Page 4 - The Effects of Confinement inside Carbon Nanotubes on Catalysis

Effects of Confinement inside CNTs on Catalysis Pan and Bao

FIGURE 4. (a) Syngas conversion to C 2 oxygenates over RhMn catalysts; (b) Raman spectra for RhMn catalysts upon in situ CO adsorption.

confinement in CNTs appears to be a general phenomenon by 13% after 125 h time on stream. 31 The above results
because a similar trend was observed for ruthenium 20 and indicate that the FTS activity difference between inside
cobalt. 28 and outside catalysts is sensitive to pressure and space
Raman spectroscopy showed that the FeO band red- velocity.
shifted for the outside Fe 2 O 3 particles with respect to bulk Confinement of a bicomponent RhMn catalyst in CNT
Fe 2 O 3 . 22 However, it blue-shifted for the CNT-confined channels (RhMn-in) was also found to enhance its catalytic
Fe 2 O 3 particles compared with the outside oxide. The blue activity for syngas conversion to C 2 oxygenates with respect
shift became stronger with decreasing inner diameter to the outside catalyst (RhMn-out) (Figure 4a). 24 TEM indi-
(Figure 3b). Although the particle size declined with the cated that the particle size was not the only factor affecting
downsized CNT channels, the stepwise blue shift of the the activity because the two fresh catalysts had similar size
FeO band contrasts with the red shift generally reported distributions. The catalyst supported on SBA-15, which has a
for nanosized Fe 2 O 3 . Confinement of manganese oxide pore diameter (68 nm) similar to that of the used CNTs,
also resulted in a blue shift of the MnO frequency by exhibited a much lower activity and a higher selectivity to
6cm 1 compared with the outside MnO 2 particles. 29 the byproduct methane under the same reaction conditions.
These observations imply there might be interactions This suggested that the spatial restriction may not be the
between the confined metal oxides and the CNT surface. sole effect. In addition to likely increased local density of
In situ XRD (Figure 2b) showed that the improved reduci- reactants within small channels, Raman spectra (Figure 4b)
bility of Fe-in favored the formation of more iron carbides indicated that the activation of CO may have been modified
under reaction conditions, which are generally accepted as inside CNTs. 24 Two Raman bands corresponding to the
the catalytically active phase for FTS synthesis. 30 Therefore, RhC(∼550 cm 1 ) and MnO(∼640 cm 1 ) bonds were
the higher activity of Fe-in could be attributed to the con- observed for RhMn-in but only one for RhMn-out when both
finement inside CNTs, which prevented iron particles from catalysts were exposed to CO. This was likely the result of the
severe sintering and facilitated the reduction of iron oxide to different interaction of the metals with CNTs. The oxophilic
form iron carbides. Mn may remain in a more reduced state inside the CNT
A better reducibility and an enhanced catalytic perfor- channels with respect to the outside Mn species. Thus Mn
mance of CNT-confined iron (Fe-in) were also observed by could attract the O of CO adsorbed on adjacent Rh sites,
Dalai and co-workers. 31 Their Fe-in catalyst had 7080% Fe leading to a tilted adsorption with the C atom bonding to Rh
particles located inside CNTs with a particle size of 611 nm, andOtoMninsideCNTs,whichcouldfacilitateCOdissociation.
while their Fe-out had slightly smaller particles (59 nm). In contrast, the tendency of outside Mn to attract the O of CO
Both catalysts exhibited a similar initial CO conversion at could be weaker and hence a lower CO dissociation activity.
2 MPa and space velocity of 2 L/(g h). However, Fe-in was Ammonia Synthesis and Ammonia Decomposition.
much more stable as CO conversion was 6%10% higher Confinement inside CNTs can also have a negative impact
than that over Fe-out in the temperature range of 265285 °C, on the catalytic activity, as we have observed for ammonia
and selectivity to C 5þ hydrocarbon products was higher synthesis over Ru catalysts. 20 The turnover frequency (TOF)

556 ’ ACCOUNTS OF CHEMICAL RESEARCH ’ 553–562 ’ 2011 ’ Vol. 44, No. 8