Hydrogen Partitioning as a Function of Time-on-Stream for an Unpromoted Iron-Based Fischer − Tropsch Synthesis Catalyst Applied to CO Hydrogenation

.


INTRODUCTION
−3 The hydrocarbon product output is dependent on not only the reaction conditions employed but also the catalyst used. 4Current commercial processes operate with iron-and cobalt-based catalysts. 3Iron FTS catalysts convey some favorability over cobalt catalysts on grounds of low cost and tolerance to poisons.Furthermore, through the addition of promoters or changes to reaction conditions, they are responsive to manipulation of the product slate to produce olefins and oxygenates. 3,5ron-based FTS catalysts are known to go through an "evolutionary" period in the early stages of reaction.−5 The active phase of iron FTS catalysts is still under debate, but several studies have independently demonstrated a clear correlation between FT activity and the formation of iron carbides, 8−11 in particular, the Hagg carbide.Concerning the duration of this initial catalyst conditioning period, Peŕez-Alonso et al. report that catalytic activity required a period of ca.120 h to reach the steady state, with the changes attributed to variations in the composition of the iron catalyst in the initial stages of the reaction. 12The duration of these changes is in general agreement with Buker and co-workers, who report an iron FTS catalyst to require approximately 100 h time-onstream (T-o-S) to attain steady-state operation. 13However, it is noted that the extent of the initial conditioning period will be dependent on several factors including reactor configuration, GHSV, gas consumption, pressure, and temperature.It is the iron-based FTS catalyst evolutionary phase that is the principal matter of interest in this communication.
−19 In particular, a recent communication by Warringham et al. examined the evolution of the hydrocarbonaceous overlayer over a reaction period of 24 h T-o-S that revealed the overlayer to be dominated by polyaromatic moieties alongside a less prominent aliphatic component. 18With reference to sections of the selective hydrogenation literature, 20 it is further suggested that the overlayer defines the active sites on the iron carbide surface. 18After 24 h T-o-S, the olefinic/aromatic This is an open access article published under a Creative Commons Attribution (CC-BY) License, which permits unrestricted use, distribution and reproduction in any medium, provided the author and source are cited.
moiety of the overlayer continues to increase, while the population of the aliphatic feature appears to saturate after approximately 6 h. 18If the hydrocarbonaceous overlayer does indeed play an active role in determining catalytic performance, it is opportune to consider how the nature of the overlayer is modified for reaction times beyond 24 h and to concentrate on the ∼100 h T-o-S duration, as highlighted by Peŕez-Alonso and co-workers and Buker and co-workers. 12,13−18 Interestingly, the INS spectrum for the hydrocarbonaceous overlayer that forms for extended periods of T-o-S for the iron-based FTS reference catalyst examined here (CO hydrogenation, ambient pressure, 623 K, 24 h T-o-S) is remarkably similar to that of a technical catalyst that has been extracted from a large-scale commercial FTS unit operation. 18This outcome signifies that the hydrocarbonaceous overlayer is indeed relevant to the large-scale Fe-based FT catalyst and FT process.
Given the industrial significance of iron-based FTS catalysis, 3 this article seeks to build on the work of Warringham et al. 18 and to apply the INS technique in order to discover, for ambient pressure CO hydrogenation at 623 K, how hydrogen is partitioned within an unpromoted iron FTS catalyst for extended periods of T-o-S (0−240 h).Postreaction ex situ characterization includes elemental analysis, Xray diffraction (XRD), Raman spectroscopy, and temperatureprogrammed oxidation (TPO) measurements.This combination of techniques determines how the form and nature of the overlayer is modified during the transition from a hematite precatalyst to a catalyst stabilized for CO hydrogenation activity.Collectively, this novel series of measurements over this extended reaction time reveal different aspects of the complex catalyst evolutionary process to be indirectly connected with catalytic turnover.

EXPERIMENTAL SECTION
2.1.Catalyst Preparation.The iron oxide sample, with no promoters or modifiers, utilized for this experiment was prepared using a co-precipitation of iron nitrate (Sigma-Aldrich, 99.99%) and sodium carbonate (Sigma-Aldrich, 99.99%), using a Mettler Toledo LabMax batch reactor apparatus to ensure reproducibility of catalyst synthesis. 18he resulting slurry was filtered, washed with deionized water, and calcined using the same stepwise programme described elsewhere. 18Samples were ground and sieved to a particle size range of 250−500 μm.In a separate experiment (not presented here), this material was examined for FTS activity on a semitechnical test facility operating with a syngas pressure of 25 Barg and a temperature range of 300−350 °C: The catalyst displayed sustained FTS activity for a test period duration of 6 days.It is noted that although this catalyst was prepared in exactly the same way as the catalyst examined for 0−24 h by Warringham and co-workers, 18 it is a different batch of catalyst.
2.2.INS Measurements.For INS measurements, approximately 10 g of the catalyst was loaded into an Inconel reactor cell and attached to a custom-built sample preparation apparatus. 21As the present study concentrates on using INS to characterize the form of the hydrocarbonaceous overlayer over an extended 240 h period of reaction, only the relatively larger scale INS reactor is employed in this work.Previously, the authors have considered differences in the gas/solid exchange dynamics that can lead to retardation of reaction processes in the INS reactor compared to a more conventional microreactor arrangement. 15,17Crucially, for this study, as all reactions are undertaken using the INS reactor, post-reaction temporal profiles using ancillary analytical methods can be directly compared to the INS spectra.
For CO hydrogenation measurements, the iron oxide catalyst samples were heated to 623 K at 5 K min −1 under a flow of carbon monoxide (75 mL min −1 , CK Gas, 99.9%) and hydrogen (150 mL min −1 , CK Gas, 99.9%) in a carrier gas (He, 600 mL min −1 , CK Gas, 99.9%, total WHSV of 1.27 h −1 ) and held at this temperature for a predetermined length of time.It is noted that the INS measurements were not consecutive, and each time period analyzed has utilized a different charge of the catalyst.Thus, an approximately 10 g charge of the catalyst was reacted for 48 h, the reaction quenched, and the catalyst transferred to the INS spectrometer for spectral acquisition.In the case of the 96 h run, a fresh 10 g charge of the catalyst was reacted for 96 h before stopping the reaction and transferring the catalyst to the spectrometer.Likewise for the 240 h run.
The gas products were analyzed by an in-line mass spectrometer (Hiden Analytical, HPR20 QMS Sampling System).Once the reaction was complete (t = x h), the reactant gases were stopped, and the sample was left to cool to room temperature in a flow of carrier gas.The reactor cell was isolated and transferred to an argon-filled glovebox (MBraun UniLab MB-20-G, [H 2 O] < 1 ppm, [O 2 ] < 2 ppm) before being loaded into an aluminum INS cell and sealed via an indium wire gasket for INS analysis. 22All INS measurements were performed using the MAPS direct geometry spectrometer. 23Spectra were recorded at 20 K at incident energies of 600 and 250 meV using the A chopper package. 23uantification of the ν(C−H) modes was achieved following a calibration protocol described elsewhere. 18,24For ex situ characterization, reacted samples were subjected to a passivation procedure involving the introduction of small amounts of O 2 to the reactor, gradually increasing to atmospheric levels (i.e., 20% O 2 in the gas feed). 25The samples prepared in the large-scale INS reactor are referred to as LR-x, where x indicates the T-o-S in hours.Three reaction times are considered: 48, 96, and 240 h T-o-S.
2.3.Post-Reaction Analysis.TPO studies of the largescale samples were carried out ex situ on a microreactor catalyst test line.The apparatus is composed of 1/8″ diameter stainless steel Swagelok tubing; a full description of which can be found elsewhere. 17,18Approximately, 10 mg of the catalyst was loaded into a 1/4″ quartz tube reactor with the reactor plugged with quartz wool.The reactor is housed within a tube furnace (Carbolite MTF 10/15/30) equipped with PID control and a thermocouple that is positioned within the catalyst bed to ensure accurate temperature reading during measurements.Oxygen (5% in He, 70 sccm, BOC Ltd, 99.5%) was introduced to the sample, and the reactor was heated to 1173 K at 5 K min −1 using the mass spectrometer to monitor the eluting gases.Quantification of the CO 2 peak area was achieved by measuring the CO 2 response from the in situ TPO of known masses of graphite (Sigma-Aldrich, 99.9%). 26revious ex situ TPO measurements have used larger catalyst loadings (∼40 mg), but 15−19 for the extended reaction times under consideration here that resulted in significant carbon retention; such masses caused exhaustion of the oxygen cofeed, compromising carbon quantification.TPO measurements of a lower catalyst mass of ∼10 mg ensured that the carbon oxidation process remained within quantitative limits for all samples examined.
Ex situ powder XRD was performed using a Panalytical X'Pert PRO MPD equipped with a Co Kα radiation.Ex situ Raman scattering was performed using a Horiba Jobin Yvon LabRam HR confocal Raman microscope and a 532 nm laser source at <20 mW power.Measurements were performed for approximately 5 min.Elemental analysis was performed on an Exeter CE-440 Elemental Analyzer (∼detection limit 0.1 wt %).

RESULTS AND DISCUSSION
3.1.Reaction Testing Prior to INS Measurements.The catalyst under consideration here is the same as that used in the 24 h INS study undertaken by Warringham and coworkers, with XRD confirming the phase of the iron oxide sample to be hematite (α-Fe 2 O 3 ).The sample has been additionally characterized by N 2 physisorption, XRD, Raman, scanning electron microscopy, transmission electron microscopy, and inductively coupled plasma-optical emission spectrometry that are described elsewhere. 17−19 Figure 1 displays a representative mass spectrometric profile for a 0−96 h reaction and, as reported for the 24 h reaction sequence, it is divisible into two subsections: 18 (I) the simultaneous production of H 2 O, CH 4 , and CO 2 as the reaction reaches the temperature, followed by (II) the reaction approaching steady-state operation.The latter phase is deemed to be achieved within approximately 9 h T-o-S; thereafter, the product distribution is stable.Under these conditions, CO conversion approximates to <1%, with catalytic activity approximating to an iron time yield of 6.39 × 10 −6 mmol (CO) g Fe −1 s −1 , similar to values reported in the literature for bulk iron catalysts operating under similar conditions. 27It is noted that CO 2 production drops quite dramatically over the reaction period.This was also noted by Warringham et al., who suggested the main route of CO 2 production to be predominant reduction of α-Fe 2 O 3 → Fe 0 . 18With reference to Figure 1, the decrease in CO 2 correlates with the CO returning to a steady-state regime, suggesting that by approximately 9 h on stream, the catalyst is fully reduced.Figure 1 highlights the initial conditioning period and the steady-state operation at extended reaction times (0−10 and 85−100 h, respectively).Figure S1 presents an extended reaction profile that covers the range 0−100 h T-o-S.
3.2.Post-Reaction Characterization.Elemental analysis of the catalyst before and after exposure to CO hydrogenation conditions for 48, 96, and 240 h T-o-S enable the degree of carbon and hydrogen retention within the catalyst to be determined; the results are presented in Table 1.Minimal C and H are detected in the unreacted sample.On reaction, there is a systematic trend of increasing C retention for increasing periods of T-o-S, with the C retention rate decreasing for times ≥100 h.Table 1 additionally shows hydrogen retention to be evident but to a much lesser degree than observed for C and also in a less systematic fashion.Collectively, Table 1 indicates increased retention of C (major) and H (minor) up to approximately 100 h T-o-S.Indeed, the degree of hydrogen retention is just above the detection levels of the elemental analyzer (∼0.1 wt %).
Figure 2 presents the ex situ diffractograms of the catalyst pre-and post-reaction.Samples LR-48, LR-96, and LR-240 h are displayed in Figure 2b.In contrast to Figure 2a, which distinctly shows hematite (α-Fe 2 O 3 ) reflections, Figure 2b reveals the loss of these features on reduction over the 240 h period and the emergence of various reflections in the range of 40−60°, signifying the formation of iron carbides.Rietveld refinement of the catalyst samples, details of which are presented within the Supporting Information section (Figures S2 and S3), establishes the presence of a pure Hagg carbide (Fe 5 C 2 ) phase for the post-reaction samples.Previous work has indicated cementite (Fe 3 C) to be present in the early stages of the reaction, 15 but Figure 2 is interpreted as indicating the formation of the Hagg carbide, the proposed active phase, to be completed by 48 h on stream.The broad reflections at ca. 30°are attributed to amorphous carbon; the intensity of this feature increases as T-o-S extends from 48 to 240 h.
Figure 3 presents ex situ Raman spectra of the catalyst prior to reaction (Figure 3a) and after 240 h reaction (Figure 3b).−30 Figure 4b includes a fit to the G, D1, D3, and D4 states of carbon; consistent with the build-up of an amorphous carbon component.
Figure 4 presents the post-reaction TPO profiles.Previous measurements have identified the presence of three carbonaceous species from TPO measurements on iron-based FTS  catalysts and are assigned as follows: α-reactive carbon, βamorphous carbon, and γ-bulk iron carbides. 17,18From the profiles observed within this investigation using the large-scale reactor (Figure 4), only the β and γ features with respective peak maxima of ca.630 and 680 K are observed for reaction times of 48, 96, and 240 h.The absence of the α peak (assigned as a precursor for the formation of the hydrocarbonaceous overlayer) 19 at these reaction times is considered further in Section 3.3.Quantification of the features in the TPO profiles of Figure 4 are tabulated in Table 2 and displayed in Figure 5.
Table 2 shows both the β and γ features to increase quite steeply up to ca. 100 h TOS; thereafter, the retention rate is retarded up to 240 h.Alongside the integrated TPO intensities for the 48, 96, and 240 h data sets, Figure 5 also includes intensity information taken from the earlier 0−24 h data set for CO hydrogenation over the same promoter-free catalyst under examination here. 18It is noted that from 48 to 240 h, a different batch of catalyst was used.In this way, Figure 5 presents a more comprehensive perspective on the dependency of the α, β, and γ TPO features to reaction time using the large-scale INS reactor.The data set is normalized per gram of Fe in each case.Figure 5 shows that the amorphous carbon component (β) displays continued growth up to 96 h on stream, after which it saturates.The iron carbide component (γ) on the other hand exhibits an intense growth period up to 48 h on stream but, thereafter, shows a reduced rate of growth up to 240 h TOS.Overall, as a function of TOS utilizing the large-scale reactor, Figure 5 shows the following: (i) the α feature grows and then declines within the first 50 h, and (ii) the β and γ features progressively increase up to a region of 100−200 h, where their populations appear to saturate.The apparent "dip" at 50 h in the β profile of Figure 5 is thought to reflect some variance in the testing and/or post-reaction measurement of different batches of the hematite precatalyst.
Comparing the TPO profiles of Figure 5 with the XRD data of Figure 2 indicates that the previously adopted TPO assignments may be over-simplistic.Specifically, XRD unequivocally shows saturation of the iron carbide structure by 48 h T-o-S, yet the intensity of the γ feature in the TPO plots progressively increases significantly beyond that reaction time.Adding further uncertainty on the validity of the original assignments is the observation that the intensity profile of the TPO α species of Figure 5 more correctly correlates with the carbide feature detected by XRD (Figure 2).The matter of previously unrealized hidden complexity within the TPO spectrum will be revisited in the Discussion section (Section 4).
3.3.Inelastic Neutron Scattering. Figure 6 presents the INS spectra of the reacted samples and the sample that was dried and not exposed to reaction conditions (black), recorded at incident energies of 650 and 250 meV.The spectra for 3 and 6 h have previously been reported and are included here to convey the full hydrocarbonaceous overlayer evolution. 18At  650 meV, Figure 6a shows that the spectra are defined by the presence of a ν(C−H) feature consisting of a main peak at 3048 cm −1 with a noticeable shoulder at ca. 2932 cm −1 .These bands are respectively assigned to the ν(C−H) modes of sp 2and sp 3 -hybridized carbons. 16,17In contrast, the unreacted sample (0 h) is practically featureless, except for a small but noticeable peak at 3633 cm −1 that is attributed to the ν(O−H) mode of terminal hydroxyl groups.With T-o-S, the intensity of the ν(C−H) features increase, while the ν(O−H) mode is not present from 3 h onward, indicating a progressive deposition of the hydrocarbonaceous material and the removal of the small population of surface hydroxyls.
At 250 meV (Figure 6b), the clean, dehydrated sample has a significant peak at 797 cm −1 .Following previous studies, this can be attributed to a spinon mode of the hematite precatalyst. 17The feature is no longer present at 3 h reaction time, signifying the loss of the hematite structure as the carburization process advances with extended T-o-S.The 3 h spectrum is characterized by the emergence of two weak features at 591 and 941 cm −1 that are, respectively, assigned to the A 1g Fe−O phonon mode of Fe 3 O 4 and a combination of alkenic δ(C−H) and a possible magnetic interaction associated with Fe 3 O 4 . 17,31−17 Evolution of each of these features is prominent up to 96 h, after which there is effectively little change in the spectral profile.
A significant advantage of using INS is the ability of the technique to determine the concentration of hydrogen within the analyzed samples. 34Through previously established calibration procedures, 18,24 Figure 7 presents the degree of   hydrogen retention for the two resolved ν(C−H) modes.It is noted that Figure 7 is a combination of two sets of INS data from two separate batches of the same catalyst: 3−24 h on stream as previously reported by Warringham et al. 18 and 0, 48−240 h, as obtained from this investigation.Linking in with comments made in Section 3.2, as no characteristic features were observed in the MS profile at 48 h (Figure S1), the "dip" in intensity of the INS spectra at 48 h in Figure 7 is thought to reflect a degree of variance in the testing and post-reaction measurement of different batches of the hematite precatalyst rather than any structural dynamics of the catalyst.−18 Therefore, it is deduced that the anomalously low value for the 48 h sp 2 -hybridized ν(C−H) mode in Figure 7 is reflective of some complication in that particular run.It is possible that this could be an unknown problem in the reaction testing procedure but, more likely, it could be due to poor alignment of the sample in the neutron beam.Nonetheless, with the exception of that one data point, Figure 7 essentially conveys a continuous profile.Figure 7 is characterized by two distinct trends: first, the integrated intensity of the aliphatic ν(C−H) stretch at 2932 cm −1 saturates at approximately 24 h T-o-S; second, the sp 2hybridized ν(C−H) stretch at 3048 cm −1 , representing olefinic and/or aromatic hydrocarbonaceous moieties, approaches saturation after 100−200 h T-o-S.This is important and new information, which indicates a degree of previously unrealized complexity in the evolutionary phase of the hydrocarbonaceous overlayer.

DISCUSSION
In order to better understand the INS profiles, it is first necessary to address the uncertainty connected with the original TPO assignments, as considered toward the end of Section 3.2.Concentrating first on the TPO α peak, this has been assigned to a carbonaceous species that is a precursor for the formation of the hydrocarbonaceous overlayer. 18The coincidence of its intensity profile with that of the sp 3hybridized ν(C−H) mode (Figure 7) more specifically links it to the aliphatic component of the overlayer.However, as exposed in Section 3.2, the TPO α peak could also be linked to iron carbide formation, as XRD (Figure 2b) shows iron carbide formation to have saturated over the period for which the TPO α peak is generated and consumed.One further possibility for the α feature is that part of its consumption could reflect this species additionally contributing to the growth of the TPO β and/or γ peaks for extended T-o-S (Figure 5), that is, in accordance with Niemantsverdriet's competition model; 18 the α feature could simultaneously be a precursor to the TPO β and/or γ features.
The possible assignment for the TPO γ peak is less convoluted than that discussed above for the α peak but, nonetheless, it necessarily requires revision.As noted in Section 3.2, it cannot exclusively be attributed to carbide formation.Given its profile (Figure 5) is broadly coincident that of the sp 2 -hybridized ν(C−H) mode of the hydrocarbonaceous overlayer (Figure 7), an additional contribution of polyaromatic carbon is proposed.
These considerations indicate that the post-reaction TPO plots are more complicated than originally envisaged. 18Indeed, it appears that certain peaks can be associated with various chemical transformations.Table 3 presents the revised assignments for the three TPO features connected with ambient pressure CO hydrogenation over an iron-based FTS catalyst at 623 K. Further analysis is clearly required to  ) of the 2932 cm −1 (triangles) and 3048 cm −1 (circles) features identified in Figure 6.Data previously reported by Warringham et al. for the period 3−24 h are included in the figure 18 in order to convey the full evolutionary profile of the hydrocarbonaceous overlayer over a duration of 240 h.determine which of the proposed pathways the α component follows.This important matter constitutes "work in progress".
With an improved awareness of the TPO features presented in Figure 5, it is informative to concentrate on connections to the partitioning of hydrogen within the catalyst matrix as a function of T-o-S.Specifically, how does the temporal profile of the hydrocarbonaceous overlayer correspond to that observed for the reaction test data over this extended reaction coordinate?First, the CO hydrogenation test data in Figure 1 indicate the reaction system to be effectively operating in a steady-state regime from ca. 9 h.This is the benchmark time against which other temporal trends may be usefully compared.Interestingly, elemental analysis (Table 1) shows C and H values to continue to accrue way beyond that period, not achieving saturation values until 100−200 h T-o-S.
The XRD diffractograms (Figure 2) indicate the transition from hematite to Hagg carbide and a contribution from amorphous carbon; this is consistent with the Raman spectra presented in Figure 3.However, while the diffractograms show the extent of amorphous carbon to increase on increasing T-o-S, the intensity of the Hagg carbide features has saturated at 48 h.−11 The intensity profile for the TPO β and γ features (Figure 5) broadly matches that observed for the sp 2 -hybridized carbon ν(C−H) mode (Figure 7).This leads to the proposal that, over a period of up to ca. 200 h T-o-S, the amorphous carbon (β) and/or iron carbide and polyaromatic carbonaceous deposits (γ) evident in TPO measurements are responsible for the formation of the olefinic and/or aromatic component of the hydrocarbonaceous overlayer.If the α species is in fact a precursor to the β or γ as suggested above, then the α species is suggested to be the initial stage in a sequential pathway in the development of the hydrocarbonaceous overlayer, that is, α species → β and/or γ species → sp 2 -hybridized ν(C−H) mode.This proposal is merely speculative at this stage.Nonetheless, each of the suggested pathways indicate the α species to have an integral role within the development of the evolving catalyst matrix.Indeed, these considerations illustrate the dynamic nature of the evolutionary phase of iron-based FTS catalysts undergoing CO hydrogenation at elevated temperatures and atmospheric pressure over extended periods of T-o-S.
Returning to the matter of catalytic performance, Table 4 contrasts the reaction times that correspond to completion of a particular component of the catalyst evolutionary process.Most notably, sustained catalytic turnover is attainable over short periods of time (≥9 h T-o-S), which can be loosely correlated with the formation of Hagg carbide features (≤48 h T-o-S).However, against this background, there is a progressive carburization process that leads to distinct profiles for specific entities detectable by TPO and INS.Indeed, modification of the carbonaceous and hydrocarbonaceous entities appears to undergo a relatively slow transitional phase that, seemingly, does not disturb catalyst performance.This slower modification of the catalyst matrix approaches completion over the period 100−200 h T-o-S.Collectively, this work shows that the journey from the hematite precatalyst to a stabilized material capable of supporting sustained syngas turnover is complex and involves a variety of molecular entities, that is, iron oxides, iron carbides, carbonaceous entities, and a hydrocarbonaceous overlayer.

Figure 1 .
Figure 1.Reaction profile for the sample exposed to CO hydrogenation conditions at 623 K for 96 h.Roman numerals indicate different stages within the reaction profile (see the text).The vertical dashed line signifies 9 h T-o-S.

Figure 2 .
Figure 2. XRD diffractograms of (a) the hematite precatalyst and (b) the catalyst sample after exposure to CO hydrogenation conditions for 48 (black), 96 (red), and 240 h (blue) T-o-S.The black sticks in (a) are indicative of reflections corresponding to the α-Fe 2 O 3 reference diffractogram from the Highscore Plus software package.

Figure 4 .
Figure 4. Ex situ TPO MS profiles with oxygen-associated consumption for samples exposed to CO hydrogenation conditions at 623 K for (a) 48 h, (b) 96 h, and (c) 240 h.(d) Stacked ex situ TPO MS profiles for 48 (black), 96 (red), and 240 h (blue) reactions.

Table 1 .
Elemental Analysis of the Catalyst before and after Exposure to CO Hydrogenation Conditions for 48, 96, and 240 h T-o-S a a The standard deviation from measurements performed in duplicate is shown in brackets.

Table 2 .
Quantified Peak Areas and Peak Maximum Temperatures from the TPO Studies Involving Samples from the Large-Scale INS Reactor a LR-large-scale reactor.b C content in mmol C g Fe −1 .c T max in K.

Table 3 .
Revised Probable Associations for the Three TPO Features Connected with Ambient Pressure CO Hydrogenation over an Unpromoted Iron-Based FTS Catalyst at 623 K 3-and sp3-hybridized C−H species, with quantification of the ν(C−H) modes indicating the concentration of aliphatic C−H species to saturate after

Table 4 .
Values of T-o-S That Correspond to Completion of a Particular Component of the Catalyst Evolutionary Process as Determined by (a) Reaction Testing, (b) Elemental Analysis, (c) XRD, (d) TPO, and (e) INST-o-S.In contrast, the concentration of olefinic/aromatic C−H entities progressively increase up to ∼200 h before appearing to reach saturation.• The development of the amorphous carbon (β) and iron carbide + polyaromatic carbonaceous (γ) entities that occurs over a catalyst conditioning period of ca.200 h correlate with the formation of the olefinic and/or aromatic component of the hydrocarbonaceous overlayer.• The TPO α species is suggested to play an integral part in the catalyst evolutionary process and may be associated with the following three pathways: (i) a precursor to iron carbide formation; (ii) a precursor to the formation of the aliphatic component of the hydrocarbonaceous overlayer; and (iii) a precursor to either the β or γ TPO features.• From inspection of TPO and INS intensity profiles, it is concluded that under the stated reaction conditions, the overall catalyst evolutionary phase requires approximately 200 h T-o-S to achieve completion.• Steady-state operation of the catalytic system is not directly coupled to the TPO β and γ entities, nor is it connected to the development of the olefinic/aromatic component of the hydrocarbonaceous overlayer.S Supporting Information The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.iecr.9b04636.Full reaction profile for the sample exposed to CO hydrogenation conditions at 623 K for 96 h; powder XRD with Rietveld refinement of the calcined hematite catalyst before exposure to reaction conditions; Rietveld refinement of the hematite catalyst after exposure to ambient pressure CO hydrogenation at 623 K for 24, 96, and 240 h T-o-S (PDF) *