Platinum Metals Rev., 1998, 42, (3), 90
Ruthenium-Mediated Electrochemical Destruction of Organic Wastes
No industrial process is 100 per cent efficient and the generation of waste, both organic and inorganic, is an unavoidable reality. The recycling of waste is becoming increasingly important as concerns about the environment and the availability of resources come more to the fore, but there remain many waste streams which are not yet suitable for recovery of their reusable content, on the grounds of cost or practicality. Historically, the disposal of such wastes has been either via incineration or release into the environment after neutralisation or immobilisation, in a landfill site or into a drain or sewer, or the like. These options are becoming more restricted as regulations tighten and public perception of any but the most benign discharges worsens. This paper sets out reasons for using electrochemical oxidation with a ruthenium electrocatalyst to provide an effective and environmentally friendly alternative to other technologies. Electrochemical oxidation can effect mineralisation of toxic organic species with minimal generation of secondary waste and efficient recovery of the ruthenium mediator and is particularly suited for the treatment of highly chlorinated and aromatic compounds.
Organic wastes, of the types which cannot be converted into benign biodegradable species, are often treated by oxidation, usually incineration, although other chemical treatments are sometimes used. Many of the oxidants which could be used are ruled out by cost or toxicity.
The use of air for such oxidations, normally by high temperature incineration, is the cheapest in terms of reagents. The current best-available technology is capable of achieving essentially complete destruction of even refractory materials. However, the combustion products from many organic materials, in particular chlorinated species, contain acidic gases (for example, hydrogen chloride, sulphur oxides and nitrogen oxides) which must be scrubbed to very low levels before discharge. As excess air is always used, the volumes of gases for treatment tend to be large, leading to large scrubbers and other related equipment, and associated increased costs. Incomplete combustion, possibly due to maloperation, can lead to the production of secondary species, such as dioxins and dibenzofurans which are highly toxic and limited, by Regulatory bodies, to low ppb levels in the discharges. Toxic waste incineration has thus tended to become very ‘high-tech’ in recent years, with only a few highly specialised facilities treating most of the waste destined for combustion. Public perception of incinerators, particularly those treating toxic wastes, is poor or hostile, and the licensing of new facilities in most countries in the developed world is becoming very difficult and sometimes almost impossible.
However, the use of electrochemical oxidation to treat organic industrial waste gives a cheaper, cleaner and renewable alternative: the ‘electron’. The ‘electron’ is also cheaper than chemical oxidants on a per/mole basis, see Table I.
|Oxidant||Cost/mole of [O] equivalent, £|
|Potassium dichromate (CrVI)||0.18|
|Ceric nitrate (CeIV)||8.68|
In principle, electrochemical oxidation has a number of advantages over oxidation using conventional oxidising agents. In practice, however, direct electrochemical oxidation of organic species is often difficult due to mass transfer and kinetic restraints at the electrode surface and competing reactions with the electrolyte. Many of the restrictions can be overcome by the use of a mediator which is easily oxidised at the anode surface to a species which then reacts in the bulk solution with the organic substrate. The mediator can be re-oxidised at the anode for further use and can thus be termed an electrocatalyst as it is not consumed in the overall reaction.
The silver-mediated electrochemical oxidation of organics (SILVER II™) has been under development by AEA Technology for some time as a possible process for the complete oxidation of a number of types of organic waste or the organic waste component in a mixed waste. However, in the silver-mediated system, the presence of significant amounts of halogen (usually chlorine) in the organics being oxidised can result in the formation of substantial amounts of insoluble silver halide(s), which are, at best, a nuisance.
Here an alternative, ruthenium-mediated electrochemical system is described, see Figure 1, which has many similarities to the SILVER II™ system but which is not affected by the presence of halogen, and in particular by chlorine.
In order to understand the internal workings of the ruthenium-mediated oxidation system, it is necessary to consider some aspects of the aqueous chemistry of ruthenium. Ruthenium can form compounds with a multiplicity of oxidation states, ranging from −2 to +8. Its chemistry in aqueous solution mainly relates to the +3, +4, +6, +7 and +8 oxidation states, with the +3 and +4 states being the most stable. Ruthenium is normally supplied as the trichloride, “RuCl3.(H2O)n” (n = 2 or 3), a very soluble dark red-brown solid which contains a complex mixture of chloro and chlorohydroxo complexes, many of which are polymeric and contain mainly Ru(IV) (1).
Although “ruthenium(III) trichloride” is the form in which ruthenium is initially added to the ruthenium-mediated electrolytic system, Ru(IV), in the form of hydrated RuO2 is of much greater significance. The great insolubility of this oxide in water ensures that its production is favoured when ruthenium oxo-anions in high valent states are reduced (in a similar manner to the easy formation of insoluble MnO2 when MnO4- is used as an oxidant).
The hydrated oxide can be dissolved oxidatively (for example, using oxygen, persulfate or permanganate) in alkaline solution (pH > 12) to yield a solution of the orange ruthenate ion, RuO42- (Ru(VI)):
Ruthenate is only stable at high pH and in the absence of reducing agents. If the pH is lowered, RuO42- disproportionates to give hydrated Ru(IV)O2 and green perruthenate, Ru(VII)O4-:
Perruthenate solutions are unstable and decompose, either with the formation of O2 and RuO2:
or, if the pH is below ∼ 7.5, the decomposition can produce RuO4:
Important points to note from the above are:
(i) The favoured reduced form of uncomplexed ruthenium in near-neutral aqueous solution is Ru(IV)O2.
(ii) Uncomplexed ruthenium exists in aqueous solution as neutral or anionic species.
(iii) Partial oxidation of Ru(IV) in neutral or acid solution, followed by disproportionation, forms ruthenium tetroxide.
Volatile RuO4 (b.p. ∼ 130°C), like isostructural OsO4, is a powerful and unstable oxidising agent. It may be prepared from an acidic solution or suspension of any one of a number of ruthenium salts or hydrated RuO2 by treatment with a powerful oxidant, such as MnO4-, Cl2, IO4-, Ce4+ or NaOCl (bleach). Pure RuO4 is not particularly stable in either the liquid or vapour phase and tends to decompose, eventually forming RuO2. It is freely soluble in organic solvents, although its reactivity limits the solvents which can be used (typically CCl4 or CHCl3, plus some polar solvents with little nucleophilic character). These solutions of RuO4 are reasonably stable if kept in the dark and with a slight excess of the oxidant which formed the tetroxide. It is not normally prepared in bulk quantities, but is generated for immediate use.
Organic Oxidation by Ruthenium Oxo-Species
Ruthenate, perruthenate and, in particular, ruthenium tetroxide have all been studied as oxidants for organic synthesis. RuO4 has been studied fairly comprehensively and preparative oxidations in both the aqueous and organic phases (mainly in unreactive solvents like CHCl3 and CCl4) have been described (2, 3), with the exact conditions being tailored to suit the particular synthesis.
One of the first investigations of organic oxidations using RuO4 included the following description using only 10 mg of tetroxide with some common solvents:
“Benzene – vigorous explosion; pyridine – no explosion, only flame; anhydrous ether – small explosion, followed by yellow flame” (4).
Of particular interest in the context of destructive, rather than synthetic, oxidation by RuO4 are the following properties:
The solubility of RuO4 in both polar and non-polar organic solvents, allows oxidation to take place within an immiscible organic phase rather than only at the aqueous/organic interface.
RuO4 cleaves many C=C bonds, usually yielding carbonyl compounds in the first instance.
RuO4 is able to oxidise saturated hydrocarbons under mild conditions (5).
Aromatic nuclei are attacked and are completely degraded.
Destruction of Organics by Mediated Oxidation
Direct electrochemical oxidation is in principle easily capable of oxidising any organic species to carbon dioxide, and other inorganic products, based on thermodynamic, electrode potential considerations alone. However, in practice this is usually ruled out by a combination of slow electrode kinetics, difficult mass transfer of substrate to the electrode and competition from side reactions, such as oxygen evolution, so that mediated oxidation must be used. A suitable electro-active mediator must:
be soluble in the electrolyte to an extent which minimises any restrictions on mass transfer to the anode;
have fast electrode kinetics to minimise activation overpotential at the anode;
react rapidly in the bulk solution with a wide range of organic substrates, forming a reduced species which is easy to re-oxidise at the anode;
form no insoluble or complex species which remove the mediator irreversibly from solution.
Based on electrode potential considerations alone, a number of possible electrochemical couples may be suitable, but most are ruled out by one or more of the above criteria. However, the simple oxidation of chloride can be used to regenerate RuO4 in a double -mediated, continuous system.
The electrolysis of sodium chloride, NaCl, solution to produce chlorine, Cl2, and sodium hydroxide, NaOH, is one of the best understood electrochemical processes. This electrolysis is carried out industrially in a variety of cell types and is the basis for a worldwide industry with a capacity of millions of tonnes/year (8).
The principal process at the anode surface is oxidation of chloride ions (with very minor oxidation of water to form oxygen):
At the low pH normally encountered in chlorine cells, the product is gaseous chlorine. As the pH is raised, the hydrolysis of the chlorine to form HOCl begins to predominate and at pH 5.5 HOCl is the sole product, from hydrolysis of the Cl2 formed at the anode:
The hypochlorite (HOCl or OCl-, depending on the pH of the solution) generated electrochemically can be used to drive mediated organic oxidations and replace the continuous additions of, for example, NaOCl solution. As a mediating couple, the chloride/hypochlorite system has a number of advantages:
NaCl is soluble up to ∼ 5M in water, so there is little mass transfer limitation on the anode reaction.
The production of chlorine at a suitable anode is a fast reaction and the overpotential, even at high current density, is only a few tens of mV.
The reaction of the chlorine either directly with Ru(IV) or with water to form HOCl is a fast reaction and the reduced product (Cl-) is easily re-oxidised.
The complexation or reaction of Cl- to form species which cannot be re-oxidised is unlikely.
The anodic production of chlorine in a NaCl anolyte containing a small amount of dissolved ruthenium or hydrated RuO2 thus produces highly reactive RuO4 as long as current is passed. The RuO4 in turn attacks organic material fed to the anolyte, ultimately oxidising it to carbon dioxide, carbon monoxide and water (plus sulfate, phosphate or nitrate from heteroatoms such as S, P and N, for example).
It is this continuous coupling of electric power with oxidation which makes the ruthenium oxidation system so useful for the treatment of organic wastes, in particular wastes with a high chlorine content, which are often toxic and difficult to dispose of. The integration of the ruthenium-mediated oxidation into a complete system is described next.
At the heart of the ruthenium-mediated oxidation process, see Figure 1, is an electrochemical cell, very similar to the type used for chlorine manufacture. The anode, where chloride is oxidised, has to favour the production of chlorine over oxygen and in practice this restricts the material (or the coating) to platinum or platinum/iridium (or RuO2, but not the hydrated form). Modern cell designs have a titanium anode which is coated with a few microns of the noble metal or oxide (9). Development trials of the ruthenium process have employed anodes consisting of either platinum mesh (lab-cell scale) or 2.5μm platinum coated on titanium (flow-cell scale).
As the anolyte and catholyte chemistries are very different, the cell is divided with a Nafion™ cation exchange membrane which permits the passage of cations to carry the cell current but prevents bulk mixing of the two solutions. Nafion™ is a sulfonated perfluoropolymer which combines extreme chemical resistance with selective permeability to cations and is widely used industrially in cells for chlorine production. The selective permeability has a further advantage in that it prevents migration of soluble ruthenium species (which tend to be anionic under operating conditions) from anolyte to catholyte.
A number of materials can be used for the cathode – stainless steel has been used during development trials, but nickel, or one of the proprietary types of chlor-alkali cathode, would be a better choice for a full-scale plant.
The following processes occur in the anolyte:
Cl2 production at the anode,
Cl2 hydrolysis to HOCl,
RuO4 production (with reduction of the HOCl to Cl-, which is available for further oxidation to Cl2 at the anode),
The reaction of the RuO4 with the organic wastes fed to the anolyte (with the formation of insoluble hydrated RuO2) forming CO and (mainly) CO2, plus inorganic anions from, for example, S and P,
RuO2 re-oxidation to RuO4 by further HOCl.
The overall process thus consists of two linked cycles, (a)-(c) and (c)-(e). The oxygen which ultimately appears as CO+CO2 is derived from water molecules in the anolyte, via the formation and reduction of HOCl and RuO4. The overall anode/anolyte reaction thus becomes:
with the electrons passing to the external cell circuit.
Current through the cell is carried by Na+ ions which cross the membrane to the catholyte, usually in association with 1 to 3 molecules of water per Na+ ion. The Na+ ions do not react at the cathode, but merely transport charge to the catholyte (NaOH solution) while the cathode reaction is hydrogen evolution:
The transfer of Na+ ions and water to the catholyte and the formation of hydroxyl ions produces NaOH solution, exactly as in a chloralkali cell. The catholyte thus becomes more alkaline while the production of hydrogen ions in the anolyte causes a shift to a more acid pH. These changes are controlled by returning NaOH solution from the catholyte to the anolyte at the same rate as Na+ transfers from the anolyte. The overall reaction across the cell then becomes:
The process is thus a net consumer of water, rather than a producer as is the case with most organic oxidations, and the organic hydrogen fed to the anolyte appears from the cathode as hydrogen, rather than as water. A water feed is thus needed to replace that lost as H2 and CO+ CO2 or, alternatively, dilute organic streams can be treated with equal ease. Combustion of the hydrogen can be used to recover about 30 per cent of the power fed to the cell as useful heat, if this is deemed cost-effective and/or necessary.
If heteroatoms, such as Cl, S or P, are present in the organic feed additional NaOH has to be fed to the system (usually to the recycle stream from the catholyte) to balance the additional formation of H+ in the anolyte, for instance:
The net product from organic heteroatoms is the appropriate sodium salt (sulfate or phosphate, for instance). Note that organic chlorine produces chloride, which is one of the process reagents. When heteroatoms are present the overall reaction becomes:
Electrolysis of chloride-containing solutions, like the ruthenium process anolyte, can lead to chlorine evolution if the pH is allowed to fall. Also, RuO4 is volatile and could be flushed from the anolyte along with the CO2 and CO formed from organic oxidation. In order to prevent either species from escaping, some of the recycle stream of NaOH is used to maintain alkaline conditions in a scrubber through which the off-gases from the anolyte pass. Any Cl2 is absorbed as NaOCl while the RuO4 is absorbed as “hyper-rudienic acid” (H2RuO5) or its sodium salt. In this manner, the oxidising power of both the Cl2 and the RuO4 is returned to the anolyte and the escape of valuable ruthenium is prevented.
The process is typically operated with an electrolyte temperature of 50 to 80°C. The anolyte consists of a NaCl solution held near neutral pH and usually buffered to limit pH changes. The catholyte consists of NaOH solution, typically 35 wt.% NaOH or less (the limit being set by the properties of the membrane – the process will operate chemically at up to 50 wt.%). Conditions in the anolyte are highly aggressive to materials like stainless steel, so titanium or polymers, such as PTFE or PVDF, must be used there. The catholyte is much less aggressive and cheaper materials, such as polypropylene can be used.
The ruthenium-mediated system is surprisingly reactive compared to literature reports of organic oxidations using RuO4. This may be because the oxidation is carried out in the aqueous phase rather than in an inert solvent, and thus the formation of reactive radicals from water, for example, OH., is possible.
In Table II is shown some of the compounds which have been successfully oxidised during development trials, together with the composition of the gases leaving the anolyte. The percentage of CO+CO2 in the anolyte offgas gives a good indication of the electrochemical efficiency (that is the percentage of the current flowing which is driving useful oxidation, rather than oxygen evolution, at the anode).
|Substrate||Temperature °C||Carbon dioxide (2), maximum per cent||Carbon monoxide (2), maximum per cent|
|(2-Chloroethyl)ethyl sulphide (1)||50||73-95||2.6-6.7|
|PCBs (Aroclor 54)||90||67.6||1.2|
The amount of current which has to flow per mole of organic material oxidised is very dependent on the chemical nature of the organic material and the composition of the offgases. For instance, in the slightly extreme case of tetrachloroethylene, C2Cl4, oxidation to CO2 requires 4 faradays/mole:
while production of CO is actually hydrolysis rather than oxidation:
Oxidation to CO always requires less current to flow than oxidation to CO2, with a consequent saving in the power consumed by the cell.
The ruthenium process is thus able to oxidise highly chlorinated substrates, such as arise as waste from pesticide and herbicide manufacture, organophosphates (insecticides) and aromatic nitro-compounds (explosives manufacture) and can do so when these are in association with chloride ion or other salts in an aqueous stream. The ruthenium process is therefore seen as complementary to the related SILVER II™ electrochemical oxidation process in terms of feeds.
Economic and Practical Aspects
Like most electrochemical processes, the economics of ruthenium-mediated oxidation are closely tied to the cost of the electric power supplied to the cell. Electrochemical cells and DC power supplies are also expensive items of equipment and cells suffer from a relatively poor economy of scale compared to many process plant items (doubling the plant throughput requires twice as many cathode-anode pairs, while pipework and tankage would only have to be increased by a factor of ∼ 20,6). The treatment of bulk quantities of relatively benign organic wastes is therefore not likely to be very economic unless driven by environmental and regulatory pressures which eliminates incineration, for instance.
When organic wastes arise in conjunction with salts and water, as is the case with many industrial processes, particularly the manufacture of pharmaceuticals, dyestuffs and explosives, ruthenium-mediated oxidation becomes a much more attractive prospect. The incineration of salty wastes is often not very practical and the toxicity/salinity of the waste may rule out more ‘aqueous’ treatments such as reed beds and bio-degradation. The dilution of the organic in the water is not a particularly limiting factor.
The ‘ideal’ feed for the process shown in Figure 1 would consist of a mixture containing the correct amount of water to replace that consumed to form the CO+CO, and H2. However, more dilute feeds can be treated on a batch recycle basis, and the waste stream itself functions as the anolyte with additions of ruthenium (and possibly NaCl if it is not already present) being made, see Figure 2. If Na+ ions are already present in the waste liquor, a ‘product’ of catholyte NaOH solution can be withdrawn for recycle/reuse, if desired.
The ruthenium is only required at low concentrations as its role is principally catalytic, but its cost probably precludes its disposal, albeit as insoluble and inert RuO2, with the treated waste. However, the low solubility of the RuO2, which is the form the ruthenium takes when the cell is switched off and the electrolyte is allowed to stand, means that simple filtration can be used for recovery.
Alternatively, the ruthenium can be swept from the anolyte by allowing the pH to fall so that some chlorine is evolved. This (possibly assisted by an air sparge) sweeps the ruthenium from the anolyte as RuO4 which is then absorbed into NaOH (formed simultaneously at the cathode) using the anolyte offgas scrubber. In this way a fresh batch of anolyte or reagent is effectively reformed, and can be used for dosing the next batch of waste liquor requiring treatment, see Figure 3.
The incoming waste liquor may contain metal ions which would precipitate in the membrane and these could therefore be usefully removed from the waste before the oxidative step, for separation and disposal. The continuous production of NaOH at the cell cathode provides a convenient source of caustic with which many metals can be precipitated as their insoluble hydroxides. An alternative use for the catholyte NaOH is that refractory materials (such as highly chlorinated, hydrophobic solids, for example hexachlorobenzene) may be partially hydrolysed using the catholyte recycle stream (10). Figure 4 shows how the additional NaOH that must be added to balance the halogen content of the organics, may be added to the recycle stream to give the highest practical concentration for the hydrolysis.
Double mediated electrochemical oxidation using the Cl(-I)/Cl(I) and Ru(IV)/Ru(VIII) couples in a cell very similar to a chlor-alkali cell has been shown to have promise for the treatment of organic waste streams, particularly halogenated or dilute, salty liquors. Currently, studies are concentrating on optimising the core process and examining the application of the system to a variety of ‘real’ wastes with an industrial origin.
- 1F. A. Cotton and G. Wilkinson, “ Advanced Inorganic Chemistry ”, Wiley Interscience, New York, 1968
- 2J. L. Courtney and K. F. Swansborough, Rev. Pure Appl. Chem ., 1972, 22, 47
- 3D. G. Lee, and M. van den Engh, “ The Oxidation of Organic Compounds by Ruthenium Tetroxide ” in “ Oxidation in Organic Chemistry ”, ed. W. S. Trahanovsky, 1973, Academic Press, New York
- 4C. Djerassi and R. R. Engle, J. Am. Chem. Soc ., 1953, 75, 3838
- 5U. A. Spitzer and D. G. Lee, J. Org. Chem ., 1975, 40, 2539
- 6D. C. Ayres, Nature, 1981, 290, 323
- 7C. S. Creaser,, A. R. Fernandes and D. C. Ayres, Chem. lnd ., 1988, 499
- 8A. M. Couper,, W. N. Brooks and D. A. Denton, “ Modern Chlor-alkali Technology ”, 1989, 4, 71
- 9P. C. S. Hayfield, Platinum Metals Rev ., 1998, 42, ( 2 ), 46
- 10J. K. Beattie,, R. W. Kaziro and P. A. Lay, World Appl . 89/05172
In a cell (193000Cb, 54 Ah) in which oxidation occurs, with cell voltage of 3V, 2 moles of electrons carry out the oxidation process, equivalent to one mole of [O] and costs ∼ 1p. The same amount of oxidation delivered using hydrogen peroxide (one of the cheapest chemical oxidants) at a nominal cost of £600/te for a 35 wt. % solution (£ 1700/te H2O2, 29411 moles [O]) costs ∼ 5.8 p. High-valent metal-based oxidants such as CrVI and CeIV are much more expensive. Bulk K2Cr2O7 costs £1835/te (3400 moles, 10200 moles [O]) while CeIV nitrate ∼ £10000/te (2300 moles CeIV, equivalent to 1150 moles [O]) and the equivalent oxidation, carried out using this reagent costs £8.68 (all costs assume 100 per cent efficiency for simplicity). Like peroxide, the electron (the oxidant itself) leaves no residues which are toxic or difficult to dispose of, while oxidants like CrVI and CeIV leave large quantities of spent oxidant for disposal.