ACS Catalysis
Research Article
at 530 K only through extrapolation; however, the activity of
this catalyst increased dramatically with increasing calcination
temperature. After heating to 1073 K, the temperature at which
the rate was 1019 CH4/(s·g Pd) was 485 K, indicating that
50ZrO2−PdO/Al2O3(1073) catalyst was more active than even
fresh Pd/Al2O3. Because the STEM results indicated that the
Pd particles were covered by the ZrO2 film following ALD, it is
likely the increased rates are due to breaking up of that ZrO2
film. Again, changes in the rates were not reflected by changes
in the dispersions measured by CO adsorption, shown in Table
2. The relatively high activity of the 50ZrO2−PdO/Al2O3
(1073) catalyst must be due in part to interactions with ZrO2.
Methane oxidation rates for a catalyst prepared by 50 ALD
cycles of ZrO2 onto reduced Pd/Al2O3 are shown in Figure 6c.
Rates for the unmodified Pd/Al2O3 after calcination at 773 and
1073 K are shown for comparison. Interestingly, rates on the
50ZrO2−Pd/Al2O3 samples were the same when calcined at
either 773 or 1073 K. Rates were somewhat lower than for the
unmodified Pd/Al2O3 calcined at 773 K but were stable. The
measured dispersions on this catalyst were higher than that
measured on 50ZrO2−PdO/Al2O3 catalysts and did not change
with calcination temperature. We suggest that, in this case, the
ZrO2 film did not grow on the metallic Pd, so there was no
need to “break up” the oxide film. However, the presence of the
ZrO2 film on the Al2O3 does appear to thermally stabilize the
Pd particles.
the oxide support.24−26 In past work on Pd@ZrO2 core−shell
catalysts, evidence was presented that the ZrO2 in contact with
the Pd is reducible and that this reducibility is somehow
responsible for stabilizing PdO and enhancing reaction rates.11
Another interesting observation from the present study is
that the ZrO2 films deposited by ALD were XRD-amorphous to
high temperatures. There was some evidence from TEM for
changes in the film at 773 K, but the crystallization is clearly
suppressed. One would normally expect that thin films would
be highly unstable and that the zirconia would try to minimize
its free energy by forming crystalline particles. Interactions with
the alumina must somehow suppress this crystallization. One
would expect these effects to be strongly dependent on the
thickness of the zirconia film. How important the crystalline
structure for zirconia might be is uncertain. With ceria, it is well
established that reducibility is affected by the crystalline
structure.29
The mechanism by which the zirconia films stabilize Pd
sintering is not completely understood. Sintering is most severe
when PdO decomposes to form Pd, which normally occurs
above 1073 K. Although the zirconia film may simply stabilize
the particles through physically covering the particles, it is also
possible that the proximity between the Pd and zirconia could
stabilize the PdO phase, as has been observed in Pd@ZrO2
core−shell catalysts.11 Although there have been many studies
of catalyst preparation by ALD, we believe this area of work is
still in its infancy. Even considering only applications in which
ALD is used to add promoters to metal catalysts, there are
many variables that have yet to be explored, including
composition of the deposits and thickness of the films. The
present work provides an example to further demonstrate the
promise of this approach.
DISCUSSION
■
ALD is an intriguing method for promoting supported-metal
catalysts.14 The structure of the oxide films prepared by ALD is
clearly very different from what could be achieved by adding an
oxide promoter using infiltration or coprecipitation methods.
Although most other methods of adding oxides will tend to
form oxide particles or are limited to forming a single oxide
monolayer, ALD allows formation of thin films of varying
thickness. The observation from this work that ZrO2 films
could grow on PdO but not on Pd also suggests that it may be
possible to deposit promoters over only selected parts of the
catalyst. The present study did not explore compositional
changes, but ALD clearly allows deposition of a wide range of
materials.
CONCLUSION
■
Modification of PdO/Al2O3 catalysts by ALD deposition of 1
nm ZrO2 films is shown to result in catalysts with a semicore−
shell structure. The ZrO2 films stabilize Pd against sintering at
high temperatures and enhance the methane oxidation activity
of these catalysts following high-temperature calcination.
AUTHOR INFORMATION
Corresponding Author
■
It is interesting to notice that the properties of the Pd/Al2O3
catalysts with 1 nm ZrO2 overlayers prepared by ALD have
many similarities to Pd@ZrO2/Al2O3, core−shell catalysts that
have been described in a previous publication.11 In that case,
the Pd@ZrO2 particles were prepared in solution to have a ∼ 2
nm ZrO2 shell covering Pd nanoparticles, then adsorbed onto
functionalized Al2O3. In both cases, the ZrO2 films were found
to stabilize the Pd particles against sintering at high
temperatures. The ZrO2-covered Pd catalysts also exhibited
enhanced activity for methane oxidation.
The reason for the enhanced methane oxidation rates on the
zirconia-promoted catalysts is uncertain. The literature shows
that methane oxidation on Pd is a complex reaction that likely
involves C−H activation on sites in the vicinity of both metallic
Pd and PdO.27 Although there is strong evidence that C−H
bond activation is structure-sensitive in Pd, there is controversy
over whether particle size influences reaction rates, with some
reporting that larger particles exhibit higher specific rates21,22
and others reporting that rates are strictly proportional to the
surface area of the Pd or PdO.28 There is agreement that PdO
must be present in the active phase; but the stability of PdO,
along with the reaction rate for methane, can be influenced by
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
T.M.Z. and R.J.G. are grateful to the Department of Energy,
Office of Basic Energy Sciences, Chemical Sciences, Geo-
sciences and Biosciences Division, Grant No. DE-FG02-
13ER16380 for support of this work. S.Z., G.G., and X.P. are
grateful to the National Science Foundation, Grant Nos.
CBET-1159240 and DMR-0723032 for support of this work.
REFERENCES
■
(1) McCarty, J. G.; Malukhin, G.; Poojary, D. M.; Datye, A. K.; Xu,
Q. J. Phys. Chem. B 2005, 109, 2387−2391.
(2) Farrauto, R. J. Science 2012, 337, 659−660.
(3) Colussi, S.; Gayen, A.; Farnesi Camellone, M.; Boaro, M.; Llorca,
J.; Fabris, S.; Trovarelli, A. Angew. Chem., Int. Ed. 2009, 48, 8481.
(4) Cargnello, M.; Jaen, J. J. D.; Garrido, J. C. H.; Bakhmutsky, K.;
Montini, T.; Gamez, J. J. C.; Gorte, R. J.; Fornasiero, P. Science 2012,
337, 713−717.
5700
ACS Catal. 2015, 5, 5696−5701