The impact of sorption on perceived indoor air quality

The impact of sorption on perceived indoor air quality

Introduction

A variety of studies have demonstrated that buildingmaterials with large surface areas, in addition to beingmeaningful emission sources of organic compounds,influence the dynamic variation of organic concentra-tions through sorptive and desorptive processes – the�sink effect� (An et al., 1999; Berglund et al., 1988;Borazzo et al., 1990; Colombo et al., 1993; Jørgensenet al., 1993; Kephalopoulos et al., 1996; Seifert andSchmahl, 1987; Singer et al., 2003, 2004; Tirkkonenand Saarela, 1997; Won et al., 2000, 2001). Althoughthe uptake of organics (sorption) will lower their peak

concentrations, the subsequent slow re-emission willprolong their presence in indoor air. In this way,temporary pollution events (painting, cleaning, smo-king, etc.) can continue to impact a room well after thepollution event has stopped (Nazaroff and Weschler,2004; Nielsen, 1987; Salthammer, 1999; Schlitt andKnoppel, 1997; Singer et al., 2004; Van Loy et al.,2001). Even building materials and furnishings that aredesigned to have low emissions can play a role inpolluting the indoor environment through sorption andsubsequent desorption of pollutants.To better determine, the impact of sorption on

indoor air quality, researchers have studied this

Abstract Sensory evaluations were used to investigate the impact of sorptionprocesses on indoor air quality. Experiments were carried out in four similar,adjacent, unfurnished offices. Samples of carpet, linoleum, painted gypsumboard, and Semia (a specially designed high-sorbing fabric) were tested indi-vidually and in combination. Additionally, to investigate the interaction betweenthe pollutants emitted from the building materials and the test room surfacesthemselves, air streams polluted by two different building materials were ventedinto an empty test office. Each experiment lasted for either 1 week (adsorptionstage only) or 10 days (adsorption and desorption stages). Untrained panelsassessed the air quality at specified times after moving the materials into or outof the rooms. The results showed that, in comparison with air in a room withcarpet or linoleum alone, the presence of painted gypsum board improved theperceived air quality. This improvement persisted throughout the 168 h of the�adsorption stage� of the experiments. A mass balance model was used toquantify the improvement. Calculated results indicate that, for the conditionsused in these experiments, pollutant removal via sorption to the relatively inertoffice surfaces was equivalent to an extra 0.4 air change per hour (ACH) ofventilation air, while sorption to painted gypsum board surfaces was equivalentto an extra 1–7 ACH of ventilation air. In the case of Semia, sorption wasequivalent to an extra 16 ACH of ventilation air. During the �desorption stage� ofthe experiments, after carpet or linoleum were taken out of a room, approxi-mately 3 days were required before the air in the test office, ventilated at 0.8ACH, was judged to be free of the sorbed pollutants.

W. Sakr1,3, C. J. Weschler1,2,P. O. Fanger11International Center for Indoor Environment andEnergy, Department of Mechanical Engineering,Technical University of Denmark, Lyngby, Denmark,2Environmental and Occupational Health SciencesInstitute, UMDNJ/Robert Wood Johnson MedicalSchool and Rutgers University, Piscataway, NJ, USA,3Department of Building, Civil and EnvironmentalEngineering, Concordia University, Montreal, Quebec,Canada

Key words: desorption; olf; sensory evaluation; Sinkeffect; surface removal.

W. SakrCornell UniversityDepartment of Design and Environmental AnalysisIthaca-NY14853-4401USATel.: 607 227 5488Fax: 607 255 0305e-mail: [email protected]

Received for review 6 June 2005. Accepted forpublication 17 October 2005.� Indoor Air (2006)

Practical ImplicationsVentilation rates in non-industrial buildings are based largely on sensory pollution sources and a desired level ofperceived air quality. This study documents that sorptive materials in a room influence the perceived air quality andshould be considered when evaluating ventilation requirements. Indeed, it may be possible to deliberately use sorp-tion/desorption to improve indoor air in a manner analogous to the way thermal storage/release is currently used inbuildings as a means of conserving energy.

phenomenon using a number of different approaches.Based on data obtained from static or dynamicchamber experiments, they have developed modelsthat address sorption of chemical compounds to avariety of indoor materials. Most of this research hasbeen carried out with single organic compounds andsingle materials (Kirchner et al., 1995; Tichenor et al.,1991). Only a few studies have investigated sorption ofmixtures of organic compounds onto single materialsor single organic compounds onto combinations ofmaterials (Jørgensen and Bjørseth, 1999; Singer et al.,2003, 2004; Wouda et al., 1997). In most of thesestudies, samples of indoor materials were exposed to anartificially polluted atmosphere (one or several organiccompounds) with concentrations usually much higherthan those found in actual indoor environments. Anumber of studies also have been performed tocompare measured concentrations in field experimentswith calculated concentrations based on data fromsmall chamber experiments (Dokka et al., 1999; Sparkset al., 1999; Zellweger et al., 1995). The predictedconcentrations (based on chamber experiments) tendto be considerably higher than those measured inactual indoor environments.Despite the development of different sorption models

and improvements achieved during the last few years inunderstanding sorption phenomena, there is still nomethod to easily and reliably estimate the sorptionproperties of different materials and the impact ofsorption on indoor air quality in actual indoor envi-ronments. In real life, where hundreds of organiccompounds co-exist with different kinds of surfacematerials, the airborne concentrations of pollutantsthat affect human senses are often several orders ofmagnitude lower than the concentrations that havebeen used to investigate sorption processes in laborat-ory experiments.The aim of the present study was to investigate the

impact of sorption on perceived indoor air quality. Thetool employed was a human panel that assessed the airquality under a variety of conditions. The experimentswere performed in real rooms where the pollutionsources and sinks were actual building materialscommonly found in indoor environments. Sensorymeasurements take advantage of the ability of humansto detect low concentrations of certain indoor pollut-ants and address the net effect of a complex chemicalmixture on human perceptions. For a subset ofchemicals found indoors, humans are more sensitivethan the analytical chemical techniques that are rou-tinely used in indoor air investigations. Furthermore,sensory measurements sidestep the analytical difficul-ties associated with simultaneously examining thesorption of multiple chemicals to multiple surfaces.Another important point is that the present ventilationguideline and standards (CEN 1752 and ASHRAE62.1) are largely based on perceived air quality.

Therefore, it is valuable to know how sorption influ-ences perceived air quality and the consequent changesin ventilation rates required to achieve acceptableindoor air quality.

Methods

Experimental plan

The experiments were performed in four similar,adjacent, unfurnished offices located at the Interna-tional Centre for Indoor Environment and Energy atthe Technical University of Denmark. The air waspolluted either by individual building materials or acombination of two materials. In combination, one ofthe materials could be considered to be primarily a�source� while the other could be considered to beprimarily a �sink�. The pollution source varied from oneoffice to another. Sensory evaluations were performedat specified times after moving the materials into or outof the rooms depending on the design of each experi-ment (see Table 1 for the design and conditions of theexperiments). For Experiment 3, two offices containedonly one material, another had a combination of twomaterials and the remaining office received a mixture ofthe exhaust air from the two offices with a singlematerial each (see Figure 1).

Facilities

Each office had a volume of 40 m3, a floor area of13 m2 covered with low polluting polyolefin, paintedbrick walls, ceiling tiles made of compressed mineralwool, and one glass window on the external wall.For Experiment 3, two holes in the two oppositewalls were equipped with two axial fans and dampersto allow controlled mixing of the exhaust air fromthe two adjacent rooms. The supply air to this room(mixing room) was solely from the two adjacentrooms (see Figure 1). The outdoor air was suppliedto the other three rooms by an axial fan and adamper mounted in each room’s window. Theoutdoor air was neither filtered nor conditioned. Inthe rooms adjacent to the mixing room, the gapsunder the doors were sealed so no air could escapeto the corridor. The air from the other two officeswas exhausted into the corridor via the gap underthe door. For Experiments 1, 2a and 2b, there wasno �mixing room� and the supply air to all rooms wasfrom the outdoors. Small fans were distributed in thefour offices to ensure that the air was well mixed.Each office had a screen, which served to hide thetest materials from the panel’s view. For Experi-ments 2a, 2b and 3, electric heaters and ultrasonichumidifiers controlled temperature and relativehumidity, which were measured during all assess-ments.


Building materials

In these experiments, samples of carpet and linoleumwere used as pollution sources, while samples ofpainted gypsum board and Semia were used as�sorbents�. In actuality, each material functioned asboth a source and a sink. However, carpet andlinoleum were much stronger pollution sources thanpainted gypsum board, the room surfaces or Semia. Inaddition the area, an important factor in sorption, waslarger for the room surfaces and painted gypsum boardthan for linoleum and carpet. The carpet (14 m2) was atufted boucle carpet with 100% polyamide fibers andlatex backing. It was approximately 24 years old andhad been taken from an office building. The linoleum(14 m2) was 5 years old. The gypsum board (23 m2)was painted (on both sides) 1 year before the start ofthe experiments. Semia is an absorbing fabric consist-

ing of fine activated charcoals particles embeddedbetween polyester-filament non-woven layers. TheSemia (24.5 m2) was delivered 3 months before theexperiments took place and was kept in its originalpackage until used. Semia sheets have a thickness of0.48 mm and come in rolls 1.05 m wide. The carpetand linoleum samples were stapled back to back toreduce emission from the backside and were hung onstainless steel racks when placed behind the screens inthe rooms. The surface areas of the samples werechosen to match the area of the floor (in the case ofcarpet) or the walls (in the case of painted gypsumboard) in the test rooms.

Sensory panel

Untrained panels, composed of different numbersof recruited students (18 for Experiment 1, 31 for

Table 1 Summary of the experiments

Test materials Sensory assessments Experimental conditions Participating subjects

Experiment 1Part 1 Room 1: carpet (14 m2)

Room 2: painted gypsumboard (23 m2)

Rooms 3: carpet and paintedgypsum board (14 m2 + 23 m2)

Assessments performed for the emptyrooms then 4, 24, 48, 72, 96,and 168 hafter the materials were placed in therooms

Ventilation rate:Room 1: Q ¼ 8.8 l/s (outdoor air)Room 2: Q ¼ 8.8 l/s (outdoor air)Room 3: Q ¼ 8.8 l/s (outdoor air)AverageT ¼ 25 € 0.6�CRH ¼ 50 € 3%

18 subjects61% male11% smokersAverage age € s.d.:30.6 € 8.1 (years)

Part 2 Room 1: emptyRoom 3: painted gypsumboard (23 m2)

Assessments performed for room 1 androom after 4, 24 and 48 h of taking thecarpet out of the rooms

AverageT ¼ 22.6 € 0.4�CRH ¼ 50 € 3%

Experiment 2aPart 1 Room 2: linoleum (14 m2)

Room 2: painted gypsumboard* (23 m2)

Room 3: linoleum andpainted gypsum board

(14 m2 + 23 m2)

Assessments performed for the emptyrooms then after 4, 24, 48, 72, 96, and168 h of placing the material in the rooms

Ventilation rate:Room 1: Q ¼ 8.8 l/s (outdoor air)Room 2: Q ¼ 8.8 l/s (outdoor air)Room 3: Q ¼ 8.8 l/s (outdoor air)


Part 2 Room 2: emptyRoom 3: painted gypsumboard (23 m2)

Assessments performed for room 2 androom 3 after 4, 24, 48 and 72 h of takingthe linoleum out of the rooms

Preset T and RH (for both parts):T ¼ 22�CRH ¼ 40%

Experiment 2bPart 1 Room 2: linoleum (14 m2)

Room 1: Semia (24.5 m2)Room 4: linoleum + Semia(14 m2 + 24.5 m2)

Assessments performed for the emptyrooms then after 4, 24, 48,72, 96, and168 h of placing the material in the rooms

Ventilation rate:Room 1: Q ¼ 8.8 l/s (outdoor air)Room 2: Q ¼ 8.8 l/s (outdoor air)Room 3: Q ¼ 8.8 l/s (outdoor air)


Part 2 Room 2: emptyRoom 4: Semia (24.5 m2)

Assessments performed for room 2 androom 4 after 4, 24, 48 and 72 h of takingthe linoleum out of the room

Preset T and RH (for both parts):T ¼ 22�CRH ¼ 40%

Experiment 3Room 1: carpet (14 m2)Room 2: exhaust air fromrooms 1 and 3

Rooms 3: painted gypsumboard (23 m2)

Rooms 4: carpet and paintedgypsum board (14 m2 + 23 m2)

Assessments performed for the emptyrooms then after 4, 8, 24,48, 72, and144 h of placing the materials in therooms

Ventilation rate:Room 1: Q1 ¼ 4.4 l/s (outdoor air)Room 2: Q ¼ 8.8 l/s ¼ 2 Q2 (Q2 ¼ 4.4 l/sfrom rooms 1 and 3)

Room 3: Q1 ¼ 4.4 l/s (outdoor air)Room 4: Q ¼ 8.8 l/s (outdoor air)Preset T and RH: T ¼ 22�C, RH ¼ 35%


*The evaluations of painted gypsum board in room 2 were performed 10 days before the start of Experiment 2a and 2b in Rooms 1–4. The assessments were performed by the same panelunder the same environmental conditions as Experiments 2a and 2b.

Sakr et al.

100

Experiments 2a and 2b and 34 for Experiment 3) wereused. See Table 1 for information concerning the panelfor each experiment. Before the start of the experi-ments, the panel was carefully instructed to assess theacceptability of the air in the test rooms by marking anacceptability scale (Figure 2). This scale has beenslightly modified from that of Gunnarsen and Fanger(1992) by splitting �just acceptable� and �just unaccept-able� at the center of the scale to avoid a neutral vote.In subsequent data analysis, numbers were assigned tothe marks as indicated in Figure 2. Subjects were toldto focus on their initial perception and not to commu-nicate with one another regarding the air qualityduring the assessments. They were asked not to eatspicy food or garlic on the day of the experiment or theday before, and not to use strong deodorant orperfume the day of the experiment.

Procedure

The two polluting materials, carpet and linoleum, andthe two sorbents, painted gypsum board and Semia,were evaluated individually and in paired combinationsof a polluting material and a sorbent. Experiment 3,designed to study only the adsorption stage, lasted1 week; Experiments 1, 2a and 2b, designed to studyboth the adsorption and desorption stages, lasted10 days. Sensory assessments took place every dayexcept on weekends. Before each experiment the roomswere cleaned and ventilated for 48 h at a high airexchange rate. The air exchange rate was then adjustedusing tracer gas (SF6) decay measurements (Bruel andKjaer Multi-gas Monitor 1302 and Multipoint Samplerand Doser 1303). See Table 1 for the ventilation ratefor each room and each experiment. The air exchangerate was chosen to be moderate, but realistic, to makeit easier to detect the impact of sorption.Before the start of the experiments the material

samples sat for 48 h in separate, rooms similar to theones used in the experiments; these rooms were highlyventilated with outdoor air to facilitate desorption ofpreviously sorbed chemicals. Before the materials wereintroduced to the rooms, the sensory panel assessed theair quality in the empty rooms. Sensory evaluationswere conducted following a pre-established time sched-ule (see Table 1). Sensory assessments were performedaccording to a randomized plan that was different fordifferent subjects and different days of the experiment.The subjects were unaware of the experimental condi-tions in the rooms where they assessed the air quality.Only one participant at a time was allowed to enter aroom for sensory assessments. The panel was seated inan office next to the offices used in these experiments.This office was naturally ventilated and the windowswere kept open to insure the freshness of its air. After

Outdoor airQ1 = 4.4 L/s Q1 = 4.4 L/s

boards

carpet

Outdoor air

14 3 2

Outdoor airQ

=

8.8 L/s

Two materials:

painted gypsum boards with carpet

One material:

painted gypsum

One material:

Mixing room

Screen

Q2 =

4.4 L/s Q2

=4.4 L/s

Fig. 1 Set-up for Experiment 3

Imagine that during your daily work you are exposed to this air.

How do you assess the air quality?

Name: ……………………………………… Room :

Date: ………………………………………

Name: ……………………………………… Room :

Date: ………………………………………

+1

Name: ……………………………………… Room :

Date: ………………………………………

Name: ……………………………………… Room :

Date: ………………………………………

0

-

Name: ……………………………………… Room :

Date: ………………………………………

Clearly acceptable

Name: ……………………………………… Room :

Date: ………………………………………

Notice the distinction betweenacceptable and unacceptable

Name: ……………………………………… Room :

Date: ………………………………………

Just acceptableJust unacceptable

Clearly unacceptable

Name: ……………………………………… Room :

Date: ………………………………………

0

-1

Fig. 2 Acceptability scale. Numbers were used in data analysisand they were not shown on the scale during the experiments


each assessment, the subject was asked to spend 2 minin this room before proceeding to the next assessment.

Statistical analyses

Data obtained from the sensory assessments wereanalyzed using the Wilcoxon Matched Pairs Test, anon-parametric test to compare two paired groups ofdata. To compare more than two groups Friedman’stwo-way analysis of variance was used (Siegel andCastellan, 1988).

Results

Experiment 1: Interaction between carpet and painted gypsum board

The first experiment was designed to examine theinfluence that sorption to painted gypsum boardsurfaces might have on indoor air polluted by a carpet.Three rooms were used in this experiment. Room 1contained 14 m2 of carpet, room 2 contained 23 m2 ofpainted gypsum board, and room 3 contained 14 m2 ofcarpet together with 23 m2 of painted gypsum board.Details of the experimental conditions are presented inTable 1. In the first part of this experiment assessmentsof the air quality in the three rooms were performed atfixed times up to 168 h after the materials were placedin the test rooms. From 24 to 168 h, the paintedgypsum board assessments averaged 0.05, the carpet/painted gypsum board assessments averaged )0.12 andthe carpet assessments averaged )0.33 (see Table 2).Figure 3 shows the results, in terms of acceptability,for the room containing carpet alone and the roomcontaining the combination of carpet and paintedgypsum board. Throughout the 168 h of evaluation theroom with carpet and painted gypsum board togetherwas perceived to have significantly better air qualitythan the room containing only the carpet sample.Presumably, when painted gypsum board is in theroom with the carpet samples, sorption to the paintedgypsum board is an additional sink for the carpetemissions. During the 168 h that assessments wereperformed in these three rooms, there was no apparentchange in the perceived air quality in the carpet/painted gypsum board room; the painted gypsumboard continued to be a strong sink for the carpetemissions throughout the period, showing no signs of�saturation�.After the evaluation at 168 h, the carpet samples

were removed from the two rooms in which they hadbeen present and three additional air quality evalua-tions were performed. These occurred 4, 24 and 48 hafter removing the carpet samples. Within 4 h ofremoving the carpet from the room that had containedonly carpet, there was a sharp improvement inperceived air quality (from )0.33 to +0.13). Incontrast, when carpet was removed from the room

that also contained painted gypsum board, the evalu-ation at 4 h was only slightly better (from )0.09 to0.00). Presumably, the carpet emissions that hadbeen sorbed by the painted gypsum board werenow desorbing, adversely affecting the air quality.The assessments 24 and 48 h after removing thecarpet showed continued improvement in both rooms(see Figure 3). At the time of the last assessment, 48 hafter removing the carpet, the room that had containedonly carpet reached an assessed air quality (0.28)almost equal to the average assessments of the emptyrooms (0.34), while the carpet/painted gypsum boardroom had reached an assessed air quality (0.07) equalto a room containing only painted gypsum board(0.05).

Experiment 2a: Interaction between linoleum and painted gypsumboard

This experiment was analogous to the previousexperiment, but linoleum was substituted for carpet.Room 2 contained 14 m2 of linoleum, and room 3contained 14 m2 of linoleum together with 23 m2 ofpainted gypsum board (see Table 1 for furtherdetails). From 24 to 168 h, the painted gypsumboard assessments averaged 0.06, the linoleum/pain-ted gypsum board assessments averaged )0.19 andthe linoleum assessments averaged )0.39 (seeTable 2). Figure 4 shows the results, in terms ofacceptability, for the room containing linoleum aloneand the room containing the combination of lino-leum and painted gypsum board. Throughout thefirst 168 h that assessments were performed in theserooms, the room with the combination of linoleumand painted gypsum board was perceived to havesignificantly better air quality than the room con-taining only linoleum. That is, the painted gypsumboard continued to be a strong sink for the linoleumemissions and, as was the case with carpet emissions,there was no indication that it was approaching�saturation�.After the evaluation at 168 h, the linoleum samples

were removed from the two rooms in which they hadbeen present and additional air quality evaluationswere performed 4, 24, 48 and 72 h later. Analogousto the previous experiment, the improvement waslarger for the room that had contained only linoleum()0.34 to )0.12 after 4 h) compared with the roomthat had also contained painted gypsum board()0.11 to )0.01). Presumably, linoleum emissionswere desorbing from the painted gypsum board thathad been exposed to linoleum. Desorption of pollut-ants from the painted gypsum board in combinationwith the room surfaces had a more deleteriouseffect on air quality than desorption from theroom surfaces alone. The assessments 24 and 48 hafter removing the linoleum showed continued

Sakr et al.

102

improvement. By 72 h the room that had containedonly linoleum reached an assessed air quality equalto its initial assessment as an empty room, while thelinoleum/painted gypsum board room had reachedan assessed air quality equal to a room containingonly painted gypsum board (see Figure 4).Note that the assessments made in Experiments 1

(carpet) and 2 (linoleum) cannot be directly comparedsince the temperature and relative humidity were lowerduring the linoleum experiment.

Experiment 2b: Linoleum and Semia

�Semia� is a product deliberately designed to sorbpollutants from the air; it consists of sheets containing

activated charcoal. At the same time that linoleum andlinoleum/painted gypsum board were evaluated, Semiaand linoleum/Semia were evaluated in adjacent lowpolluting offices. Room 1 contained 24.5 m2 of Semia,and room 4 contained 14 m2 of linoleum together with24.5 m2 of Semia (see Table 1 for further details).Figure 5 shows the results, in terms of acceptability,for the room containing Semia alone, the roomcontaining the combination of linoleum and Semia,and the room containing linoleum alone. The airquality in the room containing Semia alone wasequivalent to the air quality of the room when it wasempty (average acceptability of +0.38). The room withthe combination of linoleum and Semia was perceivedto have much better air quality (average acceptability

Table 2 Acceptability, perceived air quality and pollution load for the assessments performed during Experiments 1, 2a and 2b

Experiment 1

Acceptability € s.e. % Dissatisfied Perceived air quality (decipol) Pollution load (olf)

CarpetACH ¼ 0.8

PGBACH ¼ 0.8

Carpet and PGBACH ¼ 0.8 Carpet PGB

Carpet andPGB Carpet PGB

Carpet andPGB Carpet PGB

Carpet andPGB

Empty rooms 0.43 € 0.09 0.39 € 0.09 0.41 € 0.1 8 10 9 0.48 0.59 0.53 0.42 0.52 0.47Part 1. Averageassessment

)0.33 € 0.09 0.05 € 0.09 )0.12 € 0.1 83 39 61 18.2a 3.6a 8.9a 16.1a 3.2a 7.9a

Part 2. Carpet out Empty PGB Empty PGB Empty PGB Empty PGB4 (172) 0.13 € 0.16 0.00 € 0.14 30 46 2.0a 4.6a 1.8a 4.1a

24 (192 0.15 € 0.14 0.02 € 0.16 27 43 1.7a 4.1a 1.5a 3.6a

48 (216) 0.28 € 0.13 0.07 € 0.16 16 37 0.6a 3.0a 0.5a 2.6a

Experiment 2a


LinoleumACH ¼ 0.8

PGBb

ACH ¼ 0.8Linoleum andPGB ACH ¼ 0.8 Linoleum PGBb

Linoleumand PGB Linoleum PGBb

Linoleumand PGB Linoleum PGBb

Linoleumand PGB


)0.39 € 0.07 0.06 € 0.07 )0.19 € 0.07 87 38 69 20.2a 2.8a 11.9a 17.8a 2.5a 10.4a

Part 2. Linoleumout

Empty PGB Empty PGB Empty PGB Empty PGB

4 (172) )0.12 € 0.07 )0.01 € 0.08 61.1 47 8.3a 4.8a 7.3a 4.3a

24 (192) 0.13 € 0.09 0.014 € 0.07 29.6 44 1.6a 4.2a 1.4a 3.7a

48 (216) 0.13 € 0.09 0.08 € 0.07 29.6 35.4 1.6a 2.7a 1.4a 2.4a

72 (240) 0.34 € 0.08 0.09 € 0.07 12.2 34.2 0.0a 2.5a 0.0a 2.2a

Experiment 2b


LinoleumACH ¼ 0.8

SemiaACH ¼ 0.8

Linoleum andSemiaACH ¼ 0.8 Linoleum Semia

Linoleumand Semia Linoleum Semia

LinoleumandSemia Linoleum Semia

LinoleumandSemia


)0.39 € 0.07 0.38 € 0.08 0.16 € 0.08 87 10 26 20.2a 0.0a 1.4a 17.8a 0.0a 1.2a

Part 2. Linoleumout

Empty Semia Semia Empty Semia Semia Empty Semia Semia Empty Semia Semia

4 (172) )0.12 € 0.07 0.37 € 0.1 0.21 € 0.1 61.1 10.6 21.6 8.3a 0.1a 0.9a 7.3a 0.1a 0.8a

24 (192) 0.13 € 0.09 0.33 € 0.09 0.24 € 0.09 29.6 12.8 19.0 1.6a 0.2a 0.6a 1.4a 0.2a 0.6a

48 (216) 0.13 € 0.09 0.31 € 0.08 0.34 € 0.08 29.6 14 12.2 1.6a 0.3a 0.1a 1.4a 0.3a 0.1a

72 (240) 0.34 € 0.08 0.31 € 0.09 0.30 € 0.08 12.2 14 14.6 0.0a 0.3a 0.2a 0.0a 0.3a 0.2a

aValues for the empty room have been subtracted from the reported values.bThe evaluations of painted gypsum board in room 2 were performed 10 days before the start of Experiment 2a and 2b in rooms 1–4.The assessments were performed by the same panel under the same environmental conditions as Experiments 2a and 2b.


–1.0

–0.8

–0.6

–0.4

–0.2

0.0

0.2

0.4

0.6

0.8

1.0

0 24 48 72 96 120 144 168 192 216 240 264Time (h)

Acc

epta

bili

ty

LinoleumLinoleum & painted gypsum board Linoleum out of both

rooms

Average T = 22.3°CAverage RH = 40%

Empty room

P < 0.001P < 0.001

P < 0.01P < 0.001


Fig. 4 Mean assessments of acceptability for the air in the rooms that contained linoleum and a combination of linoleum and paintedgypsum board (Experiment 2a). Included are assessments that were made after the linoleum was removed

Carpet out ofboth rooms

Average T = 24.6 ºCAverage RH = 48 %

Empty room

Average T = 22.6 ºCAverage RH = 47 %

0 24 48 72 96 120 144 168 192 216 240Time (h)

–1.0

–0.8

–0.6

–0.4

–0.2

0.0

0.2

0.4

0.6

0.8

1.0

Acc

epta

bili

ty

Carpet

Carpet & painted gypsum board

P < 0.001 P < 0.05P < 0.05 P < 0.01 P < 0.001

Fig. 3 Mean assessments of acceptability for the air in the rooms that contained carpet and a combination of carpet and paintedgypsum board (Experiment 1). Included are assessments that were made after the carpet was removed

–1.0

–0.8

–0.6

–0.4

–0.2

0.0

0.2

0.4

0.6

0.8

1.0

0 24 48 72 96 120 144 168 192 216 240 264

Time (h)

Acc

epta

bili

ty

LinoleumSemiaLinoleum & Semia

Linoleum out of bothrooms


Empty room

P < 0.0001 for all assessments

Average T = 22.1°C

Average RH = 40%

Fig. 5 Mean assessments of acceptability for the air in the rooms that contained linoleum, Semia and a combination of linoleum andSemia (Experiment 2b). Included are assessments that were made after the linoleum was removed

Sakr et al.

104

of +0.16) than the room containing only linoleum (asnoted earlier, average acceptability of )0.39) or theroom containing linoleum and painted gypsum board(as noted earlier, average acceptability of )0.19, seeFigure 4). Semia remained a strong sink for linoleumemissions throughout the first 168 h.Within 48 h of removing linoleum, the air quality in

what had been the �linoleum/Semia� room matched thatin the room containing Semia alone.

Experiment 3: Mixed exhausts

The third experiment was designed to better evaluatethe influence of sorption by room surfaces on the levelsof carpet and painted gypsum board emissions (seeDiscussion). Room 1 contained 14 m2 of carpet; room2 was empty; room 3 contained 23 m2 of paintedgypsum board; and room 4 contained 14 m2 of carpettogether with 23 m2 of painted gypsum board. Room 4was ventilated with outdoor air. In contrast, room 2received the exhaust air from rooms 1 and 3; this wasthe only source of supply air to room 2. Rooms 2 and 4were ventilated at the same rate (8.8 l/s or 0.8/h) asused in prior experiments. Rooms 1 and 3 wereventilated at half the rate [4.4 l/s or 0.4 air changeper hour (ACH)] used in prior experiments. The intentwas that room 2, receiving the exhaust air from rooms1 and 3, would be ventilated at the same rate as room 4,facilitating direct comparisons between rooms 2 and 4.Figure 6 shows the results, in terms of acceptability,for the room containing the mixed exhausts and theroom containing the combination of carpet and pain-ted gypsum board. For assessments performed at 24,48, 72 and 144 h, the room with the combination ofcarpet and painted gypsum board was perceived tohave significantly better air quality (average assessmentof +0.18) than the room containing the mixedexhausts (average assessments of +0.02). This suggeststhat in the room with both materials, painted gypsumboard was acting as a meaningful sink for carpet

emissions (and carpet may have been acting as a sinkfor painted gypsum board emissions). Throughout theperiod that assessments were performed there was noapparent change in the air quality of either of theserooms; neither the room surfaces nor the paintedgypsum board were close to �saturation�.The assessed air quality in the rooms containing

carpet alone or painted gypsum board alone cannot bedirectly compared with that of the room with mixedexhausts or the carpet/painted gypsum board roombecause of the lower ventilation rate in the former twocases compared with the latter two cases.

Discussion

The air quality assessments made in Experiments 1–3were used to find the �percent dissatisfied� (Gunnarsenand Fanger, 1992) and then converted to �decipols�(Fanger, 1988), which can be viewed as a measure ofconcentration (i.e. mass per unit volume). In turn, bytaking the ventilation rate into consideration, thedecipol values were converted to �olfs� (Fanger, 1989),which can be viewed as a measure of emission rate (i.e.mass per unit time). The following equations were usedto effect these conversions:

PD ¼ expð�0:18� 5:28:Acc:Þ1þ expð�0:18� 5:28:Acc:Þ

� �� 100 ð1Þ

where PD is the percent dissatisfied (%) and Acc themean acceptability vote.

PAQ ¼ 112½lnðPDÞ � 5:98��4 ð2Þ

where PAQ is the perceived indoor air quality (deci-pol).

Esp ¼ 0:1QðPAQ� PAQoÞ ð3Þ

where Esp is the sensory pollution load (olf); Q themeasured outdoor air supply rate to the room (l/s);PAQo the perceived outdoor air quality (decipol). Inthis study PAQo is the perceived air quality of theempty room (background)Tables 2 and 3 show the results of performing such

conversions on the sensory assessment data reported inFigures 3–6.In this analysis, we are assuming that the sensory

pollutants emitted from linoleum, carpet and paintedgypsum board have the same exposure–response rela-tionship as the bioeffluents that constitute the founda-tion of the olf-decipol concept. We recognize that thisis a less-than-perfect assumption (see Knudsen et al.,1997), but we judge that the present imperfectapproach still provides useful insights into sorptionprocesses.To evaluate the impact of sorption on perceived air

quality, the converted results from the aforementioned

–1.0

–0.8

–0.6

–0.4

–0.2

0.0

0.2

0.4

0.6

0.8

1.0

0 24 48 72 96 120 144 168Time (h)

Acc

epta

bili

ty

Carpet & painted gypsum board

Mixed exhausts


Empty room

Fig. 6 Mean assessments of acceptability for the air in the roomthat contained carpet and painted gypsum board together andfor the room where the exhaust air from the rooms with carpetand painted gypsum board were mixed (Experiment 3)


https://www.researchgate.net/publication/222917221_Introduction_of_the_OLF_and_the_DECIPOL_units_to_quantify_air_pollution_perceived_by_humans_indoors_and_outdoors?el=1_x_8&enrichId=rgreq-e872fe28-4ee4-4bb5-b45e-61a4dc8b5803&enrichSource=Y292ZXJQYWdlOzcyNzA4MDQ7QVM6OTc1NDAxNjI3MTk3NTFAMTQwMDI2Njc5NTQxNw==

https://www.researchgate.net/publication/236393754_The_new_comfort_equation_for_indoor_air_quality?el=1_x_8&enrichId=rgreq-e872fe28-4ee4-4bb5-b45e-61a4dc8b5803&enrichSource=Y292ZXJQYWdlOzcyNzA4MDQ7QVM6OTc1NDAxNjI3MTk3NTFAMTQwMDI2Njc5NTQxNw==

experiments were used in a simple mass balance model.Under steady-state conditions, the first-order rateconstants that describe the sorptive removal of carpetor linoleum emissions by the room surfaces or thematerial surfaces can be calculated from Equation 4:

PAQ ¼ ðEsp=VÞ=ðkVent þ kSurÞ ð4Þ

where V is the volume of test room (40 m3); kVent theremoval rate constant for ventilation (h)1); kSur theremoval rate constant for sorptive surfaces withinthe room, including room surfaces andmaterial surfaces(h)1). Depending on the materials within the room, kSuris the sum of removal rate constants for kRmSur, kCarp,kPGB, kLin, and kSemia where kRmSur is the removal rateconstant for sorption to room surfaces (h)1); kCarp theremoval rate constant for sorption to carpet surfaces(h)1); kPGB the removal rate constant for sorption topainted gypsum board surfaces (h)1); kLin the removalrate constant for sorption to linoleum surfaces (h)1);kSemia the removal rate constant for sorption to Semiasurfaces (h)1). The results are summarized in Table 4(note that Esp/V has units of decipol h)1 or olf m)3).The tabulated rate constants that describe sorptive

removal to different surfaces (i.e. kRmSur, kCarp, kPGB,

kLin and kSemia) can be viewed in terms of theequivalent amount of ventilation that would berequired to achieve the same pollutant removal. Forexample, in Table 4 for Experiment 1 the value of�kCarp + kPGB� is listed as 1.7 h)1. This indicates thatin this experiment sorptive removal of paintedgypsum board emissions by carpet and of carpetemissions by painted gypsum board surfaces im-proved the perceived air quality by the same amountthat would have been achieved with an extra 1.7 ACHof ventilation.In discussing Table 4, it is best to begin with

Experiment 3. This experiment was performed so thatthe contribution from sorption onto room surfacescould be separated from sorption onto the buildingmaterial surfaces. The average perceived air quality inthe room with carpet alone was 7.6 decipols larger thanwhen the room was empty, and the average perceivedair quality in the room with painted gypsum boardalone was 2.3 decipol larger than when the room wasempty. In the mixing room, which received air onlyfrom these two adjacent rooms, the average perceivedair quality was 3.3 decipol larger than before theexperiment began. This is smaller than predicted if

Table 3 Acceptability, perceived air quality and pollution load for the assessments performed during Experiment 3

Experiment 3


CarpetACH ¼ 0.4

PGBACH ¼ 0.4

Carpet andPGBACH ¼ 0.8

MixedexhaustACH ¼ 0.8 Carpet PGB

CarpetandPGB

Mixedexhaust Carpet PGB

CarpetandPGB

Mixedexhaust Carpet PGB

CarpetandPGB

Mixedexhaust

Empty rooms 0.27 € 0.06 0.27 € 0.06 0.32 € 0.07 0.25 € 0.06 16.7 16.7 13.3 18.2 1.1 1.1 0.85 1.25 1 1 0.7 1.1Averageassessment

)0.11 € 0.08 0.08 € 0.07 0.18 € 0.07 0.02 € 0.08 59.5 35.5 24.7 42.9 7.6a 2.3a 1.1a 3.3a 6.7a 1.9a 1.0a 3.0a

aValues for the empty room have been subtracted from the reported values.

Table 4 Perceived air quality (PAQ), as well as the volume-normalized emission rates (Esp/V) and removal rate constants for the sensory pollutants emitted by carpet, painted gypsumboard (PGB) and linoleum in Experiments 1 through 3

PAQ* decipol Esp/V decipol h)1 kVent h)1 kRmSur h

)1 kCarp + kPGB h)1 kLin + kPGB h

)1 kSemia + kLin h)1

Experiment 1 (24.6�C)Carpet 18.2 21.8 0.8 0.4PGB 3.6 4.25 0.8 0.4Carpet + PGB 8.9 26.1 0.8 0.4 1.7Experiment 2a (22.1�C)Linoleum 20.2 24.25 0.8 0.4PGB 2.8 3.37 0.8 0.4Linol. + PGB 11.9 27.63 0.8 0.4 1.1Experiment 2b (22.1�C)Semia 0 0 0.8 0.4Linol. + Semia 1.4 24.25 0.8 0.4 16.1Experiment 3 (22.3�C)Carpet 7.6 6.12 0.4 0.4PGB 2.3 1.87 0.4 0.4Carpet + PGB 1.0 8.0 0.8 0.4 6.7Mixed Exhaust 3.3 4.0 0.8 0.4

* Values for the empty room have been subtracted from the reported values.

Sakr et al.

106

there were no sorption on the room surfaces; withoutsorption to room surfaces the anticipated perceived airquality in the mixing room would be five decipol largerthan when the room was empty (i.e. 1/2 of 7.6decipol + 2.3 decipol) [recall that the ventilation ratein the mixing room (8.8 p/s) was twice as large as that(4.4 p/s) in the rooms with carpet alone or paintedgypsum board alone]. Hence, from Equation 4, thefirst order rate constant for pollutant removal bysorption onto the room surfaces, kRmSur, was 0.4 h)1,equivalent to an extra 0.4 ACH of ventilation (seeTable 4).Continuing the discussion of Experiment 3, if there

had been no sorption onto the surfaces of the carpetand painted gypsum board, the anticipated perceivedair quality in the room that contained both carpet andpainted gypsum board would have been 6.7 decipollarger than that in the empty test room. Instead theperceived air quality in the room with carpet andpainted gypsum board was 1 decipol larger than whenthe test room was empty. Again from Equation 4, thefirst order rate constant for pollutant removal bysorption onto the painted gypsum board and carpetsurfaces, kPGB + kCarp, was equivalent to an extra6.7 ACH of ventilation.The same approach can be used to evaluate the

results from Experiment 1 for the test room containingcarpet and painted gypsum board (recall that inExperiment 1 all three of the test rooms were ventilatedat 0.8 ACH). When a test room contained only carpet,the perceived air quality was 18.2 decipol larger thanwhen the test room was empty, and when a test roomcontained only painted gypsum board, the perceivedair quality was 3.6 decipol larger than when the testroom was empty. Hence, if a test room containedcarpet and painted gypsum board together and if therewere no sorption of the emitted pollutants to thematerial surfaces, we would expect the perceived airquality to be 21.8 decipol larger than when the testroom was empty. Instead, the perceived air quality was8.9 decipol larger than the empty test room. FromEquation 4, the first order rate constant for pollutantremoval by sorption onto the painted gypsum boardand carpet surfaces, kPGB + kCarp, was equivalent toan extra 1.7 ACH of ventilation.Using the results from Experiment 1, Figure 7 con-

trasts the pollution load for carpet when it was presentalone with the pollution load when painted gypsumboard was also present. The difference (8 olf) is a clearindication of the impact of sorption on perceived airquality.The same procedure used to evaluate the impact of

sorption in Experiments 1 and 3 was used to evaluatethe data from Experiments 2a and 2b where linoleumwas the major pollution source and the room surfaces,painted gypsum board and Semia were the sinks. Theaverage perceived air quality in the room that con-

tained only linoleum was 20.2 decipol larger than theempty room. When a test room contained linoleum andpainted gypsum board together, the perceived airquality dropped to a value 11.9 decipol larger thanthe empty test room. From Equation 4, removal ofpollutants by sorption was equivalent to an extra1.1 ACH of ventilation.Using the results from Experiment 2a, Figure 8

(linoleum, PGB and linoleum + PGB) contrasts thepollution loads for linoleum when it was present alonewith the pollution load when painted gypsum boardwas also present. The reduction in the linoleumpollution load (7.4 olf) indicates the impact of sorptionon perceived air quality.In a similar fashion, the rate constant for sorption

onto Semia surfaces was calculated using Equation 4and data from Experiment 2b (Table 2). The resultsindicate that sorption onto the Semia surfaces isequivalent to 16.1 h)1 extra ventilation (see Table 4).Figure 8 (linoleum, Semia and linoleum + Semia) andhighlights the impact of sorption on pollution loads in

Carpet PGB Carpet + PGB

Impact of sorption

0

2

4

6

8

10

12

14

16

18

Po

lluti

on

load

(o

lf)

Fig. 7 Results from Experiment 1: average pollution loads forcarpet and painted gypsum board when tested separately andtogether

0

2

4

6

8

10

12

14

16

18

20

Linoleum PGB Semia Linoleum + Semia Linoleum + PGB

Impact of sorption

Po

lluti

on

load

(o

lf)

Fig. 8 Results from Experiments 2a and 2b: average pollutionloads for linoleum, painted gypsum board and Semia whentested separately and together


Experiment 2b, linoleum and Semia. Sorption dramat-ically reduces the pollution load when linoleum andSemia are present together.In evaluating the different experiments, it is import-

ant to note that the sensory assessments measured inExperiments 1 and 3 cannot be directly compared,although the same building materials were used (carpetand painted gypsum board). The reasons for thisinclude: 1) Compared with Experiment 3, the higherenthalpy of the room air in Experiment 1 amplified theinfluence of sensory pollutants on perceived air quality(Fang et al., 1998). 2) Compared with Experiment 3,the higher temperature in Experiment 1 resulted inlarger emission rates from the test materials. 3)Compared with Experiment 3, the higher temperaturein Experiment 1 resulted in less sorption by equivalentsurfaces (Tichenor et al., 1991; Van der Wal et al.,1998). Consistent with these differences, the combina-tion of carpet and painted gypsum board is lessacceptable in Experiment 1 ()0.12) than in Experi-ment 3 (+0.18). Carpet alone, even though it wasevaluated at a higher air exchange rate in Experiment 1than in Experiment 3, was less acceptable in Experi-ment 1 ()0.33) than in Experiment 3 ()0.11). This isalso the case, but to a lesser extent, for painted gypsumboard alone.The emissions from carpet and linoleum are repre-

sentative of typical indoor air pollution sources. Thesorption of these emissions on gypsum board wasstriking. The perceived air quality in a room thatcontained gypsum board together with carpet wassignificantly better than in a room that contained onlycarpet. The same was true when linoleum was substi-tuted for carpet. Depending on conditions, theimprovement was comparable to that which might beachieved with an extra 1–7 ACH of ventilation air. Thepainted gypsum board continued to function as a sinkthroughout the 168 h that constituted the first part ofthese experiments, with no indication that it wasapproaching �saturation�. Painted gypsum board thathad shared a room with carpet or linoleum had a morenegative effect on perceived air quality than paintedgypsum board that had not been exposed to suchpollution sources. Approximately 2 days were requiredfor the painted gypsum board to desorb most of thesensory pollutants that it had sorb during the periodthat it shared a room with carpet or linoleum. Semia,consistent with its design intent, was very effective atsorbing the sensory emissions from linoleum; for theconditions tested, sorption to Semia was equivalent toan extra 16 ACH of ventilation.It is obvious that, under certain circumstances,

sorption can improve indoor air quality significantly,reducing the ventilation rate required to meet certainstandards. However, at some point, materials thatinitially sorb sensory pollutants will reach equilibriumwith their environment. At equilibrium the rate at

which sensory pollutants are sorbed by the materialmatches the rate at which they are desorbed by thematerial. In effect, the concentration of sensorypollutants in a room is the same as it would be if nosorbing materials were present. Further experimentsare necessary to determine how long it takes forpainted gypsum board or Semia to reach equilibriumwith their environment over a range of conditions.However, the fact that such an equilibrium was notreached in Experiments 1–3 after 168 h indicates thatsorption to painted gypsum board (and Semia)involves diffusion into the material as well as sorptiononto surface sites.This paper has focused on the impact of sorption

processes on perceived air quality. However, weacknowledge that other processes may be affectingperceived air quality in Experiments 1–3. These includesurface reactions between chemicals present in the carpetor linoleum and ozone in the ventilation air (Morrisonand Nazaroff, 2000, 2002; Weschler et al., 1992). Addi-tional experiments would be necessary to determine thecontribution of such processes to the observed effects.However, it is our judgment that ozone-initiated oxida-tion processes are of less importance than sorption inExperiments 1–3. This reflects the fact that outdoorozone levels during the period of the experiments wererelatively low (typically <40 ppb).

Conclusions

Existing ventilation guidelines and standards (ASH-RAE, 2004; CEN CR 1752, 1998) are largely based onperceived air quality. The present study has demon-strated that sensory evaluation is an effective tool forevaluating the impact of sorption on the acceptabilityof indoor air. The results show that sorption ofcommon indoor air pollutants onto surfaces such asgypsum board and Semia can markedly improve theperceived air quality, decreasing the ventilation raterequired to achieve a given degree of acceptability.It is recommended that a dynamic model be devel-

oped, taking into account sorptive and desorptivecharacteristics of room surfaces, that predicts theventilation rate required in a space in order to obtaina certain perceived air quality. Such a model wouldfacilitate the optimization of building ventilation dur-ing both occupied and non-occupied hours. For such apurpose, it will be essential to study sorptive anddesorptive characteristics of other types of surfacematerials over extended periods when exposed topollution from building materials as well as people(bioeffluents).

Acknowledgements

We thank Geo Clausen, ICEE for useful discussionsthroughout these experiments. W. Sakr thanks IRSST

Sakr et al.

108

(Institut de recherche Robert-Sauve en sante et ensecurite du travail) for the 3 year scholarship thatpermitted her to conduct these studies at the TechnicalUniversity of Denmark.

ReferencesAn, Y., Zhang, J.S. and Shaw, C.Y. (1999) Measurements of VOC

adsorption/desorption characteristics of typical interior buildingmaterial surfaces. International Journal of HVAC and R Research5, 297–316.

ASHRAE (2004) Ventilation for acceptable Indoor Air Quality,Atlanta, GA, American Society of Heating, Refrigeration andAir Conditioning Engineers (ASHRAE Standard 62.1–2004).

Berglund, B., Johansson, I. and Lindvall, T. (1988) Adsorption anddesorption of organic compounds in indoor materials. In: Ber-glund, B and Lindvall, T. (eds.), Proceedings of Healthy Build-ings� 88, Vol. 3, Stockholm, Swedish Council for BuildingResearch, 299–309.

Borazzo, J.E., Davidson, C.I. and Andelman, J.B. (1990) Sorptionof organic vapor to indoor surfaces of synthetic and naturalfibrous materials. In: Walinshaw, D.S. (ed.) Proceedings ofIndoor Air 90, Vol. 3, Toronto, International Conference onIndoor Air Quality and Climate, 617–623.

CEN CR 1752 (1998) Ventilation for Buildings: Design Criteria forthe Indoor Environment, Brussels, European Committee forStandardization.

Colombo, A., De Bortoli, M., Knoppel, H., Pecchio, E. and Vissers,H. (1993) Adsorption of selected VOCs on a carpet, a wallcoating and a gypsum board in a test chamber. Indoor Air, 3,276–282.

Dokka, T.H., Jørgensen, R.B., Bjørzeth, O. and Malkiv, B. (1999)Comparison of field experiments in a refurbished bedroom withsmall chamber experiments. In: Raw, G., Aizlewood, C. andWarren, P. (eds.), Proceedings of Indoor Air �99, Vol. 5, Edin-burgh, Construction Research Communications Ltd., 93–98.

Fang, L., Clausen, G. and Fanger, P.O. (1998) Impact of tem-perature and humidity on perception of indoor air quality.Indoor Air, 8, 80–90.

Fanger, P.O. (1988) Introduction of the olf and the decipol units toquantify air pollution perceived by humans indoors and out-doors. Energy and Buildings, 12, 1–6.

Fanger, P.O. (1989) The new comfort equation for indoor airquality. ASHRAE J., 31, 33–38.

Gunnarsen, L. and Fanger, P.O. (1992) Adaptation to indoor airpollution. Environ. Int., 18, 43–54.

Jørgensen, R.B. and Bjørseth, O. (1999) Sorptive behavior of VOCson material surfaces- the influence of combinations of com-pounds and materials compared to sorption of single compoundson single materials. Environ. Int., 25, 17–27.

Jørgensen, R.B., Knudsen, H.N. and Fanger, P.O. (1993) Theinfluence on indoor air quality of adsorption and desorption oforganic compounds on materials. In: Jaakkola, J., Ilmarinen, R.and Seppanen, O. (eds.), Proceedings of Indoor Air �93, Vol. 2,Helsinki, Indoor Air 1993, 383–388.

Kephalopoulos, S., Knoeppel, H. and De Bortoli, M. (1996) Testingof sorption and desorption of selected VOCs on/from indoormaterials. In: Yoshizawa, S., Kimura, K., Ikeda, K., Tanabe, S.and Iwata, T. (eds.), Proceedings of Indoor Air �96, Vol. 2,Nagoya, International Conference on Indoor Air Quality andClimate, 61–66.

Kirchner, S., Maupetit, F., Quenard, D., Rouxel, P. and Giraud, D.(1995) Characterization of adsorption/desorption of volatileorganic compounds on indoor surface materials. In: Maroni, M(ed.) Proceedings of Healthy Buildings 95, Vol. 2, Milano,International Society of Indoor Air Quality and Climate,953–958.

Knudsen, H.N., Clausen, G. and Fanger, P.O. (1997) Sensorycharacterization of emissions from materials. Indoor Air, 7, 107–115.

Morrison, G.C. and Nazaroff, W.W. (2000) The rate of ozone up-take on carpets: experimental studies. Environ. Sci. Techn., 34,4963–4968.

Morrison, G.C. and Nazaroff, W.W. (2002) Ozone interactions withcarpet: Secondary emissions of aldehydes. Environ. Sci. Techn.,36, 2185–2192.

Nazaroff, W.W. and Weschler, C.J. (2004) Cleaning products andair fresheners: exposure to primary and secondary air pollutants.Atmospheric Environ., 38, 2841–2865.

Nielsen, P.A. (1987) Potential pollutants-their importance to thesick building syndrome and their release mechanism. In: Seifert,B., Esdorn, H., Fischer, M., Ruden, H. and Wegner, J. (eds.),Proceedings of Indoor Air� 87, Vol. 2, Berlin, InternationalConference on Indoor Air Quality and Climate, 598–602.

Salthammer, T. (1999) Organic Indoor Air Pollutants: Occurrence,Measurement, Evaluation, Weinheim, New York, Wiley-VCH.

Schlitt, C. and Knoppel, H. (1997) Human exposure to volatileorganic compounds from household products: the influence ofsorption phenomena. In: Proceedings of the InternationalConference �Volatile Organic Compounds in the Environment–Risk Assessment and Neurotoxicity�, Vol. 3, 5–7 October 1997,Pavia, Advances in Occupational Medicine and Rehabilitation,3–15.

Seifert, B. and Schmahl, H. (1987) Quantification of sorptioneffects for selected organic substances present in indoor air. In:Proceedings of Indoor Air 87, Vol. 1, Berlin, International Con-ference on Indoor Air Quality and Climate, 252–256.

Siegel, S. and Castellan, N.J. (1988) Nonparametric Statistics forBehavioral Sciences, New York, McGraw-Hill.

Singer, B.C., Hodgson, A.T. and Nazaroff, W. (2003) Gas-phaseorganics in environmental tobacco smoke. 2. Exposure relevantemission factors – Effect and indirect exposures from habitualsmoking. Atmospheric Environ., 37, 5551–5561.

Singer, B.C., Revzan, K.L., Hotchi, T., Hodgson, A.T. and Brown,N.J. (2004) Sorption of organic gases in a furnished room.Atmospheric Environ., 38, 2483–2494.

Sparks, L.E., Guo, Z., Chang, J. and Tichenor, B.A. (1999) Volatileorganic compound emissions from latex paint-part 2 test housestudies and indoor air quality (IAQ) modeling. Indoor Air, 9,18–25.

Tichenor, B.A., Guo, Z., Dunn, J.E., Sparks, L.E. and Mason,M.A. (1991) The interaction of vapor phase organic compoundswith indoor sinks. Indoor Air, 1, 23–35.

Tirkkonen, T. and Saarela, K. (1997) Adsorption of VOCs oninterior surfaces in a full scale building. In: Woods, J.E.,Grimsrud, D.T. and Boschi, N. (eds.), Proceedings of HealthyBuildings �97, Vol. 3, Washington, DC, Healthy Buildings/IAQ’97, 551–559.

Van der Wal, J.F., Hoogeveen, A.W. and Van Leuwen, L. (1998) Aquick screening method for sorption effects of VOCs on indoormaterials. Indoor Air, 8, 103–112.

Van Loy, M.D., Rilley, W.J., Daisey, J.M. and Nazaroff, W.W.(2001) Dynamic behavior of semi volatile organic compounds inindoor air.2. Nicotine and Phenanthrene with carpet and wall-board. Environ. Sci. Techn., 35, 560–567.

Weschler, C.J., Hodgson, A.T. and Wooley, J.D. (1992) Indoorchemistry – ozone, volatile organic compounds, and carpets.Environ. Sci. Techn., 26, 2371–2377.

Won, D., Corsi, R.L. and Rynes, M. (2000) New indoor carpet asan adsorptive reservoir for volatile organic compounds. Environ.Sci. Techn., 34, 4193–4198.

Won, D., Corsi, R.L. and Rynes, M. (2001) Sorptive interactionbetween VOCs and indoor materials. Indoor Air, 11, 246–256.

Wouda, P., Bluyssen, P.M., Cornelissen, H.J.M., Hoogeveen, A.W.and Van der Wal, J.F. (1997) Chemical and sensory evaluation of


sorption effects of chemicals on combined indoor materials incomparison to sorption effects on the individual materials. In:Woods, J.E., Grimsrud, D.T. and Boschi, N. (eds.), Proceedingsof Healthy Buildings �97, Vol. 2, Washington, DC, HealthyBuildings/IAQ ’97, 225–230.

Zellweger, C., Gehrig, R., Hill, M. and Hofer, P. (1995) VOCemissions from building materials: comparison of chamberemission data with concentrations in real room. In: Maroni, M(ed.) Proceedings of Healthy Buildings 95, Vol. 2, Milano, In-ternational Society of Indoor Air Quality and Climate, 845–851.

Sakr et al.

110

The impact of sorption on perceived indoor air quality

Documents

Transcript of The impact of sorption on perceived indoor air quality