Article

pubs.acs.org/est

Long-Term Formaldehyde Emissions from Medium-Density

Fiberboard in a Full-Scale Experimental Room: Emission
Characteristics and the Eects of Temperature and Humidity
Weihui Liang, Shen Yang, and Xudong Yang*
Department of Building Science, Tsinghua University, Beijing 100084, P. R. China

Downloaded by UNIV VERACRUZANA on August 26, 2015 | http://pubs.acs.org

Publication Date (Web): August 21, 2015 | doi: 10.1021/acs.est.5b02217

S Supporting Information
*

ABSTRACT: We studied formaldehyde emissions from the

medium-density berboard (MDF) in a full-scale experimental
room to approximate emissions in actual buildings. Detailed
indoor formaldehyde concentrations and temperature and
humidity data were obtained for about 29 months. Temperature,
relative humidity (RH), and absolute humidity (AH) ranged over
10.931.4 C, 46.583.6%, and 1.123.1 g/kgair, respectively.
Annual cyclical seasonal variations were observed for indoor
formaldehyde concentrations and emission rates, exhibiting
entirely dierent characteristics than those in an environmental
chamber under constant environmental conditions. The maximum concentration occurred in summer rather than at initial
introduction of the material. The concentrations in summer
could be a few up to 20 times higher than that in winter, depending on the indoor temperature and humidity conditions.
Concentrations decreased by 2065% in corresponding months of the second year. Indoor formaldehyde concentrations were
positively correlated with temperature and AH but were poorly correlated with RH. The combined eects of temperature and
AH on formaldehyde emissions from MDF in actual buildings were veried. These detailed long-term experimental results could
be used with environmental chamber measurement data to scale up and validate emission models from chambers held at constant
conditions to actual buildings.

INTRODUCTION
Formaldehyde is one of the most ubiquitous gaseous pollutants
indoors.1 Acute exposure to formaldehyde could cause nose,
eye, and throat irritation;24 chronic exposure could increase
respiratory illness and even cancer.57 The World Health
Organization has recommended that formaldehyde concentration in air should be below 0.1 mg/m3 as a 30 min average.8
This value is also adopted by the Chinese national standard for
the 1-h exposure limit.9 However, indoor formaldehyde
concentration in 70% of the newly built or remodeled houses
exceeded the national standard in China,10 with a mean value of
0.238 mg/m3. This was much higher than that in developed
countries, which were 0.05 mg/m3 in Germany and 0.037 mg/
m3 in Denmark.1 In actual buildings, indoor temperature and
humidity are generally not constant11,12 and vary with outdoor
environmental conditions as well as occupant behavior.13,14
Consequently, formaldehyde is emitted under variable temperature and humidity conditions. Understanding the emission
characteristics of building materials in a realistic indoor
environment is needed to model indoor exposures and mitigate
health risks.
Emissions from building materials have previously been
studied under conditions quite dierent from those in actual
buildings. The most common approach is the use of an
environmental chamber or eld and laboratory emission cell in
XXXX American Chemical Society

which emission measurements are conducted at constant

temperature, relative humidity (RH), and ventilation rate.15,16
Generally speaking, when a material is tested in a ventilated
environmental chamber, the concentration in the air of the
chamber typically increases to a maximum value in the rst few
hours and then continuously decreases, with the highest
emission rate at the beginning of the experiment. Empirical
models17 and diusion mass transfer models1820 have been
developed to describe the emission behaviors of materials based
on these studies. However, under the variable environmental
conditions in actual buildings, emission characteristics of
building materials may be dierent. Thus, emissions under
realistic conditions may better replicate emission characteristics
in the built environment.
More realistic conditions are typically evaluated through eld
studies, which are time-consuming, expensive, and more
dicult to carry out than studies using environmental
chambers. Most eld studies conduct short-term (days to
weeks) measurements, with few evaluating long-term (more
than one year) concentrations as would be found in actual
Received: May 11, 2015
Revised: August 5, 2015
Accepted: August 11, 2015

DOI: 10.1021/acs.est.5b02217
Environ. Sci. Technol. XXXX, XXX, XXXXXX

Article

Downloaded by UNIV VERACRUZANA on August 26, 2015 | http://pubs.acs.org

Publication Date (Web): August 21, 2015 | doi: 10.1021/acs.est.5b02217

Environmental Science & Technology

buildings. Wolko et al.21 measured formaldehyde concentrations in two semi-detached twin apartments for one year.
They found that formaldehyde concentrations were higher in
summer in the vacant apartment, while decreased and then
increased during autumn in the occupied one. Pollutant
transport occurred between these two apartments, which
complicated the situation. Brown22 measured formaldehyde
concentrations four times over eight months in an occupied
dwelling. Formaldehyde concentrations were found to decay
double exponentially. However, the above studies could not
directly characterize emissions from materials because human
activities may also have contributed to the observed formaldehyde concentrations, and multiple emission sources may
have been present in the studied spaces. Crump et al.23
measured formaldehyde in four unoccupied test houses for 24
months to investigate the relationship between sources and
indoor concentrations. As multiple materials were present in
the house, it was dicult to distinguish emission from each
individual source. Thus, long-term measurements conducted in
actual houses without interference by human behavior and
multiple sources are needed to understand realistic materials
emission characteristics. Existing eld studies also have not
continuously measured environmental conditions. The frequency of concentration measurements and the lengths of the
studies may also have been insucient considering the longterm eects of formaldehyde emissions on indoor air quality.
The main objectives of this study are to investigate the longterm formaldehyde emission characteristics of building
materials under variable temperature and humidity conditions,
to explore the extent to which formaldehyde emissions vary by
season, to identify the eects of temperature variations on
emissions, and also to explore the importance of humidity on
formaldehyde emissions. We measured formaldehyde emissions
from a medium-density berboard (MDF) for more than 29
months using a full-scale experimental room. Indoor formaldehyde concentrations and air ventilation rates were
measured frequently throughout the study. Temperature and
humidity were monitored continuously over the entire time
period.

Figure 1. Digital images of the interior and exterior of the full-scale

experimental room and the schematic illustration of the wall assembly:
(a) interior image, (b) exterior image, and (c) wall assembly.

materials. There were no formaldehyde emissions from

the ceramic oor tiles, and emissions from the latex wall
paint were negligible by the time the experiment began.
Thus, the only formaldehyde emission source in this
room was the MDF introduced. Formaldehyde is a small
molecule compound whose adsorption capacity is not as
strong as other VOCs and aldehydes.24 Previous
experimental studies indicated that sorption of formaldehyde by wall paint was insignicant.25
(3) Temperature and humidity were allowed to vary
naturally in the room, without the use of heating or air
conditioning equipment. In this way, temperature and
humidity ranges were much larger than that commonly
observed in actual homes. This would allow better
investigation of the possible impact of temperature and
humidity on material emissions.
(4) A fan was operated to ensure good mixing of air in the
room, with the same design purpose as the fan in an
environmental chamber.26
Studied Material. The studied material was a typical MDF
directly purchased from the manufacturer. It was transported
and placed inside the room immediately after being received.
Hence, it could be considered as a new material at the
beginning. The boards were leaned against the interior walls at
an angle. Both the front and back surfaces of the boards emitted
formaldehyde, which could be simplied to a one-dimensional
material with half thickness and double emission area to the
original board. Key features of the studied MDF are listed
below.
(1) Dimension of the original full-size MDF is 2.44 m 1.2
m 0.012 m (length width thickness). Three fullsize boards together with a small piece (dimension of
0.18 m 1.2 m 0.012 m) were placed in the room,
resulting in a loading ratio of 0.5 m2/m3. This value was
comparable to that of densely furnished spaces such as a
small room or kitchen but higher than most living or
bedrooms in actual residential buildings.27 A relatively
large loading ratio was selected to allow higher room
concentrations over a long period of time to minimize
the error caused by measurement uncertainty.

MATERIALS AND METHODS

The Full-Scale Experimental Room. Dimension of the
full-scale experimental room was 4 m 3 m 3 m (length
width height). It was located in the corner of an experimental
base in a rural district of Beijing. The total area of the
experimental base was 5000 m2, and a 2.5 m high enclosure was
built around it. There was only one door in the south wall of
the room, which remained closed during the entire measurement period. Figure 1 gives the digital images and schematic
illustration of the experimental room and its wall assembly. The
room layout and interior surface materials shared similar
features as a typical room in China, but the room possessed
several key features to better serve the goal of this particular
study, as described below.
(1) The room was unoccupied. Thus, all human-related
interferences to emissions could be eliminated. Another
advantage was that measurements could be conducted at
any time without disturbing occupants or requiring
permission.
(2) The construction of the room was completed two years
before the experiment begun, and only latex wall paint
and ceramic oor tiles were used as interior construction
B

DOI: 10.1021/acs.est.5b02217
Environ. Sci. Technol. XXXX, XXX, XXXXXX

Article

Downloaded by UNIV VERACRUZANA on August 26, 2015 | http://pubs.acs.org

Publication Date (Web): August 21, 2015 | doi: 10.1021/acs.est.5b02217

Environmental Science & Technology

measurement period was 9 October 2012 to the end of

February 2015. The indoor background formaldehyde concentration was also measured before introducing MDF. Outdoor
formaldehyde concentrations were also occasionally measured.
Environmental Measurements. Indoor temperature and
RH were measured by an automatic data logger (WSZY-1
sensor, Beijing, China) placed in the center of the room. The
instrument measurement ranges were from 40 to 100 C for
temperature and 0100% for RH with measurement errors of
0.5 C for temperature and 3% for RH. The sampling interval
was 10 min. The ventilation rate was measured monthly in
2013 using the CO2 decay method31 to analyze annual
variations.

(2) The MDF was directly exposed to air dierent from its
typical use as chairs surrounded by foam or tables and
countertops laminated by a plastic cover. This could
simplify the emission process and better t the main
purpose of this study. More complicated and realistic
features could be added later after the fundamental
emission characteristics of the material were understood.
(3) Headspace analysis of the MDF indicated that formaldehyde was the dominant emitted compound. Headspace concentrations of other aldehydes and ketones
were 1 or 2 orders of magnitude lower than formaldehyde, and the sum of other aldehydes and ketones
was less than 5% of the formaldehyde mass. The total
VOC concentration in the MDF headspace was as low as
15.7 g/m3. Thus, potential generation of formaldehyde
from chemical oxidation of terpenes, alkenes, or
hydrocarbons could be eliminated.28,29
The above features of the experimental room and studied
material will benet the understanding of long-term formaldehyde emission characteristics as well as the quantication of
temperature and humidity eects. On the other hand,
formaldehyde concentrations in the experimental room were
much higher than that in actual homes due to large loading
ratio, elimination of the retard resistance from the cover
surrounding the MDF, and extreme summertime environmental conditions. Concentration dierences in dierent
seasons could also be more pronounced due to larger indoor
temperature and humidity ranges across seasons. The formaldehyde concentrations measured in this experimental room
should not be considered as typical in actual homes.
Formaldehyde Measurements. Entry into the experimental room could cause air exchange between outdoor and
indoor air, aecting the indoor formaldehyde concentration. To
eliminate human interference in the concentrations, a sampling
portal was placed near the door frame so that samples could be
collected from outside the room without opening the door and
entering. The portal was located 1.2 m above the oor.
Formaldehyde was measured using the 3-methyl-2-benzothiazolinone hydrazone (MBTH) spectrophotometric method, in
accordance with the Chinese national standard.30 The accuracy
of this method was evaluated and is described in Section S1 of
Supporting Information. To collect formaldehyde samples, a
stainless steel tube was extended into the center of the room
through the sampling portal. Indoor air samples were pumped
through the stainless steel tube into a glass sampling tube using
a QC-2 pump (Beijing Institute of Labor Protection) at a ow
rate of 100200 mL/min. Formaldehyde in the air was
absorbed into a solution in the glass tube containing 5.0 mL of
50 g/mL MBTH. The sampling time was 530 min,
depending on the estimated formaldehyde concentration in
the air, with less volume collected at higher concentrations. To
analyze the formaldehyde concentrations, 0.4 mL of 10 g/L
ferric ammonium sulfate solution was added to the sampling
tube. The tube was shaken and then held for 15 min, during
which formaldehyde was converted into a blue cationic dye by
the MBTH. The light absorbance was measured using a
spectrophotometer at 630 nm (Unic 7200, China). Quality
control and assurance of the method is detailed in Section S2 of
the Supporting Information.
Frequent sampling (ranging from a few days to a few weeks)
was conducted to evaluate variations and trends over time.
Three duplicate samples were taken for each measurement. The

RESULTS AND DISCUSSION

Formaldehyde Concentration. The measured indoor
formaldehyde concentrations are illustrated in Figure 2.

Figure 2. Indoor formaldehyde concentration and emissions rate

proles in the experimental room. Aut., Win., Spri., and Sum. are
abbreviations for autumn, winter, spring, and summer, respectively.
Error bars are the standard deviations of the formaldehyde
concentration results.

Relative standard deviations (RSD) were within 10% for

most measurements, indicating good repeatability of the results.
The mean value for each test was used for data analysis. The
year was divided into spring (MarchMay), summer (June
August), autumn (SeptemberNovember), and winter (DecemberFebruary) seasons. Indoor formaldehyde concentration ranges and means for each season are listed in Table 1.
Indoor formaldehyde concentrations did not decrease continuously, instead having dierent trends in dierent seasons. A
monotonic increase in the concentration occurred in spring,
reaching a maximum value of 4.78 mg/m3 in summer 2013 and
3.19 mg/m3 in summer 2014. Concentrations then decreased
throughout midlate summer and autumn and reached a
minimum in winter. Both the levels and ranges diered in
dierent seasons, with summer > spring > autumn > winter.
The maximum concentration was 3050 times the minimum,
exhibiting signicant dierences in the same year. The mean
concentration in summer was 20 times that in winter.
Therefore, indoor formaldehyde concentrations and trends
were strongly seasonal.
Both the formaldehyde concentrations and ranges decreased
from one year to the next. The mean concentration was 3.59
mg/m3 in the summer of 2013 and decreased to 2.52 mg/m3 in
the summer of 2014. Similar concentration decreases were
measured in other seasons from 2013 to 2014. Generally
speaking, the formaldehyde concentration decreased 2065%
C

DOI: 10.1021/acs.est.5b02217
Environ. Sci. Technol. XXXX, XXX, XXXXXX

Article

8.2
3.4
13.5
23.1
14.7
3.4
14.5
21.1
15.1
2.8

Measurement data for 2012 only partially covered the autumn season (9 October30 December). These data were not used for comparison with data for autumn 2013 and 2014.

ER(t ) = VR (C(t ) Camb)/L

(1)

where ER(t) is the emission rate (mg/m h) at time t, VR is the

ventilation rate of the experimental room (h1), C(t) is the
measured indoor formaldehyde concentration (mg/m3) at time
t, Camb is the outdoor formaldehyde concentration (mg/m3),
and L is the loading ratio of the source material (0.5 m2/m3).
The background formaldehyde concentration in the
experimental room was 0.021 mg/m3. The outdoor formaldehyde concentration varied only slightly within 0.012
0.010 mg/m3 during the entire study period. Thus, there was
no signicant dierence between the outdoor and background
formaldehyde concentration, and the assumption that there
were no other formaldehyde emission sources in the
experimental room was conrmed. Additionally, no seasonal
trends were observed in the outdoor formaldehyde concentration (Camb).
The monthly measured ventilation rates for 2013 are
illustrated in Figure 3. It varied randomly between 0.210.35
2

Figure 3. Ventilation rate results for the experimental room in 2013.

h1. The relatively stable ventilation rate was ideal for direct
comparison purposes. The main reason for the small variation
was that no heating or cooling equipment was operated in the
experimental room; thus, indooroutdoor temperature dierences were small during all of the seasons. Moreover, the eect
of outdoor wind pressure on the ventilation rate was minimized
by an enclosure around the room. No seasonal variations in
ventilation rate would be expected. The mean and standard
deviation (SD) of the ventilation rate throughout the year were
0.29 h1 and 0.046 h1, respectively. Ventilation conditions
over the entire test period were assumed to follow the pattern
observed in 2013.
Thus, the loading ratio (L) was constant, and outdoor
formaldehyde concentrations (Camb) were much lower than

max

in corresponding months of the second year. Concentration

ranges also narrowed, for example, from 1.324.78 mg/m3 in
the summer of 2013 to 2.173.19 mg/m3 in the summer of
2014. The above results indicate that formaldehyde emissions
decreased annually within the seasonal trends noted above.
Formaldehyde Emission Characteristics. Emission Rate
Calculation. In the time-scale of several hours or a day, indoor
temperature and humidity varied in a small range. Indoor
environmental conditions were relatively stable. Consequently,
a short-term steady state emission rate could be assumed. The
formaldehyde emission rate from MDF at a certain time t was
estimated based on the following equation:

2.3
1.1
2.9
10.2
1.9
1.3
2.7
10.8
2.8
1.5
4.6
1.9
6.7
17.2
7.8
2.1
7.2
15.6
7.4
1.9

min
mean
max

71.2
73.6
69.1
83.6
78.2
68.3
62.8
72.9
76.1
59.9

min

55.3
58.7
49.1
55.1
49.9
50.9
46.5
51.4
55.8
49.4
63.5
66.3
60.1
75.2
67.3
58.6
55.3
64.3
66.7
54.8

mean
max

19.3
4.3
26.9
31
24.7
4.5
30
31.4
26.1
4.8

min

0.3
10.9
2.5
19.1
0
5.6
3.2
22.7
2.6
4.0

mean

8.7
4.18
14.7
27.0
14.2
1.3
17.1
28.1
14.0
1.2
1.02
0.23
1.84
2.76
1.38
0.06
1.28
1.84
0.77
0.04

max
min

0.18
0.04
0.16
0.76
0.08
0.05
0.09
1.25
0.05
0.02
0.57
0.10
0.80
2.07
0.45
0.06
0.61
1.46
0.38
0.03

mean
max
min

0.33
0.09
0.28
1.32
0.15
0.1
0.17
2.17
0.1
0.05
1.00
0.19
1.38
3.59
0.79
0.11
1.07
2.52
0.66
0.07

season

autumna
winter
spring
summer
autumn
winter
spring
summer
autumn
winter

time period

9/10/201230/11/2012
1/12/201228/2/2013
1/3/201331/5/2013
1/6/201331/8/2013
1/9/201330/11/2013
1/12/201328/2/2014
1/3/201431/5/2014
1/6/201431/8/2014
1/9/201430/11/2014
1/12/201428/2/2015

mean

1.77
0.38
3.18
4.78
2.4
0.12
2.21
3.19
1.33
0.08

AH (g/kgair)
RH (%)
temperature (C)
emission rate (mg/m2h)
concentration (mg/m3)

Table 1. Indoor Formaldehyde Concentrations, Formaldehyde Emissions Rates, Temperature, Relative Humidity (RH), and Absolute Humidity (AH) in Dierent Seasons

Downloaded by UNIV VERACRUZANA on August 26, 2015 | http://pubs.acs.org

Publication Date (Web): August 21, 2015 | doi: 10.1021/acs.est.5b02217

Environmental Science & Technology

DOI: 10.1021/acs.est.5b02217
Environ. Sci. Technol. XXXX, XXX, XXXXXX

Article

Downloaded by UNIV VERACRUZANA on August 26, 2015 | http://pubs.acs.org

Publication Date (Web): August 21, 2015 | doi: 10.1021/acs.est.5b02217

Environmental Science & Technology

indoor formaldehyde concentrations (C(t)). Uncertainty or the
RSD of the emission rate for each measurement may be larger
than that for the concentration considering the variations in the
ventilation rate (VR). However, the overall trends in emission
rates would not be aected by this factor. Thus, the mean
emission rate for each measurement would vary in a similar
manner as did the indoor formaldehyde concentration, as
illustrated in the right-hand y-axis of Figure 2. Means and
ranges for emission rates in the dierent seasons are listed in
Table 1.
Comparison of Emission Characteristics between Environmental Chambers and the Experimental Room. The
emission rate from MDF decreased from the beginning of
the measurement until 1 December 2012, consistent with that
observed in an environmental chamber with constant environmental conditions. One might draw the conclusion that MDF
emission in actual buildings would be similar to those in the
chambers based on such data. However, this trend did not
continue over time, and an entirely dierent emission pattern
occurred in the experimental room compared to that in the
environmental chamber, as discussed below.
(1) Material emission rates in an environmental chamber
should decrease continuously. 32 However, in the
experimental room, an apparent increase in the emission
rates occurred in spring and summer. Annual cyclical
seasonal variations were observed.
(2) In an environmental chamber, the highest emission rates
usually occur at the beginning when materials are
introduced.17 However, in the experimental room,
maximum emissions occurred in summer instead of at
the beginning of the study. For example, the highest
emission rates in the summer of 2013 and 2014 were
2.76 and 1.84 mg/m2h, respectively, much higher than
the 0.93 mg/m2h initial emission rate. These results
suggest that environmental conditions should be
considered in estimating formaldehyde emission rates
from MDF.
(3) When materials are tested in an environmental chamber,
a steady state could be achieved with prolonged
emission time.33 The emission rates change little after
that.34 However, when the MDF was placed in the
experimental room, no such steady state could be
assumed for the long time scale.
Due to apparent dierences in emission characteristics
observed between environmental chambers and the experimental room, emission models developed based on constant
environmental conditions should not be directly applied to
actual buildings. Adjustments to the model or model
parameters are required to represent the more complex
situation in actual buildings.
Inuencing Factors. The main dierences between the
experimental room and environmental chambers were the
temperature and humidity conditions. The impacts of these two
factors are examined below.
Inuence of Temperature. The overall trends in formaldehyde concentration and temperature were consistent
(Figure 4). Obvious seasonal variations were also observed in
indoor temperature. Indoor formaldehyde concentrations
increased in spring and decreased in autumn, as did the
temperature. The same patterns in concentration and temperature were observed in summer and winter. On the basis of this
qualitative analysis, temperature was likely one of the key

Figure 4. Indoor formaldehyde concentration and temperature proles

for the experimental room.

factors inducing seasonal variations in formaldehyde concentrations. The Pearson correlation coecient between the
formaldehyde concentration and temperature was 0.84,
indicating a strong positive correlation between these two
parameters.
Temperature ranges for each season are listed in Table 1.
The annual variations were from 10.9 to 31 C in 2013 and
5.6 to 31.4 C in 2014, respectively. This broad range goes
beyond all possible temperature situations in actual buildings
and inclusive. In particular, it is interesting to compare
emissions at high- and low-temperature extremes. The
magnitude of indoor temperature variations in summer were
nearly the same as or smaller than those in winter (19.1 to 31
C vs 10.9 to 4.3 C); however, the formaldehyde
concentration ranges were much larger in summer than in
winter (1.32 to 4.78 mg/m3 vs 0.09 to 0.38 mg/m3). Thus,
formaldehyde emissions were more sensitive to temperature
changes at higher temperatures.
Temperatures and ranges over the next year did not change
substantially from the rst year (Table 1, Figure 4). However,
formaldehyde concentrations and ranges decreased signicantly
(0.09 to 4.78 mg/m3 vs 0.10 to 3.19 mg/m3). The same
correlations between temperature and concentration in each
season were observed. These results demonstrate the reduced
eect of temperature on formaldehyde emissions over the long
term, which was not revealed in previous studies using
environmental chambers.3436
Inuence of Humidity. Ranges for RH in each season are
presented in Table 1, and indoor RH and formaldehyde
concentration proles are illustrated in Figure 5a. Indoor RH
mainly varied between 55 and 70%. RH began to increase at the
end of May 2013 and reached a maximum of 83.6% in the
middle of August 2013. This increase in RH was due to rainy
weather during that period. Indoor RH decreased and returned
to almost the same level as before after 15 October 2013.
Unlike formaldehyde concentration and temperature, RH was
relatively stable without strong seasonal variations. The Pearson
correlation coecient between the indoor formaldehyde
concentration and RH was 0.33, indicating that seasonal
variations in the formaldehyde concentration were not strongly
associated with RH.
The absolute humidity (AH) prole, on the other hand,
exhibited clear seasonal variations (Figure 5b). The maximum
AH occurred in summer and the minimum in winter, with a
E

DOI: 10.1021/acs.est.5b02217
Environ. Sci. Technol. XXXX, XXX, XXXXXX

Article

Environmental Science & Technology

Downloaded by UNIV VERACRUZANA on August 26, 2015 | http://pubs.acs.org

Publication Date (Web): August 21, 2015 | doi: 10.1021/acs.est.5b02217

the standpoint of the hydrolysis reaction mechanism, AH is a

more appropriate metric for evaluating the inuence of
humidity on formaldehyde emissions than RH.
Verication of the Eects of Temperature and AH on
Formaldehyde Emissions through Small Environmental
Chamber Tests. On the basis of the above results, the trends
in formaldehyde concentration, temperature, and AH were
quite similar. To further verify temperature and humidity
eects, additional small-scale environmental chamber tests were
conducted by varying these two parameters individually.
Small environmental chamber conditions and procedures are
detailed in Section S3 of Supporting Information. Time trends
for formaldehyde concentrations in the chamber air are shown
in Figure 6. The maximum concentration occurred in the rst

Figure 6. Formaldehyde results of the small-scale environmental

chamber tests.

Figure 5. Indoor formaldehyde concentration and humidity proles

for the experimental room: (a) formaldehyde concentration and
relative humidity (RH) and (b) formaldehyde concentration and
absolute humidity (AH).

few hours and then decreased continuously, unlike the annual

seasonal variations observed in the experimental room. By
comparing the results for Cases 1 and 2, it was found that at the
same temperature, formaldehyde emissions increased with
increasing AH. A similar trend was observed by comparing the
results for Cases 1 and 3, demonstrating that increasing
temperature also increases emissions.
Comparison with Other Studies. The measurement
settings and results of this and several other studies are briey
compared in Table 2. The unique features and advantages of
this study are summarized as follows.
(1) Human-related emissions, as well as disturbances by the
testing personnel, were eliminated in this unoccupied
experimental room. Other formaldehyde sources and
interactions were not present, so that the measured
concentrations were solely due to emissions by the MDF
studied. In other studies, the measurement results may
represent the cumulative contributions of several factors.
(2) The overall study period using the experimental room
was 29 months, far longer than previous controlled
laboratory or eld studies. Formaldehyde concentrations,
as well as temperature, humidity, and ventilation rate
data, were all obtained over this long study period.
Annual cyclical seasonal variations in formaldehyde
concentrations and emission rates were observed along
with annual attenuation. Positive correlations between
the formaldehyde concentration and temperature and the
formaldehyde concentration and AH were also clearly
identied based on the detailed long-term data. Due to

range from 1.1 to 23.1 g/kgair. The Pearson correlation

coecient between the formaldehyde concentration and AH
was 0.89, indicating a strongly positive correlation between
them.
Previous studies on the inuence of humidity on formaldehyde emissions were mainly conducted in environmental
chambers and focused on analysis of RH. The inuence of
humidity on formaldehyde emissions has also been neglected in
eld studies.2123 However, Parthasarathy et al.35 found that
the emission rate increased when RH increased from 50% to
85%. Similar conclusions were obtained in other studies.36,37
Given the apparent discrepancies between the results of this
and previous studies, it is important to consider the
fundamental mechanism of formaldehyde emissions from
MDF. A common explanation about the mechanism of
humidity eect on formaldehyde emissions is that free
formaldehyde would increase due to ureaformaldehyde
depolymerization and hydrolysis in the presence of free
water.3638 Competition between pollutant and water molecules for free adsorption sites of the building material is another
possible mechanism.3941 However, it is the molar amount of
free water that determines the amount of hydrolysis. The
absolute quantity of free water in the air (AH) is a better
approximation of the molar amount of free water than the RH.
Moreover, RH changes with temperature even when the molar
amount of free water remains the same in the air. Thus, from
F

DOI: 10.1021/acs.est.5b02217
Environ. Sci. Technol. XXXX, XXX, XXXXXX

Article

Environmental Science & Technology

Table 2. Summary of Measurement Conditions and Results of This Study with Those of Previous Studies
studies
house occupancy
testing person position
building materials number
other VOCs existences
time duration
total number of concentration tests
duplicated samples
temperature interval
RH interval
temperature range (C)
RH range (%)
AH range (g/kgair)
ventilation measurement period
ventilation measurement frequency

Downloaded by UNIV VERACRUZANA on August 26, 2015 | http://pubs.acs.org

Publication Date (Web): August 21, 2015 | doi: 10.1021/acs.est.5b02217

Wolko et al.21
occupied
indoor
multiple
yes (21)a
12 months
10 times
no duplicate
not reported
not reported
14.923.9
4056
not reported
March to May
not reported

unoccupied
indoor
multiple
yes (21)a
12 months
10 times
no duplicate
not reported
not reported
1733
2467
not reported
March to May
not reported

Brown22

Crump et al.23

this study

occupied
indoor
multiple
yes (27)a
8 months
4 times
two duplicates
not reported
not reported
2129
3254
6.610.8
not measured
not measured

unoccupied
indoor
multiple
yes (>5)a
24 months
28 times
no duplicate
not reported
not reported
2030
3340
not reported
not reported
one time

unoccupied
outdoor
only MDF
none
29 months
71 times
three duplicates
10 min
10 min
10.931.4
46.583.6
1.123.1
a whole year
monthly

Numbers in parentheses are the reported number of VOCs in the corresponding references.

the shorter durations and sometimes confounding data in

other studies, these interesting variations and correlations
were not reported.
(3) Temperature and humidity varied naturally, and the
ranges for these variables were deliberately allowed to be
much larger than typically occur in occupied buildings.
Thus, the eects of temperature and humidity on
formaldehyde emissions could be fully represented and
analyzed. In particular, the eects of low extremes in
temperature and AH were analyzed in the eld for the
rst time.
Thus, this study had several advantages, including
elimination of multiple interferences, long-term detailed
measurement of data, and incorporation of broader temperature and humidity ranges. The results provide a meaningful
contribution to characterization of long-term formaldehyde
emissions and can be used to guide factor analysis as well as
emissions model validation for actual buildings.

(1) Salthammer, T.; Mentese, S.; Marutzky, R. Formaldehyde in the

indoor environment. Chem. Rev. 2010, 110 (4), 25362572.
(2) Main, D. M.; Hogan, T. J. Health effects of low-level exposure to
formaldehyde. J. Occup. Environ. Med. 1983, 25 (12), 896900.
(3) Yang, X.; Zhang, Y.; Chen, D.; Chen, W.; Wang, R. Eye irritation
caused by formaldehyde as an indoor air pollution-a controlled human
exposure experiment. Biomed. Environ. Sci. 2001, 14 (3), 229236.
(4) Lang, I.; Bruckner, T.; Triebig, G. Formaldehyde and
chemosensory irritation in humans: a controlled human exposure
study. Regul. Toxicol. Pharmacol. 2008, 50 (1), 2336.
(5) Franklin, P.; Dingle, P.; Stick, S. Raised exhaled nitric oxide in
healthy children is associated with domestic formaldehyde levels. Am.
J. Respir. Crit. Care Med. 2000, 161 (5), 17571759.
(6) Rumchev, K. B.; Spickett, J. T.; Bulsara, M. K.; Phillips, M. R.;
Stick, S. M. Domestic exposure to formaldehyde significantly increases
the risk of asthma in young children. Eur. Respir. J. 2002, 20 (2), 403
408.
(7) Formaldehyde, 2-Butoxyethanol and 1-tert-Butoxypropan-2-ol;
IARC Monographs on the Evaluation of Carcinogenic Risk to
Humans; World Health Organization, International Agency for
Research on Cancer, 2006; Vol. 88.
(8) Development of WHO Guidelines for Indoor Air Quality World
Health Organization; WHO-ROE, Regional Oce of Europe, 2006.
(9) Indoor Air Quality Standard; GB/T18883-2002; Beijing:
Administration of Quality Supervision, Inspection and Quarantine,
2000 (in Chinese).
(10) Tang, X.; Bai, Y.; Duong, A.; Smith, M. T.; Li, L.; Zhang, L.
Formaldehyde in China: Production, consumption, exposure levels,
and health effects. Environ. Int. 2009, 35 (8), 12101224.
(11) Korjenic, A.; Teblick, H.; Bednar, T. Increasing the indoor
humidity levels in buildings with ventilation systems: simulation aided
design in case of passive houses. Build. Simul. 2010, 3 (4), 295310.
(12) Soutullo, S.; Enrquez, R.; Jimenez, M. J.; Heras, M. R. Thermal
comfort evaluation in a mechanically ventilated office building located
in a continental climate. Energy Build. 2014, 81, 424429.
(13) Saeki, K.; Obayashi, K.; Iwamoto, J.; Tone, N.; Okamoto, N.;
Tomioka, K.; Kurumatani, N. The relationship between indoor,
outdoor and ambient temperatures and morning BP surges from interseasonally repeated measurements. J. Hum. Hypertens. 2014, 28 (8),
482488.
(14) Fabi, V.; Andersen, R. V.; Corgnati, S. P.; Olesen, B. W. A
methodology for modelling energy-related human behaviour:
Application to window opening behaviour in residential buildings.
Build. Simul. 2013, 6 (4), 415427.
(15) Bohm, M.; Salem, M. Z.; Srba, J. Formaldehyde emission
monitoring from a variety of solid wood, plywood, blockboard and

ASSOCIATED CONTENT

S Supporting Information
*

The Supporting Information is available free of charge on the

ACS Publications website at DOI: 10.1021/acs.est.5b02217.
Additional details on the accuracy of the MBTH method
and comparison results with HPLC (Page S2, Figure S1).
Quality control and assurance of the MBTH method
(Page S2S3). Environmental settings and procedures of
the small environmental chamber test (Page S3S4,
Table S1). (PDF)

REFERENCES

AUTHOR INFORMATION

Corresponding Author

*Telephone: (86)10-62788845. Fax: (86)10-62773461. E-mail:

xyang@tsinghua.edu.cn.
Notes

The authors declare no competing nancial interest.

ACKNOWLEDGMENTS
This study is supported by the National Natural Science
Foundation of China Project No. 51178237 and Shenzhen
Institute of Building Research.
G

DOI: 10.1021/acs.est.5b02217
Environ. Sci. Technol. XXXX, XXX, XXXXXX

Article

Downloaded by UNIV VERACRUZANA on August 26, 2015 | http://pubs.acs.org

Publication Date (Web): August 21, 2015 | doi: 10.1021/acs.est.5b02217

Environmental Science & Technology

flooring products manufactured for building and furnishing materials.
J. Hazard. Mater. 2012, 221-222, 6879.
(16) Kim, K. W.; Kim, S.; Kim, H. J.; Park, J. C. Formaldehyde and
TVOC emission behaviors according to finishing treatment with
surface materials using 20L chamber and FLEC. J. Hazard. Mater.
2010, 177 (1), 9094.
(17) Brown, S. K. Chamber assessment of formaldehyde and VOC
emissions from wood-based panels. Indoor Air 1999, 9 (3), 209215.
(18) Yang, X.; Chen, Q.; Zhang, J. S.; Magee, R.; Zeng, J.; Shaw, C. Y.
Numerical simulation of VOC emissions from dry materials. Build.
Environ. 2001, 36 (10), 10991107.
(19) Huang, H.; Haghighat, F. Modelling of volatile organic
compounds emission from dry building materials. Build. Environ.
2002, 37 (12), 13491360.
(20) Xu, Y.; Zhang, Y. An improved mass transfer based model for
analyzing VOC emissions from building materials. Atmos. Environ.
2003, 37 (18), 24972505.
(21) Wolkoff, P.; Clausen, P. A.; Nielsen, P. A.; Molhave, L. The
Danish twin apartment study; part I: formaldehyde and long-term
VOC measurements. Indoor Air 1991, 1 (4), 478490.
(22) Brown, S. K. Air toxics in a new Australian dwelling over an 8month period. Indoor Built Environ. 2001, 10 (34), 160166.
(23) Crump, D. R.; Squire, R. W.; Yu, C. W. F. Sources and
concentrations of formaldehyde and other volatile organic compounds
in the indoor air of four newly built unoccupied test houses. Indoor
Built Environ. 1997, 6 (1), 4555.
(24) Petitjean, M.; Hantal, G.; Chauvin, C.; Mirabel, P.; Le Calve, S.;
Hoang, P.; Picaud, S.; Jedlovszky, P. Adsorption of benzaldehyde at
the surface of ice, studied by experimental method and computer
simulation. Langmuir 2010, 26 (12), 95969606.
(25) Salthammer, T.; Fuhrmann, F. Photocatalytic surface reactions
on wall paint. Environ. Sci. Technol. 2007, 41 (18), 65736578.
(26) ASTM D 6007-02. Standard Test Method for Determining
Formaldehyde Concentration in Air from Wood Products Using a Small
Scale Chamber; American Society for Testing and Materials: West
Conshohocken, PA, 2008.
(27) Yao, Y. Research on Some Key Problems of Furniture VOC
Emission Labeling System. Ph.D. Thesis, Tsinghua University, 2011.
(28) Nazaroff, W. W.; Weschler, C. J. Cleaning products and air
fresheners: exposure to primary and secondary air pollutants. Atmos.
Environ. 2004, 38 (18), 28412865.
(29) Nrgaard, A. W.; Kofoed-Srensen, V.; Mandin, C.; Ventura, G.;
Mabilia, R.; Perreca, E.; Cattaneo, A.; Spinazze, A.; Mihucz, V. G.;
Szigeti, T.; Kluizenaar, Y.; Cornelissen, H. J. M.; Trantallidi, M.;
Carrer, P.; Sakellaris, I.; Bartzis, J.; Wolkoff, P. Ozone-initiated terpene
reaction products in five European offices: replacement of a floor
cleaning agent. Environ. Sci. Technol. 2014, 48 (22), 1333113339.
(30) Methods for Determination of Formaldehyde in Air of Public Places;
GB/T18204.26-2000; Beijing: Administration of Quality Supervision,
Inspection and Quarantine, 2000 (in Chinese).
(31) Van Hooff, T.; Blocken, B. CFD evaluation of natural
ventilation of indoor environments by the concentration decay
method: CO2 gas dispersion from a semi-enclosed stadium. Build.
Environ. 2013, 61, 117.
(32) Cox, S. S.; Little, J. C.; Hodgson, A. T. Predicting the emission
rate of volatile organic compounds from vinyl flooring. Environ. Sci.
Technol. 2002, 36 (4), 709714.
(33) Qian, K.; Zhang, Y.; Little, J. C.; Wang, X. Dimensionless
correlations to predict VOC emissions from dry building materials.
Atmos. Environ. 2007, 41 (2), 352359.
(34) Xiong, J.; Wei, W.; Huang, S.; Zhang, Y. Association between
the emission rate and temperature for chemical pollutants in building
materials: general correlation and understanding. Environ. Sci. Technol.
2013, 47 (15), 85408547.
(35) Parthasarathy, S.; Maddalena, R. L.; Russell, M. L.; Apte, M. G.
Effect of temperature and humidity on formaldehyde emissions in
temporary housing units. J. Air Waste Manage. Assoc. 2011, 61 (6),
689695.

(36) Frihart, C. R.; Wescott, J. M.; Chaffee, T. L.; Gonner, K. M.

Formaldehyde emissions from urea-formaldehyde-and no-addedformaldehyde-bonded particleboard as influenced by temperature
and relative humidity. For. Prod. J. 2012, 62 (7), 551558.
(37) Sidheswaran, M.; Chen, W.; Chang, A.; Miller, R.; Cohn, S.;
Sullivan, D.; Fisk, W. J.; Kumagai, K.; Destaillats, H. Formaldehyde
emissions from ventilation filters under different relative humidity
conditions. Environ. Sci. Technol. 2013, 47 (10), 53365343.
(38) Myers, G. E. The effects of temperature and humidity on
formaldehyde emission from UF-bonded boards-A literature critique.
Forest Prod. J. 1985, 35 (9), 2031.
(39) Lin, C. C.; Yu, K. P.; Zhao, P.; Lee, G. W. M. Evaluation of
impact factors on VOC emissions and concentrations from wooden
flooring based on chamber tests. Build. Environ. 2009, 44, 525533.
(40) Huang, H.; Haghighat, F.; Blondeau, P. Volatile organic
compound (VOC) adsorption on material: influence of gas phase
concentration, relative humidity and VOC type. Indoor Air 2006, 16
(3), 236247.
(41) Suzuki, M.; Akitsu, H.; Miyamoto, K.; Tohmura, S.; Inoue, A.
Effects of time, temperature, and humidity on acetaldehyde emission
from wood-based materials. J. Wood Sci. 2014, 60 (3), 207214.

DOI: 10.1021/acs.est.5b02217
Environ. Sci. Technol. XXXX, XXX, XXXXXX