Ethylene Dimerization and Oligomerization to 1-Butene and Higher

Feb 5, 2018 - (1) The price of ethane in the US fell by 60% from 2011 to 2012 roughly to its fuel equivalent(2, 3) and has remained at a low level eve...
1 downloads 14 Views 2MB Size
Subscriber access provided by UNIVERSITY OF TOLEDO LIBRARIES

Article

Ethylene Dimerization and Oligomerization to 1-Butene and Higher Olefins with Chromium-Promoted Cobalt on Carbon Catalyst Zhuoran Xu, Joseph Paul Chada, Lang Xu, Dongting Zhao, Devon C. Rosenfeld, Jessica L. Rogers, Ive Hermans, Manos Mavrikakis, and George W. Huber ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.7b03205 • Publication Date (Web): 05 Feb 2018 Downloaded from http://pubs.acs.org on February 6, 2018

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.

ACS Catalysis 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 21 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

Ethylene Dimerization and Oligomerization to 1-Butene and Higher Olefins with Chromium-Promoted Cobalt on Carbon Catalyst Authors: Zhuoran Xua, Joseph P. Chadaa, Lang Xua, Dongting Zhaoa, Devon C. Rosenfeldb, Jessica L. Rogersb, Ive Hermansa,c, Manos Mavrikakisa, George W Hubera,* a.

Department of Chemical and Biological Engineering, University of Wisconsin-Madison, Madison, WI 53706, United States b.

The Dow Chemical Company, 2301 N. Brazosport Blvd, Freeport, TX 77541-3257, United States

c.

Department of Chemistry, University of Wisconsin-Madison, Madison, WI 53706, United States

Corresponding author *Email for George W. Huber Corresponding Author: [email protected] ABSTRACT Industrial dimerization of ethylene to 1-butene is achieved with catalysts bearing ligand structures and activated by a co-catalyst. Here, we report that a Cr-promoted cobalt oxide on carbon catalyst was able to produce 1-butene with 53.5% - 82.4% total selectivity at ethylene conversions between 8.9% - 31.5% without the use of a co-catalyst. The Cr-promoted catalyst showed enhanced activity and stability compared to the non-promoted catalyst. The characterization results revealed that the incorporation of Cr into the cobalt oxide on carbon catalyst was able to decrease the cobalt oxide particle size. Charge transfer from Cr to Co was also observed in the Cr-promoted catalyst compared to monometallic Co and Cr on carbon. The kinetic model built to rationalize the experimental observations predicted a 50% increase in active site dispersion in the Cr-promoted catalyst, which explained the 1.6x apparent reaction rate increase compared to the non-promoted catalyst. KEYWORDS: ethylene dimerization, 1-butene, heterogeneous catalyst, promotion effect, reaction kinetics

INTRODUCTION 1-butene, a light linear alpha olefin (LAO), is used in the production of polyethylene, polybutene, butylene oxide, and aluminum alkyls. Industrial production of 1-butene is by two main routes: the separation of crude C4 refinery streams and the dimerization of ethylene 1. The price of ethane in the US fell by 60 percent from 2011 to 2012 roughly to its fuel equivalent 2-3 and has remained at a low level ever since. The decline in the cost of ethane and ethylene has generated an increase in the interest for dimerization of ethylene to 1-butene and longer olefins. Selective dimerization of ethylene to 1-butene is commercially practiced with homogeneous catalyst systems in three different processes: 1) the AlphaButol® process with a total 1-butene production capacity of 708,000 metric tons per year (MTPY) 4-5 that operates in the liquid phase using a homogeneous catalyst system comprising Ti(OBu)4 and AlEt3; 2) the Philips process which employs a homogeneous catalyst system consists of bis(tri-nbutylphosphine) nickel dichloride and AlEtCl2 that produces high yields (nearly 90%) of 1-butene from ethylene; 6-7 and 3) a process developed by Dow to dimerize ethylene into 1-butene using an organic aluminum compound (AlR3) as catalyst in a boiling solvent 8-9. A recent review has included a detailed description of many other transition metal based homogeneous systems including Ni, Zr, Hf, V, Ta, Co and Fe in the production of LAO 10. These chemical processes utilize homogeneous molecular catalysis, where organic solvent such as alkanes and aromatics are required to stabilize the catalyst in ~ 6.9 mmol/l concentration 11 making the recycle or removal of those catalysts difficult. Additionally, the homogeneous catalysts generally require an organoaluminum compound (e.g., MAO) as

ACS Paragon Plus Environment

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

co-catalyst to form the active species, and the Al/M (M represents the active metal center) ratio is in the range of 1 – 1000. The organoaluminum co-catalysts are highly air and moisture sensitive necessitating careful handling and storage and use of dry organic solvents. Consequently, the development of an effective heterogeneous catalyst system is desirable for the large-scale manufacture of chemical commodities to reduce operating cost. Studies on ethylene dimerization with acidic heterogeneous catalysts have shown less than 20% 1-butene selectivity at above 20% conversions due to double bond isomerization producing 2-butenes. For example, ethylene dimerization catalyzed by Rh-Y at a conversions of 5-10% produces 2% 1-butene, 77% trans-2-butene and 21% cis2-butene 12. Rapid isomerization of 1-butene into 2-butene was also observed for nickel-containing MCM-41 and AlMCM-41 molecular sieves for ethylene dimerization 13. Several other Ni – containing heterogeneous catalysts including NiO-ZrO2/WO3 14, Ni/Al-KIT-6 15, Ni/SiO2-Al2O3 16, Ni-Y 17 and Ni-MCM-36 18 all produced mainly 2butene from ethylene. A nickel catalyst bearing two PEt3 ligands supported on SiO2, however, showed a 95% 1butene selectivity at 23% ethylene conversion as reported by Cai et al 19. Recently, Metzger et al. synthesized a Niexchanged MOF molecular catalyst (Ni-MFU-4l) which combines a high activity (turnover frequency (TOF)=21,000 h-1) with a high 1-butene selectivity (92.0 wt.%) 20. Nevertheless, Ni-MFU-4l requires a co-catalyst (MAO) with 50-500 equivalents to the metal center to show dimerization activity. Schultz et al. have used cobalt oxide supported on ammonia-treated carbon for the dimerization of light olefins 21-23. Ort et al. 24 improved the activity of this catalyst by approximately twofold by adding ZnO into the cobalt oxide on carbon catalyst. Addy et al.25 have reported that the incorporation of nickel, chromium, and copper into the cobalt oxide supported on carbon catalyst was able to show improved activity for olefin oligomerization. Hill 26 have demonstrated a nearly twofold improvement in activity after treating the cobalt oxide on carbon catalyst with CrO3. We have identified a cobalt oxide species to be the main phase of the active cobalt oxide on carbon catalyst (CoOX/N-C) 27, and have established a positive correlation between the pyridinic nitrogen content in the carbon support and the catalyst’s oligomerization activity 28. The selectivity of the catalyst can be explained by invoking a Cossee-Arlman mechanism 29. During ethylene oligomerization, a 1-butene selectivity of 87.0% was observed at a conversion of 10.0% with the carbon supported cobalt catalyst 29. The promotional effect of adding a second metal into the primary metal or metal oxide is commonly seen in many chemical reactions such as biomass upgrading 30-32, oxygen reduction reaction (ORR) 33, and Fischer-Tropsch synthesis 34-35. In this work, we report the enhanced activity and promotional effect of the incorporation of chromium into the CoOX/N-C catalyst and the high 1-butene selectivity achieved during ethylene oligomerization with this catalyst. Chromium itself is also a widely used active metal in the oligomerization and polymerization of ethylene both in the form of soluble metal complexes 36 (after activation with a co-catalyst) and as a SiO2-supported metal oxide 37-38. The chromium-promoted cobalt oxide on carbon catalyst via facile synthesis could provide an alternative for selective dimerization of ethylene into 1-butene without the use of a co-catalyst. EXPERIMENTAL SECTION Catalyst Preparation The 13 wt.% cobalt on carbon catalyst was synthesized based on a wetness impregnation method previously described in the literature 23, 27. The catalyst is denoted as CoOX/N-C. The chromium-promoted cobalt on carbon catalyst was synthesized as follows: 2.00 g sieved activated carbon (Norit, Darco MRXm-1721; BET surface area 600-800 m2/g; 250-600 μm particle size) was first treated with 1.8 mL 30% NH4OH solution at room temperature. The carbon was then dried at 403 K for 2 h. A solution of 1.15 g Co(NO3)2·6H2O (Sigma-Aldrich) and 0.78 g Cr(NO3)3·6H2O (Sigma-Aldrich) dissolved in 1.82 g deionized water was prepared and added dropwise onto the treated carbon. After drying at 403 K overnight, the impregnated carbon was treated with another 3.0 mL of 30% NH4OH solution. The amount of NH4OH solution was decreased from 2.5 mL/gcarbon in our previous publication 27 to 1.5 mL/gcarbon in this publication. This was done to avoid caking in the final drying step. The material was dried overnight at 403 K on a hot plate. The theoretical metal loadings are 8 wt.% Co and 5 wt.% Cr based on the

ACS Paragon Plus Environment

Page 2 of 21

Page 3 of 21 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

corresponding metal oxides on carbon. The Cr promoted cobalt on carbon catalyst with other metal loadings including 8 wt.% Co, 0 wt.% Cr; 5 wt.% Co, 8 wt.% Cr and 3 wt.% Co, 10 wt.% Cr were also synthesized and initially tested for 1-butene oligomerization, and the catalyst with 8 wt.% Co and 5 wt.% Cr demonstrated the highest catalytic activity and stability among all. In this work, focus will be given to the catalyst with the best performance (8 wt.% Co, 5 wt.% Cr, denoted as Cr-CoOX/N-C) on ethylene oligomerization. The chromium on carbon catalyst was synthesized following the same steps as above. During the synthesis, 0.78 g Cr(NO3)3·9H2O (Sigma-Aldrich) was dissolved in 1.82 g deionized water and deposited on 2.00g of carbon support previously treated with NH4OH followed by the same drying steps as mentioned above. The catalyst is denoted as CrOX/N-C with 4.7 wt.% theoretical Cr loading. The activity among the same catalysts synthesized from different batches is within ±5% in difference. Catalyst Characterization The X-ray diffraction (XRD) profiles were collected between 2θ = 5o and 2θ = 45o with Rigaku Rapid II diffractometer with a Mo Kα source at 50 kV and 50 mA. The crystallite phase identification was performed with JADE 9 software. Samples for XRD analysis were pretreated in flowing helium at 150 mL/min with 5.5 K/min ramp rate and held at 503 K for 2 h then cooled to RT prior to the XRD measurement. X-ray absorption spectroscopy (XAS) measurements were taken at beamline 12-BM of the Advanced Photon Source (APS) at Argonne National Lab (Lemont, IL). Catalyst samples were crushed and diluted with boron nitride (Sigma-Aldrich). Self-supporting pellets were pressed inside a 4 mm I.D. stainless steel cylindrical sample holder. Dilution ratios were calculated to give an edge step of ~1. The stainless steel sample holder was sealed in a 1” O.D. Kapton-windowed quartz tube fitted with Swagelok valves. The sealed sample was pretreated in a tube furnace and continually purged with He. After pretreatment, the sealed samples were transferred to the beamline without exposure to air. XAS measurements of the Cr (5.989 keV) and Co K-edge (7.709 keV) were collected in the transmission geometry. X-ray absorption was measured with gas ionization chambers before and after the sample holder. Energy calibration was performed with metallic reference foils placed between ion chambers after the sample. Data normalization and background subtraction were performed in Athena 39. X-ray photoelectron spectroscopy (XPS) was conducted on K-alpha XPS spectrometer (Thermo Scientific) with a micro-focused monochromated Al Kα X-ray source. Prior to the measurements, the samples were pretreated in 150 mL/min flow of helium at 503 K for 2 h and sealed in a stainless steel tube. The samples were then packed into a vessel (Transfer Vessel K-Alpha) and sealed under vacuum in a glove box under a nitrogen atmosphere. The XPS measurements were taken of the samples without exposure to air. Samples were analyzed at 10-7 mbar and room temperature with the flood gun on to avoid sample charging. Spectra were taken in the regions of C 1s, O 1s, N 1s, Co 2p, Cr 2p. The binding energy (BE) values were calibrated to the BE of the graphite C 1s peak at 284.8 eV. The Co 2p and Cr 2p spectra were taken over 30 scans with a pass energy of 50 eV and a dwell time of 50 ms. The energy step size was 0.2 eV for all scans in all regions. No difference was observed in the XPS spectra for two different spots for each sample. The peak fitting was performed using Avantage (Thermo Scientific) software package. Scanning transmission electron microscopy (STEM) images were collected with an FEI Titan microscope equipped with a Cs aberration corrector operated at 200 kV. High-angle annular dark field (HAADF) images were obtained in the range of 54 to 270 mrad using a 0.8 nm probe and 24.5 mrad probe convergence semi-angle. Catalyst samples were suspended in ethanol with sonication then deposited on a carbon-coated copper grid (EMS). Samples were plasma cleaned for 15 min prior to analysis. Catalytic Measurement

ACS Paragon Plus Environment

ACS Catalysis

The ethylene oligomerization reactions were conducted in a tubular down-flow fixed-bed reactor described elsewhere 27, 40. The length of the reactor tube is 30.5 cm long and the catalyst bed is between 2.07 cm (0.5 g) and 8.29 cm (2.0 g). The temperature is measured by a built-in thermocouple touching the reactor wall at the center of the catalyst bed. We have used an aluminum heating block to fill the void space between the furnace and the tubular reactor to ensure a uniform temperature profile. The reaction pressure was adjusted with a back pressure regulator. For a typical reaction, 0.5 g to 2.0 g of the catalyst sample was packed into a 3/8 in. stainless steel tube without diluent. All the catalysts were pretreated at 503 K under 150 mL/min flowing helium prior to reaction. The reactions were carried out at 353 K with ethylene diluted with approximately 50% helium in the gas phase. The overall reaction pressure was kept at 31 bar, and ethylene (UHP, Airgas) was co-fed with helium (UHP, Airgas) at 34.8 mL/min and 45.4 mL/min flow rate, respectively. The gas effluent was analyzed by an online gas chromatograph (GC-FID, Shimadzu) every 0.5 h. The liquid products were gathered from a cold trap and analyzed by a comprehensive two-dimensional gas chromatograph-mass spectrometer (2D GC-MS). The GC operation conditions and the compound standard information can be found elsewhere 27, 40. Ethylene conversion, product selectivity, product yield, α-butene distribution and weight hourly space velocity (WHSV) were calculated based on Eq. 1-5. Ethylene conversion (%) = Product selectivity (%) =

                     

                       

× 100%

× 100%

Product yield (%) = ℎ  !" #$%&" × '$"()! %!&#& α-Butene distribution (%) =

  *+                 

Weight hourly space velocity (WHSV, h-1) =

× 100%

 ,    -   ,  

(Eq. 1) (Eq. 2) (Eq. 3) (Eq. 4) (Eq. 5)

RESULTS AND DISCUSSION Selectivity Figure 1 shows the butene isomer selectivity with time on stream (TOS) for ethylene oligomerization at a WHSV of 32.6 h-1. At the initial TOS (< 2.0 h TOS), the reactor was in the transient state with below 60% carbon balance. The average yield and selectivity data reported in the later section was taken after 3.5 h TOS. After 3.5 h TOS, the catalyst did not show a high rate of deactivation with an average ethylene conversion of 20.8% and a 1-butene selectivity of 66.1%. At steady state 83.6% of the butenes were 1-butene.

100 30 80

60

20

40

15

Conversion (%)

25

Selectivity (%)

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 21

10 20 5 0 0

2

4

6

8

10

12

14

TOS (h)

ACS Paragon Plus Environment

Page 5 of 21 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 1. Ethylene conversion (■) and product selectivity vs. time-on-stream including (■) 1-butene, (●) trans-2-butene and (▲) cis-2-butene with Cr-CoOX/N-C. Reaction conditions: 32.63 h-1 WHSV, T = 353 K, p (ethylene) = 13.4 bar.

1

The product selectivity was studied at different ethylene conversions by varying the WHSV from 16.3 h-1 to 65.3 has shown in Figure 2 and

ACS Paragon Plus Environment

ACS Catalysis

Table 1. The main products were 1-butene > trans- + cis-2-butene > C6 olefins > C8 olefin and C10+ olefins. Olefins up to C20 were detected although they were less than 1% (carbon molar) of the products. The 1-butene selectivity decreased from 81.8% to 45.4% with an increasing ethylene conversion from 8.6% to 32.1%. The C6 olefin selectivity increased from 12.4% to 20.7%, and the trans- and cis-2-butene selectivity increased from 5.6% to 21.0% over this same range. The ratio of trans-/cis-2-butene was close to 1 which is far from the equilibrium ratio of 1.9. A 44.4 % loss in activity was observed during 12 h TOS at the highest WHSV (65.3 h-1) as shown in Figure S1, and the loss in activity tends to appear at longer TOS as WHSV increases. Though slightly lower in 1-butene yield at 31.5% ethylene conversion compared to the silica supported nickel complex 19, the Cr-CoOX/N-C catalyst surpasses the majority of the heterogeneous nickel-based catalysts for ethylene dimerization in overall 1-butene yield.

80 70

Product Selectivity (%)

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 21

60 50 40 20 15 10 5 0 5

10

15

20

25

30

35

Conversion (%)

Figure 2. Product selectivity including (■) 1-butene, (●) trans-2-butene, (▲) cis-2-butene, (▼) C6 olefins, (◆) C8 olefins, (◄) C10 and C10+ olefins at different levels of ethylene conversions at 353 K with Cr-CoOX/N-C. The conversion was varied by changing the WHSV between 16.3 and 65.2 h-1. The selectivity data is averaged after steady state is reached (3.5 h -12.0 h TOS). Lines were added to guide the eye.

We identified the distribution of C6 and C8 products by two-dimensional GC as shown in

ACS Paragon Plus Environment

Page 7 of 21 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 1. The most abundant C6 olefin species are linear internal hexenes (2- and 3-hexene). The 1-hexene distribution decreased, accompanied by an increase in the linear internal hexene and mono-branched hexene distributions, with increasing ethylene conversion. The 1-octene distribution also decreased from 9.3% to below 1.0% as ethylene conversion increased from 8.6% to 32.1%. No major change in the total linear octene distribution was observed at different ethylene conversions. The catalyst isomerized the linear terminal olefins into linear internal isomers with increasing conversion.

ACS Paragon Plus Environment

ACS Catalysis

Table 1. Product selectivity and distribution of different products as a function of ethylene conversions over Cr-CoOX/N-C catalyst. Reaction conditions: 353 K, 13.4 bar ethylene partial pressure balanced in helium. Data averaged between 3.5 h -12.0 h TOS. Ethylene conversion (%) WHSV (h-1) Butene Distribution (%)

Hexene Distribution (%)

Octene Distribution (%)

1-butene 2-butene Total butene selectivity 1-hexene 2- and 3-hexene 3-methyl-1-pentene Total hexene selectivity 1-octene trans-2-octene cis-2-octene 3-octene 4-octene trans-3-methyl-2-heptene trans-5-methyl-2-heptene cis-5-methyl-2-heptene trans-5-methyl-3-heptene cis-5-methyl-3-heptene Total linear octene Total octene selectivity

8.6 65.2 93.6 6.4 87.4 31.5 66.7 1.7 12.4 9.3 7.1 4.0 60.5 6.1 0.9 0.8 10.3 0.2 0.7 87.1 0.1

20.8 32.6 83.6 16.3 79.1 9.8 88.0 2.1 14.3 2.5 7.9 3.7 63.2 10.5 1.2 0.1 8.0 2.1 0.6 87.8 3.9

27.3 21.7 79.7 20.2 72.7 7.3 89.3 3.3 17.7 1.6 8.0 3.7 58.2 12.1 1.4 0.3 11.5 2.3 0.8 83.6 5.4

32.1 16.3 68.4 31.6 66.4 3.4 93.1 3.5 20.7 0.6 8.8 3.9 63.4 9.4 1.6 0.2 8.2 2.6 1.3 86.1 7.6

Cr promotion effect on catalyst activity Figure 3 shows ethylene conversion over both Cr-CoOX/N-C and CoOX/N-C catalysts. A description of the details of ethylene conversion with CoOX/N-C catalyst is reported elsewhere 29. The ethylene conversion with Cr-CoOX/NC is 1.6 times as high as CoOX/N-C at steady state (after 3.5 h TOS) at a 16.3 h-1 WHSV. The difference between the Cr-promoted and non-promoted catalyst became less (1.2 times) at higher WHSV (32.63 h-1). The Cr-CoOX/N-C catalyst increased in conversion slightly after the initial transient period, which was not seen with the CoOX/N-C catalyst. At both WHSVs, the Cr-CoOX/N-C is more stable with TOS compared to CoOX/N-C. Less than 0.1 % ethylene conversion was observed with CrOX/N-C (at 353 K, 13.4 bar ethylene partial pressure and 65.25 h-1 WHSV) as compared to 8.6% conversion at the same conditions with Cr-CoOX/N-C. This indicates that the activity of monometallic chromium on carbon is negligible.

60

60

(a)

(b)

50

50

Ethylene Conversion (%)

Ethylene Conversion (%)

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 21

40

30

20

10

0

40

30

20

10

0

0

2

4

6

8

10

12

14

0

2

4

6

8

10

12

14

TOS (h)

TOS (h)

Figure 3. Ethylene conversion with time-on-stream for (■) Cr-CoOX/N-C and (●) CoOX/N-C catalyst at (a) 32.63 h-1 WHSV, and (b) 16.31 h-1 WHSV.

ACS Paragon Plus Environment

Page 9 of 21

Catalyst characterization In our previous study of 1-butene conversion with CoOX/N-C, we have confirmed the existence of Co(III) and Co(II) on this catalyst 27. At elevated pretreatment temperatures (above 823 K) Co(0) forms in an inert gas atmosphere 27. A reduction of cobalt was observed for the Cr-CoOX/N-C compared to CoOX/N-C catalyst with both x-ray absorption near edge structure (XANES) and XPS. Upon chromium addition, the Co K-edge shifted to lower energy indicative of a partial reduction of Co(III) oxide to Co(II) as shown in Figure 4. The CoOX/N-C catalyst showed a characteristic sharpening of the main Co 2p3/2 and Co 2p1/2 peaks in the XPS at 780.9 eV and 796.0 eV, respectively. The main 2p3/2 and 2p1/2 peaks were found at 781.1 eV and 796.5 eV respectively for the Cr-CoOX/N-C catalyst. The minor differences associated with the binding energies of Co3+ and Co2+ 41 makes it difficult to accurately quantify the amount of Co3O4 and CoO within each sample by XPS. However, the satellite peaks (at 787.1 eV and 803.9 eV) for the two catalysts reveals more information regarding the Co3O4 and CoO features. A dramatic decrease in intensity of the satellite structure was observed for CoOX/N-C, whereas a much higher intensity was observed in the satellite peaks for Cr-CoOX/N-C relative to the main peaks. The high-spin character of the Co2+ oxides is unique, and it allows for strong electron correlation which leads to various d-d coupling and chargetransfer from O 2p to the Co 3d. In contrast, the low-spin octahedrally-coordinated Co3+ or tetrahedrally-coordinated Co2+ does not possess the above effects 42-43. Therefore, the relatively sharper Co 2p XPS peaks and the diminished satellite intensity observed from CoOX/N-C suggest it has more cobalt in the Co3O4 structure 44, while the cobalt oxides within Cr-CoOX/N-C possess more Co2+.

1.6

CoO standard

Co3O4 standard

1.4

Cr-CoOx/N-C Norm. Absorption (ux)

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

1.2

CoOx/N-C 1.0 0.8 0.6

Co foil

0.4 0.2 0.0 7700

7720

7740

7760

7780

Energy (eV)

ACS Paragon Plus Environment

ACS Catalysis

Co 2p

Intensity (a.u.)

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 21

2p3/2

2p1/2 CoOX/N-C S

S 2p3/2 2p1/2

S

S

Cr-CoOx/N-C 810

800

790

780

770

Binding Energy (eV)

Figure 4. Top: XANES spectra collected at the Co K edge. Pink dotted trace: CoO standard; blue dotted trace: Co3O4 standard; green dotted line: Co foil; black trace: Cr-CoOX/N-C catalyst; red trace: CoOX/N-C catalyst. The cobalt edge energy for each sample is reported in the table. Bottom: Co 2p XPS region comparing CoOX/N-C catalyst (red) and Cr-CoOX/N-C catalyst (black). S represents the satellite peak.

As shown in Figure 5 upon cobalt addition, there is a shift in the Cr K-edge to higher energy and more notably an increase in the intensity of the pre-edge features consistent with partial oxidation of the Cr(III) to higher oxidation states. The XPS Cr 2p region suggests different chromium species exist for CrOX/N-C and Cr-CoOX/N-C on the surface (Figure 5). The XPS Cr spectra are similar to what was observed by X. Zhang et al. 45 for CrOx/ZnO, and by Y. Zhang et al. 46 for Cr/Al2O3. For CrOX/N-C, two peaks representing Cr 2p3/2 with a binding energy of 577.6 eV and Cr 2p1/2 with a binding energy of 586.8 eV were observed, and they are the characteristic peaks of Cr3+ oxides 41. The deconvolution of the Cr 2p region for Cr-CoOX/N-C (Figure S2) has resulted in four peaks: Cr3+ 2p3/2, BE=576.6 eV; Cr6+ 2p3/2, BE=579.3 eV; Cr3+ 2p1/2, BE=579.3 eV; and Cr6+ 2p1/2, BE=588.5 eV. The Cr3+ and Cr6+ compositions for the Cr-CoOX/N-C surface are 55.9% and 44.1%, respectively, while CrOX/N-C contains almost entirely Cr3+, consistent with the XANES results. The appearance of Cr6+ in the pretreated catalyst is a potential health concern due to its carcinogenicity and proper care should be taken to ensure that the Cr does not leach into the environment. It is likely that Cr could be replaced with another metal with a high oxidation state which could introduce the similar promotional effects to the cobalt catalyst.

ACS Paragon Plus Environment

Page 11 of 21

1.6

CrOx/N-C

Cr(III) standard

1.4

Cr-CoOx/N-C Norm. Absorption (µ x)

1.2

Cr(VI) standard

1.0 0.8 0.6 0.4

Cr foil

0.2 0.0 5980

5990

6000

6010

6020

6030

6040

Energy (eV)

2p3/2 2p1/2

Intensity (a.u.)

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

CrOX/N-C

2p3/2 2p1/2

Cr-CoOX/N-C

595

590

585

580

575

570

Binding Energy (eV)

Figure 5. Top: XANES spectra collected at the Cr K edge. Red dotted trace: Cr(VI) standard. Blue solid trace: Cr(III) standard. Pink solid trace: Cr foil. Black dotted trace: Cr-CoOX/N-C catalyst. Black solid trace: CrOX/N-C catalyst. The chromium edge energy for each sample is reported in the table. Bottom: Cr 2p in-situ XPS spectra of CrOX/N-C (black) and Cr-CoOX/N-C (red).

The results from XANES and XPS analysis indicate a possible charge transfer from Cr to Co when both metals exist in their corresponding oxide form on the carbon support. As a group VIII metal, Co has a higher electronegativity (χ=1.88) than Cr (χ=1.66), and hence Co would appear in a more reduced form when Cr is present. The electronic effect observed from bimetallic catalysts is very common in transition metal systems 47-48. The more reduced Co does not offer any advantage during the initial step where the formation of a metal-olefin complex is favored over a metal center with more electron-accepting features 27, 49. However, the increased electron density in Co within our Cr-CoOX/N-C catalyst allows for easier product desorption 19, so that Cr-CoOX/N-C would suffer less

ACS Paragon Plus Environment

ACS Catalysis

from product olefin deposition. Our prior report hypothesized olefin product deposition as a mode of catalyst deactivation 29. The weaker binding of the products on the Cr-CoOX/N-C catalyst could be the reason for the improved catalyst stability. Figure 6 shows the XRD patterns of the non-promoted and promoted catalyst. As discussed in our earlier work 27, the CoOX/N-C catalyst contains both Co(III) and Co(II) oxide. The Co3O4 (2θ=16.79o) peak decreases in intensity with Cr addition with no other major peaks appearing. The diminished peak intensity indicates that the addition of Cr has resulted in a cobalt oxide that is primarily amorphous, and possibly with smaller crystallite size and more defects within the Co3O4 phase 50. Additionally, no Cr-containing peak is observed in the XRD patterns of CrCoOX/N-C and CrOX/N-C (Figure S3), indicating that the Cr exists in either an amorphous form or with very small crystallite size.

Cr-CoOX/N-C

CoOX/N-C

Co3O4 ,PDF # 01-071-4921

Intensity (A.U.)

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 21

Graphite, PDF # 01-071-3739

Cr2O3, PDF # 00-006-0504

Cr, PDF # 00-006-0694

5

10

15

20

25

30

35

40

45

Two Theta (o)

Figure 6. Mo-pXRD spectra for Cr-CoOX/N-C catalyst and CoOX/N-C catalyst after pretreatment at 503 K in helium.

Figure 7 shows the STEM-HAADF image of the Cr-CoOX/N-C catalyst. The high metal loading on an amorphous support resulted in a poor contrast. The particles were typically on the order of 5 nm or less. Occasionally large clusters were observed (10-50 nm). A precise particle size measurement was not obtained as it was challenging to resolve the smallest particles. The particle size observed with Cr-CoOX/N-C was smaller than the particle size previously observed for CoOX/N-C (5-10 nm) 27.

ACS Paragon Plus Environment

Page 13 of 21

Figure 7. Representative STEM-HAADF images of Cr-CoOX/N-C catalyst.

Figure 8 shows the Co K-edge EXAFS of different catalysts. The first and second shells of CoOX/N-C align with the spinel Co3O4 crystal structure. The first peak (~1.5 Å) in Figure 8b-d is assigned to the nearest Co-O bonds. The second (~2.5 Å) and third peaks (3.1 Å) in Figure 8b-c are from the two Co-Co scattering paths due to the coordination geometries of octahedral and tetrahedral Co sites 51 found in spinel structures. The spectra of CrCoOX/N-C shows similar shells although the features were too weak to definitively identify any Co-Cr scattering. The decrease in amplitude of the Fourier transformed EXAFS (Figure 8d) can be attributed to either an increase in crystalline disorder or a decrease in particle size (i.e. lower coordination numbers). The weak scattering of higher shells (>3.5 Å) is further evidence for the high dispersion of this system. The Cr EXAFS spectra with and without the addition of Co can be found in Figure S4.

Co-Co Intensity (a.u.)

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

(a) Co-O

Co-Co (O) Co-Co (T)

(b) (c) (d) 0

1

2

3

4

5

6

R (A)

Figure 8. The k3-weighted Fourier-transformed spectra from EXAFS collected at the Co K edge. Key (a): Co foil; (b): Co3O4 standard; (c): CoOX/N-C catalyst; (d): Cr-CoOX/N-C catalyst. All the catalysts were pretreated at 503 K in helium then cooled to RT prior to the measurement without exposure to air.

XRD, EXAFS and STEM results indicate a decrease in the particle size of the cobalt oxide on carbon for CrCoOX/N-C compared to CoOX/N-C. It should be noted that, the cobalt loading of Cr-CoOX/N-C (8 wt.%) is less than

ACS Paragon Plus Environment

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

the cobalt loading of CoOX/N-C (13 wt.%). This can partially explain the observed decrease in the cobalt oxide particle size with Cr-CoOX/N-C. Nevertheless, geometric and electronic influences cannot often be separated. The change in geometry could alter the nature of the exposed planes and the topology of the surface sites, while it could also result in a change in the electron bandwidth and binding energies of core electron 48. When supported and properly calcined on silica, the chromium itself can act as an effective metal site for ethylene polymerization (Philips catalyst). A number of studies have excluded Cr(VI) as an active oxidation state for ethylene conversion 38, 52-53. As reported in this study the Cr(III) on carbon was not active for ethylene oligomerization. The introduction of Cr has increased the dispersion and population of Co(II) sites.

ACS Paragon Plus Environment

Page 14 of 21

Page 15 of 21

Development of a reaction kinetics model

2.0

b

18

35

1.6

15

30

1.2

12

25

0.8

0.4

9

20

6

15

Conversion (%)

c Product yield (%)

ln(r)

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

3

Slope = 0.97 R2 = 0.993

0.0

10 0 5

1.2

1.6

2.0

2.4

2.8

20

ln(pethylene)

30

40

50

60

70

WHSV (h-1)

Figure 9. Pathway, reaction order, and kinetic model fitting of ethylene oligomerization with Cr-CoOX/N-C. (a) Proposed reaction pathway for ethylene oligomerization with Cr-CoOX/N-C. The rate constants (at 353 K) were estimated using the kinetic model. The experimental values were averaged between 3.5 h and 12 h TOS. (b) Ethylene consumption rate (gethylene/gcat/h) as a

ACS Paragon Plus Environment

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 21

function of ethylene partial pressure (bar). The dotted line represents the linear fitting result. (c) Kinetic model fit (dotted line) with product yield including 1-butene (■), 2-butene (●), hexenes (▲) and octenes (▼), together with ethylene conversion (◇) at different WHSVs.

The reaction order with respect to ethylene (7 to 14 bar, 353 K) is close to first order as shown Figure 9b for the Cr-CoOX/N-C catalyst. The catalyst suffered from deactivation under the reaction conditions, and the apparent reaction rate is averaged across the entire TOS (Figure S5). The ethylene consumption rate was averaged between 3 – 12 h TOS. A Schulz-Flory distribution of oligomer products from ethylene oligomerization was also observed with Cr-CoOX/N-C (Figure S6) at 16.31 h-1 WHSV. The chain growth probability (α) and the fitting constant (β) were found to be 0.54 and 0.22, respectively. The observations are consistent with the Cossee-Arlman mechanism, a classic mechanism to describe the carbon-carbon bond-forming step occurring by insertion of the coordinated olefin into the metal-alkyl bond 49, 54. Aside from ethylene chain propagation, we have also considered the isomerization of oligomers with 2-hexene and 3-octene as the isomer species for C6 and C8 products, respectively in the construction of the kinetic model, due to their high abundances in the product stream. The isomerization of linear alpha olefin products became more significant leading to an increase in distribution of linear internal and branched olefin products in the C4+ range. Based on this reaction mechanism, internal hexene and octene products are formed from isomerization after 1-hexene and 1-octene are produced. These isomerization steps should not affect the kinetics of 1-hexne and 1-octene formation. The product selectivity with Cr-CoOX/N-C was measured at different ethylene conversion as shown in Figure 9. The kinetic model was built based on a plug flow reactor (PFR) model, where the concentration of the reactants and products is a function of position in the reactor (catalyst loading). The model-predicted reaction rates and product yields were obtained by solving the PFR rate equations (Eq. 6) using the built-in ordinary differential equation solvers (ode15s) in the MATLAB software package where Fi is the molar flow rate of Species i per catalyst site; x is the fractional length of the reactor; rij is the turnover frequency of Species i in Reactor j. ./ 0

= ∑3 $3

(Eq. 6)

The rate constants of all the adsorption/desorption steps (kads, Steps 1, 3, 5, 7, 9, 11 and 13) were estimated using collision theory as described in Eq. 7 55 where A is the surface area per catalyst site; m is the mass of the adsorbate molecule; kB is the Boltzmann constant; T is the reaction temperature. The rate constants of the surface reaction steps (ksurf, Steps 2, 4, 6, 8, 10 and 12) were estimated using transition state theory as described in Eq. 8 56 where ν is the molecular vibrational frequency (~O(1013)), and Ea is the activation energy. Q  =

R STUVW X

Q  = # YZ/VW X

(Eq.7) (Eq. 8)

Further analyses of the reversibility of each elementary step suggested that all the adsorption/desorption steps were essentially quasi-equilibrated, and the reaction kinetics was mainly determined by surface steps. The four activation energies (Ea2, Ea4, Ea6, and Ea8, Ea10 and Ea12, reported in Table S2) plus the active site dispersion were adjusted during the model-fitting process in order to minimize the value of the objective function (obj) defined in Eq. 9, so that the predicted product yields matched those collected experimentally where yi,model is the model predicted yield of Species i, and yi,expr is the experimentally measured yield of Species i. obj = ∑

],/,^_`ab +,/,acde ] ,/,acde

(Eq. 9)

The model predicts an increased energy barrier for the reaction steps forming higher oligomers products. Step four, the 1-butene isomerization step, is kinetically more favored compared to step six, oligomerization, with a 9.4 kJ/mol

ACS Paragon Plus Environment

Page 17 of 21 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

lower activation energy. It can be inferred that once formed, the linear alpha olefin products are more likely to undergo double bond isomerization to form internal olefins, rather than further oligomerization with ethylene. Our catalyst would thus have better control over the carbon chain length of the oligomer products, and be used as an efficient catalyst to selectively dimerize ethylene into butenes. The same set of reaction rate constants were then used to estimate the active site dispersion for CoOX/N-C catalyst, by fitting the product yields data collected with this catalyst (Table S3) at reaction conditions reported in Figure 3. The thermodynamic data of different species was gathered from 57 and adjusted to the reaction conditions. The heats of adsorption of the hydrocarbon species on cobalt oxide were estimated from literature 58. The comparison between experimental measurements and modelpredicted overall product yields (grouped by carbon number) is summarized in Figure 9c and Table S3, for CrCoOX/N-C and CoOX/N-C, respectively. The model gives an excellent fit for both catalysts. The compositions of 1hexene and 1-octene in the total C6 and C8 products, respectively, are illustrated in Figure S7. An excellent agreement between model and experiment is observed, except for the highest WHSV (65.2 h-1) case. Due to the low conversion (8.9%) at 65.2 h-1 WHSV, the experimentally measured yield of C6 and C8 products are extremely low (1.0 % and 0.2 %, respectively). These experimental measurements could bear significantly larger percentage errors, which is likely reason for the 30%-50% discrepancy in the model fit. The model predicts an active site dispersion of 20.9% and 13.9% for Cr-CoOX/N-C and CoOX/N-C, respectively. Based on the characterization results discussed previously, the 50.3% increase in the active site dispersion could be due to 1) the decrease in the cobalt oxide particle size and 2) the increase in the Co(II) composition for Cr-CoOX/N-C compared to CoOX/N-C. The higher apparent reaction rate observed from Cr-CoOX/N-C can be attributed to the increase in the number of active sites per gram of Cr-CoOX/N-C as compared to CoOX/N-C catalyst.

CONCLUSIONS Cr-CoOX/N-C catalyst was able to dimerize ethylene into butenes with a 1-butene selectivity of 82.5% at a conversion of 8.9%, which is 3-4 times higher than the 1-butene selectivity obtained with other heterogeneous catalysts without the use of a co-catalyst. The 1-butene selectivity decreases with increasing ethylene conversion due to the formation of 2-butene and higher oligomers. The catalyst showed a low rate of 1-butene double bond isomerization after an initial transient period, which is different from most of the Ni-containing heterogeneous catalysts. Compared to the CoOX/N-C catalyst, Cr-CoOX/N-C demonstrated 1.2-1.6 times higher apparent rate in ethylene conversion. The XRD, STEM and EXAFS results have evidenced a decrease of the Co3O4 particle size upon addition of Cr. The XANES and XPS results have evidenced a change in the cobalt oxidation state upon Cr addition. The cobalt is more reduced with a higher proportion of Co2+ in Cr-CoOX/N-C than in CoOX/N-C. At the same time, Cr6+ was seen in the Cr-CoOX/N-C while CrOX/N-C contains mostly Cr3+. The CrOX/N-C was not active in ethylene conversion. The product selectivity observed during ethylene conversion with Cr-CoOX/N-C catalyst can be described with Cosee-Arlman mechanism. A simplified kinetic model was constructed to describe the reaction network, which predicted a nearly 50% increase in the active site dispersion that resulted in a higher apparent reaction rate based on the mass of the catalyst.

SUPPORTING INFORMATION Cr-CoOX/N-C stability with TOS at various WHSVs, Cr 2p XPS peak deconvolution and analysis, pXRD patterns comparing CrOX/N-C and Cr standards, EXAFS spectra comparing CrOX/N-C and Cr-CoOX/N-C, ethylene partial pressure effect with TOS, Schulz-Flory distribution fit of the oligomer products, kinetic model fitting result for hexene and octene isomerization, Cr-CoOX/N-C XPS survey scan result, activation energies computed from the kinetic model, kinetic model fitting results for CoOX/N-C.

ACS Paragon Plus Environment

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

ACKNOWLEDGEMENTS The authors thank Nat Eagan, Dan McClelland, Insoo Ro, Ben Reinhart with their assistance in the preparation of the samples for XAS analysis. The authors acknowledge the facilities and instrumentation for microscopy and XPS at UW-Madison supported by the University of Wisconsin Materials Research Science and Engineering Center (DMR-1121288) and the XRD facility supported by the Department of Geoscience. The authors thank the Dow Chemical Company for providing funding for this work. Part of the computational work were performed at supercomputing centers located at the Environmental Molecular Sciences Laboratory, which is sponsored by the DOE Office of Biological and Environmental Research at Pacific Northwest National Laboratory; the Center for Nanoscale Materials at Argonne National Laboratory, supported by DOE contract DE-AC02-06CH11357; and the National Energy Research Scientific Computing Center (NERSC), a DOE Office of Science User Facility supported by DOE contract DE-AC02-05CH11231.

TOC

ACS Paragon Plus Environment

Page 18 of 21

Page 19 of 21 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

REFERENCES (1) Greiner, E.O.; Blagoev, M.; Yamaguchi, Y. In Chemical Economics Handbook, Linear Alpha-Olefins; IHS Chemical, 2013; p 10. (2) McCafferty, M.; Rubin, R. US Gulf Coast Ethane Prices Fall to 10-Year Low on High Inventories. In S&P Global Platts: Houston, Nov 30, 2012. (3) Schwartzel, E. Drillers Rattled as Ethane, Propane Prices Plunge. In Pittsburgh Post - Gazette, Aug 31, 2012. (4) McGuinness, D. S. Olefin Oligomerization via Metallacycles: Dimerization, Trimerization, Tetramerization, and Beyond. Chem. Rev. 2010, 111 (3), 2321-2341. (5) Forestière, A.; Olivier-Bourbigou, H.; Saussine, L. Oligomerization of Monoolefins by Homogeneous Catalysts. Oil Gas Sci. Technol. 2009, 64 (6), 649-667. (6) Carter, C. O. Surface Conditioning in Olefin Dimerization Reactors. US Patent 4538018, Aug 27, 1985. (7) Carter, C. O. Olefin Dimerization. US Patent 4242531 A, Dec 30, 1980. (8) Maschmeyer, D. M.; Fowler, A. E.; Sims, S. A.; White, G. E. Process for Making a Mixture of Ethylene and Butene-1. US Patent 4484016 A, Nov 20, 1984. (9) Al-Jarallah, A.; Anabtawi, J.; Siddiqui, M.; Aitani, A.; Al-Sa'doun, A. Ethylene Dimerization and Oligomerization to Butene-1 and Linear α-olefins: A Review of Catalytic Systems and Processes. Catal. Today 1992, 14 (1), 1-121. (10) Nicholas, C.P. Applications of Light Olefin Oligomerization to the Production of Fuels and Chemicals. Appl. Catal. A-Gen. 2017, 543 (5), 82-97. (11) Al-Sa'doun, A. W. Dimerization of Ethylene to Butene-1 Catalyzed by Ti(OR')4-AIR3. Appl. Catal. AGen. 1993, 105 (1), 1-40. (12) Okamoto, Y.; Ishida, N.; Imanaka, T.; Teranishi, S. Active States of Rhodium Exchanged Y Zeolite Catalysts for Hydrogenation of Ethylene and Acetylene and Dimerization of Ethylene Studied with X-Ray Photoelectron Spectroscopy. J. Catal. 1979, 58 (1), 82-94. (13) Hartmann, M.; Pöppl, A.; Kevan, L. Ethylene Dimerization and Butene Isomerization in Nickel Containing MCM-41 and AlMCM-41 Mesoporous Molecular Sieves: An Electron Spin Resonance and Gas Chromatography Study. J. Phys. Chem. 1996, 100 (23), 9906-9910. (14) Sohn, J. R.; Shin, D. C. New Catalyst of NiO-ZrO2/WO3 for Ethylene Dimerization. J. Catal. 1996, 160 (2), 314-316. (15) Hwang, A.; Kim, S.; Kwak, G.; Kim, S. K.; Park, H.-G.; Kang, S. C.; Jun, K.-W.; Kim, Y. T. Low Temperature Oligomerization of Ethylene over Ni/Al-KIT-6 Catalysts. Catal. Lett. 2017, 1-12. (16) Barnett, K.W.; Glockner, P. W. Ethylene Oligomerization. US Patent 3527839 A, Sep 8, 1970. (17) Lallemand, M.; Finiels, A.; Fajula, F.; Hulea, V. Catalytic Oligomerization of Ethylene over NiContaining Dealuminated Y Zeolites. Appl. Catal. A - Gen. 2006, 301 (2), 196-201. (18) Lallemand, M.; Rusu, O. A.; Dumitriu, E.; Finiels, A.; Fajula, F.; Hulea, V. Ni-MCM-36 and Ni-MCM-22 Catalysts for the Ethylene Oligomerization. Stud. Surf. Sci. Catal. 2008, 174, 1139-1142. (19) Cai, F.; Lepetit, C.; Kermarec, M.; Olivier, D. Dimerization of Ethylene into 1-Butene over Supported Tailor-Made Nickel Catalysts. J. Mol. Catal. 1987, 43 (1), 93-116. (20) Metzger, E. D.; Brozek, C. K.; Comito, R. J.; Dincă, M. Selective Dimerization of Ethylene to 1-Butene with a Porous Catalyst. ACS Cent. Sci. 2016, 2 (3), 148-153. (21) Schultz, R. G. Olefin Dimerization over Cobalt-Oxide-on-Carbon Catalysts: III. Oligomerization of Ethylene. J. Catal. 1967, 7 (3), 286-290. (22) Schultz, R. G.; Engelbrecht, R. M.; Moore, R. N.; Wolford, L. T. Olefin Dimeirzation over CobaltOxide-on-Carbon Catalysts: II. Butene and Hexene Dimerization. J. Catal. 1966, 6 (3), 419-424. (23) Schultz, R. G.; Schuck, J.; Wildi, B. Olefin Dimerization over Cobalt-on-Carbon Catalysts: I. Propylene Dimerization. J. Catal. 1966, 6 (3), 385-396.

ACS Paragon Plus Environment

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

(24) Ort, M. R.; Schuck, J. M.; Schultz, R. G. Polymerization Process and Cobalt Oxide-Metal Oxide Catalyst Therefor. US Patent 3333017 A, Jul 25, 1967. (25) Addy, L. E.; Hill, T. Improvements Relating to the Oligomerisation of Olefines and Catalysts Therefor. GB Patent 1102298-A, 1966. (26) Hill, T. Olefin Oligomerization Catalysts. GB Patent 1183201, 1967. (27) Xu, Z.; Chada, J. P.; Zhao, D.; Carrero, C. A.; Kim, Y. T.; Rosenfeld, D. C.; Rogers, J. L.; Rozeveld, S. J.; Hermans, I.; Huber, G. W. Production of Linear Octenes from Oligomerization of 1-Butene over Carbon Supported Cobalt Catalysts. ACS Catal. 2016, 6 (6), 3815-3825. (28) Zhao, D.; Xu, Z.; Chada, J. P.; Carrero, C. A.; Rosenfeld, D. C.; Rogers, J. L.; Hermans, I.; Huber, G. W. Cobalt Oxide on N-Doped Carbon for 1-Butene Oligomerization to Produce Linear Octenes. ACS Catal. 2017, 7(11), 7479-7489. (29) Xu, Z.; Zhao, D.; Chada, J. P.; Rosenfeld, D. C.; Rogers, J. L.; Hermans, I.; Huber, G. W. Olefin Conversion on Nitrogen-Doped Carbon-Supported Cobalt Catalsyt: Effect of Feedstock. J. Catal. 2017, 354, 213-222. (30) Huber, G. W.; Shabaker, J. W.; Evans, S. T.; Dumesic, J. A. Aqueous-Phase Reforming of Ethylene Glycol over Supported Pt and Pd Bimetallic Catalysts. Appl. Catal. B - Environ. 2006, 62 (3), 226-235. (31) Simonetti, D.; Kunkes, E.; Dumesic, J. A. Gas-Phase Conversion of Glycerol to Synthesis Gas over Carbon-Supported Platinum and Rhenium Catalysts. J. Catal. 2007, 247 (2), 298-306. (32) Huber, G. W.; Shabaker, J. W.; Dumesic, J. A. Raney Ni-Sn Catalsyt for H2 Production from BiomassDerived Hydrocarbons. Science 2003, 300 (5628), 2075-2077. (33) Markovic, N. M.; Schmidt, T. J.; Stamenkovic, V.; Ross, P. N. Oxygen Reduction Reaction on Pt and Pt Bimetallic Surfaces: A Selective Review. Fuel Cells 2001, 1 (2), 105-116. (34) Iglesia, E.; Soled, S. L.; Fiato, R. A.; Via, G. H. Bimetallic Synergy in Cobalt Ruthenium Fischer-Tropsch Synthesis Catalysts. J. Catal. 1993, 143 (2), 345-368. (35) Bahome, M. C.; Jewell, L. L.; Padayachy, K.; Hildebrandt, D.; Glasser, D.; Datye, A. K.; Coville, N. J. FeRu Small Particle Bimetallic Catalysts Supported on Carbon Nanotubes for Use in Fischer-Tropsch Synthesis. Appl. Catal. A - Gen. 2007, 328 (2), 243-251. (36) Manyik, R. M.; Walker, W. E.; Wilson, T. P. A Soluble Chromium-Based Catalyst for Ethylene Trimerization and Polymerization. J. Catal. 1977, 47 (2), 197-209. (37) Agapie, T. Selective Ethylene Oligomerization: Recent Advances in Chromium Catalysis and Mechanistic Investigations. Coordin. Chem. Rev. 2011, 255 (7), 861-880. (38) Beck, D. D.; Lunsford, J. H. The Active Site for Ethylene Polymerization over Chromium Supported on Silica. J. Catal. 1981, 68 (1), 121-131. (39) Athena, A. HEPHAESTUS: Data Analysis for X-Ray Absorption Spectroscopy Using IFEFFIT. J. Synchrotron Radiat. 2005, 12 (4), 537-541. (40) Kim, Y. T.; Chada, J. P.; Xu, Z.; Pagan-Torres, Y. J.; Rosenfeld, D. C.; Winniford, W. L.; Schmidt, E.; Huber, G. W. Low-Temperature Oligomerization of 1-Butene with H-Ferrierite. J. Catal. 2015, 323, 33-44. (41) Naumkin, A. V.; Kraut-Vass, A.; Powell, C. J. NIST X-ray photoelectron spectroscopy database; NIST: Washington, DC, 1989. (42) Wdowik, U. D.; Parlinski, K. Lattice Dynamics of CoO from First Principles. Phys. Rev. B 2007, 75 (10), 104306/1 - 104306/6. (43) Shen, Z.-X.; Allen, J. W.; Lindberg, P. A. P.; Dessau, D. S.; Wells, B. O.; Borg, A.; Ellis, W.; Kang, J. S.; Oh, S.-J.; Lindau, I.; Spicer, W. E. Photoemission Study of CoO. Phys. Rev. B 1990, 42 (3), 1817-1828. (44) Petitto, S. C.; Marsh, E. M.; Carson, G. A.; Langell, M. A. Cobalt Oxide Surface Chemistry: the Interaction of CoO(100), Co3O4(110) and Co3O4(111) with Oxygen and Water. J. Mol. Catal. A-Chem. 2008, 281 (1), 49-58.

ACS Paragon Plus Environment

Page 20 of 21

Page 21 of 21 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

(45) Zhang, X.; van den Bos, C.; Sloof, W.; Terryn, H.; Hovestad, A. Effect of Drying Temperature on Chromate Conversion Coatings on Zinc. J. Corrosion Sci. Eng. 2003, 6, Paper no. 57. (46) Zhang, Y.; Kumar Megarajan, S.; Xu, X.; Lu, J.; Jiang, H. Catalytic Abatement of Nitrous Oxide Coupled with Ethane Oxydehydrogenation over Mesoporous Cr/Al2O3 Catalyst. Catalysts 2017, 7 (5), 137, 1-11. (47) Hu, S.; Scudiero, L.; Ha, S. Electronic Effect on Oxidation of Formic Acid on Supported Pd-Cu Bimetallic Surface. Electrochim. Acta 2012, 83 (30), 354-358. (48) Coq, B.; Figueras, F. Bimetallic Palladium Catalysts: Influence of the Co-Metal on the Catalyst Performance. J. Mol. Catal. A-Chem. 2001, 173 (1), 117-134. (49) Hartwig, J. F. In Organotransition metal chemistry: from bonding to catalysis; Edward Brothers, Inc.: Sausalito, CA, 2010; p 47. (50) Li, W.-Y.; Xu, L.-N.; Chen, Co3O4 Nanomaterials in Lithium-Ion Batteries and Gas Sensors. J. Adv.Funct. Mater. 2005, 15 (5), 851-857. (51) Wang, H.-Y.; Hung, S.-F.; Chen, H.-Y.; Chan, T.-S.; Chen, H. M.; Liu, B. In Operando Identification of Geometrical-Site-Dependent Water Oxidation Activity of Spinel Co3O4. J. Am. Chem. Soc. 2016, 138 (1), 36-39. (52) Zecchina, A.; Groppo, E. Surface Chromium Single Sites: Open Problems and Recent Advances. Proc. R. Soc. A 2012, 468 (2143), 2087-2098. (53) Myers, D. L.; Lunsford, J. H. Silica-Supported Chromium Catalysts for Ethylene Polymerization: The Active Oxidation States of Chromium. J. Catal. 1986, 99 (1), 140-148. (54) Arlman, E. I.; Cossée, P. Ziegler-Natta Catalysis II. Surface Structure of Layer-Lattice Transition Metal Chlorides. J. Catal. 1964, 3 (1), 89-99. (55) Kolasinski, K. W. In Surface science: foundations of catalysis and nanoscience, 3rd ed.; John Wiley & Sons: 2012; p 267-303. (56) Eyring, H. The Activated Complex in Chemical Reactions. J. Chem. Phys.1935, 3 (2), 107-115. (57) Yaws, C. L. In Chemical properties handbook. McGraw-Hill: 1999; p 264-288. (58) Moro-oka, Heats of Adsorption for C4, C5 and C6 Hydrocarbons on Cobalt Oxide Determined by Gas Chromoatography. Y. Faraday Soc. 1971, 67, 3381-3387.

ACS Paragon Plus Environment