and economic evaluation of processes to utilize shale gas NGLs at the gas gathering station and gas processing plant scales.  In addition, promising processes that convert NGLs to liquid fuel are being analyzed for lifecycle greenhouse gas emissions. Reduced-order models are systematically developed and calibrated using data from microkinetic simulations; these models allow detailed process simulation and reactor optimization. Evaluation of scenarios to decarbonize fossil resource-based CISTAR fuels through regionally specific pathways including enhanced oil recovery and integration of CISTAR processes and processes to make biofuels are in progress. Opportunities for process intensification through use of renewable electricity is being explored.

Thrust 6

Ruilan Guo, University of Notre Dame 

Thrust 6 is actively researching new high performance membrane materials to meet the demanding gas separation needs in CISTAR, including separations of H2/HC, C1/C2+, olefin/paraffin and O2/N2. In the project of developing microporous polybenzoxazole (PBO)-based polymer membranes, CISTAR researchers at Notre Dame (Guo group) have successfully expanded the materials spectrum via incorporating specifically designed functionality in the precursor polymers. Such structural manipulation has significantly improved gas permeabilities while maintaining high selectivity (e.g., H2 permeability of 3674±110 Barrer and H2/CH4 selectivity of 119), outperforming the 2015 upper bound. In parallel, high temperature pure-gas permeation tests have been initiated on crosslinked PBO membranes. Preliminary results showed that at 180 oC crosslinked PBO membranes still maintained high separation performance exceeding the 2008 upper bound for H2/CH4 separation. Further investigation is underway to fully assess their high temperature separation performance. The CISTAR team at the University of New Mexico (UNM, Brinker group) has completed the setup and validation of a high temperature, high pressure gas separation and detection system. This system will facilitate permeation testing of ceramic/hybrid membranes up to 800 °C and 100 PSIA. In the effort of developing stable facilitated transport membranes (i.e., supported ionic liquid membranes, SILMs) for separations of C2+/C1 and olefin/paraffin, CISTAR researchers at UT, in collaboration with UNM researchers, have advanced on surface modification of the ceramic membrane support of SILMs, which allows operation at high transmembrane pressure (>15 bar), resulting in effective membrane thickness of less than 1 micron for high flux. The chemical and mechanical stability afforded by these CISTAR innovations is a major step forward in making olefin/paraffin membrane separation a commercial reality.

Thrust 6 also welcomed two new PIs who joined CISTAR this spring: Prof. Casey O’Brien from Notre Dame and Prof. Gabriel Sanoja from UT Austin. Prof. O’Brien’s group is exploring a new strategy to enhance the long-term stability of facilitated transport membranes for olefin/paraffin separations by incorporating thiolate-protected silver nanoclusters into polymer-based nanocomposite membranes. Prof. Sanoja’s CISTAR project will focus on incorporating damage-activated probes into polymeric membranes to predict failure under operating conditions for olefin/paraffin separations.

 

Testbed

Raj Gounder, Purdue University; Justin Notestein, Northwestern University; and Fernando Garzon, University of New Mexico

 

 

We implemented a matrix organization to ensure that CISTAR progresses towards its technology goals. Now, each researcher and project are aligned to a scientific research Thrust and a technology-focused Testbed. To help researchers understand how projects are interrelated in the broader context of candidate CISTAR processes, we developed a graphical organization (termed the ‘Schemata’) of how candidate CISTAR unit operations were related. One can now immediately see the left-to-right flow from candidate feedstocks to products. Each of the three Testbeds encompasses all routes and unit operations that connect one feedstock (e.g. methane) to one product (e.g. an aromatics blend). The form of the Schemata evokes a familiar block flow diagram, but it is not meant to represent an actual process flowsheet. Instead, each Testbed is currently evaluating 2 or 3 broad options. For each option, we are performing a technoeconomic analysis (TEA) to define benchmarks and milestones to be demonstrated by the Technology Modules (Thrust 5), and to identify new research targets to guide Thrust-level research at the fundamental knowledge plane. Periodic Testbed-focused meetings bring together researchers across the involved Thrusts and Technology Modules, to identify promising research leads that may prompt revising the Testbed options or TEA, and generating more detailed process models.

Testbed 1: Natural gas conversion to aromatics

Testbed 1 is focused on converting natural gas components to aromatics. Aromatics represent a large, fungible market as chemicals and fuels components. Producing a high octane, C7-C9 aromatics mix will allow substantial upgrading of local natural gasoline and butane to gasoline value. This will result in a substantial gross margin uplift relative to ethane. Producing aromatics also allows entry into the growth market of chemicals production, especially for p-xylene, should CISTAR wish to pursue that route. Three broad options are being considered for this Testbed: (1) soft oxidative coupling of methane, (2) electrochemical methane activation, and (3) alkane/alkene dehydroaromatization. These Testbed options integrate the latest research advances from Thrusts 3, 4 and 6.

Testbed 2: Natural gas liquids conversion to linear hydrocarbons

Testbed 2 is focused on converting natural gas liquids (C2+) to a product slate of C10-C18 linear and mono-branched paraffins. Methane conversion is not currently considered as part of this Testbed, but methane (as a natural gas feedstock) may be carried through a process as a nominally inert compound. These paraffins represent a large, fungible market as diesel fuel components and as lubricant basestocks. Diesel and lubricant markets are projected for steady growth in the future. This conversion also results in a significant gross margin uplift relative to ethane. Two broad options are being considered for this Testbed using either a (1) hydrocarbon thermal cracking unit or a (2) catalytic hydrocarbon cracking/dehydrogenation catalyst unit, coupled with a downstream catalytic alkene oligomerization unit (selective to linear hydrocarbons). These Testbed options integrate the latest research advances from Thrusts 1, 2, 4 and 6.

Testbed 3: Natural gas liquids conversion to branched hydrocarbons

Testbed 3 is focused on converting natural gas liquids (C2+) to a product slate of C6-C7 di- branched alkanes as a gasoline blendstock. Although gasoline is projected to slowly decline in the US, there remains a substantial need – and a substantial gross margin uplift – in the near and medium-term. As for Testbed 2, methane conversion will not be considered in Testbed 3, but unseparated natural gas may be used as a feedstock. Two broad options are being considered for this Testbed using either a (1) hydrocarbon thermal cracking unit or a (2) catalytic hydrocarbon cracking/dehydrogenation catalyst unit, coupled with a downstream catalytic alkene oligomerization unit (selective to branched hydrocarbons). These Testbed options integrate the latest research advances from Thrusts 1, 2, 4 and 6.