(487e) Final Step for CO Syngas Clean-Up: Comparison Between CO-PROX and CO-SMET Processes | AIChE

(487e) Final Step for CO Syngas Clean-Up: Comparison Between CO-PROX and CO-SMET Processes

Authors 

Specchia, V. - Presenter, Politecnico di Torino
Specchia, S., Politecnico di Torino
Ashraf, M. A., Politecnico di Torino
Ercolino, G., Politecnico di Torino



Polymer Electrolyte Fuel Cells (PEM-FC) are the most promising candidate for electric powertrain in transportation applications due to higher energy efficiency and lower greenhouse gas emissions as compared to the internal combustion engines. Hydrogen is normally used as fuel for PEM-FCs. Because of lack of infrastructures and low storage capability on board vehicles, a hydrogen-rich gas can be produced through suitable reforming reactions of fossil fuels followed by water Gas Shift reactors to reduce CO up to 0.5–1%. Considering the actual tolerable level of 10 ppmv CO for PEM-FC’s Pt electrocatalyst, the hydrogen-rich gas requires a final CO clean-up step. There are several suitable methods to remove CO from synthesis gas. For PEM-FC application, the low operating pressure processes of preferential CO oxidation (CO-PROX) or CO selective methanation (CO-SMET) are the feasible choice because of the limited spaces available on board vehicles.

CO-PROX reaction has been extensively studied experimentally because of its reliable capability to lower CO to less than 10 ppmv. For example, in a fixed bed reactor the catalyst 1% Rh on zeolite 3A, tested at a WSV of 0.66 NL min–1 gcat–1, was able to reduce the inlet CO concentration below 2 ppmv within a temperature range of at least 80-120 °C [1]. However, this technology requires a closely controlled low O2 supply by expensive mass flow meters, to keep lowest the H2 parallel oxidation, and a wide operating temperature range for control purpose thus it makes costly and complex the process for low power PEM-FC application.

CO-SMET reaction, instead, being able to reduce the CO concentration below 10 ppmv without adding any co-reagent, does not present the above mentioned shortcomings making the process inherently more easily controllable than CO-PROX as, moreover, the selective methanation reaction is less exothermic than CO and H2 oxidations. For example, in a fixed bed reactor the catalyst 4% Ru γAl2O3, tested at a WSV of 0.33 NL min–1 gcat–1, was able to reduce the inlet CO concentration below 2 ppmv within a temperature range of at least 300-340 °C [2]. Even if the H2 consumption by the selective methanation reaction is higher compared to the hydrogen consumed by the parallel hydrogen oxidation inherently existing in the CO-PROX process, CO-SMET produces methane, inert compound within a fuel cell, but it can be burnt in the afterburner, together with the unreacted anodic hydrogen. In this way, more heat and thermodynamically more useful (since at higher temperature) is generated for the fuel processor compared to heat recovered with the CO-PROX, resulting, therefore, for CO-SMET lower fuel consumption and higher fuel processor efficiency.

Since the two processes for the final CO removal present both pros and cons, the main goal of the present study is to compare and evaluate a Steam Reforming Fuel Processor Unit (SR-FPU) integrated with a PEM-FC for H2 production for automotive applications, with two different options for the final CO removal: a CO-PROX or a CO-SMET reactor. The integrated SR-FPU consists of a catalytic steam reformer SR, integrated with an afterburner AFB, where the combustion of the H2 exhaust gas (or H2 and CH4 exhaust gas) from the PEM-FC anode takes place to provide the necessary heat to the system. The water gas shift WGS reactor together with CO-PROX catalytic reactor, or CO-SMET catalytic reactor, come to get to completion the transformation of the primary fuel in the secondary one by enabling both the CO clean-up process and the simultaneous increase of the H2 flow rate. The auxiliary units necessary to properly operate the FPU (heat exchangers for the internal heat recoveries, water recovery radiators, air compressor, water and fuel pumps) complete the BoP of the whole system. Finally, the PEM-FC produces electric power from the secondary fuel. Methane and propane are the considered fuel feedstocks.

The evaluation of SR-FPU performance is done by simulations carried out with a model implemented with the AspenPlus software. The following variables are considered: the steam-to-carbon ratio (SCR = 3.5), a oxygen-to-carbon ratio to avoid coke formation (OCR = 0.07), the SR reactor inlet and outlet temperature (Tin = 550 °C, Tout = 780 °C), the H2 utilization coefficient in the PEM-FC anode (ηAD = 0.65), the cathode stoichiometry (CS = 1.8), various efficiencies (FPU and APU), the powers (gross and net) and the overall water balance. Various assumptions were implemented into the model [3,4]: (i) complete conversion of the primary fuel; (ii) the thermodynamic equilibrium in the reformer outlet gas, according to the WGS reaction; (iii) no methanation reaction when the CO-PROX unit is considered; (iv) methanation reaction limited to a value of R ratio (R = mol formed CH4/mol reacted CO) equal to 1.3 when the CO-SMET unit is considered. The above assumptions are generally confirmed in practice if the reformer volume is sufficiently large and the temperature is adequately high to boost reaction kinetics. As major constrains implemented into the model, the thermal self-sustainability of the SR-FPU and the closure of water balance with no make-up were used. The global thermal power requested to fulfill the Tin value of the steam reformer feed is recovered from the hot flow rates of the APU system by suitable heat exchangers. Also the requested water is recovered from the system itself by using suitable air radiators which control the dew point temperature of some gaseous rates rich of water vapor.

Preliminary calculations performed up to now allowed enlightening that the SR-FPU with the CO-SMET reactor presents higher fuel processor efficiency as compared to the SR-FPU with the CO-PROX unit. Details will be provided within the full manuscript.

[1] Galletti C., Specchia S., Saracco G., Specchia V. Ind. Eng. Chem. Res. 47 (2008) 5304-5312.

[2] Galletti C., Specchia S., Saracco G., Specchia V. Chem. Eng. Sci. 65 (2010) 590-596.

[3] Cutillo A., Specchia S., Antonini M., Saracco G., Specchia V. J. Power Sources 154 (2006) 379-385.

[4] Specchia S.; Cutillo A., Saracco G., Specchia V. Ind. Eng. Chem. Res.45 (2006) 5298-5307.