Inadequate Laboratory Ventilation: An All Too Common Reason for Major Safety and Operational Problems Down the Road

Laboratory exhaust ventilation is expensive. Estimates range from $20-70/CFM to install and $3-$12/CFM to operate depending on local costs and system design. Hoods are an added expense ranging from $200-$400/linear foot to purchase and install. Hood controls add another $1,000-$3,000 per hood. This means it is not uncommon for ventilation systems to account for 15-40% of the total cost of a new laboratory. Hence it is not surprising that designers, contractors, and management all want to minimize the amount of exhaust provided to lower the cost of a new or modified laboratory.

Realistically, cost considerations always must enter into the design of any new facility. On a personal level, we would all like that extra bedroom, bathroom, or garage or that upgraded kitchen or finished basement when we are looking for a new home but usually must settle for something less to remain on budget (or even solvent). So, we should not be surprised that most new laboratories end up having less space, less hoods, and less exhaust than the researchers would like in an ideal, non-cost constrained world[1].

Unfortunately, the long term impacts of reducing the amount of exhaust ventilation provided are often unrecognized and frequently result in safety and operating issues later. These can range from employee and visitor exposures, fires, or explosions, to significant operating restrictions leading to reduced efficiency and effectiveness and even – in some cases – an inability to do some desired work safely. I suspect that most families of 4 or more would recognize that buying a new house with only one bathroom is probably not going to be their best decision and lead to many problems later; most will try awfully hard to get at least two bathrooms if not more. They might decide to look further out from their desired location, compromise on other areas of lower impact, or perhaps accept a smaller size home. They recognize that the reduced number of bathrooms is likely to be a key factor in how usable their home will be long term. Too often, however, the impact of a lower laboratory exhaust capacity goes unrecognized until it is too late, is too casually evaluated during the planning and budgeting process or is accepted based on a grossly overly optimistic belief of how minor the long term impact will be. And, invariably, the long term impact is extremely high.

A colleague of mine always reminds everyone that “laboratory safety is spelled V-E-N-T-I-L-A-T-I-O-N". Without enough laboratory exhaust ventilation numerous problems will always emerge.

·        The reduction in the number of hoods supplied results in hoods become overcrowded and failing to work properly. Hoods are designed to work effectively empty and when filled struggle to work as effectively as we think they will. This can result in exposures to the operating personnel when the hood does not capture all releases properly or even fires or explosions inside the hood because of inadequate or slow dilution to safe levels[2].

·        The lack of enough hoods leads to attempts to deal with the problem by using less effective approaches such as ventilated enclosures, local exhaust, canopies, and snorkels all of which rarely work as well or as effectively (if at all) long term[3]. Everyone agrees the work is best and most safely don in a hood but there simply are not enough hoods to allow that. So some work moves out to other, less effective, less safe alternatives. If you don’t have enough ladders, invariably someone will stand on a chair or a bucket.

·        Operations tend to migrate work to open benches due to the lack of adequate hood space. Smaller procedures (viewed as less risky), intermittent and/or infrequent operations (viewed as less frequent) , fast work (viewed as short duration) , and tasks viewed – usually incorrectly – as less hazardous just take place in the open just relying on general laboratory exhaust. Invariably, this leads to exposures, fires, or explosions as there is no capture of the hazardous gases or vapors at the source as required by NFPA-45 Fire Protection for Laboratories Using Chemicals. Argue that it will not happen. Argue that a hazard analysis and risk assessment will stop it. Argue that company policy prohibits this. Without enough hoods all these arguments will fail and it will inevitably occur with some frequency.

·        Safety oversight becomes much more difficult and challenging as the lack of the primary mitigation, exhaust ventilation, encourages unsafe to at least less safe alternatives, sub-rosa operations, and general frustration. “I can’t do this in a hood as one is not available right now!” Inspections routinely identify problems and bad practices. This leads to them being more frequent and requiring more resources. Practices and procedures become more involved and complex to try and address the issues uncovered eating into productivity. And, despite the best efforts, safety usually remains adversely affected since the researchers have no really viable choices.

·        The decision is made to limit how hood openings either as part of the design to lower the initial capital costs or later when new requirements or actual operations identify concerns that cannot be mitigated by any other means than stealing existing ventilation. Calling it “repurposing” to make this shift in exhaust should more professional and less innocuous does not change the basic principle in the least. Yet limiting a hood opening to 28”, 24”, or even less results in significant operating difficulties, unergonomic activities, and incredibly difficult set up. This results in most hoods invariably being open too far and leads to the potential for exposure or releases[4].

·        Operational efficiency decreases as personnel must wait their turn to use hoods, pour solvents, empty waste, or dozens of other typical practices. Perceptions that the wasted time is minimal or could be spent profitably doing other things is usually wrong, reflecting wishful thinking rather than any real analysis. How often will you go back to your office to do work rather than waiting 5-10 minutes for someone to finish using the copier or for the printer to finish a longer report?

 Certainly, a valid counterpoint to all the above is that organizations have limits on how much they can afford which has to set some caps on how much exhaust a new laboratory can have provided. What can one do to try to reconcile the budget and the needs and yet improve long term safety and operability by providing enough exhaust? Sadly, I do not have a magic bullet or a one size fits all solution. I can suggest some ideas that should be considered and can often help to close the gap between the ventilation desired and the organization’s budget.

1.      Avoid bad initial estimates of what exhaust will be required. If you get a bad estimate of the exhaust to start, the cost estimate will be low and you will be starting on a (losing) uphill battle when the actual exhaust requirements and their associated costs are determined. Getting funding for perceived increases due solely to terrible preliminary estimates is always harder than getting them for even gulp producing real estimates. Review your estimated exhaust ventilation needs realistically with the actual users and company safety personnel well before any cost estimates are made and try your best to ensure they are accurate. Adding 10-20% at this stage as contingency for predictable but currently unrecognized needs is prudent. Challenge assumptions. (“Yes we agree the work in that area is shrinking but they have been very constrained by lack of enough hoods and assuming all of them will be surplus does nt seem reasonable.”) Relook at current practices and procedures and try to validate any estimates or assumptions, particularly those that seem to promise a much lower exhaust capacity. (“Yes we agree training everyone to always close their hood sashes when not in use may help lower our average exhaust but we see no real proof to validate the 20% savings you claim.”) Once you have the required capacity don’t let it be reduced unless a reduced basis is approved. (“If we need to reduce the exhaust costs by 20% then we will need to look at a 20% smaller laboratory facility and decide what operations or research we will plan to eliminate.”) Do not reduce your ventilation capacity by a fixed amount to reduce costs assuming you will somehow find a way to make it work. You will not. Do not assume a much lower diversity (the number of hoods that could be operating all at once) without very careful validation. Your maximum simultaneous usage is normally more, and not less, than what many design organizations recommend or predict. (My classic story is one recommendation based on “actual observations”. Upon checking further, several labs surveyed had personnel away traveling, and in others the laboratory personnel felt so intimidated by the hovering survey taker that they admitted to avoiding doing anything they could during the survey. Hence the much lower diversity.)

2.      Work with your design firm to investigate all viable alternatives such as horizontal versus vertical sashes (this can reduce the exhaust by 30-70%), providing fixed ventilation to some equipment (which removes the need for a hood and often reduces its requirements by 80-90%), piping some instruments to vent (saving 100-200 CFM/instrument), and revisiting practices and procedures to see if their required ventilation could be safely reduced through modifications. (“All transfer operations must take place in a hood” is a good policy but probably is not necessary for a water treatment tests which involves -at most- ppb levels of hazardous materials. Modifying procedures to use closed transfers may allow then to take place on a bench.) Many of these approaches are not considered because no one takes the time and effort to recognize that the sum of all these potential small savings might be a way to safely reduce the total exhaust requirements and alleviate some of the problem. All the problem? Rarely, but some of the problem almost certainly. There will be tradeoffs. Horizontal sashes will require more frequent positioning than vertical sashes (usually, in my experience unnoticed after a month), instruments piped to vent may have to be disconnected to open (trivial if you think to make the last few inches plastic tubing), ovens may have to be segregated based on the exhaust provided, and similar issues. All are generally much less onerous than being short of ventilation. Almost all will fade into the background after personnel grow used to them.

3.      Recognize that it is difficult to identify every exhaust need exactly and, worse, impossible to accurately predict the future. Demand some surplus exhaust for those needs that are currently unknown but sure to arise. My studies have shown that (1) most ventilation surveys miss 5-10% of actual current needs despite the most diligent efforts (the reasons range from the very subtle to the hilarious were the consequences not so severe) and (2) most new labs require 5-20% more ventilation with 2 years of construction (new operations almost invariably require new ventilation). Relocating to a new facility always involves some level of re-review of current operations. And these reviews inevitably highlight some things that slipped under the radar and require more ventilation before they are permitted to continue. Small organization or large, strong safety culture or weak, industry or academia, they will arise. If you do not plan on having at least 10% extra exhaust included in the base design you will have problems from the second day of operation. And the problem will only get worse in time. It is possible with care (and added design work) to design a laboratory exhaust system that can be economically expanded later. It is difficult and often fails to work as envisioned because the design firm has not thought through and addressed all the issues associated with the addition (shutdowns, access, impact on power and controls, etc.). It also will invariably be more expensive than doing it during the initial construction. Worse, how many organizations will approve $50,000 to provide more exhaust when all they need is one more hood? Instead they will try and get by with what they have no matter how many problems it produces.

4.      Ask that someone present a realistic detailed analysis of the estimated operating versus construction costs. In many cases a more energy efficient variable exhaust system with some viable energy recovery may pay back in less than 2 years. In other cases, the operating costs (or conversely savings) will strongly encourage organizations to swallow hard and authorize the higher one time construction costs. (Another wise colleague always noted that “spending capital is painful but operating costs are the gift that keeps on taking forever”.) While the argument to spend more initially may be difficult and undesirable, it is often winnable with enough effort if the annual savings are identified and are credible.

And when these more obvious tactics fail to close the budget gap what should you do? The only answer is – usually – to find the required funding required to support the required exhaust capacity. While securing this extra funding may prove possible in some cases, in more I think it will prove impossible. Hence the only viable method to secure the required exhaust capacity is eliminating or reducing an equivalent other costs to balance the equation in favor of the required exhaust. Here are some suggestions for that.

5.      Revisit decisions about other less critical, although usually much more desirable, requirements that are adding costs to the project. Smaller or less well furnished offices, fewer or more spartan meeting rooms, bland versus cutting edge exteriors may not be what everyone wants but can often yield significant savings that can be applied to offset the costs for the required ventilation. The researcher waiting an hour for their turn at a hood or unable to do their work until a way can be found to give up in use hood space will rapidly learn to rue their nicer office or fancier conference room. Recognize that dissenting opinions will often arise from those with no or limited need for actual laboratory operations. The modeler who works at their PC, the supervisor who only tours a laboratory, the manager long removed from the actual work will probably view these other non-laboratory areas as more important. This requires enough courage to question how important their opinions are to the primary needs the laboratory is intended to serve. Alternately, it may allow a discussion and support for the additional funding to have both (although I would not count on that outcome too often.)

6.      Casework, the benches, hoods, and similar laboratory furnishings, are a major cost component of a laboratory. Manufacturers have different models and grades ranging from the spartan to the opulent. Manufacturers tend to start their suggestions from the high end (where they make the most profit). Designers tend to recommend the same high end (as they look great and lets them spend less time and effort on the specification). Management – and often actual researchers – are impressed by the glossy pictures, high end 3D models and extensive list of features. All of these come with a price. And how often do all, or even most, of your benches need to be raised or lowered, rolled around, or act as showcases to visitors? Consider lower quality furnishings (fixed versus mobile casework, fixed versus adjustable desks and benches, lower end office and conference room furniture). Consider if every bench needs to be installed now versus later if – and when – the need arises. Reduce the number of stools and add more if it becomes a problem. All of these and many similar small deletions can free up a lot for exhaust. The costs saved can be used to offset the higher exhaust requirements.

7.      Talk to the design firm about building construction options. “Cheapening the building”, an acerbic term architects like to drop when this subject is raised, is often anathema to them and to management. The desire to be nominated as a laboratory of the year, the dramatic post construction photo, the wow effect upon visitors is always present even as an undercurrent. Everyone would like a project they could be proud of but, again, keep the primary purpose of the entire project, a safe, efficient, and well-functioning laboratory, in mind and push back hard. Consider lower end construction (concrete block versus brick exteriors, low cost VFT or even bare concrete floors versus ceramic tile). Consider lower cost and more mundane signage, both interior and exterior, limit wall treatments and special finishes, avoid architectural or high end fixtures and components for more traditional (but less dramatic) off the shelf components.

8.      Consider eliminating some things often assumed but not necessary for a safe laboratory operation. These are often things that are difficult to accept but which can be made to work satisfactorily. Examples include cutting out suspended ceilings (paint everything up high a solid color and no one will notice after a week), avoiding costly diffused lighting with individual adjustable levels (use standard lighting and add a dimmer only if proven necessary later), avoiding modern handsfree fixtures (most personnel will be wearing gloves), the always desired wall cabinets for even more storage (accepting some things will need to be disposed of and supplies limited to a reasonable minimum), and similar areas. Does every door need a lock? Does every bench drawer or cabinet need a key? (How many are locked today if you do an unannounced off hours tour.) Does every chair need to be ergonomic even those whose intent is to allow someone to sit for 10 minutes at a desk to make some fast notes?

9.      Consider, sadly, eliminating – or at least deferring to a later date when more funding is available – these areas not a critical to a safe operation. Fitness centers, onsite day care, libraries, high tech auditoriums and meeting rooms, board rooms, well presented lobbies, outdoor eating or resting areas/patios, fancier cafeterias, extensive landscaping, and numerous similar areas are all very desirable but add up to a lot of extra costs. If it is decided they are needed to hire and retain the desired personnel (an unsubstantiated but prevalent claim of some HR personnel) than suggest that these costs be funded outside the project or ask them to justify additional funding provided to cover this need. If it is suggested that the feature will pay for itself in improved performance (or the ever nebulous “employee satisfaction”) suggest that that should be enough justification for he added funding.

10.  Revisit and validate the need for the extra sink, bench, storage shelf, solvent cabinet, and refrigerator, stool, the additional utility taps, the weighting table versus a bench, and similar items. When asked what they need, researchers invariably overstate their requirements. Sometimes it is an honest belief they will be doing more and so need more facilities. (“If we move into that area, we will need another sink and bench.”) Other times it is due to a failure to really analyze their own usage. (“I sense we are sometimes backed up at the sink so we probably need another.”) Less obviously most will overstate their needs assuming that after the inevitable cost cutting they will end up with their actual requirements. (“Better ask for three so we get the two we need.”) Many times, these things will be needed and cannot be cut without operational impacts but some will always be potential candidates with little or no impact. Many of these can be added later if required. Big savings? Usually no but every little bit can help.

11.  Consider if a different procurement strategy might allow you to eliminate or significantly reduce the size and cos, of solvent storage rooms, waste holding rooms, general storage rooms, and similar support spaces. Contracting with a vendor to provide glassware and small laboratory items the next business day, provide chemicals the next day upon demand, empty waste weekly versus monthly, and similar approaches can dramatically reduce these inventory needs and their associated space requirement. Less space translates into less costs making more available for exhaust.

12.  Be open to using prefabricated and less expensive solutions for waste, solvent and sample storage. Prefabricated buildings and enclosures are not cheap but are usually less expensive than fixed construction. Proper placement and some attention to access can minimize (but never completely eliminate) issues with having to go outside. More exhaust can go far to accept the occasional trip in the rain or cold. And a canopy and proper snow removal can reduce the potential hazards to very acceptable levels.

13.  Be careful to comprehensively question the space required for storage including consumables, spare parts, replacement glassware, gas cylinders, chemicals, waste, and samples. Everyone always wants more than they have now. Don’t forget to include files and manuals many of which could be digitized or stored in lower cost alternatives or, better yet, thrown out. I am sorry to say but over 45 years of experience have repeatedly shown me that 50% of all storage is garbage and is rarely missed if discarded. An extra hood is worth a lot more than the one time you must buy a new bath since you tossed the 20 year old back up away. Storage spaces are often the biggest empty or underutilized areas I encounter when I inspect existing laboratories. There is a strong tendency to deal with the pack rat, the unregulated sample storage area, the solvent room loaded with old and used chemicals by simply specifying larger spaces in the new facility. Few end up being needed after the old spaces are cleaned up (and out) for the relocation. Consider smaller spaces with more oversight. (Another classic story is a project to provide more “critically needed” on site storage of spare parts. I convinced the manager to institute a policy of having anything stored for to be approved by a supervisor at 3 month intervals. Within a year the current space was half empty. I have numerous other similar stories.) Again, less space translates into less costs making more available for exhaust.

14.  Consider a “cold eyes” review of the design at various stages. Cold eyes are person(s) who have no vested interest in the project and can give a more unbiased opinion. Unbiased in this case means not only free of any organizational pressures but also free of the blinders owners often place on themselves. “I understand that you would like extra space behind the benches for access but you realize this adds 5% to the cost of the lab, is prone to becoming a cubbyhole for storing junk, and is only used a few times a year. Are you sure it is worth losing 10 hoods for that?” They are also more likely to challenge design firms. “Assuming you can cut hood use by 50% based on this survey of one lab seems to be very overly optimistic. I believe further surveys are required.” These reviews also frequently highlight design problems that the owners do not have the time nor expertise to identify. “People have repeatedly mentioned continued odors in your existing labs but I don’t see any measures to address this in your design.” Cold eyes can also help owners decide which of several competing options are best for their particular needs. The wider the laboratory experience of the reviewer (they can offer ideas that worked in other fields), the further away from the organization (so the less they share the same preconceived ideas), and the more knowledgeable in the field usually the more valuable their input.

15.  There is no such thing as a free lunch. Low flow hoods use less exhaust resulting in lower capital and operating costs but they do not work well when filled with equipment. So choosing to get 3 low flow hoods instead of 2 regular flow hoods may be the wrong choice if each of the three low flow hoods can only be filled to 50% of what you planned. Ductless hoods promise to eliminate the need for exhaust. However, they must be designed for a limited and fixed set of chemicals, require significant monitoring to ensure their capacity is not exceeded, require detailed procedural controls with rigorous enforcement, and are a major job to safely change out the filters. And no reputable designer will install them without adequate general laboratory exhaust which often claws back much of the purported cost savings. Snorkels and ventilated enclosures promise reduced exhaust needs but require careful design and often expensive components to ensure proper operation. A cheap, hanging hose snorkel requires a way to hold it in place often resulting in the need for a second person or improper placement leading to a failure to capture effectively. So a flexible arm is often required which is not cheap. A canopy hood rarely captures anything but the lightest vapors and even them not very well. Hence an enclosure often has to be provided which often fails to work well without a plenum[5]. While use of these and similar lower exhaust equipment is sometimes useful, my experience has continually shown that assuming it will address all your problems without a careful consideration for the potential issues and limitations leads to major safety and operating problems later when correcting them is usually totally uneconomic.

16.  Be wary of designer claims that a new control system can “fine tune” and lower the exhaust, that a new device can lead to “significant savings”, that a new energy recovery scheme can “address the budget gap”, and similar claims. Make sure their claims of code compliance are confirmed and checked particularly where the results of a miss interpretation could have a major impact. A good designer can always find ways to save you money while meeting your needs but there are limits on what they can do. This who promise excessive results usually cannot deliver and often are clueless about the effects and limitations of their grandiose proposals. (The number I have tested or analyzed only to find them useless to inadequate is legion.)

17.  Make sure you have identified all the operations you routinely do and ensure the design allows you to do them safely and effectively. The intermittent catalyst production operation that generates significant combustible or hazardous dust, the occasional need to evaporate out large amounts of condensable liquids, the need to bring in drums of solvent for a longer term or larger test, the desire to install a pilot plant for operating efficiency, and many other areas can produce requirements poorly understood yet often far reaching and difficult to impossible to accommodate later[6]. Conversely, that once a year operation may have to be recognized to be difficult and very inefficient to avoid a large expenditure for a rare event. It is, admittedly, a fine balancing act. The key to success is understanding the tradeoffs.

18.  Make sure to evaluate your long term needs. This is probably the most difficult and the most poorly performed part of the design process. Fail to identify a long term need and you may not have the design to accommodate it safely or efficiently. Identify too many potential needs and your facility costs will grow to totally unsupportable levels usually resulting in cuts that adversely impact the ability of the laboratory to do its current routine operations. My advice on this subject is simple in principle but difficult in application. Consider only defined short term (<3-5 year) needs. Past that your crystal ball is worthless. Looking at past problems is useful but be highly skeptical of the potential for them to repeat. If the same issue arises every 1-2 years it probably needs to be addressed. If it was a disaster 10 years ago then perhaps one can assume you have learned enough to avoid thinking dirigibles are likely to return and ignore the issue. Remember that the last problem, i.e. the one that happened yesterday, last week, or last month, assumes a disproportionate level of importance and adjust your evaluation accordingly. And be wary of the entertaining doomsayers who can at the drop of a hat resurrect horrible issues that have arisen in the far past. They can easily lead designers down expensive rabbit holes for no good reason.

Many of you may argue that a lot of these suggestions do not relate directly to laboratory exhaust and I will agree. However, almost all laboratories cost more than organizations expect or budgeted[7]. These ideas can generate cost savings to be used to provide the extra budget for the exhaust you need. My imaginary family may not like having to get into a cold car in the winter to drive to work since they gave up a garage but will be glad they avoided not having to deal constantly with one less bathroom. It is all about priorities and, for a laboratory, exhaust ventilation must be the highest and most overriding priority.

Will these approaches always work and resolve your mismatch? No but they will always help make it more manageable. The key to a good laboratory is understanding your needs (as best you can), making sure you get a good design that meets those needs, carefully reviewing all aspects of the design to make sure you have not overlooked the potential issues and downsides, getting a realistic cost estimate and them building your wonderful new laboratory[8].

For more information on this and other areas of laboratory design you may want to consider the University of Wisconsin course on Successful Laboratory Design: Grass Roots, Renovations, and Relocations (see https://epd.wisc.edu/courses/successful-laboratory-design-grass-roots-renovations-and-relocations/ for more details although the dates for this fall will not be available until later this spring) or Laboratory Hood Operation, Selection, and Use for Safe Installations (see https://epd.wisc.edu/courses/laboratory-hood-operation-selection-and-use-for-safe-installations/ for more information).

[1] See The Practical Effects of Budget Constraints at https://www.dhirubhai.net/pulse/practical-effects-budget-constraints-richard-palluzi/ for a further discussion.

[2] See Ventilation Dilution: A Safe Way to Avoid A Fire or Explosion or a Placebo? at https://www.dhirubhai.net/pulse/ventilation-dilution-safe-way-avoid-fire-explosion-placebo-palluzi/,“Why Can’t We Put It In the Hood?” at https://www.dhirubhai.net/pulse/why-cant-we-put-hood-richard-palluzi/ and “Why Don’t We Just Put It in the Hood?”: Issues with Degrading Hood Effectiveness Due to Equipment Placement in ACS J. Chemical Health & Safety, Jan, 2020,  for a discussion of this problem.

[3] See Ventilated Enclosures for Lowering Area Electrical Classification at https://www.dhirubhai.net/pulse/ventilated-enclosures-lowering-area-electrical-richard-palluzi/, Ventilated Enclosures: Why Do They Often Fail to Work Properly at https://www.dhirubhai.net/pulse/ventilated-enclosures-why-do-often-fail-work-properly-richard-palluzi/, and The Ten Most Common Laboratory Ventilation Mistakes at https://www.dhirubhai.net/pulse/ten-most-common-laboratory-ventilation-mistakes-richard-palluzi/ for more discussion.

[4] See Limiting Hood Openings: A Bad Idea at https://www.dhirubhai.net/pulse/limiting-hood-openings-bad-idea-richard-palluzi/ for a discussion.

[5] . See Ventilated Enclosures: Why Do They Often Fail to Work Properly at https://www.dhirubhai.net/pulse/ventilated-enclosures-why-do-often-fail-work-properly-richard-palluzi/ for more information on this problem.

[6] See “We Have Ruined the Hood”: The Potential Hazards of Vaporizing Liquids In Hoods at https://www.dhirubhai.net/pulse/we-have-ruined-hood-potential-hazards-vaporizing-liquids-palluzi/, Laboratory Area Electrical Classification at https://www.dhirubhai.net/pulse/laboratory-area-electrical-classification-richard-palluzi/, “But It Is Only A Laboratory”: The Often Overlooked Hazards Inherent in Laboratory Operations at https://www.dhirubhai.net/pulse/only-laboratory-often-overlooked-hazards-inherent-richard-palluzi/, Can I Put a Pilot Plant in a Laboratory? at https://www.dhirubhai.net/pulse/can-i-put-pilot-plant-laboratory-richard-palluzi/, Can you apply NFPA-45 to a Pilot Plant? at https://www.dhirubhai.net/pulse/can-you-apply-nfpa-45-pilot-plant-richard-palluzi/ and The Ten Most Common Laboratory Safety Issues at https://www.dhirubhai.net/pulse/ten-most-common-laboratory-safety-issues-richard-palluzi/  for further discussion.

[7] See “But That Means We Don’t Have Enough Money for Our New Laboratory!” The Dangers of Budget Estimates for New Laboratories at https://www.dhirubhai.net/pulse/means-we-dont-have-enough-money-our-new-laboratory-dangers-palluzi/ for a discussion of this issue.

[8] For a discussion of some of the issues with ensuring these last steps see "But What Will It Cost?" The Trials and Tribulations of Pilot Plant Cost Estimating at https://www.dhirubhai.net/pulse/what-cost-trials-tribulations-pilot-plant-estimating-richard-palluzi/, and Pilot Plant and Laboratory Unit Cost Estimates: Should We Progress the Idea or Not? at https://www.dhirubhai.net/pulse/pilot-plant-laboratory-unit-cost-estimates-should-we-progress/ .