Selected Highlights of the Labs21 2009 Annual Conference

Jason Rifkin, Equilibrium and Patrick Goetz, Southland Industries, Inc.

HVAC Efficiency Within a Controlled Current Good Manufacturing Practice (cGMP) Environment

In the summer of 2007, a strategic decision was made to decommission an existing GMP Cell Banking facility and replace it with a new GMP Cell Banking facility. Historically, GMP facilities are exponentially more expensive to design and build than standard research and development (R&D) laboratories. Similarly, they are more expensive to operate.

GMP facilities are designed and built under the regulation of the Food and Drug Administration (FDA), and are codified within the Code of Federal Register, Section 21, Parts 210 and 211. As a result, the federal government decides how these facilities are designed, built, operated, and maintained. The reason for this regulation is simply an effort to manufacture and distribute safe and effective drugs and biologics.

A key design element of the new facility was to improve on the existing inefficient and unreliable HVAC system, while creating a more environmentally stabile and robust facility containing mechanical redundancy. In this effort, the existing design parameter of a 100 percent outside air unit was replaced with several, smaller and redundant fan coil units all served by a dedicated outdoor air system. In addition to HVAC redundancy, air segregation for multiple products, and greater temperature and humidity controls were all key design parameters for this project.

The new facility encompasses four class 10,000 (ISO 7) process rooms, with a class 10,000 (ISO 7) "clean" corridor on the entrance side of the process rooms and a class 10,000 (ISO 7) "dirty" corridor on the egress side of the process rooms. Access and egress for the clean and dirty corridors is via a class 100,000 (ISO 8) gown-in and gown-out room, respectively. The above rooms make up the classified cGMP spaces of the project and were constructed with modular wall/ceiling technology. A floor plan of the facility is shown below.


Figure 1: Floor plan of the facility

Figure 1: Floor plan of the facility

One of the major inefficiencies in GMP facilities is achieving the classifications that are described in the above design program. The classification corresponds to the amount of particulates, greater than 0.5 micrometer within a cubic foot of air. Therefore, a classification of "Class 10,000" limits the amount of 0.5 micrometer particulates to 10,000 with a cubic foot of air. To achieve this, the FDA and industry standards recommend 40 air changes per hour, through HEPA filters. The lower the amount of particulates, the higher the air change rate that is recommended. In addition, these facilities are typically run all day and all night to maintain controlled conditions. Other HVAC inefficiencies include the fan energy required to push this high number of air changes through high efficiency filters and the cooling required to maintain below normal operating temperatures. These temperatures are critical to allow for worker comfort. Uncomfortable workers are both inefficient and shed more particulates to the clean spaces.


Figure 2: Cell banking finishes and terminal HEPA filters

Figure 2: Cell banking finishes and terminal HEPA filters

Outdoor air is provided via a dedicated outdoor air handling pre-treatment unit. This outdoor air unit is right sized to pressurize the cell banking suites. This amount of outdoor air is also sufficient to ventilate the spaces. This feature is a key differentiator between the old and new facilities. A dedicated outdoor air system (DOAS) system provides reliable environmental control with the added benefit of reduced energy costs.

Each room, as well as the corridors and gowning rooms are served via independent fan-coil units. These units recirculate air, which significantly increases energy efficiency, while maintaining room air segregation, as well as enhances the steady and reliable environmental control aspects of temperature and humidity. These attributes are particularly critical in a multiple product GMP environment and are also important given the program complexity of maintaining 65 degrees Fahrenheit room air within close humidity specification tolerance.


Figure 3: AHU 9 Temperature Data

Figure 3: AHU 9 Temperature Data

The old and new facilities both encompassed approximately 2,000 square feet of actual cGMP production space. Calculations were performed for air moving, cooling, and reheat costs across both facilities. These factors cost $117,228 in the old facility versus $40,034 in the new facility, which equates to an annual savings of $77,194.

These figures are based on the following calculations:

  • Old Facility
    • 10,665 CFM central AHU– 7 in SP (as balanced)
    • 82 Tons of cooling
    • 45 GPM of 20 degrees Fahrenheit reheat
    • Total suite operating all day and all night
  • New Facility
    • 6,000 CFM Pre-Treat – 1.3 in TSP (as balanced)
    • 11,165 CFM Indiv. AHU's – 2.0 in TSP (as balanced)
    • 44 Tons of cooling (53 percent of old facility capacity)
    • Individual suite cooling controls to 65 degrees Fahrenheit only when needed
    • 0 GPM reheat (as recorded for one full year)

Based on both facilities operating all day and all night, the projected energy costs are as follows. Note that the new facility has the flexibility to offer further savings if individual rooms are not required to be operational based on contract demand:

  • Original Facility Costs
    • Air Moving $19,915
    • Cooling $59,687
    • Reheat $37,626
    • TOTAL $117,228
  • New Facility Costs
    • Air Moving $8,030
    • Cooling $32,004
    • Reheat $0
    • TOTAL $40,034

The key items are obvious. The lack of required reheat because of the use of a DOAS system represents an enormous energy savings feature. This is not just calculated or projected; it is also proven via design and trending information. The design of the fan coil units does not include reheat coils; the heating coils are in a pre-heat position. Also, the trending data indicates that those heating valves are always closed.

The air moving costs are much less because of the actual balanced static pressures associated with the fans. Because of the longer duct runs required from the large air handler, the old facility unit was operating at 7 inches of static pressure. The new facility has close coupled rooms and fan coil units, which operate at 2 inches of static pressure.

Finally, cooling is obviously operating at a lower cost since the outdoor air quantity is reduced by almost 50 percent.

These costs may appear astronomical for such a small production facility, but we hope that they provide a dramatic, yet classic, example of the inefficiencies experienced in many highly specialized laboratory and GMP production environments. Often, the easy answer of 100 percent outside air is utilized in these facilities, when there are alternatives to this paradigm. Others are encouraged to challenge what is done elsewhere and look beyond the easy answer to more efficient alternatives. This small example offers an excellent opportunity for far greater savings in many other GMP, or other controlled environments.

On that note, of challenging the established industry standards, it would be interesting to take this case study a step further by experimenting with lower air change rates, while achieving the same cleanliness levels, i.e., air classification levels. By reducing air change rates, it would be possible to significantly reduce HVAC energy consumed in GMP clean room environments. Efforts to challenge air change rates have been under way for some time and are proliferating in the area of computational fluid dynamics; however, references in current FDA, ISPE and WHO documents to air change rates are typically carried as codified requirements. As with most other established guidelines and regulations, this may be prove to be an "uphill climb", but we believe it to be possible to achieve established air classification levels with a substantial reduction in air change rates. We have recently built a facility utilizing calculated air change rates to predict particulate counts in lieu of using historical standards. We hope to be able to prove this point in a subsequent paper.


Jason Rifkin has worked in the life science industry for the past 15 years and brings extensive experience in the biotechnology and life science sector previously working for Celera Genomics as a Quality Control Supervisor and NeuralStem as a Researcher and Laboratory Manager. Mr. Rifkin has worked with Turner Construction Company as a life sciences construction market consultant and for Scheer Partners as a senior vice president in charge of development and construction of life science facilities.

Mr. Rifkin is the principal of Equilibrium and oversees program management and design-build projects for life science facilities. Mr. Rifkin holds a Bachelors of Science in Biology from the University of Maryland at Baltimore County, a Masters of Science in Neurobiology from Montana State University, and a Masters of Business Administration from the University of Baltimore. Mr. Rifkin has also published articles in peer reviewed scientific journals.

Patrick Goetz has more than 20 years experience in the life sciences industry where he has prepared master plans, prepared energy analyses, prepared contract documents, estimated and provided construction, commissioning and project management services for single buildings as well as campus facilities. Projects have ranged from research and development to manufacturing facilities for research, biotechnology and pharmaceutical clients.

Mr. Goetz's current duties at Southland include performing engineering analyses to determine work programs that satisfy customer work requests, producing proposals, and negotiating contracts. He also acts as the engineering leader for the life sciences group. In this role, he prepares basis of design and red line documents and he reviews engineering calculations and design drawings. Mr. Goetz also remains involved during construction to review submittals as well as work progress and conformance with the program.