Measuring source reduction of laboratory hazardous wastes.

One of the highest priorities for hazardous waste management is to reduce the volume of generated wastes. This goal requires extensive analysis of waste streams and source reduction approaches (1, 2, 3, 4). The benefits of source reduction are well documented and include reduced disposal costs, decreased liability, improved worker safety, and better public relations (5).

Most of the literature on source reduction relates to industrial generators. This paper, however, focuses on academic and research laboratories that generate hazardous waste. Such labs usually present a greater challenge in measuring source reduction than their industrial counterparts. The waste streams they generate are typically more diverse. Lab processes vary more over time than industrial processes. Also, many labs do original work with no accepted methods for source reduction or measurement of waste streams (5, 6). Therefore, laboratories require a more inventive approach to source reduction and measurement of waste streams (7, 8, 9, 10). Our discussion emphasizes the unique legal, financial and technical problems that academic and research labs have with hazardous waste source reduction (11).

In the next section of this paper, we discuss several recently enacted laws that address source reduction, including the federal Pollution Prevention Act of 1990, and U.S. EPA's voluntary 33/50 program. Many states have passed similar legislation, and we focus on California's Hazardous Waste Source Reduction and Management Act of 1989 (SB-14). In our third section, we review ways to measure source reduction. Other authors have suggested various normalization factors to provide a clearer picture of source reduction (5, 6). In contrast, the thesis of this paper is that normalization factors fail in the dynamic setting of academic and research laboratories. In the fourth section, we suggest input/output analysis as a more effective measure of source reduction, and we provide appropriate caveats for such an approach. In the last section, we discuss source reduction measures in the larger context of hazardous waste management.

Emerging legislation

The EPA is targeting source reduction through at least two major developments: the Pollution Prevention Act of 1990 (PPA), and the voluntary 33/50 program. The PPA requires generators to file reports on source reduction. The voluntary 33/50 program encourages generators to achieve a 33 percent reduction of certain listed chemicals by 1992 (compared to 1988 levels), and a 50 percent reduction by 1995. In the area of hazardous waste management, the Hazardous Pollution Prevention Planning Act of 1991 proposed that filers of the Toxic Release Inventory (TRI) prepare reports on source reduction of hazardous wastes. Although most research institutions do not fall into these categories, these developments show a definite direction that the EPA is taking in source reduction.

More than 15 states have passed laws that require companies to prepare written plans for pollution prevention (12). For example, California is regulated by the Hazardous Waste Source Reduction and Management Act (13). Passed in 1989, this law is one of the first of its kind. We focus on this law as an example of laws passed or proposed by other states. The California Environmental Protection Agency (Cal-EPA) implements this source reduction law (14). Its purpose is to reduce hazardous waste at its source and to document such progress to the state. The law covers generators who routinely generate more than 12,000 kilograms of hazardous waste (or 12 kilograms of Extremely Hazardous Waste as defined in the California Code of Regulations) in a calendar year. Certain waste streams such as asbestos and infectious waste are exempt (15).

A major feature of the California law is its documentation requirements. Generators must prepare a source reduction evaluation review and plan (and plan summary), and a performance report (and report summary). These documents were due September 1, 1991 and every four years thereafter. Before submission to Cal-EPA, these reports must be certified by a registered engineer, a registered environmental assessor, or an individual who understands and is responsible for site operations (16).

The first document, the Source Reduction Review and Plan (and Plan Summary), requires generators to assess their institution's operations and waste streams. Source reduction options must then be identified and analyzed for their viability. Viable options must be implemented by a defined schedule. If a generator does not implement a source reduction option, they must document their reasons to the satisfaction of the state.

Table 1

Schedule for a hypothetical research institution

             No. of   No. of      Research     Square   Output
Quarter      Units    Personnel   Dollars      Feet     (kgs.)

1. January
  to March   10       10          $2,500,000   10,000   5,000

2. April
  to June     8        5          $1,333,000   10,000   4,000

3. July
  to Sept.    6        5          $  750,000   10,000   3,000

4. October
  to Dec.     5        5          $  500,000   10,000   2,500

The second document, the Hazardous Waste Management Performance Report (and Report Summary), evaluates the source reduction approaches used from the baseline year to the current reporting year for each waste stream. This report describes the processes that generate each hazardous waste. It evaluates the implemented source reduction approaches by considering input changes, operational improvements, and various administrative steps.

It is this second report that presents the fundamental problem of how to measure progress in source reduction. Such a measure should normally include the percent change in volume generated over a reporting period. However, each generator must adjust these measurements for factors such as growth, changes in activity, and changes in waste classifications. Without these adjustments, measures of waste reduction are inaccurate. For example, start-up of a new research project can increase the total waste generated. This could inaccurately indicate a failure of a waste reduction program. Conversely, the shutdown of a key project can inaccurately indicate a waste reduction success. The tracking and reporting requirements for SB-14 and other such laws do not specify how to quantify these factors. Also, the current literature offers limited guidance on how to track and report source reduction of waste streams from academic and research laboratories.

Methods for measuring source reduction

The first step to tracking source reduction is to develop an appropriate baseline. Data may come from manifested shipping weights or internal documents. Initial adjustments include one-time lab cleanups or removal projects. To reflect ongoing operations, such weights must always be subtracted before proceeding with the analysis.

The next step is to measure the percent change in output over the reporting period. In order to reflect the change brought about by source reduction approaches, we must adjust these measures for extraneous factors that affect hazardous waste generation. This includes such factors as growth and changes in activity. Unfortunately, it may be difficult to separate out all the influences on waste output.

Due to these difficulties, some researchers suggest the use of normalization factors (5, 6). This process divides the waste output by a chosen unit (i.e., normalization factor). For research institutions, these factors may include: research dollars, number of researchers, square feet of lab space, and number of research units (a research unit is any laboratory operating within the larger jurisdiction of an organization).

The problem is that this approach assumes the relationship between waste generation and normalization factors is constant over time. In research and academic labs, this is often not true. To demonstrate the effect of these issues, the strongest evidence would come from documented data on source reduction efforts. However, the availability of such data is limited by the relative newness of legislation and the reluctance of institutions to release their data outside agency mandates. Therefore, we present hypothetical data that we will argue is a realistic representation of current trends. Table 1 shows a hypothetical schedule of a research institution. Table 2 shows the resulting values for different normalization factors. We analyze the different adjustment units as follows.

Waste output per research unit

Waste output per research unit is inappropriate because a research unit may close down, reduce or increase its activity. This is not uncommon in the dynamic environment of university and research labs. For example, Table 1 shows a decrease in research units over a single year. As a result, Table 2 shows that the normalized output remains steady throughout the year.

This particular measure ignores a critical development in this institution. Moreover, research institutions may develop inactive units for no other purpose than to show "progress" in source reduction.

Waste output per research personnel

Measuring the waste per researcher can also be inaccurate, since researchers may change their activity or experimental procedures. For example, Table 1 shows a one time cut in personnel during the second quarter. As a result, Table 2 shows an increase in normalized output that returns to its original condition.

This situation contrasts with the static picture in the previous section. It may even encourage institutions to name as many "researchers" as possible to keep this normalized measure as low as possible. For example, some researchers may be part-time employees. Others may carry multiple appointments within the institution that are inappropriate. The question of what constitutes a viable "researcher" may be as difficult as measuring the waste itself. Moreover, it may be unnecessary to introduce this added layer of complexity.

Waste output per research dollars

Table 1 shows an institutional decrease in research dollars. Sadly, this is increasingly common in the current economic environment. As a result, Table 2 shows an increase in normalized output. Also, research dollars may cover many items unrelated to actual lab activity. For example, institutions have different rates for indirect costs and salaries.

This particular example invites enforcement activities precisely when the institution may be fighting for survival. More ominously, enforcement activities may be unrelated to actual risk. Finally, it encourages the type of financial manipulation that can be extremely difficult to diagnose.

Waste output per square feet of lab space

Table 1 shows a static condition for lab space. As a result, Table 2 shows a steady decrease in normalized output. This normalization factor presents the temptation to maintain lab space, even if it does not support research activities. As universities and research labs struggle to maintain their competitiveness, this normalization factor may ultimately tie up finances in needless capital (i.e., poorly used lab space).

All the above measures fall short of being accurate since there may not be a linear relationship between the normalizing factor and hazardous waste generation. Indeed, when viewed together, these measures present a conflicting picture of source reduction efforts.

Input/output analysis

A better approach to this problem is to measure percent change of waste output over materials input. We divide the waste generated by the hazardous materials input (output/input) over successive periods for a particular waste stream. For academic and research labs, wastes could be grouped by experimental procedures used at the institution. These experimental procedures would be equivalent to industrial processes producing a waste stream. For example, many types of research involve electron microscopy. Various chemicals used to process tissues for electron microscopy can generate hazardous waste. Institutions would define all wastes from electron microscopy as a single waste stream, even though the exact chemicals used for each procedure would differ. Changes in this waste stream could be measured over time and reported as required.

Table 2

Normalization of schedule for a hypothetical research
institution

              Output/   Output/     Output/    Output/
Quarter       unit      personnel   $100,000   1,000 sq. feet

1. January
   to March    500       500         200        500

2. April
   to June     500       800         300        400

3. July
   to Sept.    500       600         400        300

4. October
   to Dec.     500       500         500        250

Other examples of procedures which could be identified as one waste stream would be electrophoresis, histology, and scintillation experiments. Under the California law (SB-14), many procedures would remain unique. However, the waste from these procedures would typically not add up to 5 percent of the total waste volume, and thus would not require analysis under the California law.

Table 3 shows the input/output results for a second hypothetical lab. Notice that the increased activity in the second quarter is reflected in both increased input and output, but the input/output ratio shows progress in source reduction. Again, these ratios can be further analyzed by different experimental practices. The measurement of inputs is certainly not trivial, but it is readily manageable by maintaining records of purchased and distributed supplies.

Of course, any measurement approach has limitations, and input/out analysis has at least three major disadvantages. First, it imposes extra data demands in requiring information on inputs. However, this effort will become more justified by the increased regulatory emphasis on source reduction. Second, the unique inputs to a given research lab may make it difficult to compare input/output ratios among institutions. However, such comparisons are almost always problematic. Third, this approach does not account for system throughputs such as chemicals released to the environment during processing. Unfortunately, the dynamic nature of research laboratories may preclude any comprehensive accounting of such throughputs. Despite these limitations, input/output analysis is closest to the underlying goals of source reduction. That is, it requires generators and agencies alike to consider the entire generation process from a systems view. It also reduces the temptation of institutions to concoct organizational tricks to show "progress" in source reduction.

Discussion

Hazardous waste management has entered a new era. With recent legislation, we have seen increased emphasis on source reduction. Simultaneously, the law emphasizes accurate tracking of hazardous wastes. Aside from the obvious motivation of generator accountability, this approach introduces problems of how to measure source reduction. It is not a trivial problem. Academic and research laboratories are most problematic on this issue. They have waste streams that are difficult to define, but they can still reduce hazardous wastes. The key to such progress is better chemical management and improved lab practices. This requires careful design of experiments with waste minimization considered throughout the process. It also requires a measurement approach that is resistant to abuse.

Table 3

Input/output analysis for a hypothetical research institution

Quarter        Input (kgs)   Output (kgs)   Output/Input

1. January
   to March      5,000         5,000          1.0

2. April
   to June      10,000         8,000          0.8

3. July
   to Sept.      8,000         4,800          0.6

4. October
   to Dec.       6,000         3,000          0.5

We reiterate that all measures are subject to abuse. However, agencies should specify a measurement approach that is most consistent with underlying legislative goals. We believe that input/output analysis is closest to these goals and is ultimately easier to measure. If more agencies mandated these procedures, we could gain a more accurate measure -- and with it a deeper understanding -- of source reduction approaches.

References

1. California Dept. of Health Services (1988), Waste Audit Study -- Research and Educational Institutions, Sacramento, CA.

2. U.S. Environmental Protection Agency (1990), Guides to Pollution Prevention -- Research and Educational Institutions, Doc. No. EPA/625/7-90/010, U.S. Government Printing Office, Washington, D.C.

3. Newton, J. (1990), Setting up a waste minimization program, Poll. Engineering 22(4):75-80.

4. Rice, S.C. (1988), Environmental review strategy for R&D activities, Environ. Progress 7(1):46.

5. Freeman, H.M. (1990), Hazardous Waste Minimization, McGraw Hill, New York, NY.

6. Rice, S.C. (1988), Minimizing waste from R&D activities, Chem. Engineering 95:85-88.

7. ... (1985), Less is Better -- Laboratory Chemical Management for Waste Reduction, American Chemical Society, Washington, DC.

8. Armour, M.A. (1987), Hazardous Chemicals -- Information and Disposal Guide, University of Alberta Press, Edmonton, Alberta.

9. Backus, B.D. (1990), "University of Minnesota Waste Minimization Strategies," 8th Annual College and University Hazardous Waste Conference, Minneapolis, MN.

10. Hollinsed, C.W. (1987), Waste reduction through minimization of reagent usage, Hazardous Waste and Hazardous Materials, 4(4):357-361.

11. Kaufman, J.A. (1990), Waste Disposal in Academic Institutions, Lewis Publishers, Chelsea, MI.

12. Pojasek, R.B. and L.J. Cali (1991), Measuring pollution prevention progress, Poll. Prev. Rev. 1(4):91-95.

13. ... (1992), California Health and Safety Code, Section 25244.12, Sacramento, CA.

14. ... (1992), California Code of Regulations, Section 55520, Sacramento, CA.

15. ... (1992), California Health and Safety Code, Section 66521, Sacramento, CA.

16. ... (1992), California Health and Safety Code, Sections 66523.3 and 66524.3, Sacramento, CA.

Thomas H. Hatfield, Dr. P.H., Dept. of Health Science, California State University, Northridge, CA 91330.