J . Chem. Tech. Biotechnol. 1990, 49, 331-343

Biosorption of Radionuclides by Fungal Biomass" Christopher White1 & Geoffrey M. Gaddg Department of Biological Sciences, University of Dundee, Dundee DD1 4HN, UK (Received 29 September 1989; accepted 21 December 1989)

ABSTRACT Four kinds of bioreactor were evaluated for thorium removal by fungal biomass. Static-bed or stirred-bed bioreactors did not give satisfactory thorium removal probably because of poor mixing. An air-lijit bioreactor removed approximately 90-95 % of the thorium supplied over extended time periods and exhibited a well-dejned breakthrough point afer biosorbent saturation. The air-lift bioreactor promoted efjicient circulation and effective contact between the thorium solution and the mycelial pellets. Of several fingal species tested, Rhizopus arrhizus and Aspergillus niger were the most effective biosorbents with loading capacities of 0.5 and 0.6 mmol g respectively (116 and 13%mg g - ' ) at an inflow thorium concentration of 3 mmol dm-3. The efjiciency of thorium biosorption by A. niger was markedly reduced in the presence of other inorganic solutes while thorium biosorption by R. arrhizus was relatively unaffected. Air-lift bioreactors containing R.arrhizus biomass could effectively remove thorium from acidic solution (1 mol dm-3 HNOJ) over a wide range of initial thorium concentrations (0-1-3 mmol dm- 3). The biotechnological application and significance of these results are discussed in the wider context of fungal biosorption of radionuclides. Key words: biosorption, fungi, radionuclides, thorium, bioreactors, mycelial pellets.

* Paper presented at the meeting 'Recovery/Removal of Metals by Biosorption-A Commercial Reality or a Scientist's Dream?), organised by the Solvent Extraction and Ion Exchange Group of the Society of Chemical Industry and held in London on 18 May 1989. $ Present address: Department of Biotechnology, South Bank Polytechnic, 103 Borough Road, London SEl OAA, UK. 5 To whom correspondence should be addressed. 33 1

J. Chem. Tech. Biotechnol. 0268-2575/90/$03.50 0 1990 SCI. Printed in Great Britain

C. While, G. M . Gadd

332

1 INTRODUCTION

Microbial biomass has a high affinity for the actinide elements, heavy metals and also other radionuclides, as observed both in laboratory studies and in natural environments.'. Binding to biomass is the main route of sedimentation of actinides in oceanic ecosystems4 while metal-loaded bacterial cells can act as nuclei for the formation of a variety of crystalline metal deposits including phosphates, sulphides and organic-metal complexes in aquatic sediments.' Such sequestration of metals and radionuclides by microorganisms is a significant component of biogeochemical cycling and is also relevant to the fate of potentially toxic heavy metals/ radionuclides when discharged into the environment. The ability of microorganisms to accumulate such elements can result in their immobilization and removal from solution but ingestion by other organisms can result in transfer along food-chains, ultimately to These properties of microorganisms have given rise to a considerable interest in the use of microbial biomass and derived products to remove metals and radionuclides from industrial Since the uptake and accumulation of metals and radionuclides by microbes may involve both a number of physiological and physicochemical processes, which are not always possible to distinguish in practice, the overall process is often termed biosorption, although it is more appropriate to use the term in connection with those processes which are independent of metabolism.'*'^' ' Where the microorganisms are physiologically active, advantage can be taken of metabolismdependent accumulation of metals,'*' ' extracellular precipitation by excreted or of specific enzymic reactions to immobilize metal ions.' 2 * 1 Microbial biomass can also remove large amounts of heavy metals from solution by adsorption and related processes without the involvement of metabolic activity, allowing its use, for instance, in the absence of nutrients or other 'non-physiological' '~ conditions. This process has been applied to the removal of nickel and ~ o p p e rand to radionuclides such as thorium, uranium and p l ~ t o n i u m . ' ~ ~ ~ ~ Fungi and yeasts can accumulate significant amounts of heavy metals and radio nuclide^.^ Although this is a feature of microbial cells in general, fungi possess many attributes of interest and many species have received detailed study. Two significant advantages in relation to industrial exploitation are the range of morphological types available, which include unicellular and filamentous forms and the availability of large amounts of fungal biomass and derived products as waste from industrial processes and fermentations.'.'' The actual mechanisms of uptake can vary considerably between species and also depend on whether organisms are living or dead. With heavy metal cations, metabolism-independent biosorption to cell walls or other extracellular material may be followed by entry into the cell by transport across the cell membrane or as the result of an increase in cell membrane However, these phases may not occur in all species or with all metals and radionuclides. Furthermore, both metabolism-independent and -dependent processes may be affected by the excretion of complexing agents or other substances that may bind or precipitate metals and by changes in the physical and chemical characteristics of the growth or suspending m e d i ~ r n . ~ The uptake of radionuclides has been examined in a wide range of fungi and

Biosorption of' radionuclides b y fungal biomass

33 3

yeasts and this appears to mainly comprise metabolism-independent biosorption. The main site of uptake of actinides in fungi is the cell wall,20 although or detergentsz3 can increase permeabilisation of the cells with carbonates2 uptake, indicating that intracellular sites are also capable of binding. The mechanism of biosorption varies between elements. Both adsorption and precipitation of hydrolysis products occur with uranium,24while coordination with cell wall nitrogen was the main mechanism of thorium b i o s o r p t i ~ n . ~ ~ * ~ ~ Precipitation or crystallization of radionuclides within or on the cell walls may be a significant feature in some circumstances. In Saccharomyces cereuisiae, uranium was deposited as a layer of needle-like fibrils on cell walls, reaching up to 50% of the dry weight of the individual cells. That such a large amount was taken up implied that additional uranium had crystallized on that already b o ~ n d . Such ~ ~ , ~ ~ precipitation has also been observed for t h o r i ~ m . ~ . ' ~ . ~ ' Biosorption of uranium and thorium may be affected by the external pH. Initial rates of uranium uptake by yeast increases above pH 2.5.28At pH values below 2.5, the predominant species is UO:', but at pH values above 2.5, hydrolysis products include (UO,),(OH)i+, UO,(OH)+ and (UO,),(OH)i. The accompanying reduction in solubility favours biosorption.' 7 * 2 4 Similar phenomena occur with thorium where, at pH values below 2, Th4+ is the main species present. At higher pH values, Th(0H)i and other hydrolysis products form, which are taken up more eflkiently than Th4+.'6,25 Despite these effects, significant quantities of radionuclides may still be accumulated at low pH values. It has recently been shown that biomass from a wide range of fungal species was able to remove thorium from acidic solutions at pH values below l.0.23 Such acidic conditions are characteristic of many industrial process streams and this attribute of fungi may be particularly useful in an industrial context. This paper, as well as outlining and discussing previous work on fungal biosorption, examines the use of filamentous fungal biomass in several designs of column bioreactor as a biosorbent for thorium in acidic solution and also in a simulated acidic process stream. The strains of filamentous fungi used were those which gave the best performance as biosorbents in previous batch studies under acidic condition^.'^ 'vZ2

+

2 METHODS 2.1 Organisms and culture

Rhizopus arrhizus IMI 57412, Penicillium italicum, P . chrysogenum IMI 26211 and Aspergillus niger were routinely maintained on malt extract agar (Oxoid) at 25°C. For experimental purposes, these strains were grown in a liquid medium comprising (g dm-3); D-glucose, 20.0; (NH4),S04, 5.0;KH,PO,, 05; MgSO, .7H,O, 0.2; CaCl, .2H,O, 0.05; NaCl, 0.1; FeCI,, 0.0025; ZnS0,.7H,O, 0.004; MnS0,.4H20, 0.004; CuSO,. 5H,O, 0.0044. Fungi were cultured in a 2dm3 glass air-lift fermenter equipped with a Gallenkamp pH controller. The pH was maintained in the range 5.5 to 6.5 by addition of 4 mol dm-3 KOH in all cases. All of the cultures were inoculated using

C . White, G . M . Gadd

334

conidial suspensions which were prepared by shaking 100 cm3 of sterile liquid medium with a well-grown culture on approximately 200 cm3of malt extract agar in a 500 cm3 medical flat bottle. The conidial suspension was added aseptically to the fermenter to a final density of approximately lo5 ~ m - Mycelial ~ . pellets were harvested after 96 h culture at 25°C by filtration through a 63 pm nylon mesh (Staniar Ltd, Manchester, UK). The biomass was washed on the sieve with 4 dm3 of distilled water. Excess water was removed by draining, followed by pressing with three changes of absorbent paper. 2.2 Thorium assay This was carried out using the Arsenazo I11 colorimetric determination. Glassware and cuvettes were washed in l.0moldm-3 HCl followed by three changes of distilled, deionized water and dried before use. All preparations used analytical grade reagents and were made up in distilled, deionized water. A volume of the solution to be assayed containing 5-50 nmol of thorium was pipetted into a test tube and 2.4 cm3 of concentrated HC1 was added, followed by 1.0 cm3 of 8.0% (w/v) oxalic acid and 0.8 ml of 0.5% (w/v) arsenazo I11 (Sigma Chemical Co., Poole, Dorset, UK). The volume was then made up to 10cm3 with distilled, deionized water. The sample was thoroughly mixed by vortexing between additions. The optical density at 655 nm was measured against a reagent blank using a Pye Unicam SP600 spectrophotometer (Pye Unicam Ltd, Cambridge, UK). Known quantities of thorium as thorium nitrate (BDH Ltd, Dagenham, Essex, UK) were used as standards. Where additional solutes were present in the process stream, these were also added to the reagent blank and standards. 2.3 Column reactors Detailed descriptions of the column designs used are given in Section 3. In all cases the column contained 10 g pressed weight of biomass. The thorium solution was supplied as thorium nitrate from a 5 dm3 conical flask acting as reservoir and a 1 dm3 Buchner flask acting as a header-tank. The solution was pumped between these with a peristaltic pump. Overflow back to the reservoir from the header was by gravity. Solution flow through the column was maintained at 200 cm3 h- using a peristaltic pump supplied with a Pharmacia Frac-100 fraction collector (Pharmacia LKB, Uppsala, Sweden) which was used to collect the efluent. 2.4 Other methods Dry weights were determined by drying to constant weight at 80°C in tared, aluminium foil cups. Conidial suspensions were counted using a modified FuchsRosenthal haemocytometer. 3 RESULTS 3.1 Bioreactor column designs Four patterns of bioreactor were evaluated. These were packed-bed reactors using upwards and downwards flow, a bed agitated by stirring and one in which an

Biosorption of radioriuclidrs by Jingo1 biomass

335

internal secondary circulation was supplied by air-lift. All of the columns contained 10 g pressed weight of R. arrhizus biomass but in a variable column volume. The two packed-bed columns had total volumes of 50 cm3, while the stirred and airlift columns had volumes of 200 cm3 each to allow for circulation. These bioreactor designs are shown in Fig. 1. Neither of the static-bed designs nor the stirred bed gave satisfactory thorium removal. In these reactor designs the proportion of thorium removed from the process stream was low, decreasing gradually with no well-defined breakthrough (Fig. 2). This suggested that mixing in these designs was poor and passage of dyes such as methylene blue through the bed also showed that some channelling occurred. However, the air-lift reactor gave efficient removal of approximately 9095% of the thorium present over a prolonged period followed by a rapid breakthrough at saturation of the biosorbent (Fig. 3(a)).This indicated that the air-

(b)

(a) inflow

outflow

inflow

(d)

(C)

air out inflow outflow

sleeve outflow air in

n

inflow

0

glass wool plugs

Fig. 1. Diagrammatic representation of column bioreactors used in this study. The designs are: (a) packed bed with downwards flow; (b) packed bed with upwards flow; (c) stirred bed, and (d) air-lift reactor. The biomass is not shown. The total volumes were: (a) and (b) 50 cm3, (c) and (d) 200 cm3.

C . White. G . M . Gadd

336

Fig. 2. Removal of thorium from a 2.0 mmol dm-' solution in 1.0 mol dm-3 HNO, by (0) packed-bed reactor with downwards flow, ( 0 )packed-bed reactor with upwards flow and (0) stirred-bed reactor. The data are derived from two separate experiments (four replicates) and the bars indicate the SEM.

-

2.0

1 ?

I?

'E D

- 1.0

E

-E C

0

c

+ F

s

0 2.0

C

8 c

s

2 r

1.0

W

0

500

1000 0

500

1oc

Volume passed (cm 3,

Fig. 3. Removal of thorium from 2 mmol dm- solution in 1 mol dm- HNO, by an air-lift reactor. The biomass used was: (a) R. arrhizus, (b) A . niger, (c) P. italicurn and (d) P . chrysogenum. The data are derived from two separare experiments (four replicates) and the bars indicate the SEM.

lift mechanism promoted an effective contact between the thorium solution and the biosorbent because of efficient secondary circulation and mixing of the bed. The airlift design of bioreactor was therefore used in subsequent experiments.

3.2 Variation between fungal species There was considerable variation between the performance of biomass derived from

Biosorption of' radiotiuclides by futigal hiomuss

331

different fungal species in the air-lift bioreactor. The time course of thorium uptake by the four strains used is shown in Fig. 3. While biomass from A . niger gave high removal and a well-defined breakthrough similar to R. arrhizus in the air-lift reactor, the performance of biomass from the two Penicillium strains was significantly worse than these. Penicilliurn biomass removed less thorium in total during the course of the run, and the amount of thorium removal decreased gradually, showing no clearly defined breakthrough. The total amount of thorium removed by both R. arrhizus and A . niger was consistent with the biomass of these species attaining equilibrium with the inflow solution at similar loadings to those previously determined in batch experiment^.'^ These were, at an inflow concentration of 3 mmol dm-,, 0.5 mmol g-' (116mgg-')for R.arrhizusand06mmol g-' (138 mgg-')forA. nigerin termsof dry weight. The two Penicillium species showed very similar uptake to these strains in batch experiments', so that their poor performance in column reactors was clearly related to the operation of the column system. One possible mechanism which reduced uptake by these strains was the breakdown of the mycelium which occurred in the Penicillium strains in the air-lift reactor. This did not occur during stirred incubation with 1 mol dm-, HNO, and was thus apparently the direct consequence of air-lift agitation in the column. The reduced uptake observed may thus have resulted from the subsequent release of thorium taken up by the mycelium.

3.3 Thorium biosorption in the presence of inorganic solutes The simulated acid process liquor used in this experiment comprised 200mmol dm-3 NaNO,, 10 mmol dm-, Mg(NO,)', 1Ommol d m P 3 Ca(NO,),, 20 mmol dm-, Al,(SO,), and 10 mmol d m - 3 FeCl, in addition to either 200pmol dm-, or 2.0 mmol dm-3 thorium in 1 mol dm-, HNO,. Where R. arrhizus was used as a biosorbent, these solutions had no apparent effect on thorium uptake at an initial concentration of 200pmoldm-3 (Fig. 4). At an initial concentration of 2.0 mmol dm- ,, however, the efficiency of removal was reduced and the transition was more gradual at breakthrough in the presence of these solutes than in their absence (Fig. 5 ) . A similar effect was also observed in batch experiments and it appeared that these solutes did not greatly affect the affinity of the biomass for thorium but reduced its capacity.23 The effect of these solutes on thorium uptake by A . niger biomass in an air-lift reactor was much more pronounced, resulting in greatly reduced uptake and more gradual breakthrough (data not shown). As the effects of solutes on equilibria in batch experiments using biomass of this species were almost identical to those on R. ~ r r h i z u s , ' ~other factors were clearly contributing to the reduced uptake in the airlift reactor, although the identity of these factors is currently uncertain. However, it was clear from these results that species differences were extremely significant in determining the effects of solutes on thorium uptake and that A . niger biomass was very much reduced in its efficacy as a biosorbent in the presence of these solutes, while R . arrhizus biomass was almost unaffected.

C . White, G . M . Gadd

338

Volume passed (dm3)

Fig. 4. Removal of thorium from a 200prnol dm-’ solution in (0) 1 mol dm-’ HNO, and ( 0 )in the presence of solutes simulating process liquor (see text). The data are derived from two separate experiments (four replicates) and the bars indicate the SEM.

Lo-0 4 /

0

500

1000

Volume passed (an3)

Fig. 5. Removal of thorium from a 2.0 mmol dm-’ solution in (0) 1 mol dm-’ HNO, and ( 0 )in the presence of solutes simulating process liquor. The data are derived from two experiments (four replicates) and the bars indicate the SEM.

3.4 Operation of columns at varying initial thorium concentrations Air-lift columns containing R. arrhizus biomass were able to remove thorium from acidic solution over a wide range of initial concentrations between 0.1 and 3 mmol dm- (Fig. 6). The proportion of thorium removed was high in all cases and the total amount of thorium taken up at saturation of the biomass was consistent

Biosorption of radionuclides by fungal biomass

1.0 20 3.0 Volume passed (dm3)

339

4.0

Fig. 6. Removal of thorium from solution in 1 mol dm-, HNO, at initial thorium concentrations of (0) 100 pmol dm-3, ( 0 )400 pmol dm-,, (0) 1.0 mmol drn-,, ( W ) 2.0 mmol dm-’ and (A)3.0 mmol dm-3. The data presented are representative examples.

with uptake isotherms obtained in batch experiment^.^^ The transition at breakthrough was also sharp at all of these concentrations (Fig. 6).

4 DISCUSSION

In order to achieve maximum efficiency in the use of any kind of sorbent to remove solutes from a solution flowing through a contactor, it is necessary that the effluent concentration should remain low and constant for as long as possible, and that the transition on exhaustion of the sorbent should be as rapid as possible. This ensures In a previous paper the that the sorbent is as close to saturation as po~sible.~’ present authors reported that fungal biomass can act as a biosorbent with both high affinity and high capacity for thorium in batch experiments carried out under acidic condition^.^^ An objective of the present study was to ascertain whether the fungi used met the above criteria in use in contactors with a continuous flow of thorium solution. Several factors were found to affect the performance of the biosorbents in the bioreactors. The column design was clearly the most significant variable, with a very

3 40

C . White, G . M . Gadd

poor performance being obtained from static beds or those stirred at one point in the column. A very much better performance was achieved by using the air-lift contactor. A major factor was undoubtedly bulk mixing, but the air-lift probably also improved transfer within the mycelial pellets, since these are flexible structures whose pore space may be greatly reduced by settling or compaction. In an air-lift reactor, the most significant variable was the biological nature of the biomass used. Rhizopus arrhizus and A . niger biomass both met the abovementioned criteria for removal of thorium from a pure solution, while that of the Periicillium species did not. The affinity and capacity of the former two species for thorium in batch experiments was also greater23and this clearly contributed to their better performance in a column system. These differences between species were, however, not completely adequate to account for the extent of the difference observed and it is likely that the disintegration of Penicillium biomass in an air-lift reactor was also significant. Furthermore, there were considerable differences between R . arrhizus and A . niger biomass in the effect on thorium uptake by other solutes characteristic of thoriumcontaining process streams. These only slightly reduced the uptake capacity of R . arrhizus biomass, but had a very much greater effect on A . niger, to the extent of preventing it from acting as a useful biosorbent in the column system. The form of the biosorbent is an important consideration when microbial material is used in a metal removal process. For real or simulated industrial application, freely suspended fungal biomass has several disadvantages which include low density and mechanical strength which may make biomass/effluent separation difficult.' The unicellular form of many microbes, such as bacteria, yeasts or microalgae, presents a problem in any reactor system where the treated process stream flows continuously through the reactor, as they may be subject to washout, compaction or clogging.'*'9 Immobilized or pelleted biomass can surmount such problems and has the beneficial attributes of easy separation of cells and effluent, minimal clogging, high biomass loadings and flow rates as well as a better capability for regeneration and r e u ~ e . ' * ~ - ' ~Several . ' ~ - ~ kinds ~ of immobilization treatment have been applied to fungi and other microbes, including entrapment within organic or inorganic matrices and attachment to inert surfaces or support particle^.^^'^*^'*^^ However, treatment of a highly acidic emuent, such as that studied here, may lead to complications with organic matrices and immobilization within or on inert matrices may be more appropriate. Despite the advantages listed, it should be borne in mind that biosorbent immobilization may constitute an additional and significant economic cost which may preclude its use under certain conditions. Clearly, economic considerations are paramount in the exploitation of biomass in metal removal/recovery systems. In this study, thorium removal is primarily for decontamination prior to safe storage or environmental discharge. There is little or no economic benefit in thorium recovery, unlike the situation with rare or precious metals.33Thus, as simple a process as can be devised is desirable. An advantage with some filamentous fungal systems that has been exploited here is that many can be grown in the form of pellets which have analogous properties to immobilized particles. Furthermore, the use and application of fungal pellets in

Biosorption of radionuclides by fungal biomass

341

other industrial applications is well e ~ t a b l i s h e d .All ~ ~the . ~ fungi ~ used in this study were used in pellet form; those of the Penicillium species and A . niger were spherical while those of R . arrhizus were more irregular. Pellets of A . niger have previously been used for uranium removal in a fluidized-bed bioreactor. This system was more efficient than the commercial ion-exchange resin IRA-400.36 However, the pellets eventually showed some disintegration resulting in an increased resistance to liquid flow, and similarities in density between the biomass and the liquid medium made continuous operation Similar disintegration of Penicillium pellets was observed in the present study, but not for R . arrhizus or A . niger which were more robust, even under the acidic conditions employed. Species selection is obviously highly important for these kinds of systems. Certain kinds of pelleted fungal biomass are, therefore, effective biosorbents for thorium in a flow-through reactor. Effective removal requires an adequately agitated biosorbent bed, achieved here by means of an air-lift but probably also achievable by a fluidized-bed design on a larger scale. The physical and chemical conditions of operation greatly affected uptake by some of the fungi examined but selection of suitable species was able to largely overcome these problems. In conclusion, biosorption using fungal biomass is a technically feasible method for the removal of thorium from acidic solutions similar to those encountered in industrial process streams. However, the practicality of this approach is very much dependent on the economics of the process. Such factors as the availability of waste biomass or cheap fermentable growth substrates may favour such a process while others, such as the need for handling and transport of a bulky biosorbent, or the need for largescale, on-site fermentation facilities would act against it. The balance of these factors could only be determined in individual cases by detailed costings.

ACKNOWLEDGEMENTS The authors gratefully acknowledge financial support from BNFL plc and Dr H. Eccles for helpful discussion.

REFERENCES 1 . Gadd, G . M., The uptake of heavy metals by fungi and yeasts: The chemistry and physiology of the process and applications for biotechnology. In Immobilisation of Ions by Bio-sorption, ed. H. Eccles & S. Hunt. Ellis Horwood, Chichester, 1986, pp. 149-58. 2. Gadd, G . M., Fungal responses towards heavy metals. In Microbes in Extreme Environments, ed. G . A. Codd & R. A. Herbert. Academic Press, London, 1986, pp. 83-1 10. 3. Gadd, G . M. & White, C., Heavy metal and radionuclide accumulation and toxicity in fungi and yeasts. In Metal-Microbe Interactions, ed. R. K. Poole & G. M. Gadd. IRL Press, Oxford, 1989, pp. 19-38. 4. Fisher, N. S.,Cochran, J . K., Krishnaswami, S. & Livingston, H. D. Predicting the oceanic flux of radionuclides on biogenic debris. Nature, 335 (1988) 622-5. 5. Beveridge, T. J., Interactions of metal ions with components of bacterial cell walls and their biomineralization. In Metal-Microbe Interactions, ed. R. K. Poole & G . M. Gadd. IRL Press, Oxford, 1989, pp. 65-84.

342

C . White, G. M . Gadd

6. Gadd, G . M. & Grifiths, A. J., Microorganisms and heavy metal toxicity. Microbial Ecology, 4 (1978) 303-1 7. 7. Duxbury, T., Ecological aspects of heavy metal responses in microorganisms. In Advances in Microbial Ecology, ed. K. C. Marshall. Plenum Press, New York, 1985, pp. 185-235. 8. Ochiai, E. I., General Principles of the Biochemistry of the Elements. Plenum Press, New York, 1987. 9. Gadd, G. M., Accumulation of metals by microorganisms and algae. In BiotechnologyA Comprehensive Treatise, Vol. 66, Special Microbial Processes, ed. H.-J. Rehm. VCH Verlagsgesellschaft , Weinheim, 1988, pp. 40 1-3 3. 10. Macaskie, L. E. & Dean, A. C. R., Microbial metabolism, desolubilization and deposition of heavy metals: metal uptake by immobilized cells and application to the detoxification of liquid wastes. In Biological Waste Treatment, ed. A. Mizrahi. Alan R. Liss Inc., New York, 1989, pp. 159-201. 11. Shumate, S. E. & Strandberg, G. W., Accumulation of metals by microbial cells. In Comprehensive Biotechnology, Vol. 4, ed. M. Moo-Young, C. N. Robinson & J. A. Howell. Pergamon Press, New York, 1985, pp. 235-47. 12. Hansen, C. L., Zwolinski, G., Martin, D. & Williams, J. W., Bacterial removal of mercury from sewage. Biotechnol. Bioeng., 26 (1984) 133C3. 13. Macaskie, L. E. & Dean, A. C. R., Cadmium accumulation by a Citrobacter sp. J. Gen. Microbiol., 130 (1984) 5 3 4 9 . 14. Norris, P. R. & Kelly, D. P., Accumulation of metals by bacteria and yeasts. Develop. Indust. Microbiol., 20 (1979) 299-308. 15. Horikoshi, T., Nakajima, A. & Sakaguchi, T., Studies on the accumulation of heavy metals in biological systems. XIX. Accumulation of uranium by microorganisms. European J . Appl. Microbiol. Biotechnol., 130 (1984) 53-69. 16. Shumate, S. E., Strandberg, G. W. & Parrott, J. R., Biological removal of metal ions from aqueous process streams. Biotechnol. Bioeng., 8 (1978) 13-20. 17. Tsezos, M. & Volesky, B., Biosorption of uranium and thorium. Biotechnol. Bioeng., 25 (1983) 583-604. 18. Tsezos, M. & Keller, D. M., Adsorption of radium-226 by biological-origin adsorbents. Biotechnol. Bioeng., 25 (1983) 201-5. 19. Tsezos, M., Adsorption by microbial biomass as a process for removal of ions from process or waste solutions. In Immobilisation of Ions by Bio-sorption, ed. H. Eccles & S. Hunt. Ellis Horwood, Chichester, 1986, pp. 201-19. 20. Weidemann, D. P., Tanner, R. D., Strandberg, G. W. & Shumate, S. E., Modelling the rate of transfer of uranyl ions onto microbial cells. Enzyme Microbial Technol., 3 (1981) 3340. 21. Galun, M., Keller, P., Feldstein, H., Galun, E., Siegel, S. & Siegel, B., Recovery of uranium(V1) from solution using fungi. 11. Release from loaded Penicillium biomass. Water, Air Soil Pollut., 20 (1983) 277-85. 22. Galun, M., Keller, P., Malki, D., Feldstein, H., Galun, E., Siegel, S. & Siegel, B., Removal ofuranium(V1) from solution by fungal biomass: Inhibition by iron. Water, Air Soil Pollut., 21 (1984) 41 1-14. 23. Gadd, G. M. & White, C., Removal of thorium from simulated acid process streams by fungal biomass. Biotechnol. Bioeng.,33 (1989) 592-7. 24. Tsezos, M. & Volesky, B., The mechanism of uranium biosorption by Rhizopus arrhizus. Biotechnol. Bioeng., 24 (1982) 385401. 25. Tsezos, M. & Volesky, B., The mechanism of thorium biosorption by Rhizopus arrhizus. Biotechnol. Bioeng., 24 (1982) 95549. 26. Tsezos, M., The role of chitin in uranium biosorption by Rhizopus arrhizus. Biotechnol. Bioeng., 22 (1983) 2025-40. 27. Gadd, G . M. & White, C., Uptake and intracellular compartmentation of thorium in Saccharomyces cerevisiae. Environ. Pollut., 61 (1989) 187-97. 28. Strandberg, G. W., Shumate, S. E. & Parrott, J. R., Microbial cells as biosorbents for

29. 30. 31. 32. 33. 34. 35. 36.

heavy metals: accumulation of uranium by Saccharomyces cerevisiae and Pseudomonas ueruyinosa. Appl. Enoiron. Microbiol., 41 ( 1 981 ) 23745. Weber, W. J. & Morris, J. C., Equilibria and capacities for adsorption on carbon. J . Sanitary Eng., Division of the American Society of Civil Engineers, 90 (1964) 79-107. Hutchins, S. R.,Davidson, M. S., Brierley, J. A. & Brierley, C. L., Microorganisms in reclamation of metals. Ann. Rev. Microbiol., 40 (1986) 311-36. Kiff, R.J. & Little, D. R.,Biosorption of heavy metals by immobilized fungal biomass. In Immohilisarion of Ions by Bio-sorption, ed. H. Eccles & S. Hunt. Ellis Horwood, Chichester, 1986, pp. 71-80. Townsley, C . C., Ross, I. S. & Atkins, A. S., Copper removal from a simulated leach effluent using the filamentous fungus Trichoderma viride. In Immobilisation of Ions b y Bio-sorprion, ed. H. Eccles & S. Hunt. Ellis Horwood, Chichester, 1986, pp. 159-70. Brierley, J. A., Goyak, G. M. & Brierley, C. L., Considerations for commercial use of natural products for metal recovery. In Immobilisation of Ions by Bio-sorption, ed. H. Eccles & S. Hunt. Ellis Horwood, Chichester, 1986, pp. 105-17. Anderson, J. G., Immobilized cell and film reactor systems for filamentous fungi. In The Filamentous Fungi, Vof. 4 , Fungal Technology, ed. J. E. Smith, D. R. Berry & B. Kristiansen. Edward Arnold, London, 1983, pp. 145-70. Whitaker, A., Fungal pellets-present and potential applications. Int. Indust. Biotechnol., 7 (1987) 285-9. Yakubu, N. A. & Dudeney, A. W. L., Biosorption of uranium with Aspergillus niger. In Immobilisation o j Ions by Bio-sorption, ed. H. Eccles & S. Hunt. Ellis Horwood, Chichester, 1986, pp. 183-200.