Prudent Planning of Experiments
The cornerstone of a sound program for prudent laboratory practices is a process designed to comprehensively review the operations and potential hazards associated with each experiment1 over its life cycle. This review should take place before any work is conducted. The diverse nature of research and development activities makes it advisable to have such a process in place as part of the scientific method of experimentation. In laboratories where this preliminary survey is routinely practiced, it has proved to be useful in both the maintenance of safe laboratory operations and the minimization of chemical exposure and waste generation. Because of the diversity of types of researchers and laboratory work, such processes—both formal and informal—can help individuals associated with new, modified, or unfamiliar experiments or procedures to plan and work safely, responsibly, and productively. By first evaluating the work area, materials, equipment, and procedures in depth, hidden hazards may be identified and addressed. The pre-experiment review process can also help to ensure that every experiment and laboratory operation complies with all applicable laws, regulations, and other policies. Moreover, by addressing all relevant health, safety, and environmental issues when an experiment is first conceived, further research, scale-up, or development based on it can be made safer and more effective.
In this chapter the concept of experiment planning is addressed as the first step in ensuring that prudence is exercised in conducting laboratory operations. Because of the diversity in experiments, laboratory workers, facilities, and hazards, experiment planning can be a complex process for which it is impractical to structure rules. The committee's approach has been to consider the likely steps involved in conducting any chemical experiment. These range from developing a clear understanding of the goals and objectives of the project to providing for the acquisition and handling of materials and equipment all the way to the storage and ultimate disposal of all chemicals, both desired products and waste.
The steps to be considered in planning an experiment can be described by a flowchart such as the one depicted in Figure 1.2. (section 1.G), which shows the individual steps in a laboratory experiment. These steps, in turn, correspond to the different chapters of this book. For the experienced laboratory worker, these steps are understood intuitively, and the pre-experiment review is primarily a thought process, with perhaps a brief written description of the experiment plan. For others, the review of some or all steps on such a flowchart (and the corresponding chapters in this book), along with a more formal documentation process, is in order.
2.B LEVELS OF FORMALITY IN EXPERIMENT PLANNING
As mentioned above, the types of laboratory experiments are diverse and may be conducted by a wide range of practitioners whose skills and backgrounds may be enormously varied, even within a single discipline or institution. Thus, the degree of formality and documentation necessary for prudent planning is a matter of judgment. Decisions to be made in experiment planning will be affected by the knowledge and skills of the personnel, the scale of the experiments, the particular hazards of the materials or operations being contemplated, the institution's policies for planning and conducting experiments, and the regulatory environment in which the experiments are to be performed.
Because of this diversity among practitioners, what may pose a significant and unfamiliar challenge to one laboratory worker may be second nature to another. For example, the very limited skills of high school or undergraduate chemistry students may demand that extensive written planning of basic laboratory procedures take place before an experiment is attempted. For graduate students or seasoned research chemists, those same operations may be sufficiently familiar that pre-experiment review of an informal "mental checklist" and a single line entry in a laboratory notebook may, in such a circumstance, be entirely prudent in laboratory practice. Similarly, what may be a straightforward distillation of organic solvent to a Ph.D. graduate student in synthetic organic chemistry may be an unfamiliar and, thus, rather more hazardous procedure for even an experienced professor of theoretical chemistry.
To allow for these differences in level of experience, many institutions have established general guidelines for several different levels of formality, which are scaled according to the estimated risks. Depending on the situation, these might include (1) simple mental evaluation of hazards for straightforward experiments
by experienced practitioners, (2) more formal discussion of the experiment and options with experienced peers for more hazardous operations, or (3) a formal hazard review process with complete written documentation of the procedures to be employed for new, unfamiliar, or intrinsically hazardous operations. For example, the following special areas of laboratory work will almost always require some approval steps: work with radioactive materials; experiments involving pathogens that cause serious or lethal infection; high-and low-pressure work; research involving especially hazardous materials; and experiments being scaled up.
Diversity in local, state, and federal regulatory issues and institutional policies also enters into the planning of an experiment. Simply "thinking things through" and recording a description of a procedure in a laboratory notebook might be a fully prudent practice for handling one chemical, while the identical procedure for another compound subject to specific regulations or for a large quantity of the same compound might demand a detailed written experiment plan, review by others, authorizing signatures, and accounting of material balances.
It is clear that no single, universally applicable description of "good" experiment planning exists and that the level of formality to be considered prudent in pre-experiment planning is a matter of judgment. In an area where potential hazards exist, more attention to planning is clearly better than less.
2.C INDIVIDUAL RESPONSIBILITIES FOR PLANNING EXPERIMENTS
Implementation of effective pre-experiment review programs must be initiated and backed by the highest level of leadership in an organization. Primary responsibility for day-to-day implementation of such programs should rest with individuals who supervise particular laboratory activities. While the experiments may be prepared and conducted by the laboratory workers, it remains the responsibility of the laboratory supervisor to determine what level of experiment planning is appropriate and to be accountable for necessary training, documentation, and compliance with regulations.
The laboratory workers involved with the experiment or procedure should participate actively and monitor the planning process carefully. When planning for new or unfamiliar procedures or experiments, the workers should review the literature and consult experts to assist with the review. These experts may be outside the regular chain of leadership in the organization or may even be outside the organization altogether. They could include program leaders, co-workers, and safety, health, toxicology, and industrial hygiene personnel who are associated with chemical research. Experimenters should also consult appropriate sections of this book and any other available safety, toxicology, and industrial hygiene reference materials that might aid in planning the experiment. At the completion of the pre-experiment review process, the workers should have complete familiarity with the planned activities, their associated risks, all protective measures needed, and contingency plans to deal with unexpected events or accidents. The protection of the individual worker and the public is paramount. When conducting laboratory activities, workers not only must have the knowledge necessary to ensure their own safety and that of co-workers and society, but also must be willing to accept the responsibility for that safety.
2.D INSTITUTIONAL POLICIES AND EMERGENCY RESPONSE PLANNING
Just as those proposing an experiment have responsibilities for safety in laboratory work, the institution in which the experiment is to be conducted is also responsible for certain aspects of experiment planning. This is the case for academic departments in universities and national laboratories as well as for private corporations. Because of the scope of institutional responsibilities, it is generally less effective for an institutional bureaucracy than for the experienced professionals directly involved in the work to attempt to set guidelines for specific experiments. However, the institution shares the ethical, legal, and financial burden of ensuring that experimental work is carried out safely and responsibly; thus, the institution must establish general guidelines for what constitutes prudence in laboratory work practices. The institution is responsible for setting standards and keeping records of any necessary training of laboratory workers. Moreover, in specific circumstances the institution may spell out guidelines for working with specific hazards, as in the case of an especially toxic compound or a federally regulated drug intermediate.
In addition to setting the general tone for work practices, the institution is responsible for developing and implementing laboratory policies and standards for emergency response procedures and training. This responsibility is best handled at the institutional level by a central environmental health and safety office. Activities of such an office might include developing contingency plans for handling injuries, chemical spills, explosions, fires, natural disasters, the loss of
power or water pressure, and other emergencies. Such a central office should also make all laboratory workers familiar with applicable laws as well as the institution's policies and plans so that considering emergency response becomes a natural aspect of experiment planning for all laboratory workers. Many institutions that have worked with their local response agencies (e.g., hospital emergency rooms and fire stations) have found that such planning helps all parties develop a better understanding of the requirements for effective emergency response and a clearer appreciation of the potential magnitude and likelihood (or frequency) of the services that might be needed. These policies should include the broad types of emergency response mentioned above, as well as the maintenance of, and training in the use of, fire extinguishers, first aid kits, spill cleanup kits, eyewashes and safety showers, self-contained breathing apparatus, and so forth.
It is also the responsibility of the institution to determine the level of documentation appropriate for different laboratory operations, including experiment planning, as mentioned above, as well as emergency response planning. In some instances, minimum standards have already been spelled out by regulatory agencies, and the institution must ensure compliance. In other circumstances, such as for emergency response planning, it may simply be a good idea to establish laboratory standards for documenting the location of flammable solvent storage, the best routes for laboratory and building evacuation, the decision trees specifying the contact person, and so forth. Prudence also requires that appropriate levels of documentation be provided to emergency responders. Although some regulatory environments may specify that all facilities that handle chemicals must provide local response agencies with complete lists of chemical storage by location (with accompanying Material Safety Data Sheets (MSDSs)), this type of detail is rarely appropriate for a laboratory where small quantities of thousands of different chemicals (many new and/or not yet fully characterized) are used. Even if a laboratory were able to comply with such a stipulation, the local response agency could easily be inundated with the enormous volume of paperwork and be unable to easily find the needed information in an emergency. A multipage inventory of complex names and structures of research chemicals, many of which never have been published in the literature, would be of little value to a public safety response team facing a laboratory fire. In planning for emergencies, therefore, the actual needs of the responders should be the highest priority, and the appropriate level of information should be provided to ensure effective emergency response.
Finally, some institutions, particularly those in private industry, require extensive documentation of the planning process for all experiments, while others do not. Similarly, undergraduate teaching laboratories often require detailed written experiment plans before work can commence, whereas this level of written experiment planning is much less common in graduate research programs. In the end, the extent of documentation is probably much less important than establishing a culture in which workers think through the potential hazards of experiments they plan to conduct and seek out the resources necessary to ensure that experiments are conducted safely. Documenting the experiment planning process is one tool that may help to build this proactive safety culture, but it will probably not be sufficient to ensure safe work if other drivers for prudent practices are absent.
2.E STEPS FOR PLANNING AN EXPERIMENT
One mechanism to facilitate effective planning is to consider the steps of an experiment in a flowchart. When the fundamental steps in the research process and the flow of work through each step are understood, the critical issues for laboratory work can be addressed in the sequence in which they are likely to be encountered. Once the goals and objectives of the experiment have been clearly formulated, the planning can begin. Consideration must be given, in turn, to risk assessment, acquisition and storage of chemicals, handling of chemicals and equipment, and disposal of waste. Other customized flowcharts, with more or less detail, should be considered by laboratories that employ different procedures or use only a few of these steps.
The general steps a laboratory worker must consider in planning an experiment are highlighted in the following paragraphs. The actual execution of each step is discussed in much greater detail in the following chapters.
Just as a clear understanding of research goals and objectives is an essential part of any scientific investigation, so also is a clear understanding of the goal of "safety first," and how it meshes with the research goals and objectives, an essential part of planning. The research goals and objectives should be stated clearly in order to generate unambiguous data and to facilitate consideration of such matters as source reduction and the substitution of benign alternatives to some reagents. Pollution prevention methods, in turn, can minimize exposure to hazards and the potential risk to the researchers while also minimizing the cost and waste disposal requirements associated with the experiment.
Following the philosophy described in Chapter 1, the development of a laboratory culture that empha-
sizes safe work practices and workplaces should be an important goal of the leadership of an institution. An inspection system to audit work practices is recommended to ensure that laboratory activities are conducted prudently and comply with all regulatory requirements and local policies. Owing to the diversity and number of activities conducted in a typical laboratory, inspections commonly focus on particular experiments or procedures that may contain special hazards. However, random checking of more routine operations may uncover laxity or hidden problems not envisaged when the experiment was first planned. The nature and structure of an inspection will vary with circumstances, but clearly inspections should not be conducted in circumstances where they may actually increase laboratory hazards. However, the basic concepts used by a financial auditor to define a scope, conduct an audit, document the findings, require a response to the findings, and ensure that these responses address the findings properly can apply quite readily to the audit of pre-experiment reviews and general laboratory safety inspections.
2.E.1 Chapter 3: Evaluating Hazards and Assessing Risks in the Laboratory
Complete assessment of hazards should be made for all materials and suspected products associated with the experiment or procedure. Chapter 3 provides the basis for interpreting and applying much of the available hazard information. Both literature resources and knowledgeable contacts inside the local research community and in other institutions can provide additional information. If risks are determined to be unacceptable, experiments can be redesigned to minimize the volumes of chemicals used or to employ less hazardous alternatives that might do the job equally well. Some important considerations include volumes and flow rates to be employed, amounts required, physical properties of materials to be used, potential for exposure, regulatory concerns, and emergency response for unexpected events.
The Material Safety Data Sheet (MSDS) for each hazardous chemical is one of the resources that should be incorporated into experiment planning. However, because of the inconsistent quality of information found in MSDSs, Laboratory Chemical Safety Summaries (LCSSs), which are compiled in Appendix B, should be consulted or developed for the materials involved. In any case, the experiment planner needs to be aware that the existing regulations do not necessarily represent the full complement of prudent practices for handling hazardous materials and that other input is therefore essential.
In many experiments, new materials are produced whose physical properties and toxicity are unknown. Product mixtures should, therefore be regarded with suspicion of hazard until their compositions can be determined and they can be proven safe. Some provision for protecting those involved in the analysis of these product compositions from potential hazards must also be considered in experiment planning. Moreover, not all experiments proceed in the expected manner. A critical analysis should also involve consideration of the accidents that could occur in even simple experiments.
2.E.2 Chapter 4: Management of Chemicals
The experiment plan should include provisions for acquiring and storing chemicals and equipment to be used in the procedures. Some considerations for management of materials include effective labeling; inventory maintenance and reagent tracking; source reduction and materials sharing; compound shelf life; monitoring of reactive chemicals; hazards associated with storage of incompatibles, flammables, reactive chemicals, and so on; and the regulations governing shipping and storage of chemicals.
2.E.3 Chapter 5: Working with Chemicals
While the subject of Chapter 5 is at the heart of every experiment, because of the diversity of possible laboratory procedures, it is impossible to anticipate all of the potential issues that should be specified for a "generic experiment." Instead, the worker who is planning the experiment needs to rely on judgment and consultation with the literature and fellow scientists in determining which factors require particular attention. In any case, the proposed experimental procedure should be considered in adequate detail before any laboratory operations begin. Certainly these preparations should include steps such as sample preparation, equipment assembly and commissioning, start-up and calibration of equipment, data acquisition, product isolation and characterization, and storage and disposal of materials after the work is completed. Special consideration should be given to planning for unattended operations, novel equipment that is to be purchased or fabricated, and experiments that are undergoing significant scale-up.
In experimental work, it is important to recognize that although accidents can be minimized, the nature of gaining new knowledge suggests that they can never be eliminated completely. Any good experimental design process should identify hazards and develop
contingency plans to deal with the unexpected so that people are not hurt, facilities are not damaged, and the public and the environment are protected. But unexpected situations can develop despite the best experiment planning. With good contingency plans, the worker might be able to change reaction conditions or procedures to obtain valuable information from experiments that do not proceed as expected, while, at least, preventing them from becoming harmful accidents.
2.E.4 Chapter 6: Working with Laboratory Equipment
A complete assessment should be made of the equipment proposed for the experiment to highlight any associated hazards. The location of the equipment within the work space should also be noted. The equipment hazards to be considered include those associated with reactors, tubing, relief devices, pumps, refrigerators, glassware, heat sources, electrical devices, lasers, ultrasound generators, photochemical equipment, compressed gases, and equipment for working at temperature extremes. Consideration should be given to whether proper maintenance procedures have been followed and documented for all equipment. The proper use of personal protective equipment such as aprons, face shields, gloves, safety glasses, and respirators should also be planned. Certain equipment will require the use of warning signs, lights, barriers, equipment monitors, alarms, and safety interlocks, particularly when temperature extremes, pressurized gases, or extremely hazardous substances are involved. The use of certain materials might also require industrial hygiene monitoring and/or special occupational health reviews.
Various institutions and local, state, and federal agencies may require certain considerations, documentation, or training for some laboratory operations, particularly (but not restricted to) those involving especially hazardous materials, equipment, or procedures. The laboratory worker and supervisor are responsible for understanding and ensuring compliance with such mandates, but they need to be aware that not all hazards are regulated and not all regulations are sufficient and that other safety measures may be necessary.
2.E.5 Chapter 7: Disposal of Waste
Environmental and waste disposal issues for source reduction, waste minimization, and recycling of materials must be considered in any experiment plan. The chemical composition of all products and waste materials generated by the experiment should be considered, and appropriate handling and disposal procedures for each of these materials should be evaluated in advance. Careful attention to regulatory requirements is essential for waste disposal. Special issues to consider include the frequency and amount of waste generated, methods to minimize waste, steps to neutralize waste or render it nonhazardous, procedures for dealing with unstable waste or waste that requires special storage and handling, and the compatibility of materials being accumulated. During the planning stage, particular attention should be given to the minimization of multihazardous waste, such as waste that represents both a chemical and a biological hazard.
Precautions should be taken to minimize the release of hazardous chemicals to the environment. A fume hood is a safety device and not a waste disposal facility. Therefore, fume hoods should not be used to dispose of volatile hazardous materials—to do so could cause toxic materials to be released. Special ventilation and exhaust systems, scrubbers, filters, or some other control equipment for discharges to the air or chemical sewer systems may be required under some circumstances.
2.E.6 Chapter 8: Laboratory Facilities
The facilities proposed for an experiment should be assessed completely to identify any associated hazards and to determine if the facilities are adequate for the purposes of the experiment being planned. The location of the equipment in the work space relative to the location of emergency response facilities should be considered. Work with hazardous chemicals should be carried out with fume hoods, elephant trunks, and glove boxes for some operations. The use of certain materials might also require industrial hygiene monitoring and/or special occupational health reviews. General consideration of the type of work space, its layout, and infrastructure may be appropriate. Special needs for bench space, storage, ventilation, shielding, and so forth might also affect the planning of the experiment.
2.E.7 Chapter 9: Governmental Regulation of Laboratories
Regulations are an intrinsic part of modern laboratory work that cannot be separated easily from other matters and should be considered at each step of experiment planning. It is only prudent for laboratory workers and supervisors to ensure regulatory compliance in conducting laboratory experiments. However, the responsibility of leadership goes beyond compliance to the protection of individual laboratory workers,