Laboratory water systems: Cost-effective generation and distribution Part 1: Quality requirements and demand By Norman H. Toussaint, PE, and Lauren M. Goodfellow Regardless of the research mission, every laboratory aims to provide accurate and repeatable results, executed in consistent environmental and physical surroundings. High-purity water is a universal component for research, spanning needs from sample preparation to feedwater for glassware washing. These needs are diverse even in simple laboratories, but in today’s interdisciplinary science buildings, the lab water systems must meet especially varied criteria. As research pushes the envelope of molecular manipulation and device fabrication, and as construction budgets tighten, designers are faced with a greater challenge to provide cost-effective yet high-performing lab water systems. If not carefully designed, water systems can be costly to install and operate, and may even adversely affect the research they are designed to support. This two-part article outlines the challenges and some solutions for designing a robust yet cost-effective lab water system. The first installment discusses quality requirements and demand calculation; the second installment, to be published in January, reviews specific design choices, equipment, and operational issues. Fig. 1: A breakdown of the four common water specifications and the associated categories. All figures: HDR. Click to enlarge. Industry guidelines and standards While high-purity water is critical to scientific pursuits, no consistent standard exists for its production and delivery. “Lab water” is a concept that involves numerous definitions, subsets, guidelines, and standards, ranging from accredited consensus to committee-aligned authoring and approval over a diverse user community. Depending on the intended research, scientists may have a multitude of contamination concerns, ranging from endotoxin units to trace metal content. Designing a cost-effective system requires a team to establish a baseline understanding of the definitions, categories, parameters, and recommended applications. While lab water is commonly referred to in many different ways, each name in fact entails a unique definition. The most common are as follows: n High-purity water: a generic term for the output water of a water treatment process that removes recordable amounts of ions, organics, and bacteria. • UPW or ultra-pure water: a term used specifically in manufacturing and research in the electronics and semiconductor industry. UPW is water that is both highly filtered and deionized to a specified standard. • DI or deionized water: water that has been filtered and treated to remove metallic ions and impurities, and kill microorganisms. • RO or reverse osmosis water: water that is processed via a membrane filtration technique removing more than 90% of ions, organics, and bacteria. • Reagent water: used specifically as a component of an analytical measurement process. The water meets or exceeds defined specifications. • Water for injection (WFI): water that is purified by distillation or reverse osmosis, and used in pharmaceutical production. As indicated in Fig. 1 (above), four common research guidelines and standards for laboratory water exist. These break down into 27 subcategories, all used to catalog laboratory water grades. Needless to say, the potential for confusion is great. Fig. 2: Definitions of common monitoring points. Click to enlarge. Despite the multiplicity of “approved and governing” water quality specifications, there is some lack of consistency in the practical measurement or processing components involved in meeting these standards. (Additional standards can be found that analyze and critique the above-noted standards, comparing and contrasting the measurement criteria as well as the approval process; for the purposes of this discussion, we’ve chosen to omit these standards.) In short, when planning a water system, definition of terms will necessarily be a key initial task for the entire design team. It is critical to understand not only the industry definitions of different types of water but also the definitions of measured contaminants. Fig. 2 (above) lists the definitions of common measurement parameters shared among various water quality standards. Understanding the impact of each recordable quantity is key to devising the most appropriate qualitative and cost-effective method of delivery. Fig. 3 (below, left) outlines the respective monitoring points associated with each standard. Fig. 3: Monitoring points for each specification. Click to enlarge. While each standard entails a unique list of parameters, there is a significant amount of overlap between the measured criteria and the measurement levels. A common measurement among standards with an inconsistent baseline is biological contamination (Fig. 4, below). Measurement of biological contamination is integral to each standard but is quantified using varied definitions and limits. Lab procedures such as PCR (polymerase chain reaction), which is used to build a large sample of DNA from a very small sample, require the use of small quantities of water that is not contaminated with protease or nuclease material. In this example, the CLSI C3-A4 Recommended Standard would not be acceptable since it does not appear to measure protease contaminants. Many other procedures such as PCR cleanup, BAC cleanup, gel extraction, nucleic acid blotting, and a host of proteomics research tasks, such as antibody sample preparation, have unique requirements that are best met by providing Type I or II lab water with a local filtration or purification system specific to its process. Successful planning tactics The most basic planning issue that affects design of lab water systems is identification of the laboratory function. This can range from teaching labs (chemistry, biology) and inter-disciplinary research labs to cleanroom labs to incubator/ pilot plant labs used to develop and scale-up manufacturing processes. Each function has different implications for water quality and consumption. It is also useful to determine the future directions the laboratory may take (for example, is the nature of the research likely to result in frequent equipment/process upgrades or changes?). Fig. 4: A graphical representation of differing bacteria measurements and limits between standards. Click to enlarge. A thorough programming phase will define user needs and aid in the detailed design of a “right-sized” system. This is accomplished through a well-thought-out approach, concentrating on the required water quality, system size, and scope of distribution. The variety of water quality standards applicable to laboratory systems was discussed previously. Unless the design team and user groups have a deep understanding of these specifications, they should communicate in terms of contamination types and levels rather than a given ASTM or CLSI specification (see Fig. 5, below, for a graphic presentation of common procedures and related specs). This approach helps achieve data-driven decisions. It allows for the contamination concern to be addressed in a different specification that is applicable to the whole user population, leading to a cost-effective, consolidated system rather than a patchwork. Many research groups are trending toward collaborative disciplines like nanobiotechnology, a blend of physical and life sciences resulting in not only overlapping but also unique requirements not necessarily addressed by any single existing quality standard. Fig. 5: Recommended listing of common procedures as they align to the respective specification. Click to enlarge. Calculating demand Once the quality requirements have been established, the respective system demand must be calculated. Design decisions depend on correctly estimating demand, including sizing equipment (e.g., makeup or polishing systems) and distribution piping. Underestimating demand has obvious consequences. Overestimating demand can also have negative impacts on system performance and cost. Diversification and utilization are two calculation-based approaches used to determine required capacity. Diversification factors and utilization rates are based on peak loading, equipment up-time, and duty factor of service in use. As represented in Fig. 6 (below), a thoughtful diversification and utilization scheme can result in a 23 to 43 cost saving. These metrics must be evaluated with the users for each specific project, but on average R&D facilities lean toward a 20 to 30% utilization rate, with industrial facilities performing at a 65 to 75% rate. Diversification is best applied to standard laboratory sink usage, while utilization is best applied to unique laboratory equipment, or requirements exceeding a peak flow rate of ~1 gal/min (gpm). With the incorporation of diversification factors, clean processing lab flow rates on average run 1 to 2 gpm per 1,000 ft2 (gross). Non-classified wet labs are directly correlated to quantity of sinks, and typical usage is ~0.3 gpm per faucet. These metrics include a safety capacity of 20% that affords future expansion or equipment installation. Fig. 6: Cost impact of a “right sized” system. Click to enlarge. The need for redundancy is another decision that the design team must make, ideally during the early planning stages since the outcome affects area requirements as well as budget. For production facilities, there is usually a simple case to be made for redundant distribution loop feed pumps, filters, and even redundant treatment modules (for example, demineralization units), since any down time for planned or unplanned maintenance can be costly. For laboratories, where the demand may not be continuous, the case for redundancy is less clear. A decision on the need for (or degree of) system redundancy should include a review of criticality of supply, impact on research that would result from loss of service, and alternative sources of an acceptable water supply during planned outages. After determining required system capacity, the next step is deciding the approach to generation and distribution. These aspects, as well as equipment choices and operational issues, will be discussed in Part 2 of this article, which will appear in January. Norman H. Toussaint, PE, LEED AP, is a senior chemical engineer in the London, U.K., office of HDR Architecture Inc. Lauren M. Goodfellow is a laboratory planner in the firm’s Mountain View, Calif., office. HDR offers architectural, engineering, consulting, and management services for various sectors, with science and technology as a core market group.