(550e) Copper Modified Activated Carbon for Hydrogen Purification

Authors: 
Relvas, F., University of Porto
Whitley, R. D., Air Products and Chemicals, Inc
Mendes, A., Faculty of Engineering - University of Porto
Silva, C. M., University of Aveiro

Copper modified
activated carbon for hydrogen purification

Frederico Relvas1*,
Carlos M. Silva2, Roger D. Whitley3, Adélio Mendes1**

1LEPABE -
Laboratory for Process Engineering, Environmental, Biotechnology and Energy,
Faculdade de Engenharia, Universidade do Porto, Rua Dr. Roberto Frias, 4200-465
Porto, Portugal

2CICECO – Aveiro Institute of Materials, Department
of Chemistry, University of Aveiro, Campus Universitário de Santiago 3810-193
Aveiro, Portugal

3Air Products & Chemicals, Inc., 7201 Hamilton
Boulevard, Allentown, Pennsylvania 18195, United States

    (*)
Presenting author: frelvas@fe.up.pt                                                         (**)
Corresponding author: mendes@fe.up.pt

Keywords: Adsorption; Kinetics; Hydrogen purification



1.     Introduction

Several studies have been proposing
the use of π-complexation adsorbents for different applications, including
paraffin/olefin separations and hydrogen purification. In a recent work, the
authors modified a commercial activated carbon with CuCl2∙2H2O,
which showed an enhanced capacity for carbon monoxide adsorption. The adsorbent
was tested in a lab-scale pressure swing adsorption unit and proved to be very
effective for improving the PSA performance for producing fuel cell grade
hydrogen from reformate.1 However,
using such adsorbents, the double contribution of the adsorption on the activated
carbon surface and π-complexation bonds formation turns the process
difficult to simulate using a single kinetic model.

In the current work, the
contributions of both the activated carbon surface and the copper adsorption
sites are assessed. A dual kinetic model was proposed based on the linear
driving force (LDF) approximation, attributing a fraction of the total capacity
to the activated carbon and the remaining fraction to the π-complexation bonds
formed with copper salt, each presenting distinct adsorption kinetics. The
model was used to describe experimental fractional uptake and breakthrough
curves.

 
2.   
Experimental
2.1  
Adsorbent
preparation and characterization

Several samples of commercial
activated carbon Kuraray 2GA-H2 (AC) were modified by wet impregnation of CuCl2∙2H2O
using different copper loadings. After drying, the samples were activated under
hydrogen atmosphere for reducing the copper state to Cu+, to which
improved adsorption capacities are attributed. Adsorption equilibrium isotherms
of H2, CO2, CH4, N2 and CO were measured
afterwards by volumetric method. The adsorbents were subsequently exposed to
oxygen and the adsorption isotherms were measured again. The objective of this
step was to oxidize the copper to Cu2+, which does not form π-complexation
bonds, thus isolating the contribution of the pristine AC.

The initial adsorption isotherms
(before oxygen exposure) were fitted by dual-site Langmuir model according to eq.
(1),

                                                                                            

where the first set of sites was attributed to the
contribution of the activated carbon surface, which corresponds to the
equilibrium assessed after oxygen exposure, and the second set of sites was endorsed
to the presence of the copper salt.
2.2  
Kinetic model

The analytical solution proposed by
Kočiřík et al.2 for
describing simultaneous mass and heat transfer under variable pressure
conditions was used to fit the experimental uptake curves by nonlinear
optimization. The model was adapted for considering two parallel processes
according to eq. (2), thus obtaining the effective LDF mass transfer
coefficientsand ,

                                           

Where  is the fractional uptake, is the effective
LDF mass transfer coefficient, is
the average adsorbate concentration in the solid at instant , and are the are the
solid concentrations in equilibrium with the initial gas () and with feed (), respectively, and
is the
contribution factor of each mechanism. The subscripts AC and Cu-AC denote the
activated carbon and copper contributions, respectively. 

 
3.   
Results
and conclusions

Figure 1 presents the adsorption
capacity of the studied gases on the pristine AC and the modified adsorbent Cu-AC-2
(2 mol kg-1 of copper loading). A substantial increase in the carbon
monoxide adsorption was observed for the modified adsorbent, while the opposite
effect was observed for the other gases. As reported elsewhere,1 such
effect is even more evident with copper loading increase. The improvement is
attributed to π-complexation bonding of the Cu+ with CO, which
does not happen for the other gases under study. This fact is confirmed by the adsorption
experiments after O2 exposure: when contacting oxygen, the copper is
oxidized to its valence state Cu2+, which cannot form chemical bonds
with CO and thus decreases the CO adsorption capacity. Since the other gases
cannot form π-complexation bonds, their adsorption capacities remain approximately
the same before and after O2 exposure.

The same conclusions can be drawn
from the fractional uptake curves and breakthrough curves, Figures 2 and 3
respectively. While the adsorption of CO2, CH4, N2
and H2 can be described using a single-kinetic model, carbon
monoxide exhibits a bi-modal behavior, suggesting the presence of two different
mechanisms. The dual-kinetic model suggested in this work proved to efficiently
describe the adsorption dynamics inside the adsorbent material.

The development of a simulator
based on stable adaptive multiresolution methods and comprising this kinetic
approach is ongoing to simulate the breakthrough curves and further PSA
applications. 

Figure 1 -
Experimental adsorption isotherms in commercial (AC) and modified (Cu-AC)
adsorbent at 40 °C for H2, CO2, CH4, N2
and CO. Dashed-lines denotes the dual-site Langmuir model.

Figure 2 -
Fractional uptake curves of CO on the pristine AC and Cu-AC-2.

 
Figure 3 - Experimental
monocomponent adsorption breakthrough of CO (95 % He; 5 % CO) and CO2
(75 % He; 25 % CO2). Dashed-lines – Pristine AC; Solid lines:
Cu-AC-2. Primary axis: CO2; Secondary axis: CO.
Acknowledgements

This work was partially supported by projects POCI-01-0145-FEDER-006939
(Laboratory for Process Engineering, Environment, Biotechnology and Energy –
UID/EQU/00511/2013) funded by the European Regional Development Fund (ERDF),
through COMPETE2020 - Programa Operacional Competitividade e
Internacionalização (POCI) and by national funds, through FCT -
Fundação para a Ciência e a Tecnologia. F. Relvas would also
like to thank to PhD grant NORTE-08-5369-FSE-000028 supported by North Portugal
Regional Operational Programme (NORTE 2020), under the Portugal 2020
Partnership Agreement and the European Social Fund (ESF). Authors also thank project CICECO-Aveiro Institute of Materials
(Ref. FCT UID/CTM/50011/2013), financed by national funds through the FCT/MEC
and when applicable co-financed by FEDER under the PT2020 Partnership
Agreement.

 
References

 

 

1.           Relvas, F.; Whitley, R. D.;
Silva, C. M.; Mendes, A., Single-stage pressure swing adsorption for producing
fuel cell grade hydrogen. Industrial & Engineering Chemistry Research 2018.

2.           Kočiřík, M.; Struve,
P.; Bülow, M., Analytical solution of simultaneous mass and heat transfer in
zeolite crystals under constant-volume/variable-pressure conditions. Journal
of the Chemical Society, Faraday Transactions 1: Physical Chemistry in
Condensed Phases
1984, 80 (8), 2167-2174.

Topics: