Engineering aspects of electrochemical hydrodehalogenation of 2,4-dichlorophenol in a solid polymer electrolyte reactor

Applied Catalysis A: General 261 (2004) 1–6

Engineering aspects of electrochemical hydrodehalogenation of 2,4-dichlorophenol in a solid polymer electrolyte reactor H. Cheng, K. Scott∗ , P.A. Christensen School of Chemical Engineering & Advanced Materials, University of Newcastle upon Tyne, NE1 7RU Newcastle, UK Received 12 August 2003; received in revised form 20 October 2003; accepted 21 October 2003

Abstract A solid polymer electrolyte reactor is designed for the electrochemical hydrodehalogenation (HDH) of 2,4-dichlorophenol (DCP) in paraffin oil. The reactor has been evaluated at 0.05, 0.1, 0.2, 1 and 2 dm3 scales in terms of cathode materials and structure and ratio of the cathode surface area to the solution volume. A cathode of titanium mini mesh with a palladium electrocatalyst produced by electrodeposition was particularly effective. Current efficiencies of up to 80%, percentage of DCP removal of up to 50%, space-time yields of up to 5.0 kg DCP m−3 h−1 and energy consumption as low as 1.5 kW h (kg DCP)−1 , were achieved. The reactor showed stable operation for periods of up to 170 h. © 2003 Elsevier B.V. All rights reserved. Keywords: Electrochemical hydrodehalogenation (HDH); Solid polymer electrolyte reactor; Ti mini mesh; Carbon cloth; 2,4-Dichlorophenol (DCP); Paraffin oil

1. Introduction Halogenated organics are a major source of pollutants, which can adversely affect the environment. Treatment of such organics presents technical challenges and can involve high cost. Approaches based on bioremediation and incineration have limitations in terms of effectiveness and cost and thus alternative approaches are desirable. An attractive approach is to dehalogenate the organics and then look to use a more effective bioremediation. One method of dehalogenation involves electrochemical reduction in an hydrodehalogenation (HDH) process [1–5]. In principle, electrochemical technology is environmentally compatible because the main reagent, the electron, is a ‘clean reagent’. In an electrochemical HDH process, toxic materials are removed from gases and liquids using the electron as a main reagent rather than using redox chemicals. A typical example is the HDH of 1,2,3,5-tetrachlorobenzene (TCB) and chlorobenzene (CB) in methanol or dimethylsulphoxide and acetonitrile (with 0.25 M tetraethyl ammonium bromide) at a cathode potential of −3.3 V versus Ag/AgCl [2]. Greater than 95% conversion of CB at an ∗ Corresponding author. Tel.: +44-191-222-5207; fax: +44-191-222-5292. E-mail address: [email protected] (K. Scott).

0926-860X/$ – see front matter © 2003 Elsevier B.V. All rights reserved. doi:10.1016/j.apcata.2003.10.021

initial concentration of 12 mM was achieved, with a current efficiency of 15–20% using carbon cloth or Pb cathodes. On the other hand, Pt, Ti, and Ni cathodes gave low current efficiencies of ca. 5% at lower conversions. The feasibility of electrochemical hydrodehalogenation (HDH) in aqueous solutions has been demonstrated [6–9]. The key technology involved in the electrochemical HDH process is to use a solid polymer electrolyte reactor. The anolyte and catholyte chambers were separated by an ion exchange membrane, and Titanium mini mesh-based porous electrodes were attached to each face of the membrane, forming a membrane-electrode assembly (MEA), similar to that employed in conventional solid polymer electrolyte fuel cells (Fig. 1a). Electrochemical reactions occur at the interfaces between the ion exchange membrane and electrochemically active layers of the electrodes. The membrane employed in the reactor was Nafion® , and protons (H+ ions) generated at the anode from the aqueous anolyte migrate through the membrane under the influence of the applied electric field, and are then involved in the hydrogenation reaction or reduced to atomic and molecular hydrogen at the cathode. The membrane serves both as reactor separator and conductive electrolyte, providing the ionic conductivity required without additional supporting electrolytes. Most systems with halogenated organic compounds are characterised by low or non-conductive media and low reactant

2

H. Cheng et al. / Applied Catalysis A: General 261 (2004) 1–6

scale-up the electrochemical HDH process to a commercial process. The paper concentrates on several special aspects of the reactor, including cathode material, cathode structure and ratio of the cathode surface area to the waste volume etc. The design has been evaluated during the HDH of DCP: Cl2 C6 H3 OH (DCP) + 2e− + H+ → ClC6 H4 OH (CP) + Cl−

(1)

ClC6 H4 OH (CP) + 2e− + H+ → C6 H5 OH (Phenol) + Cl−

(2)

2. Experimental 2.1. Materials and chemicals

1

2

3

4

The following materials and chemicals were used as received: Ti mini mesh (99.6%, open area 37%, wire diameter 0.2 mm, Goodfellow), carbon cloth (GC-14, E-Tek Inc.), 2,4-dichlorophenol (DCP, 99%, Aldrich), 2- or 4-chlorophenol (CP, 99%, Lancaster Synthesis), phenol (99.9%, Aldrich), PdCl2 (99%, Aldrich), H2 PtCl6 (99%, Johnson Matthey) and H2 SO4 (98%, AnalaR, BDH). All oil solutions were prepared using light paraffin oil (Aldrich) and all aqueous solutions were prepared using a Millipore-Q water (18.2 M cm). 2.2. Solid polymer electrolyte reactor rig

5

6

7

8

Fig. 1. (a) Solid polymer electrolyte reactor. 1, Catholyte output; 2, reactor end plates (stainless steel); 3, screw for current collector; 4, stainless steel mesh turbulence promoter; 5, PTFE gasket; 6, Tiron® sealing ring; 7, anolyte output; 8, catholyte inlet; 9, cathode; 10, Nafion® 117 membrane; 11, anode; 12, anolyte inlet. The cell dimension is 22 cm × 14 cm × 3 cm. (b) Flow circuit of the solid polymer electrolyte reactor rig. 1, Anolyte reservoir; 2, condenser; 3, power supply; 4, catholyte (halogenated organic compounds+paraffin oil) reservoir; 5, pump; 6, valve; 7, the solid polymer electrolyte reactor; 8, flow meter.

concentrations [6–9]. Thus, catholytes consisted of paraffin oil and 2,4-dichlorophenol (DCP) with different concentrations were chosen as model systems. For a DCP-paraffin oil system, the conductivity provided by the membrane was insufficient to allow both the anolyte and catholyte streams to be conducting enough for the HDH process. This was overcome by employing an aqueous acidic solution as anolyte, which was in no way problematic to the operation or application of the technology. To achieve a reasonable HDH rate in such systems, a high and uniform rate of mass transport and proper operation conditions are required and, obviously, it is necessary to address engineering aspects of an HDH process. This work focuses on an effective reactor design based on a solid polymer electrolyte reactor in order to enable

A compact auxiliary system is designed to constitute a reactor rig, as shown in Fig. 1b. The designed solid polymer electrolyte reactor is inserted into a circulation loop consisting of anolyte and catholyte peristaltic pumps (Cole-Parmer) and reservoirs (0.1, 1 or 2 dm3 ) placed in two heating mantles (Electrothermal® Flask/Funnel, Cole-Parmer). The rig is operated in a batch recirculation mode. A glass condenser was mounted on top of each reservoir to condense organic vapour from the gases exiting the reservoir before passing through a concentrated alkaline aqueous solution. Non-condensible gases were vented to the atmosphere. In operation, catholyte and anolyte, each with a volume of 0.05, 0.1, 0.2, 1 and 2 dm3 , were pumped through the cell and then returned to the reservoirs for recycle by the pumps. Flow rate, electrolyte temperature and applied currents were controlled using the control unit. The solid polymer electrolyte reactor has already been shown schematically in Fig. 1a, which consisted of an electrode membrane (Nafion® 117) assembly placed between stainless steel blocks with machined flow channels. The principle of the reactor was described in the Introduction. 2.3. Electrodes Palladised cathodes (2 mg Pd cm−2 , 20 cm2 in the geometric area) and platinised anodes (2 mg Pt cm−2 , 20 cm2

H. Cheng et al. / Applied Catalysis A: General 261 (2004) 1–6

in the geometric area) were prepared by electrodeposition. The substrate for both electrodes was titanium mini meshes. During the deposition, the Ti mini mesh was first abraded with emery paper and rinsed thoroughly with water. After drying, the Ti mini mesh was rinsed in acetone. Following etching with boiling 37% HCl solution for 1 min, the mesh was put into the deposition cell in which a N2 -saturated deposition solution (0.02 M PdCl2 or H2 PtCl6 aqueous solution) was filled and stirred magnetically. The catalyst was electrodeposited onto the substrate at a controlled potential, which was chosen according to the linear sweep voltammograms. The amount of charge required to deposit the catalyst was monitored through a computer-controlled potentiostat. A number of electrodes were produced and tested under identical conditions to check reproducibility. More details regarding the deposition, including pre-treatment of substrates and after-treatment of electrodes, are detailed elsewhere [9]. Change in ratio of electrode surface area to waste volume (RAV) was achieved using varying waste volumes in a reactor with fixed electrode surface area. 2.4. Batch electrochemical HDH Batch electrolyses were performed in the solid polymer electrolyte reactor using a FARNELL LS60-5 power supply. All electrolyses in the solid polymer electrolyte reactor were carried out at constant current density, ranging from 5 to 100 mA cm−2 , for periods between 2 and 170 h. The longer-term testing was performed to investigate the durability of the reactor components. There were several short shutdown periods during the long-term operation to ensure continued operation of the experiments by changing the anolyte feeds. The concentrations of DCP, intermediates and phenol during the electrolysis were monitored using HPLC.

3

2.6. Parameter definitions Percentage of DCP removal (PR), space time yield (STY), current efficiency (φ) and energy consumption (ECN) were used to evaluate the process performance and efficiency. The percentage of DCP removal (PR) is expressed as: PR =

C0 − Ct × 100% C0

(3)

where C0 and Ct are DCP or DBP concentrations at start and at electrolysis time, t, respectively. The HDH capacity is expressed, in terms of the space time yield (kg m−3 h−1 ), using the following formula [10]: STY =

3600ajφMFW nF

(4)

where a is a specific area (m−1 ), defined as a ratio of the electrode area to the waste volume in the reactor, j is the current density (A m−2 ), φ is the current efficiency, n is the number of electrons in the concerned reaction, F is the Faraday constant (96 500 C mol−1 ) and MFW is the molar mass (kg mol−1 ). The current efficiency was calculated as that part of current (or charge) passed to convert the starting DCP to phenol and 2- or 4-CP. Total current efficiency is the sum of the individual current efficiencies of forming phenol and 2- or 4-CP according to Eqs. (1) and (2). Energy consumption for HDH processes was calculated according to the following equation [10]: ECN =

nFECell φMFW

(5)

where ECell is the cell voltage. The total energy consumption included the energies used for the HDH processes, heaters and pumps, etc.

2.5. Product analysis High-performance liquid chromatography (HPLC) was performed in a DIONEX HPLC system, which consisted of a P 580 Pump and a Softron 2000 UVD 170S/340S UV-Vis detector with an Econosphere C8 column (5 ␮m particle size and 25 cm × 0.46 cm, Alltech Associates, Inc.). The wavelengths used in HPLC measurements were determined using UV-Vis spectroscopy (UV-160A UV-Vis Recording Spectrophotometer, SHIMADZU, Japan). Normally, the UV detector was set to 270 nm for phenol and 290 nm for DCP and 2- or 4-CP. The mobile phase was an acetonitrile–water mixture (52:48 v/v) with a flow rate of 1.0 ml min−1 . The peaks for phenol (retention time, tr = 1.92 min), CP (tr = 2.05 ∼ 2.15, it was not possible to discriminate between 2and 4-CP under the analytical conditions) and DCP (tr = 2.96 min) were characterised by using standard paraffin oil solutions. Quantification of the product distribution during the electrolysis was accomplished by the use of calibration curves with the authentic samples. A sample volume of 20 ␮l was generally employed.

3. Results and discussion Palladised cathode has been demonstrated as an effective material for an electrochemical HDH process [6–9] and thus the evaluation of an HDH reactor was focused on palladised cathodes with different substrates and structures. 3.1. Effect of electrode substrate Two electrode substrates, Ti mini-mesh and carbon cloth, were examined in a solid polymer electrolyte reactor during the HDH of DCP in oil media. The catalyst loading of these electrodes was fixed at 2 mg Pd cm−2 . Typical results collected in the paraffin oil solutions are shown below. Fig. 2 shows the change in DCP and phenol concentration with time for HDH in paraffin oil using palladised electrodes. The DCP concentration decreased more rapidly and more phenol was formed at the Ti mini mesh than at the carbon

4

H. Cheng et al. / Applied Catalysis A: General 261 (2004) 1–6 200

Organic concentration/mM

160

120

80

40

0 0

30

60

90

120

Electrolysis time/min Fig. 2. Effect of cathode substrate on the product distribution during the electrochemical HDH of DCP in paraffin oil media using a Nafion® 117 membrane reactor. (䊉) DCP, Ti mini mesh. (䉱) DCP, carbon cloth. (䊏) Phenol, Ti mini-mesh. (䉬) Phenol, carbon cloth. Cathode: Pd/Ti mini mesh or carbon cloth (20 cm2 , 2 mg Pd cm−2 ). Anode: Pt/Ti mini mesh or carbon cloth (20 cm2 , 2 mg Pt cm−2 ). Controlled current density: 10 mA cm−2 . Catholyte: 200 mM DCP or DBP in paraffin oil (50 cm3 ). Anolyte: 0.5 M H2 SO4 aqueous solution (50 cm3 ). Flow rate: 200 ml min−1 . Temperature: 19.5 ± 0.5 ◦ C.

cloth. After 2 h HDH, the DCP concentration fell from 200 to 81.4 or 162.7 mM and the phenol concentrations were 80.2 or 35.5 mM at the Ti mini mesh or the carbon cloth cathodes, respectively. Cathode substrate affects HDH capacity, i.e. space-time yield, as shown in Fig. 3 collected during the HDH of 200 mM DCP in paraffin oil using a Nafion® 117 membrane reactor. A large difference in space-time yield was observed, e.g. 3.0–3.4 and 4.5–5.0 kg DCP m−3 h−1 for the HDH of

-3

Space-time yield/kg m h

-1

6

5

4

3 20

50

80

110

140

Electrolysis time/min Fig. 3. Effect of cathode substrate on the space-time yield during the electrochemical HDH of DCP in paraffin oil media using a Nafion® 117 membrane reactor. (䊉) Ti mini mesh. (䉱) carbon cloth. Conditions: same as those in Fig. 2.

DCP at the carbon cloth and the Ti mini-mesh cathodes, respectively. Fig. 4 shows the influence of cathode substrate on current efficiency and energy consumption obtained during the HDH of 200 mM DCP in paraffin oil media using a Nafion® 117 membrane reactor. Using a Ti mini mesh as the cathode substrate, the current efficiencies were above 74%, and are approximately 20% higher than observed with the carbon cloth substrate. The energy consumptions were 1.5–2.4 and 3.0–3.4 kW h (kg DCP)−1 using the Ti mini mesh and the carbon cloth, respectively. This is a consequence of the higher current efficiencies and also the lower resistance of the Ti mini mesh, compared with the carbon cloth. Better mass transport of DCP at the Ti mini mesh can be expected because of its very open structure and attributed to better HDH performance, compared with the carbon cloth. In addition, during the HDH of DCP using carbon cloth cathode, some black (Pd) particles were observed in the catholyte, suggesting that the palladised Ti mini mesh cathode was more stable than the palladised carbon cloth cathode. 3.2. Electrode structure (single- or three-layer electrodes) Preparation of high effective electrodes can be approached by several methods, e.g. increased catalyst loading and using a three-dimensional structure. In this work, multiple layer meshes were used in an attempt to enhance HDH rates and efficiency. The results were compared with those achieved at a single layer electrode. Fig. 5 shows data for the percentage of DCP removal and the current efficiency achieved during the HDH of 200 mM DCP in paraffin oil using the reactor with threeor single-layer electrodes. The use of the three-layer electrodes produced higher percentage of DCP removal than the single-layer electrodes, although the improvement was small, e.g. only 1.1% after 3 h HDH. Using the three-layer electrodes to replace the single-layer electrodes increased the current efficiency, i.e. 62–83 or 56–66% at the threeor single-layer cathode, respectively. The energy consumption decreased by approximately 0.3 kW h (kg DCP)−1 by using the three-layer electrodes rather than the single-layer electrodes. The decrease was mainly due to the increase in current efficiency because the cell voltages in the reactors with the three- or single-layer electrodes showed only small difference, i.e. less than 50 mV. The improvement achieved with the three-layer electrodes is relatively small in relation to the increase in total electrode area. A contributing factor to this will be the lower overall utilisation of the area due to a poor current distribution in a direction normal to the membrane surface. Bear in mind that the catalyst amount used at the three-layer electrodes is three times higher than that with the single-layer electrodes, so the improvement achieved with multiple-layer electrodes must be balanced against catalyst cost.

H. Cheng et al. / Applied Catalysis A: General 261 (2004) 1–6

5 3.5

73%

3

64%

2.5

55%

2

46% 20

50

80

Energy consumption/kW h kg

Current efficiency/%

-1

82%

1.5 140

110

Electrolysis time/min Fig. 4. Effect of cathode substrate on the current efficiency and the energy consumption during the electrochemical HDH of DCP in paraffin oil media using a Nafion® 117 membrane reactor. (䊉) Current efficiency, Ti mini mesh. (䉱) Current efficiency, carbon cloth. (䊏) Energy consumption, Ti mini mesh. (䉬) Energy consumption, carbon cloth. Conditions: same as those in Fig. 2.

85%

15%

80%

12%

75%

9%

70%

6%

65%

3%

60%

0% 0

30

60

90

120

150

44%

62%

33%

52%

22%

42%

11%

32%

Current efficiency

18%

Current efficiency

Percentage of DCP removal

The cathodes used in the HDH of DCP worked well for a short-term operation. It is necessary to know the stability of the cathodes in a long-term operation because a practical waste treatment process requires a continuous run. Pre-

liminary long-term HDH of DCP was carried out in paraffin oil at 1-l level and at a current density of 10 mA cm−2 . Fig. 6 shows the data in percentage of DCP removal and current efficiency obtained during the HDH of 200 mM DCP in the paraffin oil. The percentage of DCP removal increased rapidly first then slowed at the later stage. Since no obvious damage of the electrodes was observed during the HDH of DCP, the reduction in the percentage of DCP removal was

Percentage of DCP removal

3.3. Long term electrolysis

55% 180

Electrolysis time/min

0% 0

Fig. 5. Effect of cathode structure on the destruction percentage and the space-time yield during the electrochemical HDH of DCP in paraffin oil media using a Nafion® 117 membrane reactor. (䊉) Percentage of DCP removal, single layer. (䉱) Percentage of DCP removal, three layers. (䊏) Current efficiency, single layer. (䉬) Current efficiency, three layers. Cathode: single layer or three-layer Pd/Ti mini mesh (20 cm2 , 2 mg Pd cm−2 ). Anode: single layer or three-layer Pt/Ti mini mesh (20 cm2 , 2 mg Pt cm−2 ). Controlled current density: 10 mA cm−2 . Catholyte: 200 mM DCP in paraffin oil (100 cm3 ). Anolyte: 0.5 M H2 SO4 aqueous solution (100 cm3 ). Flow rate: 100 ml min−1 . Temperature: 20 ± 0.5 ◦ C.

30

60

90

120

150

22% 180

Electrolysis time/hr Fig. 6. Long-term evaluation on the percentage of DCP removal and current efficiency during the electrochemical HDH of 200 mM DCP in paraffin oil media using a Nafion® 117 membrane reactor. (䊉) Percentage of DCP removal. (䊏) Current efficiency. Cathode: Pd/Ti mini mesh (20 cm2 , 2 mg Pd cm−2 ). Anode: Pt/Ti mini mesh (20 cm2 , 2 mg Pt cm−2 ). Catholyte: 200 mM DCP in paraffin oil (1000 cm3 ). Anolyte: 0.5 M H2 SO4 aqueous solution (1000 cm3 ). Controlled current density: 10 mA cm−2 . Flow rate: 100 ml min−1 . Temperature: 18.5 ± 1.5 ◦ C.

6

H. Cheng et al. / Applied Catalysis A: General 261 (2004) 1–6

thus due to reduction in DCP concentration. The current efficiency fell correspondingly, i.e. from 60 to 25% after 170 h HDH. The data demonstrate that the cathodes can be subjected for a relatively longer HDH operation. To satisfy the requirements of a commercial HDH process, the electrode stability will need to be examined in much longer experiments, e.g. thousands of hours, which may require more effective procedures for preparation of electrodes.

4. Conclusion The HDH of DCP has been successfully carried out in a solid polymer electrolyte reactor with current efficiencies of up to 86%, percentage of DCP removal of up to 50%, space-time yields of up to 5.0 kg DCP m−3 h−1 and energy consumptions as low as 1.5 kW h (kg DCP)−1 in the concentrated DCP– paraffin oil solutions. The HDH rates were doubled with 20% higher current efficiencies and the energy savings of up to 50% using a Ti mini mesh rather than a carbon cloth as the cathode substrate. The engineering aspects of the reactor design has been evaluated at 0.05, 0.1, 0.2, 1 and 2 dm3 scales in terms of the reactor components, particularly cathode materials and cathode structure. The long-term HDH of DCP in oil media, up to 170 h, displayed reasonable HDH rates, acceptable efficiency and electrodes showed stability during operation.

Acknowledgements The authors thank the United Kingdom Engineering and Physical Sciences Research Council (EPSRC) for funding. The work was performed in research facilities provided through an EPSRC/HEFCE Joint Infrastructure Fund award (No. JIF4NESCEQ). References [1] F. Bonfatti, S. Ferro, F. Lavezzo, M. Malacarne, G. Lodi, A. De Battisti, J. Electrochem. Soc. 146 (1999) 2175. [2] S.M. Kulikov, V.P. Plekhanov, A.I. Tsyganov, C. Schlimm, E. Heitz, Electrochim. Acta 41 (1996) 527. [3] I.F. Cheng, Q. Fernando, N. Korte, Environ. Sci. Technol. 31 (1997) 1074. [4] S. Zhang, J.F. Rusling, Environ. Sci. Technol. 29 (1995) 1195. [5] D. Schmal, J. van Erkel, P.J. van Duin, Ichem. Symp. Ser. No. 98 (1986) 259. [6] K. Scott, H. Cheng, P. Christensen, in: E.W. Brooman, C.M. Doyle, C. Cominellis, J. Winnick (Eds.), Energy and Electrochemical Processes for a Cleaner Environment, PV 2001-23, San Francisco, California, USA, Fall 2001. p. 45. [7] H. Cheng, K. Scott, P. Christensen, J. Electrochem. Soc. 150 (2003) D17. [8] H. Cheng, K. Scott, P. Christensen, J. Electrochem. Soc. 150 (2003) D25. [9] H. Cheng, K. Scott, P. Christensen, Hydrodehalogenation of 2,4-dibromophenol by electrochemical reduction, J. Appl. Electrochem. 33 (2003) 893. [10] F. Goodridge, K. Scott, Electrochemical Process Engineering, Plenum Press, New York, 1995.