Treating organic waste

It is well-known that large volumes of liquid organic wastes can easily be incinerated. Even radioactive liquid organic wastes can be incinerated and this is the baseline treatment route throughout the UK SLC (Site Licence Company) waste management programmes. However, each incinerator has a permit from the relevant body (in the UK it is the Environment Agency) that limits the amount of radioactive waste that can be treated according to the type of radiation that it emits. The limits are usually less stringent for beta and gamma radiation compared with the limits for alpha [1]. Therefore the cost and timescales for the incineration of a tonne of alpha-emitting waste is far higher than the equivalent of beta- or gamma-emitting waste. Consequently, an organic containing alpha-emitting species will use up the permits of an incinerator across a long time period, meaning that the incinerator cannot accept any other alpha-emitting waste in the interim.

The Arvia™Organic Destruction Cell can minimise this "blocking up" of the incinerator, by facilitating the treatment of high alpha wastes within the permits of the SLC. If implemented holistically, this could free up the incinerators to treat larger volumes of other low level wastes, thus providing a complimentary waste treatment approach and ultimately minimising the amount of waste placed in interim storage and clearing up current waste inventories.

Originally developed for the water industry [2], the technology operates on the fundamental principles of adsorption and electrochemical oxidation. Adsorption is the adherence of molecules to a surface, in this case, organics onto a graphitic material. Electrochemical organic oxidation is the conversion of hydrocarbons into water and carbon dioxide, via electron transfer. These principles have been extensively investigated and engineered to create a stand-alone treatment unit for the effective and efficient oxidation of a range of different organics.

Arvia's founder and technical director, Nigel Brown, has an extensive history in waste water engineering consultancy. After inventing the technology, he was commissioned with a series of research funding grants to validate and mature its potential. Working at the University of Manchester with Edward Roberts of the School of Chemical Engineering and Analytical Sciences, (who would later become Arvia's research director), Brown designed and prototyped the first generation of the Arvia ODC. The University of Manchester Intellectual Property encouraged the incorporation of the university spin out, along with the filing of a number of intellectual property applications and registrations. Arvia won GBP 3.8 million in a Series A funding round in early 2012 from venture capital firms Parkwalk. MTI and Sustainable Technologies Investment Ltd.

Arvia is coming to the completion of the design phase of a full-scale plant for the treatment of 2000L of oil at the Magnox Limited Trawsfynydd SLC. This is a follow on from previous pilot scale, active trials with Magnox. Arvia is also working with the Department of Energy in the US to trial its technology for some of its legacy wastes. Arvia is also working with AREVA in France to identify opportunities to help it treat wastes that are not suitable for the French incinerators. Another key project is the trialling of a second generation Arvia technology for the treatment of plutonium contaminated organics located at Sellafield in the UK. This project is being delivered in collaboration with the National Nuclear Laboratory and is funded by the Nuclear Decommissioning Authority via a Technology Strategy Board funding call.

There are a number of radioactively-contaminated organics both across the UK Nuclear Decommissioning Agency estate and internationally, for example fluorinated oils (Fomblin), chelating agents (EDTA) and solvents (Odourless Kerosene). Each organic has different physical and chemical properties which influence its behaviour. Arvia has ascertained the optimal treatment parameters for a large range of these organics, allowing for a tailored service offering according to the specific waste problem.

The Process

The ODC is operable in a batch or a continuous manner [3] and the fundamental principles for each are similar. The first phase, adsorption, is the exposure of Arvia's proprietary adsorbent, Nyex™, to an aqueous solution of the organic waste. The organic components in the solution are adsorbed and therefore concentrated onto the adsorbent. During the second phase, separation, the adsorbent is allowed to settle under gravity to form a bed between two current feeders (or in larger units, an array of current feeders). The specific gravity of the adsorbent is fundamental to this step, because no additional forces are used for settlement and separation. Once a bed of adsorbent is formed, the third stage, regeneration, is initiated: an electric current is passed across the bed and the adsorbent acts as a three-dimensional electrode. Electron transfer at the adsorbent-organic surface brings about oxidation of the organic according to Equation 1:

The average power consumption per litre of neat, radioactive organic is 42.5 kWh (samples of that volume with low concentrations of oil will be lower). If the adsorbed organic contains radionuclides, the majority of radionuclides are transferred into the aqueous phase upon organic oxidation [4]. Thus the treatment process destroys a hazardous, untreatable waste stream, leaving clean, active water, which can be disposed of via conventional means (an active effluent treatment plant). The build-up of radionuclides upon treatment materials is monitored and controlled so that the unit itself does not become a problematic secondary waste. The adsorbent is regenerated in-situ when the organic is oxidised, so as soon as the regeneration phase is complete, the adsorbent can be re-fluidised to adsorb more organic. This regeneration process is a key differentiator compared to for example activated carbon regeneration, which usually occurs off-site via thermal means.

Properties of the adsorbent

A key influencing factor of organic behaviour is the adsorption capacity of the adsorbent for the organic, which is associated to the interaction of the properties of the organic with the properties of the adsorbent (Figure 1).

The adsorbent is a non-porous graphite-based material with a high electrical conductivity and a high specific gravity; each characteristic is independently important. It is of a flake consistency (Figure 2) and is composed of a series of layers of graphite sheets of decreasing basal-plane diameter. The basal plane itself is hydrophobic (repulsive to water), adopting an electronic structure similar to that of graphene but with a sparse distribution of functionalised molecules. This structure contributes to the conductivity of the adsorbent because there is a sufficient level of electron mobility across the plane. The edge plane is hydrophilic (attractive to water) due to the dense distribution of functionalised groups at its surface. The flake formation provides a large surface area of hydrophobic basal-plane and a small surface area of hydrophilic edge-plane. Accordingly, the adsorbent is primarily hydrophobic in nature, which is a key driver behind its interaction with organics.

High hydrophobicity

If an organic has a high hydrophobicity (that is, does not readily dissolve in water), the enthalpy of hydration of the organic is less favourable than the enthalpy of interaction with the adsorbent. This will lead to a high thermodynamic loading of the organic on the Nyex, attributed to multilayer adsorption. Multilayer adsorption is due to self-association of the organic with other organic molecules that have initially formed a monolayer on the adsorbent. The degree of this multilayer build-up is dependent upon the concentration of the organic in the solution. This must be kept at an optimum so as to provide a trade-off between obtaining a maximum organic loading on the adsorbent and preventing total water/organic separation due to self-association during the adsorption stage.

An interesting intricacy with multilayer adsorption is the implication for bulk density. For the adsorbent to settle prior to electrochemical regeneration, it must have a higher bulk density than the solution in which it is dispersed. Bulk density is a function of the specific densities of each component (i.e. the adsorbent particle and the adsorbed organic). It is calculated according to Equation 2:

Organics tend to have a lower specific density than water (and thus float). The adsorbent has a higher specific density than water (and thus sinks under gravity). Consequently as the oil component of the adsorbent-oil particle increases, its bulk density decreases, decreasing the efficiency of the settlement process. This can be mitigated by closely controlling the degree of self-association of the organic on the adsorbent.

Low Hydrophobicity

If an organic has a low hydrophobicity (that is, readily dissolves in water), the enthalpy of hydration of the organic is more favourable than the enthalpy of interaction with the adsorbent. Thus there is a low thermodynamic loading of the organic on the Nyex, which tends only to be monolayer in thickness and can be attributed to factors outside of hydrophobicity, for example, interactions with functional groups on the surface of the adsorbent. Interactions could include ionic and covalent bonds and hydrogen bonding, but the nature of the interaction depends on the nature of the organic. If the aqueous concentration of an organic with a low hydrophobicity is increased, the thermodynamic loading will generally only increase until a full monolayer is reached. Self-association, and therefore multilayer coverage, is not observed.

This understanding of adsorbent/organic interaction behaviour, combined with parallel electrochemical performance characteristics, has led to the optimization of the ODC operating parameters to achieve minimal operating expense.

Electrochemical influencers

The efficiency of electrochemical performance is associated with the mechanism of oxidation. There is a host of theoretical oxidation mechanisms and the broader academic field of electrochemical oxidation at carbonaceous surfaces is not fully understood. Nevertheless, the two fundamentally influencing mechanisms are indirect oxidation and direct oxidation.

Indirect oxidation is the oxidation of dissolved/suspended organics via exposure to aqueous species formed during the redox reactions of bulk water and other components within the solution. Indirect oxidation is less efficient than direct oxidation because it relies upon the spontaneous association of oxidising species and organic (an essentially random process that is affected by physical and chemical factors such as diffusion rates and lifetime of oxidizing species). The occurrence of indirect oxidation is dependent upon the concentration of organics in the solution. The higher the concentration of organics in the solution, the greater is the likelihood of spontaneous association.

There are a number of electrochemical processes available which depend upon the achievement of indirect oxidation. They generally employ only solution between the electrodes, rather than a conductive solid, and consequently have relatively high operational energy demands.

Direct oxidation is the transfer of electrons at the adsorbent/organic boundary. When an adsorbent bed is formed between two current feeders, the path of least resistance for the electric current is at this boundary, providing sufficient electron transfer for complete organic oxidation. The occurrence of direct oxidation is more likely if there is a high concentration of organic adsorbed onto the adsorbent, and vice versa.

The ratio of direct oxidation to indirect oxidation within the ODC system is therefore dependent upon the organic loading of the adsorbent and the concentration of organic in the solution. Consequently, the different organic tendencies for adsorption, and the difference in starting composition of the treatment waste will influence the overall efficiency of the treatment. To mitigate this fact, the lengths of the adsorption, separation and regeneration phases are tailored according to the intricacies of each independent waste stream.

Maintaining process efficiency

For organics at a high concentration within the solution and with a high hydrophobicity (that is, neat organic wastes that have been diluted to a desired concentration, for example legacy waste oil from decommissioning activities), there will be a high thermodynamic loading on the adsorbent and so direct oxidation is favourable. For the same nature of organic but at a lower concentration (that is, trace oil contamination from hydraulic systems service and maintenance) there will be a fraction less organic on the adsorbent, but there will still be enough to favour direct oxidation and therefore the length of the regeneration cycle is not limited by adsorption.

For organics at a low concentration and with low hydrophobicity (that is, trace solvent contamination from heavy water moderators and reactor coolant systems for example ethylene glycol), there will be a low thermodynamic loading on the adsorbent. For the same nature of organic but at a higher concentration (that is, cleaning and chelating agents from service maintenance and outage activities for example EDTA and morpholine), a higher thermodynamic loading on the adsorbent (up to monolayer) is achieved. In these cases the length of the regeneration and adsorption cycles can be manipulated to maintain as high an occurrence of direct oxidation as possible.

A batch treatment process (in a plant taking up three ISO containers) is used for the treatment of high concentrations of organics in water, or neat, diluted organics wherein a longer regeneration time is required. A continuous process is used for low concentration or trace organics whereby only small regeneration times are required. A major project for the scaling up of the continuous process is about to commence, which will end with a plant of 500m3/day, as well as a smaller, interim plant with a capacity of 200L/h.

Experimental results

Following on from successful laboratory and pilot scale trials, a 10-cell, site-based demonstrator unit was commissioned at the Magnox Trawsfynydd decommissioning site to destroy LLW and ILW radioactive oils. More than 99% of the oil was removed and destroyed at a rate of about 30 mL/hr [4].

References

1. For example, the Environment Agency's Certificate of Authorisation and Introductory Note for the Accumulation and Disposal of Radioactive Waste, Authorisation Number CD8155/CE4007

2. N.W. Brown, and E.P.L. Roberts, (2007), Electrochemical pre-treatment of effluents containing chlorinated compounds using an adsorbent, Journal of Applied Electrochemistry, 37, 1329-1335.

3. F.M. Mohammed, E.P.L. Roberts, A. Hill, A.K. Campen, and N.W. Brown, (2011), Continuous water treatment by adsorption and electrochemical regeneration, Water Research, 45, 3065-3074.

4. N.W. Brown, D.A. Wickenden, and E.P.L. Roberts, (2012), On-site destruction of radioactive oily wastes using adsorption coupled with electrochemical regeneration. Proceedings of WM2012 Conference, February 26 - March 1, 2012, Phoenix, Arizona, USA.

5. Warwick P. E. Radiochemical analysis of Nyex, effluent, catholyte, bubbler and electrode plates. (Sample ID: MO4 and T101-106 samples), GAU Radioanalytical Laboratory Analytical report no. 2337-2431, National Oceanography Centre, Southhampton, UK October 2011.

About the author

Alex Adams; Nigel Brown, technical director, Arvia Technology Ltd, Daresbury Innovation Centre, Daresbury, WA4 4FS.