May 25, 2007

Summary

This paper summarizes a Battelle report, “Self-Assembled Mercaptan on Mesoporous Supports” (SAMMS®) Technology for Mercury Removal and Stabilization,” PNNL-11691 UC-2030, Sept. 1997. The report discusses regeneration of thiol-SAMMS and shows SAMMS survives at least 2 regeneration cycles with 100% mercury recovery and steady 50% mercury loading capacity relative to the virgin capacity. Regeneration uses 37% (wt.) HCl. The experimental work was conducted at bench scale as a demonstration. The process has not yet been optimized such as by using reducing agents or other steps which would tend to increase the regenerated material’s loading capacity above 50%. To demonstrate complete manufacturing effectiveness and cost benefit analysis of the SAMMS™ materials, Steward is providing this preliminary regeneration procedure to GE with the request that it not be considered as the final tested and economically optimized procedure for regeneration.

Introduction

The Thiol-SAMMS product manufactured by Steward Environmental Solutions, LLC is a higher quality SAMMS than was used in the Battelle report cited above. SAMMS manufactured by Steward has higher surface coverage of silane and thiol groups and can be expected to yield the same or better performance than reported by Battelle. Because the thiol groups are directly attached to the silica substrate, SAMMS does not behave in the same manner as polymer thiol-impregnated polymeric resins such as GT-73, nor is it subject to the pH limitations and swelling when used in liquids.

Details of Battelle Report on Regeneration of SAMMS

The SAMMS used in the early Battelle work was prepared by wet chemical methods described in one of the patents on this technology. Duplicate samples of SAMMS with about 80% surface coverage of silane tethered thiol groups were tested. In each case the contacting solution was aqueous 0.1M NaNO3 containing 330 mg/L of Hg (330 ppmw). The sample of SAMMS was stirred for four hours to saturate the thiol groups, then filtered out and dried. The mercury concentration of the contacting solution was measured again. The method of measuring mercury is not stated but is almost certainly CVAA. Capacity is calculated using the starting and finishing concentrations and the known weight of SAMMS.

The SAMMS material was regenerated by holding the spent SAMMS in highly concentrated HCl (37% by weight) for four hours. The SAMMS was removed by filtration and the concentration of mercury in the regeneration solution measured. Using the known weight of SAMMS and its mercury content, the percent recovery was calculated.

These two steps were repeated for three regeneration cycles. A second experiment attempted to strip mercury using a 2N HCl solution using the same procedure.

Results

1. Initially 0.5163g of SAMMS#2 (~80% surface coverage with thiol) was added to 1 liter of the spiked NaNO3 solution. After recovery of the SAMMS and measuring the final concentration of the contacting solution, the mercury loading was calculated to be 445 mg/g.
2. First regeneration. A sample (0.2g) of the recovered and dried SAMMS was added to 10ml aqueous solution of high purity HCL (37% HCl by weight) and held for four hours. The SAMMS was separated by filtration and the mercury concentration of the remaining solution was determined to be the concentration corresponding to 100% removal of the adsorbed mercury. No dissolved silica was measured indicating no attack of the silica support.
3. A sample of the regenerated SAMMS in (2) was used to test effectiveness of the regenerated adsorbent. A sample (0.075g) of regenerated SAMMS was placed in 150ml of a mercury solution, agitated and tested as in (1). The resultant concentration difference yielded a calculated mercury capacity of 0.210 mg/g, approximately 50% of the capacity of virgin SAMMS.
4. Second regeneration. The SAMMS of (3) contacted as in (1) and then regenerated as done in (2) and tested as in (1). The resultant concentration yielded a calculated 100% recovery of the adsorbed mercury.
5. The second regenerated sample of SAMMS was recovered and regenerated as in (2) and a sample (0.069g) was added to 137 ml of a mercury solution, agitated and tested as in (1). The capacity of the second regeneration SAMMS was calculated to be 233 mg/g, again about 50% of virgin capacity.
6. A test at lower HCL concentration was conducted. A sample (0.2g) of the recovered and dried SAMMS was added to 20ml aqueous solution of high purity 2 N HCl and held for four hours. The SAMMS was separated by filtration and the mercury concentration of the remaining solution was determined to be the concentration corresponding to <14% removal of the adsorbed mercury.

Table 1. Summary of Results

Conclusion

This investigation was conducted at the bench scale. It showed that “first generation” SAMMS survived two regeneration cycles, recovering 100% of adsorbed mercury each cycle and recovery approximately constant capacity. The adsorbent capacity for mercury dropped about 50% after the first regeneration cycle but did not change afterwards. The capacity of the material used in the third regeneration recovered 100% of the capacity in the second cycle. This work shows that SAMMS can survive at least two cycles of regeneration with 100% recovery of mercury. It also shows lower concentration of 2N HCl is not very effective.

Steward notes that even with a capacity of 0.2 to 0.3 g Hg/g SAMMS, this regenerated material is significantly larger than competitive adsorbents. The study also suggests that the capacity could likely be recovered to near 100% of the virgin capacity by using reducing treatments to recondition the thiol groups. Steward has planned experimental work to optimize the regeneration process.

Steward also notes our process for making the Thiol-SAMMS achieves higher surface functionalization than the materials of the referenced Battelle report. SAMMS may show improved performance compared to the SAMMS in the Battelle report. Steward has reason to believe SAMMS may achieve full regeneration to 100% recovery using other agents.