It is well understood heating, ventilation and air-conditioning (HVAC) systems’ cooling coils are reservoirs of microorganisms typically identified with poor IAQ and Hospital Acquired Infections. In addition to poor IAQ, these microorganisms develop a biofilm on HVAC coils resulting in poor mechanical performance. Poor performance can be measured in reduced airflow, increased pressure differential and poor air flow uniformity. ASHRAE document “HVAC Design Manual for Hospitals and Clinics 2013” states “. . . an uneven air velocity distribution across the coil face can result in loss of capacity, moisture carryover or freeze up problems”. ASHRAE recognizes Ultraviolet Germicidal Irradiation (UVGI) as an effective tool to treat biofilms growing on HVAC surfaces. ASHRAE Handbook: HVAC Applications suggests 50-100μW/cm² (0.394 in²) of UVC intensity, at the coil’s surface, can be an effective coil treatment. However, the document does not quantify the effectiveness nor does it address the UVC intensity required for coil penetration dependent upon fin spacing or coil depth. This paper will present data from theoretical modeling and laboratory measurements of the UVC intensities at the surface of typical HVAC coils. To understand how effectively the UVC penetrates the coil’s interstitial spaces, measurements were taken at the coil’s surface, 2” (50.8 mm) and 4” (101.6 mm) depths respectively. UVC intensities were measured and recorded with a calibrated radiometer capable of producing results that are traceable to NIST and through NIST to the International System of Units (SI), ANSI/NCSI Z540.1 – 1994 and ANSI/NCSI Z540.3 – 2006. Theoretical reductions of typical coil biofilms are presented with varying UVC intensities at varying coil depths. The comparative analysis is demonstrated at intensities ranging from 50-1000 μW/cm². The comparisons are based upon laboratory analysis, published data, and data from applied field studies quantifying the microorganism concentrations on cooling coil surfaces (Leach and Scheir 2014; Ryan, et al., 2011). Field data are presented from case studies of two tertiary care hospitals. The two hospitals were experiencing severely underperforming HVAC systems after repeated chemical coil treatments. Both hospitals were able to mitigate the coil’s biofilm and restore optimum performance by applying a minimum UVGI intensity of ≥1000 μW/cm² (0.394 in²). Contamination from coils was analyzed pre and post UVGI installation. Based upon these data, standards can be established for minimum coil UVC intensities to effectively treat varying coil depths and fin spacing’s.
Timothy Leach is the Director of Healthcare Solutions at Steril-Aire, Inc. Burbank, CA. Graham Taylor is the Director of Engineering at Steril-Aire,
Inc. Burbank, CA
Bacterial and fungal contamination of heating, ventilation and air-conditioning (HVAC) cooling coils is a widespread phenomenon which leads to poor indoor air quality (IAQ) (Huegenholtz 1992; Levitin et al., 2001; Apter 1994) and Hospital Acquired Infections (HAIs) (Walter 1969; Ryan et al., 2011). The downstream or supply side of the cooing coil is typically where the highest concentration of microorganisms exist, typically in concentrations of 105 – 106 colony forming units (CFU) of microorganisms per cm² of coil urface area (Hugenholtz and Fuerst 1992; Ryan et al., 2011; Leach and Scheir 2014). In addition to poor IAQ the biofilms growing on cooling coils can lead to underperforming HVAC systems often resulting in reduced airflow, increased pressure differential and uneven airflow uniformity (ASHRAE 2013; Kowalski 2011; Sigel and Nazaroff 2002).
Ultraviolet Germicidal Irradiation (UVGI) is recognized as an effective tool to treat biofilms growing on HVAC surfaces (ASHRAE 2015; ASHRAE 2016; Leach and Scheir 2014; Ryan et al., 2011). The ASHRAE Handbook for HVAC Applications suggests 50-100μW/cm² (0.394 in²) of UVC intensity, at the coil’s surface, can be an effective coil treatment. However, the document does not quantify the effectiveness nor does it address the UVC intensity required for coil penetration dependent upon fin spacing or coil depth. This paper will present data from laboratory modeling and measurements of the UVC intensities at the surface of typical HVAC coils. Additional analysis is presented to demonstrate the effective UVC penetration of the coil’s interstitial spaces and theoretical microbial reduction on coil surfaces.
There are two alternative verification methods to test the effectiveness of coil cleaning and disinfection are: A.) sampling for bacteria and fungi pre versus post coil treatment or B.) determining if the coil performance characteristics are returned to design conditions (Kowalski 2011). This paper demonstrates results using both verification methods and will conclude with presentation of field data from case studies of two tertiary care hospitals. The two hospitals were experiencing severely underperforming HVAC systems after repeated chemical coil treatments. Both hospitals were able to mitigate the coil’s biofilm and restore optimum performance by applying a minimum UVGI intensity at the ccoil’s surface of ≥1,000 μW/cm². Contamination from coils was analyzed pre and post UVGI installation. Based upon these data, standards can be established for minimum coil UVC intensities to effectively treat varying coil depths and fin spacing.
LABORATORY MODELING & THEORETICAL MICROORGANISM REDUCTION
Methods and Materials
A simplified two-step model was developed and implemented to better understand UVGI impact on biofilms growing several inches within the coil’s interstitial spaces. The first step modeled and measured UVC energy (μW/cm2) penetration through a typical HVAC cooling coil. Once the UVC energy penetration was established it was and then theorized an expected reduction of biofilm.
The UVC modeling used two different depths of coils (Rahn Industries Whittier, CA), 30” (762 mm) H x 40” (1016 mm) W x 2” (508 mm) D and 30” (762 mm) H x 40” (1016 mm) W 4” (1016 mm) D respectively. Coils were new and unused and constructed of aluminum fin and copper tubing. Fin spacing for both coils was 10 fins per 1” (25.4 mm). The tests were performed at room temperature with coils not charged with refrigerant. Two separate UVC energy sources (Steril-Aire, Inc. Burbank, CA) were used, one 24” (610 mm) and one 36” (914 mm) UVC lamp respectively. A UVC source was positioned 20” (508 mm) from the UVC radiometer’s sensor (Gigahertz-Optik, GmbH Pucheim, Germany) and 12” (305 mm) from the face of the coil as shown in Figure 2. The UVC sensor was mounted on a tripod in a fashion as to be centered on the face of the coil as shown in Figure 1. At the start of each test the UVC lamp was activated and run for 5 minutes for the temperature stabilization. Three measurements were recorded for each UVC energy source. The first measurement was recorded without a coil and one each with the 2” (50.8 mm) and 4” (101.6 mm) coils between the source and the radiometer’s sensor as shown in Figure 2.
Results and Discussion
UVC Modeling & Theoretical Microorganism Reduction. The UVC measurements indicate that only a fraction of energy irradiating the face of a coil make it deep into the coil’s interstitial space. This can be attributed to several factors including energy loss based upon Newton’s inverse square law, reduced line of sight due to tight fin spacing and bends and aluminum fin adsorption of approximately 26% of energy (Kowalski 2011) with every reflection. This simplified test (without considering operational conditions such as moisture, broad spectrum temperatures and contamination) demonstrate a 1-Log (10-1) loss of UVC energy for every 1” (25.4 mm) of coil penetration as shown in Table 1. Based upon this model, a 50-100 μW/cm² (ASHRAE 2015) of UVC intensity at the coil face would yield conservatively 0.5-1 μW/cm² (0.394 in²) at 2” (50.8 mm) of coil depth.
The second step of the analysis considered the theoretical effectiveness of 100 μW/cm²(0.394 in²) coil face intensity (ASHRAE 2015) on a monolayer of biofilm at 2” (50.8 mm) of depth within the coil. A typical monolayer of biofilm will contain 105 – 106 colony forming units (CFU)/cm2 (Leach and Scheir 2014; Ryan et al., 2011; Moreau-Marquis et al., 2010). For purposes of demonstration we selected Pseudomonas sp., a bacterium commonly isolated from HVAC systems (Leach and Scheir 2014; Ryan et al., 2011). Bacteria will double approximately every 20 minutes. The UVC dose required for a 1-Log or 90% reduction of Pseudomonas sp is 5,495 μJ/cm² (Zelle 1955). The expected cumulative 20-minute dose for the Pseudomonas sp. on the coil surface at the 2” coil depth would be 1μW/cm² (0.394 in²) intensity x 1,200 seconds equaling 1,200 μJ/cm² (0.394 in²). This equates to approximately a 22% reduction of the Pseudomonas sp every 20-minutes. The result: 105 or 100,000 CFUs of Pseudomonas sp x 22% reduction = 78,000 CFU survival. The surviving 78,000 CFU of bacteria are doubling through the 20-minute period. This equates to an increase of the concentration CFUs within the biofilm of by ≥50% or approximately 156,000 CFUs.
An increase of 1-Log or a factor of 10 of UVC intensity at the coil face would produce significantly different results. For example, an intensity of 1,000 μW/cm² (0.394 in²) at the face of the coil would yield 10 μW/cm² (0.394 in²) at 2” (5.08 cm) of coil depth. Over a 20-minute period the expected cumulative dose for the Pseudomonas sp. on the coil surface at 2” coil depth would be 10μW/cm² (0.394 in²) x 1,200 seconds or 12,000 μJ/cm²(0.394 in²). The result: 105 or 100,000 CFUs of Pseudomonas sp x 99% reduction = 1,000 CFU survival. The surviving 1,000 CFUs plus the bacteria doubling through the 20-minute period would still reduce the biofilm CFU concentration to only 2,000 CFUs. The next 20-minute period would result in an additional 99% reduction of the surviving 2,000 CFUs to approximately 40 CFUs.
This scenario demonstrates the 50-100 μW/cm² (0.394 in²) of UVC is not effective in removing a biofilm growing within the depth of a coil. Conversely an increase from 100 μW/cm² (0.394 in²) to 1,000 μW/cm² (0.394 in²) of UVC intensity at the coil face will produce the intended results of cleaning the interstitial spaces of the coil’s fins.
UVGI FIELD APPLICATION & ANALYSIS
Methods and Materials
Two hospitals were selected for installation of UVGI. Hospital A is a for profit hospital located in Central California and Hospital B is a University teaching hospital located in Central Tennessee. Hospital A and Hospital B were experiencing increased delta pressure, reduced airflow and uneven airflow uniformity across the HVAC system’s cooling coils. Hospital A had an eight (8) row coil and Hospital B had a ten (10) row coil. The fin spacing for the coils at both Hospitals were ten fins per 1” (2.54 cm). Hospitals A’s and B’s initial process of mitigation was to power wash the coils upstream and downstream with an industrial grade coil cleaning surfactant. Hospital A cleaned the coils on four occasions over the span of 30-days, using hospital engineering personnel. Hospital B cleaned coils on two occasions over the span of 14-days, using an outside commercial coil cleaning contractor.
Both Hospitals realized no measurable airflow improvements from the chemical cleaning as can be seen in Table 4.Based upon these results Hospital A and B opted for an alternative mitigation process. The alternative process selectedby both Hospitals was the installation of UVGI systems (Steril-Aire, Inc.) at the air effluent side of the cooling coils. ASHRAE has identified UVGI as a technology that can prohibit the growth of biofilms on HVAC systems cooing coils(ASHRAE 2105). The UVGI systems, installed by Hospitals A and B were designed to deliver a minimum UVC intensity of ≥ 1,000 μW/cm2 (0.394 in²) across the entire face of the air effluent side of the cooling coils. The UVC intensity was selected in order to effectively penetrate several inches into the cooling coils’ fins. The intention was to destroy and remove the biofilm believed to be the cause poor airflow performance.
Hospital A and B determined to use a two-step analysis to quantify coil performance improvement. The first step was isolation and identification of the coil contaminants. The contamination analysis entailed culturing microbial samples from the coil surfaces with sterile swabs (Healthlink, Inc.) and collection of a hard black material coming from within the interstitial space of the cooling coil fins. This was done pre and 60-day post UVGI installation. The second step involved delta pressure (ΔP) across the cooling coils.
Coil Biofilm Analysis. HVAC coil samples were collected using methods standardized for inanimate surface (AIHA: AIHA Field Guide. 1996) and were analyzed by an independent laboratory (Pure Earth Environmental Laboratory, Inc., Pennsauken, NJ). Each sample collected was one square inch (2.54 x 2.54 cm) from the air effluent side of condensate cooling coils of the HVAC systems. Surface wipe samples were obtained using a BBL culturette (Becton Dickinson, Franklin Lakes, NJ) with a sterile rayon-tipped swab that was moistened with a modified Stuart’s transport medium before sampling one square inch of surface area. The completed swab specimen was placed back into its original container, sealed and shipped via next day air to the laboratory for identification and quantification of fungi and bacteria to the species level. Upon arrival at the laboratory, each surface wipe was immersed in a sterile test tube containing 10 ml of sterile distilled water. The test tube sample was kept at room temperature for 10 minutes and then placed in a rotary shaker (3.81 throw, 220 rpm) for one minute. The resulting suspension or dilution was then inoculated (0.1 ml aliquots) on a 2% malt extract agar (MEA for fungal growth) and a trypticase soy agar (TSA for bacteria growth). The results provided estimates of the total number of viable propagules per ml of suspension. The samples were immediately incubated at 34.5° to 35.5°C, along with laboratory controls.
Coil Bulk Contamination Analysis. Coil Bulk Contamination Analysis. Contamination analysis was performed on a black, hard bulk contamination from the coils as can be seen in Figures 4 and 5. After samples were collected they were sent to a University Laboratory (State University of New York at Buffalo Instrument Center) for Scanning Electron Microscopy (SEM) analysis.
HVAC Mechanical Analysis. Delta pressure readings were recorded across Hospital A and Hospital B’s cooling coils post chemical cleaning and post UVGI installation. The measurements were recorded with an electronic micro manometer (Shortridge Instruments, Inc. Airdata Multimeter ADM-870C). Hospital A’s used hospital personnel to record coil pressure readings, while Hospital B used an outside air test and balancing contractor.
Results and Discussion
Hospital A and Hospital B realized no measurable airflow improvements from the chemical cleaning process. Both Hospitals then installed UVGI systems (Steril-Aire, Inc.) at the air effluent side of the cooling coils as an alternative means of mitigation. The UVGI systems, installed by Hospitals A and B were designed to deliver a minimum UVC intensity of ≥ 1,000 μW/cm2 (0.394 in²) across the entire face of the air effluent side of the cooling coils
Hospital A and B used a two-step analysis to qualify success: A.) quantify contamination reduction and B.) monitor airflow performance improvements. The first step was isolation and identification of the coil contaminants. Coil cultures were collected along with bulk contamination. Airflow performance was measured by means of coil differential monitoring.
Coil Cultures. The coil cultures were collected post chemical treatment, pre UVGI installation and post UVGI installation at Hospital A and Hospital B. The samples were collected at Hospital A 30-days post coil chemical cleaning and pre UVGI installation. Hospital B’s sampling was conducted 60-days post chemical cleaning and pre UVGI installation. Hospital A had nine species of bacteria and fungi and Hospital B eight as seen in Table 2. The Hospitals shared six similar species of microorganisms indicating the biofilms were similar in nature. This is noteworthy because the Hospitals are in two very different locations within the continental US, Hospital A Central California and B Central Tennessee (Acerbi 2016). The concentrations of microorganisms are impressive considering the sampling was conducted post chemical coil cleaning.
The 60-day post UVGI installation samples resulted in a 5-Log reduction of microorganisms on the surfaces of the supply side of the HVAC coils for Hospital A and B. Microbial counts were reduced from 79,775 to 0 at Hospital A and from 327,285 to 2 CFUs at Hospital B, respectively. One of the methods of Based upon reduction of microorganisms on the coil surfaces, as noted in Table 3, it could be determined the UVGI is functioning as intended.
Bulk Contamination Analysis. The contamination collected from the coils in Hospitals A and B’s coil fins and condensate drain pan can be seen in Figures 4 and 5. The material was collected and sent to the State University of New York at Buffalo Instrumentation Lab for analysis using a SEM EDX. The physical description of the material was black in color, hard and porous resembling activated carbon as can be seen in Figure 5.
Figure 4. Contamination from Coils
Figure 5. Contamination collected from Coils
The analysis performed at the State University of New York at Buffalo Instrumentation Lab indicated the bulk contamination as seen in Figure 5. to be a biofilm. The biofilm was predominantly made up of fungal spores and hyphae as can be seen in Figure 6.
Figure 6. SEM Identification of Black Contamination – Fungal Spores & Hyphae
HVAC Mechanical Analysis. Airflow measurements were recorded with an electronic micro manometer (Shortridge Instruments, Inc. Airdata Multimeter ADM-870C). Hospital A’s coil differential pressure readings were reduced by 2.19” 52-days post UVGI installation and Hospital B’s coil differential pressure was reduced by 2.6” ΔP post 74-days post UVGI as can be seen in Table 4. The disparity in the number of days to improve the coil differential pressures may be attributed to the depth of the cooling coils. Hospital A seeing results in 52 days had an eight row coil and Hospital B taking 74 days to see results had a ten row coil.
The intent of the UVGI installation was to return the coils to as close to design conditions as possible. Both Hospital A’s and B’s design conditions were 1.25” Δ pressure. The data from the monitored results indicate a successful airflow performance improvement and fouled coil mitigation process.
Two methods of analysis were used to validate the effectiveness of coil cleaning by UVGI treatment. The first was taking pre cleaning and post cleaning swab samples to culture for the presence of bacteria and fungi. This analysis determined the effectiveness of disinfection. The second method monitored the airflow performance improvement to determine if coil characteristics returned to design conditions. (Kowalski 2011). The data presented in this paper demonstrate a minimum UVC energy field at the face of the cooling coil of ≥1,000 μW/cm² (0.394 in²) resulted in coil disinfection and improved airflow performance. This paper also brings into question the effectiveness of suggested 50-100 μW/cm² (0.394 in²) as stated in the ASHRAE handbook for HVAC application.
With that being stated the the methods used in this study described in this paper do have limitations. More extensive and multidiscipline studies are needed to better understand the impact of biofilms on cooling coil performance. If UVGI is to be universally accepted as a standardized HVAC device, there needs to be a more refined understanding as to the levels of UVC energy to optimize cooling coil performance. This refinement should be predicated on coil size, configuration, fin spacing, and materials of construction. Lastly ASHRAE should define performance levels in terms of “efficiencies” similar to those established for filtration industry. These “efficiency” standards should be performance based, application specific designs that can be both measured and commissioned.
We would like to acknowledge Vanderbilt University Medical Center and Kaiser Permanente Healthcare for participating in the field studies.
Acerbi, E., Chenard, D., Miller, D., Gaultier, N.E., Heinle, C.E., Chang, V.W-C., Uchida, A., Drautz-Moses, D.I., Schuster, S.C., Lauro, F.M. 2016. Ecological succession of the microbial communities of an air-conditioning cooling coil in the tropics. Indoor Air doi: 10.1111/ina.12306
Apter, A., Bracker, A., Hodgson, M., Sidman, J., Leung, W-Y. Epidemiology of the sick building syndrome. The Journal of Allergy and Clinical Immunology. 1994 Vol 94, Issue2, Part 2, Pg. 277-288
American Industrial Hygiene Association (AIHA): Field Guide for the Determination of Biological Contaminants in Environmental Samples. Fairfax, VA: AIHA, 1996
ASHRAE. 2013. ASHRAE Handbook—HVAC Design Manual for Hospitals and Clinics, Atlanta: ASHRAE.
ASHRAE. 2015. ASHRAE Handbook—HVAC Applications, Atlanta: ASHRAE.
ASHRAE. 2016. ASHRAE Handbook—HVAC Systems and Equipment, Atlanta: ASHRAE.
ANSI/ASHRAE-SPC-185.2-2014. Method of Testing Ulratviolet Lamps for use in HVAC%R or Air Ducts to Inactivate Microorganisms on Irradiated Surfaces.
Hugenholtz, P., Fuerst, J. Heterotrophic Bacteria in an Air-Handling System. Applied and Environmental Microbiology, Dec. 1992, P. 3914-3920
Kowalski, W. UVGI for Cooling Coil Disinfection, Air Treatment and Hospital Infection Control. American Air & Water, January 24, 2011
Leach, T., Scheir, R. Ultra Violet Germicidal Irradiation (UVGI) in Hospital HVAC Decreases Ventilator Associated Pneumonia. ASHRAE, NY-14-C023
Levetin, E., Shaughnessy, R., Rogers, C.A., Scheir, R. Effectiveness of Germicidal UVRadiation for Reducing Fungal Contamination within Air-Handling Units. Applied and Environmental Microbiology, Aug. 2001, p. 3712-3715
Moreau-Marquis, S., Redelman, C.V., Stanton, B.A., Anderson, G.G. Co-cultures of Pseudomonas aeruginosa biofilms grown on line human airway cells, Journal of Visualized Experiments. DOI:10.3791/2186
Ryan, R.M., Wilding, G.E., Wynn, R.J., Holm, B.A., Leach, C.L. Effect of enhanced ultraviolet germicidal irradiation the heating ventilation and air conditioning system on ventilator-associated pneumonia in a neonatal intensive care unit. Journal of Perinatology. (2011), 1-8
Siegel, J.A., Nazaroff, W.W. (2003). Predicting particle deposition on HVAC heat exchangers. Atmospheric Environment 37(39),5587-5596. doi:10.1016/j.atmosenv. 2003.09.033
Walter, C.W., (1969). Ventilation and air conditioning as bacteriologic engineering. Anesthesiology. 31, 186-192
Zelle, M.R., Hollaender, A. (1955). Radiation Biology Volume II. McGraw-Hill, New York
2017 ASHRAE Winter Conference
LEAFPOWER CO., LTD.
54, 56, 58, 60 Soi Pattanakarn 64, Prawet Sub-district, Prawet District, Bangkok 10250 THAILAND
Tel : 02-130-6371
Fax : 02-130-6372