Corrigendum to “Steam reforming of ethylene over manganese-chromium spinel oxides” [J. Catal. 380 (2019) 224–235] (Journal of Catalysis (2019) 380 (224–235), (S0021951719305056), (10.1016/j.jcat.2019.10.006))

Document Type


Publication Date



Department of Chemical Engineering


The authors regret that the activity results associated with Cr2O3 for ethylene steam reforming reported in Fig. 9 and Table 3 were recently determined to be incorrect. The original paper indicates Cr2O3 is inactive for ethylene steam reforming at the experimental conditions tested. Subsequent experiments with new Cr2O3 samples revealed this conclusion to be incorrect. We speculate that the originally reported inactive Cr2O3 catalyst was contaminated by Si deposited on the catalyst surface during the reaction due to the degradation of silicone O-rings associated with the reactor assembly. Multi-spot sample analysis of the inactive spent catalyst by XPS revealed the presence of Si on the surface but not on the fresh powders. To the best of our knowledge, these O-rings were only used in this study during the evaluation of Cr2O3. This corrigendum reports the new findings regarding Cr2O3 as presented in Fig. 9, Table 3 and related text. The revised Fig. 9 and Table 3 are shown below. The following text in the original paper has been modified to reflect the correct results: Abstract Original: “Whereas Cr2O3 was nearly inactive for ethylene steam reforming, Mn3O4 was substantially more active than the spinel catalysts, but Mn3O4 deactivated significantly upon in situ reduction to MnO.” Corrected: “While being initially more active than the spinel catalysts, Cr2O3 and Mn3O4 (which reduced to MnO) both deactivated during ethylene steam reforming.” 3.2. Steam reforming of ethylene over Mn oxide, Cr oxide and Mn-Cr-O spinel catalysts Original: “Compared to the other catalysts, Cr2O3 exhibited negligible steam reforming activity at the tested conditions.” Corrected: “The Cr2O3 catalyst exhibited a higher initial ethylene reforming activity than the spinel catalysts but deactivated continuously after about 3 h on stream. The Cr2O3 catalyst and Mn1.5Cr1.5O4 spinel catalyst were further tested in ethylene steam reforming under similar partial pressures of H2 for an appropriate comparison of catalyst stability. As shown in Fig. C1, despite a higher initial reforming rate, the Cr2O3 catalyst rapidly deactivated after 3 h on stream and eventually became less active than the Mn1.5Cr1.5O4 spinel catalyst, which remained stable for over 10 h on stream. Since no degradation of the Cr2O3 phase was detected by X-ray diffraction and the initial reforming activity of the deactivated catalyst could be fully regenerated by re-oxidation, the deactivation of the Cr2O3 catalyst was likely the result of coke deposition evident by the weight loss (0.22 wt%) during temperature-programmed oxidation of the spent catalyst sample.” 4. Conclusions Original: “The spinel catalysts outperformed the pure metal oxides since under the same reforming conditions Cr2O3 was nearly inactive, whereas Mn3O4 deactivated substantially due to in situ reduction to MnO.” Corrected: “The spinel catalysts outperformed the pure metal oxides since under the same reforming conditions Cr2O3 and Mn3O4 both deactivated, with Mn3O4 reducing to MnO in situ.” Original: “Despite the presence of Cr2O3 in the Mn0.5Cr2.5O4 sample, the similar performance of Mn0.5Cr2.5O4 and MnCr2O4 suggests the stoichiometric spinel is the active component on the surface of both catalysts.” Corrected: This statement is removed from the Conclusions because of the observed activity of Cr2O3 for ethylene steam reforming. The authors would like to apologize for any inconvenience caused.

Publisher's Statement

Original article:

Publication Title

Journal of Catalysis