Tobacco Documents Online

90-92.6 TWO INDOOR AIR QUALITY INVESTIGATIONS -- OCEANS APART Simon Turner Healthy Buildings International Inc. (HBI) Fairfax, Virginia AiR & WASTE MANAGEMENT A S S 0 C I A T t 0 N Sirra 1907 For Presentation at the 83rd Annual Meeting & Exhibition Pittsburgh, Pennsylvania June 24-29, 1990 TI DN 0004448

90-92.6 INTRODUCTION A commonly seen method of indoor air evaluations in buildings perceived as "sick" involves an engineering study of a building in conjunction with a questionnaire administered to its occupants. As a result we now have evidence that tighter, air conditioned buildings generate higher incidence of upper respiratory complaints than naturally ventilated buildings.1•Z However, this combination of engineering evaluation and occupant questionnaire as a diagnostic method for sick buildings has not provided us with the ability to predict the cause of indoor air problems from occupant symptoms. One major reason for this appears to be that many common irritants found in indoor air result in a similar range of upper respiratory symptoms. Many diagnosed "sick" buildings are reported to have their causes rooted in ventilation, filtration or hygiene problems which allow a mixture of low level pollutants to build up.3•4 These irritate occupants in a complex and possibly synergistic manner, the mechanism of which is largely unknown to us. To demonstrate the range of engineering characteristics which can produce similar complaints of sick building syndrome, this paper presents two case histories of identically sized buildings, the studies of which were completed in July of 1989. One of these buildings (designated as Building A) was located in Washington D.C., and the other (designated as Building B) was located in San Diego, California. Both buildings are 600,000 square feet in size and a full description of their respective characteristics is found in the results section. Symptomology The following symptoms were reported in Building A during the year previous to this study: sneezing and/or coughing; sinus congestion; dry nasal passages; post nasal drip; rashes and dry skin; headache; breathing difficulties; sore throat; drowsiness; fatigue; eye irritation. In Building B the following complaints were noted over the same period of time: odors; headaches; shortness of breath; eye irritation; dizziness. These symptoms in both buildings compare well with the classic symptomology of the sick building syndrome as defined in the Commission of the European Communities Report -- Sick Building Syndrome, A Practical Guides which lists the symptoms as follows: nasal irritation with nasal stuffiness; dryness and irritation of the mucous membrane of the eye; dryness and irritation of the throat; skin dryness and irritation; headaches, generalized lethargy and tiredness. 2 TI DN 0004449

METHODOLOGY 90-92.6 Although a standard approach was used to survey each building, it required flexibility to cope with the different types of buildings examined. Initial Walk Through Since one of the objectives of this study was to assess maintenance standards, each building evaluation commenced with an interview with the personnel responsible for maintenance of the building. Questions were designed to elicit operative details such as system on/off times; outside air, return air and exhaust settings; scheduled maintenance routines; and complaint areas. This part of the survey did not include questioning of the occupants themselves, however. There was a walk through of each building to identify obvious building configurations or design features which could influence air quality in the occupied areas. This was followed by a visual inspection of the internals of the building's ventilation system. This consisted of the internal chambers of the air handling units including the condition of the coils, fan chambers, humidifiers and condensate trays, and a detailed assessment of the type, installation and condition of the filters in use. A visual inspection was done of the internals of the main air supply ductwork leaving each air handling unit. Where necessary, access was gained to this ductwork by the installation of a small access port and the insertion of a fiber optic borescope. Qualitative Sampling In each air handling unit and main air supply duct a series of samples were also collected on cellulose ester filters for light microscopy analysis and surface microbe samples were collected on Random Organism Detection and Counting (RODAC) agar plates to be subsequently incubated, counted, and identified. A laser particle counter with a size selective inlet for sampling particles with an aerodynamic diameter of 0.5 microns and above (Met One Inc., Oregon) was used to count particles inside the ductwork. At least two points were sampled inside each major run of ductwork. This qualitative information on the building, along with the location of the samples and the building engineer questionnaire was prepared on a set of standard field notes to ensure consistency. Quantitative Air Sampling Due to time constraints, a quantitative assessment of air volume flows in, around and out of each building was outside the scope of this work. Instead, other measurements, such as that for carbon dioxide, were used to assess ventilation rates. A set of locations were identified in each building to be used for quantitative airborne sampling. These locations were spread evenly throughout the study area of each building with a minimum of two locations per floor, as 3 TI DN 0004450

90-92.6 well as an outdoor control sample point. The following parameters were measured at each location where relevant and appropriate: o Respirable airborne particle counts using a piezoelectric microbalance (TSI, Inc., St. Paul, Minnesota). This microbalance measures particles in the 0.01 to 3.5 micrometer size range. It has a nominal sensitivity of 180 micrograms per Hz and was used in the 120 second mode. Flow rate through the piezobalance was periodically checked at one liter per minute with a bubble flow meter, and the sensor was cleaned with alcohol swabs after every five measurements. The unit is factory calibrated with diluted welding fumes which have shown equivalence to indoor RSPs to ±10%. The lower detection limit was set at 10 Ngm"3. o Carbon dioxide levels using a non-dispersive infrared absorption portable gas analyzer, sensitive to 50 ppm (C02). Accuracy is ±29 over full scale (CEA Instruments, Inc., Emerson, New Jersey). Periodic calibration of the instrument was with a factory supplied span gas of 5.000 ppm COZ. Zero was set with dry nitrogen gas and the lower detection limit was set at 50 ppm. o Carbon monoxide concentrations using a controlled potential electrolysis detector, accurate to 10% full scale (Sensidyne Inc., Largo, Florida). Periodic calibration of the instrument was with a factory supplied span gas of 50 ppm carbon monoxide. The minimum detection limit was set at I ppm. o Airborne nicotine (after Ogden et al)6 with a personal universal flow sampling pump (SKC Inc., Eighty-Four, Pennsylvania) drawing air at one liter per minute for a period of one hour through unfiltered XAD4 absorbent resin tubes which collected a portion of particulate as well as gaseous phase nicotine. Each tube contained an 80 mg front and 40 mg rear portion of resin to detect any sample break-through. Samples showing reduced collection efficiency, where nicotine was found in the rear tube above our detection limit, were rejected. Samples were desorbed into ethyl acetate containing 0.01% triethylamine and analyzed with gas chromatography (fused silica DS-5 column and thermionic-specific nitrogen-phosphorus detector, Hewlett Packard Model 5880A). Sampling pumps were calibrated daily to one liter per minute with a"Gilibrator" bubble flow meter (Gilian Instrument Corporation, Wayne, New Jersey). The gas chromatograph was calibrated with freshly prepared nicotine standards. All samples and blanks were doped with an internal standard. Results are expressed in total micrograms converted to µgm-3, and the detection limit for our sampling rate of lmin'1 for a one hour period was 1.6 µgm-3 of air. Supplemental measurements of carbon dioxide, temperature, and relative humidities were also made at each location, so that an assessment of environmental tobacco smoke levels as a whole could be made. o Temperature using a miniature platinum Pt 100 resistance sensor, conforming to 1/2DIN 43760, Class A (The Dickson Company, Addison, Illinois). Resolution was to 0.06•C and accuracy was ±0.3°C. The unit was factory calibrated and checked periodically against dry bulb thermometers. 4 T1 DN 0004451

90.92.6 1 o Relative humidity using a chromed layered capacitative electrode (Dickson Company, Addison, Illinois). Resolution was 0.18 relative humidity and accuracy was to ±0.28 relative humidity at 23°C. Temperature drift was ±0.05% relative humidity per 1.1"C. The unit was pre-calibrated with a factory supplied salt solution containing lithium chloride (12% relative humidity) and sodium chloride (75% relative humidity). The following parameters were measured in at least two selected locations in each building: o Miscellaneous gases using Gastec calibrated detector tubes (Sensidyne Inc., Largo, Florida) as follows; nitrogen dioxide (DL - 0.2 ppm); low range hydrocarbons (DL - 0.01%); high range hydrocarbons (DL - 20 ppm); ozone (DL - 0.05 ppm); ammonia (DL - 0.15 ppm); sulphur dioxide (DL - 0.1 ppm). An outdoor control was taken at each building. These screening measurements have an accuracy of approximately ±25%. 0 Airborne microbial counts using a centrifugal air sampler employing impaction onto an agar lined drum (Biotest Diagnostics, Frankfurt, W. Germany). The sampler had a separation volume of 40 lmin'1 and each agar strip was exposed for a two minute period. A GK-A culture medium strip was used which is suitable for total counts. Incubation was at 30 to 35°C for 48 hours, followed by counting and species identification. An outdoor control was taken at each building. o Formaldehyde using midget impingers containing sodium bisulphite followed by spectrophotometric analysis, after NIOSH Method P6CAM 125N (now NIOSH Method 3500). Accuracy is approximately ±10%, and detection limit was set at 0.01 ppm. Standards The visual inspections provided evidence of the physical state of the air handling plant and how it was being operated and maintained. The levels of the airborne pollutants measured were compared with recognized standards where they were available, where none existed they were compared with levels which in HBI's experience have been found to be satisfactory. Table I shows the parameters measured and standards used to assess air quality in the buildings. Microbes There are no established standards for airborne microbes in the indoor air of office or other commercial buildings, although over many years attempts have been made to set acceptable upper limits.7,8•q Based on the experience gained in more than 63 million square feet of building space HBI suggest an upper acceptable limit of 750 colony forming units per cubic meter of air with the provision that if the total airborne count is lower but species known to cause allergies or infections are identified even this figure may be unacceptable and steps should be taken to control them. From the results of more than 200 studies world-widelo members of the same four fungal genera, namely cladosporium, penicillium, aspergillus and alternaria, have been found to account for the highest mean percentages isolated and from data based on skin reactivity studies" the same four types constituted the most prevalent 5 %~W TI DN 0004452

90-92.6 Table I. Parameters measured, standards and sources. Parameter Standard ou ce Carbon dioxide 800 ppm HBI* Carbon monoxide 9 ppm EPA12 Respirable Suspended Particulates 75 ug/m3 HBI** Microbes 750 cfu m3 ~ HBI** Nicotine 50 ug/m HBI (ASHRAE***) Ozone 0.05 ppm ASHRAE Formaldehyde 0.1 ppm HBI (ASHRAE***) Temperature 20 to 26°C ASHRAE Relative Humidity 20 to 70% ASHRAE * HBI modification of American Society of Heating, Refrigerating and Air Conditioning Engineers Standard, ASHRAE 62-1989" ** Suggested upper acceptable limit based on HBI experience.4 *** ASHRAE modification of American Conference of Governmental Industrial Hygienists (ACGIH) recommendation14 fungi involved in allergic respiratory diseases with 85% of patients found to be allergic to one or more members of these fungal genera. The bacterial species which have been isolated from surface and air samples in the indoor environment and which can cause infections in people include members of the staphylococcus, pseudomonas and flavobacterium groups. Ventilation Good ventilation for commercial office buildings is defined by ASHRAE, who in their recently published Standard 62-1989, 'Ventilation for Acceptable Indoor Air Quality,"13 call for either a minimum intake of 20 cubic feet per minute of outside air per person at all times of building occupancy, or, evidence by measurements that no indoor air pollutants are accumulating in the building air in unacceptable amounts. An analysis of HBI Indoor Air Quality Investigations over the period from 1981 to 1988,4 given in Table II, shows that where the ventilation status of a Table II. Pollutant ranges for well and poorly ventilated buildings from HBI experience.4 Pollutant i1IlLEs Ventilation Status Good Poor Carbon dioxide ppm 400-700 800-2,500 Carbon monoxide ppm 0-5 10-25 Respirable Suspended Particulates 20-60 70-200 Microbes cfu/m3 50-600 700-2,000 Nicotine ug/m3 0-10 50-100 Formaldehyde ppm 0-0.05 0.06-0.25 6 TI DN 0004453

90-92.6 building is good the levels of airborne pollutants are found in low and acceptable levels; where ventilation is inadequate then they are found in correspondingly higher and unacceptable levels. RESULTS Characteristics of Building A S,,wstem Descristion. Building A is approximately 50 years o1d with six occupied floors, five above ground and one underground basement level. The building comprises an area of approximately 605,000 square feet and houses about 3,500 government employees. Heating, air conditioning and ventilation of the building is provided for by approximately 103 air handlers. Many of these units are located throughout the floor space of each level and generally serve adjacent or proximal areas close to the air handler. The larger air handlers, those serving designated areas on a number of floors, are located in the penthouse. A number of other large air handlers, serving large spaces on the first floor, are located in various fan rooms in the basement. Our study focused on a representative number of both the penthouse and basement air handling units and the associated systems. The design of the air handling system was such that relatively large volumes of outside air could be drawn in through either the penthouse or the basement units, through banks of poor quality one inch spun glass filters. The penthouse units were typical of air handling systems of this age in that they were constructed with dual chambers, one each for heating and cooling. Vertical air supply shafts then deliver air either to ceiling-mounted supply diffusers in the interior areas of the building, or forced air induction units (FAIU) around the perimeter. The return system for this building consists of 3'x3' return grilles set in corridors which lead back to large vertical return shafts connecting with either the basement or penthouse units. Exhaust is provided by large exhaust fans located in the penthouse which draw ducted air from grilles in the ceilings on each floor. This is supplemented with toilet exhaust. The air volumes in this building were once controlled by static pressure sensors in the floor spaces which worked in conjunction with vortex vanes on the air handling unit air supply fans, along with an air relief mechanism. This volume control system had been disconnected, allowing the building to run essentially as a constant volume system. The system was set to run 24 hours a day. The original air handling system described above has been added to as the building underwent renovations over the course of time. System Condition. Both the penthouse and the basement units were in need of some maintenance work to bring them up to a satisfactory condition. The preheat coils in the penthouse unit were not fitted with filters and were found to be clogged with dirt. Other filters were found to be loosely fitted and therefore operating at severely reduced efficiency. Virtually all of the units were found to be in need of a vacuuming out of loose dirt. The condensate trays and humidifier reservoirs were often in poor condition and 7 TI DN 000445,

( 90-92.6 contained stagnant water. Some badly damaged internal insulation was also noted which was releasing loa:e fibrous glass into the airstream. A number of the FAIUs were inspected and found to be very dirty. Visual inspection of the supply duct internals showed them to be moderately deposited with mixed granular dirt. They would prove difficult or impossible to clean due to poor access and the use of asbestos insulants in many areas. Airborne sampling results are shown in Table III. These show low levels of gases measured, such as carbon dioxide, carbon monoxide, formaldehyde and also nicotine and respirable sized particles. Higher levels of airborne microbes were found in some areas. Table III. Selected quantitative sampling results. Type test Buildine A Buildine B Respirable sized Outdoors 40 10 particles (pg/m3) Mean - Indoors 31 19 Range 20-50 10-85 Standard deviation 9.2 16.0 Carbon dioxide (ppm) Outdoors 350 350 Mean - Indoors 509 608 Range 350-750 450-800 Standard deviation 91 75 Carbon monoxide (ppm) Outdoors 3 2 Mean - Indoors 2.6 1.9 Range 2-3 1-2 Standard deviation 0.48 0.35 Formaldehyde (ppm) Outdoors (assumed) <0.01 <0.01 Mean - Indoors 0.016 0.048 Range <0.01-0.02 0.01-0.10 Standard deviation 0.008 0.03 Temperature ('F) Outdoors 89.5 76.0 Mean - Indoors 76.8 73.0 Range 73-79 72-74 Standard deviation 1.3 0.69 Relative humidity (!) Outdoors 87 44 Mean - Indoors 62.7 37.1 Range 50-69 32-48 Standard deviation 3.9 3.9 Nicotine (pg/m3) Outdoors (assumed) <1.6 <1.6 (discretionary Mean - Indoors <1.6 2.9 smoking areas) Range <1.6 2.1-3.7 Standard deviation 0 0.8 8 i.i TI DN 0004455

90-92.6 Table III. Seleited quantitative sampling resul-s (continued). Tvoe test Airborne microbes (cfu/m3) Building A Buildine B Outdoors >1250 700 Mean - Indoors 707 457 Range 138->1250 185-816 Standard deviation 499 153 Fungal species Penicillium Aspergillus identified from Aspergillus Cladosporium surface and airborne Penicillium sampling Scopulariopsis Acremonium Candida Oospora Chrysosporium Miscellaneous gases (nitrogen dioxide, lower and higher range hydrocarbons, ozone, ammonia, and sulphur dioxide); none of these were found above the detection limit of the method used and are, therefore, not included in this table. Rev: ppm - parts per million µg/m3 - micrograms per cubic meter cfu/m3 - colony forming units per cubic meter In summary, this building reflected the age in which it was designed in that it was built long before the energy crisis which forced ventilation system designers to create more energy efficient systems. As a result, it is equipped with a large number of air handling units bringing in relatively high volumes of outside air, and expelling equally large volumes of air via a powerful exhaust system. This is reflected in the carbon dioxide measurements which indicate the building, overall, was well ventilated. The powerful exhaust system also appeared to be removing internally generated particulates which were not at excessive levels. The building, however, was fitted with a very poor filtration system, and as a result of this along with its age, the air handling systems were dirty. These high dirt loads subsequently gave rise to relatively high airborne microbe counts and the presence of some species of fungi which are known to cause allergenic reactions in sensitive individuals, Most other commonly found indoor pollutants, including carbon monoxide, environmental tobacco smoke, and oxides of nitrogen were found at low levels in most areas, primarily because of the satisfactory ventilation rates. Recommendations for improvements in this building included a substantial upgrade of the filtration system to at least filtering rated to 35% by ASHRAE's atmospheric dust spot test (ASHRAE 52-76);15 cleaning work in each air handling unit, and FAIU repair and biocidal treatment of the condensate trays; decommissioning of the humidifier reservoirs during the humid east coast summer months; repair to loose internal insulation; and the treatment of cleaned surfaces with a broad spectrum biocidal spray. 9 TI DN o004456

90-92.6 Building B Svstem Descrintion. Building B is approximately eight years old and stands 24 stories tall. There is a basement level, a mezzanine above the first floor, and a penthouse that houses one of two mechanical equipment rooms, the other being in the basement. The building comprises approximately 600,000 square feet of space and housed close to 1,200 commercial office tenants at the time of our inspection. Conditioned air is provided by two main air handler units (AHUs), one located in the basement and the other in the penthouse. Each are served by a cooling tower mounted on the roof and chillers located in the basement. The system is a variable air volume (VAV) type delivering air via 14 VAV boxes located in the ceiling voids of each floor. Heating requirements are met by perimeter located reheat boxes in the ceiling voids, there being no requirement to heat the core of the building. The two main air handlers operate in a similar manner. Fresh air is drawn to the building and fed to the air handlers via openings in the second floor and on the rooftop. One obvious potential problem with the second floor AHU concerned the location of its fresh air intake. The intake was located directly over the loading bay and was also immediately adjacent to the garage. During high activity in the garage and high use of the loading dock, it is probable that vehicle exhaust fumes (especially hot diesel fumes from trucks using the loading area) will be induced into the air intake. This could result in complaints on any of the floors, 1 through 13, served by this unit. In each air handling unit, the air then passes into a mixing chamber and then through a set of high quality two inch pleated panel filters. The single chamber heating and cooling system is followed by a twin variable pitch air supply fan arrangement, which feeds supply air into round vertical air supply ducts to the various floors. On each floor the final supply to the occupied areas was via 14 variable air volume boxes per floor. The VAV boxes were designed with a minimum set point for the dampers. However, this setting resulted in too powerful an air jet at the diffusers so the units were modified to do away with the minimum setting and subsequently they can close completely. The perimeter areas on each floor were equipped with reheat boxes which draw in return air from the ceiling void through a spun glass furnace filter, then an electric heating coil, and then back into the occupied areas. Return air from the office space enters the ceiling void through light troffers. The ceiling void acts as a return air plenum and thus it is important that these plenums are not compartmentalized. Air moves across the void to stub ducts linking each void to the return air shaft located on the east side of the building. The low rise AHU draws air down this return shaft into the basement. The return air can be either exhausted through a set of eight exhaust fans per AHU, or it is mixed with a portion of fresh air in a prefilter chamber to start the cycle over again. It was noted at this inspection that the tenth floor ceiling voids had been partitioned from slab 10 .J TI DN 0004457