Introduction >

Midterm >

Conclusion >

Footnotes >

Section 9:
Measuring Performance with Environmental Performance Indicators

This section looks at the performance of the project through Environmental Performance Indicators (EPIs) set forth in the project's FPA. The universities committed to conducting a baseline assessment of environmental performance prior to the implementation of the Laboratory EMS, based on representative data, within the first six months of the effective date of the Final Rule with a report within nine months. The baseline assessment was completed on June 28, 2000. The progress of each school in meeting many of the EPIs should be measured in comparison to the baseline assessment. From 2000, the schools have been producing annual reports that detail each institution's progress in meeting the EPIs. Data collected for the baseline EPI, and 2000 and 2001 data are grouped according to the EPI. In general, the baseline information was more robust for certain indicators than others. In order to ascertain the level of progress, efforts should be made to ensure that the baseline information collected is complete and reflects quality data.

Because UVM implemented their EMP in December 2000, the December 28, 2000 First Year Status Update for the project did not contain much data and a decision was made to supplement the report. In April 2001, EPA New England met with the three XL universities and Nexus Environmental Consultants to discuss the format and information needs for the supplemental report. The report covered the topics of HCOCs, laboratory audits, pollution prevention, laboratory waste reduction and chemical reuse and recovery goals. In June 2001, the supplemental report was submitted and in September 2001, a follow-up meeting was held with participants from the three participating schools, Nexus Environmental Partners, MA DEP, VT DEC, EPA-New England, and EPA Headquarters. The September meeting focused on the EPIs and at the meeting, it was decided that EPA Headquarters, to assist the project in achieving its long-term outcomes, would conduct a mid-course correction evaluation. Additionally, a subcommittee formed to specifically evaluate different ways to look at compliance. That review results in a scorecard that can be used to assess EPI #9, which is presented in the discussion below.

The results presented below are based on the 2000-2001and 2001-2002 annual reports completed by the three institutions. The following sections detail the results of the mid-term review of the EPIs and presents findings and recommendations for the EPIs. Similar EPIs are grouped by relationship.

9.1 EPI #1 Goal: Outdated Chemicals of Concern and
EPI #2 Goal: Hazardous Chemicals of Concern Inventory

Hazardous Chemicals of Concern (HCOC) on shelf that exceed institution defined "shelf life" (EPI #1). The goal of EPI #1 is to ensure that outdated hazardous chemicals of concern are appropriately removed from laboratory shelves and disposed. This EPI is a result of the observation that good housekeeping is an important hazardous waste minimization strategy for laboratories. A laboratory that tracks its chemical inventory carefully enough to prevent accumulation of outdated chemicals is very likely to avoid purchasing excess chemicals. EPI #1 is discussed jointly with EPI #2, below, in this evaluation.

Annual Surveys of Hazardous Chemicals of Concern (HCOCs) (EPI #2). EPI #2 examines each university's efforts to develop a methodology for conducting baseline HCOC risk assessment inventories. The exact HCOC lists are developed on a university-by-university basis, as the types of hazardous chemicals at a particular university vary with the types of research work performed there.

The schools have completed a good deal of work on these two EPIs. Meeting these EPIs is easier with the EMP in place as it clearly defines responsibilities of Principal Investigators (PIs) and laboratory personnel. However, this is an area where the universities have not yet accomplished the goals of the FPA (to have 100 percent completion of the HCOC surveys and to only have HCOC on the shelves that are within the university defined chemical shelf lives) for a variety of reasons that are outlined below. It is clear that there is a lot of ambiguity on the definition of outdated chemical shelf lives as expiration dates supplied by the manufacturer are often too conservative to encourage future purchases of chemical stocks and researchers make individual decisions on how long to use chemicals based on research needs. In addition, implementing a common HCOC procedure for the academic institution is difficult due to frequent turnover of staff, faculty and students.

Boston College: EHS developed the HCOC list to identify the shelf lives of certain chemicals that have specific shelf-lives (i.e., ethers and picric acid), however, laboratories determine shelf-lives for most of their chemicals. EHS also advises laboratories to examine chemicals from a housekeeping perspective and to remove chemicals not used for quality reasons. These materials (which are few in number) are wastes. The remaining materials are potentially useful. EHS then asks the laboratories to evaluate a chemical's degree of hazard versus the utility of having it on hand. The Boston Fire Department requires a complete inventory of hazardous chemicals in the laboratories so that they have a good understanding of chemicals stored and used in the laboratories in case of emergencies. Boston College completed its lists of HCOC in August 2001 and it was introduced in training in August 2001. The HCOC survey was conducted in two phases-the high hazard chemicals sweep and inventory audit.

In spring 2000, EHS requested that laboratories provide complete chemical inventories in accordance with requirements of the Boston Fire Department and to identify quality or need-based decisions on keeping each chemical in stock. Ninety percent of the laboratories were able to comply with the request at that time. The remaining laboratories were involved in a renovation project and were to complete their inventories after the moves. After the EPA Audit in April 2001, BC readjusted its approach to HCOCs. In addition to providing more specific information to labs about particular HCOCs, EHS began listing chemicals that require annual review. The list is based on the criteria of degree of hazard or stability or quality over time. In February 2002, Boston College hired a chemist from Onyx Environmental who did a physical audit of all laboratories to identify potentially unstable chemicals. The chemist identified approximately 40 containers that were recommended for disposal due to their ages or conditions. During summer 2002, the laboratories will be audited again to determine if the disposals took place.

The second phase of the HCOC audit was to develop a baseline for the laboratories to identify highly hazardous, though not necessarily reactive or unstable chemicals, through a scan of the inventories submitted by the laboratories. A student worker in the EHS department reviewed the inventories on file and highlighted the chemicals that were on the HCOC list. The laboratories were notified that certain chemicals should be assessed for ongoing usefulness, proper storage, and safety considerations. This will be undertaken along with the full inventory submission for Boston and Newton Fire Departments.

University of Massachusetts Boston : UMB is required by the Boston Fire Department to have complete chemical inventories for all laboratories and to conduct an annual inventory of chemicals. This list is reviewed on an annual basis and updated to ensure it covers an appropriate breadth of hazardous materials. Additionally, Principal Investigators are asked to evaluate peroxide-forming chemicals and nitro compounds when completing the Monthly Laboratory Self-Inspection Checklists. UMB has designated the following chemicals as HCOCs:

* EPA P-listed wastes
* OSHA special carcinogens
* OSHA teratogens/reproductive toxins
* OSHA designated highly toxic substances
* Explosive nitroarenes
* Peroxide-forming chemicals
* Pyrogens
* Shock-sensitive explosives

As of June 2001, UMB has not directly tracked the absence of outdated chemicals on laboratory shelves. Instead, EHS requires laboratories to conduct comprehensive inventories of all laboratories with which EHS highlights generic categories of HCOCs in training sessions. EHS believes that it has seen fewer outdated materials remaining on shelves and that there have been a decrease in the disposal amounts of these types of materials. In June 2001, the PIs were also asked to evaluate peroxide-forming chemicals and nitro compounds when completing the Monthly Laboratory Self-Inspection Checklists. These compounds are the most prevalent and problematic HCOCs on campus as once they are opened they usually have short self-lives unless they are regularly monitored and tested for peroxide presence. Nitro compounds must be monitored to insure that they contain at least 10 percent water or they become unstable.

UMB has identified eight classes of chemicals in its CH/EM plan and laboratory workers receive guidance with respect to the management of these chemicals during training. EHS has tagged or highlighted these materials on inventory sheets for each laboratory. In June 2001, the current system of conducting the chemical inventory underwent significant change. Under the old system, the EHS office generates a chemical inventory list for each laboratory from its database and sends the list to all PIs in August. PIs have one month to update lists, sign them, and return them to EHS for input into a central database. In the past, this manual process has taken an enormous amount of time for the PIs and EHS staff. The typical update time period from start to finish has taken as much as 18 months. To minimize this problem and create more accurate inventories, in March 2002 EHS implemented the ChIM 5.2, a new chemical bar code based tracking system on a lab-by-lab basis. UMB believes that the bar code system has speeded up collection of the inventories and provided EHS with more accurate and reliable data. The tracking system is anticipated to enhance the ability of EHS to identify pollution prevention opportunities. The new system should be more efficient and allow EHS to track chemicals from laboratory to laboratory.

EHS is testing the efficacy of the software with a pilot project based on the laboratories under the supervision of one professor, who oversees five active chemistry labs. The pilot was implemented in the fall of 2001, and has returned promising results, as EHS was able to monitor all materials and update the inventory as necessary with lab personnel. This suggests that the barcoding system will achieve the anticipated benefits of inventory management. The manual inventory will not be updated in order to complete the barcoding effort.

The next step for UMB is to network the program so that individual departments will have access to the inventories, which will allow them to update the system with new materials and search for chemicals when needed. By the end of summer 2002, for specific PIs, the EHS office will take inventory from each laboratory and generate Operational Material Safety Data Sheets for each laboratory. In addition, the inventory list will have HCOC's marked an explanation of what HCOCs will be included with each information package.

University of Vermont: UVM based its HCOC inventory on the requirements of the Superfund Amendments and Reauthorization Act (SARA) Title III reporting, which is now commonly known as the Emergency Planning and Community Right-to-Know Act (EPCRA). EPCRA was designed to inform emergency planners and the public of potential chemical hazards. The regulations were developed to provide the quantity of regulated chemicals at a facility, the specific hazards presented by the chemicals, the fate of chemicals (i.e., used, discharged, sold, etc.), and any unplanned releases.

The UVM HCOC survey process includes laboratory workers identifying and disposing of outdated materials while completing the form on an annual basis. A variety of regulatory chemical lists were reviewed in 1990 to generate a list of approximately 400 hazardous chemicals considered to be of potential environmental or safety risk and likely to be found at UVM. The list is distributed to laboratories every January, and the laboratories report the quantity of each chemical on the list that is stored there on a daily basis, which are then rolled up into cumulative totals. This process provides a way for emergency responders to plan for potential responses to campus buildings by identifying those buildings with significant amounts of hazardous laboratory chemicals.

Both internal and external audits have revealed that a large number of excess chemicals were in storage in the Chemistry Department stockrooms and in the Agricultural Biochemistry stockrooms. UVM contracted with Heritage Environmental, Inc. to inventory, package and dispose of these chemicals. This work was completed in July 2001 and cost UVM more than $25,000.

In June 2001, ESF staff also focused on the management of outdated HCOCs in the College of Medicine. Approximately 50 laboratories within the College of Medicine were decommissioned, moved or renovated. As these rooms emptied, ESF staff provided clean-out assistance to the laboratory workers to facilitate excess chemical disposal. Twenty-three laboratories from the College of Medicine asked for this assistance.

In June 2001, UVM had a HCOC survey procedure in place, but had not yet determined how best to use the survey process to measure the numbers of outdated chemicals on laboratory shelves. ESF staff have found that the concept of "outdated chemicals" as it is ambiguous to lab workers, who often find reliable ways of using chemicals beyond manufacturer's expiration dates. ESF is investigating ways of developing a more quantitative approach to tracking outdated chemicals. The 2001 UVM HCOC survey was administered between February 1, 2001 and March 31, 2001. For 2001, ESF changed its survey procedures and distributed survey forms on a room-by-room basis, along with other EMP implementation forms, rather than giving survey forms to the laboratory supervisor. ESF originally organized laboratories for the HCOC survey based on lab supervisors primarily because laboratory chemicals are commonly assigned to a particular laboratory supervisor and move between laboratories under his/her control. ESF decided to change its approach in 2001 because it believed that more laboratories would be included under the room-by-room distribution approach. Participation in the 2001 HCOC survey was disappointing for ESF. Only 251 labs out of 538 laboratories submitted HCOC forms in time to be included in the SARA Title III submission. This number represents 45% of the universe of UVM's labs and is below the historical HCOC survey return rate of 60 to 80 percent. ESF believes that the lower response rate was due to distribution of five new forms (needed to implement the EMP) at the same time as the HCOC survey form and that the information requests detracted attention to the HCOC survey. In addition, many laboratories in the College of Medicine did not complete the inventory forms as they were anticipating a move within three months and were expecting to conduct significant chemical clean-outs as part of their efforts. Lastly, many supervisors preferred the older approach and combined their laboratories into one form. For 2002, UVM returned to its laboratory supervisor management approach for HCOC inventories.

UVM has been tracking the trends in chemical inventories per UVM laboratory for the years with available data since 1994 (see Table 3)9. Of interest for EPI #1 are the columns labeled "Chemical Count per Lab," which presents the average number of different chemicals found in labs and "Total Pounds of HCOC" shows the total weight of these chemicals. These two columns indicate that both the average number of chemicals and the total amount of chemicals being stored in UVM laboratories have dropped by approximately one-third since the implementation of the EMP. This drop is statistically significant within the variation shown by these numbers over the history of the inventory. The increase in these numbers from 2001 to 2002 is not outside the historical standard deviation for this measurement.

ESF attributes the decrease in the amount of HCOCs in the laboratories to the "chemical safety surveys" (1998-2000) and safety audits (2001) conducted by ESF staff that increased attention to the chemical inventory management process. Specifically, these lab visits emphasized increasing lab workers' understanding of the problems associated with outdated chemicals. ESF attributes the increased survey response rate for 2002 to increased training, better survey distribution methods, and increased follow-up with laboratories. For example, in 2002, ESF sent two email reminders to laboratory supervisors, and this effort helped double the return rate in 2002 over 2001.

Table 3: HCOC Inventory Trends at UVM Year10 Forms Distributed Rooms Reporting Lab Supervisors Reporting Supervisor Response Rate Chemical Count per Lab Total Pounds of HCOC per Lab 1994 228 85 72 32% 24 207 1995 224 121 112 50% 32 276 1998 244 109 101 41% 25 175 1999 235 97 88 37% 25 207 200111 453 220 Unknown 49% 16 134 2002 220 205 160 73% 19 153 1995-1999 average 234 103 NA 40% 26 216 2001-2002 average 337 213 NA 61% 18 143 % change 44% 107% NA 53% -31% -34%

One possible approach to improving laboratory response to the HCOC survey is through the implementation of a web-based version of the HCOC survey form to facilitate data input by laboratories and to improve the survey response rate. The EMP forms will also be distributed separately to increase emphasis on the HCOC surveys and inventories. In addition, ESF plans to use the ESF Compliance Audits to measure progress in removing outdated HCOCs from laboratory shelves. This will be done by making special note on the audit forms of any laboratories that have outdated time sensitive chemicals in storage. Using this system, ESF believes that it will be able to track the number of laboratories with this problem using 2002 as a baseline. ESF's goal for the 2003 survey is to increase the response rate to 85 percent, in order to continue to move the project goal of 100 percent participation. ESF plans to accomplish this by increasing follow-up with laboratory supervisors.

Findings: Disposing of outdated chemicals of concern is a top health and safety priority in the laboratory setting. Increased new domestic security issues around terrorism have heightened the awareness of colleges and universities to determine what hazardous materials are present on their campuses and develop proper housekeeping and management strategies to deal with chemicals both used and stored. The EMP establishes a good foundation for future improvements in this area. The three universities are working to ascertain a baseline or an inventory of outdated chemicals. It is apparent that laboratory moves and relocations have made this process lengthier and greatly increased the need for EHS presence in the laboratories to assist in the clean-outs. During the group discussions, laboratory staff and faculty at all schools universally expressed that because of the EMP and training, staff knew who to call in EHS and when to call them as it related to chemical clean-outs. Another result of the clean-out process under the EMPs is that it made laboratory staff and PIs more aware of what outdated chemicals were being held by individuals and thereby aided in the removal of unwanted chemicals. One PI at Boston College noted that only at the time of the clean-out was he made aware of the large quantities of chemicals that had been kept for a long period of time and were no longer needed in the laboratory. As a result, he was able to find another department to use the chemicals that were still in good condition (see EPI #4 on re-use).

Disposing of outdated chemicals seems to be a slow moving effort (partly due to the size of some of the relocation efforts involved at Boston College and UVM) and baseline values for this EPI have yet to be finalized. A lack of baseline values makes it difficult to measure progress in meeting this EPI.

The problem with measuring EPI #1 is that it is difficult to define what outdated means in the laboratory setting unless the chemical is determined to be a waste. According to the FPA, the EPI for HCOC chemicals is based on an "institution defined shelf-life." The schools have not been able to define what shelf-life means at each respective school-if a PI chooses not to use a chemical for 10 to 12 years, should the chemical be considered a waste? A universal problem also seems to be that certain PIs and researchers will not dispose of chemicals for any reason-making it difficult for EHS to get an accurate assessment of how well this EPI is being met. Additionally, older chemicals tend not to have expiration and "best-used-by" dates on the bottles. On the other hand, some researchers find uses for chemicals beyond their expiration dates. Therefore, the biggest challenge for EHS staff with this EPI is to define shelf-lives for chemicals and then to push faculty and PIs to remove larger quantities of unwanted chemicals over time so that EHS can aid the laboratory in chemical reuse. Clarifying definitions will help to define expectations.

The universities have obviously invested time and energy into setting up a system for inventorying HCOCs in laboratories, per EPI #2. As of 2002, these systems are either on the verge of full-scale implementation (i.e., UMB's bar-coding system) or are up and running (i.e. UVM's inventory process as part of EPCRA). Again, according to the baseline audits, HCOCs were an area that all three schools needed to pay more attention to. Across all the schools the HCOC approach is under refinement and this can be viewed as an important benefit. At Boston College and UMB, the Boston Fire Department is an important stakeholder in determining whether the HCOC approach adopted at either institution is effective or acceptable and both have been readjusting their approaches to accommodate the Fire Department. At UVM, the HCOC survey process in 2001 was disappointing to ESF staff. However with the planned changes of a web-based form and better timing for form distribution, UVM should expect better HCOC return rates.

The EMP is designed to be flexible and responsive to change, and HCOC survey process should be similarly adaptive. All the schools responded well to some disappointments and shortcomings in creating a baseline system and are attempting to collect results by tweaking their original plans and schemes. Hopefully these interim changes will continue to produce results over the remainder of the project and will allow for continual process improvements to be made over time.

Recommendations: It would help if the universities established a baseline value for their outdated chemicals of concern so that there is some way to measure improvement, which can give EHS better leverage to urge faculty and staff to adhere to the EPI. The participants have formulated their surveys to meet multiple needs, local emergency response regulations and federal reporting requirements. They should make clear what works as a most efficient system to define what outdated generally entails for laboratory staff. Given that there may be some cultural barriers and it may not best to strive for removal of 100 percent of outdated chemicals from laboratory shelves, there is some middle ground from which to measure future progress. EHS can set a best-estimate baseline for clean-outs that occurred in the last two years based on chemicals removed. Therefore, if certain laboratories have recently conducted clean-outs, EHS can track those laboratories over the remainder of the project to make sure that no additional outdated chemicals remain in those laboratories. EHS can track which chemicals are being stored and for what purposes in those laboratories that consistently hold on to outdated chemicals. Although this may be time consuming and may not change the behavior of already intransigent faculty or staff, EHS can provide laboratories with recommended holding times for certain types of chemicals. There is a financial cost to excessive chemical hoarding as evidenced by UVM paying more than $25,000 to a contractor in 2001 to dispose of unwanted chemicals from the Chemistry Department stockrooms. EHS can try to use examples like this to increase Administration support to generate a change in behavior.

Reduction in source chemicals will help prevent having outdated chemicals remain on the shelves. The ChemSource (discussed in EPI #4) program initiated at UVM before EMP implementation is a good way to promote efficient chemical purchasing by the laboratories and prevents unnecessary stockpiling of large quantities of chemicals within the laboratories. Boston College and UMB should investigate initiating a similar program that would be tailored to the size and potential demand at each school.

Once baseline assessments are complete at Boston College and UMB, it may be easier for these two schools to have their HCOC inventory methodologies approved by the Boston Fire Department. In general, the schools should collaborate on methodologies for completing HCOC inventories. For example, after UVM tests the web-based form, and it is deemed effective, the other two schools may want to explore a similar approach and can perhaps adapt the form as necessary. As the schools are partners in this effort, sharing of information and resources, such as the web-based form can help overall project performance. In addition, the HCOC inventory has served another purpose in providing information to laboratories that are interested in chemical sharing. During the group discussion, one graduate student from Boston College noted that she was comfortable sharing chemicals from trusted laboratories and would walk down the hall and look up the HCOC inventories for the chemicals that she needed. Having the HCOC lists computerized (as at Boston College) or posted in a known accessible area to encourage this type of ad hoc chemical sharing should be encouraged.

With the EMP now fully implemented at all the schools and there is more familiarity with its purpose and general concepts (as evidenced through the group discussions), the schools should see greater response to the HCOC inventory process. EHS staff at all three schools can reach out more to their graduate students to be the champions of the HCOC inventory process. During the group discussion, a graduate student at Boston College stated that she thought that students would take more responsibility for proper laboratory management if they had a better understanding of the direct environmental, health and safety impacts of the requirements. Perhaps more emphasis during training needs to be placed the "why" aspect of the EMP, in particular to the HCOC process, in addition to the process of how to achieve compliance.

9.2 EPI #3 Goal: Pollution Prevention Assessments

Pollution Prevention Assessments.Completed (EPI #3). According to the FPA, the universities outlined the goal for this EPI as identifying one P2 opportunity assessment per laboratory per year. Some of the P2 projects, where indicated, are taken from the June 2001 and May 2002 Project XL Annual Report. Anecdotal evidence obtained through group discussions conducted in March 2002 indicate that the NEU Labs project has engendered renewed awareness in pollution prevention on the three campuses. However, the three schools have fallen short of satisfying this particular environmental performance indicator. Reasons for lack of P2 activity can be attributed to a variety of reasons discussed in the proceeding findings section. Suggestions for improving P2 are provided in the recommendations section.

Many of the P2 projects discussed below took place at the schools prior to EMP implementation. This EPI goal as stated might be missing much of the pollution prevention work that goes on at the university level for certain programs, by focusing on P2 on a project-by-project basis.

Boston College: Two committees collaborated in the spring of 2001 to develop a list of P2 activities in progress as well as those planned for 2001-2002 academic year. P2 opportunities explored in 2000-2001 were the collection and reuse of computers and electronic equipment, a mercury thermometer swap initiative and the recovery of silver wastes from photographic operations. The Committee was focused on the potential of silica gel recycling, the reuse/redistribution of laboratory waste and the mercury thermometer swap program. In April 2001, Boston College sold approximately 75 used computers. Also, EHS has been working with The Institution Recycling Network on developing markets for electronic equipment, including discarded laboratory equipment, either for resale or for components. These aforementioned P2 activities are ongoing.

Boston College has one silver recovery unit for the Photography Laboratories. Small photographic laboratories in the Biology Department have individual silver recovery units attached to the plumbing of the automatic photo-processors. When cartridges reach capacity, they are replaced and silver is extracted from the used cartridges and sold by the vendor. The units are renewed as necessary with the generation of approximately two pounds of silver.

Boston College worked with Triumverate Environmental to find a recycling source for silica gel. Plans were made to ship gel to SiliCycle, Inc., in Canada. However, the paperwork requirements involved in shipping wastes across the border caused serious delays in this transaction. Boston College has yet to find another recycling source for silica gel. Boston College generates approximately 1,000 pounds of silica gel per year.

The proposed activities for the 2001-2002 academic year were designated as follows:
* Complete the administrative process for recycling of silica gel by September 2001. o Progress: Cancelled
* Promote replacement of mercury thermometers in laboratories through training, email, the EHS web site, and personal communications. o Progress: Ongoing
* Investigate less toxic glassware cleaning alternatives to propose to users of chromic acid and nitric acid. o Progress: Ongoing
* Analyze solvent generation; promote collection of certain organic solvents (e.g. acetone) as a "pure" waste stream, which can be distilled and recycled. EHS will contact CBG Biotech, a company that sells solvent to recyclers. o Progress: As of March 2002, Boston College was beginning to investigate possible opportunities to reduce and reuse acetone wastes. It was determined that CBG Biotech would not meet Boston College's needs. Acetone is used in chemistry laboratories to clean equipment and represents approximately 40 percent of all solvent wastes generated on campus. During the on-campus group discussions, students and faculty noted that changes in lab practices could reduce acetone usage and better segregation of acetone may yield recycling opportunities.

Boston College will be working on two additional initiatives for 2002-2003:
* Continuing discussions with waste vendors to search for an outlet for recycling acetone (one of the largest wastes by volume) and other solvents at the quantity levels generated by universities; and
* Hosting with EPA, C2E2 and other participants, a workshop in Fall 2002 to discuss development of P2 strategies for the type of research being done at Boston College.

EHS can report the following successes based on its efforts to date:
* Waste volume from the chemistry teaching laboratories has decreased by 67 percent due to the use of microscale chemistry procedures. * EHS continues to educate laboratories about alternatives to chromic acid cleaning solutions and has found that another laboratory is using a safer alternative.
* EHS has partnered with the Bakery Department (part of the Boston College Dining Services) to provide empty HDPE (high density polyethylene) containers to use as secondary containment in the laboratories.

University of Massachusetts-Boston: The focus for EHS and the Chemical Hygiene Committee is to place emphasis on pollution prevention through training. During training, emphasis is placed on pollution prevention and researchers are encouraged to incorporate product substitution, limited purchasing and other waste minimization strategies into their experimental design. In addition, EHS stresses the importance of purchasing only those chemicals that are needed and determining whether a treatment method can be incorporated as the final step in an experiment.

As of June 2001, the Chemical Hygiene Committee was developing a campus-wide program to replace mercury thermometers and a registration process for any remaining mercury containing devices on campus. As of June 2001, all six departments with mercury containing thermometer have replaced the mercury-containing thermometers. The Chemical Hygiene Committee is documenting the replacement activities and insuring that all mercury thermometers are replaced. In those instances where replacement is not possible, or the device is not a thermometer, the mercury containing device and its location will be registered with the Committee and the information maintained in a database.

UMB also has a Ph. D. program in Green Chemistry, which works to develop more environmentally benign chemical processes and products with in-depth knowledge of industrial operations and natural systems. A new Green Chemistry Laboratory for Research and Education in Sustainable Innovation is also in operation. This laboratory receives grants from private industrial organizations to find new alternatives for industrial processes. Opportunities to use Green Chemistry in laboratory methodologies and operations are currently being explored. As of 2001, a new campus sub-committee was formed tasked with "greening research." This sub-committee is part of a larger campus-wide Sustainability Committee. The sub-committee will be examining pollution prevention opportunities in the research community on campus. Additionally, the committee will use the results of the P2 surveys (survey attached in Appendix 4) and encourage PIs to explore new P2 ideas to investigate. Approximately 65 PIs received the P2 surveys. As of March 2002, EHS received 27 completed surveys back (approximately 40 percent return rate). The results of the survey are presented in Appendix 4. In general, the results indicate that the majority of PIs have not heard of any P2 opportunities that they require assistance in investigating or pursuing, and that the majority of them are not seeking assistance or resources to help reduce laboratory wastes.

University of Vermont: Before the EMP was implemented at UVM, ESF focused its efforts on three areas: (1) Photographic Chemical Initiative, (2) Chemicals in the Art Department, and (3) Mercury Thermometer Swap.

(1)ESF has been working with staff responsible for photographic darkrooms in an effort to reduce the hazardous waste generated as spent photochemicals.

In this ongoing program, the option selected for each darkroom depends on specifics of that darkroom's operation. ESF personnel offer assistance, as needed, with these efforts including educating users and collecting samples for analysis.

(2) Chemical wastes from Art Department studios are managed under UVM's EMP and therefore the benefits of previous P2 successes continue to be felt under the EMP.

(3) In 1997, UVM instituted a voluntary mercury thermometer replacement program. ESF staff swap environmentally friendly thermometers for the mercury thermometers at no cost to university staff and faculty. In November 2000, UVM was recognized for this program and was awarded the "Governor's Award for Environmental Excellence in Pollution Prevention."

P2 opportunities that have been addressed post-EMP implementation include the following:

(1) Replacement of formaldehyde-based preservative for tissue samples with preservative solutions containing much lower concentrations of formaldehyde. The alternative solution was developed by the UVM gross anatomy teaching laboratories in order to reduce formaldehyde exposures to students in the 1970s. A newly identified use is for historical samples in the Pathology Department that are retained for long periods of time, and whose preservative solutions must be changed regularly. The old solutions are hazardous waste due to formaldehyde content. The alternative solution has a much lower level of formalin in it. This P2 approach has potential application in a wide variety of medical laboratories that use similar preservative solutions.

(2) The chemistry department is assessing several introductory chemistry laboratory exercises to determine whether they can be redesigned to achieve a goal of "zero waste." Specific chemicals are being considered for replacement. If this effort is successful, similar methods can be used to assess other experiments in the Chemistry Department, and potentially other departments as well.

(3) The Agricultural Testing Laboratory produces significant amounts of corrosive wastes in the course of their analytical testing. Members of the ESF staff are meeting with laboratory management to determine whether process changes to reduce these amounts are feasible.

ESF staff through informal contacts with laboratory workers identified the P2 exercises described above. In order to more systematically identify P2 opportunities in the laboratories, the ESF laboratory audits will include a P2 questionnaire in 2002. This questionnaire (presented in Appendix 5) will provide the data necessary to identify which P2 opportunities provide the most potential for effective hazardous waste minimization.

Other potential P2 opportunities for 2002 may arise from UVM's Green Chemistry projects. The professor in charge of these projects is taking a novel approach to creating direct laboratory applications of green chemistry principles. The professor is working with a student environmental group and chemistry students to redesign introductory Chemistry courses to make laboratory activities more environmentally benign. This idea has shown some success and more activities are expected in the 2002-2003 academic year.

ESF's goal for this EPI for 2002 is that 60 percent of the UVM laboratories (counted on a supervisor basis) return the P2 questionnaires. Future year goals will be to increase this participation rate until 100 percent of the laboratory supervisors have returned the survey form by the end of the project. To encourage participation, the 2002 ESF laboratory compliance audit form will specifically give credit to those laboratories that participate in this program.

Findings: Although the P2 EPI goal reads as following, "complete one P2 assessment per laboratory per year," this wording suffers from confusion over what defines a 'laboratory'. Many of the discussions during the proposal phase of this project defined 'laboratory' as all the rooms under a single PI. If the goal is restated in these terms, it is slightly more achievable for the schools. However, much work needs to be done in this area across all three schools. This is a very difficult EPI for all schools, as P2 is not factored into research, with the exceptions of the Green Chemistry activities at UVM and program at UMB. Yet, this is the area where the schools will make the most environmental gains to attain superior environmental performance, a requirement for Project XL.

The P2 activities documented at the schools were activities that should be considered to be baseline-they existed or were implemented prior to the EMP. Therefore, now that the EMP is fully implemented, the universities must concentrate on generating new P2 opportunities and engaging the right individuals in P2 studies. Pollution prevention success will most likely result when there are champions of P2 approaches and there is increased environmental awareness on the part of laboratory workers-students, faculty and staff. The group discussion participants at all the schools indicated that the champions exist and that environmental awareness was more heightened with EMP implementation. The setting is opportune then for P2 assessments to emerge. Because this project is still in its early stages, it is useful to consider how P2 efforts are likely to evolve in the remaining years of the project. The discussion below reviews cultural changes that may enhance P2, systems that the schools have for promoting P2, and emerging P2 opportunities.

First, project participants note how the primary influence of the project on P2 may be a subtle and gradual cultural shift. On-campus discussion participants frequently noted how the project has made interactions between researchers and EHS more supportive and collaborative. Likewise, the process of developing and implementing the EMPs has raised the waste management knowledge and awareness of lab users. These changes may dovetail with pre-existing pollution prevention efforts (such as those in the Green Chemistry department at UMB and Green Chemistry projects at UVM) to produce P2 innovation. For example, a faculty member at UVM indicated that this project would help share waste management information among the whole lab community on campus, possibly allowing departments such as green chemistry to redirect their research in response to the most significant waste management problems. Similarly, Boston College laboratory users note that creation of chemical inventory sheets (a product of the NEU Labs project) has facilitated informal chemical sharing between labs in close proximity to one another.

While no formal P2 incentive programs currently exist at the participating schools, as mentioned above in the results section, each school is developing and implementing procedures for soliciting P2 ideas from lab users in the future:

* Boston College plans to integrate P2 into its hazardous materials and environmental awareness training. The trainers will use the training sessions to highlight the importance of P2 and encourage researchers to come forward with P2 suggestions.
* At UMB, EHS staff will be implementing a survey of faculty and students to identify potential P2 opportunities. For instance, this survey will explore whether lab users are receptive to reusing chemical bottles that have already been opened.
* UVM is promoting P2 through the school's green chemistry projects. For instance, one initiative will seek to redesign the introductory chemistry curriculum to encourage P2 and gather specific P2 suggestions.
* The three schools, EPA, and C2E2, will be hosting a workshop in Fall 2002 to discuss development of P2 strategies at colleges and universities.

In general, the data and the discussion notes suggest that the participating schools are in the early stages of implementing P2 in response to the regulatory changes introduced under the NEU Labs project. To characterize their progress, it is useful to think of P2 efforts along three basic "tiers", as suggested by an EHS staff member at Boston College:

* Tier 1: Easily implemented, small scale, product-specific waste reduction measures such as elimination of mercury thermometers.

* Tier 2: Reuse and recycling of significant chemicals and waste streams.

* Tier 3: More fundamental source reduction achieved through changes in research methods and lab practices (e.g., substitution of a less toxic chemical for a more toxic one already in use).

In the broadest terms, schools have made progress on Tier 1 (prior to the XL project) and have begun making progress on Tier 2; at least some of this progress is attributable to the NEU Labs project. However, the most significant P2 opportunities lie in Tier 3 and have not yet been fully explored.

A recurring theme in the discussions was how institutional factors can impede P2 innovations; especially design/research changes that yield reduced chemical usage, substitution of less harmful chemicals, or other source reduction. While EHS actively pursues waste minimization, most of their influence is limited to Tiers 1 and 2 described above and their focus is on management of waste once it is generated. In contrast, Tier 3 source reduction opportunities can be achieved only with input and support from researchers, especially principal investigators in charge of research plans. Faculty, particularly individuals at Boston College, noted the absence of a clear incentive system that would fuel academic interest in P2 innovation. While quality research is typically rewarded with publishing opportunities and other forms of professional advancement, the relevant academic disciplines do not offer such incentives for laboratory-level P2 research12.

Practical constraints associated with laboratory settings may also limit design-stage source reduction. First, discussion participants noted how a researcher who has successfully implemented an experiment would be hesitant to change the approach in the interest of exploring P2 possibilities. Second, it may be difficult to identify widely applicable P2 measures because laboratories vary greatly in terms of the chemicals used and the wastes generated. In general, there are no large batch processes that generate standard waste streams, as found in industrial operations. Similarly, research technology develops rapidly, making it difficult for P2 innovation to keep pace.

The awareness created by the NEU Labs project has combined with pre-existing incentives (e.g., cost saving, risk reduction) to generate new P2 plans at the schools. Participants called attention to emerging P2 opportunities during the on-campus discussion. At Boston College, discussion participants highlighted reduction of acetone wastes as a potential P2 opportunity. Acetone is used extensively in chemistry labs to clean equipment and represents roughly 40 percent of all waste solvents generated. Participants, particularly the graduate students present, noted that changes in lab practices could reduce acetone usage and better segregation of acetone may yield recycling opportunities. At UVM, researchers have developed a non-toxic substitute for formaldehyde, a chemical used extensively in the medical school labs. These kinds of innovations suggest the schools may be integrating P2 more explicitly into research protocols and lab practices.

Overall, the NEU Labs project appears to have laid the foundation for P2 innovations at the participating schools, although only limited P2 progress has been realized to date. It is difficult to provide systematic suggestions for improving P2 performance, since this is a technical subject that is heavily dependent on the research program and other institutional factors at each individual school. However, based on the information discussed above, and on other observations offered in the on-campus discussions, the broad recommendations in the following section may warrant examination.

Recommendations: First, the participating schools should continue finding ways to improve communication between EHS and the lab users as a means of promoting P2. In general, EHS staff should send the message to faculty and students that P2 is an important aspect of complying with the school's EMP. The schools may benefit from considering each other's P2 promotion strategies. These include surveying lab users on potential P2 opportunities (as at UMB); featuring P2 in the training offered to lab users (as at Boston College); and working through green chemistry projects to redesign curricula to incorporate P2 (as at UVM). Second, communication with the students and staff conducting the research is imperative-the group discussions suggest that a whole group interested in P2 may exist, however individuals do not have the opportunity to meet regularly and share ideas.

At a more general level, EHS and lab users should work together to identify opportunities to share knowledge and promote university-level P2 in the academic community. The project participants could use forums such as the 2001 Green Chemistry Research Symposium (held at University of Massachusetts-Amherst) to gather ideas for P2 measures applicable at the NEU Labs schools and to promote the project with researchers at other schools. As noted above, the participants are hosting a workshop in November 2002 to discuss the development of P2 strategies. This kind of interaction may help publicize the need for waste minimization in university laboratories and gather suggestions from researchers working outside of the three participating schools. It might also help make in-lab P2 research a more visible field; while most green chemistry research currently focuses on refinement of industrial and commercial processes, many of the scientific principles may be equally applicable in a research laboratory setting. UMB and UVM can collaborate on the Green Chemistry activities taking place on their respective campuses. The professors engaged in these activities are already champions of P2 and would most likely be the most open to working together to further the goals of this project. For example, UVM's Green Chemistry professor is engaging a student group to develop a Green Chemistry approach to basic chemistry classes. A similar approach can be tested at UMB as well.

As an administrative action-it may be best to restate this EPI as the universities intended it and not leave it as it currently exists as stated in the FPA. In a regulatory experiment such as this project, all parties should endeavor to clarify goals and expectations so that the results clearly reflect the best efforts to attain those goals. The universities are leaving themselves open to questions as to why they are falling so short of a stated goal, when in fact the stated goal is not what they set out to accomplish.

Finally, EPA can help promote P2 among the NEU Labs schools (and elsewhere) by taking on a technical assistance and facilitation role. Options include the following:

* Consistent with preliminary ideas offered during the Boston College group discussion, EPA could sponsor and facilitate a workshop with chemical vendors, equipment manufacturers, and researchers to examine waste minimization opportunities. This idea would seek to implement P2 further up the supply chain for universities. In November 2002, EPA is sponsoring with Boston College a Pollution Prevention Conference.

* EPA could explore a variety of incentive mechanisms for encouraging lab-level P2 among the XL participants (or on a wider scale). For instance, the agency could offer a competitive grant, soliciting proposals for P2 from the participating schools. The award criteria could favor P2 measures that address especially large or toxic waste streams, and measures that are widely applicable at university labs across the country. To enhance the incentive, EPA could coordinate with professional journals to plan for publication of an article on the winning P2 innovation.

* EPA could assist the interested schools with an application for a Green Chemistry research grant and encourage the schools to apply for funding with other grants and scholarships offered through organizations such as the National Environmental Technology Institute, The Green Chemistry Institute, and the Center for Process Analytic Chemistry. More information about these and other organizations can be obtained at http://www.epa.gov/greenchemistry/grants.htm.

* EPA could assist with a pilot of the Chemical Management Program being run and coordinated by EPA's Office of Solid Waste. This program is designed to do centralized chemical purchasing with an eye on tracking chemical movements to highlight areas for reuse and reduction. The idea would be to work with a consortium of colleges to have a more robust program, given that schools' chemical waste streams are small and not concentrated. This program would require an investment of financial resources in order to help implement the program.

9.3 EPI #4 Goal: Increase chemical reuse/redistribution by 20 percent from baseline, EPI #5 Goal: Reduce hazardous waste generation by 10 percent

Amount reused or redistributed within the institution (normalized and compared with and without RCRA in the lab) and cost savings (EPI #4). The assumption behind EPI #4 is that relieving laboratories of the requirement for making a RCRA hazardous waste determination will remove certain laboratory chemicals from the waste stream and result in more redistribution and reuse of laboratory chemicals on campus. This EPI is therefore tied to total laboratory wastes per institution and cost savings (EPI #5), to reduce hazardous waste. The goal of a 20 percent increase (over baseline values) in reuse/redistribution of hazardous chemicals collected from laboratories over the life of the project, would help to meet the goal of a reduction in waste disposal of 10 percent (from baseline values) (see Table 4). These goals were expected to result from better management and more time availability to devote to chemical reuse and recycling under the EMP. The only existing data on chemical reuse in an academic setting is derived from the Campus Safety, Health and Environmental Management Association figure that approximately only 1 percent of chemicals are re-used or recycled. Therefore, these goals were meant to be far-reaching. The baseline data was derived from numbers generated in conformance with RCRA reporting requirements.

Although laboratory waste reduction is a meritorious goal for the universities, each school has encountered practical constraints with achieving this EPI. It is worth stating that the universities have made good faith efforts in trying to reduce laboratory wastes. The reasons that have made this goal unattainable for the universities are noted in the findings section.

Boston College: The Chemical Redistribution Program is explained in detail in the Standard Operating Procedure of the EMP. The Redistribution System began in March 2001 with an influx of chemicals from laboratories that were being relocated. During March 2001, EHS redistributed numerous cleaning supplies, four containers of lighter fluid, a Coleman fuel cylinder, a liter of hydrochloric acid, some salts, acids, bases and ethanol that will be used in EHS's waste identification program. In August 2001, Boston College distributed an electronic chemical inventory list to all laboratories, printed the list in the newsletter and posted it on the EHS website. EHS collected data on materials received and distributed.

Since the laboratory moves were completed in 2001, EHS notes that there have been no new chemical additions to the redistribution program. Virgin chemicals are not a regular part of the waste stream at Boston College. However, EHS has learned that chemical redistribution happens informally between laboratories in two ways: (1) As graduate students, post-doctorate students and other laboratory workers leave the school; their chemicals are inherited by new workers who take over the projects or by other personnel in the laboratories, and therefore prevent orphaned chemicals. (2) Laboratories conducting similar research and are typically located close to one another and will share chemicals across laboratories. Chemical sharing in this manner happens when the other laboratory is considered a "trusted source." Compared to waste generation numbers for 1999, the amount of waste generated in 2001 by all laboratories increased by 55 percent. In-depth analysis of these data has shown that 80 percent of the laboratory waste at Boston College is generated by six laboratory groups, which comprise 15 percent of all laboratories. These laboratories are in the Organic Chemistry Division and one laboratory in Biochemistry. * * * * Since Boston College established the baseline, the Chemistry Department has received numerous research grants, especially in Organic Chemistry and Biochemistry, which enabled students and PIs to increase the research conducted. Naturally, waste increased by a very large amount. Furthermore, Boston College has stated as an academic mission that it will become a top research institution, and the Chemistry Department Master Plan includes the addition of five faculty members (and laboratories) to the department in the areas of Organic Chemistry and Biochemistry. Although Boston College is unable to meet the goal for waste minimization, EHS views the EMP as a valuable tool as it allows for better management of the volume of waste produced and better scrutiny of waste generation.

EHS has interviewed PIs and faculty to try and determine what options are available to address the waste volume problem. EHS is going to concentrate on the following issues to minimize waste generation:

* Implementation of a training program to better separate solvent wastes, in order to maximize the material going to fuel blending, and minimize the volume of material contaminated with halogenated compounds that goes straight to incineration.
* Purchase containers to assist in the solvent separation.
* Continue discussions with waste vendors to search for an outlet for recycling acetone and other solvents at the quantity levels generated by universities.
* Continue training focused on Pollution Prevention, and include the concept of Green Chemistry.

University of Massachusetts Boston: As determined from university manifests and the RCRA biennial report in 2000, the university generated approximately 3,711 pounds of hazardous waste generated in laboratories14. This decrease in hazardous waste generation was an 11.76 percent reduction in waste generation compared to 1999 (5,585 pounds). EHS attributed this reduction to smaller numbers of acutely hazardous wastes, organic peroxides, pyrophorics, flammable liquids and compressed gases. There were slight increases in overall amounts of corrosives, flammable solids and oxidizers. The EMP was implemented in October 2000; therefore it is not possible to link the reduction in chemical waste generation to the EMP.

In January 2001, EHS sent out a pamphlet to all principal investigators describing the purpose of a re-use and redistribution program. A formal reuse and redistribution was not in place prior to the XL project. The pamphlet also contained a tear-off sheet for PIs to fill out and return to EHS if they had any material available. EHS also introduced and promoted the program during training sessions.

In June 2001, EHS collected approximately 20 liters of materials. In May 2002 EHS completed an inventory list of excess chemicals. EHS published the list materials available for redistribution on its website (http://omega.cc.umb.edu/%7Eehs/labindex.htm) so that it is easily accessible once the hazardous waste accumulation area is completed by summer 2002. EHS notified all PIs about the list via email. When materials are identified as potentially reusable, they are labeled with the date. Each time the materials are used, they are tracked by EHS. If materials are in storage for more than two years, they will be disposed of. EHS obtains information concerning redistribution possibilities from direct mail, email, departmental postings, laboratory decommissioning and laboratory waste pickups. EHS expects to have use data of the excess chemicals in 2003. University of Vermont: UVM's hazardous waste generation for 2000 was 38,269 pounds from research and teaching. UVM also has a Part B storage facility regulated under RCRA's Treatment, Disposal, and Facilities regulations, at which laboratory waste is sorted and repackaged for more economical disposal. The amount of waste shipped from campus has been fairly steady over time. Under pre-EMP conditions, the amount of hazardous waste disposed of in 2000 was 4 percent more than in 1999. This was well within the standard deviation around the average amount of laboratory waste generated during the 1990's (36,800 +/- 13 percent).

Table 4 for 2001 shows the laboratory waste generated by UVM laboratories, less the amount generated by the 2001 clean-out of the Chemistry and Agricultural Biochemistry Departments. The data demonstrates that laboratory waste generation dropped significantly in 2001, when chemical clean-out data is not included in the total laboratory waste generation value for UVM. ESF believes that the decrease in laboratory waste generation can be attributed to the ongoing presence of ESF staff in laboratories as they conduct laboratory audits and increased awareness of inventory management. These audits, which began in 1998 as "chemical storage surveys" generally, result in the disposal of chemicals that are recognized as surplus. ESF expects to see a continued decrease in the amount of waste generated. Other factors also affect the amount of laboratory chemical waste generated in 2002. A new medical research building was opened on campus, resulting in the movement of a significant number of laboratories between and within medical college buildings. These moves resulted in clean-outs of individual laboratories, which were processed, along with routine waste, at the Environmental Safety Facility. Similar clean-outs of campus laboratories occurred outside the medical college in preparation for the EPA/VT DEC audit. Most of these wastes are included in the amounts show in Table 4 (some of this waste was disposed of in 2002 and will be accounted for in the 2002 reported numbers).

ESF believes that based on current trends, it is possible for UVM to meet the goal for this EPI.

The baseline waste generation values are based on UVM's hazardous waste annual reports generated for VT DEC. Because UVM's Environmental Safety Facility is a Part B storage facility, UVM is required to file two reports-one for the waste streams generated on campus and another for those shipped out of the ESF. The numbers reported here are the amounts shipped from the campus to the facility. UVM used these numbers because at this point in the waste handling process, laboratory waste streams are easily differentiated from other campus waste streams. The amount of hazardous waste shipped from UVM's laboratories (about 550) has been reasonably consistent from 1995-1999, with an average amount of just over 36,000 pounds during that period. It should be noted that the 1996 number does not include a large chemical clean out of the Chemistry Department that took place that year. This clean out produced more material than expected (about 11,000 pounds) and was not representative of a single year's waste production. The annual variation from average of laboratory wastes (less than 10%) is much less than that observed for other campus wastes, whose totals are often driven by large construction and renovation projects which produce oil contaminated soils, lead paint debris, and other sporadic hazardous waste streams. The most significant hurdle that ESF has found in instituting a laboratory waste reuse program has been that most laboratory workers are reluctant to use materials of uncertain quality. This trend is universal to all three institutions and is not unique to UVM alone. If researchers receive a chemical from a known, trusted source, he/she is more likely to use it. This process is not formal and therefore difficult to track or document. Most laboratory workers prefer to use chemicals directly purchased from chemical suppliers. Therefore, ESF combined its chemical recycling program with a chemical distribution program called ChemSource prior to the implementation of the XL program.

ChemSource, which has been operating for six years, involves ESF staff buying new chemicals in case lots and breaking down those case lots in individual containers so that laboratories can obtain necessary chemicals at a cheaper cost without purchasing them in excess. This aspect of ChemSource works in combination with the redistribution of chemicals discarded by laboratories. ESF measured the activity for this program from 1996 to 2000 (prior to EMP implementation). Based on the data collected (Table 5) ESF believes that strong patterns or trends for ChemSource use prior to XL had not yet developed. Table 6 shows the 1998 through 2001 results of the ChemSource program. This is expressed in the number of ChemSource orders delivered, because there is no common unit of measurement for the various chemicals delivered as part of this program. While the sale of new chemicals continues to grow as more laboratories participate in the program, the amount of recycled chemicals has not.

In 2001, ChemSource publicity efforts included representation at the UVM purchasing fair, at the scientific vendors fair, and the UVM Environmental Fair, a letter to the Chemistry Department in August and to chemical buyers in November, and joint projects with vendors to meet specific needs of chemical buyers on campus. ESF has established a goal for the ChemSource program and would like to see new chemical deliveries grow by 10 percent for 2002 and that the amount of reused chemicals delivered by the program increase by 50 percent. Progress towards these goals will be achieved by continuing outreach activities similar to those described above.

In 2002, ESF will focus on increasing both the amount of new chemicals and reusable chemicals redistributed. The expansion of the program will help minimize the amount of reusable chemicals generated by laboratories. Findings: The universities have had and will continue to have a difficult time promoting, documenting, and achieving EPIs #4 and #5. The cultural barriers-voiced at all campuses visited-are a stumbling block to making official progress on these indicators. At all three universities, students and researcher faculty alike echoed the sentiment that chemical purity and quality assuredness is understandably of utmost priority for scientific research. Therefore, it is very unlikely for a researcher to use a previously opened or used chemical liquid, although he/she may consider using an opened chemical powder if the purity can be affirmed. Similarly, all stated that there was great hesitancy about re-using chemicals that had been taken by EHS. Although all the schools are beginning to institutionalize some formal chemical redistribution program-it does not seem likely that there will be a great deal of usage of these programs. What is promising, however, is that there are more informal chemical sharing opportunities that seem to be occurring between laboratories, and the participants should promote and capitalize on these opportunities.

The successful element of EPI #5 is that the schools generated baseline values of waste generation and they have been tracking their waste generation in comparison to the baseline. However, by looking at the waste generation numbers, it is clear to see there is fluctuation in the waste generation numbers and that it is difficult to characterize the average amount of waste generation for a lengthy period of time at each school. Although UMB was able to meet its reduction goals, all three schools have been struggling with the need to meet this EPI, for laboratories to conduct chemical clean-outs of outdated chemicals per EPI #1, and to complete laboratory clean-outs prior to laboratory relocations and moves. These are conflicting goals, as clean-outs and removal of outdated chemicals of concerns will increase waste generation. Another cultural barrier to achieving this EPI is that certain established research protocols require heavy chemical inputs-and there is currently no readily available alternative to researchers for certain protocols. Therefore, if an increase in research occurs, there will be a correspondent increase in research waste. Given the research culture and the need to do laboratory clean-outs, it is difficult to reconcile the need to meet the waste reduction goal in its current form.

Recommendations: EHS should promote informal chemical sharing opportunities by using the HCOC inventories. First, for example, if the HCOC is web-based or in an electronic database (currently being explored by UVM), EHS can match laboratories and send a notice alerting them to the fact that those similar chemicals are being used in a variety of laboratories. Again, the chemical sharing must occur with chemicals on shelves and not with chemicals tagged for EHS pickup. The easier the process is made for a researcher, the more likely he or she will make efforts to share chemicals. Second, more chemical sharing should be encouraged for student teaching. In basic science courses, laboratory curricula are defined well in advance that chemical sharing can be maximized. For these laboratory exercises, chemicals should first be pulled from the EHS cache of used chemicals before new chemicals are purchased, as the experimental purity is not of higher import than the learning process for the students. Third, chemical sharing may increase, where deemed appropriate by the researcher, as it is stressed in training.

The goal for EPI 5 as it stands does not meet the cultural research needs or the other EPI goals of this project. For the next two years of this project, it may be a better environmental goal for the schools to pursue a source reduction strategy. Given that there may be little room for improvement with more advanced research taking place at each university-EHS staff should focus on those processes where there is flexibility in research protocols. For example, a switch to microchemistry or green chemistry at each school in introductory Chemistry classes, might results in larger and more lasting environmental gains. Additionally, students in introductory classes will be taught about the benefits of these new approaches and will have an environmental awareness that they will carry with them throughout their academic experiences. Perhaps, if the goal of the EPI was to assess and implement at least one source reduction initiative, it is possible that the schools can see more lasting measurable effects of an EPI that is not affected by the clean-outs, is not research dependent, and raises the environmental awareness of its students.

9.4 EPI #6 Goal: Assess and demonstrate improvement in environmental awareness by using an environmental awareness survey

Survey Scores (EPI #6). The purpose of the survey is to provide a standard by which to evaluate the success of hazardous materials and environmental awareness training. The survey also helps compare environmental awareness across campuses. The Environmental Awareness Survey developed for the project was a cooperative effort among the three universities. A survey specialist worked with the Environmental Health and Safety Offices at each of the universities to develop and finalize the survey instrument. The survey tests laboratory worker awareness in the following four major categories:

(1) awareness of appropriate disposal regulations;
(2) awareness of appropriate laboratory practices identified in each school's EMP;
(3) awareness of the environmental impact of laboratories; and
(4) awareness of the public health/safety impact of laboratories.

The survey was administered in 2000 to obtain the baseline values and consisted of 16 questions total.

The following table presents the questions asked on the survey broken down by the four categories. There are three different sets of survey results referenced below

(1) the baseline survey administered in 2000 before the implementation of the training program;
(2) the first survey administered after the first year of training (referred to as the post-XL survey) in 2001; and
(3) the 2002 survey. The original survey questionnaire is presented in Appendix 6.

It is difficult to differentiate a pre-XL survey and a "post-XL survey" for this EPI as the three schools had training programs in place prior to the XL project and ongoing training while the EMP was implemented. In general, the data show that the post-XL training has enhanced environmental awareness at all three schools, although variable data collection and analysis methods should be taken into account when examining the results. Table 6 presents summary statistics on the population surveyed in both rounds, the number of participants representing relevant university populations, the survey delivery method, and response rates. Table 7 presents the baseline and post-XL survey scores for relevant questions. Again, based on the results, it is possible only to say that there appears to be a heightened environmental awareness on all three campuses as time elapsed between the first survey and the second survey. The survey distribution and target populations varied at each school. The variation in survey administration produced results from which we can only create a general picture of environmental awareness at each school and affects the way in which we analyze the comparative results of the environmental awareness survey. Because of the different test populations, it is not possible to attribute to the general environmental awareness improvement to the training.

In general, across the three universities in the post-XL survey, most laboratory workers did not have a good understanding of laboratory environmental impacts or pollution prevention concepts. However for the 2002 data, the surveys show that despite some improvements in these areas, the awareness of the surveyed population seems to have leveled off. Based on these results, the schools are faced with a number of the following challenges associated with both the training and the administration of the survey:

(1) what is the best way to track and train undergraduate students and transient laboratory workers?;
(2) should the physical impacts of laboratory activities be emphasized or can this information be distributed through alternative communication channels?
(3) and what is the relationship between the survey scores and EMP compliance?

Boston College: The baseline survey was sent to all science faculty and a random selection of graduate students from lists supplied by the departments. The surveys were delivered to people through a combination of mailing and hand delivery, and respondents were asked to return the survey through the mail. A gift "give-away" raffle of a $50.00 gift certificate was provided to encourage participation and improve the response rate. However, EHS received complaints about not having the opportunity to win the raffle (since certain individuals did not receive the survey) so a second round of surveys were sent out. It is not known how many surveys were mailed in this second round. A graduate student also surveyed an undergraduate class of 25 students. Although 88 surveys were returned, a response rate was not estimated due to uncertainty about the total sample population that received the survey. EHS staff noted that because of the "give-away," some individuals who were not originally targeted for the survey obtained photocopies of the original survey and returned them to EHS.

The survey delivered post-XL training utilized a similar administration method to deliver 100 post-XL training surveys. All science faculty and a random selection of graduate students (chosen from lists supplied by the departments) received the survey. Again, EHS offered a $50 gift certificate as part of a raffle. Through a hand-count of returned surveys, approximately 19 surveys were returned through the mail, generating a response rate of 19 percent. The post-XL survey population was different than the baseline survey population, as the survey did not target solely those individuals who completed the baseline survey.

For the survey administered in 2002, EHS used a student worker to canvass laboratories and staff. The student went to each department and dropped off surveys with people he encountered, and later the same day went back collect the completed surveys. The student distributed 63 surveys over two days and collected 45 completed surveys for a return rate of 71 percent. As incentives for completing the survey, each person received a "BC Labs XL" pen, and names were collected for a $50 raffle. As an additional measure, the students handed out the survey answer key when people returned the survey. The student worker noted that in some laboratories, the surveys appeared to be group efforts and that access to certain labs was difficult due to locked laboratories and no staff present.

The survey in 2002 shows no great improvement in scores from those obtained in the post-XL training surveys. The 2002 survey population included six undergraduates, a group that does not receive training. The 2001 survey had a very small sample size and was completely voluntary, so the returns received may reflect better scores from a small population that has an unusually high interest in the EMP. For 2002, the survey was distributed to a larger group, including many people who may have not completed the survey if not personally approached. The change in methodology and sample sizes does not lend any meaningful comparisons or conclusions. EHS plans to use the survey distribution methodology used in 2002 so that the 2003 results will be more meaningful. Additionally, in the past correct answers to the survey were not distributed so participants were unable to know the correct answers and see where they made errors. The distribution of the correct survey answers in 2002 was received with interest and may have more educational value in the long run.

University of Massachusetts at Boston: An initial master list of all past individuals who had been trained by EH&S was used for the baseline survey. The list included a number of individuals who were no longer at the university. The survey was initially sent through the mail to 150 individuals who were asked to complete the survey and then send it back to the EHS office. After receiving a poor response rate, EHS sent an unknown quantity of additional surveys to others on the list in order to encourage more participation. The response rate cannot be determined due to uncertainty about the total number of surveys administered through the mail, however 88 completed surveys were returned.

In 2001, UMB's training survey was randomly administered through the mail to 250 individuals-including those who had not received training. Approximately 54 individuals responded to the survey, generating a response rate of 21 percent. UMB conducted hand-counts of the second survey and summarized the findings on its website. As with Boston College, the post-XL survey population differed from the baseline survey population.

Approximately 60 people responded to the survey in 2002. EHS used the same survey distribution method as in the previous years. Although there is little change in the responses, there seems to be better responses related to the general environmental awareness questions. Correct answers on three questions-waste generation, fume hood emissions, and environmental impacts of laboratory work-rose 6 percent, 15 percent, and 10 percent respectively, from the 2001 scores. An interesting result of the survey in 2002 was that while the percentage of respondents trained in the EMP decreased by 20 percent, the percentage of those respondents who could identify the document governing the university's laboratory waste regulations increased by 3 percent. This may indicate that perhaps those who have been trained in the EMP may be more environmentally aware than those who have not.

University of Vermont: UVM's baseline survey relied on a directory of lab users (including faculty and staff) to randomly identify respondents, selecting a target sample of 100 individuals. ESF staff visited laboratories within the 21 academic departments and located the individuals or co-workers and asked them to participate. Individuals either (a) answered questions orally (i.e., in-person administration); (b) completed the survey on their own and returned it to the surveyors (i.e., self-administered); or (c) referred to the surveyor to a separate individual in charge of environmental safety for that lab. To encourage participation, ESF staff provided an incentive gift to all survey participants.

The first post-XL training survey was distributed using a similar master list of laboratory personnel. Again, 100 surveys were completed, both in-person and self-administered. UVM conducted hand-counts of the post-XL survey data and summarized the findings on the website. In addition, UVM entered the post-XL results into a Microsoft Excel spreadsheet along with the baseline results. This database was used to derive the individual results found in Table 8. The availability of data on the number of post-XL survey participants that had previously received XL training allowed for a supplemental analysis, presented in the "% Trained Respondents Only" column in Table. Figures in this column represent the percentage of correct responses for each question among those who received the post-XL training. Of the 100 post-XL training respondents at UVM, 86 had received training.

In 2002, UVM completed the post-XL survey for the second time. UVM used similar survey methods to the first time. In general, the responses show little change from 2001, although there is still significant improvement from the 2000 results. Improvement in 2002 was noticeable on specific questions, generally related to those about general environmental awareness. Correct answers on these (wastewater treatment, fume hood emissions, labeling requirements) increased between 3 to 5 percent. It is interesting to note that less than 50 percent of the population can recall the phrase "Environmental Management Plan" as the name for UVM's waste management program. This may indicate that retention of information is higher through hands-on applied procedures in the laboratory rather than through the distribution of information on the overall management structure and process.

Findings: In general, the data show that the post-XL training has enhanced environmental awareness at all three schools, although variable data collection and analysis methods should be considered when examining the results. Since survey distribution and analysis methodologies differed at each school the findings for this EPI are separated by school.

Boston College: The general upward trend in overall environmental awareness at Boston College laboratories is similar to the results found at the other universities. All key questions in each category of awareness demonstrate improvement over the baseline, despite a wide range of baseline understanding. For example, while nearly two-thirds of the respondents correctly identified EPA as the Federal agency that regulates the disposal of chemical wastes (Question #1), the post-XL training results indicate further improvement, as 88 percent correctly answered the question in 2001. Similarly, the baseline (Question #12), fewer than 10 percent of the respondents were unable to identify that the largest environmental impact of laboratory is high-energy use. Following XL training, the correct response rate increased to 25 percent. The Boston College results should be interpreted carefully, however, because Boston College's post-XL training survey includes only 16 respondents, which may not provide statistically significant findings.

University of Massachusetts Boston: The post-XL training results indicate noteworthy improvements at UMB as well. For example, while only seven percent of respondents could identify collection for hazardous waste disposal as the required disposal method for strong mineral acids (Question #6), nearly two-thirds of the respondents identified this answer following XL training. Likewise, less than a quarter of the respondents could initially identify the correct treatment method for laboratory wastewater (Question #9), but the majority selected the correct answer in 2001.

Consistent with the results of the survey analysis at the other universities, UMB respondents portray a poor baseline understanding of the environmental impacts of laboratories. For example, in both the baseline and post-XL training survey, approximately one out of eight respondents was able to identify energy use as the largest environmental impact (Question #12). One unpredictable survey result is the apparent decline in respondents' understanding of chemical waste treatment. The percent of respondents that correctly identified incineration as the most common chemical waste treatment dropped from 31 percent in the baseline to 17 percent following XL training (Question #2). There does not appear to be a clear explanation for this downward trend. IEc confirmed that approximately 31 percent of the respondents correctly identified incineration for Question #2 in the baseline, but could not confirm the apparent decline in understanding as indicated by the post-XL training results because the raw UMB survey data are not available.

University of Vermont: The availability of survey information in spreadsheet form provides an opportunity to conduct a more thorough and reliable analysis of the UVM awareness survey. The data generally show that the XL training yielded an increased understanding of the environmental and human health impacts of laboratories. The results for the post-XL population indicate that improved environmental awareness occurs across both the trained and untrained respondents. In other words, for the post-XL training survey results, the difference in correct responses between those that had received training that year and those that did not, were not significant15. One possible reason for this trend is that environmental awareness across the targeted population may have increased due to a general dispersion of knowledge from those who received training to those who did not. If this hypothesis is correct, EHS departments that face the logistical challenge of providing training to a laboratory population with a high turnover rate may still achieve the lasting benefits of improved environmental awareness.

Improved environmental awareness is demonstrated in two different ways. First, where the baseline survey indicates poor understanding prior to training, an improvement in awareness is evident. For example, the baseline survey indicates that less than one-third of respondents could identify the threshold amount of acutely hazardous waste that can legally accumulate in the laboratory (Question #7), but post-XL training results show that number nearly doubled. Second, where the laboratory population seems to exhibit significant prior knowledge (i.e., at least 50 percent of the baseline respondents could identify the correct answer), awareness also appears to improve. For example, more than two-thirds of the respondents already understood EPA's role in regulating hazardous waste (Question #1), but that awareness improved following XL training, with 84 percent of the respondents selecting the correct answer.

In other areas-particularly in the category that covers awareness of the environmental impact of laboratories-respondents still stand to make significant improvement over the course of the XL pilot. Approximately three-quarters of the respondents had trouble identifying incineration as the most common disposal method for laboratory hazardous material (Question #2); a similar number could not identify energy use as the largest environmental impact of laboratories (Question #12). While respondents demonstrated relative improvement on both of these questions following XL training, fewer than half could identify the correct response.

Recommendations: To improve the clarity and reliability of the findings, the participants should consider refinements to current survey administration and data management practices. First, to improve the effectiveness of the survey as a measurement tool, the schools may want to clarify the survey's intent. Given that the FPA does not dictate detailed objectives for the survey, participating schools should be sure to address this question. If the project managers feel that the survey is primarily a tool to assess overall environmental understanding among lab users, then the basic survey approach used thus far is generally adequate and can be refined through a variety of steps outlined in greater detail below. In contrast, if the survey specifically seeks to measure the effectiveness of XL training, then more fundamental changes to the survey instrument and survey method may be appropriate in future survey rounds. The discussion below provides a separate set of recommendations for this scenario.

If the intention of the survey is measure overall changes in environmental awareness, the following modifications should be initiated:

* For several reasons, Boston College and UMB should consider in-person administration of the survey (as done by UVM). Although this method requires additional time and resources, it is more likely to generate a large and statistically significant sample of respondents and therefore provide more robust environmental survey results16. The results would also be more robust because in-person administration would discourage collaboration on answers and therefore measure individual environmental awareness. To encourage participation, the schools may wish to offer material incentives to targeted respondents, but may want to avoid a "lottery" type giveaway that encourages unintended participation (as evidenced by what occurred at Boston College). UVM found that providing a small incentive to all participants was effective in encouraging survey participation.

* To ensure meaningful findings, the schools should distribute the survey more systematically across lab user sub-populations (e.g., workers, students, faculty, etc.). In-person administration will enable this kind of targeting because it avoids the response bias that can arise in voluntary mail surveys. For example, it would help avoid the problem Boston College encountered when surveys were duplicated and distributed to whole undergraduate classrooms. The resulting data will allow more reliable analysis of awareness changes among sub-populations and may provide findings useful for refining training or other environmental awareness enhancement actions.

* To facilitate future data analysis, schools should practice better data management. First, all hardcopies of the completed surveys should be stored carefully; loss of records from earlier survey rounds (as in the case of UMB) will undermine future analysis of awareness changes. Second, schools should enter baseline and subsequent survey responses into electronic databases (e.g., Microsoft Excel or Microsoft Access), as was done at UVM and Boston College. Appendix 7 presents a print out of Boston College's Excel spreadsheet containing post-training survey data, which can serve as a useful template for future survey data management.

* Schools should use the electronic databases to pursue more thorough analysis of the survey data. As noted, one area of interest might be analysis across different respondent attributes. Likewise, electronic data may allow more systematic analysis of open-ended survey questions17. For example, schools could analyze the frequency of terms or concepts provided in response to open-ended questions by electronically searching text fields in the database. As noted by the survey specialist that aided in developing the survey instrument, open-ended questions provide an effective way to measure respondent recall of certain issues (as opposed to multiple-choice, which tests recognition), and may also provide information on common misunderstandings among the laboratory population.

As noted above, if schools determine that the primary intent of the survey is to measure the effectiveness of the XL training, additional survey changes may be appropriate. Most fundamentally, schools may want to consider administering a pre-training (i.e., baseline) survey as well as a post-training survey, rather than relying on existing data for baseline information. Although performing the full survey sequence would potentially demand significant resources, it would allow schools to pursue several refinements:

* First, new questions could be added to the survey. For example, if the training is modified, new questions could be added to track the effectiveness of new training elements. The sections of the survey that might be added would be useful for comparison to if previous survey baseline data, i.e., new questions will have no point of comparison in the old data.

* Second, the survey administration method could be changed to a panel design. In a panel survey, the same individuals would receive the survey before and after the XL training. This design allows more direct measurement of the training's effectiveness, at the group as well as individual level. While turnover in the lab-user population may present some challenges to this approach, it may be possible to administer the pre-training survey early in the school year and follow up with the post-training survey the following spring. Alternatively, the schools may choose to target graduate students because this subgroup is frequently in charge of day-to-day operations in the laboratory and generally remains in the lab-user population for longer periods of time (e.g., 2 to 6 years).

* If a panel design is too complex, schools could at least restrict the post-training surveys to lab personnel who have received the training18. Currently, schools survey the broader population of all lab users, including those who have and have not received training19. This will require that schools compile contact information for trainees; based on discussions with the university EHS representatives, schools have already begun collecting this type of information.

All of the recommendations discussed above can be summarized as a "protocol" for future survey rounds. Specifically, this evaluation suggests that schools adhere to the following practices:

* Administer the survey in-person rather than through the mail. * Ensure that a minimum number of surveys (e.g., 100) are completed to allow meaningful and statistically significant data analysis. * Ensure that respondents represent a cross-section of the target population. If general awareness is considered, respondents should include a proportional mix of lab users. If training effectiveness is considered, respondents should include trainees only. * Retain and store hardcopies of all completed surveys. * Enter all survey data into an electronic database format. * Analyze all data thoroughly to address key questions.

9.5 EPI #7 Goal: Increase the percentage of students and laboratory workers receiving training

Students in teaching laboratories and laboratory workers receiving training (EPI #7). The goal of the training EPI is to increase the number or percentage of students and lab workers receiving training. There was no baseline assessment for this EPI, however with the EMP implementation came a more institutionalized training system. Each EMP details the training methodologies employed. Training laboratory workers in laboratory safety, environmental management, and regulatory compliance issues is of foremost importance in creating and sustaining a laboratory management system under the EMP. In the college and university setting, tracking laboratory workers and then administering training is extremely difficult as laboratory workers, staff, researchers and students are extremely transient. Therefore over time, as the schools continue to build effective training infrastructure, the number of laboratory workers trained each year may begin to stabilize or decline, depending on whether refresher training is required by the institution or the department. Given this potential trend, it may be best that other indicators, such as EMP compliance results and the environmental awareness survey serve as good measures of progress in addition to gathering data on the numbers trained.

Boston College: Training is managed in the different departments with various degrees of systematization. The EHS office coordinates and/or provides training and maintains a central record of who has been trained. Every laboratory has an EHS Contact Person who has received training and is asked to give new laboratory workers information and specific on-site training prior to attending formal training. Due to the changing population in the laboratory, the department administrators manage the training lists. The EHS office provides a list of people trained and the administrator must crosscheck the lists of those working in the labs and those trained.

Some department administrators have established the following training policies for their laboratories:

* Geology and Geophysics-no individual is allowed in the chemical laboratories unless his/her name is posted on a list that states that they have completed the training requirement;
* Chemistry-mandatory training during orientation for new graduate students prior to the start of school in August. New post-docs and staff may also be trained at that time, or will attend training scheduled by EHS