Portal de Periódicos da CAPES

Impact of Surface Area in Evaluation of Catalyst Activity

2018; Elsevier BV; Volume: 2; Issue: 6 Linguagem: Inglês

10.1016/j.joule.2018.05.003

2542-4785

Shengnan Sun, Haiyan Li, Zhichuan J. Xu,

Catalysis and Hydrodesulfurization Studies

Z.J. Xu is an associate professor in the School of Materials Science and Engineering, Nanyang Technological University. He is a member of the International Society of Electrochemistry (ISE) and The Electrochemistry Society (ECS), and a Fellow of the Royal Society of Chemistry (FRSC). He served as the Vice President of ECS Singapore Section.S. Sun is a postdoctoral research fellow in the School of Materials Science and Engineering, Nanyang Technological University. He received his PhD from the same institution.H. Li is a PhD student in the School of Materials Science and Engineering, Nanyang Technological University. She obtained her BS degree from the Huazhong University of Science and Technology. Z.J. Xu is an associate professor in the School of Materials Science and Engineering, Nanyang Technological University. He is a member of the International Society of Electrochemistry (ISE) and The Electrochemistry Society (ECS), and a Fellow of the Royal Society of Chemistry (FRSC). He served as the Vice President of ECS Singapore Section. S. Sun is a postdoctoral research fellow in the School of Materials Science and Engineering, Nanyang Technological University. He received his PhD from the same institution. H. Li is a PhD student in the School of Materials Science and Engineering, Nanyang Technological University. She obtained her BS degree from the Huazhong University of Science and Technology. Electrocatalysis has been one of the hottest research fields in recent years due to the global effort in exploring sustainable energy technologies. The central effort is to develop highly active electrocatalysts for those reactions playing key roles in fuel cells, electrolyzers, and rechargeable metal-air batteries, such as oxygen evolution, oxygen reduction, and hydrogen evolution reactions. A highly active electrocatalyst should be able to promote the reaction rate of the corresponding reaction. In kinetics, the activation energy should be lowered, and, electrochemically, a lowered overpotential should be observed at a certain current density. Tafel plots are often used as a standard approach to present and compare the activities of catalysts, in which the applied potential or the overpotential at identical current density can be read. A low overpotential is usually the indicator of a good catalyst. For example, Figure S1 shows Tafel plots of oxygen evolution reaction (OER) on spinel MnCo2O4 oxides synthesized by several methods.1Wei C. Feng Z. Scherer G.G. Barber J. Shao-Horn Y. Xu Z.J. Cations in octahedral sites: A descriptor for oxygen electrocatalysis on transition-metal spinels.Adv. Mater. 2017; 29: 1606800Crossref Scopus (417) Google Scholar The dashed line refers to the benchmark catalyst IrO2.2Lee Y. Suntivich J. May K.J. Perry E.E. Shao-Horn Y. Synthesis and activities of rutile IrO2 and RuO2 nanoparticles for oxygen evolution in acid and alkaline solutions.J. Phys. Chem. Lett. 2012; 3: 399-404Crossref PubMed Scopus (2565) Google Scholar In general, the plots (OER catalysts) located at the upper left region are less active than those located at the lower right region. Therefore, this figure basically shows that the activity of these spinel MnCo2O4 oxides is lower than those of IrO2. Tafel plots are generated based on the cyclic voltammetry (CV) curves, which is the method to record the response of the working electrodes (catalysts) in current (the charge transfer at the interface of electrode and electrolyte) to the applied potential. For a typical OER measurement, the recorded current includes the OER-reaction-associated charges as well as the capacitance contribution. After capacitance correction (and optional iR correction),2Lee Y. Suntivich J. May K.J. Perry E.E. Shao-Horn Y. Synthesis and activities of rutile IrO2 and RuO2 nanoparticles for oxygen evolution in acid and alkaline solutions.J. Phys. Chem. Lett. 2012; 3: 399-404Crossref PubMed Scopus (2565) Google Scholar the CV curve at the OER region (the reaction kinetic current versus the applied potential) then can be plotted as a Tafel plot in a selected potential window. It is interesting to find that the Tafel plots in many articles are different in current density. Some articles use the current density normalized by the catalyst surface area, while some of them use the current density normalized by geometric surface area of working electrodes (disk surface area) in their Tafel plots. This actually generated some confusion within the community. Which one should be used or which one makes more sense? Before answering the above question, one has to look at the difference between these two surface areas. If the catalyst being tested is a flat surface electrode (i.e., a single crystal surface electrode), the surface area of the catalyst can be treated the same as that of the geometric surface area of the electrode.3Stoerzinger K.A. Qiao L. Biegalski M.D. Shao-Horn Y. Orientation-dependent oxygen evolution activities of rutile IrO2 and RuO2.J. Phys. Chem. Lett. 2014; 5: 1636-1641Crossref PubMed Scopus (394) Google Scholar If the catalyst is in particles, especially in the form of nanoparticles, usually they are drop-cast on a flat electrode (i.e., glassy carbon electrode [GCE]) and then used as the working electrode. In this case, the roughness factor (the ratio of the catalyst surface area to the geometric surface area) can be very high and these two surface areas are significantly different.4Sheng W. Chen S. Vescovo E. Shao-Horn Y. Size influence on the oxygen reduction reaction activity and instability of supported Pt nanoparticles.J. Electrochem. Soc. 2012; 159: B96-B103Crossref Scopus (119) Google Scholar Consequently, the Tafel plots can be significantly different. To demonstrate this difference, we designed an experiment with a series of spinel Co3O4 powders with different particle sizes and investigated their OER performance using surface areas obtained by different approaches: (1) geometric surface area of GCE (disk surface area), (2) Co3O4 particle surface area by Brunauer-Emmett-Teller (BET) measurement, and (3) electrochemical surface area by non-Faradic double-layer capacitance (carbon contribution excluded5Wei C. Feng Z. Baisariyev M. Yu L. Zeng Li. Wu T. Zhao H. Huang Y. Bedzyk M.J. Sritharan T. Xu Z.J. Valence change ability and geometrical occupation of substitution cations determine the pseudocapacitance of spinel ferrite XFe2O4 (X= Mn, Co, Ni, Fe).Chem. Mater. 2016; 28: 4129-4133Crossref Scopus (83) Google Scholar). The Co3O4 powders are synthesized by the nitrate decomposition method, followed by annealing at different temperature (300°–600°C). Figure 1A shows the X-ray diffraction (XRD) patterns of as-synthesized Co3O4 powders. They all are pure in the spinel phase. Figure 1B shows the OER CV curves. The inset is the close-up view of the redox region at 1.25–1.55 V versus the reversible hydrogen electrode (RHE). The mass loading of Co3O4 on GCE is kept at 50 μg. These CV curves show that Co3O4 produced at a lower annealing temperature gives higher redox peaks and OER currents. It seems that Co3O4 made at 300°C gives the highest OER activity for Co3O4 and, with the increase of annealing temperature, the OER currents of the Co3O4 exhibit a gradual drop. Note that the annealing temperature difference actually caused the difference in particle size. This is shown in the particle size measured by the BET approach. The BET specific surface areas of samples were measured with an ASAP Tristar II 3020 surface area and porosity analyzer. Nitrogen gas is used as the probe gas, which is a standard approach and allows measurement of the surface areas as low as 0.01 m2 g−1. The BET approach with nitrogen standard may be best suited to particulate catalysts, and it was used as early as 1977 to determine the surface area of a ruthenium dioxide catalyst by Burke et al.6Burke L.D. Murphy O.J. O'Neill J.F. Venkatesan S. The oxygen electrode. Part 8. Oxygen evolution at ruthenium dioxide anodes.J. Chem. Soc., Faraday Trans. 1. 1977; 73: 1659-1671Crossref Google Scholar As a result of particle size decrease, the specific surface area increased gradually from 4.78 to 23.98 m2 g−1 according to BET measurements. The same trend is found in electrochemical surface area measured by non-Faradic double-layer capacitance. The electrochemical surface area of these Co3O4 electrodes increased gradually from 2.07 to 8.33 cm2 (from 4.14 to 16.66 m2 g−1) along with the decrease in particle size. Figure 2A shows the Tafel plots of OER for Co3O4 using the current density normalized by geometric surface area of GCE. It clearly shows the size-dependent OER activity; i.e., Co3O4 with small particle size gives higher OER activity and vice versa, like the size effect reported elsewhere.7Esswein A.J. McMurdo M.J. Ross P.N. Bell A.T. Tilley T.D. Size-dependent activity of Co3O4 nanoparticle anodes for alkaline water electrolysis.J. Phys. Chem. C. 2009; 113: 15068-15072Crossref Scopus (458) Google Scholar However, the Tafel plots using the current density normalized by BET and electrochemical surface areas do not show such remarkable size dependence (Figures 2B and 2C). In fact, the size dependence observed in Tafel plots by geometric surface area of GCE should be a consequence of the particle surface area variation, which is caused by the particle size difference. Figure 2D shows the positive correlation between BET surface area, OER currents by geometric surface area of GCE, and the charge associated with the reduction of Co4+ to Co3+ at the potential range of 1.25–1.55 V versus RHE. Such positive correlations strongly indicate that the increased OER activity (by GCE geometric area) should be ascribed to the smaller size of Co3O4 particles, which gives a higher surface area of the catalyst and more surface Co. However, according to what has been observed in oxygen reduction reactions on Pt,4Sheng W. Chen S. Vescovo E. Shao-Horn Y. Size influence on the oxygen reduction reaction activity and instability of supported Pt nanoparticles.J. Electrochem. Soc. 2012; 159: B96-B103Crossref Scopus (119) Google Scholar the size effect refers to the influence of particle size on the intrinsic activity (the activity normalized by the catalyst surface area, instead of electrode geometric area, which is usually used to exclude the influence of mass loading variation on electrodes). Here, the Tafel plots in Figures 2B and 2C should represent the intrinsic activity of the Co3O4 since the catalyst surface area is used (each represents one approach to obtain catalyst surface area). It is reasonable to say that there is no size-dependent OER for Co3O4. This is probably true for oxide catalysts according to the reported OER mechanism, where the molecular orbital theory of a single metal cation is employed,8Suntivich J. Gasteiger H.A. Yabuuchi N. Nakanishi H. Goodenough J.B. Shao-Horn Y. Design principles for oxygen-reduction activity on perovskite oxide catalysts for fuel cells and metal–air batteries.Nat. Chem. 2011; 3: 546-550Crossref PubMed Scopus (2081) Google Scholar which says the particle size of oxide catalysts has little influence on surface cations' electronic structure. Some exceptions should be noted, such as particle-size-induced spin-state change of active cations on particle surfaces.9Zhou S. Miao X. Zhao X. Ma C. Qiu Y. Hu Z. Zhao J. Shi L. Zeng J. Engineering electrocatalytic activity in nanosized perovskite cobaltite through surface spin-state transition.Nat. Commun. 2016; 7: 11510Crossref PubMed Scopus (253) Google Scholar Figure 3 shows the Tafel plots of OER on Co3O4 made at 300°C, in which the OER currents are normalized by the three surface areas as mentioned above. It can be seen that the three Tafel plots are aligned parallel with the same Tafel slope and the only difference is the current density value. This is not surprising since only the current density is affected by using different surface area values. However, surprisingly, the plot using geometric surface area of GCE is located at the right side of the figure with a distance of ∼0.084 mA cm2 from the other two plots using BET surface area and electrochemical surface area. Such a location difference, based on the principle of the Tafel plot, gives the activity difference. However, the three plots are for the same material. This difference can be also seen in the Tafel plots of Co3O4 made at 400°C, 500°C, and 600°C (not shown). All of them show that the plots using the geometric surface area of GCE will give a high OER activity. As mentioned above, the activity normalized by the geometric surface area of electrodes cannot represent the intrinsic activity of a material. It can only show the activity of the tested electrode from an engineering aspect (more practically meaningful in developing devices). In Figure 3, the Tafel plots of benchmark IrO2 extracted from the literature2Lee Y. Suntivich J. May K.J. Perry E.E. Shao-Horn Y. Synthesis and activities of rutile IrO2 and RuO2 nanoparticles for oxygen evolution in acid and alkaline solutions.J. Phys. Chem. Lett. 2012; 3: 399-404Crossref PubMed Scopus (2565) Google Scholar are also added for comparison purposes. It can be seen that the activity of Co3O4 can be much higher than that of IrO2 if its activity is normalized by GCE geometric area. However, this does not truly reflect the intrinsic activity of these materials. If normalized by oxide particle surface area, a reasonable comparison can be given; i.e., the activity of Co3O4 is lower than that of IrO2. In summary, we have briefly discussed the influence of surface areas on the Tafel plots using the example of Co3O4 particles. We can see that using geometric surface area of electrodes can boost the OER activity to a very high value in Tafel plots. Such boosted activity is an artificial effect and does not reflect the intrinsic activity of a catalyst material. However, it more a reflection of the activity of the tested electrode and is more practically meaningful for water electrolysis devices. This reminds us that the overpotential at 10 mA cm−2 (where disk area is used) should not be used as an activity evaluation standard for OER catalysts, which are, however, is very popular now in the literature. To compare the performance of two catalysts, the activity normalized by catalyst surface area (intrinsic activity) is recommended. Besides BET and non-Faradic capacitance measurements, there are a few other methods, such as under-potential deposition, for measuring the surface area of electrocatalysts electrochemically.10Trasatti S. Petr II, O.A. Real surface area measurements in electrochemistry.Pure Appl. Chem. 1991; 63: 711-734Crossref Scopus (1498) Google Scholar However, to determine the catalyst surface area may be challenging with some novel materials (developing novel materials to catalyze OER is a major effort currently), where the BET method is not applicable and electrochemical surface area is not measurable. More effort may be necessary in exploring alternative approaches to measure the surface area of “novel” catalysts. Z.J.X. acknowledges the funding support from the Singapore Ministry of Education Tier 2 Grants (MOE2017-T2-1-009) and the Singapore National Research Foundation under its Campus for Research Excellence and Technological Enterprise (CREATE) program. Z.J.X. also thanks The First International Symposium on Electrocatalysis and Electrosynthesis, held in Changsha, China, from March 30 to April 1, 2018, in which the issue of activity evaluation in the literature was raised by valued colleagues and students. Download .pdf (.17 MB) Help with pdf files Document S1. Figure S1