Mercury Work Group
Phase II Reports >> Hg Management Guidebook

 
APPENDIX F

CASE STUDIES

Case Study Overview Industry Reference

  1. A semiconductor manufacturer investigates A, E mercury in process wastewater

  2. An electronics product main component contains A mercury as a primary ingredient

  3. A hospital investigation of mercury sources in a D pathology laboratory

  4. A battery manufacturer investigation of mercury C in process wastewater

  5. A hospital incinerator scrubber wastewater mercury B, D, E control process

  6. A bulb/lamp manufacturer cleaning operation causes A mercury in wastewater A clinical testing laboratory achieves compliance through D, E source reduction and pretreatment

INDUSTRY REFERENCE CODES:

Electronics - A
Incineration/Solid Waste - B
Battery Manufacturer - C
Hospital/Medical - D
Wastewater Pretreatment - E

Disclaimer

These case studies are intended only for illustrative purposes to show by example the ways that several industries or institutions addressed specific mercury compliance issues. The case studies are not intended to be endorsements of particular courses of action or recommendations of specific effective versus ineffective practices. Similarly, any references to outside companies or vendors of either products or pretreatment technologies are not intended to be endorsements by the Work Group, the Subcommittee, the Subgroup, or any of its participating persons and institutions, including MASCO and the MWRA.

Since all facilities and situations are unique, individual courses of action should always be developed as pertinent and needed to achieve compliance.

CASE STUDY 1 - Semiconductor Manufacturer

Introduction
A manufacturer conducts research and development of advanced electronic materials and semiconductors where mercury in wastewater was traced to a chemical vapor deposition process on a mercury-cadmium-tellurium substrate and to the "wet" operations of a laboratory in the facility. A program for identifying mercury sources, evaluating process substitution, cleaning infrastructure systems, and examining various pretreatment technologies was carried out at the site. While the deposition process itself had been discontinued for some time, infrastructure residuals continued to affect compliance with the MWRA industrial wastewater discharge prohibition for mercury.

Laboratory wastewater flowed to a central industrial wastewater pretreatment system that originally included only neutralization. The system was modified to include metals removal and filtration of suspended solids. The facility now uses off-site disposal for the wastewater from one process that remained a problem. Overall, daily sewer discharges were reduced from 3,000 to 2,000 gallons per day.

Description of the Problem
Before installation of a mercury pretreatment system, the levels of mercury were measured between 50-100 µg/L (ppb). The facility was issued a cease-and-desist order by MWRA. In response, an investigation of the sources of mercury that were affecting wastewater discharge was conducted. Once the sources were identified, options were evaluated for reducing the mercury discharged to the wastewater pretreatment plant, removing residual mercury within the plant piping infrastructure, and pretreating mercury-bearing wastewater before discharge to the MWRA sewer system.

Methods Investigated/Implemented
The program to obtain compliance with the MWRA mercury prohibition was organized into the following areas: problem confirmation and data collection; awareness and management of the issue; reduction of mercury-containing wastewater; substitution for mercury-containing process chemicals; assessment of mercury introduction via equipment; decontamination of piping system infrastructure; and assessing wastewater pretreatment technologies.

As a first step, the wastewater discharge was resampled several times to confirm that the measured elevated level of mercury was not due to some one-time abnormal event. Occasionally, analytical test result variability exceeded 40 percent, affecting whether or not compliance was achieved. An inter-laboratory study of split samples sent to different DEP-certified laboratories suggested that some labs generated higher results than others. Once, the cleanliness of the vendor's automatic sampling device was called into question when analytical test results for samples collected by the vendor were higher those for samples collected by the facility that had cleaned its automatic sampling device before use.

Employee awareness played a critical role in controlling the discharge of certain sources of mercury into the sewer system. A meeting of all employees was held by the plant manager to emphasize the serious nature of the cease-and-desist order and to stress the need for adherence to the mercury reduction program. To reinforce the message, signs were posted above laboratory sinks noting that chemicals were not to be discharged down sink drains. In addition, training given to new employees was revised to address the issue of mercury in wastewater.

Once the problem was confirmed, the discharge was rerouted for several months from the sewer to a holding tank for off-site disposal until an engineering solution could be installed. Roughly 100,000 gallons of wastewater were collected and disposed of at a cost of $ 75,000. All "wet" processes were examined to reduce water usage and discharge to absolute minimum amounts. This was accomplished largely by the following:

  • C adding flow restrictors on each lab sink water supply,

  • C installing still/dragout rinses after the acid clean step and before the running rinse that went to the collection tank, and

  • C using still/dragout rinses as makeup to the process chemical baths.

  • Water supply, various process chemicals, and "wet" process equipment were investigated to see if they could be sources of mercury. Several test results on the incoming water between 1991 and 1994 showed that this was not a contributing source. Few mercury sources were found in process chemicals since the laboratories only used "ultrapure" semiconductor grade chemicals. However, the wastewater treatment chemicals (sodium hydroxide and bleach) were found to have up to 1.0 ppm of mercury. These mercury sources were addressed by switching from standard grade to "membrane" grade chemicals, which have mercury below 50 µg/L (ppb), at an added cost of less than $10 a drum.

    The examination of equipment as potential mercury sources focused on the exhaust ventilation and fume scrubber system. The small volumes of condensate from drip collection lines in ventilation systems were found to be high concentration sources of mercury. Inside surfaces of the exhaust fan were cleaned after they were found to have residues with very high levels of mercury. Open top wastewater treatment tanks were covered to prevent dust and other solids from the sludge filter press operation from entering the system.

    A key aspect of the mercury issue was removal of residual mercury from the facility waste piping infrastructure. Sink traps in known mercury-containing areas were removed and replaced, whereas all other sink traps in "wet" areas were dismantled and cleaned. The entire industrial waste piping system was flushed and cleaned in two separate events. The piping system was first flushed with water. Wastewater generated from the flushing activities was directed to a collection tank for offsite disposal.

    After the water flushing, an emergency response contractor familiar with handling acid chemicals was retained to flush and clean the piping by filling the piping system with 20-30 percent nitric acid for 12 hours. The cost of this activity was roughly $10,000. However, the high unit weight of the acid solution caused leaks in some gravity piping. (Note: The use of acid is not recommended for waste piping cleaning because of handling hazards, corrosion potential, waste disposal issues, and cost. Refer to Appendix B for information on recommended waste pipe cleaning procedures and associated cautions.)

    Since there was a general perception that source reduction alone would not reduce mercury levels to the MWRA prohibition of less than 1.0 µg/L (ppb), an engineering firm was hired to evaluate pretreatment options. The technologies evaluated were ion exchange, sulfide precipitation, evaporation, and microfiltration. One technology alone would apparently not consistently meet the MWRA mercury prohibition. The engineering evaluation recommended modification of the pretreatment system with the addition of sulfide precipitation followed by an ultrafilter (to remove small suspended particulate) and activated carbon (to reduce remaining levels of dissolved mercury). Capital cost of the system for 3,000 gallons per day was roughly $150,000.

    Initially, the ultrafilter achieved excellent results but, as the unit became clogged and required cleaning, subsequent performance levels decreased. Waste liquid (concentrate) from the ultrafilter was collected with filter press sludge for off-site disposal. Influent and effluent samples from the activated carbon unit did not show any appreciable removal of mercury.1

    Although it did achieve compliance a significant portion of the time, the pretreatment system could not consistently meet the MWRA limit of 1.0 µg/L (ppb). Repeated efforts at minor adjustment of system control set points and routine maintenance were unsuccessful in bringing the system to a point of continuous compliance. Finally, the waste stream from the mercury deposition process was segregated for alternate disposal to eliminate that stream from the discharge to the sewer.

    Conclusion
    The problem of removing mercury from wastewater to a level of 1.0 µg/L (ppb) was a difficult one and required that the problem be approached from several angles. Although wastewater pretreatment and fume scrubber chemicals were contributing mercury to the discharge, the main source of mercury was the water rinse after acid cleaning of equipment from the known mercury-containing process. The waste acid from the cleaning process was collected and managed for off-site disposal.

    Several rounds of analytical data confirmed that mercury was present above the regulated limit and that the existing wastewater pretreatment system alone could not remove the mercury. Because consistent results could not be obtained with one technology, a combination of technologies was needed. Between 1991 - 1994, overall expenditures to achieve compliance was between $400,000 and $600,000. Since 1995, mercury concentrations in the sewer discharge from the facility have been between nondetectable (less than 0.2 µg/L) and 3 µg/L (ppb). Recent performance information has not been examined.

    A key aspect of managing the issue was employee awareness, so that the staff members could know the consequences of their actions. Attempts at reducing the mercury in materials and chemicals used also yielded significant improvement. Additionally, cleaning of the waste piping infrastructure (subject to the recommendations and cautions in Appendix B) was a critical action because residual mercury within the waste piping system had caused additional violations.

    CASE STUDY 2 - Electronics Manufacturer

    Introduction
    This case study documents the activities performed by an electronics manufacturer in its attempt to reduce/eliminate its mercury discharge to the MWRA sewer system. This facility is engaged in the development, design and low volume production of Electro-Optical Systems for Infrared Sensors and Seekers. A main component of the product is a mercury-cadmium-tellurium wafer.

    Mercury Reduction/Elimination Activities
    Sampling
    Various sample points showed mercury levels from 0.2 to 75 µg/L (ppb). Mercury-contaminated debris consisted of Chem-wipes, Q-tips, cotton swabs, protective gloves and glassware. The sodium hydroxide (NaOH) used in the manufacturing process was found to have a mercury content of 0.005 mg/l (5 µg/L or 5 ppb). Although this mercury content was fairly low, the facility eliminated the use of this product to help reduce the mercury found in the facility discharge.

    Sampling was done at the discharge point of potential sources, rather than at the end-of-pipe, to identify actual mercury sources throughout the facility. In addition, a baseline of sample data was developed (at the end-of-pipe) to characterize the discharge accurately.

    Formation of Environmental Health and Safety Team
    An environmental health and safety compliance team was assembled to meet biweekly to discuss and actively address applicable issues. The team serves as the platform for internal training and awareness to lab operations. The team simplified the exchange of information needed for process change and modification to achieve compliance with discharge standards.

    Collection of Wastes
    All lab procedures require that all spent chemicals, such as organics and acids, be collected. All process containers and beakers are rinsed with DI water and have the first and second rinses collected and stored for hazardous waste disposal. Mercury-contaminated debris is separated in the labs satellite accumulation area and later compacted with a specific hazardous waste compactor. Quartz tube cleaning and the photo array process in the Research and Development area have been disconnected from the drain. Currently all water from these operations is collected for off-site disposal.

    Water Use Reduction
    Water use was reduced in several ways. Filter saw wastewater is recirculated back to mercury wafer dicing equipment. The water in the quartz dicing and sawing research and development operation is also recirculated, using a closed loop filter system. All nonessential process sinks in the research and development area have been eliminated. The facility is currently seeking a water recycling filter media. DI water will be used for makeup.

    Consolidation of Discharge Points
    All wastewater discharges were piped to two 3,200 gallon wastewater holding tanks. Following MA-DEP permitting requirements for offsite disposal of the wastewater, both are equipped with alarms that sound when the tanks reach 75 percent of capacity. In addition, both tanks are plumbed in series to the MWRA system. The pH of the tank contents is monitored by the facility. Before discharging to the sewer system, the facility analyzes the wastewater for all pollutants required to be analyzed by its Sewer Use Discharge Permit. If the waste stream meets all MWRA limits, the contents of the tank are released to the sewer system. If it is found that any MWRA limits are not met, the waste stream is hauled offsite according to the MA-DEP permit.

    Pretreatment
    Currently no pretreatment is done at this facility. The facility is currently doing bench tests on a pretreatment system that will treat for Hg, Cd and Te in the following waste streams, which are currently collected: a) quartz glassware etch rinse, and b) lapping, polishing and dicing.

    The pretreatment system under investigation includes: pH neutralization, carbon treatment for organics and a 3-stage ion-exchange resin canister for the removal of Hg, Cd and Te.

    Summary
    There is now a better understanding of the sources of mercury in the facility. The wastewater discharged to the sewer system has been reduced to a manageable level. The facility now has "control" of what pollutants it allows to discharge to the sewer system. The last analysis of wastewater showed that the facility was in compliance.

    CASE STUDY 3 - Hospital Pathology Laboratory

    First all chemicals used in the Pathology Department were identified and a database was established using MSDS information. In the database, each chemical was listed along with the following information:

    • quantity discharged, and
    • mercury content (provided by manufacturer or by analytical testing).

    Levels of compliance with hazardous waste disposal policies and regulations were tightened through employee education and posted disposal listings at laboratory benches for all waste reagents.

    It was decided that in all areas where chemical reagents were used, analytical evaluations were necessary. (For example, a grab sample of one reagent showed mercury at 42 µg/L (ppb) or 0.042 mg/L (ppm) - a "noncompliant" level). In retrospect, sampling and analysis should have been performed for all benches and waste streams in the laboratory - not just in "chemical reagent areas."

    To account for wastes as sources, all waste materials were segregated by department and screened for "hidden" mercury:

    • Large drums were deployed for collection of all waste in all departments. Even if the waste sample was initially considered free of mercury, aliquots were tested for mercury.
    • The drums were used to eliminate the waste piping as a possible source of mercury contamination.
    • In retrospect, an ideal situation would have been a laboratory plumbed to allow segregation of individual department waste streams.

    (Winter 1993 to Spring 1994)
    After the analytical results were in, all positive bulk streams were segregated by bench within each department. Collection stations were placed at every bench and representative samples were taken and submitted for analysis. The bench areas where mercury was found were identified and a thorough evaluation was done to find its sources. MSDSs were examined, manufacturers were questioned, and all suspect reagents were analytically tested.

    The goal was to work with all the pathologists to eliminate mercury compounds in the laboratory where possible. The results of initial testing are shown in the following table.

     

    Laboratory

     

    Bench

    Measured Hg Concentration
    (µg/L or ppb)

    Chemistry

    TDX Benches

    < 5

    Chemistry

    IMX Benches

    < 1

    Chemistry

    Array

    < 2

    Chemistry

    Flex Benches

    < 5

    Chemistry

    E4A Bench

    < 2

    Chemistry

    Specials Bench (Electrophoresis)

    8,800 2

    Chemistry

    Plating Solution (leads)

    600

    Cytology

    EOSIN Stain

    < 2

    Cytology

    OG-6 Stain

    < 5

    Bacteriology

    CSpor

    10,000

    Bacteriology

    Hepatitis (Abbott)

    150

    Bacteriology

    Probes

    2

    Bacteriology

    Methylene Blue

    2

    Bacteriology

    Basic Fuchsin

    2

    Bacteriology

    Iodine

    BQL 3

    Bacteriology

    Crystal Violet

    BQL

    Bacteriology

    Auramine-Rhodamine

    BQL

    Bacteriology

    Parasitology

    25,000

     

    Hematology

    Coagulation (MLA + Dade Reagents)

    BQL

    Hematology

    IRIS

    BQL

    Hematology

    Technicon H1

    BQL

    Hematology

    Hematoxylin

    BQL

    Hematology

    Coulter T890

    BQL

    Hematology

    Semen Analysis (Hematoxylin)

    40

    Hematology

    Naphthol and NAP AS-D

    BQL

    (January 1995)
    The histology department eliminated mercury-containing fixatives for tissue processing (some brands of hematoxylin are mercury-free), and a process was established to screen incoming products carefully for the presence of mercury. Results of testing all benches in the histology department showed that all wastes collected and tested were below the quantitation limit (BQL) with the detection limit at 0.5 µg/L. Results from follow-up testing of the 8,800 µg/L mercury level found in the electrophoresis waste were as follows:

    Electrophoresis Reagents

    Reagent

    Measured Hg Concentration
    (µg/L or ppb)

    HGBA1C Buffer

    28

    IFE Buffer

    200

    Electrophoresis Stain

    1,800

    These results could not explain the 8,800 µg/L mercury level found in electrophoresis waste. Further evaluation led to the blood bank saline and blood bank reagents. Results from testing of blood bank saline and reagents and candidate alternative products were as follows:

    Blood Bank Reagents

    Reagent

    Measured Hg Concentration
    (µg/L or ppb)

    Reagent Mix

    14,300

    Waste from Drain System

    5,800

    Dade Immusal (saline)

    44,200

    Dade Certified Blood Bank Saline

    29

    0.9% Sterile Saline Bags

    BQL

    Immu Add (LISS)

    206

     

    Candidate Alternative Reagents

    Reagent

    Measured Hg Concentration
    (µg/L or ppb)

    Ortho Antibody Enhancement

    70

    Gamma N-Hance

    < 0.5

    BCA EM-X

    138

    As a result of the analytical testing effort, the blood bank switched to the saline with the lowest amount of mercury, continued to seek an alternative, and replaced the Liss reagent with Gamma N-Hance.

    (March 1995)
    Although not in compliance with the MWRA discharge enforcement limit of 1.0 µg/L (ppb) at all times, significant strides were made in reducing discharge mercury concentrations.

    • All laboratory waste was still being collected for offsite disposal.
    • All new reagents and waste streams were tested for mercury, even for reagents whose MSDSs stated no mercury was present.
    • Alternative reagents were sought for the blood bank and microbiology.
    • The hospital established a hospital-wide mercury policy.
    • Elimination and reduction of sources were highly successful means of bringing processes under control.

    Source reduction techniques used were:

    Input Substitution - Several sections of the Pathology Laboratory converted to mercury-free reagents after lengthy clinical trials.

    Operating and Maintenance Procedures - A Standard Operating Procedure was developed and implemented for reducing mercury use and discharge at the facility.

    (1997)
    Goals for the year in the Pathology Department included continued work toward elimination of mercury and a reduction in the use of toxic chemicals. Toxic use reduction was accomplished by:

    1. Reduction in chemical purchases and improved inventory control using the Meditech Chemical Database.
    2. Purchase of equipment that used small reagent volumes and generated less hazardous waste.
    3. Elimination of Hg PVA in Microbiology by replacement with a low Zn PVA system.
    4. Replacement of manual staining procedures in Histology with kits and elimination of dry chemicals and toxic concentrated stains.
    5. Increased levels of employee training in hazardous materials management and mercury source reduction.

    At the end of the year, the amount of chemicals nearing expiration dates were cut in half from 1996 levels. Safety inspection reports showed that staff members had increased knowledge of chemical handling and spill procedures.

    Notes on the Histology Laboratory:
    It was not possible to replace all manual staining procedures in the Histology Laboratory with kits because of the unavailability of kits for certain staining procedures. As kits become available, the manual staining procedures will be eliminated.

    Contamination of equipment in the Histology Laboratory also became an issue. Although the Laboratory had eliminated the use of Hg-containing reagents, measurable amounts were still being discharged from the area. It was discovered that the water from the flotation water baths (microtome stations) and VIP processor was contaminated with mercury. Therefore, all flotation baths, VIP filters, baskets and tubing in the processor were replaced to rectify this problem.

    (1998)
    Non-hazardous liquid laboratory waste volumes decreased this year by about 11 percent although the workload increased by as much as 21 percent. Hazardous waste drums shipments showed no change from 1997 though the workload in Histology alone showed an increase of 23 percent. Currently, formaldehyde and xylene recycling systems are being evaluated under criteria that include minimal employee exposures and waste generation.

    CASE STUDY 4 - Battery Manufacturer

    Mercury was a common constituent in some types of dry cell batteries. A manufacturer in New England eliminated mercury from in their batteries in 1988 because of growing concern over its hazardous nature. The mercury content of the batteries was steadily reduced over several years. After the use of mercury was ended, the facility went on to conduct a thorough clean-out to become truly "mercury-free."

    Manufacturing activities included chemical mixing, battery assembly, storage, and shipping, and wastewater pretreatment. Mercury was used in the chemical mix area as a component of the battery slurry. In July 1988, the facility conducted a cleanup effort using an outside contractor to remove all residual mercury from the chemical mix area. Cleanup involved thorough cleaning out of all wastewater pretreatment system sumps and trenches, repiping of all pretreatment system sources, treatment and cleaning of grossly contaminated surfaces and equipment, and a final steam cleaning.

    In detail, steps were:

    • Pretreatment sumps and trenches were acid washed (hydrochloric)
    • All piping to the pretreatment facility replaced
    • Noticeable gross elemental mercury: vacuumed using mercury vacuum
    • Residual elemental mercury: residual elemental mercury was converted to metal/mercury amalgam using a mercury absorbent powder that was vacuumed using mercury vacuum
    • All areas: final high pressure water wash

    All areas were tested for mercury contamination after cleaning. If mercury was detected, the area was recleaned. Equipment used in processing and distributing mercury-containing battery slurry was also thoroughly cleaned. If mercury was detected in equipment after cleaning, it was recleaned. Larger and unnecessary equipment was cut out of the processing system but not necessarily removed from the area. The battery assembly machines, production areas, loading dock, storage areas, and trades shops were not cleaned.

    The mercury source was also found in the facility infrastructure. Although mercury has been removed from the process, it was still found occasionally in equipment and material samples. The thorough cleaning and removal of contaminated piping, vessels, floors, and walls removed the source and, therefore, essentially eliminated the potential for mercury discharges.

    CASE STUDY 5 - Hospital Incinerator

    Introduction
    Some hospitals operate onsite incinerators for disposal of infectious (or Ared bag) wastes. These wastes frequently contain mercury (from various tissues and diagnostic reagents). Some mercury can be transferred to wastewater generated by air pollution control systems (wet scrubbers) on the incinerator fume discharge.

    In a medical waste incinerator, typically operating at temperatures of 1800-2200 0F, any organic material containing mercury in any form will be oxidized and most of the mercury will be released in gaseous metallic (ionic) form. For mercury in an amalgam or other inorganic compound (e.g., as a chloride or oxide), it is possible that either the compound or the metallic form will be released. As a vapor, mercury may pass through pollution control equipment when gas temperatures are elevated (perhaps at about 300 0F or more).

    Wet scrubbers, however, can remove much of the mercury vapor from the gas stream. The removal rate varies with the partial pressure of mercury at the effective scrubber temperature. The mercury is removed from the gas stream by the recirculating water stream in the scrubber and appears in the wastewater (i.e., the blowdown) discharged from the scrubber. Typically, the blowdown stream from a medical waste incinerator scrubber cannot meet the current MWRA prohibition of mercury.

    Source Reduction Measures
    Identification and removal of all mercury-bearing substances before incineration could potentially eliminate the need for pretreatment of the wastewater discharge from the wet scrubbers that serve these medical waste incinerators. For example, one hospital set up a policy banning mercury thermometers. However, a full year after the ban was in place and a Acomplete roundup of all mercury thermometers had been conducted, some mercury thermometers could still be found in use.

    The hospital integrated other mercury control measures into routine daily operations including a program of spent battery collection. Satellite collection containers for used batteries were placed at nursing stations throughout the hospital. Two 55 gallon drums of used batteries were accumulated annually from beepers, flashlights, calculators, and medical equipment. The collected batteries were sent offsite for proper disposal. Other potential mercury sources eliminated from the incinerator were vacuum cleaner bags and used fluorescent light bulbs.

    Even after these source reduction measures were fully carried out, however, the incinerator scrubber wastewater stream could not meet the MWRA prohibition of mercury.

    Description of the System
    The hospital medical waste incinerator was equipped with a scrubber having a blowdown rate of six gallons per minute or about 5,000 gallons per day (gpd). The blowdown was being sent to the sewer system. Mercury concentrations in samples of the blowdown varied from nondetect to nearly 900 µg/L (ppb) as a function of quantity of mercury contained in the materials being incinerated. The hospital retained an incinerator pollution control company to help achieve compliance with the MWRA prohibition for mercury.

    Pretreatment Methods Investigated/Implemented

    Activated Carbon Adsorption
    An activated carbon unit was considered for installation in the blowdown line from the water recirculating section of the incinerator scrubber. However, over the long term, the mercury removed from the blowdown stream by the activated carbon could be re-released with variations in the blowdown pH and temperature. After trying various activated carbon materials, the use of an activated carbon was rejected as a solution to the problem.

    Sulfur-impregnated Activated Carbon
    Many mercury spill control kits rely upon elemental sulfur to react with the free mercury to produce a solid product, mercuric sulfide, that is easy to recover and remove. Mercuric sulfide is a very stable form of mercury. Therefore, an activated carbon material impregnated with elemental sulfur was seen as a possible solution to the problems of activated carbon alone. The pollution control company decided, therefore, to test a sulfur-impregnated activated carbon in place of standard activated carbon.

    The impregnated carbon is manufactured by a process that impregnates the pore structure of an activated carbon with elemental sulfur. Where activated carbon itself physically adsorbs the mercury from a wastewater stream, sulfur-impregnated activated carbon first physically adsorbs the mercury and then the adsorbed mercury chemically bonds with the sulfur. The chemical bond between the adsorbed mercury and the impregnated sulfur provides good stability in wastewater environments of varying pH and temperature.

    At the medical waste incinerator, substitution of a sulfur-impregnated activated carbon for activated carbon resulted in blowdown samples with mercury levels frequently below detection limits. A pilot wastewater pretreatment system using a sulfur-impregnated activated carbon achieved an average mercury removal efficiency of 99.92%.

    Because of the strong chemical bond within the medium, regeneration of the sulfur-impregnated activated carbon is not a possibility. Depending upon the ultimate mercury loading of the medium, however, recovery and recycling of the adsorbed mercury may be possible.

    Conclusion

    Hospital incinerators equipped with wet scrubbers often have compliance problems with the MWRA prohibition of mercury in permitted wastewater discharges. Sulfur-impregnated activated carbon is being used at one hospital facility to reduce mercury concentrations in the incinerator scrubber blowdown. To reduce initial and operating costs of the blowdown treatment system, source reduction measures should be carried out and aggressively monitored.

    CASE STUDY 6 - Bulb/Lamp Manufacturer

    Introduction
    The facility is involved in the manufacture of bulb/lamp lighting components. One stage of the manufacturing process, known as the alumina reclaim process, uses acid (hydrochloric acid, HCl) to remove internal metal components from alumina arc tubes. Mercury is one metal removed during this cleaning process. Spent acid is collected for disposal as a hazardous waste. Rinsewater flows to an on-site wastewater treatment system (pH adjust only) before discharge to the local sewer. Monitoring data from the rinsewater showed that mercury was present above the regulated limits. In 1994, work practice and technical measures aimed at reducing the carryover of acid into the rinsewater were set up. The facility is not on the MWRA system and has a local limit of 0.1 ppm or 100 µg/L (ppb).

    Description of Problem
    The acid cleaning process consists of the following two steps: (1) removing frit left over from the process of sealing the arc tube electrodes to each end of the tube; and (2) final dissolution of remaining internal metallic components/deposits, one of which is mercury. The first key step in addressing the problem of controlling mercury levels in wastewater was to reduce the carryover of acid from the cleaning process into the rinse. The problem involved the need to evaluate both work practice and technical measures.

    Methods Investigated/Implemented
    In 1994, sampling data showed exceedances of the facility's mercury permit limit. A more detailed sampling program from the end-of-pipe back up into the process was conducted to identify areas where mercury loading to the system was highest. Values ranged from below 1.0 ppm in treated effluent wastewater to 1,500 ppm in spent process acids collected for off-site disposal as a hazardous waste. An issue that arose during review of the sources discharging to the sewer was that the dilution effect from combining liquids both regulated and unregulated by the local sewer authority called into the question the end-of-pipe compliance of the combined flow. Adjustment of the end-of-pipe data to reflect this inadvertent dilution was recommended. The Acombined wastestream formula was used to adjust the end-of-pipe data to reflect the inadvertent dilution.

    A Quality Assurance/Quality Control review of the laboratory work was done to identify if laboratory procedures had the potential to affect the mercury being reported. Independent review of the laboratory confirmed that reported values were not affected by laboratory procedures.

    Examination of the processes contributing mercury to the wastewater linked the probability of the facility exceeding its mercury limit to the timing of sample collection relative to the acid baths life cycle. An opportunity became available to analyze raw aliquots of unused and used acids to identify a point at which the accumulation of mercury in the process acid approached a level that would affect compliance with the sewer limits.

    The facility did not expect mercury to exist in the HCl acid because chemical handbooks said that mercury was A insoluble in weak HCl acid solutions. The term A insoluble means that only relatively A small concentrations of a material will dissolve. However, A small concentrations may be high enough to result in a wastewater discharge compliance problem. Here, mercury may be more soluble in the concentrated acid used.

    Observation of the process steps identified that under the current work practices, acid solutions were not being completely drained before removal of the tube and immersion in the rinse step. Work practice modifications to increase drainage rates were reviewed. Also, small quantities of elemental mercury particles were settling out on the rinse tank trough. The trough was sloped toward the drain, and it was recommended that the trough be sloped away from the drain to retain the settled mercury particles.

    In an additional attempt to control the mercury transferred to the wastewater, a recommendation was made that a stagnant rinse tank be added as a process step between the acid clean and running rinse operations. This tank will rinse off most of the acid carried out of the process bath and result in a clean part being immersed into the running rinse that ultimately flows to the sewer. The periodic need to change the stagnant rinse could be addressed by treatment, collection, and off-site disposal, or by reuse of the stagnant rinse as makeup to the process cleaner. The option of using the stagnant rinse as makeup to the process cleaner was not advised because of accumulation of mercury in the stagnant rinse.

    Mercury was found in a component of the wastewater treatment system where supposedly no alumina reclaim processing wastewater was discharged. One possible explanation is that elemental mercury was present in the tank and was dissolving into the wastewater. Since the facility discharged on a batch basis, a recommendation was made to consider a larger tank for flow equalization that could attenuate the discharge of mercury over a longer period.

    Conclusion
    Current work practices required modification to increase the amount of acid solution drained before the rinse step. This modification and the installation of a stagnant rinse tank between the acid clean and running rinse operations will help reduce the mercury transferred to the wastewater. A mathematical model showed that these measures alone may be sufficient to bring the facility into compliance with the local mercury standard of 0.1 ppm.

    CASE STUDY 7 - Clinical Testing Laboratory

    Introduction
    Under a compliance order from the MWRA, a 250-employee clinical testing laboratory reduced its mercury discharges from 0.3 mg/L (ppm) to less than 0.001 mg/L (ppm) or 1.0 µg/L (ppb). This reduction was achieved through a two-pronged approach:

      1. Source reduction techniques were used to reduce the mercury entering the facility's wastewater treatment system by approximately 90 percent; and

      2. A sophisticated treatment system was then installed to remove the residual mercury.

    These steps enabled the facility to meet the 1.0 µg/L (ppb) mercury discharge standard several months ahead of the MWRA compliance deadline. The Office of Technical Assistance (OTA), a branch of the Massachusetts Executive Office of Environmental Affairs, provided confidential, nonregulatory assistance with the project at no charge.

    Background
    After the laboratory received an MWRA compliance order to eliminate mercury from its wastewater discharge, it was found that the source of the mercury was thimerosal, a mercury salicylate salt used as a bacteriostat/fungistat in many clinical tests and could not readily be replaced. Clinical test equipment and test kit manufacturers, aware of the problem and under pressure from their customers, are working to develop reagents with alternative preservatives. However, the removal of mercury from a test kit involves revalidation of the test kit with subsequent approval by the US Food and Drug Administration. This is a time-consuming process that can require several years to complete.

    Source Reduction Efforts
    Having determined that thimerosal was the source of the mercury, the facility undertook a program to identify which analytical instruments generated wastewater discharges that contained mercury. Samples from all point-of-source discharges were sent to an environmental testing laboratory for mercury quantification. Approximately 50 potential sources were identified, sampled, and tested. About 30 percent of the potential sources were found to contain measurable quantities of mercury, some as high as 1.0 mg/L (ppm) or 1,000 µg/L (ppb).

    Once the analyses were completed, several source reduction efforts (including toxics use reduction (TUR) and wastewater sequestering) were carried out to prevent mercury from entering the wastewater. The primary TUR technique was to contact test kit manufacturers to find the availability of suitable mercury-free alternatives. Some manufacturers said that revalidation of reformulated kits would take a minimum of two years. Other manufacturers, including Technicon and Hybritech, were already aware of the mercury issue and could supply alternative kits. The effort resulted in the replacement of four of the 15 test kits used by the facility after validation studies were performed (a one to two month process).

    At OTAs prompting, the facility pursued several other TUR options, including worker training and improved housekeeping techniques, to prevent mercury from entering the facilities wastewater. All employees were informed of the problem and the efforts being taken to correct it. Signs and labels are now posted throughout the facility describing the proper handling and disposal of mercury-containing materials, with emphasis on what should not be discharged down the drain. These efforts are particularly important since only a very small amount of mercury (about 0.2 grams) could raise mercury concentrations in the facility wastewater to about 0.3 mg/L (ppm) or 300 µg/L (ppb).

    Because of the investigations, about five gallons per day of wastewater is now sequestered at the sources, collected in containers, and transported off-site as hazardous waste. This sequestered wastewater consists of all equipment discharges containing mercury that can be easily collected at the sources in small containers. While the sequestered wastewater is not a large volume of water, it does include many instrument discharges containing high concentrations of mercury.

    From these source reduction efforts, the mercury level in the wastewater discharge was reduced to about 0.03 mg/L (ppm) or 30 µg/L (ppb). Although this represented a significant reduction in mercury concentration, it was not sufficient to comply with the MWRA enforcement limit. Consequently, the facility investigated additional wastewater sequestering and offsite disposal (up to and including the entire facility discharge) and wastewater pretreatment. The pretreatment options explored included evaporation, ion exchange, precipitation, and carbon adsorption. Based on an economic analysis of the various approaches, carbon adsorption was chosen as the most cost-effective technique.

    Wastewater Pretreatment
    A pilot study was conducted to learn the effectiveness of carbon adsorption. In this study, about 450 gallons per day of wastewater were pretreated using DisposorbTM carbon. This is a reactivated carbon sold by Calgon in plastic drums. When the carbon bed reached saturation, the entire drum would be sent offsite for disposal. Carbon adsorption was found effective in removing mercury: mercury levels were reduced from about 0.06 mg/L (ppm) or 60 µg/L (ppb) upstream of the carbon bed to nondetectable levels downstream.

    Based on these results, the facility decided to treat approximately 1,800 gallons per day of wastewater in a full-scale carbon adsorption pretreatment system. The system consisted of three parallel trains of carbon beds, with each train consisting of two drums of DisposorbTM carbon in series. The system was effective in removing mercury but only to a concentration of approximately 0.02 mg/L (ppm) or 20 µg/L (ppb) - well above the MWRA enforcement limit. The facility also noted extensive bacterial growth on the carbon beds.

    At this point, the facility asked OTA for assistance with the adsorption system. Following a site visit, OTA prepared several recommendations intended to optimize system performance:

    • Operate the system with constant optimum flowrates through the carbon beds by using an equalization tank and pumps upstream of the carbon beds to provide storage capacity. The system had been installed without any provision for control of flow rates, and this was resulting in channeling in the beds, which led in turn to poor mercury removal. OTA suggested that the facility contact Calgon to learn optimal flow rates.
    • Install bag filters upstream of the carbon beds to avoid plugging with solids that decrease the adsorptive capacity of the carbon.
    • Investigate the use of an ultraviolet (UV) light sterilization unit or silver-impregnated carbon to control bacterial growth in the carbon adsorption system. (Silver is a bacteriostatic material, i.e., it inhibits bacterial growth).

    Based on these recommendations, the facility installed a 500-gallon equalization tank upstream of the carbon beds and flow control valves on each of the three trains to maintain optimum flow through the carbon beds. The valves were sized based on information supplied by Calgon. Calgon also suggested that the adsorptive capacity of the carbon is greatest when the pH of the water is maintained between 4.0 and 5.0. The pH of the influent wastewater is now adjusted to this range in the equalization tank. Bag filters (15 micron rating) were added upstream of the equalization tank, and ultraviolet lights were installed between the filters and the equalization tank to control bacterial growth. The pH of the activated carbon system effluent is adjusted to the range of 5.5 to 10.5 before it flows into an existing final neutralization tank.

    Results
    Once these changes were made, effluent mercury levels of less than 1.0 µg/L (ppb) were achieved - meeting the MWRA enforcement limit. The facility successfully completed the required compliance testing several months before a MWRA-stipulated deadline.

    While toxics use reduction (TUR) efforts were not themselves sufficient to achieve compliance, they did lead to a 90 percent reduction in the mercury that had to be removed in the pretreatment system. This reduction translated into lower pretreatment system capital and operating costs.

    Capital expenditures to achieve compliance were more than $60,000. This figure includes the costs of both the pilot and full-scale carbon adsorption systems and of replumbing the facility. According to facility calculations, the operating costs of the new system are more than $7,500 per month. However, the facility would have faced higher costs for offsite disposal of the wastewater and could have been liable for fines of $10,000 per day if it had not met the MWRA deadline for elimination of the mercury from the wastewater.

    For further information about this case study, other OTA case studies, or OTA's technical services, contact:

    Office of Technical Assistance
    100 Cambridge Street, Room 2109
    Boston, MA 02202
    Telephone (617) 727-3260
    Facsimile (617) 727-3827

     

    RETURN TO HG MANAGEMENT GUIDEBOOK
    TABLE OF CONTENTS