Tobacco Documents Online

James E. Woods and Brian C. Krafthefer FILTRATICN AS A METHOD FOR AIR QUALITY CONTROL IN OCCUPIED SPACES REFERENCE: Woods, J.E. and Krafthefer, B.C. "Filtration as a Method for Air Quality Control in Occupied Spaces," Fluid Filtration: Gas, Volume I, ASTM STP 975, R. R. Raber, Ed., American Society for Testing and Materials, Philadelphia, 1986. ABSTRACT: Filtration, or removal control, is one of three methods currently available to provide acceptable indoor air quality in occupied spaces. While the other two methods, source and dilution control, are primarily employed for occupant needs, filtration has conventionally been used for protection of components within the heating, ventilating, and air conditioning systems. In this paper, filtration tech- nology is reviewed with respect to current ventilation stand- ards for occupants. The difference between ventilation and air quality control is discussed in terms of acceptability criteria and control methods. Parameters that relate these terms are identified and control strategies are proposed that can be used to optimize removal and dilution methods for occupant acceptability and cost-effectiveness. KEYWORDS: Ventilation, Indoor Air Quality, Control Strategies, Removal Control, Filtration Although filtration technology has been applied in heating, venti- lating, and air conditioning (F1UAC) systems for more than fifty years, its primary function has not been provision of health or comfort for the occupants. Rather, filters are usually installed to protect the components (e. e. , coils, fans, ductwork) within the HVAC systems f rom dust and particulate loading. Only in special cases, such as critical care areas in hospitals, have filters been installed in the systems primarily for the protection of the occupants.. Otherwise, ventilation with outdoor air has been depended upon to dilute particulate and gaseous contaniiiants in occupied spaces to concentrations below those Dr. Woods is Senior Engineering Manager, Honeywell Energy Products Center, Golden Valley, MN 55422. Mr. Krafthefer is Principal Research Scientist, Honeywell Physical Sciences Center, Bloomington,. MN 55420. 193

194 FLUID FILTRATION: GAS considered to te deleterious or annoying. However, with the advent of energy conservation practices of the last decade, ventilation rates have been generally reduced. As a result, many ventilation systems no longer provide acceptable control of particulate and'gaseous contam- inants. The objective of this paper is to describe control strategies that can be used to optimize filtration and dilution for occupant acceptability ar(d cost-effectiveness. VENTILATION CONTROL A'schematic of conventional HVAC systems, Figure 1, identifies the catmon placements of filters (1]. In residential systems, filters are usually placed in the return air duct as these systems normally use 1001; recirculated air. If outdoor air is supplied mechanically to the system, a"pre-filter" may be installed, but usually dilution depends on infiltration (i.e., uncontrolled air leakage through the building envelope). Mechanical systems for commercial or institutional fa- cilities may have filters installed either in the return air duct or the "mixed-air" plenum. Note that the placement of these filters is upstreamlof the major mechanical components in the system. This prac- ti'ce is encouraged to achieve protection of the system from contamin- ation. The effectiveness of this practice has recently been docu- mented, and results indicate that significant energy and cost savings can accrue over the lifetime of the system (2]. AIR. EXHAUST CONDITIONING INFILTR UNIT ATION'. OTHER AIR CLEANER AIR CLEANER ATION' LOC LOCATION r'~~~~ _ ~ 1 OUTDOOR- AIR I i (MAKEUP ~. AIRI ALTERNATE _- I--t r _.~ F~ PAf11SFOR t ~ OTHER HCCI/IL'UTATED - - AI/tYILANER AIR -LOCATIONS LOCAL SUPPLY AIR ~ r/.-MAKEUP CONDITIONED SPACE L_4 LOCAL EXI/AUST 1 OCAI VCNf11.AIlON GENERAL \...- RETURN AIR EXHAUST l • EXFILrRAT1ON Figure 1. Heating, ventilating, and air conditioning,system schematic indicating variables for indoor air quality control. (From Ref. 1) Only In special cases have filters also been installed downstream of the major components to provide clean air to the occupied space. For example, in critical areas of hospitals, two filters are speci- fied. The upstream filter (pre-filter) and the downstream (final) filter must have minimum ASHRAE dust-spot efficiencies of 25S arid 85%, respectively [3, y]•

WOODS AND KRAFTHEFER ON OCCUPIED SPACES FILTAATION 195 While common practice indicates that filtration is used for con- tamination control in only special cases, recognition of its general application for ventilation control has existed for more than a de- cade. In 197', the ASHRAE ventilation standard defined ventilation air as: "that portion of supply air which comes from outside (out- doors) plus any recirculated air that has been treated to maintain the desired quality of air within a designated space" [5]. In that stand- ard, minimum and recommended ventilation rates of acceptably clean outdoor air were specified for numerous occupied spaces. That stand- ard also allowed recirculation if adequate temperature and filtration control were employed'. For the latter case, the outdoor air require- ments could~be reduced to 33% of the tabulated values if particulate filters were employed, and to 15% of the values if adsorption or other gas removal equipment were employed. However, two constraints were imposed: 1. the outdoor air quantity could not be less than 2.5 1/s (i.e., 5 cubic feet per minute) per person; and 2. the maximum allowable concentrations of contaminants in the supply air to the occupied space could not exceed the specified values in section 3 of that standard. The two constraints imposed in those criteria were seldom fol- lowed, and as a result, the standard'was of ten misinterpreted during the energy shortages of the 1970s. In 1981, the revised standard cl arified the recrrculation criteria by providi ng an eq uation w hich relates a selected filter efficiency to a corresponding recirculation rate to provide indoor air quality equivaI ent to that expected'by ven- tilation with 100% outd'oor air [1]: Vo - Vm e where: l. _ recirculation rate (cfm or 1/s per person). Outdoor ai r rate speci f'ied i n Tabl e 3 of' ASIIRAR Standard 62-1981 (cfm or I/s per person). [1] The minimum rate of outdoor air that can be used in the recirculated air ventilating system to provide acceptable indoor air quality, but never less than 2.5 1/'s (5 cfm) per person. Efficiency of the contaminant removal (air cleaner) device. The efficiency should be derived'f rom the most relevant parameters of the contaminant involved (e.g., ppm, or Ing/m3). This revision addresse6 two short-comings in the previous stand- ard: 1) it defined the air cleaner efficiency, quantitatively; an62) i t speci f ied a reci rcul atiom ai r fl ow rate as a functi on of the f il ter efficiency. In the previous standard, as long as any particulate f il ter were i ncor porated i nto the sy stem, the amount of outdoor ai r could be reduced and no ccmpensation in recirculation was specified.

196 FLUID FILTRATION: GAS i Thus, common practice was to reduce the total supply air flow rate to that only required for thermal loads. As energy conservation efforts were increased, lighting and envelope loads were decreased and'vari- able air volune systems became popular. As a result, less supply air was needed for thermal loads and complaints about poor air quality began to increase. Application of Eq 1 requires the recirculation rate to compensate- for the reduction in outdoor air, as well as the air cleaner efficiency. Moreover, it requires the total air supply rate, Vs (cfm or 1/s per person), to be the sun of the recirculation and minimun outdoor air flow rates: llS = Vm + Vr (21 For exampl e, ASH RAE 62-1981 speci f ies the req ui red' amount of out- door air for "meeting and waiting spaces" in offices as 17.5 1/s (35 cfm) per person if smoking is allowed'in the space. If it is desired' to reduce the outdoor ai r fl ew ra te to the mi nimum ot' 3.5 1/s (7 cfm) per person and to compensate for the smoking by installing air clean- ers, the required recirculation rate for a system with an air cleaner efficiency of 100% for "tobacco snoke" ( i. e. , for particulates, gases, and vapors) would be 17.5 - 3.5 = 14 1/s (35 - 7= 28' efm) per person. Thus, the total suppl y ai r fl ow rate woul d be 3.5 + 14 = 17.5 1/ s(7 + 28 = 35 cfrn) per persori, the sane as with 100% outduor air, but only 20% of the originally required outdoor air worrld require thermal treatment. For a more realistic air cleaner efficiency of 50%, the required recirculation rate would' be 14r'0.5 - 28 L/s (28/0.5 = 56 cfm) per person, and the required total supply rate wou13 be 28 + 3.5 = 31.5 1/s (56 + 7= 63 cfm) per person. In this case, the thermal load for the minimua outdoor air would be the sane as for the 100% efficient air cleaner, but the fan power and space requirements for the additional recirculation and total air supply rates would' be significantly larger. Thus, Equations 1 and 2 provide criteria for evaluating ventilation options which are expected to provide equivalent indoor air quality. AIR QUALITY CCNTROL Equations 1 and 2 do not directly address the quality of the indoor air. Rather, they provide an indirect procedure to achieve the same indoor air quality with some recirculation air as would be expected by ventilating with 100% outdoor air, but the value of that quality remains unspecified. In ASHRAE Standar662-1981, the indirect method is known as the "Ventilation Rate Procedure". Another method is also specified in that standard as the "Indoor Air Quality Procedure." The latter method specifies objective and'subjective criteria with which to evaluate the environment, but does not specify the means to achieve the required control. The Ventilation Rate Procedure may be considered a prescriptive standard, and the Air Quality Procedure may be considered a performance standard. These two procedures may be rel ated by consideri ng a simpl e one compartment model, Fig. 2. The concentration within this compartment is a function of the rate of contaminant generation within the

WOODS AND KRAFTHEFER ON OCCUPIED SPACES FILTRA r ION 197 I I occupied space, the rate of dilution with outdoor air, the contaminant concentration of the air transported into the occupied space, and the rate of removal of the contaminant by the air cleaning system. Conceptually, this Figure indicates that three methods exist to control the concentration of a contaminant within the occupied space: o suppression of the generation rate, or source control; o reduction of indoor air contaminant concentration by outdoor air exchange, or dilution control; and o removal of indoor air contaminant concentration by air cleaners, or removal control. ' BLOVIER V~ Co N NET DILUTION Vo, GENERATION. CI V. RATE 4 RATE f ~ c. ICa' 11,e) Ca L_j CI' C., OCCU~IEDSPACE ' I AIRCLEANER. E • V't C. REMOVAL RATE In steady-state within a well-mixed space, a relationship among these three control methods may be expressed as: N F < Figure 2. One compartment, unif ormly mixed, steady-state model for indoor ai r quality. [3) AC = where: ! C = Ci; the difference between the uniformly mixed indoor air concentration, C•, and the outdoor air concentration, Co (tlg/r3) . N= Q- S= the net generation rate of,the contaminant in the occupied space, where Q is the source strength ( i. e. , emission rate) and S is the sink strength (i.e., settling or sorption rate), (tig/hr) . E = VreCu = the removal, rate of a contaminant in the air cleaner (lig/fir); where Vr is th recirculation rate through the air cleaner (m~/hr), e is the efficiency of the air cleaner rated in terms of the contami nant removed ( i. e. , s= 1 - Cd/CU) , Cu is the contaminant concentration upstream of the air cleaner (i.e., Cu = Ci for a well-mixed system), and Cd is the contaminant concen- tration downstream of the ai:r cleaner. ou door air fl ow rate f or dilution control (m~/hr).

198 FLUID FILTRATION: GAS In this model, the diiution rate, Vo, represents inf il tration, natural ventilation, or mechanical ventilation with outdoor air. The removal rate, E, represents the rate at which ttre contaminant is ac- cunulated in the air cleaner. For the simple case in this model, the removal rate ( i. e. , E= VrECu could be achieved by fan-fil ter mod- ules now aviilable as c:rnsunrzr products, or by filtered recirculated air commonly used in residential forced air systems. Note that the removal rate is a function of three factors. Thus, for a given up- strean concentration, the sane rernoval rate can be expected fran two devices; one which has an air cleaner efficiency of' one-half the other if its recirculation rate is twice the other's. Moreover, as the in- door concentration decreases ( i. e. , Cl, and Cu approach zero) , the removal rate of a device with a fixed product of IrE decreases. For the case where the air cleaner is located in the mixed air, Fig. 1, the rel ationship among the three control methods may still be expressed by Eq 3, but the removal rate must be defined as E_(VoCo + VrCr)e. The advantage of this configuration is that contaminated outdoor air is also treated by the air cleaner before it enters the occupied space. This feature is especially important when it is necessary to control the indoor air concentration belaw that of the outdoor air. Although Eq 3 was derived fran a simple model, it serves to ident- ify sane basic control strategies and' their limitations t'or indoor air qual,ity control!: o If removal controli is not employed~ (i.e., E= 0), the indoor concentration will exceed,the outdoor concentration urrless the source is removed ( i. e. , N= 0) or an inf ini te dilutiurn rate is provided (i.e., Vo = -). o If the outdoor concentration is to be controlled belar that of' the outdoor air and the dilution rate is finite,.the removal rate must exceed the rret gener•atiore rate (i.e., E) N)l o Outdoor air required for dilution contrul may be reduced with- out affectiing AC, if' the difference betweera the net eerierati<rn rate and the r•u,roval rate is corre::{,und'int;ly rcducecJ. o To achieve an acceptable ,~C econanically, a cuutiirred strateCy of sourcu, r(srroval, arid dilution contrci probably will t,e re- quired. ACCEPTABILITY CRITERIA Only source control' canieliminate occupant exposure to a cunl.ar.,r- nant; dilution and removal cor,trol require mixture cf the currtan.irurnt within the occupiled space before their mechaudisms N.:cume roefrtl. Thrr~, both of these control methods expose occupants lo ttre cuntansiro rits, arid stwrui d only be used ii' some 1 evel of exE,osur-e i s"acce{,taLl e". Air quality may be defirred, getierally, as "the nature of air lhatt affects your health arid well-being". In Uiisdet iniliun, the Wvr1cI Health Ur•ganization's concept of healthi is imt,lied: "He«lU1, is a: :.tatc of compiete physical, mental and social weil-beira;, and! nut rucrtly the

WOODS AND KRAFTHEFER ON OCCUPIED SPACES FILTRATION 199 absence of disease or infirmity" [6]. These definitions offer criteria for evaluating beneficial, as well as deleterious effects of indoor environments. However, use of these definitions to ascertain accept- able indoor air quality requires both objective and subjective cri- teria. Objective criteria may be expressed as quantitative values of environnental stress which result in measurable physiological or psy- chological (i.e., behavioral) strains on the occupants. Thus, these criteria may be used to ascertain compliance with environmental con- ditions that should not cause measurable disease, disability, or dysfunction. Subjective criteria may be expressed in terms of af- fective responses of the occupants (e.g., comfort, annoyance, dis- comfort). When stress results in a complete state of well-being, a comfortable (:i.e., healthy) strain may exist. The amount of deviation frcm these ideal conditions that can be accepted without discomfort or adverse health effects is d'ependent upon the occupant's abilities to adapt to the deviations. As the ability to adapt diminishes, sus- ceptibility to the adverse effects of the stress increases. Thus, susceptible populations within occupied spaces must be considered when "acceptable ranges" of environmental stressors are selected. Traditionally, ASIIRAE has defined'acceptable environments as those in whictr 80% of the occupants find satisfactory: o Acceptable air quality is defined as "air in which there are no known contaminants at harmful concentrations and with which a substantial majority (usually 80%) of the people exposed do not express dissatisfaction" (1]. o Acce table thermal environment is defined as "an envirorment in which at least 80% of the occupants would fin6thermaily ac- ceptable" [7]. For purposes of evaluating control strategies, a more technical def'inition of indoor air quality has been proposed which incorporates the above concepts [8]: The quality of the air iman enclosed space is an indicator of haw we1L the air satisfies three criteria: o Thermal conditions of the air must be adequate to provide ther- mal acceptability for the occupants as defined by ASHRAE Stand- ard 55-1981. o The concentrations of oxygen and carbon dioxide must be within acceptable ranges to allow normal functioning of the respir- atory system. o The concentration of gases, vapors, and particulates should be below lievels that can have deleterious effects, or that can be perceived as objectionable by the occupants. This definition addresses three important factors needed to achieve acceptable indoor air quality control. First, it recognizes that ai r quali ty cc:~ntrol should be considered to be integral wi th con- trol of We thermal environment. While some aspects of source control can be achieved without reliance on thermal control (e.g., product substitution)~, suppression of tlie net generation rate by contairment

200' FLUID FILTRATION: GAS or isolation requires interface with the thennodynamic state of the air in the occupied space. Moreover, neither dilution nor removal control can be achieved without control of air movement, a thenno- dynamic process. Second, it suggests that responses to indoor air quality can be quantified in tenns of a subjective scale, not unlike those for predicted mean vote (PMV), percent people dissatisfied (PPD) or standard effective temperature (SEM [9, 10]. Third, and maybe most important, it implies that simultaneous control of the three methods is required, if satisfactory responses are to be achieved. IMPROUED CONTROL STRATE1jIES A classical conflict has developed between concepts of energy con- servation and environmental acceptability. During the energy crises of the last. decade, ventilation systems were de-activated, building envelopes were "tightened", and temperatures and relative humidities were allowed to decrease in winter and increase in summer. As a re- sult, environmental quality was degraded, sometimes to the extend that the health of occupants was jeopardized. Terms tc describe these con- ditions are now in literature, such as "Sick Building Syndrome", "Tight Building Syndrome", and "Building Related Illness" [11]. Con- versely, data also indicate that this conflict need not exist. Rather, if environmental control i's approached intelligently, energy efficient operation will result (12, 133. /A4 Figure 3. System schematic for micro- and mini-environmental, closed-loop control for thermal and air quality acceptability.

WOODS AND KRAFTHEFER ON OCCUPIED SPACES FILTRATION 201 A composite of these control strategies is shown schematically in Fig. 3. In this Figure, the region of primary concern within the room (i.e., micro-envi'ronment is shown as Compartment 1, and the remainder of the robm (i.e., mini-environment) is shown as Compartment 2. The relationship between the mini- and micro-environments is represented by a filtered room-air supply (1 - E 1)m2, and a room-coupling co- effi'cient, a, a factor which is similar to the concept of "Venti- lation Efficiency" (12, 14). Roan Ventilation Control The thermal and air quality control interaction between the HVAC system and the mini-environment can be expressed in terms of the fol- lowing steady-state equations: o The mass balance of the contaminant in the mini-environment: msxs + mixo + Nx2 = (ms + mi)x2 (4) where: m = mass air flow rate (gdry air/hr) C Nx2 = the generation rate of the contaminant i'n the mini- env i ronment. x = mass concentration of contaminant (Ngcont./gdry air) o The fraction of mini-environmental supply air flow to system air flow: z = ms/mm [5] o'Ihe mass balance of the contaminant in the tN AC system, down- stream fran the air cleaner:' %xm(1 - E2) + Ns = rt4nxs where Ns is the generation rate of the contaminant in the IIVAC system, downstream fran the air cleaner which has an efficiency of L2. (ugcont./hr). [6] o Itie mass balance of the air in the mixed air system: mo + mr = mm 171 o The mass balance of the contaminant in the mixed air system: moxo + mrxr = R4nxm [8l o And the energy balance of the mixed air system in terms of specific enthalpy, h (J/gdry air): rtb1k) + mrli~ = mnh4n• [9]

202 FLUID FILTRATION: GAS In tfiese equations, the subscripts represent: i= infiltration and natural ventilation rates into the mini- environment (i.e., psychrometric condition 5 in Fig. 3). m.= mixed air in HVAC system (i.e., psychrometric condition 6 in Fig. 3). o- outdoor ai r( i. e. , psychrometric condi tion 5 in Fi g. 3). r= recirculated air into"fNAC system ( i. e., psychrometric condi- tion 4 in Fig. 3). s- supply air to mini-environment ( i. e. , psychrometric condition 8 in Fig. 3). Note that it is necessary to express these balances in terms of mass flow rates of "dry air" rather than volumetric rates, as iso- thermal conditions can no longer be assumed'and changes in air den- si'ties must be considered due to the psychrometric processes. Equations 4 through 9 may be combined to provide an expression for the "Room Acceptability Ratio," K2, defined as the ratio of roan air to outdoor air contaminant concentrations: x2 (H + (1 - H)x /x )(1 - E ) + M + p_ fC2= = r o 2 2 [10] x 1 + M 0 where: H= mo/m~ _( hr - h~)/( hr - ho) [11] M = mi/ zmm [12] Q = W zrt x (13) 2 o M where: N=zNs+ Nx2 If the contaminant concentration in the mini-envirorment is identical to that in the recirculated' air (i.e., x2 = xr), Eq 10 simplifies to: K2_ x2 - H(1 - e2 ) + M + Q2 [14'] xo H(1 - e2) + M+12 Functional relationships from Fq 14 are shown in Figures 4-7 be- tween the "Roan Acceptabili'ty Ratio," K2 ( i. e., x2/xo), and the "Room Contamination Factor," Q2, with the "Air Cleaner Efficiency," e-2, the "Passive to Active Air Exchange Ration " M, and the "Outdoor Air Ratio," H, as parameters. In each of Ehese Figures, four sets of graphs are presented for air cleaner efficiencies at 0, 0.2, 0.5, 0.9, at each of four outdoor air ratios of 0.2, 0.5, 0.75, and 1.0. Also shown in each set of graphs is the reference condition of =2 = 1.0,