Using Subcritical Water for Decarboxylation of Oleic Acid into Fuel

Mar 7, 2017 - Current interest in renewable fuel production is focused on high-performance fuels such as jet fuel because of their premium value in th...
0 downloads 14 Views 1MB Size
Subscriber access provided by University of Colorado Boulder

Article

Using subcritical water for decarboxylation of oleic acid into fuel range hydrocarbons Md Zakir Hossain, Anil Kumar Jhawar, Muhammad B.I. Chowdhury, William Z. Xu, Wei Wu, David Hiscott, and Paul A. Charpentier Energy Fuels, Just Accepted Manuscript • DOI: 10.1021/acs.energyfuels.6b03418 • Publication Date (Web): 07 Mar 2017 Downloaded from http://pubs.acs.org on March 14, 2017

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 free 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 accessible to all readers and 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.

Energy & Fuels 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 36

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

Energy & Fuels

213x160mm (96 x 96 DPI)

ACS Paragon Plus Environment

Energy & Fuels

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

Using subcritical water for decarboxylation of oleic acid into fuel range hydrocarbons Md Zakir Hossain, Anil Kumar Jhawar, Muhammad B.I. Chowdhury, William Z. Xu, Wei Wu, David Hiscott, Paul A. Charpentier*

1

Department of Chemical and Biochemical Engineering, The University of Western Ontario, London, Ontario, Canada N6A 5B9

* Corresponding author E-mail: [email protected] The University of Western Ontario London, Ontario, Canada N6A 5B9 Phone: 1 (519) 661-3466 Fax: 1 (519) 661-3498

ACS Paragon Plus Environment

Page 2 of 36

Page 3 of 36

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

Energy & Fuels

ABSTRACT Current interest in renewable fuel production is focussing on high performance fuels such as jet fuel due to premium values in the marketplace. Currently, lower value fuels such as biodiesel can be provided via a variety of feedstocks, yet contain significant amounts of oxygen, hence lowering its fuel value. In this work we examined a one-pot catalytic hydrothermal process for the decarboxylation of oleic acid as a model compound for free fatty acids with an activated carbon catalyst. Temperature (350-400 oC), water to oleic acid ratio (2:1 to 4:1) (v/v), catalyst, catalyst to total feed ratio (0.15 to 0.75) and residence time (1 to 2 h) were key factors for removing oxygen from oleic acid. The complete removal of the carboxylic group from the upgraded liquid phase was achieved at 400°C with a water to oleic acid ratio 4:1(v/v) and 2 h residence time as confirmed by FTIR and 13C NMR results. The pseudo first order reaction rate constant was found to follow Arrhenius behaviour with the activation energy determined to be 90.6±3 kJ/mol. The GC-FID results showed a high selectivity to heptadecane conversion, while the GC-TCD results indicated that decarboxylation was the dominating chemical reaction. High heating values and fuel densities were obtained using this approach, in the range of commercial jet fuels, without adding high pressure hydrogen or a hydrogen donor solvent.

KEYWORDS: Hydrothermal decarboxylation, Subcritical water, Activated carbon, Oleic acid, Fuel like hydrocarbons.

ACS Paragon Plus Environment

Energy & Fuels

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 36

1. INTRODUCTION The depletion of fossil fuel reserves and increasing greenhouse gas emissions have encouraged researchers to explore biomass as a renewable feedstock to produce liquid transportation fuels. 1 Esterification or transesterification of long-chain fatty acids or their derivatives for liquid fuel production such as biodiesel is a well-established process.

2, 3

However, the poor cold flow

properties such as high pour and cold filter plugging points are major drawbacks for using biodiesel in Northern climates. 4 Although the esterified products have a lower higher heating value (HHV) than petroleum diesel, the high carbon number long-chain fatty acids can give sufficiently high heating values if properly deoxygenated. The deoxygenated hydrocarbons have higher energy density, lower nitrogen oxide (NOx) emissions, lower acidity, lower viscosity, higher oxygen stability and are better suited for the existing infrastructure for distribution and vehicles.

5

Deoxygenated products also have a much higher Cetane number (>70) than that of petroleum diesel fuel (∼45), while the boiling point range is comparable to typical petroleum based-diesel. 6 To produce deoxygenated fuels, free fatty acids or their derivatives can be decarboxylated or decarbonylated. 7 Both reactions are thermodynamically favorable at 300 oC: ∆Grxn = -83.5 kJ/mol and -17 kJ/mol for decarboxylation and decarbonylation, respectively, 8 but complete removal of oxygen can only be achieved through the decarboxylation reaction. Decarboxylation requires less hydrogen compared to other processes (such as hydrodeoxygenation), thereby reducing production costs. In a previously proposed decarboxylation mechanism of fatty acids, the carboxylic acid group was found to adsorb to the catalyst surface, with the -COOH group removed through a CC cleavage to release CO2, thereby forming a hydrocarbon with one carbon less than the original fatty acid chain. 9 Heterogeneous catalytic decarboxylation of fatty acids has previously been done either in organic solvent such as dodecane or mesitylene or without any solvent.

ACS Paragon Plus Environment

10-13

However,

Page 5 of 36

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

Energy & Fuels

use of sub or supercritical water has been shown to enhance the decarboxylation mechanism.

14

Water at high temperatures and pressures, is an environmentally benign solvent, which has intriguing physico-chemical properties making its usage both challenging and potentially useful. For example, water at supercritical conditions loses its hydrogen bonding ability, becoming more like a non-polar solvent. Also its density and dielectric properties vary widely depending on the utilized T, P. 15-17 Decarboxylation of fatty acids or their derivatives provides low hydrocarbon yield at moderate temperatures ( 95% for this experiment, ranging for all the experiments performed in this study from 90 to 95%. Table 1 shows a typical selectivity or % composition of the products at 400 oC, 2h reaction time and 4:1 water to OA ratio at 800rpm in presence of 5 g activated carbon catalyst. As shown from this Table, the formed product can be used as aviation fuel which consists of C8 to C17 alkanes, alkenes and aromatic hydrocarbons. 25, 31

ACS Paragon Plus Environment

Page 15 of 36

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

Energy & Fuels

Table 1. Product distribution of oleic acid decarboxylation at 400 oC, 2 h reaction time and 4:1 water to OA ratio at 800 rpm

Compounds Octane (C8H18) Nonane (C9H20) Decane (C10H22) Undecane (C11H24) Dodecane (C12H26) Tridecane (C13H28) Tetradecane (C14H30) Pentadecane (C15H32) Hexadecane (C16H34) n-Heptadecane (C17H36) Heptadecene (C17H34) Octadecane (C18H38) Nonadecane (C19H40) Icosane (C20H42) Oleic acid

% Selectivity (composition) 0 0 0 0 2.2 5.1 2.8 4.4 4.0 81 0.6 0 0 0 0

3.2 Fuel Quality Specific gravity is an important parameter for any liquid because it helps to determine the usability of the fuel product. The value of specific gravity of the decarboxylated product at different temperatures and some of the commercial fuels is shown Table 2. If we compare the experimental data with the conventional fuels, our experimental decarboxylated product falls within typical diesel range.

ACS Paragon Plus Environment

Energy & Fuels

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 36

Table 2. Specific gravities data for the decarboxylated product at 400 oC, 2 h reaction time and 4:1 water to OA ratio at 800 rpm (in presence of 5 g catalyst) and commercial fuels Temperature (oC) 15.6 21.6 25 40 15.6 15 15.6

Compounds Decarboxylated product

Kerosene 32 Jet fuel 33 Diesel 32

Specific gravity 0.798 0.798 0.792 0.780 0.78-0.82 0.78-0.84 0.80-0.96

The heating value of a fuel is the amount of heat released during the combustion of a specified amount, which is characteristic for each fuel. Table 3 shows the HHV values of the feed, decarboxylated product and some of the commercial fuels, respectively. If we compare HHV of decarboxylated product is slightly higher than that of jet fuel and diesel, respectively whereas the value is slightly lower than that of kerosene, which means that the product quality lies within jet fuel, kerosene and diesel ranges. Table 3. High heating values of feed, product and commercial fuels Compounds

HHVs (MJ/kg)

Oleic acid Decarboxylated product

39.22 *

45.73

Jet fuel 34 43.54 32 Kerosene 46.20 Diesel 32 44.80 *product at optimum conditions

ACS Paragon Plus Environment

Page 17 of 36

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

Energy & Fuels

3.3 Mechanism of Decarboxylation 3.3.1 ATR-FTIR Analysis of Decarboxylated Liquid Products To examine the degree of decarboxylation of oleic acid under different reaction conditions, ATRFTIR spectra of oleic acid and the formed products were measured and are compared in Figure 3. The spectrum of oleic acid (Figure 3a) shows several major peaks at 3004, 2921, 2852, 1707, 1463, 1412, 1284, 934, and 722 cm-1. The peak at 3004 cm-1 is ascribed to the alkene CH stretching mode. The peaks at 2921, 2852, and 722 cm-1 are attributed to asymmetric stretching, symmetric stretching, and rocking modes of CH2, respectively. The peak at 1463 cm-1 is assigned to CH2 scissoring and CH3 asymmetric bending modes. The peak at 1707, 1412, 1284, and 934 cm-1 are attributable to C=O stretching, combination of C-O stretching and O-H deformation, C-O stretching, and OH out of plane bending modes, respectively. After the decarboxylation reactions in the presence of 5 g of activated carbon, all the peaks related to the C=O, C-O, O-H, and alkene C-H vibrations decrease (Figure 3b-e and g-h) or completely disappear (Figure 3f). By comparing Figure 3b, c and f, it is found that the degree of decarboxylation increased with increasing ratio of H2O to oleic acid from 2:1 to 4:1 in the reaction for 2 h at 400 oC. By comparing Figure 3d, e and f, it is seen that the degree of decarboxylation increased with increasing reaction time from 1 h to 2 h at 400 oC with the ratio of H2O to oleic acid being 4:1. By comparing Figure 3f, g and h, it is observed that the degree of decarboxylation increased with increasing reaction temperature from 350 oC to 400 oC for 2 h with the ratio of H2O to oleic acid being 4:1. By varying reaction conditions, complete decarboxylation and conversion of alkenyl group of oleic acid were achieved by running the reaction at 400 oC for 2 h with a ratio of H2O to oleic acid of 4:1, as evidenced by the absence of the peaks of C=O, C-O, O-H and the alkene C-H peak in Figure 3f.

ACS Paragon Plus Environment

Energy & Fuels

Absorbance

1707, C=O stretching

Fig 3 350, 375, 400_4-1_2h, 1.esp 2921, CH2 asym. stretching

1412, O-H and C-O combination

2871, CH3 sym. stretching 2852, CH2 sym. stretching

1463, CH2 scissoring & CH3 asym. bending

2955, CH3 asym. stretching

4.0

1377, CH3 umbrella 722, CH2 rocking 1284, C-O stretching 934, OH out of plane bending

3004, CH stretching

a) oleic acid (OA) 3.5

b) 400oC, 2h, H2O:OA=2:1

Absorbance (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 18 of 36

3.0

c) 400oC, 2h, H2O:OA=3:1

2.5

d) 400oC, 1h, H2O:OA=4:1

2.0

e) 400oC, 1.5h, H2O:OA=4:1

1.5

f) 400oC, 2h, H2O:OA=4:1 1.0

g) 375oC, 2h, H2O:OA=4:1 0.5

h) 350oC, 2h, H2O:OA=4:1 0 3400

3200

3000

2800

2600

2400

2200

2000

1800

1600

1400

1200

1000

800

-1

Wavenumber (cm )

Figure 3. ATR-FTIR spectra of (a) oleic acid and (b-h) the formed products under different reaction conditions with 5 g of catalyst loading.

In order to study the effect of catalyst on decarboxylation, the reaction was then conducted at 350, 375, and 400 oC for 2 h with a ratio of H2O to oleic acid of 4:1 but without using any catalyst. The FTIR results showed that the peaks of the formed products were almost identical to those of oleic acid (see Figure S2), indicating that no reaction took place without using a catalyst under the testing conditions. To study the effect of catalyst on decarboxylation of oleic acid, reactions were performed by varying the ratio (w/w) of catalyst to feed from 0.15 to 0.75. As shown in Figure 4, by increasing the ratio of catalyst to feed from 0.15 to 0.75, the peaks of C=O, C-O, and O-H

ACS Paragon Plus Environment

Page 19 of 36

decrease gradually and eventually disappear, suggesting that the catalyst played a key role in the

Absorbance

decarboxylation of oleic acid.

1707

Fig 4b catalyst 0.75, 0.45, 0.15, OA.esp

2921 2.5

2852 1412 1463

2.0

Absorbance (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

Energy & Fuels

722 1284 934

3004 a) oleic acid

1.5

b) catalyst : feed=0.15

1.0

c) catalyst : feed=0.45 0.5

d) catalyst : feed=0.75 0 3400

3200

3000

2800

2600

2400

2200

2000

1800

1600

1400

1200

1000

800

Wavenumber (cm-1)

Figure 4. A comparison of ATR-FTIR spectra of (a) oleic acid and (b) the formed products after reaction at 400 oC for 2 h by using different ratios (w/w) of catalyst to feed.

To observe the catalytic activity of the spent catalyst for decarboxylation, a sample of the used catalyst was examined under the optimum reaction conditions, i.e, 400 oC, 2h reaction time, water : OA = 4:1 and rpm = 800. Figure S3 compares the FTIR results of decarboxylation of oleic acid using both the fresh and spent catalysts. In comparison to almost complete removal of carboxylic group by using the fresh catalyst, only about 70% of the carboxylic group was removed using the

ACS Paragon Plus Environment

Energy & Fuels

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 36

spent one. To obtain higher decarboxylation efficiency using the spent catalyst, regeneration of the catalyst will be required. This will be studied in our future work.

3.3.2 NMR analysis of Decarboxylated Liquid Products In order to better understand the conversion of alkenyl group and decarboxylation of oleic acid, the decarboxylated liquid products were further characterized by 1H and

13

C NMR. Figure 5

compares the 1H NMR spectra of oleic acid and the formed liquid products. In the spectrum of oleic acid (Figure 5a), there are several proton peaks located at 10.74 (a broad peak, not shown), 5.35, 2.36, 2.04, 1.64, 1.30, and 0.89 ppm, which are attributed to carboxylic acid (1), alkenyl (10 and 11), methylene (3), methylene (9 and 12), methylene (4), methylene (5-8 and 13-18), and methyl (19) protons. In the spectrum of the formed products (Figure 5b), all the peaks related to the carboxylic group (1, 3 and 4) and the alkenyl group (9-12) disappear, suggesting complete decarboxylation and conversion of alkenyl group.

ACS Paragon Plus Environment

Page 21 of 36

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

Energy & Fuels

O

4 HO

1

2

3

5

10

8

6 7

11

9 12

13 14

CDCl3

1.0 h

15 16

17 18

5-8,13-18 19

19

0.5

3

9,12

10,11

a

4

0

-0.5

-1.0

b 8.0

7.5

7.0

6.5

6.0

5.5

5.0

4.5

4.0

3.5

3.0

2.5

2.0

Chemical Shift (ppm)

Figure 5. 1H NMR spectra of (a) oleic acid and (b) the formed products.

Similarly, in the 13C NMR spectrum of oleic acid (Figure 6a), there are several peaks located at 180.6, 129.7-130.0, 34.1, 31.9, 29.0-29.8, 27.2, 24.6, 22.7 and 14.1 ppm, which are attributed to the carboxylic (2), alkenyl (10 and 11), methylene (3), methylene (18), methylene (5-8 and 13-16), methylene (9 and 12), methylene (4), methylene (17), and methyl (19) carbons, respectively. After the decarboxylation reaction, the peaks related to the carboxylic group (2-4) and the alkenyl group (9-12) disappear, indicating complete decarboxylation and conversion of alkenyl group. Therefore, all these NMR results confirm complete conversion of carboxylic and alkenyl groups after the

ACS Paragon Plus Environment

Energy & Fuels

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 22 of 36

decarboxylation reaction under the selected reaction conditions, being in good agreement with the FTIR results discussed above. O

4 HO

1

2

3

5

10

8

6 7

11

9 12

13 14

15 16

1.0 c

5-8,13-16

17 18

19

3

10,11

0.5

18

2

9,12 4 17 19

CDCl3

a 0

-0.5

-1.0

b 180

160

140

120

100

80

60

40 Chemical Shift (ppm)

Figure 6. 13C NMR spectra of (a) oleic acid and (b) the formed products.

3.3.3 GC-TCD Analysis of Gaseous Products The decarboxylation of oleic acid under all the experimental conditions tested in the presence of catalysts provided the formation of gaseous products. Carbon dioxide and carbon monoxide were mainly detected in the gaseous products under the reaction conditions investigated. Figure 7 shows

ACS Paragon Plus Environment

Page 23 of 36

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

Energy & Fuels

the mole percentage of CO and CO2 in the gas fraction of oleic acid decarboxylated products at 400 oC, 2h reaction time and 800 rpm in presence of 5g catalyst. The result follows the general trends under the reaction conditions investigated. Previous studies have suggested the formation of these carbonaceous species through decarboxylation/decarbonylation of free fatty acids.

35, 36

Several researchers have also reported the presence of both CO and CO2 in the gas fraction of thermal and catalytic pyrolysis of fats and oils as well as model triglycerols. 36, 37 In our case, at 400 oC, the amount of CO2 is much higher than the amount of CO in the gas mixture. It has been shown by Akgul et al.29 that the water gas shift reaction is still possible in presence of less CO and excess water in the reaction mixture. GC-TCD results are consistent with ATR-FTIR and NMR results. Some other compounds such as methane and some other lighter fractions of hydrocarbons were also confirmed by GC-TCD (data not shown).

Figure 7. Percentage of CO and CO2 in the gas fraction of oleic acid decarboxylated products.

ACS Paragon Plus Environment

Energy & Fuels

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 36

3.3.4 Proposed Reaction Mechanism A proposed reaction mechanism for the hydrothermal decarboxylation of oleic acid in subcritical water based on our experimental and catalyst characterization results is shown in scheme 1. The insitu hydrogen produced during either the thermal cracking of oleic acid, 30 and/or the water gas shift reaction29, will hydrogenate the feedstock oleic acid (C17H33COOH) into stearic acid (C17H35COOH). 26 Higher temperatures such as at 400oC promoted the thermal cracking reaction, which produced larger amounts of hydrogen, leading to complete hydrogenation even in the blank experiment without activated carbon (see Figure S2). The formed stearic acid was finally decarboxylated in the presence of activated carbon to form heptadecane (C17H36). It was found by ATR-FTIR that the catalytic performance of the fresh activated carbon is higher than that of the spent one (Figure S3) while the XPS results revealed more C-O/C=O structures on the surface of the fresh catalyst than the spent one (Table S2). This indicates that the electrophilic carboxylic C from stearic acid was attacked by the nucleophilic O on the surface of the activated carbon, creating a tetrahedral intermediate. Subsequent proton transfer from the O+ to the O- and elimination of heptadecane resulted in the formation of hydrogen carbonate, which further decomposed back to activated carbon by releasing CO2.

ACS Paragon Plus Environment

Page 25 of 36

C17H33COOH

thermal cracking

lighter HCs + CO + CO 2 + H2

water gas shift

CO + H2O

CO2 + H2

C17H33COOH + H2

hydrogenation

O

C17H35COOH -

O

O C17H35

H H

H H

O

C17H35

O

O AC

O

C17H35

+

O

H

AC H O

:

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

Energy & Fuels

O

O AC

C17H36

H AC

O

CO2 HCs: hydrocarbons; AC: activated carbon (catalyst)

Scheme 1. Proposed reaction mechanism of hydrothermal decarboxylation of oleic acid in subcritical water on activated carbon (AC).

3.4 Catalyst Characterization 3.4.1 BET surface area analysis and pore size distribution Table 4 compares the BET surface area, pore volume and pore size of fresh and spent catalysts. The results show that the spent activated carbon catalyst slightly loses its activity due to decreasing surface area and pore volume. The drop in surface area and pore volume of spent activated carbon may be due to pore blockage by product molecules that were not completely recovered. The increase of pore size indicates pore breaking during the decarboxylation reaction.

ACS Paragon Plus Environment

Energy & Fuels

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 36

Table 4. BET surface area, pore size and pore volume of fresh and spent activated carbon Samples Fresh Activated Carbon Spent Activated Carbon

Total surface area (m2/g) 851 544

Total pore volume (cm3/g) 0.56 0.45

Average pore size (nm) 2.6 3.6

Figure S4(a) depicts the N2 adsorption-desorption isotherms of fresh and spent activated carbon at -193oC exhibiting a type IV isotherm with a type III hysteresis loop in the relative pressure range from 0.4 to 1.0. This indicates plate type particles or slit shaped mesopores. 38, 39 Activated carbon has a high BET surface area attributed by mesopores and micropores, respectively. The spent catalyst shows a drop in surface area although a similar pore volume and pore size after the decarboxylation reaction. Figure S4(b) shows the pore size distributions of the activated carbon. Both fresh and spent catalyst show a narrow pore size distribution centered at 4 nm which is in the mesoporous region (2 to 50 nm).

3.4.2 XRD analysis Figure 8 shows the XRD patterns of fresh and spent activated carbon. The appearance of broad diffraction peaks in the range of 2θ ∼15–35◦ and ∼40–50◦ ascribes the randomly arranged amorphous carbon structures containing low content of crystalline graphite

40

. There is no

significant difference in the XRD patterns between fresh and spent activated carbon except shifting of the first peak slightly to lower 2θ values, which may be due to the accumulation of impurities during the decarboxylation reaction. Coke deposition on a catalyst’s surface is quite common when exposed to high temperatures, which will dramatically reduce the catalytic activity and hinder the reusability of the catalyst. As previously reported, deposited coke (graphitic) usually shows XRD peaks at 29.84° and 61.92° on

ACS Paragon Plus Environment

Page 27 of 36

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

Energy & Fuels

the spent catalyst surface

41

. However, our XRD pattern of the spent activated carbon does not

show any of these peaks, indicating no graphitic coke deposition occurred during decarboxylation of oleic acid in the present study.

Figure 8. XRD patterns of fresh and spent activated carbon.

3.4.3 SEM analysis The surface morphologies of the fresh and spent activated carbon were examined by SEM analysis. SEM images of fresh and spent activated carbon are shown in Figure S5. The SEM image of the fresh catalyst shows porous structure whereas the spent catalyst shows slightly deactivated structure due to agglomeration. The porous structure of the fresh catalyst contributes to the large BET surface, as provided in Table 4.

ACS Paragon Plus Environment

Energy & Fuels

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 36

3.4.4 ATR-FTIR and XPS analysis Functional groups present on the surface are very crucial for activated carbon. Surface functional groups identify the surface properties of carbon and play a critical role in its catalytic activity. 42 The FTIR spectra of fresh and spent activated carbon samples were found to provide limited information (data not shown), which we attribute to the low concentration of functional groups at the carbon surface. Hence, we further examined the surface chemistry of the fresh and spent activated carbon by XPS. The survey XPS spectra of the fresh and spent catalysts is shown om Figure S6. The survey spectra of fresh and spent activated carbon show very distinct peaks of C and O and traces of Al, Fe, N, S and Si. The atomic percentage of these elements and the relative contents of the carbon species are calculated and summarized in Table S1 and Table S2. It was found that the concentration of Al, Fe, and Si on the surface of the activated carbon increased from 0.3%, 0.1%, and 0.6% to 0.8%, 1.2%, and 1.3%, respectively after the decarboxylation reaction, suggesting adsorption of impurities during the reaction. The surface functional groups present in the activated carbon were identified by high resolution C1s spectra as shown in Figure 9. The C1s spectra of fresh and spent activated carbon contain four peaks corresponding to C-C/C=C (284.5 eV), C-OH and C-O-C (286.5), C=O (287.9) and O-C=O (289.0).43 Among all the carbon peaks, the C-C/C=C peak is predominant for both the fresh and spent catalysts, which accounts for 86.4% and 94.2% of the carbon species, respectively. Moreover, the carbon species of O-C=O, C=O, and C-OH/C-O-C decreased from 4.2%, 2.6%, and 6.8% to 1.9%, 1.2%, and 2.7%, respectively after the decarboxylation reaction. These changes may be attributed to partial deactivation of the catalyst as evidenced by lower catalytic performance of the spent activated carbon than the fresh one discussed above.

ACS Paragon Plus Environment

Page 29 of 36

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

Energy & Fuels

Figure 9. High resolution C1s spectra of (a) fresh and (a) spent activated carbon.

3.4.5 Raman analysis of catalysts Raman spectroscopy is a powerful nondestructive technique to study carbonaceous materials. Figure 10 compares the Raman spectra of the fresh and the spent activated carbon. In the spectrum of the fresh activated carbon, the prominent peak commonly denoted as the G band centered at 1595 cm-1 corresponds to the first order scattering of sp2 hybridized carbon atoms. The D band centered at 1293 cm-1 arises from the defect or disorder sites of the sp3 carbon atoms. 44 Under the same measurement conditions, the intensities of both the D and G bands of the spent activated carbon are much lower than those of the fresh one, suggesting surface contamination by/adsorption of impurities after the decarboxylation reaction. This is in good agreement with the XPS results as discussed above.

ACS Paragon Plus Environment

Page 30 of 36

Spent and Fresh AC.esp

8500 8000 7500

1293 7000

1595 6500

Counts (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

Counts

Energy & Fuels

6000 5500

a

5000 4500

b 4000 3500 1920

1840

1760

1680

1600

1520

1440

1360

1280

1200

1120

1040

960

880

-1

Wavenumber (cm )

Figure 10. Raman spectra of (a) fresh and (b) spent activated carbon.

3.4.6 TGA Thermo-gravimetric analysis (TGA) was further used to compare the fresh and spent catalyst surfaces. Figure 11 shows the % weight loss along with derivative weight loss curves as a function of temperature for fresh and spent catalyst under N2 atmosphere. Weight loss observed in TGA profile before 200 oC are assigned to the removal of adsorbed water or gases from the environment or any easily removable carbonaceous species. The weight loss associated with fresh catalyst belongs to the removal of adsorbed water. The weight loss