Novel Chemical Solulions Remove Heavy Metals from Wastewaler. industrial Waslcwalcr.

(August/September), p53-56.; Apte, N. 1993. Hazardous Waste Treatment Technologies - An inventory. Hazmat World. (August), p27-32.; Zagula, S. and Beitinger, E. 1994. Gelting Down and Dirty. Industrial Wastewater. (January/February), p20-25.). These technologies make use of classical ion exchange, chemical precipitation, electrochemical precipitation, membrane concentration and evaporation. In virtually all cases, metal selectivity is limited with the result that, for complex wastes or groundwaters, the sludges and residucs generated for disposal or recycle include substantial amounts of nonhazardous, nonregulated metals such as iron, calcium and magnesium. The applications of biological processes are less advanced than chemical approaches, but have also been reviewed (Brierley, C.L. 1991.

Bioremediation of Metal-Contaminated Surface and Groundwaters. Geomicrobiology Journal. 8:201-223.; Means, J. L. and R. E. Hinchee, eds. Emerging Teclmology for Bioremediation of Metals. CRC Press, Inc. Boca Raton. 1994.). Even though efforts continue on the development of physio-chemical treatments that are less hazardous than current technologies for metals removal, they have met with limited success. Bioprocessing, however, especially with biosorbents, is intrinsically less hazardous. Moreover, biosorbents promise to be able to treat lower influent concentrations, reach lower effluent levels, and act more selectively toward transition and heavy metals thus extending this technology to a broader array of applications (Smith, L.A.; Alleman, B.C. and Copley-Graves, L. 1994.

Biological Treatment Options in Emerging Technologies for Bioremediation of Metals. J.L.

Means and R.E. Hinchee, eds., pl-12.; Mattison, P.L. Bioremediation of Metals - Putting It to Work. Cognis. Santa Rosa, CA. 1994).

A number of (bio)adsorbent technologies based on bacterial, fungal, algal, cellulosic and related materials are reported to be commercially available. The performance of many of these materials has been summarized and extensively reviewed (Brierley, 1991). While more selective than classical ion exchange matrices, these materials are often subject to interference by metals such as aluminum and iron. In addition, these structures are described as simple cation exchangers with good functionality over a relatively limited pH range. Optimal performance is seen on metal ion feed concentrations ranging from 10 ppb to 10 ppm with a significant fall off in performance at lower concentrations even though capacities are still about a factor of ten higher than synthetic ion exchangers at low feed concentrations. The data summarized by Brierly indicates that these adsorbents have high theoretical capacities for metals, but in most cases, demonstrate operational capacities that are significantly lower.

For example, dry weight capacities for adsorbents like BIOCLAIMTM and BIOFIXTM are found to be - 2-3 % for metals such as cadmium and copper vs. their theoretical capacities of 16 % as calculated from their Cation Exchange Capacity (CEC) values of 5 meq/g. One potential explanation for this behavior is that the heavy metals of interest are coordinated at 4- 6 sites per metal ion rather than the 1 site per proton/sodium ion on which the CEC titrations are based.

Summary of the Invention The examples presented in this application demonstrate that crude cell-exopolysaccharide (EPS) complexes from marine microorganisms and especially those of the genus Hyphomonas including the strain designated MHS-3 (ATCC No. 43965) and other Hyphomonas strains such as ATCC Nos. 33882, 33884, and 43869, and similar species as well as various crosslinked and derivatized structures possess exceptional heavy metals chelation, adsorption and removal capabilities. It has been demonstrated for crude MHS-3 cell mass that: 1) it is a more effective chelation agent than several purified EPS fractions; 2) it is an effective adsorbent for most metals at acidic pH's from at least as low as about 4.4 to at least as high as about 6.6 (where metal oxide/hydroxide precipitation becomes a significant removal mechanism); 3) its capacity for metals is exceptional and is in the range of about 15+ % on a dry weight basis for mercury and lead and of about 3-7 % for most other metals; 4) it is able to reduce the concentrations of copper, cadmium, lead and mercury from about 50-100 parts per billion (ppb) levels to levels at or below U.S. EPA drinking water standard MCL's (Maximum Contaminant Levels); 5) its relative metals <BR> <BR> <BR> <BR> binding strengths are: Cu2+>Hg2+>Pb2+>Cd2+>Zn2+; , 6) it exhibits excellent selectivity for targeted heavy metals even in the presence of environmental contaminant metals such as iron and aluminum and hard water metals such as calcium and magnesium and only begins to lose effectiveness when those ions reach concentrations substantially in excess of the target metal and their normal environmental levels; 7) its metals adsorption capacity can be completely regenerated at least four times; 8) it exhibits higher capacity than available commercial materials such as the ForagerTM sponge, 9) it can be formed into granular, easily processed and handled resin-like materials by a number of simple chemical crosslinking procedures which maintain and enhance its metals binding capacity, and 10) these resin-like adsorbents are easily regenerated.

Other modification procedures provide useful variations in metal binding and result in adsorbents with unique properties.

A Brief Description of the Drawings Figure 1: Shows the pH dependence of metal binding to MHS-3 cells. One mg dry weight (dwt) cells was incubated in 101lg metal in 1.7 mL 1 mM Acetate-buffer or 1 mM MES buffer adjusted to the appropriate pH for 24 h. After removal of the cells by centrifugation, the metal concentration in the supernatant was determined by AAS and used to calculate the amount of metal bound to the cells.

Figure 2: Shows the metal binding capacity of MHS-3 cells. One mg dwt cells was incubated for 24 h in 1.7 mL 1 mM MES buffer pH 6.0 containing 200 pg of metal. Analyses were performed at least as duplicates (Fe); for most of the metals 6 individual values were averaged, and in the case of copper and cadmium the data represent the average of 36 individual samples.

Figure 3: Shows the comparison of the metal binding performance of each 1 mg dwt of MHS-3 cells and ForagerTM Sponge, a commercially available heavy metal removal material. Bound metal amounts were calculated after an incubation time of 24 h by measuring the remaining metal concentration of the supernatant by AAS.

Figures 4a-f: Show select graphs from competitive binding experiments. One mg of dwt MHS-3 cells was incubated in a total volume of 1.7 mL at pH 6.0 with various amounts of target metal in addition to various amounts of competitor metal. After an incubation time of 24 h, the metal content of the cells was determined as described previously. Fig. 4a shows the influence of Al on Pb binding; Fig 4b of Fe2+ on Pb binding, Fig 4c of Hg on Pb binding, Fig. 4d of Pb on Hg binding, Fig. 4e of Zn on Cd binding, and Fig. 4f shows the influence of Ca and Mg on Cu binding.

Figures 5a-b: Show the displacement of bound metal by Awl3+ 10, 50, and 200 pg Cd2+ (5a) or Cu2+ (5b) were incubated in 1.7 mL lmM buffer pH 6.0 with 1 mg dwt MHS-3 cells for 24 h. After determining the amount of metal bound to the cells using the standard method, cells were washed and incubated again in various amounts of A13+, after an additional incubation time of 24 h, cells were removed by centrifugation and the supernatant was analyzed by AAS to determine the amount of target metal displaced.

Figure 6 a-d: Show the efficiency of the removal of low concentrations of metals by MHS-3 cells. A total of 0.588 mg/mL dwt MHS-3 cells were incubated in 5880, 588, 294, or 58.8 ppb of the target metal in 1 mM MES buffer pH 6.0. After an incubation time of 24 h remaining metal concentrations were determined by graphite furnace AAS. Fig 6a shows the data obtained for Cd, Fig. 6b for Cu, Fig 6c for Hg, and Fig. 6d for Pb. In the case of Pb, the remaining concentrations for the lowest 3 concentrations were found to be lower than the method detection limit (M.D.L.) of 5 ppb, therefore the dotted line indicates the more likely performance. The dashed line in the graphs indicates the current EPA drinking water threshold for each of the metals.

Figure 7: Shows the regeneration of the metal binding performance of MHS-3 cells by 10 mM HC1 after subsequent incubations in 10 or 50 ,ug Cu. As usual, 1 mg dwt cells were incubated in a total volume of 1.7 mL 1 mM MES buffer pH 6.0 for 24 h. Volumes of regenerating solution and washing buffer were 1.5 mL respectively.

Figure 8: Shows the efficiency of various strategies on the regeneration of the metal binding performance of epichlorohydrin/hexanediamine crosslinked MHS-3 cells after incubation in 10 or 200 pg Cu, the strongest binding target metal. The material was treated for 24 h with the regenerating solution; in the cases where two different treatments had been used, the first treatment was done for 18 h, the second for 6 h. After washing the regenerated material in 20 mM MES buffer pH 6.0, the material was incubated again using the previous amounts of metal for 24 h in order to determine the extent of regeneration.

Detailed Description of the Invention The examples presented in this application demonstrate that crude cell-exopolysaccharide (EPS) complexes from marine microorganisms and especially those of the genus Hyphomonas including the strain designated MHS-3 (ATCC No. 43965) and similar species as well as various crosslinked and derivatized structures possess exceptional heavy metals chelation, adsorption and removal capabilities.

The starting microbial strain for this work, Hyphomonas sp. MHS-3 (ATCC No. 43965), was cultured in full-strength Marine Broth (MB, 37.4 g/L) (Difco Laboratories, Detroit, MI).

Isolated colonies were picked from Marine Agar plates (MB, 2 % agar w/v), inoculated into 100 mL of sterile MB and incubated at 25 "C for 96 h. Loopfuls of the copious biofilm that formed on the walls of the culture flask were used to inoculate 100 mL seed cultures, 1 mL aliquots of 4-5 day old seed cultures containing biofilm material were used to inoculate 1 L MB flasks, which were grown at 25"C for 4-5 days. At this stage, copious flocculation and biofilm formation could be observed in the flasks. MHS-3 cell mass and adherent EPS were harvested during the stationary phase of growth. To harvest the cultures, biofilm material was removed from the walls of the flask cultures with a rubber policemen, and the broth centrifuged at 16,000- 20,000 x g for 15 min. The supernatant was discarded and the pellets processed for metal scavenging experiments or for crosslinking.

Harvested MHS-3 cells/biofilm pellets were resuspended in 10 mM citric acid (pH 2.0) + 3 % NaCI with a Brinkman Homogenizer for 10 seconds to remove metals acquired from Marine Broth from binding sites in the material. The material was centrifuged, the supernatant discarded, and the pellet resuspended in the appropriate 1 mM buffer to a concentration of 10 mg (dry wt)/mL working buffer (the dry wt of MHS-3 cells is 15 % of their wet weight; cell suspensions were prepared using 66 mg wet weight cells/mL) to prepare the MHS-3 cell stock suspensions.

Crude cell mass was evaluated over the pH range of 4 to 7 on well characterized aqueous solutions of Cu2+, Cd, Pb2+, Hg2+, Zn2+ and C+ in order to: 1) define its operational pH range on target heavy metals, and 2) determine the approximate pH optimum for use. Crude cell mass was found to be an effective adsorbent over the entire pH range studied of 4.4 to 6.6.

A maximum in removal efficiency/binding capacity in the region of pM 6-6.5 was noted for most metals and was most pronounced for Zn2+ and least pronounced for Pb2+ and Hg2+. Crude cell mass as well as its crossed linked derivatives can be used in media between pH 1 and 11, and temperatures between 0 "C and 200 "C and preferably between pH 4 and 9, and temperatures between 5 "C and 50 "C.

In subsequent standardized batch incubation procedures, known, standardized amounts of fresh, acid-treated crude cell mass were exposed to amounts of Zn2+, Pb2+, Cu2+, Cd2+, Hg2+, Cr3+ and (Cr2O7)2-, Fe2+ and Fe3+ ranging from 10pg to 200pg per mg of dry cell mass (concentrations of these metals were 6 ppm to 120 ppm) for various lengths of time at pH 6 in order to determine metal uptake rates and maximum binding capacities. These experiments indicated cells possessed exceptional dry weight capacities for Hg2+ and Pb2+ of ~ 15 % with Cu2+ at - 6 %, Fe2+ and Fe3+ at - 4 %, Cd2+ at 3.5 %, C+ and Zn2+ at - 2 % and (Cr2O7 at 1 %. The rate of metals uptake was rapid and typically complete within one-half hour.

Similar batch incubation experiments were carried out with very low concentrations ¼60 to 600 ppb) of a subset of targeted metals (Cu2+, Cd2+, Hg2+ and Pb2+) in order to assess the ability of MHS-3 cell mass to clean up contaminated water to a quality level which meets state/federal drinking water MCL's for the tested heavy metals. Crude cells achieved 30 fold reductions in metals concentrations for Cu2+, Cd2+ and Hg2+ while Pb2+ concentrations were reduced by a factor of >100. At initial metals concentrations of 600 ppb and less, these reductions resulted in the generation of water which met or surpassed federal drinking water standards for Cu2+ (5 ppb vs.

1.3 ppm standard) , Cd2+ (2 ppb vs. 5 ppb standard) and Pb2+ (< 5 ppb vs. 15 ppb standard) while Hg2+ levels were reduced to within 0.5 ppb of the 2 ppb standard.

The relative binding strengths of the various metals to crude cell mass were determined from a matrix of competitive binding and displacement binding experiments. In the competitive binding experiments, fresh cell mass was incubated with mixed solutions of target metals at varying absolute and relative concentrations. These experiments indicated relative binding <BR> <BR> <BR> <BR> strengths on the order of: Cu2+ > Hg2+ > Pb2+ > Cd 2+> Zn 2+. The effects of varying metal concentrations on binding within this group were modest in most cases. When competitions included lower atomic weight metals, i.e., Awl3+, Ca2+, Mg2+, Fe2+ and Fe3+, more complex effects of concentration on binding were noted. In these cases, the lower atomic weight metals impaired the binding of target heavy metals only when present at relatively high concentrations which typically exceeded their normal environmental levels, e.g., Ca2+ and Mg2+ reduced the binding of Hg2+ by only 10-15 % at normal hard water concentrations of - 150 ppm but reduced binding by 30-60 % at 600 ppm concentrations.

To assess the relative displacement potential of target metals, cell mass was preloaded with varying amounts of one metal then incubated with varying amounts of a second metal.

Included in this study were both targeted heavy metals (Cu2+, Cd2+, Hg2+, Pb2+, Zn2+) as well as potentially problematic but environmentally common metals including Ca2+, Mg2+, Awl3+, Fe2+ and Fe3+. These experiments gave results similar to the competitive binding experiments and indicated the following displacement potential: Awl3+ > Fe2+> Cu2+ > Pb2+ > Hg2+ > Ca2+, Mg2+> Cd2+. As in the competitive binding experiments, the effects of A13+ and Fe2+ on the displacement of targeted heavy metals were only important when these metals were present at relatively high concentrations which typically exceeded their normal environmental levels.

Another group of experiments was designed to determine the ability of crude cell mass to be stripped of adsorbed metals and regenerated for reuse. In this series, MHS-3 cell mass was subjected to a total of four metal adsorption/dilute aqueous acid treatment regeneration cycles (10 millimolar hydrochloric acid at ambient temperature). Metals used in this study included the strongly adsorbed Cu2+ ion and the more weakly adsorbed Cd2+ ion. These experiments indicated excellent reuse potential. After an initial loss of 5-10 % of adsorption capacity for both Cu2+ and Cd2+ after the first regeneration cycle, adsorption capacities after each of the next three use/regeneration cycles were within experimental error of each other.

Efforts were made to compare crude MHS-3 cell mass to all available, commercial bioadsorbents. The only material that could be obtained for comparative evaluation was the ForagerTM sponge, a material recently evaluated in the EPA SITE program. These materials were compared on a equal weight basis in the removal of both Cu2+ and Cd2+ at challenge levels between 10 and 200 ppm. While both materials bound Cu2+ well at low concentrations, MHS- 3 cell mass had >2 fold higher capacity at high, saturating concentrations. In the case of Cd2+, MHS-3 bound 2 times as much metal at low concentrations and almost 4 times as much at saturating concentrations.

Crude cell mass can be likened to and resembles a mucoid material. It is hard to contain it and to get a handle on it. A variety of approaches were successful in converting crude MHS-3 cell mass into a more easily handled, processed and used form. The most effective of these approaches involved crosslinking fresh cell mass with bifunctional reagents to create cross-linked granules. Such bifunctional reagents include dialdehydes, halohydrins, aryldiisocyanates, alkyldiisocyanates, and dialkyldimidates . These reagents can be used in association with coreactants such as alkyl diamines, aryl diamines, aminomethylated tannin or lignin (and other Mannich reaction products possessing reactive amino functionality), polyethylenimine, polyols and carbohydrates. Glutaraldehyde-hexanediamine and epichlorohydrin-hexanediamine are preferred combination of reagents and coreactants.

Cross linking with glutaraldehyde-hexanediamine gave a red-brown granular resin with excellent physical properties and capacity for Cu2+ at high concentrations but one which had reduced affinity for Cu2+ at low concentrations and which appears to bind copper so tightly that mild acid or base treatment could not remove it. These data indicate that the use of crosslinking reagents can in some cases bind sites important to high affinity/low metals concentration adsorption. The size of the cross-linked particles is a function of the speed or amount of agitation used on the cross-linking solution. If agitation is conducted slowly larger particles are produced.

Faster rates of agitation produce smaller particles. If no agitation is performed a cross-linked mass is produced. The particles are spherical in shape and average about between 0.5 mm and 3.0 mm in diameter.

While not suited for many applications, such a material can be used where irreversible metals removal is needed. Epichlorohydrin-hexanediamine crosslinking also gave granular, easily handled resin-like materials which provided excellent removal of Cu2+ at both low and high concentrations (6 ppm to 120 ppm). It is crucial to note here that the use of epichlorohydrin diamine-based crosslinking chemistry with MHS-3 cells does not compromise the low metals concentration removal efficacy intrinsic to the MHS-3 cell mass and in fact increases overall capacity (maximum Cu2+ binding is increased to 10 % from the 6 % seen with MHS-3 cell mass alone). Apparently, crosslinking and the introduction of additional free amino groups leads to a significant change in the coordination chemistry of the metal binding sites in these cells. As presented in Figure 8, experiments to evaluate different regeneration strategies confirmed this assumption. Most importantly, these materials could be regenerated to original adsorption capacity by treatment with 1 molar sodium hydroxide solution.

Such crossed-linked material can be loaded into porous cartridges which allow a fluid media to pass there through while retaining the cross-linked product of the invention. The ability to load such cartridges allows one to insert the cartridges in charging and discharging pipes for ease in removing metals from liquid streams flowing through said pipes. Contact with the fluid media can be conducted in batch processes or continuously. Continuous processes can use both up-flow and down-flow reactors as well as a continuous stirred tank.

Reagents, buffers and experimental conditions All reagents were purchased from Sigma Chemical Co. (St. Louis, MO), Aldrich Chemical Co. (Milwaukee, WI) or Fisher Scientific (Pittsburgh, PA). The metals used in all experiments were obtained as high purity salts: All3, CdCl2, MgCl2, MgCl2, CaCl2 2H2O, ZnCl2, Pb(NO3)2, Fe(NO3)3, CuSO4, FeSO4, and K2(Cr207). All solutions were prepared using in deionized, distilled M2O (dd H20), HC1 and HNO3 were metal analysis grade. Acetate buffer was used for pMs between 4.0 and 5.5, 2-(N-morpholino)ethanesulfonic acid (MES) buffer was used for pMs between 5.5 and 6.6 and N-Tris(hydroxymethyl)methyl-2-aminoethanesulfonic acid (TES) buffer was used for pH 7.0.

Atomic Absorption Spectrophotometry (AAS) Low ppm, high ppb analysis of metal content in samples was performed using a Model 460 Perkin-Elmer Atomic Absorption Spectrophotometer equipped with the appropriate Hollow Cathode Lamps and operated in the air/acetylene flame mode. The instrument was calibrated using "Groundwater and Wastewater Pollution Control Check Standards" purchased from SPEX Industries Inc. (Edison, NJ) diluted with the appropriate acid solution.

For the determination of Hg2+, 10 mL of metal solution was mixed with 5 mL of NaBH4 solution (3 % NaBH4 in 1 % NaOH) in a gaseous metal hydride generator attached to the AA spectrophotometer and used for flameless determination of metal concentration.

Samples containing low metals concentrations were sent out for analysis by Furnace Atomic Absorption Spectrophotometer at CT&E Environmental Services Inc. (Baltimore, MD), a laboratory certified by the Environmental Protection Agency (EPA). The samples were analyzed following EPA protocols described in "Methods for Chemical Analysis of Water and Waste" (EPA-600/4-79-020, March 1983): Cu, method EPA220.2; Cd, method EPA213.2; Pb, method EPA239.2; Hg, method EPA245.1.

EXAMPLE 1 Bacterial strains, media and growth conditions.

The marine bacterium strain ATCC No. 43965 (MHS-3) was cultured in full-strength Marine Broth (MB, 37.4 g/L) (Difco Laboratories, Detroit, MI). Isolated colonies were picked from Marine Agar plates (MB, 2 % agar w/v), inoculated into 100 mL of sterile MB and incubated at 25 "C for 96 h. Loopfuls of the copious biofilm that formed on the walls of the culture flask were used to inoculate 100 mL seed cultures, 1 mL aliquots of 4-5 day old seed cultures containing biofilm material were used to inoculate 1 L MB flasks, which were grown at 25 "C for 4-5 days. At this stage, copious flocculation and biofilm formation could be observed in the flasks. To harvest the cultures, biofilm material was removed from the walls of the flask cultures with a rubber policeman, and the broth centrifuged at 16,000-20,000 x g for 15 min. The supernatant was discarded and the pellets processed for metal scavenging experiments or for crosslinking.

EXAMPLE 2 Cell preparation.

Harvested MHS-3 cells/biofilm pellets (see procedure described in EXAMPLE 1 above) were resuspended in 10 mM citric acid (pH 2.0) + 3 % NaCl with a Brinkman Homogenizer for 10 seconds. to remove metals acquired from Marine Broth from binding sites in the material.

The material was centrifuged, the supernatant discarded, and the pellet resuspended in the appropriate 1 mM buffer to a concentration of 10 mg (dry wt)/mL working buffer (the dry wt of MHS-3 cells is 15 % of their wet weight; cell suspensions were prepared using 66 mg wet weight cells/mL) to prepare the MHS-3 cell stock suspensions.

EXAMPLE 3 Metal binding studies reaction set-up.

Eppendorf tubes were used to set up metal binding experiments. The tubes held a volume of a 1500 L, plus 200 L capacity in the lid, which provided a 1700 L working volume.

Depending on the pH requirements of particular experiments, different buffer solutions were used, all prepared as 10 mM stocks, diluted to 1 mM working concentration. The volumes of metal stock solution, MHS-3 stock suspension and dd H2O were varied according to experimental requirements. Most experiments were set up as described below: Metal concentration 10 g (5.9 ppm) 50 g (29.4 ppm) 200 g (117.6 ppm) Metal stock soln. 100 L (100 ppm) 500 L (100 ppm) 200 L (1000 ppm) Buffer (10 mM) 170 L 170 L 170 L MHS-3 stock susp. 100 L 100 L 100 L ddH2O 1330 L 930 L 1230 L Typically, tubes were incubated for 24 h. at room temperature on an platform shaker. The tubes were centrifuged at 13000 rpms for 10 min in a bench top microfuge, 1500 L of supernatant harvested, mixed with 1500 L of acid (1.6 % HCl or 3 % HNO3) and stored for AAS analysis (sample dilutions were prepared as required in the appropriate acid solution). In samples processed for the detection of mercury binding, the final volume of the samples was adjusted to 50 mL. For each reaction vial, 2-5 experimental replicates were tested.

Metal binding Calculations for reaction chamber set ups: <BR> <BR> <BR> <BR> <BR> <BR> Me,, = Me"ed - ([Meharvested] x VOlsted) - ([Me,,="es,edd x Volnot esled) where the volumes harvested were 1500 L, and the volumes not harvested were 200 L in most cases.

EXAMPLE 4 Effects of pH on metals adsorption.

To characterize the effects of pH on metal binding behavior of crude MHS-3 cell mass, reactions chambers were set up to test a range of pHs from 4.4 to 6.6, with 0.2 pH unit intervals.

Stock buffer solutions (10 mM) from pH 4.4 to 5.4 were prepared with acetate buffer, and from pH 5.6 to 6.6 with MES buffer. The experiments were performed as described above. Metals tested include: Cu2+, Cd2+, Hg2+, Pb2+, Zn2+ and Cr3+. The data are summarized in Figure 1 which clearly demonstrate the metal binding capacity of MHS-3 across a wide range of pH values.

EXAMPLE 5 Metal binding capacity.

The maximal metals binding capacity of crude MHS-3 cell mass was determined from reaction chamber incubation data obtained from experiments carried out on cells at pH 6.0 with 200 µg amounts of metals according to the procedures described in the reaction chamber setup section above. The metals used in this experiment included: Pb2+, Hg2+, Cu2+, Fe2+, Fe3+, Cd2+, Cr3+, Zn2+, and Cr(VI) as (Cr2O7)2-. The data defining metal binding capacities of MHS-3 are summarized in Figure 2.

EXAMPLE 6 Time course of metals adsorption.

The amount of time required by the MHS-3 material to bind the metal present in the reaction chamber was determined by setting reaction chambers and harvesting them at 30 minute intervals. The supernatants were processed for AAS analysis, and these data used to determine when the metal binding reaction had reached equilibrium. Metals tested include Cu2+ and Cd2+.

EXAMPLE 7 Comparative adsorption studies.

The MHS-3 material metal binding capacity was compared to ForagerTM sponge material, a commercially available metal chelating material marketed by Dynaphore, Inc. (Richmond, VA). The material (1 mg dry wt) was placed in reaction chambers plus 100 L dd H2O; the amounts of metal and buffer solutions and additional dd M2O were as described previously. Cu2+ and Cd2+ were the metals tested in this study. The data which consistently demonstrate higher binding capacity of MHS-3 material are summarized in Figure 3.

EXAMPLE 8 Metal competition and displacement experiments.

Metal binding competition experiments were carried out using the mixed chamber set ups as described previously, but a second, competing metal was added to the tubes, and the volumes adjusted to 1700 L with dd H2O. The absolute amounts of most competing metals used were 10, 50, and 200 g. In the case of calcium and magnesium, the amounts added were 10, 100, 260 and 1000 g. The 260 g of Ca2+ in a volume of 1700 L is equal to 153 ppm of calcium, which is equal to the Ca2+ concentration in Synthetic Chicago Tap Water (SCTW), a water treatment industry standard for moderately hard water. The tubes were incubated and harvested as described previously.

For the metal displacement experiments, the reaction chambers were set up first with one metal, incubated and harvested. The supernatant was then mixed with acid and stored for AAS analysis (this allowed determination of the total amount of the first metal bound by the MHS-3 material). The MHS-3 pellets were washed with buffer, and used to set up second reaction chambers containing the second metal ion solution then incubated overnight and harvested. The supernatant was diluted in acid and stored for AAS analysis in order to determine how much of the first metal bound to MHS-3 was released by exposure to various concentrations of the second metal. Zn2+, Pb2+, Cu2+, Cd2+, Hg2+, Al3+, Fe2+, Fe3+, Ca2+ and Mg2+ were used in various binary combinations in these studies. Select data are summarized in Figures 4 and 5.

EXAMPLE 9 Metal binding at low concentration.

To determine the efficiency of removal of low concentrations of metals to levels at or below EPA Drinking Water Standards by the MHS-3 material, reaction chambers were set up in 50 mL conical tubes, the initial concentrations of metal tested were 5.88 ppm, 588 ppb, 294 ppb and 58.8 ppb; 0.59 mg/mL dry wt of MHS-3 material was used. For Cu, Cd and Pb, 17 mL reaction chambers were set up and incubated overnight, 15 mL of supernatant were harvested and mixed with 15 mL of 2 % HNO3; for Hug, 25.5 mL reaction chambers were set up, 25 mL harvested and mixed with 25 mL 2 % MONO3. Hg2+, Pb2+, Cu2+ and Cd2+ were the metals used in this study. The data are summarized in Figure 6.

EXAMPLE 10 Crosslinking of MHS-3 cells using glutaraldehyde and hexanediamine.

Briefly, 18 mL of 1 % hexanediamine solution (w/w, in 1:1 hexane/chloroform) were placed in a 50 mL Erlenmeyer flask, under vigorous vortexing, and 4 mL of a glutaraldehyde- treated, concentrated MHS-3 stock suspension (40 mg dry wt/mL, 8 % glutaraldehyde) were added dropwise to the flask. After all the cell suspension had been added, the mixture was vortexed for 5 min (at this point there was no longer a discernible aqueous phase) and allowed to cure overnight at room temperature. The organic solvent was evaporated, the "wet' material weighed (a fraction was dried and used to determine yields), washed with dd H2O and 10 mM MES buffer (pH 6.0), and used in metal binding experiments. The reaction chambers were set up in Eppendorf tubes as described earlier with 15 mg of wet crosslinked material plus 100 L of dd H2O. Cu2+ was the metal used to assess binding in this study.

EXAMPLE 11 Crosslinking of MHS-3 cells using epichlorohydrin and hexanediamine About 2.4 g of cells (wet weight) were dispersed in 20 mL of 1 N NaOH (Brinkman Homogenizer PT 10/35, dial power setting 4 for 10 s) and transferred to a 250 mL round-bottom flask to which 20 mL of epichlorohydrin was added. The reaction was performed in a heated waterbath at 65 "C with the flask rotating at 200 rpm for 3 h. Treated cells were removed from the reaction mixture by centrifugation (3000 g, 5 "C, 10 min) and washed four times with 40 mL 1 N NaOH. Cells were resuspended in 10 mL of 0.1 M potassium phosphate buffer at pH 6.3 and added to a thoroughly agitated solution of 5 g hexanediamine in 50 mL acetone. Due to the reduced polarity of the solvent, cells aggregated readily to particles of 0.5 to 3 mm diameter, which greatly facilitated the crosslinking reaction. After an incubation time of 24 h at ambient temperature, the resulting particulate material was washed with 10 mL 1 mM MES buffer pH 6.0, freed from excess buffer on a disc of filter paper (Whatman No. 2), and used for metal binding studies. Cu2+ was the metal used to assess binding in this study.

EXAMPLE 12 Regeneration of the metal-binding capacity of MHS-3 cells.

To test the regenerability of the MHS-3 material, reaction chambers were set up and cells incubated, using standard procedures, with 10 and 200 pg Cu2+ (the strongest binding target metal) for 1 h. Cells were then removed by centrifugation and the supernatants processed for AAS analysis. The MHS-3 pellets were resuspended in 1500 L of 10 mM HCl (pH 2.0) for 1-2 min and centrifuged, the supernatant (containing the released metal) was set aside. The MHS-3 pellet was washed in 1 mM pH 6.0 MES buffer for about 1 h and centrifuged again, and then used to set up subsequent metal binding experiments. This regeneration process was repeated four times using the same MHS-3 pellets. The metal used for this study was Cu2+. The data are summarized in Figure 7.

EXAMPLE 13 Regeneration of the metal-binding capacity crosslinked cells.

Glutaraldehyde/hexanediamine: Although the glutaraldehyde-hexanediamine treated cells exhibited good metal binding properties, this material could not be regenerated using 10 mM, 100 mM or 0.5 M HCl, or 1 M NaOH. The metal appeared to be tightly bound to the matrix probably by the same kinds of forces and interactions responsible for the tight binding seen in [Cu2+(gly2j.

Epichlorohydrin/hexanediamine: Figure 8. shows an experiment to evaluate different regeneration strategies. Only about 60 % of the initial binding performance at 10 ,ug Cu and about 30 % at 200 ,ug Cu could be regenerated by using 10 mM HCl, a treatment that led to full restoration of the initial performance using crude cells. The use of 100 mM HCl (pH 1.0) improved regeneration success to about 80 % of initial capacity for 10 pg Cu and 40 % for 200 llg Cu. A pH of 0 (1 M HCl), however, led to an almost complete breakdown of the binding performance, suggesting that at this pH chemical degradation ofthe matrix and EPS material may occur. On the other hand, treatment with 1 N NaOH before or after acid treatment led to 100 % regeneration of the low concentration removal performance and improved the performance at high concentrations significantly. Interestingly, skipping the acid step and regenerating with NaOH alone led to full, overall regeneration of the initial binding performance of the material which is both significant support for the hypothesis of altered coordination chemistry in this material, and also important for the potential commercial use of this crosslinked form of metal adsorbent. The data are summarized in Figure 8.

While the invention has been described with reference to a preferred embodiments, it will be understood by those skilled in the art that various changes may be made and equivalents may be substituted for elements thereof without departing from the scope of the invention. In addition, many modifications may be made to adapt a particular situation or material to the teachings of the invention without departing from the essential scope of the invention. For instance in addition to metals disclosed in the embodiments discussed above, all of the following metals can be removed using the compositions of the invention Ag, Al, As, Au, Ba, Cd, Co, Cr, Cs, Cu, Fe, Hg, Mn, Mo, Ni, Pb, Pu, Sb, Se, Sn, Sr, Tc, Th, Ti, Tl, U, V and Zn. Metals in the following oxidation states can also be removed: Ag(I), Al(III), As(III),(V), Au(III), Ba(II), Cd(II), Co(II), Cr(III),(VI), Cs(I), Cu(II), Fe(II),(III), Mg(II), Mn(IV),(VII), Mo(VI), Ni(II), Pb(II), Pu(IV), Sb(III),(V), Se(IV),(VI), Sn(II),(IV), Sr(II), Tc(VII), Th(IV), Ti(IV), Tl(I).(III), U(VI), V(V), and Zn(II). As well as metals in the following ionic states: Ag+, A13+, As3+'5+, Au3+, Ba2+, Cd, Co2+, Cur3+6+, Cs1+, Cu2+, Foe2+3+, Hg2+ My4+'7+, Mo6+, Ni2+, Pb2+, Pu4+, Sub3+5+ Sue4+'6+, Sun2+'4+, Sr2+, Tc7+, Th4+, Ti4+, To+ 3+, U6+, V3+, and Zn2 Therefore, it is intended that the invention not be limited to the particular embodiment disclosed as the best mode contemplated for carrying out this invention, but the invention will include all embodiments and equivalents falling within the scope of the amended Claims.