César
Catizane
,
Ying
Jiang
* and
Joy
Sumner
School of Water, Energy and Environment, Cranfield University, College Rd, Cranfield, Wharley End, Bedford, UK. E-mail: y.jiang@cranfield.ac.uk
First published on 20th December 2023
Electrochemical hydrogenation (ECH) is a novel route for the upgradation of pyrolysis oil from both biomass and plastic feedstocks. Compared with conventional routes, including thermal cracking, ECH can be performed under mild conditions (<80 °C and 1 atm) and without the requirement of additional H2 supply. The successful demonstration of this application can be a critical step to enabling a circular plastic economy and low-carbon fuel production. In this review we provide a critical overview of the recent advancements in understanding the variables that influence the ECH process. In addition, we debate how this technology could be optimized and applied to plastic waste pyrolysis oil, assessing concerns such as the selection of cathode material, which needs to be resilient enough to address the complex nature of bio-oil. In addition, we present ideas on how to circumvent the challenge where the commonly used water-based electrolytes are unlikely to be suitable for pyrolysis oil treatment. Finally, we discuss the possible utilization of this product and scalability of this process.
Fig. 1 Mismanaged plastic waste trend for three possible scenarios.3 |
Many of these plastic products are designed for single-use, often discarded when they are still in pristine condition.2,3 In the EU for instance, in 2020, 29.5 million tonnes of post-consumer waste (PCW) plastics were collected and only 10% was sent to recycling plants, the majority of which via mechanical recycling.4–6 This presents a significant challenge, as most of the plastic waste is either consigned to landfills or incinerated. There is considerable potential for expanding thermochemical recycling processes, with pyrolysis emerging as the most notable option.5 Furthermore, the degradation and depolymerisation of plastic through weathering process leads to the production of micro and nano-plastics. These particles are ingested by animals, thereby being introduced into the food chain, and cause food contamination for both animals and humans.7
Fast pyrolysis is a thermochemical process that occurs in an inert environment, in the range of 450 to 600 °C and can be used for chemical recycling of biomass or plastic.8 This process products are separated into three categories: gas, wax, and oil. The resulting pyrolysis oil is a complex mixture composed of acids, alcohols, aldehydes, esters, ketones, sugars, phenols, guaiacols, syringols, furans, lignin-derived phenols, and extractible terpene with multi-functional groups.9 This is a product that potentially can be utilized as a biofuel, an alternative to petroleum-based fuels, or as feedstock for the chemical industry, which could help the creation of a circular economy10 (Fig. 2).
Fig. 2 Closed loop life cycle of plastic.11 |
Pyrolysis oil, however, usually has a high viscosity, low-density, low pH, high-water content (as high as 15–30 wt%),9 high oxygen-to-carbon (O:C) ratio and low hydrogen-to-carbon (H:C) content.12 All these factors make the storage, transport, and utilization of the product more difficult, due to corrosion and instability, as well as lower high heating values (HHV).8 For effective and widespread usage, the pyrolysis oil needs to be upgraded.13
Property (ASTM standard) | Value (average) |
---|---|
C | 56 wt% |
H | 6 wt% |
O | 38 wt% |
N | 0.2 wt% |
S | 0.02 wt% |
Water content (D95, E203) | 25 wt% |
pH acidity (D974, D664, D3339) | 2.5 |
Specific gravity (density compared to water) | 1.2 |
High heating value (D240, D4809) | 17 MJ kg−1 |
Viscosity (D88, D445, D2170) | 40–100 mPa |
Solids (char content) | 0.10 wt% |
Density (D1298, D4052) | 1.2 kg L−1 |
Pre-treatment is common practice in the industry in which the biomass goes through several physical–chemical processes before being pyrolyzed, such as torrefaction and trituration.18,19 Pyrolysis oil from a lignocellulosic (Acacia nilotica) raw biomass (PO-RAW) and torrefied biomass (PO-TB) are composed of furan derivatives, phenol derivatives, acids, alcohols, aldehydes, ketones, sugar derivatives, esters, and other compounds. Table 2 summarizes the most abundant species.19
Compound | Relative content (peak area (%)) | |
---|---|---|
PO-TB | PO-RAW | |
Furan derivatives | ||
2-Furancarboxaldehyde | 10.46 | — |
2-Furanmethanol | 5.62 | 6.05 |
2(3H)-Furanone, 5-methyl | 3.52 | 5.02 |
2-Furancarboxaldehyde,-5-methyl | — | 6.45 |
Phenol derivatives | ||
Phenol, 2-methoxy | 5.78 | — |
Phenol, 4-methoxy | — | 8.72 |
Creosol | 3.84 | 6.05 |
Phenol, 2,6-dimethoxy | 9.46 | 9.11 |
3,5-Dimethoxy-4-hydroxytoluene | 4.63 | 4.61 |
Phenol, 2,6-dimethoxy-4-(2-propenyl) | 6.68 | 0.82 |
Acids | ||
n-Hexadecanoic acid | 0.60 | 0.54 |
Linoelaidic acid | — | 1.90 |
Alcohols | ||
1-Penten-3-ol | 0.80 | 0.65 |
1,2-Benzenediol,3-methyl | — | 2.15 |
1-Octen-3-ol,acetate | 0.82 | — |
Aldehydes | ||
Benzaldehyde, 4-hydroxy-3-methoxy | — | 0.8 |
Benzaldehyde, 4-hydroxy-3,5-dimethoxy | 0.64 | 0.64 |
Ketones | ||
Alpha, beta-crotonolactone | 2.16 | 2.16 |
3-Methylcyclopentane-1,2-dione | 2.86 | 3.93 |
6-Methoxycoumaran-7-ol-3-one | 3.61 | — |
Sugar derivatives | ||
1,4:3,6-Dianhydro-alpha-D-glucopyranose | 0.30 | 1.47 |
Beta.-D-glucopyranose, 1,6-anhydro | — | 6.39 |
Esters | ||
2-(Acetyloxy) ethyl acetate | 0.53 | — |
Diethyl phthalate | — | 0.55 |
Di-n-octyl phthalate | 0.35 | 0.31 |
Other compounds | ||
1H-Pyrazole, 3,5-dimethyl | — | 9.34 |
Benzene, 1,2,3-trimethoxy-5-methyl | 3.18 | 1.91 |
Temperature is a critical factor during the pyrolysis process, significantly affecting product yields and altering the composition of the resulting oil.18,20–23 For example, during the pyrolysis of food waste (FW) and food waste solid digestate (FWSD), increasing the temperature from 500 to 800 °C led to a reduction in oil yield, decreasing from 19.5% to 8.2% for FW and from 15.9% to 8.3% for FWSD. Additionally, the yield of aliphatics in the bio-oil decreased from 18.91% to 16.86% for FW and from 33.78% to 25.64% for FWSD. Notably, as the temperature increased, carbon and hydrogen content in the oil increased, while oxygen content decreased. This desirable outcome for improving the oil stability and fuel quality.22 Rice husk pyrolysis oil composition was also influenced by temperature. Acids concentration varied from 7.07 wt% at 400 °C to 5.69 wt% at 600 °C, although it peaked at 8.46 wt% at 450 °C, ketones presented a similar behaviour, going from 8.65 wt% to 4.14 wt% for the same temperature increase. Aldehydes, phenols and nitrogenated compounds all presented a reduction in the concentration.23 Using a heavy metal contaminated mangrove as the biomass (temperatures range: 300–700 °C) the main compounds found in all cases were phenol derivatives, carboxylic acids, and alcohols. Analysis by GC-MS showed that independently of the temperature, the most abundant alcohol and phenol derivative were methylpropan-1-ol and catechol, respectively, although concentrations varied. The lower temperature (300 °C) led to a higher concentration of 2-methylpropan-1-ol, catechol, syringol and germanicol. For the higher temperatures (500–700 °C), the most abundant compounds were the same ones. A slight variation of the concentration of 2-methylpropan-1-ol and syringol occurred (0.59 and 0.76% variation, respectively), for catechol and germanicol, the lower temperature resulted in a noticeably higher concentration (2.80 and 2.09% variation, respectively).24
Phenol derivatives, lignin derivatives, benzene derivatives, alcohols and carboxylic acids are the main components of bio-oil, and account for the high oxygen concentration and instability of the oil due to low pH, when compared to conventional oils.19,24–27 The viscosity is also highly dependent on the feedstock, as reported by Luo et al.25 when comparing lignocellulosic biomass fast pyrolysis bio-oils. The results of the characterization show a total phenolic, lignin derivative and benzene derivative compound content of 32.26%, 32.36%, 15.78% respectively. All main compounds reported are oxygenated, most of which are aromatics. The results are summarized in Table 3.
Compound | Relative content (%) |
---|---|
Furfural | 9.06 |
Acetoxyacetone, 1-hydroxyl | 1.21 |
Furfural, 5-methyl | 1.82 |
Phenol | 2.55 |
2-Cyclopentane-1-one, 3-methyl | 1.58 |
Benzaldehyde, 2-hydroxyl | 2.70 |
Phenol, 2-methyl | 5.04 |
Phenol, 4-methyl | 0.51 |
Phenol, 2-methoxyl | 0.27 |
Phenol, 2,4-dimethyl | 9.62 |
Phenol, 4-ethyl | 2.18 |
Phenol, 2-methoxy-5-methyl | 4.15 |
Phenol, 2-methoxy-4-methyl | 0.55 |
Benzene, 1,2,4-trimethoxyl | 3.80 |
Phenol, 2,6-dimethyl-4-(1-propenyl) | 4.25 |
1,2-Benzenedicarboxylic acid, diisooctyl ester | 1.80 |
2-Furanone | 5.70 |
Levoglucosan | 6.75 |
Phenol, 2,6-dimethoxy-4-propenyl | 3.14 |
Furanone, 5-methyl | 0.49 |
Acetophenone, 1-(4-hydroxy-3-methoxy) | 2.94 |
Vanillin | 6.35 |
Benzaldehyde, 3,5-dimethyl-4-hydroxyl | 4.54 |
Cinnamic aldehyde, 3,5-demethoxy-4-hydroxyl | 2.19 |
Due to its highly complex nature, it is common to separate the bio-oil into hydrophilic, also referred to as water soluble bio-oil (WSBO) and hydrophobic shares. This process can increase the number of quantifiable compounds in the analysis, although it remains difficult to identify all components.26,27 Naturally, the water-soluble fraction presents a higher concentration of water, alcohols, and carboxylic acids, due to their polar nature. Sipila et al. reported that WSBO from straw, pine and ensyn (hardwood) oils show a high water content (19.9, 11.1 and 23.2% respectively), low pH (3.7, 2.6 and 2.8 respectively), and high carboxylic acid concentration (10.80, 7.30 and 6.20% respectively). Acetic and formic acids were the most significant components and the main reason for the low pH of the oils.27 When two commercial wood pyrolysis oils (BTG and Amaron) were separated in water-soluble (WS) and hydrophobic (oil), the authors reported only 34 directly quantifiable compounds, and the main components are shown in Table 4.26
Compound | BTG oil | Amaron oil | BTG WS | Amaron WS |
---|---|---|---|---|
Glycolaldehyde | 5.60 | 1.0 | 16.10 | 2.90 |
Acetic acid | 3.90 | 5.50 | 0.40 | 0.70 |
Acetol | 5.60 | 5.50 | 1.30 | 1.20 |
Levoglucosan | 3.50 | 3.00 | 0.20 | 0.20 |
Syringaldehyde | 0.10 | 0.10 | 8.10 | 6.80 |
Proximate analysis (wt%) | HDPE | LDPE | PET | PP | PS |
---|---|---|---|---|---|
a As received. b Dry. c Dry ash-free (HDPE: high-density polyethylene; LDPE: low-density polyethylene; PET: polyethylene terephthalate; PP: polypropylene; PS: polystyrene).28 | |||||
Moisture a | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 |
Ashb | 4.1 | 0.3 | 42.7 | 0.8 | 1.6 |
Volatilec | 98.1 | 100.0 | 94.6 | 100.0 | 100.0 |
Fixed carbonc | 1.9 | 0.0 | 5.4 | 0.0 | 0.0 |
Elemental analysis (%) | |||||
Cc | 82.5 | 83.3 | 69.0 | 81.0 | 87.8 |
Hc | 13.1 | 12.6 | 5.0 | 11.4 | 9.5 |
Oc | 1.8 | 0.7 | 32.7 | 0.6 | 0.0 |
Nc | 0.1 | 0.1 | 0.0 | 0.1 | 0.0 |
Plastic waste pyrolysis oil usually presents a highly aromatic compound share, when compared to bio-oil, in which can be found monoaromatics, diaromatics, naphthenoaromatics and triaromatics (51.3, 7.5, 5.9 and 2.3 wt% respectively). Some of the main components of the aromatic share of the oil are benzene, toluene, ethylbenzene, xylene, and styrene. The detailed composition of the oil can be found in Table 6.29
Component | Composition (wt%) |
---|---|
Paraffins | 5.0 |
Iso-paraffins | 8.2 |
Olefins | 12.3 |
Aromatics | 67.1 |
The feedstock variation is the main factor in plastic-oil composition In a process treating PET as the feedstock, the resulting pyrolysis oil contains a high concentration of benzene derivatives (10.6 wt%), ketones (5.90 wt%) and acids/esters (62.0 wt%). The most abundant compound in PET pyrolysis oil is benzoic acid and its derivatives, constituting 49.9 wt% of the oil.30 Cit et al. found similar results when comparing polyolefins (PP and LDPE) and PET, where PET pyrolysis oil main components were benzoic acid derivatives (benzoic acid = 23.7, 4-methyl-benzoic acid = 1.6, 4-ethylbenzoic acid = 1.3 and 4-acetylbenzoic acid = 10.3 wt%).31 LDPE pyrolysis oil has a high share of carbon doble bonds, as 39% of its composition are aromatic compounds and other unsaturated carbon molecules.32 A mix of LDPE and HDPE led to naphthalene derivates as the main components (approximately 24 wt%) in the plastic-oil, followed by p-xylene, toluene, naphthalene, benzene, 1-ethyl-4-methyl-benzene, m-xylene, 1,2,4-trimethyl-benzene, indane, indene and 1-methyl-3H-indene (approximately 19, 14, 9, 6, 3.5, 3, 2, 1.5, 1.5 and 1 wt% respectively).33 For PS, monoaromatics are the main components, followed by polyaromatics and aliphatics (77.62, 12.86 and 0.01 wt% respectively). As expected, the main component is styrene (73.60 wt%), other noticeable compounds being 2,4-diphenyl-1-butene (7.80 wt%), 2,4,6-triphenyl-1-hexene (2.21 wt%), α-methylstyrene (1.61 wt%), and toluene (1.58 wt%).34 When conducting a study on the performance of plastic pyrolysis oil as fuel, Kalagaris et al. reported on the composition of the oil. The feedstock was a mixture of styrene butadiene, polyester, clay, ethylene-vinyl acetate, rosin, polyethylene, and polypropylene (47, 26, 12, 7, 6, 1 and 1 wt% respectively). The main components of the oil are presented in Table 7.35
Compound | Composition (wt%) |
---|---|
Xylene | 8.80 |
Benzene | 5.30 |
Chlorinated phenols | 4.20 |
Naphthalene derivatives | 4.10 |
Cresols | 3.90 |
Naphthalene | 3.70 |
Cyclopentene | 2.90 |
Toluene | 2.70 |
Benzene derivatives | 1.00 |
Cyclobutane | 0.90 |
The addition of different feedstocks may also change the composition of the oil; that is, the oil composition of a PS/PP mixture may not be just the sum of PS and PP oils. Many of the components found in PP pyrolysis oil, such as methylnaphthalene (8.4 wt%), xylene (7.8 wt%) and phenanthrene (7.6 wt%), completely disappeared when the feedstocks were mixed. Fig. 3 shows the oil composition for several different feedstock compositions. The ratio of each plastic in each mixture was 50/50%, and in the last mixture (PS, PP, PE and PET) the ratio was 40/20/20/20%, respectively.36
Fig. 3 GC-MS results of liquid oils from pyrolysis of different types of plastic wastes. Reprinted with permission from ref. 36 Copyright 2017 Elsevier. |
Both temperature and pressure are vital factors in the pyrolysis process. Pyrolysis of LDPE under different conditions (425 °C/1.60 MPa, 450 °C/2.45 MPa and 500 °C/4.31 MPa) led to variation in both the classes of compounds and their concentration. The carbon chain length also changed, significantly reducing at higher temperatures/pressures. Table 8a shows the variation of the macro composition and Table 8b the specific compounds.37
Composition | 425 °C/1.60 MPa | 450 °C/2.45 Mpa | 500 °C/4.31 Mpa |
---|---|---|---|
(a) | |||
Naphthenes | 2.69 | 5.56 | 1.50 |
Alkanes | 46.20 | 31.7 | 17.8 |
Alkenes | 12.40 | 13.1 | 3.58 |
Aromatics | 12.00 | 22.9 | 68.0 |
Unknowns | 19.10 | 17.9 | — |
(b) | |||
Heptane | 2.26 | 3.16 | 0.46 |
Toluene | 0.35 | 3.05 | 24.30 |
Octane | 2.45 | 3.78 | 0.19 |
Ethylbenzene | 0.25 | 0.89 | 3.06 |
p/m-Xylene | 0.62 | 1.79 | 3.90 |
Allylbenzene | 0.21 | 1.02 | 4.84 |
Methylnaphthalene | 0.90 | 0.36 | 4.22 |
C21–C30 alkanes | 9.98 | 1.85 | 1.20 |
The same variation can be observed in the pyrolysis of PS (process conditions: 400 °C/1.14 MPa, 425 °C/1.26 MPa, 450 °C/1.47 MPa, and 500 °C/1.60 MPa). Results are summarized in Table 9.
Compound | 400 °C/1.60 MPa | 425 °C/2.45 MPa | 450 °C/4.31 MPa | 500 °C/4.31 MPa |
---|---|---|---|---|
Benzene | 0.38 | 0.73 | 1.27 | 1.63 |
Toluene | 21.70 | 23.10 | 27.00 | 28.40 |
Ethylbenzene | 32.60 | 39.30 | 39.00 | 36.60 |
Styrene | 1.09 | 0.80 | 0.39 | 0.61 |
Cumene | 10.20 | 9.10 | 9.36 | 9.60 |
Propylbenzene | 0.60 | 0.77 | 0.90 | 1.29 |
Methylstyrene | 1.37 | 0.39 | 0.33 | 0.55 |
Diphenylmethane | 0.91 | 0.83 | 1.00 | 1.61 |
1,2-Diphenylethane | 1.44 | 1.16 | 0.60 | 0.37 |
Phenylnaphthalene | 0.98 | 1.02 | 1.01 | 1.25 |
1,2-Diphenylbenzene | 1.27 | 1.26 | 1.87 | 1.89 |
Triphenylbenzene | 6.30 | 5.13 | 3.83 | 3.18 |
Long chain alkanes and alkenes are common products in plastic-oil, and the carbon chain lengths can reach as high as 55 carbons (C55) when the feedstock is a mixture of PP, LDPE, HDPE and styrene.14,15 The carbon number range of alkanes (Table 10a) and alkenes (Table 10b) are as follows:
Feedstock | C11–C20 | C21–C30 | C31–C40 | C41–C50 |
---|---|---|---|---|
(a) | ||||
Polypropylene | 2 | 63 | 0 | 0 |
LDPE | 6 | 54 | 12 | 13 |
HDPE | 1 | 67 | 18 | 0 |
Styrene | 0 | 4 | 27 | 8 |
(b) | ||||
Polypropylene | 9.9 | 8.9 | 0 | 0 |
LDPE | 3.3 | 4.2 | 0 | 0.8 |
HDPE | 0.5 | 6.2 | 1 | 1 |
Styrene | 0 | 0 | 0 | 10 |
Plastic waste (polyethylene film and mixed polyolefin) pyrolysis oils blended with fossil naphtha led to an even higher carbon number range. For polyethylene film (PE-film), was C5–C55 and for mixed polyolefin (MPO) was C5–C52. The composition groups of the resulting pyrolysis oils consisted of paraffins, iso-paraffins, α-olefins, iso-olefins, diolefins, naphthenes, monoaromatics and naphthenoaromatics.15 Both oils also present a substantial amount of several halogen and metal contaminants, among which are found Cl, Br, Al, Ca, Cr, Cu, Fe, Li, Na, Ni, Sb, Si, Sr and Zn. Both Cl and Br can be explained from the initial composition of the feedstock, as polyvinyl chloride (PVC) is a common plastic and bromine is an element that is used in fire retardants. Metal contamination can be traced to plastic enhancers, as metals are commonly used as additives.15 The detailed composition of the plastic-oil and the distribution of paraffins, olefins, naphthenes and aromatics over the light, medium and heavy product fractions of all post-consumer plastic waste pyrolysis oils can be found in Table 11.16
(a) | ||
---|---|---|
Composition | MPO pyrolysis oil | PE-film pyrolysis oil |
a Carbon number up to C20 and boiling point up to 250 °C. b Carbon number up to C30 and boiling point up to 450 °C. c Carbon number >C30 and FBP >450 °C.16 | ||
Paraffins | 32.2 | 41.9 |
Iso-paraffins | 3.7 | 3.5 |
α-Olefins | 23.5 | 33.0 |
Diolefins | 5.3 | 4.5 |
Naphthenes | 12.4 | 5.5 |
Monoaromatics | 2.7 | 1.0 |
Naphthenoaromatics | 0.2 | 0.1 |
For utilization as biofuel, HHV is a vital factor. Bio-oil has a lacklustre HHV when compared to transportation fuels (bio-oil HHV: 13–19 MJ kg−1 compared to gasoline: 44–46 MJ kg−1 or coal: 25–35 MJ kg−1). Plastic-oil, on the other hand, presents a high HHV (41–47 MJ kg−1).8
The main components from both bio and plastic oil are summarized in Fig. 4.
Upgrading routes can be separated into chemical (hydrotreatment, catalytic cracking, hydrocracking, steam reforming, hydrodeoxygenation and hydrogenation) and physical (distillation, supercritical fluid extraction, liquid–liquid extraction, and emulsification) processing. There are several advantages and disadvantages of each upgrading route; some of the disadvantages of the more traditional routes are the operational conditions, high costs and coke production. Catalytic cracking is accomplished at medium temperature and atmospheric pressure, but it leads to catalyst deactivation and the formation of carbon double bonds, which would require further improvement.51 Decarbonylation and decarboxylation leads to the loss of valuable carbon. Hydrocracking, hydrodeoxygenation, and hydrogenation require an outside source of H2, which is costly and difficult to store (see Fig. 5).43,44
Fig. 5 Examples of reactions associated with catalytic bio-oil upgradation. Reprinted with permission from ref. 44 Copyright 2011 Elsevier Inc. |
During ECH, water is oxidised at the anode (eqn (1)), the resulting H+, that can flow through a cation exchange membrane into the cathode cell. The H+ is then chemisorbed by the catalytic cathode (eqn (2)), where M is the metal surface and H(ads)M is the chemisorbed hydrogen.54,55
(1) |
H(aq)+ + e− + M ↔ H(ads)M | (2) |
In the cathode chamber, the organic unsaturated compound is chemisorbed onto the cathode (eqn (3)), where the hydrogenation process occurs (eqn (4)), followed by the desorption of the stabilized product (eqn (5)). Eqn (6) is the net reaction.54,55
(Y = Z)aq + S ↔ (Y = Z)(ads)S | (3) |
2H(ads)M + (Y = Z)(ads)S ↔ (YH–ZH)(ads)S + 2 M | (4) |
(YH–ZH)(ads)S ↔ (YH–ZH)(aq) + S | (5) |
Unsaturated compound + nH+ + ne−1 ↔ stabilized product | (6) |
2(H(ads)M) ↔ 2H2 + 2 M | (7) |
H(ads)M + H(aq)+ + e− ↔ H2 + M | (8) |
ECH can be performed under mild conditions (<80 °C and 1 atm) (Table 12). Lower temperatures can also avoid catalyst deactivation by coke formation and membrane fouling, potentially reducing the costs associated with catalyst purchase and recycle.56 However, due to high viscosity and low conductivity, it requires the utilization of membranes, catalysts, and electrolytes to facilitate the reaction.10,13,38,39 That leads to a vast spectrum of different arrangements that can be organized to optimize the process.
Method | T (°C) | P (bar) | H2 need |
---|---|---|---|
Cracking | 330–700 | 1.0 | No |
Hydrocracking | >400 | 68–96 | Yes |
Hydrodeoxygenation | 300–600 | 80–300 | Yes |
Electrochemical hydrogenation | <80 | ≈1 | No |
(9) |
This equation can also be simplified:
(10) |
(11) |
(12) |
(13) |
The choice of cathode can greatly interfere with the process, as early transition metals (e.g. Ti, Mn, Nb) and platinum group metals (Ru, Rh, Ir, and Pt) have higher affinity toward hydrogen formation, leading to more H+ in solution, which increases collision probability and, in turn, leads to higher conversion rates.51 Cathode surface coverage (or adsorption rate) can lead to the same problem of low collision probability, and is dependent on cathode/compound pairing, for example, the adsorption of phenol on Pt is stronger compared to guaiacol.42,62
Cathode selection can also influence the selectivity of the process and the desirable product yield. For the ECH of glucose, late transition metals (e.g. Fe, Cu, Pd, Ag and Au) led to higher sorbitol formation and post-transition metals (In, Sn, Sb, Pb, and Bi), Zn and Cd (d metals) favoured d 2-deoxysorbitol formation. Ni had the lowest and Pb the highest sorbitol formation potential.63
Platinum group metals (Pt, Pb, Ru, Rh and Ir) are commonly used in hydrogenation processes, due their stability and high affinity towards hydrogen production and have been employed under several conditions and for different compound classes.63 For the ECH of indigo, a highly unsaturated compound that contains both oxygen and nitrogen (reaction conditions: 1 M NaOH, 50 °C, 20 mA cm−2) using RANEY® nickel electrodes doped with different noble metal catalysts (Pt, Pd and Pd/C), Pt coating led to both optimal conversion and FE (10.2 and 8.16% respectively), followed by Pd (8.6, 6.88%), Pd/C (4.7, 3.76%) and RANEY® nickel without coating (4.4, 3.5%).64 Nobel metals are also capable of hydrogenating phenolic compounds at high conversion rates and FE. ECH of phenol on several cathodes (Pt/C, Rh/C, and Pd/C) demonstrated that Rh/C exhibited the highest hydrogenation rate, the order was Rh/C > Pt/C > Pd/C.65 Du et al. reported on the ECH of phenol on several different electrocatalysts: Pt, Ru, Pt3Ru, Pt3Sn, Ru3Sn, and Pt3RuSn supported on carbon cloth (reaction conditions: 0.2 M H2SO4 containing 10 mM phenol at −100 mA, 50 °C for 100 min). Pt3Ru/CC presented the highest conversion rate, at 96.3%, followed by Pt3RuSn/CC and Pt/CC (91.5 and 90.2% respectively).66 The concentration of the metal on the surface of the carbon can also lead to different results, although graphite alone was unable to react (reaction conditions: phenol, electrolyte 0.1 mol L−1 H2SO4 (30 mL), 30 mA, Pt sheet as the anode). 1% Pt/G and 1.5% Pt/G presented the highest conversion rates, both 95%. The highest yield of cyclohexane and the highest FE were 30.4% and 20.3%, respectively, both occurred with the same cathode (1.5% Pt/G). The results are summarized in Table 13.55 As shown in previous sections, lignin pyrolysis oil can contain phenolic compounds (phenol, p-cresol, 4-ethylphenol, and 4-propylphenol, as well as guaiacol) which can be hydrogenated into their corresponding alkylcyclohexanols at high conversion rates using Ru/ACC as the catalytic cathode. The same cathode is effective for transforming model lignin monomers into cyclohexanol.56
Cathode | Conversion (%) | F.E. | 1 | 2 | 3 | 4 |
---|---|---|---|---|---|---|
Where [1] is cyclohexane, [2] cyclohexanol, [3] cyclohexanone and [4] benzene. | ||||||
Pt | 0 | 0 | 0 | 0 | 0 | 0 |
Ni/G | 0 | 0 | 0 | 0 | 0 | 0 |
Rh/G | 90 | 0 | 0 | 65.7 | 24.3 | 0 |
Pd/G | 18 | 0 | 0 | 12.6 | 5.4 | 0 |
0.5% Pt/G | 92 | 11.1 | 16.6 | 57 | 12 | 5.5 |
1% Pt/G | 95 | 15.2 | 22.8 | 52.3 | 16.2 | 3.8 |
1.5% Pt/G | 95 | 20.3 | 30.4 | 58.9 | 5.7 | 0 |
2% Pt/G | 90 | 9.1 | 13.5 | 60.3 | 9 | 7.2 |
Graphite | 0 | 0 | 0 | 0 | 0 | 0 |
Metals outside the platinum group can also be utilized as the catalytic cathode, although it still requires more investigation. The ECH of guaiacol and related lignin models utilizing a nickel cathode and a Co/phosphate anode presented high selectivity towards cyclohexanol formation (>99%), with only traces of 2-methoxycyclohexanol. Other experiments were done utilizing cobalt and copper as catalyst cathodes, and only Co formed trace amounts of cyclohexanol and phenol, Cu completely failed in producing these products.58 The ECH of aldehyde compounds shows that aromatic aldehydes have a higher conversion rate than aliphatic aldehydes. Au, Ag, Cu, and C catalysts exhibit the highest conversion to alcohol products.67
Electrolytes are needed during ECH to increase the conductivity of the system, although they can interfere with the process. Acids, alkalis, and salts can achieve these results, and the selection of electrolytes should take into consideration both the organic compound and the cathode. Another challenge during this selection is that the electrolyte that presents the highest conversion rate and the one that leads to the highest FE might not be the same one. For the ECH of guaiacol (reaction conditions: 3-CE-NH3 cathode, 80 °C and 100 mA), the conversion rates, FE and product selectivity varied greatly. The highest conversion rates were NaOH (62 ± 2.4%), NaCl (64 ± 8.8%), and HCl (75 ± 3.9%). For FE, the order was NaCl (20 ± 2.2%) < NaOH (28 ± 1.0%) < HCl (30 ± 4.3%). Similar results were reported for phenol and syringol.39 The pH of the system can interfere with the results, as demonstrated by the ECH of guaiacol where HClO4 as the electrolyte led to the highest guaiacol conversion rates, HCl and H2SO4 led to moderate conversion rates (PtNiB/CMK-3 cathode, 0.2 M HClO4, 0.2 M HCl and 0.2 M H2SO4 as electrolytes). In addition, HClO4 achieved optimal selectivity for the desirable product (KA oil). However, increasing the concentration of sulfuric acid from 0.2 M to 0.5 M reduced the ECH activity.42
Different catholyte/anolyte pairings can also lead to vastly different results. Acid/acid pair (catholyte/anolyte respectively) lead to higher guaiacol conversion. (Conditions: T = 50 °C, I = 300 mA (j = −109 mA cm−2), catalyst = 5 wt% Pt/C (0.133 g), reaction time = 2 h, concentration of all electrolytes = 0.2 M) followed by salt/acid and acid/alkali. Alkali/acid completely failed in generating products. ECH of phenol led to similar results.62
Cathode | Feedstock | Conversion | Electrolyte | Temperature (°C) | J (mA cm−2) | F.E. | Duration (h) | Ref. |
---|---|---|---|---|---|---|---|---|
a Reported as mA. | ||||||||
Ru/ACC | Guaiacol | 64 ± 8.8 | NaCl | 20 ± 2.2 | ||||
75 ± 3.9 | HCl | 80 | 100a | 30 ± 4.3 | 2 | 39 | ||
Phenol | 89 | HCl | 29 | |||||
Syringol | 58 | HCl | 29 | |||||
PtNiB/CMK-3 | Guaiacol | >99% | HClO4 | 60 | 20a | 86.2 | 2 | 42 |
RANEY®-Ni | Guaiacol | >99% | BK3O3 | 75 | 8 | 26 | 6 | 58 |
Ru/ACC | Phenol | 100 | 20 | 4 | ||||
4-Methylphenol | 100 | HCl | 80 | 100a | 9 | 56 | ||
4-Ethylphenol | 98 | 9 | 6 | |||||
4-Proprylphenol | 100 | 6 | ||||||
RANEY®-Ni | 4.4 | 3.5 | 64 | |||||
RANEY®-Ni with Pt | Indigo | 10.2 | NaOH | 50 | 20 | 8.16 | — | |
RANEY®-Ni with Pd | 8.6 | 6.88 | ||||||
RANEY®-Ni with Pd/C | 4.7 | 3.76 | ||||||
Iron | Furfural | 97 ± 1.4 | NH4Cl | — | 6 | 29 ± 3.5 | 5 | 54 |
Pt/CC | 90.2 | 34.6 | ||||||
Ru/CC | 73.4 | 28.2 | ||||||
Pt3Ru/CC | Phenol | 96.3 | H2SO4 | 50 | 100a | 37.6 | 1.67 | 66 |
Pt3Sn/CC | 73.3 | 29.7 | ||||||
Ru3Sn/CC | 62.3 | 26.7 | ||||||
Pt3RuSn/CC | 91.5 | 39.5 |
Feedstock | Cathode | Catholyte | Anode | Anolyte | T (°C) | Duration (h) | J (mA cm−2) | Ref. |
---|---|---|---|---|---|---|---|---|
a Reported as mA. | ||||||||
Guaiacol | Boron-doped PtNiB/ordered mesoporous carbon (CMK-3) | 0.2 M HClO4 | IrO2/C | 0.2 M HClO4 | 60 | 5 | 40a | 42 |
Phenol derivatives | Ru/ACC | 0.2 M HCl | Pt wire | 0.2 M phosphate | 80 | — | 100a | 56 |
WSBO | Ru/ACC | 0.2 M NaCl | Pt wire | 1 M H2SO4 | 27 | 6.5 | 480a | 72 |
WSBO | Ru nanoparticles on mesoporous carbon–Ru@OMC | WSBO | Fe(III)/Fe(II) + carbon felt | Wastewater or lignin | — | 3 | — | 41 |
Resin wood (ponderosa pine) bio-oil | Metal-free (carbon) | 1 M Na2SO4 | Ti sheets/Pt coating | Purified water | 35(±2) | — | — | 40 |
Fig. 7 Schematic of ECH of guaiacol into KA oil integrated with O2 production, equipped with PtNiB/CMK-3 and IrO2/C as the cathode and anode, respectively. Reprinted with permission from ref. 42 Copyright 2019 Wiley. |
Faradaic efficiency (FE) of the process was 86.2%, in comparison to the 6.3% that was found when the reaction was carried out in a PtNi/CMK-3 cathode, that is, without boron doping. The conversion rate was respectively <99.0% and 11.7%. The study also reported on the ECH of other phenolic compounds using PtNiB/CMK-3. All achieved high (<95%) conversion rates and high (<80%) FE. The temperature increase effect on conversion rate and FE on range of 20–60 °C was positive (conversion rate: from 58% to 98.9% and FE: from 45% to 86.2%), but further heating to 80 °C reduced these parameters to 78% and 57% respectively42 (Fig. 8).
Fig. 8 ECH of other phenolic compounds using PtNiB/CMK-3. Reprinted with permission from ref. 42 Copyright 2019 Wiley. |
Garedew et al. studied the ECH of phenolic compounds. Several model compounds were selected, including catechol, anisole, phenol, guaiacol, 3-methoxyphenol, 4-methoxyphenol, syringol, 1,3-dimethoxybenzene, o-cresol, m-cresol, p-cresol, creosol, vanillin, syringaldehyde, eugenol, 4-ethylphenol, 4-propylphenol, 4-propylguaiacol and 4-ethylguaiacol.56
The cell setup was a H cell, separated by a Nafion 117 membrane. The cathode was Ru/ACC and the anode a Pt wire. Catholyte was a 0.2 M HCl solution (30 mL) and the anolyte a 0.2 M phosphate buffer (30 mL). Temperature was maintained at 80 °C and the experiment was run at 100 mA. The cell setup is shown on Fig. 9.56
Fig. 9 Electrocatalytic hydrogenation in a two-chambered H cell. Reprinted with permission from ref. 56. Copyright 2019 American Chemical Society. |
The authors reported that increasing guaiacol concentration from 5 to 20 mM increased the FE and product selectivity. Further increases from 20 to 60 mM had led to a steep decline in the conversion rate, although this was mitigated by increasing the duration of the process (Fig. 10).
Phenol, p-cresol, 4-ethylphenol, and 4-propylphenol, as well as guaiacol, were transformed into their corresponding alkylcyclohexanols at high conversion rates (>98%). Carbon chain length was a determinant factor in reaction speed, as increased lengths resulted in slower conversions.56
Fig. 10 Conversion, product selectivity, and faradaic efficiency of guaiacol reduction using different substrate concentrations (5, 10, 20, 40, and 60 mM) at 100 mA and 80 °C. Error bars are standard error. Reprinted with permission from ref. 56. Copyright 2019 American Chemical Society. |
Electrochemical hydrogenation of WSBO was carried out in a two-chamber glass H-cell, separated with a DuPont® Nafion117 membrane (see Fig. 11). The reaction temperature was 27 °C and the pressure was 1 atm. Anolyte consisted of 1 M sulfuric acid and Pt wire was used as the anode. A mixture of WSBO with 0.2 M NaCl served as the catholyte. Current was 480 mA.72
Fig. 11 Electrochemical hydrogenation cell setup for ECH of water-soluble bio-oil.72 |
After the 6.5 h treatment, most of the organic compounds were depleted or fully converted by hydrogenation, the proposed mechanisms are shown in Fig. 12. Removal of >50% of acetic acid is an expressive result, as it is a major contributor to bio-oil corrosiveness. No coke formation was observed. The same experiments were done using only ACC as the cathode and showed that Ru is an essential part of the proposed ECH process.39
Fig. 12 Reductions via ECH of organic compounds in water-soluble bio-oil.72 |
Although a noble metal catalyst (e.g., Pt and Ir) as the anode is normally used to ensure effective water oxidation, Zhang et al. found that a Fe(III)/Fe(II) redox pair can be utilized as a substitute when combined with carbon felt as the anode. This change can make the process less expensive and can further its usage. The authors presented a Ru-based ECH cathode catalyst by uniformly embedding Ru nanoparticles into the surface of the ordered mesoporous carbon (OMC) support (noted as Ru@OMC). The catholyte solution was WSBO, duration was 3 h.41 The cell setup is shown in Fig. 13.
Fig. 13 Schematic illustration of low-energy ECH of bio-oil: experimental setup. Reprinted with permission from ref. 41. Copyright 2018 American Chemical Society. |
The EHC treatment led to an increase in the H:C ratio and a decrease in the O:C ratio. Another result was the content reduction in several organic groups, such as acids, esters, carbonyls, phenols, sugars, and furans. The alcohol content increase was substantial, and the process elevated the selectivity of polyhydric alcohols. The results infer that this technique provides de ability to saturate unsaturated components in WSBO.41
Fig. 14 Spatial chemical representation (a) and diagram of the separation−hydrogenation process (b). Reprinted with permission from ref. 40. Copyright 2018 American Chemical Society. |
The cell temperature was held at 35 °C (±2 °C), the fluids flow rate was 2 mL min−1. A metal-free carbon catalytic cathode was chosen to prevent hydrogen evolution reaction (HER). The process achieved an increase in the pH from 2.6 to >4 and a significant reduction in total acid number (TAN). There were no signs of coke formation, improving carbon efficiency and catalyst lifetime. Although membrane degradation was noticed.40
Carbon loss during the ECH process was reported by Zhenglong et al., as only 80% of the carbon was retained in the cathode chamber. 6.0% was found in the anode chamber, due to migration through the membrane, 2.6% was adsorbed by the cathode and 0.2% was trapped in the downstream water trapping system. The system's total carbon loss was 11.2%. The authors suggest that the rest could be inside the membrane, as it turned black after the process.72 This exemplifies how different membranes could lead to a better result, as the carbon loss could possibly be avoided. Also, the process cost is increased by constantly changing the system's components. Zhang et al. showed that a cheaper anode, the Fe(III)/Fe(II) redox pair, can be utilized as a substitute for the commonly used noble metal (e.g. Pt) as a way to make the overall process cheaper and more competitive.
The work of Zhou et al. demonstrates how different catalyst cathodes can interfere with the final product, as the difference in FE and conversion rates were quite significant. The authors also proposed that the guaiacol upgrading was favoured by a chemisorption selectiveness to this compound. This indicates that different feedstocks need different catalytic cathodes to be more effective. Gardew et al. exemplified how substrate concentration can affect the FE and selectivity of the product, although it is not a direct relationship. Coke formation was not a major problem, although Zhenglong et al. reported finding decanted materials on the cathode chamber.
To avoid the problems related to the commonly used H-type cell, Lister et al. proposed a carbon catalyst cathode and DEM as a cheap and effective way to achieve upgrading. Results were promising, and the product had a higher economic value, as it could be used as commodity chemicals, but the authors found that further upgrading would be necessary for utilization as biofuel.
One area where a process of generating feedstock could be helpful includes plastic recycling, where the waste product can be returned to oil and then upgraded. Taking into consideration the highly variable nature of real plastic waste, Kusenberg et al. analysed municipal plastic waste and concluded that, to achieve industrial standards, the nitrogen concentration must be reduced by 87.7%, and oxygen by 98.0% (from 825 to 100 and 5200 to 100 ppmw respectively).73
Therefore, the focus of the upgrading process is the oxygenated compounds, that is, reducing the O:C ratio of the oil. Oxygenated compounds can lead to lower stability and heating values, as well as a more corrosive product, due to carboxylic acids. Most of the oxygen content comes from PET but to accomplish an effective plastic waste recycling process, PET cannot be excluded. Benzoic acid and its derivatives are the most common carboxylic acids in plastic-oil, and therefore should be the focus of the ECH process.
Although carboxylic acids are resistant to hydrogenation, ruthenium shows high activity for this process.75 As shown in the previous section, Ru was also the chosen metal as the cathode for Zhenglong et al. (Ru/ACC), Zhang et al. (Ru nanoparticles on mesoporous carbon) and Gardew et al. (Ru/ACC), which proves its effectiveness in hydrogenating WSBO and phenolic compounds, setting it as a prime candidate for the hydrogenation of whole plastic-oil. The possible mechanism for the hydrogenation of carboxylic acids is shown in Fig. 16.
Fig. 16 Mechanism for the hydrogenation of carboxylic acids. Edited with permission from ref. 75. Copyright 2020 Wiley. |
The hydrogenation of toluene into methylcyclohexane was carried out in a proton exchange membrane (PEM) system by Fukazawa et al. and found that PtRu/C was a better catalyst than individual metals such as Pt and Ru and presented high FE (>90%).53 In a later study, the same group reported on the ECH of benzoic acids into cyclohexane acids (PtRu/C catalyst) and presented high FE (99%) at a current density of 1.5 mA cm−2. The authors also reported that in less mild conditions, the carboxyl group could also be hydrogenated (see Fig. 12). As mentioned in previous sections, Du et al. also reported on the activity of different cathodes and found that Pt3Ru/ACC led to a higher conversion rate than Pt and Ru separately. The authors suggested that this could be the result of a synergistic effect between the two metals, combining platinum's excellent hydrogenation potency and the predominant direct hydrogenolysis ability of ruthenium.66 In addition, Salakhum et al. suggested that combining both metals can facilitate the cleavage of C–O and O–H bonds, as it reduces their heterolytic activation energies.76 This indicates that PtRu/ACC is a prime candidate for the ECH of carboxylic acids, and therefore, a possible choice of cathode for the hydrogenation of plastic-oil.52
Possible solutions consist in either increasing oil conductivity via additives (e.g., tetrabutylammonium hexafluorophosphate (NBu4PF6); LiCl; ammonium carbonate) or using a solvent that is capable of dissolving both polar and nonpolar compounds (e.g., ionic liquids, DES (Deep Eutectic Solvents); biomass-derived organic solvents). Studies are necessary to understand the changes in the physicochemical properties of the oil and its behavior during ECH that an additive could bring. LiCl could increase the corrosivity of the product as Li+ and Cl− are prone to corrosion. Ammonium carbonate could lead to membrane clogging via carbonate deposition, as well as increase in the nitrogen composition of the oil, and therefore, its corrosivity. NBu4PF6 is a chemically inert powder that is commercially available and could provide the much-needed conductivity for the system.
Fig. 17 2-MeTHF production via renewable sources. Edited with permission from ref. 80. Copyright 2012 ChemSusChem. |
Fig. 18 Total biofuel production in terawatt-hours (TW h) per year. Biofuel production includes both bioethanol and biodiesel. Chart constructed by Our world in data.83 |
• Noble metals are reactive species for the ECH of bio-oils. Rh, Pt and Ru loaded carbon cloth have all been reported to hydrogenate alcohols, carboxylic acids, olefins, and aromatic compounds, and presented high efficiency and FE.
• Current, potential, pH, and temperature are all essential factors in the ECH process. The fine tuning of these variables is necessary and directly linked to the quality of the final product.
• Conductivity of the oil presents a major challenge for ECH, alternative electrolytes and additives might be required to mke the process viable.
• The hydrogenation of plastic-oils is a possibility that still needs testing, although, as this technology matures, it may become one alternative for both the production of green plastics and biofuels.
This journal is © The Royal Society of Chemistry 2024 |