(658a) Simultaneous Measurement of X-Ray Absorption Spectra and Kinetics: A Fixed-Bed, Plug-Flow Operando Reactor
AIChE Annual Meeting
2009
2009 Annual Meeting
Catalysis and Reaction Engineering Division
In Situ and Operando Spectroscopy of Catalysts
Friday, November 13, 2009 - 8:30am to 8:50am
Introduction
X-ray absorption
spectroscopy (XAS) is a powerful tool that can be used to characterize
catalysts at reaction conditions. Extended x-ray absorption fine structure
(EXAFS) yields information on coordination numbers, inter atomic distances and
types of neighbors. X-ray absorption edge spectroscopy (XANES) reveals
information on the oxidation state, coordination symmetry and, in many cases,
the coverage of adsorbates. Because the structure-property of a catalyst is
most important at reaction conditions, we have developed a novel operando
reactor system which allows us to perform high quality XAS while simultaneously
obtaining kinetic data [1].
Clausen and Topsøe [2]
utilized a capillary tube as an operando XAS reactor while Bare et
al. [3]
have demonstrated operando XAS using a Be plug flow reactor (PFR).
However, catalysts are difficult to load into capillary tubes and Be is
expensive which makes these reactor designs non-ideal.
Here we describe an operando
reactor capable of probing catalysts at temperatures up to 550 °C
and 40 atm. We used a readily available borosilicate glass tube reactor loaded
with a 2% Pd, 13.7% Zn on Al2O3 catalyst to examine the
water-gas shift (WGS) reaction at catalytically relevant reaction conditions.
In addition, we show this novel reactor is capable of determining surface
coverages of H2O, CO and H2 under WGS reaction conditions
on alumina supported Pt and Au catalysts from XANES spectra. These data
correlate well to data we find with diffuse reflectance infrared Fourier
transform spectroscopy (DRIFTS).
Experimental
X-ray absorption
measurements were made on the insertion device (ID) beamline, 10-ID-B, of the
Materials Research Collaborative Access Team (MRCAT) at the Advanced Photon
Source (APS), Argonne National Laboratory. The 2% Pd, 13.7% Zn on Al2O3
WGS catalyst was the same as Bollmann et al. [4].
The Pt/Al2O3 catalyst was fabricated by using atomic
layer deposition (ALD): 1 cycle of Pt(MeCp)Me3 followed by O2
over spherical alumina (NanoDur). The 0.9% Au/Al2O3 was
purchased from AuTEK. Two types of reactors were used: (i) a borosilicate glass
tube reactor (OD: 0.25?, ID: 0.152?, 10 cm long) for the 2% Pd, 13.7% Zn on Al2O3
and ALD Pt/Al2O3 catalysts and (ii) a standard NMR tube
reactor (Wilmad-LabGlass, part #: 6.5-PP-9) for the Au/Al2O3
catalyst. The operando reactor was heated by a custom designed,
temperature controlled Al heating block. The block was outfitted with a through
hole allowing access to the catalyst bed by the x-rays. The reactant gas flow
rate was a standard WGS gas composition of 21.4% H2, 6.4% CO, 11% H2O,
and balance Ar. The rate of CO consumption was used to calculate WGS rate and
the effluent gas was analyzed with a gas chromatograph. Adsorption experiments
of CO, H2 and H2O were performed on two different
catalysts, ALD Pt/ Al2O3 and 0.9% Au/g-Al2O3,
using the operando reactor and a DRIFTS cell (CO only). The standard WGS gas
composition was used to investigate the amount of adsorption on the metal
surface. The DRIFTS experiments were performed in an in-situ cell
described elsewhere [4].
Results and Discussion
To
validate the borosilicate glass tube reactor as a true PFR, it was loaded with
a 2.0% Pd, 13.7% Zn on γ-alumina catalyst and tested under standard WGS conditions
and compared to Bollmann et al. [4].
The apparent activation energy Ea within the temperature range of
200-300 °C was 65.4 ± 1.7 kJ/mol, in agreement with 69 ± 3 kJ/mol found by
Bollmann et al. [4],
and the rate at 250 °C was 1.2x10-6 ± 0.1x106 mol s-1 g-cat-1.
The rate reported from Bollmann et al. adjusted to our conditions is
1.4x10-6 mol s-1 g-cat-1,
a difference of 15%. At RT, the PdZn has not been fully reduced, but at higher
temperatures, 250 °C to 330 °C,
the catalyst shows no signs of oxidation, as evident in both XANES and EXAFS,
and a Pd-Zn alloy forms. The quality of XANES and EXAFS spectra are excellent
and are comparable to spectra seen elsewhere [4].
The Pd-Pd peak at 2.74 Å, Pd-Zn at 2.52 Å and Pd-O at 2.00 Å are clearly
visible in the EXAFS.
Adsorption experiments
show that under WGS conditions CO covers approximately 80% of the Pt surface at
250 °C and 16% of the Au surface at 120 °C using ALD Pt/Al2O3
and Au/Al2O3, respectively. These findings match well with
what we see using DRIFTS; CO covers significantly more of the Pt surface than
it does Au. At higher CO surface coverages, as seen on ALD Pt/Al2O3,
one would expect to see a lower CO order of reaction than on lower CO
coverages, as seen on the Au/Al2O3. Indeed, this is the
case; the CO order of reaction is 0.05 and 0.78 for Pt/Al2O3
and Au/Al2O3, respectively. Moreover, the amount of
adsorbed H2 seen using XANES is minimal on both the Pt and Au
catalysts under reaction conditions.
Conclusion
It has been shown that the operando XAS
reactor is capable of simultaneously producing high quality EXAFS, XANES and
kinetic data. These types of data aide in adsorbate and structure-property
characterizations at relevant reaction conditions, liquid or gas phase. These
types of experiments are made possible by high flux 3rd generation
ID beamlines.
Acknowledgements
The authors would like
to thank the Advanced Photon Source (APS) at Argonne National Laboratory and
the Materials Research Collaborative Access Team (MRCAT) at beamline 10-ID-B.
References
1) Fingland,
B.R., Ribeiro, F.H., and Miller, J.T., Catalysis Letters (2009) Accepted for
publication.
2) Clausen,
B.S., and Topsøe, H., Catalysis Today 9 (1991) 189-196.
3) Bare,
S.R., Yang, N., Kelly, S.D., Mickelson, G.E., and Modica, F.S., Catalysis Today
126 (2007) 18-26.
4) Bollmann,
L., Ratts, J.L., Joshi, A.M., Williams, W.D., Pazmino, J., Joshi, Y.V., Miller,
J.T., Kropf, A.J., Delgass, W.N., and Ribeiro, F.H., Journal of Catalysis 257
(2008) 43-54.