ArticlePDF Available

Diel variation of formaldehyde levels and other VOCs in homes driven by temperature dependent infiltration and emission rates

Authors:

Abstract and Figures

High time resolution monitoring of formaldehyde and other volatile organic compounds in the air of four homes in winter and summer revealed diel variation of VOC levels driven by infiltration and temperature dependent whole house emission rates. In unoccupied homes, these pollutants displayed a large diel concentration variation, with an afternoon maxima and early morning minima. VOC abundance lagged about 2 h behind changes in infiltration rates measured by a tracer release method, resulting in poor correlations between VOC concentration and air change rate. The data demonstrate that VOC abundance was not in steady state with respect to whole house emission rates. Formaldehyde and other VOCs displayed a positive correlation with indoor temperature in both winter and summer. Formaldehyde sensitivity to temperature ranged from 3.0 to 4.5 ppbv per °C, a useful metric for predicting the impact of heat waves and changing regional climate on indoor air quality. Gypsum wallboard used as radiant ceiling heating product in one home was identified as source of formaldehyde and potentially mercury.
Content may be subject to copyright.
1
Diel variation of formaldehyde levels and other VOCs in homes
1
driven by temperature dependent infiltration and emission rates
2
3
Yibo Huangfu1, Nathan M. Lima1,2, Patrick T. O’Keeffe1, William M. Kirk2, Brian K. Lamb1,
4
Shelley N. Pressley1, Beiyu Lin3, Diane J. Cook3, Von P. Walden1, Bertram T. Jobson1,*
5
6
1 Laboratory for Atmospheric Research, Department of Civil and Environmental Engineering,
7
Washington State University, Pullman, WA, USA
8
2 School of Architecture and Construction Management, Washington State University,
9
Pullman, WA, USA
10
3 School of Electrical Engineering and Computer Science, Washington State University,
11
Pullman, WA, USA
12
*Corresponding author. Phone: 509-335-2692. E-mail: tjobson@wsu.edu
13
Present Address: 2001 Grimes Way, Pullman, WA 99164-5845
14
15
Declarations of interest: none
16
Color should be used for all the figures except Figure 5
17
18
DOI: 10.1016/j.buildenv.2019.05.031
19
20
©2019. This manuscript version is made available under the CC-BY-NC-ND 4.0
21
license http://creativecommons.org/licenses/by-nc-nd/4.0/
22
2
Abstract
23
High time resolution monitoring of formaldehyde and other volatile organic
24
compounds in the air of four homes in winter and summer revealed diel variation of VOC
25
levels driven by infiltration and temperature dependent whole house emission rates. In
26
unoccupied homes, these pollutants displayed a large diel concentration variation, with an
27
afternoon maxima and early morning minima. VOC abundance lagged about 2 hours behind
28
changes in infiltration rates measured by a tracer release method, resulting in poor
29
correlations between VOC concentration and air change rate. The data demonstrate that VOC
30
abundance was not in steady state with respect to whole house emission rates. Formaldehyde
31
and other VOCs displayed a positive correlation with indoor temperature in both winter and
32
summer. Formaldehyde sensitivity to temperature ranged from 3.0 to 4.5 ppbv per °C, a
33
useful metric for predicting the impact of heat waves and changing regional climate on indoor
34
air quality. Gypsum wallboard used as radiant ceiling heating product in one home was
35
identified as source of formaldehyde and potentially mercury.
36
37
Keywords: Indoor air; formaldehyde; volatile organic compounds; diel variation; infiltration;
38
PTR-MS
39
40
41
3
1. Introduction
42
Poor air quality, both indoors and outdoors, is noted as a leading cause globally of
43
non-communicable disease and mortality [1]. Exposure to chemicals in the indoor
44
environment is important for understanding the impact of air quality on non-communicable
45
diseases, such as asthma [2,3] and cardiovascular diseases [4]. Interestingly, exposure to
46
airborne chemicals including elevated carbon dioxide has been implicated in reduced
47
cognitive function and neurological health [59]. Inhalation exposure to chemicals in
48
residential dwellings arises from off gassing of chemicals from building materials and
49
furnishings [1017], from the use of household chemical products [1823], and from human
50
activities such as cooking [2429]. Chemical reactions indoors, such as the reaction of ozone
51
with materials such as carpets [30] and human skin lipids [31,32], and radical gas phase
52
reactions [33], are also potential sources of volatile organic compounds such as
53
formaldehyde. Adequate ventilation of buildings should reduce concentrations of airborne
54
chemicals that have indoor sources, but in the case of formaldehyde, a clear relationship
55
between concentration and air change rate has not been established. Residential indoor
56
concentrations of formaldehyde have been reported to display a strong negative correlation
57
with air change rates [3436] or no apparent correlation with air change rates [37,38]. This
58
lack of clarity is in part attributable to the typical temporal resolution of indoor air studies
59
where passive samplers and integrated sampling over long time periods obfuscates
60
connections between pollutant levels and physical and chemical processes, including human
61
activities that can be both a source of chemicals and variation in building air change rates.
62
Indoor air quality, ventilation, and building energy consumption are linked issues.
63
Increased ventilation rates come at the cost of a greater energy requirement for heating and
64
cooling needs. Residential buildings in the US are estimated to account for 20% of total
65
national energy consumption and related CO2 emissions [39]. In the US, ventilation
66
4
requirements for residential dwellings are covered by ASHRAE standard 62.2 that
67
recommends 0.35 air changes per hour (ACH), but it is not clear if this leads to acceptable
68
levels of air quality for homes. A conundrum exists between the goal of reducing building
69
energy needs that often involve making buildings more airtight and the public health goal of
70
reducing our exposure to chemicals that make us unhealthy and less mentally fit. Efforts to
71
mitigate climate change by reducing building energy needs through the construction of more
72
energy efficient buildings, such as net-zero energy homes, must consider the potential public
73
health cost of increased chemical exposure. In the construction of net-zero energy homes
74
where infiltration rates may be as low as 0.02 hr-1, it is important to choose building materials
75
with low outgassing rates paired with a reliable and energy efficient mechanical ventilation
76
system to control indoor air pollutant concentrations [40]. Outgassing rates of volatile organic
77
compounds (VOCs) from building materials are known to be temperature dependent [4147],
78
but there are few reports describing how indoor air VOC concentrations vary with
79
temperature. For Chinese dwellings it has been reported that formaldehyde concentrations
80
displayed a strong temperature dependence [4851], while no dependence was found in a
81
large study of US homes [37]. The difficulty in understanding the role of indoor temperature
82
comes from the fact that indoor temperature can affect both infiltration rates [52,53] and
83
VOC emission rates from materials, producing opposing effects on indoor air concentration.
84
The objective of this study was to better understand how climate change may impact
85
indoor air quality of the US residential housing stock through meteorological influence on
86
building infiltration rates and temperature dependent off-gassing rates from building
87
materials. This paper describes the results of high temporal resolution monitoring of air
88
change rates and concentrations of several air toxic compounds, including formaldehyde,
89
acetaldehyde, benzene, and methanol, from four homes in eastern Washington, USA. The
90
5
high time resolution VOC measurements allowed us to study the dependence of pollutant
91
concentration on indoor temperature and infiltration rates.
92
93
2. Materials and methods
94
2.1 Description of homes
95
Houses were chosen in this study to provide a range of ages and design characteristics
96
that represent typical detached home types listed in the NIST database of 209 representative
97
residential building types in the US [54], with the idea in mind that older houses may off-gas
98
at a lower rate or may off-gas different types of VOCs and other pollutants, and that newer
99
homes may off-gas at a higher rate and may have lower air change rates (ACH). In this paper
100
we focus on the results from homes H2, H6, H7, and H10 where clear patterns were evident
101
in the data and provided opportunities to examine relationships between VOC abundance,
102
infiltration rates, and indoor air temperature. General characteristics of the homes are listed in
103
Table 1. These homes were located in the town of Pullman, WA and nearby communities.
104
These communities would be considered rural, and outdoor air concentrations of pollutants
105
were typically low except during wildfire events when PM2.5 can be very high [55]. For all
106
homes, measurements were performed twice, for a 1-week period in summer (August or
107
September) and a 1-week period in winter (January or February). All the homes had attached
108
garages and had a range of ventilation types and heating appliances. H2 and H7 had a central
109
forced air system (CFA) that could provide heating and air conditioning. H7 also had an
110
auxiliary fresh air intake (FAI) vent that periodically provided fresh air flow. H6 and H10 did
111
not have a central air system and relied on electrical baseboard heaters for heating (H6) or
112
radiant ceiling heating (H10), common designs for homes built in the Pacific Northwest of
113
the US during the period 1960 1980. In the US it is estimated that approximately 35% of
114
homes do not have a central forced air ventilation system [56].
115
6
116
Table 1. Selected characteristics of homes investigated in this study
117
Home
Year
Built
Occupants
(adults /
children)
Living
space (ft2)
Ventilation
type
Home
Heating
H2
1963
2 / 0
1,765
CFA
G
H6
1958
2 / 0
1,804
no CFA
Baseboard
H7
2010
1 / 0
1,051
CFA -FAI
G
H10
1972
2 / 1
2,116
no CFA
Radiant
ceiling
*G = gas, E = electric, NA = none
118
119
2.2 Measurement methods
120
A suite of instruments was used to measure both indoor and outdoor air pollutant
121
concentrations. These instruments were typically located in the garage. Indoor and outdoor
122
measurements were alternated by using a three-way solenoid valve (Galtek, USA) that
123
connected the instruments to either a 1/2” OD PFA sample line from inside the house or a
124
3/8” PFA sample line that went to inlet tripod mounted on the house roof. The tripod also
125
supported a mast for a surface weather station (Airmar 200WX). Indoor and outdoor
126
sampling was switched every 15 minutes. For some houses, the garage air and attic air were
127
also sampled on a 4-hour measurement interval. The indoor inlet was run along the ceiling to
128
a location chosen to be relatively open that connected several spaces within the house (i.e.,
129
kitchen-hallway-living room). The inlet was mounted so that it was several inches from the
130
ceiling surface. From these indoor and outdoor inlets, continuous measurements of O3, CO,
131
NOx (Teledyne Inc.), CO2, H2O (LI-840a, LI-COR), and selected VOCs (PTR-MS, Ionicon
132
Analytik) were performed. Outdoor measurements of PM2.5 mass concentration (DustTrak II
133
8530, TSI) were also made from a separate copper tubing inlet mounted on the roof tripod.
134
Additional equipment was located indoors and mounted in a rack placed in the living room to
135
measure temperature, PM2.5 mass (DustTrak II 8530, TSI), PM number concentration
136
7
(DC1100, Dylos), CO2 and H2O (LI-840a, LI-COR). Five wall mounted devices measured
137
CO2 (K30, Senseair), RH, and temperature and these were distributed throughout the home.
138
Sensors for measuring window and door openings and movement of people were also
139
installed [57].
140
A PTR-MS instrument measured selected VOCs, typically 30 ions were monitored.
141
The drift tube was operated at 120 Td, with an O2+/H3O+ ratio less than 2% and a NO+/H3O+
142
ratio less than 0.2%. The drift tube chamber was maintained at 60 °C and the drift tube
143
pressure at 2 mbar. The PTR-MS response was calibrated with compressed gas standards
144
with a stated accuracy of 5% (Scott Marrin, CA and Apel-Reimer Environmental, FL, USA)
145
which were dynamically diluted to about 20 parts per billion (ppbv). A permeation source
146
(KinTek, TX, USA) was used to calibrate PTR-MS formaldehyde response as a function of
147
water vapor concentration [58]. Formaldehyde response factors used for indoor and outdoor
148
data processing were calculated separately to account for the difference in indoor and outdoor
149
water vapor levels. Formaldehyde has a small positive inference resulting from methanol and
150
ethanol reactions with O2+. This interference was accounted for (typically < 5%) but could be
151
significant at times (30%) in some homes when very high concentrations of methanol and
152
ethanol occurred as a result of human activities. The instrument background signal was
153
determined automatically by sampling zero air made by catalytically scrubbing ambient air
154
[58]. Dissociative protonation reactions are common at 120 Td [59] and can be problematic
155
for accurately determining the abundances of some compounds. Notably benzene suffers
156
from a positive interference from ethylbenzene, cumene, and n-propyl benzene, compounds
157
that are often present together due to their common source from automobile exhaust and
158
gasoline vapor. We estimate that benzene levels in our measurements, in particular of garage
159
air, may be overestimated by as much as 8% due to interferences from compounds in gasoline
160
vapor.
161
8
162
2.3 Air change rate measurement
163
The air change rate per hour (ACH) was determined from periodic release of CO2 into
164
the home via the central forced air system. Pure CO2 was injected into the furnace supply
165
duct, typically every 2 hours, by flowing 20 SLPM of CO2 gas for 2-3 minutes. CO2 was
166
continuously measured in the return duct (LiCor 840a) as a measure of the overall indoor
167
level. The furnace fan continuously ran to move and mix the air throughout the home during
168
tracer release experiments. The CO2 increase after injection was typically between 250 to 300
169
ppmv. The ACH was determined from the rate of change of the CO2 concentration with time
170
due to the exchange of air from outdoors [60]. A box model analysis was applied to the CO2
171
decay data to determine the ACH. Inputs to the model included the measured outdoor air
172
concentration of CO2, the assumption that CO2 is conserved, and a human respiration source
173
term. We used a human CO2 respiration of 31.2 g hr-1 during nonsleeping periods and 20.8 g
174
hr-1 during sleeping periods [61]. Periods, when the homes were not occupied, provided a
175
clear picture of the ACH and were used to verify that the human respiration values used for
176
occupied periods were appropriate. Given the uncertainty in the human respiration values, we
177
estimate the uncertainty of the ACH during occupied periods was ~ 20%.
178
179
2.4 Gypsum wallboard chamber test
180
Emission rates for the gypsum wallboard used in home H10 radiant ceiling heating
181
system were measured in the lab using an emissions testing chamber. The size of the gypsum
182
wallboard sample was 0.065 m2. The sample was placed in a 150-L Teflon film chamber, as
183
described in Toro et al. [62]. Zero grade air humidified to ~45% RH flowed through the
184
chamber at 10 L min-1 and fans inside the chamber provided constant mixing. VOCs, O3, and
185
SO2 were measured from the chamber outlet by PTR-MS, an O3 analyzer (Teledyne T400),
186
9
and an SO2 analyzer (Teledyne T100U). The gypsum wallboard was placed inside the
187
chamber at room temperature of 25 °C and then heated to a 31 °C surface temperature using
188
the native resistive wiring in the sample. Measured ceiling temperatures in the home using a
189
non-contact IR thermometer were around 30 30.6 °C so the chamber test results should
190
represent the actual condition in H10. This chamber test was repeated with the same setup
191
except using zero air containing 100 ppbv NO to verify the measured O3 signal.
192
193
3. Results and discussions
194
3.1 Diel variation of pollutant levels in unoccupied homes
195
Measurements made in unoccupied homes provided important data for understanding
196
whole house emission rates and resulting air pollutant levels. Human activities in the home
197
can be a source of these chemicals and can influence ventilation rates through window and
198
door openings, obscuring relationships between indoor air concentrations and ACH. Figure 1
199
displays the temporal variation of indoor levels of formaldehyde, acetaldehyde, methanol,
200
and benzene, the measured ACH, and indoor and outdoor temperatures for H2 in summer
201
during an unoccupied period with windows and doors closed. The air pollutant mixing ratios
202
clearly varied with time of day. Highest mixing ratios occurred in the early afternoons and the
203
levels decreased by about 40% to an early morning low. This diel trend in indoor pollutant
204
levels appears to be driven by the factor of 2 diel variation in infiltration rates, ranging from
205
an afternoon low of ~0.2 hr-1 to early morning high values of 0.4 hr-1. This diel variation in
206
infiltration is driven principally by the variation of the indoor-to-outdoor temperature
207
difference and is consistent with our understanding of how infiltration rates are influenced by
208
temperature gradients across the building envelope. Lamb et al. [53] measured air change
209
rates for homes in Pullman, WA and parametrized ACH with an empirical model using
210
temperature differences and wind speed as variables. This meteorological parameterization
211
10
was used to infer ACH for H2, as illustrated by the black shading in Figure 1. The shading
212
represents the 10% uncertainty range of the infiltration rate parameterization. The predicted
213
infiltration rate yields a diel variation and values consistent with our measurements. The diel
214
variation was largely driven by the difference in temperature across the building envelope.
215
Wind speeds were low and accounted for less than 10% of the overall ACH diel variation.
216
The continuously changing infiltration rates over the day created a dynamic indoor
217
environment, and indoor VOC mixing ratios were not constant with time.
218
219
Figure 1. Time series showing pollutant mixing ratio, indoor and outdoor temperature and
220
measured ACH during an unoccupied summer period in home H2. VOC data are shown as 15
221
11
minute averages with standard deviation. Expected infiltration rate due to the difference in
222
indoor and outdoor temperature and wind speed as parameterized by Lamb et al. [53] is
223
shown by black shading and is a reasonable match to observed values and their diel variation.
224
225
Close inspection of the data in Figure 1 reveals a time lag of ~2 hours between the
226
daily minimum indoor pollutant levels and the daily maximum of ACH. The lag time is
227
related to the characteristic timescale for a compound to reach a steady state concentration.
228
This characteristic time is a function of the ACH and the compound’s first order loss rate
229
inside the home due to irreversible surface uptake or homogenous gas phase removal. Of the
230
compounds shown, only formaldehyde is thought to have a significant first order loss rate, on
231
the order of 0.4 hr-1 [63,64], similar to the loss rates from dilution caused by infiltration. For
232
compounds like methanol and benzene that are not thought to have appreciable loss rates
233
inside the home, the characteristic timescale to reach steady state is determined by the
234
infiltration rate. The lag between VOC abundance and ACH results in a relatively poor
235
correlation between VOC mixing ratios and ACH as shown in Figure 2. The left hand side
236
panels of Figure 2 show the average VOC mixing ratio over the two hour ACH determination
237
period plotted against measured ACH. Averaging the VOC data over a 2 hour period
238
following the ACH determination, to account for the apparent lag time, yields an improved
239
correlation, as shown in the center panels of Figure 2. A clear negative correlation can be
240
observed for acetaldehyde, methanol, and benzene but the formaldehyde data are still quite
241
scattered. One reason for this might be that the formaldehyde physical loss rate in the home is
242
sufficiently large that it is the dominant factor in determining concentration-time trends. The
243
right hand panels of Figure 2 illustrate that a six-hour averaging period improves the
244
correlation between VOCs and ACH. This averaging period was long enough to smooth out
245
the time-lag differences. Longer time averaging hides the lag time inherent in the data and the
246
12
fact the system was never in steady state. It is a common assumption in the indoor air quality
247
literature to assume steady state conditions for measured pollutants so that concentrations can
248
be related to emission rates through a simple analytical box model [36,6569]. For homes
249
where ACH is determined by meteorologically modulated infiltration rates, steady-state
250
pollutant concentrations are unlikely to be achieved and thus the relationship between indoor
251
pollutant levels and emission rates is more nuanced. Assuming steady state conditions is
252
likely a poor assumption for most homes.
253
254
Figure 2. The correlation between VOC abundance and measured ACH for the unoccupied
255
period in home H2 as illustrated in Figure 1. Left most panels show the resulting correlation
256
when VOC data are averaged over the two hour ACH determination. Center panels illustrate
257
13
the correlation if the VOC data are averaged over a 2-hr period following the ACH
258
determination. Right most panels illustrate correlations for six hour averages that smooth out
259
the lag time difference.
260
261
Figure 3 illustrates another unoccupied period in a different house, H7, where the diel
262
variation of formaldehyde and other pollutants was observed as a result of diel variation
263
infiltration rates. In this house, the CFA had a fresh air intake (FAI) vent. This FAI was left
264
continuously open for the first part of the monitoring period then closed for a 24-hour period
265
as a deliberate manipulation of air change rates to examine the impact on pollutant levels.
266
During the unoccupied period with the FAI open, the measured ACH values were about 0.3
267
to 0.4 hr-1 greater than infiltration rates predicted from the meteorological parameterization.
268
Even with the FAI vent open, a diel variation in ACH was observed due to the contribution of
269
infiltration. The diel variation in ACH was qualitatively consistent with the time of day
270
variation in infiltration rates due to the changing temperature gradient across the building
271
envelope. Normally the FAI vent is programmed to periodically open for a few minutes to
272
allow fresh air flow into the home, so its typical impact would be much less pronounced than
273
shown here. Measured ACH values for the period when the vent was open were between 0.54
274
and 0.95 hr-1. With the FAI vent sealed shut, the measured ACH and model infiltration values
275
were in fair agreement except at night when the homeowner opened a bedroom window.
276
During the unoccupied period, the lowest formaldehyde values occurred around sunrise (~
277
05:40 PST) and maximized shortly after 16:00 PST, similar to the pattern observed in H2.
278
When the FAI vent was closed on Sept 27th, the average afternoon (14:00-16:00)
279
formaldehyde levels increased by a factor of 1.7 compared to the previous afternoon, whereas
280
methanol and benzene tripled in abundance, and acetaldehyde doubled. The much greater
281
change in the abundance of methanol, benzene, and acetaldehyde to changes in ACH
282
14
illustrates that formaldehyde levels are less sensitive to the variations in ACH than these
283
other compounds.
284
285
Figure 3. Time series showing methanol and formaldehyde mixing ratio, indoor and outdoor
286
temperature and measured ACH from home H7 in summer. Light grey shading indicates a
287
period when the house was unoccupied. Dark grey shading indicates when the fresh air intake
288
(FAI) valve was closed. Black shading indicates the range of home infiltration rates predicted
289
by the meteorological model and open circles are the measured ACH. 15-minute averages
290
every ½-hr of indoor formaldehyde (blue circles) and methanol (red circles) are shown.
291
Indoor (red) and outdoor (black) temperatures are shown in the lower plot.
292
293
294
295
15
3.2 Formaldehyde dependence on indoor air temperature
296
Diel variation in formaldehyde and a consistent day-to-day increase in abundance
297
with warming temperatures were noted in home H10 in summer. Measurements made in this
298
home again illustrated the impact of meteorological variation of infiltration rates on indoor
299
air quality. Figure 4 illustrates the variation of formaldehyde, infiltration rates, and
300
temperature. H10 did not have a CFA system so infiltration rates were estimated from the
301
meteorological parametrization of Lamb. et al. [53] as was done for H2 and H7. In those
302
homes, the measured infiltration rates matched reasonably well with the estimated infiltration
303
rates, giving us confidence that the meteorological parametrization would reasonably
304
estimate infiltration rates for H10. During the week, the outdoor afternoon air temperatures
305
climbed steadily from an average of 11.7 °C on Sept 21st to 24.8 °C on Sept 26th. Indoor air
306
temperatures increased correspondingly. Formaldehyde levels in the house tracked indoor
307
temperature and formaldehyde levels steadily built up in the home as it got warmer.
308
Formaldehyde levels were at their highest and reasonably constant from about 18:00 to 24:00
309
PST corresponding in time to when it was warmest in the house. From Sept 22nd, onward
310
there was a consistent pattern of low air change rates in the afternoon (0.30 to 0.35 hr-1) and
311
higher air change rates at night (0.40 to 0.45 hr-1). For the first 1.5 days formaldehyde levels
312
inside the house were relatively constant at 22 ppbv, attributable to little day-to-night
313
variation in infiltration rates. Indoor formaldehyde levels were about a factor of 20 or more
314
greater than outdoor levels. Outdoor levels displayed a diel variation, with afternoon
315
abundance of ~1.9 ppbv on Sept 26th compared to over 40 ppbv inside the home on that day.
316
The large decrease in formaldehyde occurring on Sept 24th was attributed to the door between
317
the house and the garage being left open as the home owner worked in the garage that
318
Saturday.
319
16
320
Figure 4. Formaldehyde time series for H10 in summer showing a steady increase in
321
formaldehyde with increasing indoor air temperature. Indoor formaldehyde (blue circles,
322
upper panel) and outdoor formaldehyde (middle panel) mixing ratios are shown as 15 minute
323
averages every ½-hr. Indoor air temperature (red trace) and outdoor (black trace) air
324
temperature are shown in the upper panel. Infiltration rates calculated from the
325
meteorological data are shown in the lower panel.
326
327
The steady increase in indoor formaldehyde was well correlated with temperature and
328
illustrates the temperature dependence of whole house formaldehyde emission rates. In
329
Figure 5, the difference between indoor and outdoor formaldehyde mixing ratios averaged
330
from 18:00 to 24:00 PST are plotted against the corresponding average indoor air temperature
331
for the six evenings illustrated in Figure 4. The corresponding estimated infiltration rates are
332
17
indicated in the plot besides the symbols. The estimated infiltration rates for the six evenings
333
were similar and essentially identical for the last four evenings of the week (Sept 23rd to Sept
334
26th). The change in formaldehyde levels was attributed to an increasing whole house
335
formaldehyde emission rates as the indoor air temperatures increased. The slope of the plot is
336
4.6 ± 0.6 ppbv formaldehyde per °C. The 4.5 °C temperature change in the house over the
337
week resulted in a doubling of formaldehyde levels inside the home. This sensitivity to
338
temperature likely reflects whole house emission rate sensitivity from materials in this home
339
including composite wood flooring and furnishings.
340
341
Figure 5. H10 formaldehyde data showing a correlation between indoor and outdoor
342
difference versus indoor air temperature averaged between the hours 18:00 to 24:00 PST
343
when indoor air temperature and formaldehyde levels were highest. Error bars indicate the
344
standard deviation of the averages. Numeric values adjacent to data indicate corresponding
345
ACH average in units of hr-1.
346
347
In general the relationship between VOC levels and indoor temperature is complex
348
due to temperature’s influence on infiltration rates and surface emission rates from materials
349
18
inside the home. These rates can vary throughout the day, display seasonal differences, and
350
can impact indoor air concentrations in opposite ways. Figure 6 illustrates the 1-minute
351
indoor formaldehyde data as a function of indoor temperature for three of the study homes
352
(H2, H6, and H10) to contrast winter and summer behavior for conditions where infiltration
353
dominated air change rates. We removed data when natural ventilation through window or
354
door openings impacted concentrations. The lowest 10% of the data associated with a
355
particular temperature range are highlighted to better reveal trends with temperature by
356
avoiding obfuscation from human activity emission events. These highlighted data thus
357
represent conditions with the highest infiltration rates and/or lowest emission rates. For these
358
homes summer time formaldehyde levels increased with indoor air temperature. Lowest
359
temperatures were associated with early morning periods when ACH values were typically
360
greatest. Higher temperatures were associated with afternoon data when infiltration rates
361
were typically lowest, and whole house emission rates would be greatest. The overall
362
temperature trend thus represents the combined influence of these two factors and potentially
363
other factors such as secondary source influence from O3 infiltration into the home and
364
indoor formaldehyde loss rates. A linear regression through the highlighted summer data
365
display similar slopes: 4.1 ± 0.2 ppbv per °C for H2; 3.0 ± 0.1 ppbv per °C for H6; and 4.5 ±
366
0.07 ppbv per °C for H10. The slope through the H10 summer data is very close to the 4.6 ±
367
0.6 ppbv per °C determined from the H10 data in Figure 5 that reflects whole house emission
368
rate sensitivity. These trends provide a useful simple predictor of the influence of indoor air
369
temperature on formaldehyde levels. A similar temperature dependence has been measured
370
from large surveys of residential dwellings in China [48,49]. In those studies the
371
formaldehyde mixing ratio sensitivity to temperature, assuming standard temperature and
372
pressure conditions, was 5.1 ppbv per oC [49] and 3.5 ppbv per oC [48], very similar to our
373
results.
374
19
375
376
Figure 6. Indoor extra formaldehyde levels (indoor-outdoor) as a function of indoor air
377
temperature for homes H2, H6, and H10. Shown are 1-minute data for summer (red circles)
378
and winter (black squares). Some H10 winter data are off-scale. Data are selected for periods
379
where infiltration dominated air exchange (natural ventilation periods excluded). Lowest 10%
380
of the data points are highlighted for trend analysis.
381
382
The winter trend for house H6 is similar to summer with formaldehyde levels
383
typically lower in winter, consistent with expected seasonal difference in infiltration rates.
384
The winter time infiltration rates estimated from the meteorological parametrization ranged
385
from 0.44 to 0.74 hr-1 compared to summer’s 0.29 to 0.48 hr-1 range. In house H10 the winter
386
data curiously lay atop the summer data with a slightly steeper slope (5.9 ± 0.2 ppbv per °C),
387
likely as a result of our sample inlet being close to an important wintertime formaldehyde
388
source in the home, the radiant ceiling, discussed in section 3.3. The steeper slope may reflect
389
the influence of the radiant ceiling as an extra source in winter compared to summer. The
390
winter data for house H2 display a much weaker dependence on temperature (0.58 ppbv
391
per °C) with overall much lower home temperature and lower formaldehyde than in summer.
392
During the winter measurement period, the home was largely unoccupied, and the thermostat
393
was set back to save energy. As a result, the home was quite cool. The home had relatively
394
20
high measured ACH that could not be explained by infiltration driven by meteorology. High
395
ACH may have been driven by unintended natural ventilation as a result of a fireplace
396
damper being left open or a “doggy door” being ajar. Low whole house formaldehyde
397
emission rates as a result of lower indoor temperatures coupled with relatively high ACH
398
resulted in much lower concentrations of formaldehyde and VOCs for this home compared to
399
other homes in winter. The slope through the winter data is a factor of 7 less than the summer
400
trend. The lower sensitivity to temperature in winter is likely a consequence of the home
401
ventilation being driven by natural ventilation rather than infiltration driven by temperature
402
gradients across the building envelope. We postulate that at some point indoor temperatures
403
might become cool enough that surface emission rates would be low enough so that the
404
primary formaldehyde sources become less important than secondary sources such as the
405
reaction of O3 with interior surfaces. This would allow for better qualification of indoor
406
secondary sources caused by ozone surface reactions in homes.
407
The correlations with temperature for the three other VOCs are illustrated in Figures
408
S-1, S-2, and S-3. The lower 10% of the data were highlighted to discern baseline trends. The
409
summer time VOC data from the unoccupied period of house H2 displayed a clear
410
temperature dependence: 13.4 ± 0.5 ppbv per °C for methanol, 9.6 ± 0.2 ppbv per °C for
411
acetaldehyde, and 0.93 ± 0.04 ppbv per °C for benzene. Again, the temperature dependence
412
for the H2 winter measurements were weaker; the slopes were a factor 10 less for all three
413
VOCs, similar to what was observed for formaldehyde. For home H6 and H10 temperature
414
trends were less clear because of large variability in mixing ratios attributed to home
415
occupancy and human activity emission events such as cooking and household chemical use.
416
For methanol in houses H6 and H10 in summer, the lower 10% of the data displayed a trend
417
with temperature: 6.5 ± 0.3 ppbv per °C for H2 and 19.8 ± 0.4 ppbv per °C for H10, values
418
within ±50% of the H2 trend. For benzene no trend was observed in house H6 in summer but
419
21
a trend was discernable in the winter data. For house H10 a similar temperature trend was
420
observed for the summer and winter data, ~ 0.4 ppbv per °C, about half that observed in H2.
421
For acetaldehyde the lowest 10% of the data defined a trend with temperature in house H6 in
422
winter and in H10 in summer. Acetaldehyde in house H6 was strongly and regularly
423
influenced by events associated with cooking, with values as high as 1.5 ppmv as a 10-minute
424
average, off scale in Figure S-2. The acetaldehyde temperature sensitivity displayed for
425
houses H2 summer, H6 winter, and H10 summer were similar, but clearly for some homes
426
human activity emissions can be much more important source of this compound than
427
emissions from building materials. The building material emissions act as a “baseline” upon
428
which the human activity emission events play out. These human activity sources may
429
overwhelm the more continuous building emission sources, obscuring trends with
430
temperature that are otherwise expected from material emission rates and infiltration rates.
431
Other factors may influence VOC emission rates from materials such as relative
432
humidity, which has been shown to influence VOC emissions in chamber testing of materials.
433
Similar to what was found by Hun et al. [37], no trends between VOC abundance and indoor
434
relative humidity was discernable in our data. Variation in VOC concentrations was much
435
larger than variations in indoor relative humidity for unoccupied homes. For example, for
436
house H2 depicted in Figure 1, the RH varied from 44% to 40%.
437
438
3.3 Formaldehyde emissions from heated gypsum wallboard
439
It has been suggested in the literature that homes with baseboard heating can have
440
higher formaldehyde levels as a result of strong local heating of wall surfaces [35]. H10 had a
441
radiant ceiling heating system. The H10 winter data for formaldehyde and acetaldehyde are
442
shown in Figure 7 and levels were much higher than expected. During this period, the home
443
was heated with the thermostat set at 20 °C (68 °F). The estimated winter infiltration rates
444
22
were larger than summer due to larger indoor-outdoor temperature differences in winter and
445
much higher winter wind speeds (up to 12 m/s) that drove much of the hour-to-hour and day-
446
to-day variability in infiltration rates. Estimated winter infiltration rates ranged between 0.5
447
to 1.5 hr-1 compared to summer’s 0.3 to 0.5 hr-1 range. Despite the greater infiltration rates in
448
winter, formaldehyde levels in H10 were typically higher than in summer, ranging from 30 to
449
35 ppbv with events to 159 ppbv, compared to summers 20 to 45 ppbv range. Winter VOC
450
data did not display a diel pattern as there was no consistent diel variation in ACH. However,
451
indoor formaldehyde and acetaldehyde tracked small variations in indoor air temperature.
452
Increases in indoor air temperature (up to 1 °C) occurred at about 18:00 during week days
453
when home owners returned home from work. Associated with these temperature increases
454
were increases in formaldehyde, acetaldehyde, methanol, and benzene. In other homes, the
455
spikes in acetaldehyde were associated with cooking, but these events in H10 must have an
456
additional source component to explain the covariation with formaldehyde and other VOCs.
457
Interestingly, when these spikes occurred, there was a large response from the ozone analyzer
458
yielding indoor ozone values as high as 143 ppbv for a 1-minute average.
459
23
460
Figure 7. Time series showing winter indoor formaldehyde, acetaldehyde, and ozone mixing
461
ratios in home H10. Middle plot shows 1-minute averages of indoor O3, top plot shows 15-
462
minute averages every 30 minutes of formaldehyde (blue circles) and acetaldehyde (green
463
circles), and 1-minute average indoor air temperature (red trace). Bottom plot shows
464
estimated ACH range due to infiltration (black shading). Highest values of formaldehyde,
465
acetaldehyde, and O3 are shown as numeric labels for events that are off the scale.
466
467
We determined these higher formaldehyde and acetaldehyde levels and O3 spikes
468
were due to the enhanced sources in the home from the ceiling radiant heating system. This
469
commercial heating system consisted of resistive heating wires inside the gypsum wallboard.
470
This type of heating system was commonly installed in Pacific Northwest homes. The
471
homeowner permitted the removal of a sample of the radiant ceiling product from a room
472
24
within the home, and this was tested in the lab. When heated, the sample emitted more
473
formaldehyde, acetaldehyde, and benzene as shown in Figure 8. It is known that temperature
474
can impact the emission rates from building materials by affecting the diffusion coefficient,
475
material/air partition coefficient, and initial emittable concentration [42,43,46,47]. The
476
emission rate in μg hr-1 increased from 1.7 to 4.0 for formaldehyde, from 3.4 to 5.8 for
477
acetaldehyde, and from 0.08 to 1.3 for benzene. SO2 increased as well, but we did not
478
measure SO2 in the homes. A response on the O3 monitor was also observed. The observed
479
increases of the VOCs and O3 response were qualitatively consistent with what was observed
480
in the home. There was no evidence of elevated aromatic compounds which at high
481
concentrations can produce a response in UV absorbance based O3 monitors [70]. The O3
482
response was also observed when we repeated the test using zero air flow containing a 100
483
ppbv NO to titrate O3 in the chamber. The addition of NO made no difference to the
484
measured O3 abundance, and we conclude the O3 monitor was not responding to O3 but to
485
some other contaminant. The O3 monitor uses a low-pressure Hg lamp as the light source,
486
and it is known that such UV absorption based instruments will respond to Hg vapor [71].
487
Tests done with similar instruments revealed that 0.04 ppbv Hg test gas mixtures yielded a 20
488
ppbv O3 response [71]. We conclude that this particular gypsum product is a strong source of
489
formaldehyde, acetaldehyde, and potentially Hg when heated. Since our inlet was positioned
490
about 8 cm from the ceiling surface, our measurements were strongly impacted by this
491
source. Other reports have stated that gypsum wallboard products are not strong sources of
492
mercury in the home [72]. However, those tests did not heat the samples as might occur with
493
resistive heating ceiling products or with baseboard heaters that increase the temperature of
494
gypsum wallboard surfaces. Wallboard emissions as a function of surface temperature should
495
be investigated in the future to provide more information on the role of these materials as
496
sources of mercury, acetaldehyde, formaldehyde, and SO2 in the home.
497
25
498
Figure 8. Testing of the gypsum wallboard radiant ceiling product. Shading indicates the
499
experimental conditions: empty chamber (grey), unheated wallboard (light blue), and heated
500
wallboard (light red). Heating the wallboard causes a response on the O3 monitor and
501
increases in SO2, formaldehyde, acetaldehyde, and benzene.
502
503
504
505
4. Conclusions
506
A strong diel variation in the indoor air concentrations of formaldehyde,
507
acetaldehyde, benzene, and methanol, was observed in unoccupied homes. The variation in
508
26
VOC abundance in the home was caused by time of day variation in infiltration rates. In this
509
study, the infiltration rate variation was principally caused by the time of day variation of the
510
temperature gradient across the building envelope. Maximum VOC concentrations typically
511
occurred in the afternoon when infiltration rates were lowest and minimized during the early
512
morning when they were highest. VOC abundance appears to lag behind changes in
513
infiltration rates so that the correlation between VOC abundance and air change rate (ACH)
514
was poor. The results demonstrate that for homes where infiltration dominates air change
515
rates, the VOC abundance is not in steady state with respect to whole house emissions and
516
ACH. The assumption of steady state is common practice to calculate whole house emission
517
rates from measured indoor air concentrations; this is probably a poor assumption in many
518
cases. We conclude that whole house emission rates need to be determined for homes where
519
constant mechanical ventilation provides the air flow to avoid the complication of the
520
diurnally varying infiltrations rates and resulting non-steady state VOC concentrations.
521
Formaldehyde and other VOCs were positively correlated with indoor temperature.
522
For one home where ACH was relatively constant and determined by infiltration, a warming
523
weather trend over the week resulted in increasing indoor levels of formaldehyde that tracked
524
increasing indoor air temperature. For this home a whole house formaldehyde emission rate
525
sensitivity of 4.6 ppbv per °C was determined. In general, indoor summertime formaldehyde
526
levels displayed a positive correlation with indoor temperature that ranged from 3.0 ppbv
527
per °C to 4.5 ppbv per °C, accounting for building materials emissions sensitivity to
528
temperature and the impact of indoor temperature on infiltration rates. This is a useful metric
529
for predicting changes in the indoor abundance of formaldehyde to changes in home
530
temperature as a result of thermostat set points, heat wave impacts, and potential impacts of
531
regional climate change on indoor air quality.
532
533
27
Acknowledgments
534
This publication was developed under Assistance Agreement No. RD-83575601
535
awarded by the U.S. Environmental Protection Agency to Washington State University. It has
536
not been formally reviewed by EPA. The views expressed in this document are solely those
537
of the authors and do not necessarily reflect those of the Agency. EPA does not endorse any
538
products or commercial services mentioned in this publication.
539
540
Appendices
541
Additional figures may be found in the supplementary material.
542
Figure S-1. Indoor extra methanol levels (indoor-outdoor) as a function of indoor air
543
temperature for homes H2, H6, and H10.
544
Figure S-2. Indoor extra acetaldehyde levels (indoor-outdoor) as a function of indoor air
545
temperature for homes H2, H6, and H10.
546
Figure S-3. Indoor extra benzene levels (indoor-outdoor) as a function of indoor air
547
temperature for homes H2, H6, and H10.
548
549
28
References
550
[1] P.J. Landrigan, R. Fuller, N.J.R. Acosta, O. Adeyi, R. Arnold, N. Basu, A.B. Baldé, R.
551
Bertollini, S. Bose-O’Reilly, J.I. Boufford, P.N. Breysse, T. Chiles, C. Mahidol, A.M.
552
Coll-Seck, M.L. Cropper, J. Fobil, V. Fuster, M. Greenstone, A. Haines, D. Hanrahan,
553
D. Hunter, M. Khare, A. Krupnick, B. Lanphear, B. Lohani, K. Martin, K. V.
554
Mathiasen, M.A. McTeer, C.J.L. Murray, J.D. Ndahimananjara, F. Perera, J. Potočnik,
555
A.S. Preker, J. Ramesh, J. Rockström, C. Salinas, L.D. Samson, K. Sandilya, P.D. Sly,
556
K.R. Smith, A. Steiner, R.B. Stewart, W.A. Suk, O.C.P. van Schayck, G.N. Yadama,
557
K. Yumkella, M. Zhong, The Lancet Commission on pollution and health, Lancet.
558
(2017). doi:10.1016/S0140-6736(17)32345-0.
559
[2] J.L.M. Rios, J.L. Boechat, M.C.K. Ramos, R.R. Luiz, F.R. Aquino Neto, J.R. Lapa e
560
Silva, Asthma Among Office Workers And Exposure To Chemical And Biological
561
Indoor Pollutants, J. Allergy Clin. Immunol. 127 (2011) AB95AB95.
562
doi:10.1016/j.jaci.2010.12.385.
563
[3] L.A.M. Smit, V. Lenters, B.B. Høyer, C.H. Lindh, H.S. Pedersen, I. Liermontova,
564
B.A.G. Jönsson, A.H. Piersma, J.P. Bonde, G. Toft, R. Vermeulen, D. Heederik,
565
Prenatal exposure to environmental chemical contaminants and asthma and eczema in
566
school-age children, Allergy. 70 (2015) 653660. doi:10.1111/all.12605.
567
[4] K.W. Kim, Y.L. Won, K.S. Ko, K.H. Heo, Y.H. Chung, The effects of hazardous
568
chemical exposure on cardiovascular disease in chemical products manufacturing
569
workers, Toxicol. Res. 28 (2012) 269277. doi:10.5487/TR.2012.28.4.269.
570
[5] P. MacNaughton, U. Satish, J.G.C. Laurent, S. Flanigan, J. Vallarino, B. Coull, J.D.
571
Spengler, J.G. Allen, The impact of working in a green certified building on cognitive
572
function and health, Build. Environ. 114 (2017) 178186.
573
doi:10.1016/j.buildenv.2016.11.041.
574
[6] A. Clifford, L. Lang, R. Chen, K.J. Anstey, A. Seaton, Exposure to air pollution and
575
cognitive functioning across the life course - A systematic literature review, Environ.
576
Res. 147 (2016) 383398. doi:10.1016/j.envres.2016.01.018.
577
[7] X. Zhang, X. Chen, X. Zhang, The impact of exposure to air pollution on cognitive
578
performance, Proc. Natl. Acad. Sci. 115 (2018) 91939197.
579
doi:10.1073/pnas.1809474115.
580
[8] J.G. Allen, P. MacNaughton, U. Satish, S. Santanam, J. Vallarino, J.D. Spengler,
581
Associations of cognitive function scores with carbon dioxide, ventilation, and volatile
582
organic compound exposures in office workers: A controlled exposure study of green
583
and conventional office environments, Environ. Health Perspect. 124 (2016) 805812.
584
doi:10.1289/ehp.1510037.
585
[9] E. Grossman, Time after time: Environmental influences on the aging brain, Environ.
586
Health Perspect. 122 (2014) A238A243. doi:10.1289/ehp.122-A238.
587
[10] J. Gunschera, S. Mentese, T. Salthammer, J.R. Andersen, Impact of building materials
588
on indoor formaldehyde levels: Effect of ceiling tiles, mineral fiber insulation and
589
29
gypsum board, Build. Environ. 64 (2013) 138145.
590
doi:10.1016/j.buildenv.2013.03.001.
591
[11] C. Jiang, D. Li, P. Zhang, J. Li, J. Wang, J. Yu, Formaldehyde and volatile organic
592
compound (VOC) emissions from particleboard: Identification of odorous compounds
593
and effects of heat treatment, Build. Environ. 117 (2017) 118126.
594
doi:10.1016/j.buildenv.2017.03.004.
595
[12] D.A. Missia, E. Demetriou, N. Michael, E.I. Tolis, J.G. Bartzis, Indoor exposure from
596
building materials: A field study, Atmos. Environ. 44 (2010) 43884395.
597
doi:10.1016/j.atmosenv.2010.07.049.
598
[13] M. Böhm, M.Z.M. Salem, J. Srba, Formaldehyde emission monitoring from a variety
599
of solid wood, plywood, blockboard and flooring products manufactured for building
600
and furnishing materials, J. Hazard. Mater. (2012). doi:10.1016/j.jhazmat.2012.04.013.
601
[14] S. Hormigos-Jimenez, M.Á. Padilla-Marcos, A. Meiss, R.A. Gonzalez-Lezcano, J.
602
Feijó-Muñoz, Ventilation rate determination method for residential buildings
603
according to TVOC emissions from building materials, Build. Environ. 123 (2017)
604
555563. doi:10.1016/j.buildenv.2017.07.032.
605
[15] X. Liu, M.A. Mason, Z. Guo, K.A. Krebs, N.F. Roache, Source emission and model
606
evaluation of formaldehyde from composite and solid wood furniture in a full-scale
607
chamber, Atmos. Environ. 122 (2015) 561568. doi:10.1016/j.atmosenv.2015.09.062.
608
[16] D. Bourdin, P. Mocho, V. Desauziers, H. Plaisance, Formaldehyde emission behavior
609
of building materials: On-site measurements and modeling approach to predict indoor
610
air pollution, J. Hazard. Mater. 280 (2014) 164173.
611
doi:10.1016/j.jhazmat.2014.07.065.
612
[17] Y.H. Cheng, C.C. Lin, S.C. Hsu, Comparison of conventional and green building
613
materials in respect of VOC emissions and ozone impact on secondary carbonyl
614
emissions, Build. Environ. 87 (2015) 274282. doi:10.1016/j.buildenv.2014.12.025.
615
[18] E. Uhde, N. Schulz, Impact of room fragrance products on indoor air quality, Atmos.
616
Environ. 106 (2015) 492502. doi:10.1016/j.atmosenv.2014.11.020.
617
[19] W.W. Nazaroff, C.J. Weschler, Cleaning products and air fresheners: Exposure to
618
primary and secondary air pollutants, Atmos. Environ. 38 (2004) 28412865.
619
doi:10.1016/j.atmosenv.2004.02.040.
620
[20] A. Manoukian, E. Quivet, B. Temime-Roussel, M. Nicolas, F. Maupetit, H. Wortham,
621
Emission characteristics of air pollutants from incense and candle burning in indoor
622
atmospheres, Environ. Sci. Pollut. Res. 20 (2013) 46594670. doi:10.1007/s11356-
623
012-1394-y.
624
[21] A. Manoukian, D. Buiron, B. Temime-Roussel, H. Wortham, E. Quivet, Measurements
625
of VOC/SVOC emission factors from burning incenses in an environmental test
626
chamber: influence of temperature, relative humidity, and air exchange rate, Environ.
627
Sci. Pollut. Res. 23 (2016) 63006311. doi:10.1007/s11356-015-5819-2.
628
30
[22] A. Steinemann, Volatile emissions from common consumer products, Air Qual.
629
Atmos. Heal. (2015). doi:10.1007/s11869-015-0327-6.
630
[23] A. Steinemann, Ten questions concerning air fresheners and indoor built
631
environments, Build. Environ. 111 (2017) 279284.
632
doi:10.1016/j.buildenv.2016.11.009.
633
[24] F. Klein, S.M. Platt, N.J. Farren, A. Detournay, E.A. Bruns, C. Bozzetti, K.R.
634
Daellenbach, D. Kilic, N.K. Kumar, S.M. Pieber, J.G. Slowik, B. Temime-Roussel, N.
635
Marchand, J.F. Hamilton, U. Baltensperger, A.S.H. Prévôt, I. El Haddad,
636
Characterization of Gas-Phase Organics Using Proton Transfer Reaction Time-of-
637
Flight Mass Spectrometry: Cooking Emissions, Environ. Sci. Technol. 50 (2016)
638
12431250. doi:10.1021/acs.est.5b04618.
639
[25] H. Wang, Z. Xiang, L. Wang, S. Jing, S. Lou, S. Tao, J. Liu, M. Yu, L. Li, L. Lin, Y.
640
Chen, A. Wiedensohler, C. Chen, Emissions of volatile organic compounds (VOCs)
641
from cooking and their speciation: A case study for Shanghai with implications for
642
China, Sci. Total Environ. 621 (2018) 13001309.
643
doi:10.1016/j.scitotenv.2017.10.098.
644
[26] S. Cheng, G. Wang, J. Lang, W. Wen, X. Wang, S. Yao, Characterization of volatile
645
organic compounds from different cooking emissions, Atmos. Environ. 145 (2016)
646
299307. doi:10.1016/j.atmosenv.2016.09.037.
647
[27] K.H. Kim, S.K. Pandey, E. Kabir, J. Susaya, R.J.C. Brown, The modern paradox of
648
unregulated cooking activities and indoor air quality, J. Hazard. Mater. 195 (2011) 1
649
10. doi:10.1016/j.jhazmat.2011.08.037.
650
[28] R. Fortmann, P. Kariher, R. Clayton, Indoor Air Quality: Residential Cooking
651
Exposures, State Calif. Air Resour. Board. (2001).
652
[29] J.M. Logue, T.E. Mckone, M.H. Sherman, B.C. Singer, Hazard assessment of
653
chemical air contaminants measured in residences, Indoor Air. 21 (2011) 92109.
654
doi:10.1111/j.1600-0668.2010.00683.x.
655
[30] G.C. Morrison, W.W. Nazaroff, Ozone interactions with carpet: Secondary emissions
656
of aldehydes, Environ. Sci. Technol. 36 (2002) 21852192. doi:10.1021/es0113089.
657
[31] A. Wisthaler, C.J. Weschler, Reactions of ozone with human skin lipids: Sources of
658
carbonyls, dicarbonyls, and hydroxycarbonyls in indoor air, Proc. Natl. Acad. Sci. 107
659
(2010) 65686575. doi:10.1073/pnas.0904498106.
660
[32] P.S.J. Lakey, A. Wisthaler, T. Berkemeier, T. Mikoviny, U. Pöschl, M. Shiraiwa,
661
Chemical kinetics of multiphase reactions between ozone and human skin lipids:
662
Implications for indoor air quality and health effects, Indoor Air. 27 (2017) 816828.
663
doi:10.1111/ina.12360.
664
[33] M.S. Waring, J.R. Wells, Volatile organic compound conversion by ozone, hydroxyl
665
radicals, and nitrate radicals in residential indoor air: Magnitudes and impacts of
666
oxidant sources, Atmos. Environ. 106 (2015) 382391.
667
31
doi:10.1016/j.atmosenv.2014.06.062.
668
[34] N.L. Gilbert, D. Gauvin, M. Guay, M.È. Héroux, G. Dupuis, M. Legris, C.C. Chan,
669
R.N. Dietz, B. Lévesque, Housing characteristics and indoor concentrations of
670
nitrogen dioxide and formaldehyde in Quebec City, Canada, Environ. Res. 102 (2006)
671
18. doi:10.1016/j.envres.2006.02.007.
672
[35] N.L. Gilbert, M. Guay, D. Gauvin, R.N. Dietz, C.C. Chan, B. Lévesque, Air change
673
rate and concentration of formaldehyde in residential indoor air, Atmos. Environ. 42
674
(2008) 24242428. doi:10.1016/j.atmosenv.2007.12.017.
675
[36] E.L. Hult, H. Willem, P.N. Price, T. Hotchi, M.L. Russell, B.C. Singer, Formaldehyde
676
and acetaldehyde exposure mitigation in US residences: In-home measurements of
677
ventilation control and source control, Indoor Air. 25 (2015) 523535.
678
doi:10.1111/ina.12160.
679
[37] D.E. Hun, R.L. Corsi, M.T. Morandi, J.A. Siegel, Formaldehyde in residences: Long-
680
term indoor concentrations and influencing factors, Indoor Air. 20 (2010) 196203.
681
doi:10.1111/j.1600-0668.2010.00644.x.
682
[38] A.T. Hodgson, D. Faulkner, D.P. Sullivan, D.L. DiBartolomeo, M.L. Russell, W.J.
683
Fisk, Effect of outside air ventilation rate on volatile organic compound concentrations
684
in a call center, Atmos. Environ. 37 (2003) 55175527.
685
doi:10.1016/j.atmosenv.2003.09.028.
686
[39] M.C.P. Moura, S.J. Smith, D.B. Belzer, 120 years of U.S. Residential housing stock
687
and floor space, PLoS One. 10 (2015) 118. doi:10.1371/journal.pone.0134135.
688
[40] D.G. Poppendieck, H.F. Hubbard, C.J. Weschler, R.L. Corsi, Formation and emissions
689
of carbonyls during and following gas-phase ozonation of indoor materials, Atmos.
690
Environ. 41 (2007) 76147626. doi:10.1016/j.atmosenv.2007.05.049.
691
[41] Y.K. Lee, H.J. Kim, The effect of temperature on VOCs and carbonyl compounds
692
emission from wooden flooring by thermal extractor test method, Build. Environ. 53
693
(2012) 9599. doi:10.1016/j.buildenv.2011.10.016.
694
[42] J. Xiong, Y. Zhang, Impact of temperature on the initial emittable concentration of
695
formaldehyde in building materials: Experimental observation, Indoor Air. 20 (2010)
696
523529. doi:10.1111/j.1600-0668.2010.00675.x.
697
[43] J. Xiong, W. Wei, S. Huang, Y. Zhang, Association between the emission rate and
698
temperature for chemical pollutants in building materials: General correlation and
699
understanding, Environ. Sci. Technol. 47 (2013) 85408547. doi:10.1021/es401173d.
700
[44] J. Xiong, P. Zhang, S. Huang, Y. Zhang, Comprehensive influence of environmental
701
factors on the emission rate of formaldehyde and VOCs in building materials:
702
Correlation development and exposure assessment, Environ. Res. 151 (2016) 734741.
703
doi:10.1016/j.envres.2016.09.003.
704
[45] S. Parthasarathy, R.L. Maddalena, M.L. Russell, M.G. Apte, Effect of temperature and
705
32
humidity on formaldehyde emissions in temporary housing units, J. Air Waste Manag.
706
Assoc. 61 (2011) 689695. doi:10.3155/1047-3289.61.6.689.
707
[46] Y. Zhang, X. Luo, X. Wang, K. Qian, R. Zhao, Influence of temperature on
708
formaldehyde emission parameters of dry building materials, Atmos. Environ. 41
709
(2007) 32033216. doi:10.1016/j.atmosenv.2006.10.081.
710
[47] Q. Deng, X. Yang, J. Zhang, Study on a new correlation between diffusion coefficient
711
and temperature in porous building materials, Atmos. Environ. 43 (2009) 20802083.
712
doi:10.1016/j.atmosenv.2008.12.052.
713
[48] X. Dai, J. Liu, Y. Yin, X. Song, S. Jia, Modeling and controlling indoor formaldehyde
714
concentrations in apartments: On-site investigation in all climate zones of China,
715
Build. Environ. 127 (2018) 98106. doi:10.1016/j.buildenv.2017.10.036.
716
[49] M. Guo, X. Pei, F. Mo, J. Liu, X. Shen, Formaldehyde concentration and its
717
influencing factors in residential homes after decoration at Hangzhou, China, J.
718
Environ. Sci. (China). 25 (2013) 908915. doi:10.1016/S1001-0742(12)60170-3.
719
[50] S. Huang, W. Wei, L.B. Weschler, T. Salthammer, H. Kan, Z. Bu, Y. Zhang, Indoor
720
formaldehyde concentrations in urban China: Preliminary study of some important
721
influencing factors, Sci. Total Environ. 590591 (2017) 394405.
722
doi:10.1016/j.scitotenv.2017.02.187.
723
[51] C. Zhou, Y. Zhan, S. Chen, M. Xia, C. Ronda, M. Sun, H. Chen, X. Shen, Combined
724
effects of temperature and humidity on indoor VOCs pollution: Intercity comparison,
725
Build. Environ. 121 (2017) 2634. doi:10.1016/j.buildenv.2017.04.013.
726
[52] ASTM, Building Air Change Rate and Infiltration Measurements, Philadelphia,
727
Pennsylvania, 1980.
728
[53] B. Lamb, H. Westberg, P. Bryant, J. Dean, S. Mullins, Air infiltration rates in pre- and
729
post weatherized houses, J. Air Pollut. Control Assoc. 35 (1985) 545550.
730
doi:10.1080/00022470.1985.10465931.
731
[54] A.K. Persily, A. Musser, D.D. Leber, A Collection of Homes to Represent the U.S.
732
Housing Stock, NIST Interagency/Internal Report (NISTIR) - 7330, Washington, D.C.,
733
2006.
734
[55] W.M. Kirk, M. Fuchs, Y. Huangfu, N. Lima, P. O’Keeffe, B. Lin, T. Jobson, S.
735
Pressley, V. Walden, D. Cook, B.K. Lamb, Indoor air quality and wildfire smoke
736
impacts in the Pacific Northwest, Sci. Technol. Built Environ. 24 (2018) 149159.
737
doi:10.1080/23744731.2017.1393256.
738
[56] EIA (United States Department of Energy Energy Information Administration),
739
Residential Energy Consumption Survey (RECS), (2017).
740
https://www.eia.gov/consumption/residential/ (accessed June 20, 2018).
741
[57] B. Lin, Y. Huangfu, N. Lima, B. Jobson, M. Kirk, P. O’Keeffe, S. Pressley, V.
742
Walden, B. Lamb, D. Cook, Analyzing the Relationship between Human Behavior and
743
33
Indoor Air Quality, J. Sens. Actuator Networks. 6 (2017) 13.
744
doi:10.3390/jsan6030013.
745
[58] B.T. Jobson, J.K. McCoskey, Sample drying to improve HCHO measurements by
746
PTR-MS instruments: Laboratory and field measurements, Atmos. Chem. Phys. 10
747
(2010) 18211835. doi:10.5194/acp-10-1821-2010.
748
[59] M. Gueneron, M.H. Erickson, G.S. Vanderschelden, B.T. Jobson, PTR-MS
749
fragmentation patterns of gasoline hydrocarbons, Int. J. Mass Spectrom. 379 (2015)
750
97109. doi:10.1016/j.ijms.2015.01.001.
751
[60] S. Dedesko, B. Stephens, J.A. Gilbert, J.A. Siegel, Methods to assess human
752
occupancy and occupant activity in hospital patient rooms, Build. Environ. 90 (2015)
753
136145. doi:10.1016/j.buildenv.2015.03.029.
754
[61] ANSI/ASHRAE, ASHRAE Standard 62-2013, Ventilation for Acceptable Indoor Air
755
Quality, Am. Soc. Heating, Refrig. Air-Conditioning Eng. Atlanta GA. 25 (2013) 34.
756
doi:ANSI/ASHRAE Standard 62.1-2004.
757
[62] C. Toro, B.T. Jobson, L. Haselbach, S. Shen, S.H. Chung, Photoactive roadways:
758
Determination of CO, NO and VOC uptake coefficients and photolabile side product
759
yields on TiO2treated asphalt and concrete, Atmos. Environ. 139 (2016) 3745.
760
doi:10.1016/j.atmosenv.2016.05.007.
761
[63] G.W. Traynor, D.W. Anthon, C.D. Hollowell, Technique for determining pollutant
762
emissions from a gas-fired range, Atmos. Environ. 16 (1982) 29792987.
763
doi:10.1016/0004-6981(82)90049-X.
764
[64] H. Plaisance, A. Blondel, V. Desauziers, P. Mocho, Field investigation on the removal
765
of formaldehyde in indoor air, Build. Environ. 70 (2013) 277283.
766
doi:10.1016/j.buildenv.2013.08.032.
767
[65] A.T. Hodgson, A.F. Rudd, D. Beal, S. Chandra, Volatile Organic Compound
768
Concentrations and Emission Rates in New Manufactured and Site-Built Houses,
769
Indoor Air. 10 (2000) 178192. doi:10.1034/j.1600-0668.2000.010003178.x.
770
[66] A.T. Hodgson, D. Beal, J.E.R. McIlvaine, Sources of formaldehyde, other aldehydes
771
and terpenes in a new manufactured house, Indoor Air. 12 (2002) 235242.
772
doi:10.1034/j.1600-0668.2002.01129.x.
773
[67] F.J. Offermann, A.T. Hodgson, Emission rates of volatile organic compounds in new
774
homes, 12th Int. Conf. Indoor Air Qual. Clim. 2011. 1 (2011) 177182.
775
http://www.scopus.com/inward/record.url?eid=2-s2.0-
776
84880558115&partnerID=tZOtx3y1 (accessed August 24, 2018).
777
[68] D.G. Poppendieck, L.C. Ng, A.K. Persily, A.T. Hodgson, Long term air quality
778
monitoring in a net-zero energy residence designed with low emitting interior
779
products, Build. Environ. 94 (2015) 3342. doi:10.1016/j.buildenv.2015.07.001.
780
[69] S.H. Shin, W.K. Jo, Volatile organic compound concentrations, emission rates, and
781
34
source apportionment in newly-built apartments at pre-occupancy stage, Chemosphere.
782
89 (2012) 569578. doi:10.1016/j.chemosphere.2012.05.054.
783
[70] E.J. Williams, F.C. Fehsenfeld, B.T. Jobson, W.C. Kuster, P.D. Goldan, J. Stutz, W.A.
784
McClenny, Comparison of Ultraviolet and DOAS Instruments for Ambient Ozone
785
Monitoring, Environ. Sci. Technol. 40 (2006) 57555762. doi:10.1021/es0523542.
786
[71] EPA (U.S. Environmental Protection Agency), Laboratory Study to Explore Potential
787
Interferences to Air Quality Monitors, (1999).
788
https://www3.epa.gov/ttnamti1/archive/files/ambient/criteria/reldocs/finalreport.pdf
789
(accessed December 1, 2016).
790
[72] S.S. Shock, J.J. Noggle, N. Bloom, L.J. Yost, Evaluation of potential for mercury
791
volatilization from natural and FGD gypsum products using flux-chamber tests,
792
Environ. Sci. Technol. 43 (2009) 22822287. doi:10.1021/es802872n.
793
794
... Air infiltration acts like a double-edged sword with both advantages and disadvantages. On the one hand, it can be regarded as a source of outdoor air in buildings without mechanical ventilation systems (such as average residences) to remove inner heats [2] and dilute indoor CO 2 /air pollutants generated by occupants/furniture [3,4], which is also known as natural ventilation. On the other hand, excessive unorganized air infiltration also leads to high energy consumption for space heating/cooling [5,6] and intake of outdoor air pollutants (such as particulate matter [7], PAHs [8], fungal/bacterial [9], etc.). ...
... 13,14 In addition to these pollutants, also for example, ultrafine particles (UFPs), volatile organic compounds (VOCs), aldehydes (eg, formaldehyde (HCHO)), ozone (O 3 ), and bioparticles (fungi and bacteria)) could be cause for health effects in some sports facilities. Besides, important comfort parameters-temperature (T) and relative humidity (RH)-may affect for example material emissions [15][16][17] and occupant's perception of the indoor air quality (IAQ) 18 and should be taken into account when evaluating exposure issues in sports facilities. ...
Article
Full-text available
The aim of this review was to investigate human exposure to relevant indoor air contaminants, predictors affecting the levels, and the means to reduce the harmful exposure in indoor sports facilities. Our study revealed that the contaminants of primary concern are the following: particulate matter in indoor climbing‐, golf‐, and horse riding facilities; carbon dioxide and particulate matter in fitness centers, gymnasiums and sports halls; Staphylococci on gymnasium surfaces; nitrogen dioxide and carbon monoxide in ice hockey arenas; carbon monoxide, nitrogen oxide(s), and particulate matter in motor sports arenas; and disinfection byproducts, in indoor chlorinated swimming pools. Means to reduce human exposure to indoor contaminants include the following: adequate mechanical ventilation with filters, suitable cleaning practices, a limited number of occupants in fitness centers and gymnasiums, the use of electric resurfacers instead of the engine powered resurfacers in ice hockey arenas, carefully regulated chlorine and temperature levels in indoor swimming pools, properly ventilated pools, and good personal hygiene. Because of the large number of susceptible people in these facilities, as well as all active people having an increased respiratory rate and airflow velocity, strict air quality requirements in indoor sports facilities should be maintained.
Article
Indian cities have some of the poorest air quality globally but volatile organic compounds (VOCs)—many of which adversely affect health—and their indoor sources remain understudied in India. In this pilot study we quantified hundreds of VOCs inside and outside 26 homes in Ahmedabad and Gandhinagar, Gujarat, in May 2019 and in January 2020. We sampled in the morning and afternoon/evening to capture temporal variability. Total indoor VOCs were measured at higher concentrations in winter (327.0 ± 224.2 µgm−3) than summer (150.1 ± 121.0 µgm−3) and exceeded those measured outdoors. Using variable reduction techniques, we identified potential sources of compounds (cooking, plastics [with an emphasis on plasticizers], consumer products, siloxanes [as used in the production of consumer products], vehicles). Contributions differed by season and between homes. In May, when temperatures were high, plastics contributed substantially to indoor pollution (mean of 42% contribution to total VOCs) as compared to in January (mean of 4%). Indoor cooking and consumer products contributed on average 29% and 10% to all VOCs indoors in January and 16% and 4% in May. Siloxane sources contributed <4% to any home during either season. Cooking contributed substantially to outdoor VOCs (on average 18% in January and 11% in May) and vehicle-related sources accounted for up to 84% of VOCs in some samples. Overall, results indicate a strong seasonal dependence of indoor VOC concentrations and sources, underscoring the need to better understand factors driving health-harming pollutants inside homes to facilitate exposure reductions.
Article
This study was conducted to determine the exposure and health risk to cooking fumes of a total of 88 volunteer kitchen staff aged between 18 and 65 years working in five different kitchens in Ankara. Gas‐ and particle‐phase polycyclic aromatic hydrocarbons (PAHs), and volatile organic compound (VOCs) concentrations were evaluated in the indoor air of 5 kitchens. Serum malondialdehyde (MDA) and superoxide dismutase (SOD) levels were analyzed to determine the oxidative damage as a result of the exposure to cooking fumes among the cooks and waiters. Significant positive relationships were found between serum MDA levels of the hot kitchen workers and indoor chrysene (Chr), indeno(1,2,3‐c,d)pyrene (Ind), and total VOC levels. Although the carcinogenic risks estimated for the exposed population were between the acceptable/tolerable levels, the hazard quotient (HQ) estimated for the exposure to indoor benzene exceeded the safe level. The results of the study revealed that exposure to organic pollutants in indoor air may be a risk factor for the development of oxidative stress, especially in hot kitchen workers. The importance of efficient ventilation in the kitchen has been pointed out to reduce health risks caused by cooking fumes.
Article
From the thermodynamic perspective, the term temperature is clearly defined for ideal physical systems: A unique temperature can be assigned to each black body via its radiation spectrum, and the temperature of an ideal gas is given by the velocity distribution of the molecules. While the indoor environment is not an ideal system, fundamental physical and chemical processes, such as diffusion, partitioning equilibria, and chemical reactions, are predictably temperature‐dependent. For example, the logarithm of reaction rate and equilibria constants are proportional to the reciprocal of the absolute temperature. It is therefore possible to have non‐linear, very steep changes in chemical phenomena over a relatively small temperature range. On the contrary, transport processes are more influenced by spatial temperature, momentum, and pressure gradients as well as by the density, porosity, and composition of indoor materials. Consequently, emergent phenomena, such as emission rates or dynamic air concentrations, can be the result of complex temperature‐dependent relationships that require a more empirical approach. Indoor environmental conditions are further influenced by the thermal comfort needs of occupants. Not only do occupants have to create thermal conditions that serve to maintain their core body temperature, which is usually accomplished by wearing appropriate clothing, but also the surroundings must be adapted so that they feel comfortable. This includes the interaction of the living space with the ambient environment, which can vary greatly by region and season. Design of houses, apartments, commercial buildings, and schools is generally utility and comfort driven, requiring an appropriate energy balance, sometimes considering ventilation but rarely including the impact of temperature on indoor contaminant levels. In our article, we start with a review of fundamental thermodynamic variables and discuss their influence on typical indoor processes. Then, we describe the heat balance of people in their thermal environment. An extensive literature study is devoted to the thermal conditions in buildings, the temperature‐dependent release of indoor pollutants from materials and their distribution in the various interior compartments as well as aspects of indoor chemistry. Finally, we assess the need to consider temperature holistically with regard to the changes to be expected as a result of global emergencies such as climate change.
Article
Purpose New Zealand’s historical housing stock comprises largely single-storey detached houses, characterised by poor winter comfort with high air infiltration. Challenges with affordability and land use are shifting New Zealand’s housing stock towards double-storey, conjoined medium-density housing (MDH). Reduced external surfaces in this typology should reduce winter heat loss and infiltration, improving winter comfort and health. New concerns arise, however, regarding summertime overheating and poor indoor air quality. Design/methodology/approach A field study was undertaken where temperature, humidity, airtightness, particulate matter (PM) and total volatile organic compounds (TVOC) were measured in two unoccupied, newly built double-storey, conjoined houses, for several weeks over summer. Findings The reduced surface area of this typology did not reduce infiltration and demonstrated significant periods of overheating. Internal PM concentrations generally exceeded outdoor concentrations but did not exceed annual average outdoor PM10 guidelines of 20 µg m-3. Infiltration factors (Finf) were closer to more traditional houses. TVOC readings varied widely, but frequently exceeded international guidelines. Research limitations/implications The small sample limits the applications of conclusions more widely. Recommendations to investigate a wider sample in different locations with more detailed VOC analysis over all seasons are made. Practical implications Improvements to internal environments cannot be guaranteed by housing typology changes alone and must still involve thoughtful environmental design. Social implications Housing typology changes may not improve internal living environments. Originality/value A move to the new MDH typology may not achieve expectations of airtightness and thermal improvement. New challenges arise from significant overheating and high TVOC levels, which may lead to new negative health effects.
Article
Quantifying speciated concentrations and emissions of volatile organic compounds (VOCs) is critical to understanding the processes that control indoor VOC dynamics, airborne chemistry, and human exposures. Here, we present source strength profiles from the HOMEChem study, quantifying speciated VOC emissions from scripted experiments (with multiple replicates) of cooking, cleaning, and human occupancy and from unperturbed baseline measurements of the building and its contents. Measurements using a proton transfer reaction time‐of‐flight mass spectrometer were combined with tracer‐based determinations of air‐change rates to enable mass‐balance‐based calculations of speciated, time‐resolved VOC source strengths. The building and its contents were the dominant emission source into the house, with large emissions of acetic acid, methanol, and formic acid. Cooking emissions were greater than cleaning emissions and were dominated by ethanol. Bleach cleaning generated high emissions of chlorinated compounds, whereas natural product cleaning emitted predominantly terpenoids. Occupancy experiments showed large emissions of siloxanes from personal care products in the morning, with much lower emissions in the afternoon. From these results, VOC emissions were simulated for a hypothetical 24‐h period, showing that emissions from the house and its contents make up nearly half of total indoor VOC emissions.
Article
In this study, a large-scale in-cabin benzene series hazard detection is firstly performed on 20 electric buses by a full-scale climate chamber. The sources of BTEX are analyzed deeply by parts detection, and a series of effective measures are performed to reduce BTEX. Firstly, the in-cabin BTEX pollution with considerations of a series of parameters, such as interior configuration, environment temperature, vehicle age, and ventilation mode, is analyzed. The result shows that: 1) The VOCs concentrations decrease with vehicle age, higher configuration level and better ventilation system (particularly, fresh wind mode reduce VOCs fastly), while increases with environment temperature; 2) BTEX in bus cabins occupy approximatively 70.1% of TVOC, thus the BTEX overproof is the main culprit which causes VOCs to exceed standard. Then, measurements on components/materials VOCs releases were performed in a small climate chamber to discriminate key species and their sources. Xylene released from glues materials is found as a key species that causes BTEX/VOC to exceed limitation. Lastly, some measures, such as optimizations of materials selection and manufacturing crafts, are adopted to improve in-cabin pollution, and positive effects are obtained. For example, ethylbenzene and xylene released from HL 125 (a polyurethane adhesive) decrease by 2456% and 1930% respectively after improvement. And in-cabin xylene and TVOC decrease by 2274% and 222%, respectively, and all of them are lower than limitation value.
Article
This study monitored for the first time the occurrence of a wide range of very volatile organic compounds (VVOCs) as emissions from wooden materials and in indoor air of prefabricated wooden houses non-occupied and occupied. Measurements were performed by using a newly developed method which enables a broadening of the analytical spectrum beyond ≤ C6. It is shown that even wooden materials are emission sources of very volatile species, but without playing a dominant role regarding VVOC concentrations in indoor air of new prefabricated wooden houses. C4- and C5-alkanes were identified as most abundant substances which are presumably released as propellants from insulation materials. Regarding propane-1,2-diol, acetaldehyde and C1–C8-carboxylic acids, German indoor guide values were exceeded in two out of four houses at delivery and/or under occupancy. In one house with natural ventilation, acetaldehyde concentrations were higher than the OEHHA reference exposure level at delivery before ventilation. The findings show that instead of formaldehyde both acetaldehyde and carboxylic acids will move into the focus of attention due to recent regulations. The results also underline that even with a mechanical ventilation system and, thus, at increased air exchange rates, a hygienic, harmless indoor air quality cannot always be achieved. The appropriate selection and application of construction materials is still important in order to obtain an acceptable indoor air quality.
Article
This paper presents pollutant concentrations and performance data for code‐required mechanical ventilation equipment in 23 low‐income apartments at 4 properties constructed or renovated 2013‐2017. All apartments had natural gas cooking burners. Occupants pledged to not use windows for ventilation during the study but several did. Measured airflows of range hoods and bathroom exhaust fans were lower than product specifications. Only eight apartments operationally met all ventilation code requirements. Pollutants measured over one week in each apartment included time‐resolved fine particulate matter (PM2.5), nitrogen dioxide (NO2), formaldehyde and carbon dioxide (CO2) and time‐integrated formaldehyde, NO2 and nitrogen oxides (NOX). Compared to a recent study of California houses with code‐compliant ventilation, apartments were smaller, had fewer occupants, higher densities, and higher mechanical ventilation rates. Mean PM2.5, formaldehyde, NO2, and CO2 were 7.7 µg/m³, 14.1, 18.8, and 741 ppm in apartments; these are 4% lower, 25% lower, 165% higher, and 18% higher compared to houses with similar cooking frequency. Four apartments had weekly PM2.5 above the California annual outdoor standard of 12 µg/m³ and also discrete days above the World Health Organization 24‐hour guideline of 25 µg/m³. Two apartments had weekly NO2 above the California annual outdoor standard of 30 ppb.
Article
Full-text available
This paper examines the effect of both cumulative and transitory exposures to air pollution for the same individuals over time on cognitive performance by matching a nationally representative longitudinal survey and air quality data in China according to the exact time and geographic locations of the cognitive tests. We find that long-term exposure to air pollution impedes cognitive performance in verbal and math tests. We provide evidence that the effect of air pollution on verbal tests becomes more pronounced as people age, especially for men and the less educated. The damage on the aging brain by air pollution likely imposes substantial health and economic costs, considering that cognitive functioning is critical for the elderly for both running daily errands and making high-stake decisions.
Article
Full-text available
In the coming decades, as we experience global population growth and global aging issues, there will be corresponding concerns about the quality of the air we experience inside and outside buildings. Because we can anticipate that there will be behavioral changes that accompany population growth and aging, we examine the relationship between home occupant behavior and indoor air quality. To do this, we collect both sensor-based behavior data and chemical indoor air quality measurements in smart home environments. We introduce a novel machine learning-based approach to quantify the correlation between smart home features and chemical measurements of air quality, and evaluate the approach using two smart homes. The findings may help us understand the types of behavior that measurably impact indoor air quality. This information could help us plan for the future by developing an automated building system that would be used as part of a smart city.
Article
Full-text available
People spend most of their time indoors in residential buildings; thus, maintaining good indoor air quality (IAQ) is essential. It is necessary to provide a suitable ventilation rate in these spaces (low-density occupation) taking into account that materials (finishes and furnishing) are one of the main sources of indoor pollution. For that purpose, this research, based on both the National Research Council (NRC) database and a specific case study, followed a method that consisted of the identification of indoor pollution sources, evaluation of the emission factor and characterization of the IAQ. It is intended to show the influence of materials (quantity, type and age), loading ratio and pollution concentration threshold in the calculation of a ventilation rate which guarantees a good IAQ. This study also reflects the need to bring together an overall pollution concentration threshold of TVOC, below which a safe level of exposure to indoor pollution is recognized.
Article
Formaldehyde has substantial adverse impacts on human health, and formaldehyde exposure primarily occurs in indoor environments. This study investigated the infiltration rates and indoor formaldehyde concentrations in 5 climate zones of China. In the winter, apartments in northern China suffer from higher indoor formaldehyde concentrations than apartments in southern China. The median indoor formaldehyde concentrations were 56 μg/m³ (25%, 75%: 38, 91 μg/m³) in northern China and 40 μg/m³ (25%, 75%: 30, 61 μg/m³) in southern China. There is a clear decrease in indoor formaldehyde concentrations in China. We also studied the relationships of the indoor air temperature, years of decoration, infiltration rate and source characteristic with formaldehyde concentrations in closed conditions. A multiple regression model that related these factors to the formaldehyde concentrations in closed conditions was constructed (R²=0.75). The optimal curve for the suitable combination of temperature and infiltration rate to maintain low formaldehyde concentrations with the lowest cost was calculated for northern and southern China. By comparing the optimal curve and the state point of each city, we can infer the suitable tendency of indoor temperature and infiltration rate for each city. In Tianjin and Shenyang, apartments are overheating, thus causing a high percentage of some homes to have formaldehyde concentrations above the Chinese national standard. In Shanxi, Xinjiang, Yunnan and the Yangtze River Delta, the infiltration rate should be increased to some extent to achieve better indoor air quality. In Hunan, Hubei and Chongqing, indoor temperatures could be increased to improve indoor thermal comfort.
Article
Efforts to improve energy efficiency in homes and buildings have led to tighter structures. However, these changes can also produce negative consequences for indoor air quality (IAQ) and human health. One of the dramatic effects of climate change and weather is the increase in destructive wildfires, such as those experienced in the Pacific Northwest during the summer, 2015. This paper presents data for measurements at two houses during periods with and without high levels of wildfire smoke outdoors. For each house, indoor and outdoor pollutant measurements were obtained for ozone (O3), fine particulate matter (PM2.5), and volatile organic compounds (VOCs) along with outdoor weather conditions and occupant activities including the use of windows and doors. The VOC measurements were obtained using a Proton Transfer Reaction Mass Spectrometer. Compounds monitored included acetonitrile (a biomass burning tracer), formaldehyde, acetaldehyde, methanol, acetone, benzene, toluene, and C2-alkylbenzenes (i.e. sum of xylenes & ethylbenzene), C3-alkylbenzenes (i.e. sum of trimethylbenzene, ethyltoluene, and propylbenzene isomers), and C4-alkylbenzenes (i.e. sum of tetramethylbenzene and its isomers). A carbon dioxide (CO2) tracer method was used to measure in-situ ventilation rates, and blower door tests were also completed to determine standard ventilation rates. For smoky periods with elevated outdoor pollutant levels, penetration factors, defined as the ratio of indoor/outdoor concentrations were quite low. Penetration factors for PM2.5 were 11% for H2 and 15% for H3, except when windows or doors were open. The penetration factors for O3 were also low at 24% for H2 and 5% for H3. Elevated indoor VOC levels were not typically associated with outdoor levels, but reflected significant indoor sources. During smoke events, acetonitrile, a biomass burning tracer compound, was elevated outdoors and indoors in both houses, and benzene was elevated outdoors and indoors in H3.
Article
Cooking emission is one of sources for ambient volatile organic compounds (VOCs), which is deleterious to air quality, climate and human health. These emissions are especially of great interest in large cities of East and Southeast Asia. We conducted a case study in which VOC emissions from kitchen extraction stacks have been sampled in total 57 times in the Megacity Shanghai. To obtain representative data, we sampled VOC emissions from kitchens, including restaurants of seven common cuisine types, canteens, and family kitchens. VOC species profiles and their chemical reactivities have been determined. The results showed that 51.26%±23.87% of alkane and 24.33±11.69% of oxygenated VOCs (O-VOCs) dominate the VOC cooking emissions. Yet, the VOCs with the largest ozone formation potential (OFP) and secondary organic aerosol potential (SOAP) were from the alkene and aromatic categories, accounting for 6.8-97.0% and 73.8-98.0%, respectively. Barbequing has the most potential of harming people's heath due to its significant higher emissions of acetaldehyde, hexanal, and acrolein. Methodologies for calculating VOC emission factors (EF) for restaurants that take into account VOCs emitted per person (EFperson), per kitchen stove (EFkitchen stove) and per hour (EFhour) are developed and discussed. Methodologies for deriving VOC emission inventories (S) from restaurants are further defined and discussed based on two categories: cuisine types (Stype) and restaurant scales (Sscale). The range of Stype and Sscale are 4124.33-7818.04t/year and 1355.11-2402.21t/year, respectively. We also found that Stype and Sscale for 100,000 people are 17.07-32.36t/year and 5.61-9.95t/year, respectively. Based on Environmental Kuznets Curve, the annual total amount of VOCs emissions from catering industry in different provinces in China was estimated, which was 5680.53t/year, 6122.43t/year, and 66,244.59t/year for Shangdong and Guangdong provinces and whole China, respectively. Large and medium-scaled restaurants should be paid more attention with respect to regulation of VOCs.
Article
Indoor VOCs still remain at high levels in many areas, which may directly impact human health. In this study, hundreds of residential indoor air samples were collected in Xi'an and Hangzhou from April 2014 to November 2015. Indoor VOC pollution level of newly renovated residences was more serious than that of old residences (approximately 10–15 times). The mean indoor VOC level of newly renovated residences in Xi'an was 0.882 ± 1.922 mg/m³, similar to Hangzhou's levels (0.862 ± 1.394 mg/m³). But part of the benzene homologues concentrations in Hangzhou were higher than those in Xi'an, and the exposure risk in Hangzhou could be 1.5 to 8 times that in Xi'an. This study aims to identify characteristics of indoor VOC pollution, quantitatively estimate the combined effects of temperature and humidity, and reveal sink mechanism associated with indoor VOCs. The results indicate that indoor VOC pollution in newly renovated residences is mainly affected by the combination of temperature and humidity as sink effect, especially for the components with large Henry's Law Constant. The results of this study also provide valuable guidance on indoor VOC pollution control, i.e., humidity control is a countermeasure to reduce potential indoor air pollution for people who live in humid areas.
Article
The Huai River and Qingling Mountain divide (H-Q) divide China into north and south with respect to public policies for building construction and operation practises. China's building energy efficiency standard mandates that air exchange rates be 0.5 h− 1 north of the H-Q divide and 1 h− 1 south of the divide. China's heating policy allows space heating systems only north of the H-Q divide. Consequently, indoor temperature and humidity differ considerably between north and south. A theoretical model using indoor temperature, humidity, and air change rate was developed to predict indoor formaldehyde concentrations. Data for 39 cities were obtained from 42 studies. There was good agreement between the literature and modelling in a theoretical reference room. The United States Environmental Protection Agency (U.S.EPA) model was applied to estimate cancer risk from formaldehyde exposure indoors. The median indoor formaldehyde concentration for renovation ever from 2002 to 2015 in Chinese cities was 125 μg/m³, which is higher than the WHO threshold, 100 μg/m³. The median indoor formaldehyde concentrations in the north were higher than in the south (0.5 times higher for dwellings renovated within the past year and 0.2 times higher for renovation ever), driven by the much higher northern winter concentrations (40–1320%). The U.S.EPA model predicts that the lifetime formaldehyde related cancer risk for people living north of the H-Q divide is 1.2 times greater than for people living south. This can be partly explained by greater indoor exposure to formaldehyde for Chinese living north of the H-Q divide.
Article
Furniture and building/furnishing materials are major sources of indoor volatile organic compounds (VOCs), and the VOC emission characteristics of these materials are essential for understanding indoor air pollution dynamics. Herein, we investigated the emission characteristics of formaldehyde (HCHO) and other VOCs from particleboard in sealed or ventilated environmental chambers at different temperatures (23, 35 or 50 °C), with a focus on the emission of odorous compounds. The emissions of HCHO and total VOCs (TVOC) from the particleboard increased significantly with temperature, and the emitted VOC mixtures had complex chemical compositions. In addition to HCHO, 44 compounds were identified, including alkanes, (chlorinated) aromatic hydrocarbons, carbonyl compounds (i.e., aldehydes and ketones), alcohols, and esters, etc. At room temperature (23 °C), n-hexane was the most abundant compound except HCHO; but at higher temperatures, concentrations of hexanal and pentanal significantly increased. Moreover, due to their low odor thresholds, aldehydes, particularly hexanal and pentanal, were identified as the major odorous compounds emitted from the particleboard. Enhanced ventilation could effectively decrease VOC concentrations in the environmental chamber at room temperature, but less effectively at higher temperatures. After heat treatment at 50 or 60 °C, the emissions of HCHO and TVOC at room temperature decreased significantly. More importantly, because the emissions of hexanal and pentanal were highly sensitive to temperature, their emission strengths were effectively reduced after heat treatment, resulting in significantly lowered odor emissions from the particleboard at room temperature. These results are helpful for the control of indoor odor problems arising from furniture materials.