NEW INSIGHT INTO THE RATE DETERMINING STEP AND ACTIVE

Subscriber access provided by Bibliothèque de l'Université Paris-Sud

Perspective

NEW INSIGHT INTO THE RATE DETERMINING STEP AND ACTIVE SITES IN THE FISCHER-TROPSCH REACTION OVER COBALT CATALYSTS Robert Pestman, Wei Chen, and Emiel J.M. Hensen ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.9b00185 • Publication Date (Web): 01 Apr 2019 Downloaded from http://pubs.acs.org on April 1, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

INSIGHT INTO THE RATE DETERMINING STEP AND ACTIVE SITES IN THE FISCHERTROPSCH REACTION OVER COBALT CATALYSTS

Robert Pestman,* Wei Chen, Emiel Hensen

Laboratory of Inorganic Materials and Catalysis, Department of Chemical Engineering and Chemistry, Eindhoven University of Technology, Groene Loper 3, 5612 AE Eindhoven, The Netherlands

Corresponding author: [email protected]

Abstract

ACS Paragon Plus Environment

1

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 29

Most studies on the Fischer-Tropsch reaction assume that the dissociation of the C-O bond is crucial in determining the overall reaction rate. However, recent experimental results show that a hydrogenation step is crucial in the overall kinetics. At low pressures, which are typically used in academic research, the structure independent termination by hydrogenation is dominating the reaction rate. This is reflected in a particle size- and structure-independent apparent activation energy and is confirmed by kinetic modelling of transient experiments. At higher (i.e., industrially more relevant) pressures, both the availability of appropriate dissociation sites and the removal of adsorbates by hydrogenation appear to limit the rate. This results in a comparable degree of rate control for CO dissociation as well as hydrogenation. At low pressures, the locus of termination by hydrogenation has been studied by selective site blocking of planar sites with graphene and by using nanoparticles exposing specific crystal planes. It is found that the termination runs mainly on planar surfaces, while corrugated surfaces contribute to chain growth, which follows the ASF distribution. Migration of CHx species between these two sites is foreseen. A shortage of stepped sites needed for monomer formation limits the yield of both methane and longer hydrocarbons. We propose that, as long as the number of stepped CO dissociation sites is sufficient, the overall rate is dominated by the structure insensitive hydrogenation. Otherwise, the turnover frequency follows the occurrence of CO dissociation sites.

Keywords: Fischer-Tropsch, CO hydrogenation, active site, kinetics, activation energy, rate-limiting step.


2

Page 3 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

1. Introduction

The Fischer-Tropsch synthesis of hydrocarbons from synthesis gas has already been discovered a century ago in an attempt to provide war-time Germany with liquid fuels. Recently, the process gained new interests as a method to produce clean liquid fuels from natural gas in remote locations, readily available coal, or biomass(1)-(2) Despite its broad applications and long history, the mechanism of the Fischer-Tropsch synthesis is still subject to intense debate due to its high complexity. A basic question for catalyst improvement is to identify the kinetically most relevant reaction steps and their corresponding reaction sites. Recent findings on the rate determining steps and active sites in the Fischer-Tropsch reaction are discussed here. It is generally known that the Fischer-Tropsch reaction is structure sensitive. This is exemplified by the dependency of the turnover frequency (TOF) on the particle size (Figure 1). This particle size dependence is found for a whole range of reaction conditions, such as temperatures from 185°C to 240°C and pressures up to 35 bar(3), This dependency is not only found for cobalt on different oxidic supports, but also for cobalt supported on inert carbon. Noticeably, metal-support interaction do not effect this trend.


3

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 29

Figure 1. Trend in turnover frequency (TOF) for the Fischer-Tropsch reaction, based on CO consumption, versus particle size for Co catalysts(4)-(15)

The complex Fischer-Tropsch mechanism consists of numerous steps, like C-O bond dissociation, chain growth by C-C coupling, several (de)hydrogenation steps, and termination, which can all be structure sensitive to some extent. The C-O bond dissociation reaction, which involves the breaking of a π-bond is well known to be structure sensitive and is generally believed to occur on stepped sites, such as B5 sites. This is supported by surface science studies showing the need of stepped sites(16),(18) and multiple DFT based theoretical studies, which show that stepped sites are preferred for CO dissociation, either in a hydrogen-assisted or a direct way(19)-(22). The requisite of stepped B5 sites is analogous to the earlier studied hydrogenation of the isoelectronic N2 to ammonia, for which the need and the topological arrangement of B5 sites


4

Page 5 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

have been described in detail before(23),(24). In contrast, the formation of a σ-bond, like the C-H bond during hydrogenations is mostly assumed to be practically structure insensitive(25),(26). The dissociation of the strong C-O bond is often assumed to be the most demanding step in the Fischer-Tropsch reaction. This would nicely comply with the structure sensitivity of the reaction. However, this assumption has never been supported directly by experimental results. Only DFT calculations seem to support it(27),(28). On the contrary, recent experimental research offers indications that not the CO dissociation, but, depending on the pressure, another reaction step could be rate determining,

2. Insight into Fischer-Tropsch kinetics Due to the presence of wax, water, and internal diffusion limitations, it is hard to acquire intrinsic kinetic information at industrial Fischer-Tropsch conditions. At low pressures (or at high temperatures) no long-chain hydrocarbons are formed and kinetics can be studied better. Moreover, kinetics of individual reaction steps can be studied at low pressure by transient kinetic analysis, such as steady state isotopic transient kinetic analysis (SSITKA). In addition, active sites can be studied by semi-ambient pressure surface science techniques. In this section results of fundamental research at low pressures will be discussed first. Afterwards, extrapolation of this knowledge to higher, industrially more relevant, pressures will be done.

2.1. Fischer-Tropsch kinetics at low pressure The oldest experimental indications that not the CO dissociation but another reaction step could be rate determining are based on single crystal research at ambient pressure(16). It was observed that although the reaction rate in the Fischer-Tropsch synthesis depends on the surface


5

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 29

structure, corrugated single crystals being more active, the apparent activation energy is independent on either the surface structure or the alkane formed (see figure 2). The higher activity of the corrugated surfaces was explained by the larger availability of adsorbed carbon species due to the increased CO dissociation on stepped sites. The equal apparent activation energy was explained by a common rate determining step, which was assigned to be the hydrogenation step needed for termination to methane or alkane. Some other publications stemming from the same period support this view(29),(30). Termination being rate determining seems obvious, as significant chain growth can only occur if termination is slow compared to all other preceding reaction steps: termination has to be slower than CO dissociation and chain growth in order to obtain long-chain hydrocarbons and slower or comparable to CO dissociation to prevent a shortage or built-up of carbon monomers on the surface. Van Santen et al. defined the former constraint as the chain growth limit regime and the latter as the transition to a monomer formation limit regime(31).


6

Page 7 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

Figure 2. Arrhenius plot of methane ( ) and ethane ( ) formation after 2.5 h run time for close-packed (0001) and zig-zag grooved (11-20) cobalt surfaces. Redrawn from reference 16, with permission from Elsevier.

More recent studies on supported catalysts at low pressures (typically below 2 bar) also point towards another rate determining step than CO dissociation(32), likely termination by hydrogenation(14). Chen et al. demonstrated that on an empty surface CO dissociation is fast enough to keep up with hydrogenation(33). Furthermore, apparent activation energy measurements performed by several groups using supported catalysts with different particle sizes demonstrate a particle size independent activation energy(10),(14),(34), suggesting a common rate determining step for all particles sizes. A likely candidate for this reaction step would be the structure insensitive hydrogenation of adsorbed alkyl species, consisting of a σ-bond formation and leading to termination. Recently, our group performed a degree of rate control (DRC) analysis based on a kinetic model obtained from transient kinetics experiments(35). A DRC analysis gives insight into which reaction steps influence the overall reaction rate most by quantifying their relative influence on the rate(36). In a complex network of reactions, multiple elementary reaction steps can contribute partly to the overall rate control. From our DRC analysis, it appeared that, depending on the conditions, multiple reactions participate in the control of the overall rate for the reaction from CO to methane. It was shown that at low CO partial pressure, as mostly used in academic research, the hydrogenation controls the reaction rate for 60%, while the CO dissociation contributed for less than 10% (other less rate controlling elementary reaction steps being


7

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 29

adsorption or oxygen removal by water formation) confirming the predominance of hydrogenation as the rate-controlling elementary reaction step. After identifying the termination by hydrogenation as the rate determining step, H-D exchange measurements can be used to gain detailed insight into the kinetics of the elementary steps involved in the hydrogenation reaction to methane. For this purpose, the experiments in Ref. (35) were extended using the same catalyst and reaction setup. The methanation reaction was run for 20 h at 260 °C, followed by a switch from the initial D2/CO feed to a H2/He feed. The transient responses of CDH3,CD3H and CH4 after the switch are shown in Figure 3. From these data, it can be calculated that, at 45 kPa H2, the time needed for the last single step of CH3* to gas phase CH4 is longer (6 s) than the time needed for all preceding hydrogenation steps together (the three consecutive hydrogenation steps from C* to CH3* take 4.5 s in total). Accordingly, the last hydrogenation step to methane and accompanying desorption can be appointed as the rate determining step in the methanation reaction.


8

Page 9 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

Figure 3. CD3H (upper), CH3D (middle), and CH4 (lower) transients after a switch from D2/CO to H2/He with varying H2 partial pressure.

: 𝑝𝐻2= 90 kPa;

: 𝑝𝐻2= 45 kPa;

: 𝑝𝐻2=27 kPa. In

all cases, 𝑝𝐶𝑂 = 9 kPa, T = 260 °C. The signals were recorded by an inline mass spectrometer (the signal of CH2D2 and CD4 could not be identified because they overlapped with the signals of H2O and D2O). Since each product containing deuterium had to be present on the catalyst surface before the switch, the time needed to fully hydrogenate and desorb an adsorbed CDx can be deduced from the response times of the corresponding CDxH(4-x) compounds. The difference between the response times can be used to calculate the reaction time of individual hydrogenation steps.


9

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 29

2.2. Fischer-Tropsch kinetics at high pressure As stated in above, the degree of rate control at low pressures is much lower for CO dissociation (10%) than for hydrogenation (60%). However, the degree of rate control contributions of each elementary reaction step changes with increasing pressure(35). For example, at the industrially more relevant total pressure of 20 bar, the degree of rate control is predicted to change to 30% for hydrogenation and about 40% for CO dissociation. Evidently, the CO dissociation becomes a more important factor in determining the kinetics at high pressure. Its degree of rate control reaches the same level as that for hydrogenation. CO dissociation is apparently hindered at higher pressures. Likely, the higher coverage of reaction intermediates resulting from a higher CO and lower hydrogen coverage suppresses the amount of vacancies needed to go from one CO adsorbate to the two O and C adsorbates. Notably, the activation energy for CO dissociation is, according to DFT simulations, practically independent of the CO coverage(36), only the amount of available sites is limited at increasing pressure. After CO is dissociated on a stepped B5 site, the carbon and oxygen fragments have to be removed in order to regenerate an empty site needed for closing the catalytic cycle. Hydrogenation of adsorbed C to CHx is needed to remove the carbon from stepped site(38). Theoretical DFT calculations claim that the hydrogenation of adsorbed carbon needed to create new vacancies is rate determining with an activation energy of 88 kJ/mol(40), which is in the same range as the experimentally found activation energy for the overall reaction (10),(14),(34). The necessity of site regeneration results in a still significant contribution to the rate control for a hydrogenation step. The presence of hydrogen is indeed found to enhance the removal of carbon or oxygen from the surface(35). As soon as a CO dissociation site is regenerated by


10

Page 11 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

hydrogenation of the adsorbed carbon, another adsorbed CO, which is plentifully available, can be dissociated on the liberated site. This differs from the situation at low pressures, when the total coverage is lower and the availability of free sites is not limiting. Under these conditions, the hydrogen is only needed for the rate determining termination by hydrogenation. With this knowledge the reaction orders in the Fischer-Tropsch reaction can be explained. Notably, reaction orders should not be confused with the degree of rate control for individual steps. They are however related to each other. At low pressure, the positive order in hydrogen for the Fischer-Tropsch reaction is a logical consequence of hydrogenation being rate determining. The negative order in CO has an indirect reason: CO adsorbs stronger than hydrogen. Hence an increase in CO pressure reduces the hydrogen coverage and thus the reaction rate. Also at high pressures a negative order in CO and a positive order in hydrogen is observed. A positive order in hydrogen is found as hydrogen is needed to remove adsorbed carbon by hydrogenation(40). The reaction order in CO is negative as adsorbed CO reduces the number of free dissociation sites as well as the coverage of hydrogen, which is needed to regenerate the dissociation sites. The need of hydrogen appears to be more relevant than the obstruction by CO. This can be deduced from the finding that the reaction order in H2 is higher than the absolute reaction order in CO(35). Please note that as hydrogenation and CO dissociation have a similar degree of rate control at high pressure, the negative order in CO can not only be explained by a decrease in free CO dissociation sites at an increased CO coverage but also by a decrease in adsorbed hydrogen. The positive effect of high pressure on the selectivity towards longer hydrocarbons can now tentatively be explained: The high coverage of CO and intermediates that is related to high pressures results in a lower hydrogen coverage and hence lower termination rates, but results


11

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 29

also in less free sites that enable reversed chain growth(41). In total, a higher chain growth probability is expected. As we are now gaining more insight into the kinetically relevant reaction steps in the FischerTropsch reaction, the challenge arises to identify the associated reaction sites. Below we will discuss briefly the most important reaction sites on a Fischer-Tropsch catalyst.

3. Insight into the active sites in the Fischer-Tropsch reaction

3.1. Propagation and termination sites. To identify the reaction site for the termination by hydrogenation, Chen et al. deposited graphene on the planar sites of cobalt particles(42). Remarkably, an increasing coverage of the surface by graphene suppresses the methane yield and the hydrogenation reactions (see Figure 4), which could either be the termination to alkanes by α-hydrogenation or the secondary hydrogenation of olefins. It was therefore concluded that planar sites are largely responsible for methane formation and hydrogenation, while stepped sites, which are blocked to a lesser extent by the graphitic layer, are responsible for the propagation reaction leading to longer hydrocarbons.


12

Page 13 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

Figure 4. Methane yield, olefin to paraffin ratio, and chain growth probability α, depending on the graphene deposits on Co particles conditions: pH2 = pCO = 200 mbar, T = 260 °C (adapted from reference (42) with permission of the American Chemical Society).

This is in accordance with previous single crystal research showing that different types of sites lead to other products: stepped sites enable the buildup of higher hydrocarbon adsorbates, while planar sites harbor mainly adsorbed C1 species(16). A recent impressive study by Qin et al. supported this view in an even stronger way: the authors performed experiments with purposively synthesized nanoparticles exposing predominantly either corrugated surfaces or close-packed planar surfaces. It appears that the particles exposing stepped sites produce mainly long-chain hydrocarbons while those exposing planar surfaces yield mostly methane(43).


13

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 29

3.2. The monomer formation site. Both methanation and chain growth need a C1 adsorbate as intermediate. Roughly, two kinds of mechanisms have been postulated for the Fischer-Tropsch reaction, one of them is the socalled carbene mechanism proposed by Sachtler and Biloen(44). In the carbene mechanism, CO is dissociated to form adsorbed carbon, which is hydrogenated to CHx and subsequently inserted into a growing alkyl chain. A second mechanism is the so-called CO insertion mechanism, proposed by Pichler and Schulz(41). In this CO insertion mechanism, adsorbed CO is inserted into the growing alkyl chain and the C-O bond is dissociated afterwards. We base our discussion on the carbene mechanism which finds most following in literature(46),(47). We therefore assume that adsorbed CHx is the C1 intermediate needed for both methanation and chain growth. On the basis of the equation for turnover frequency as applied by the group of the De Jong for describing their SSITKA measurements3, the following equation can be derived, in which k is the first-order reaction rate constant

―Ea

(1)

TOF = k ∙ θCHx = A ∙ e RT ∙ θCHx

The reaction rate constant is expressed according to the Arrhenius equation, in which A is the pre-exponential factor and Ea the apparent activation energy. In addition to the number of active sites, the factor A contains entropic effects, variations of which are negligible in the case of a hydrogenation reaction that is confined to a catalytic surface. Above it is argued that the apparent activation energy Ea is independent of surface structure or particle size. Hence, the TOF is


14

Page 15 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

linearly dependent on the surface coverage of CHx adsorbates. Consequently, it is essential to elucidate the CHx formation further in order to explain the structure dependency of the TOF. Van Helden et al.(48) calculated by using energy-optimized constructed Co nanoparticles that the relative occurrence of stepped sites like B5 does not have a maximum as would be expected based on the ideal Wulff constructions of nanoparticles(23), but levels off above a certain size (see figure 5). Agrawal and coworkers used molecular dynamics simulations with the embeddedatom-method to calculate the number of B5 sites(49). Though the amount of stepped B5 sites as calculated by Van Helden and Agrawal using dissimilar methods differ, the trend of both follows the TOF depicted in Figure 1. In combination with the single crystal and DFT based literature, mentioned in the introduction showing that CO dissociation likely occurs on stepped sites like B5, it seems reasonable that the occurrence of B5 sites can directly be related to CO dissociation generating C1 species, which on their turn can be related directly to the TOF (see equation 1) implying that the B5 site is responsible for the monomer generation by CO dissociation and that the TOF subsequently follows the occurrence of these sites. The occurrence of CO dissociation sites as predicted by Van Helden and Agrawal is based on theoretical calculations on rigid nanoparticles. It should be noted that adsorbate induced surface reconstruction can occur(39). Surface reconstruction leads to the formation of stepped sites on planar surfaces. Banerjee et al.(40) calculated that these reconstructed surface sites are still active. Hence, reconstruction does not lead to another explanation for the trend in the TOF. Large particles exhibit, independent on the particle size, some activity. This can be explained by the presence of stepped sites that are either present as calculated on rigid particles or formed by surface reconstruction of planar surfaces. Small particles below 6 nm do not have the capability of hosting B5 sites, whether they are reconstructed or not. Experimental proof of the presence of


15

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 29

active CO dissociation sites on large particles has been provided by exchange experiments between double (isotopically) labeled and non-labeled CO over 20 nm Co, confirming the presence of sites that dissociate CO on large particles(35). Hence, the TOF can be related to the occurrence of B5 sites, which dissociate CO. However, as discussed above, the CO dissociation is not the rate determining step at low pressures. The reason why the TOF still follows the occurrence of B5 sites, is thus not because they facilitate the reaction step with the highest activation energy, but because they provide the necessary carbon monomers, the availability of which controls the reaction rate according to equation 1. This implies that the TOF only follows the occurrence of B5 sites because the coverage of reactants is directly correlated to it. The correlation between the CHx coverage and the TOF has been established earlier by SSITKA measurements3.


16

Page 17 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

Figure 5. Relative occurrence of B5 sites as percentage of the total amount of surface atoms. Based on data of Van Helden et al.( )(48) and Agrawal et al.( )(49).

Notably, the chain growth probability as shown in figure 4 increases in parallel with the drop in methane formation(42). So the increasing yield of one reaction, methanation, goes at the expense of the other reaction, chain growth. This suggests that both reactions compete for the same reactant, CHx, which must thus be able to migrate over the catalyst surface in order to supply the different sites with enough reactant. Indeed, according to theoretical calculations, carbon formed on stepped sites can migrate, after hydrogenation over the surface to a neighboring site(36),(40),(50), just as observed experimentally for adsorbed hydrocarbons(53). After CHx is being formed by CO dissociation on a B5 site, it can thus either migrate to planar sites and react to methane, or stay on the stepped site and get involved into chain growth. With the same graphene deposition experiments as presented in Figure 4, Chen et al. demonstrated that extra ASF methane (ASF = Anderson-Schulz-Flory) was reduced after covering the planar sites, while the longer hydrocarbons followed the ASF distribution(42) (see figure 6) implying that the CHx species staying on the stepped sites lead to an ASF-like chain growth.


17

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 29

Figure 6. ASF distribution depending on the graphene coverage (Cgraphitic/Cosurf.) Cgraphitic/Cosurf. = 0.11 ( ); = 0.26 ( ); = 0.48( ); = 0.91( ) (adapted from reference (42) with permission of the American Chemical Society).

Based on the literature used for Figure 1, the production rates of methane or longer hydrocarbons per surface area, i.e., the respective site time yields, were plotted in Figure 7. As the rates of two reactions, which run on different sites but which both need adsorbed CHx species drop in parallel at decreasing particle size, there is evidently not a shortage of specific sites for either methane formation or chain growth, but a shortage of adsorbed CHx on smaller particles.


18

Page 19 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

Figure 7: (A) Site time yield of C5+ hydrocarbons (TOFC5 + ); (B) Site time yield of methane ( TOFCH4). Site time yields are based on total surface time yields (or TOF) and selectivities to either methane(6),(7),(12),(15),(54) or C5+(3),(7),(8),(11),(12),(15),(55)

As discussed above, an increase in pressure causes an increase in CO coverage accompanied by a decrease in hydrogen coverage. These changes are the reason for the changed kinetics. However, as the nature of the elementary reaction steps in the Fischer-Tropsch reaction is not expected to change at increasing pressure it can be stated that at all pressures stepped sites are responsible for CO dissociation and chain growth while planar sites contribute to additional methane formation.

4. Conclusions and perspective We discussed the kinetics of the Fischer-Tropsch reaction. At pressures below 2 bar, which are typically used in fundamental research, the apparent activation energy appears to be independent of particle size as well as surface structure, suggesting that a structure insensitive reaction forms the rate determining step. We propose that this rate controlling reaction step is a hydrogenation


19

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 29

step, likely termination by hydrogenation, and not the commonly assumed CO dissociation step, because the formation of σ-bonds is, in contrast to the dissociation of π-bonds, expected to be structure insensitive. A degree of rate control analysis of transient experiments confirms the dominating effect of hydrogenation on the overall reaction rate at low pressures, while at higher pressures CO dissociation controls the reaction rate in parallel with hydrogenation. The role of the hydrogenation step seemingly changes with pressure: at low pressure termination by hydrogenation is rate limiting, while at high pressure removal of adsorbed carbon by hydrogenation in order to regenerate the active site appears to be rate limiting in combination with the CO dissociation. After identifying the critical reaction steps controlling the reaction rate, we pinpointed the associated active sites at low pressures. Based on selective site blocking and information gained from single crystals as well as nanoparticles exposing predominantly specific surface structures, we argue that stepped sites are responsible for chain growth while planar sites contribute to additional methane formation. The occurrence of B5 sites follows the same trend as the TOF. In combination with literature that assigns B5 sites as the active site for CO dissociation, we propose that the TOF follows the availability of CHx adsorbates generated on CO dissociation sites, even though a structure insensitive termination by hydrogenation reaction has the highest activation energy. To account for the yield trends when covering specific sites, the monomers have to be mobile over the surface and can either migrate to planar sites where hydrogenation to methane occurs or stay on the stepped sites where they can get involved in chain growth. The consequence of a mobile monomer is that a shortage of it will lead to a reduced production of all Fischer-Tropsch products.


20

Page 21 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

With this perspective we hope to emphasize that, when kinetics of catalytic reactions are studied, not only the reaction with the highest activation energy controls the rate, but that the whole picture of surface coverages and active site availabilities, which might be pressure dependent, should be taken into account. Here, further insight into the Fischer-Tropsch mechanism was gained by combining apparent activation energies and degree of rate control analyses with the availability of reaction sites and reactants.

References: (1)

Van de Loosdrecht, J.; Botes, F. G.; Ciobîcă, J. M.; Ferreira, A.; Gibson, P.; Moodley, D. J.; Saib, A. M.; Visagie, J. L.; Weststrate, C. J.; Niemantsverdriet, J. W. Fischer-Tropsch Synthesis: Catalysts and Chemistry Compr. Inorg. Chem. II, Eds. Reedijk, J.; Poeppelmeier, K., 2013, 7, 525-557

(2)

Kelly, G.J. Fischer-Tropsch –carbon monoxide hydrogenation Hydrogenation: Catalysts and Processes, Ed. Jackson, D. 2018, 155-172

(3)

Breejen, J. P.; Radstake, P. B.; Bezemer, G. L.; Bitter, J. H.; Frøset, V.; Holmen, A.; De Jong, K. P. On the Origin of the Cobalt Particle Size Effects in Fischer-Tropsch Catalysis J. Am. Chem. Soc. 2009, 131, 7197-7203

(4)

Luo, Q.-X.; Guo, L.-P.; Yao, S.-Y.; Bao, J.; Liu, Z.-T.; Liu, Z.-W. Cobalt nanoparticles confined in carbon matrix for probing the size dependence in Fischer-Tropsch synthesis J. Catal. 2019, 369, 143-156


21

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(5)

Page 22 of 29

Bezemer, G. L.; Bitter, J. H.; Kuipers, H. P.C. E.; Oosterbeek, H.; Holewijn, J. E.; Xu, X.; Kapteijn, F.; Van Dillen, A. J.; De Jong, K. P. Cobalt Particle Size Effects in the Fischer-Tropsch Reaction Studied with Carbon Nanofiber Supported Catalysts J. Am. Chem. Soc. 2006, 128, 3956-3964

(6)

Breejen, J. P.; Sietsma, J. R.A.; Friedrich, H.; Bitter, J. H.; De Jong, K. P. Design of supported cobalt catalysts with maximum activity for the Fischer–Tropsch synthesis J. Catal. 2010, 270, 146–152

(7)

Prieto, G.; Martínez, A.; Concepción, P.; Moreno-Tost, R. Cobalt particle size effects in Fischer–Tropsch synthesis: structural and in situ spectroscopic characterisation on reverse micelle-synthesised Co/ITQ-2 model catalysts J. Catal. 2009, 266, 129–144

(8)

Fischer, N.; Van Steen, E.; Claeys, M. Structure sensitivity of the Fischer–Tropsch activity and selectivity on alumina supported cobalt catalysts J. Catal. 2013, 299, 67–80

(9)

Ho, S-W.; Cruz, J. M.; Houalla, M.; Hercules, D.M. The Structure and Activity of Titania Supported Cobalt Catalysts J. Catal. 1992, 135, 173-185

(10) Herranz, T.; Deng, X.; Cabot, A.; Guo, J.; Salmeron, M. Influence of the Cobalt Particle Size in the CO Hydrogenation Reaction Studied by In Situ X-Ray Absorption Spectroscopy J. Phys. Chem. B 2009, 113, 10721–10727 (11) Xiong, H.; Motchelaho, M. A. M.; Moyo, M.; Jewell, L. L.; Coville, N. J. Correlating the preparation and performance of cobalt catalysts supported on carbon nanotubes and carbon spheres in the Fischer–Tropsch synthesis J. Catal. 2011,278, 26–40


22

Page 23 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

(12) Cheng, Q.; Tian, Y.; Lyu, S.; Zhao, N.; Ma, K.; Ding, T.; Jiang, Z.; Wang, L.; Zhang, J.; Zheng, L.; Gao, F.; Dong, L.; Tsubaki, N.; Li, X. Confined small-sized cobalt catalysts stimulate carbon-chain growth reversely by modifying ASF law of Fischer–Tropsch synthesis Nat. Comm. 2018, 9,1-9 (13) Martínez, A.; Prieto, G. Breaking the dispersion-reducibility dependence in oxidesupported cobalt Nanoparticles J. Catal. 2007, 245, 470–476 (14) Yang, J.; Tveten, E. Z.; Chen, D.; Holmen, A. Understanding the Effect of Cobalt Particle size on Fischer-Tropsch synthesis: Surface Species and Mechanistic Studies by SSITKA and Kinetic Isotope Effect Langmuir 2010, 26, 16558–16567 (15) Eschemann, T. O.; Lamme, W. S.; Manchester, R. L.; Parmentier, T. E.; Cognigni, A.; Rønning, M.; De Jong, K. P. Effect of support surface treatment on the synthesis, structure, and performance of Co/CNT Fischer–Tropsch catalysts J. Catal. 2015, 328,130–138 (16) Geerlings, J. J. C.; Zonnevylle, M. C.; De Groot, C. P. M. Structure sensitivity of the Fischer-Tropsch reaction on cobalt single crystals. Surf. Sci. 1991, 241, 315-324 (17) Prior, K. A.; Schwaha, K.; Lambert R. M. Surface chemistry of the non-basal planes of cobalt: the structure, stability, and reactivity of Co(10-12)-CO. Surf. Sci. 1978, 77, 193208 (18) Geerlings, J. J. C.; Wilson, J. H.; Kramer, G. J.; Kuipers, H. P. C. E.; Hoek, A.; Huisman, H. M. Fischer-Tropsch technology-from active site to commercial process. Appl. Catal., A 1999, 186, 27–40


23

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 29

(19) Qi, Y.; Yang, J.; Chen, D.; Holmen, A. Recent Progresses in Understanding of Co-Based Fischer-Tropsch Catalysis by Means of Transient Kinetic Studies and Theoretical Analysis. Catal. Lett. 2015, 145, 145-161. (20) Gong, X-Q.; Raval, R.; Hu, P. CO dissociation and O removal on Co(0001): a density functional theory study. Surf. Sci. 2004, 562, 247–256 (21) Ge, Q.; Neurock, M. Adsorption and activation of CO over flat and Stepped Co Surfaces: A First Principles Analysis. J. Phys. Chem. B 2006, 110, 15368-15380 (22) Petersen, M. A.; Van den Berg, J. A.; Ciobîcă, I. M.; Van Helden, P. Revisiting CO Activation on Co Catalysts: Impact of Step and Kink Sites from DFT. ACS Catal. 2017, 7, 1984−1992 (23) Van Hardeveld, R.; Van Montfoort, A. The influence of crystallite size on the adsorption of molecular nitrogen on nickel, palladium and platinum, an infrared and electronmicroscopic study. Surf. Sci. 1966, 4, 396-430 (24) Honkala, K; Hellman, A; Remediakis, I.N.; Logadottir, A.; Carlsson, A.; Dahl, S.; Christensen, C.H.; Nørskov, J. K. Ammonia Synthesis from First-Principles Calculations Science 2005, 307, 555-558 (25) Liu, Z.-P.; Hu, P. General Rules for Predicting Where a Catalytic Reaction Should Occur on Metal Surfaces: a Density Functional Theory Study of C-H and C-O bond breaking/making on Flat, Stepped, and Kinked Metal Surfaces. J. Am. Chem. Soc. 2003, 125, 1958-1967


24

Page 25 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

(26) Van Santen, R. A. Complementary Structure Sensitive and Insensitive Catalytic Relationships Acc. Chem. Res. 2009, 42, 57-66 (27) Van Santen, R. A.; Ghouri, M. M.; Shetty, S.; Hensen, E. M. H. Structure sensitivity of the Fischer-Tropsch reaction: molecular kinetics simulations Catal. Sci. Technol. 2011, 1, 891-911 (28) Xu, Y. C.; Tian, P.; Cao, Y.; Xu, J.; Han, Y.-F. Application of DFT modeling in FisherTropsch synthesis over Co-based catalysts Chem. Modell., 2016, 12, 184-218 (29) Lahtinen, J.; Ankaru, T.; Somorjai, G.A. Carbon monoxide Hydrogenation on Cobalt Foil and on Thin Cobalt Film Model Catalysts. J. Catal. 1993,142, 206-225 (30) Geerlings, J. J. C.; Zonnevylle, M. C.; De Groot, C. P. M. Studies of the Fischer-Tropsch reaction on Co( 0001) Surf. Sci. 1991, 241, 302-314

(31) Van Santen, R. A.; Markvoort, A. J. Catalyst nano-particle size dependence of the FischerTropsch reaction Faraday Discuss. 2013, 162, 267-279

(32) Van Steen, E.; Schulz, H. Polymerisation kinetics of the Fischer-Tropsch CO hydrogenation using iron and cobalt based catalysts. Appl. Catal., A 1999, 186, 309–320 (33) Chen, W.; Zijlstra, B.; Filot, I. A. W.; Pestman, R.; Hensen E. J. M. Mechanism of Carbon Monoxide Dissociation on a Cobalt Fischer-Tropsch Catalyst. ChemCatChem 2018, 10, 136– 140 (34) Rebmann, E; Fongarland, P.; Lecocq, V.; Diehl, F.; Schuurman, V. Measurement of the Influence of the Microstructure of Alumina-Supported Cobalt Catalysts on their Activity


25

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 29

and Selectivity in Fischer–Tropsch synthesis by using Steady-State and Transient Kinetics ChemCatChem, 2017, 9, 2344 – 2351 (35) Chen, W.; Pestman, R.; Zijlstra, B.; Filot, I. A. W.; Hensen, E. J. M. Mechanism of CobaltCatalyzed CO Hydrogenation: 1. Methanation ACS Catal. 2017, 7, 8050−8060

(36) Zijlstra, B.; Broos, R.J.B.; Chen, W.; Oosterbeek, H.; Filot, I. A.W.; Hensen, E.J.M. Coverage effects in CO dissociation on Metallic Cobalt Nanoparticles. submitted for publication (37) Campbell, C. T. Finding the Rate-Determining Step in a Mechanism: Comparing DeDonder relations with the “Degree of Rate Control” J. Catal. 2001, 204, 520–524

(38) Shetty, S. G.; Ciobîcă, I. M.; Hensen, E. J. M.; Van Santen, R. A. Site regeneration in the Fischer–Tropsch synthesis reaction: a synchronized CO dissociation and C–C coupling pathway Chem. Commun., 2011, 47, 9822–9824

(39) Wilson, J.; De Groot, C. Atomic-Scale Restructuring in High-Pressure Catalysis J. Phys. Chem. 1995,99, 7860-7866 (40) Banerjee, A.; Van Bavel, A.P.; Kuipers, P.C.E.; Saeys, M. CO activation on realistic cobalt surfaces: kinetic role of hydrogen ACS Catal. 2017, 7, 5289-5293

(41) Chen, W.; Filot, I. A. W.; Pestman, R.; Hensen, E. J. M. Mechanism of Cobalt-Catalyzed CO Hydrogenation: 2. Fischer–Tropsch Synthesis ACS Catal. 2017, 7, 8061-8071

(42) Chen, W.; Kimpel, T. F.; Song, Y.; Chiang, F.-K.; Zijlstra, B.; Pestman, R.; Wang, P.; Hensen, E. J. M. Influence of Carbon Deposits on the Cobalt-Catalyzed Fischer-Tropsch reaction: Evidence of a Two-Site Reaction Model ACS Catal.,2018, 8, 1580-1590


26

Page 27 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

(43) Qin, C.; Hou, B.; Wang, J.; Wang, Q.; Wang, G.; Yu, M.; Chen, C.; Jia, L.; Li, D. Crystal-PlaneDependent Fischer-Tropsch Performance of Cobalt Catalysts, ACS Catal., 2018, 8,9447-9455. (44) Biloen, P.; Helle, J. N.; Sachtler, W. M.H. Incorporation of Surface Carbon into Hydrocarbons during Fischer-Tropsch Synthesis: Mechanistic Implications. J. Catal. 1979, 58, 95-107

(45) Pichler, H. Ein Rückblick auf 25 Jahre Forschungsarbeit über die Fischer-Tropsch-Synthese und verwandte Verfahren unter besonderer Berücksischtigung des in Frage kommenden Reaktionsmechanismus. Brennstoff-Chemie 1952, 17/18, 289-320

(46) van Santen, R. A.;. Ciobîcă, I. M.; Van Steen, E.; Ghouri, M. M. Mechanistic Issues in FischerTropsch Catalysis. Adv. Catal. 2011, 54, 27-187

(47) Jacobs, G.; Davis, B. H. Applications of isotopic tracers in Fischer-Tropsch synthesis. Catal. Sci. Technol. 2014, 4, 3927-3944

(48) Van Helden, P.; Ciobîcă, I. M.; Coetzer, R. L. J. The size-dependent site composition of FCC cobalt nanocrystals. Catal. Today 2016, 261, 48–59

(49) Agrawal, R.; Phatak, P.; Spanu, L. Effect of phase and size on surface sites in cobalt nanoparticles Catal. Today 2018, 312, 174-180

(50) Cheng, J.; Hu, P.; Ellis, P.; French, S.; Kelly, G.; Lok, C.M. Some Understanding of FischerTropsch Synthesis from Density Functional Theory Calculations. Top. Catal. 2010, 53, 326–337

(51) Swart, J. C. W.; Ciobîcă, I. M.; Van Santen, R. A.; Van Steen, E. Intermediates in the Formation of Graphitic Carbon on a Flat FCC-Co(111) Surface. J. Phys. Chem. C 2008, 112, 12899–12904

(52) Joos, L.; Filot, I. A.W.; Cottenier, S.; Hensen, E. J. M.; Waroquier, M.; Van Speybroeck, V.; Van Santen, R. A. Reactivity of CO on Carbon-Covered Cobalt Surfaces in Fischer-Tropsch Synthesis J. Phys. Chem. C. 2014, 118, 5317-5327


27

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 29

(53) Somorjai, G.A.; Marsh, A. Active sites and states in the heterogeneous catalysis of carbon– hydrogen bonds Phil. Trans. R. Soc. A 2005, 363, 879–900

(54) Borg, Ø; Eri, S.; Blekkan, E. A.; Storsæter, S.; Wigum, H.; Rytter, E.; Holmen, A. Fischer– Tropsch synthesis over γ -alumina-supported cobalt catalysts: Effect of support variables J. Catal. 2007, 248, 89–100

(55) Wang, Z.-J.; Skiles, S.; Yang, F.; Yan, Z.; Goodman, D. W. Particle size effects in Fischer– Tropsch synthesis by cobalt Catal. Today 2012, 181, 75– 81


28

Page 29 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

TABLE OF CONTENTS ONLY


29

NEW INSIGHT INTO THE RATE DETERMINING STEP AND ACTIVE

Recommend Documents