Final Report: Nitrogen-Selective Membrane for Carbon CaptureEPA Grant Number: SU834761
Title: Nitrogen-Selective Membrane for Carbon Capture
Investigators: Ozdogan, Ekin , Rochana, Panithita , Wilcox, Jennifer
Institution: Stanford University
EPA Project Officer: Nolt-Helms, Cynthia
Project Period: August 15, 2010 through August 14, 2011
Project Amount: $10,000
RFA: P3 Awards: A National Student Design Competition for Sustainability Focusing on People, Prosperity and the Planet (2010) RFA Text | Recipients Lists
Research Category: Pollution Prevention/Sustainable Development , P3 Challenge Area - Energy , P3 Awards , Sustainability
The main objective of the proposed research project is to develop N2-selective catalytic membrane technology with potential sustainability benefits of air separation, ammonia synthesis and indirect CO2 capture. The N2-selective membrane technology benefits from the driving force of N2 in air (78.08 vol.%) for air separation or flue gas (73 wt.%) streams for indirect CO2 capture as it provides atomic nitrogen on the permeate side of the membrane during separation. With these properties if H2 is provided as a sweep gas alongside this novel technology, ammonia synthesis as a byproduct of the N2 separation process may be possible.
Successful implementation of a N2-selective membrane technology has the capability of separating N2 from air with potentially lower energy requirements for oxy-combustion applications compared to traditional noncatalytic techniques such as sorbents or cryogenic separation. Also, due to the high selectivity of this method, increased oxygen purity is expected. Moreover, achieving ammonia synthesis from air or flue gas streams by using the proposed membrane technology could have critical impact on securing sufficient global food production.
To achieve the main objective of the project, fundamental investigations of molecular adsorption, dissociation, and potential subsequent atomic diffusion of small molecules (N2, O2, and H2) within the Group V metals, vanadium, niobium, and tantalum are performed by theoretical calculations and experiments. Group V metals, being on the far left of the transition metals in the periodic table are known for their property of strong-binding. Because of this phenomenon these metals are interesting to investigate for small molecule reactivity to the extent of potential atomic N, O, H, and C diffusion into their bulk crystal structures in application to gas separation. Combining superior transport properties of Group V metals with the enhanced ammonia synthesis activity of Ru would provide a N2-selective membrane with optimal characteristics. This work involves three areas of focus: (1) density functional theory (DFT) calculations to investigate mechanisms of reactivity of Group V metals and their alloys, (2) materials synthesis and testing using a membrane reactor set-up, and (3) synchrotron-based surface-science experiments to elucidate important details of the chemical bonding at the surface.
Phase I studies included the theoretical calculations which are used to predict the surface interactions and the bulk transport properties of N within V and V-Ru alloys using DFT-based electronic structure calculations.
Pure V and V-Ru alloys have been simulated as three-dimensional infinite periodic structures by defining a supercell with periodic boundary conditions in all three principal axes. Simulations have been carried out using Vienna ab initio Simulation Package (VASP). For the bulk diffusion simulations, atomic N was found to be stable in the octahedral (O-site), face-octahedral (face O-site) and tetrahedral (T-site) interstitial crystal sites. It was determined that the octahedral site is the most favorable binding site for N within bulk V, with a binding energy of -2.132 eV. Nitrogen binding in V is nearly two orders of magnitude stronger compared to the well-known H-binding case (-0.076 eV). These strong N binding energies indicate that N may have difficulty diffusing through the material. Ruthenium doping of V was investigated since Ru is a well-known catalyst for the Haber-Bosch ammonia synthesis process. Through alloying V with Ru, the binding energy could be tuned so that it approaches that of the H binding energy in the bulk V system. We found that the addition of the Ru as a doping material acts to reduce the interaction between V and N thereby weakening the binding energy to -0.889 eV.
Preliminary nudged elastic band calculations have been carried out for atomic N in the pure V crystal and in the V-Ru alloy. In pure V the activation barrier calculated with DFT is approximately 1.1 eV. This value is approximately 4 times that of the hydrogen diffusion barrier in bulk PdCu alloys investigated by Kamakoti and Sholl. In comparison, experimental measurements yield a value of approximately 1.4 eV. The discrepancy between the theory to experiment likely stems from the fact that the experimental systems have defects where the theoretical simulations represent perfectly crystalline periodic structures. The addition of a single Ru atom in the lattice reduces the activation barrier to 0.42 eV, which results in a substantially higher probability of hopping through this pathway. Additional simulations have been carried out with kMC to determine the diffusivity at high temperatures. To examine how alloying the V crystal with Ru will influence N diffusion, simulations have been carried out assuming the dominate hopping mechanism is from an O-site with only V atoms to an O-site with a single Ru on one corner. As it can be seen in Figure 1, the reduction in the activation barrier with the addition of the Ru dopant results in an increase in the diffusivity by several orders of magnitude at temperatures approaching 2000K.
A quick investigation on the relative stability of N2 versus CO2 on the V surface was carried out on a V surface at 1/4 coverage on the top site and oriented perpendicular to the surface. It does not appear that the CO2 molecule will bind to the V surface; meanwhile, N2 will bind strongly, even on the clean defect-free surfaces. Nitorgen is known to be the most difficult diatomic molecule to dissociate due to its triple bond; therefore, the energy barrier associated with N2 dissociation is a crucial parameter in determining its surface reactivity. The adsorption energies of N2 on V surfaces are stronger than the adsorption energy of N2 on Fe and Ru, which are the typical catalysts for Haber-Bosch process as can be seen from Figure 2. Based on the singlecrystal studies of the kinetics of the N2 dissociative adsorption and high pressure reactions on Fe, it has been shown that the Fe(111) is the most reactive surface, whereas the flat (110) surface is nearly non-reactive. A comparison between the proposed dissociation mechanism of N2 on Fe(111) and V(111) surfaces is demonstrated in Figure 3. It was found that molecular N2 initially adsorbs on V(111) in a perpendicular configuration. Then, the molecular N2 becomes tilted toward the V surface and eventually dissociating on the bridge-bridge site. The calculations for the minimum energy path along with the associated activation barrier and transition structures are ongoing using the climbing image nudge elastic band (CI-NEB) method.
Figure 1 High temperature diffusion coefficient for N in pure V and in RuV alloy (Pure V is on the left y-axis and V-Ru is on right)
Figure 2 N2 Adsorption energy on various metal surfaces.
Figure 3 Proposed dissociation pathway for N2 on Fe(111) (modified from) and V(111).
It has been shown that V may be a promising material for the N2-selective membrane since CO2 has not found to be stable on its surfaces. However, the electronic properties of the V have to be improved due to the low diffusivity of N in the bulk pure V. It has been found that Ru, as a doping material, can be used to tune the interaction between N and V resulting in an increase in the diffusivity by several orders of magnitude. In the next step, the interaction of O with V and V-Ru alloys will be examined to investigate the possibility of O transportation through the membrane. Generally, N2 bonding on the surfaces is stronger than any other molecules including O2, and it is likely that if N2 dissociates on a surface followed by atomic N diffusion through the membrane, O will likely also. Results of this research will enhance the current state of knowledge associated with small molecule adsorption and potential diffusion through materials that may show promise as catalysts for a variety of applications such as air separation, NH3 synthesis, and indirect CO2 capture. Integrating carbon capture with NH3 synthesis could play an important role in terms of food security strategies in the future as well as simultaneously preventing carbon emissions. This project will potentially have great impact toward improving both environmental and human sustainability.