- Email: [email protected]
Investigation of dust loading and culturable microorganisms of HVAC systems in 24 office buildings in Beijing
Accepted Manuscript Title: Investigation of dust loading and culturable microorganisms of HVAC systems in 24 office buildings in Beijing Author: Zhijian Liu Zunqiang Zhu Yexuan Zhu Wei Xu Hao Li PII: DOI: Reference:
S0378-7788(15)30085-2 http://dx.doi.org/doi:10.1016/j.enbuild.2015.06.056 ENB 5962
To appear in:
ENB
Received date: Revised date: Accepted date:
4-4-2015 19-6-2015 20-6-2015
Please cite this article as: Z. Liu, Z. Zhu, Y. Zhu, W. Xu, H. Li, Investigation of dust loading and culturable microorganisms of HVAC systems in 24 office buildings in Beijing, Energy and Buildings (2015), http://dx.doi.org/10.1016/j.enbuild.2015.06.056 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Investigation of dust loading and culturable microorganisms of HVAC systems in 24 office buildings in Beijing
ip t
Zhijian Liua, Zunqiang Zhua, Yexuan Zhua, Wei Xub, Hao Lic a
Department of Power Engineering, North China Electric Power University, Baoding, Hebei 071003, PR
China Institute of Building Environment and Energy, China Academy of
us
b
cr
China
Building Research, Beijing 100013, P.R. China c
an
College of Chemistry, Sichuan University, Chengdu, Sichuan 610064, PR China
M
The corresponding author Email: [email protected] Fax number: +86 0312 7522443;
d
Phone: +86 0312 7522242
Ac ce pt e
Abstract: To investigate the dust loading and culturable microorganisms contamination characteristics of HVAC systems in 24 office buildings, a series of field tests, which included temperature, RH, air velocity, dust loading, culturable fungi/bacteria loading, were conducted in the following: return air, fresh air, mixture air, cooling, and supply air segments. On basis of these measuring results, the culturable fungi/bacteria number per gram dust was calculated and the predominant culturable fungi/ bacteria species was identified. Statistical analysis showed that dust loading, culturable fungi/bacteria loading, culturable fungi/bacteria number per gram dust were in the range of 3.25~48.25g/cm2, 32~221CFU/cm2, 46~232CFU/cm2, 53815~124807CFU/gdust, and 63395~10383CFU/gdust respectively. The maximum value for dust loading and culturable fungi /bacteria loading appeared in fresh air segments. The predominant culturable fungi were Penicillium, Aspergillus, Cladosporium and Alternaria and the predominant culturable bacteria were Micrococcus, Bacillus,
Page 1 of 29
Staphylococcus, and Pseudomonas. There were significant positive correlations between dust and culturable fungi loading, dust and culturable bacteria loading, culturable fungi number per gram dust and RH, bacteria number per gram dust and temperature (p<0.05). Results of these field measurements indicated that dust
ip t
accumulation and/or high humidity and/or temperature should be properly controlled in HVAC systems to prevent the growth of culturable fungi and bacteria.
cr
Keywords: HVAC system; Dust loading; Culturable fungi; Culturable bacteria; RH; Temperature
us
1. Introduction
The interaction between dust and bio-aerosol retained in HVAC and indoor air quality has gained more
an
attention due to their possible relationship to irritation, health outcomes, and odors [1], [2] and [3]. The HVAC
M
systems was usually used to provide fresh and to clean air and remove the pollutants caused by the activity of occupants, the equipment of furniture and the materials of decoration[4]. Through normal indoor office activity
d
and occupation, a great deal of contaminants and air pollutants, such as dander, dust, and chemicals could be
Ac ce pt e
generated [5]. These contaminants could pull into the HVAC system and re-circulated 5 to 7 times per day, on average [6]. This re-circulation could cause a build-up of contaminants in HVAC system. When the fresh air could not be treated and purified appropriately, the HVAC system could become more seriously polluted and act as the source of indoor air pollution, which could further pose a severe threat to the office worker’s health [7] As deposition dust and culturable microorganism are suggested as the important pollution parameters in HVAC system, which might be closely associated to officers’ SBS and BRI [8] and [9], it is therefore very significant to investigate the characterization of dust loading and culturable microorganism in HVAC systems of office buildings. For culturable microorganism and dust in the HVAC systems, there has been a lot of research performed in several aspects, such as, the aerosolization of the fungi [10] and [11], the movement and dynamics
Page 2 of 29
characteristic of the dust in ducts work [12], the microorganism growth on the surface of filter [13]. Gao et al. compared the modeling deposition of dust in vertical square HVAC duct flows and concluded that the gravity of particles does not directly change the dimensionless deposition velocity in vertical flows [14]. Wu et al.
ip t
discussed the effect of ventilation duct as a particle filter and concluded that the filtration of the duct might not be ignored as the obvious loss by deposition when particle diameter is larger than 5μm [15]. Chang et.al
cr
evaluated the susceptibility of three types of ventilation duct materials (fibrous glass ductboard, galvanized
us
steel, and insulated flexible duct) to fungal growth [16]. Noris et.al discussed microbial contaminants and metals captured on HVAC filters in one commercial building and eight residential [17]. Forthomme et al.
an
examined the conditions (such as air relative, the airflow after microbial contamination of filter and the nature
M
of fibrous) leading to microbial development on to fibrous filters and to microbial release downstream of filters[18].Stephen noted that Cladosporium sp. fungi was commonly recovered in terms of growth sites and
d
deposited spores, and that were found mainly in the blower wheel fan blades, the ductwork, and the cooling coil
Ac ce pt e
fin in air handling units, and concluded that there was no relationship between mold growth and operating condition[19]. Pasanen investigated the accumulated dust on the air duct of 24 houses which had mechanical ventilation and found that Cladosporium, Penicillium, Aspergillus, and Yeasts made up more than 90% of fungi[20]; Li et.al investigated the effect of air-conditioning parameters (including temperature, relative humidity, and air velocity) and the deposition dust on microbial growth in supply air ducts[21]. Chen et al. from Tsinghua University indicated that 82% HVAC systems were moderately and seriously polluted [22]. Lu et.al investigate microorganisms and particles in AHU systems, and the results found that the microbial growth on surfaces of duct and equipment was noticed , and that it may be transferred into indoor air, which could lead to adverse health effect[23]. The previous researchers typically paid close attention to the dust deposition mechanism or microorganism
Page 3 of 29
growth phenomenon in only one segment, such as duct, filter, fan, and coil, to date and to our best knowledge, but nearly no related studies concerning the dust loading and culturable microorganism pollution characteristics in the integrated HVAC systems were conducted. Indeed, almost no comprehensive research has been
ip t
conducted on the dust loading and culturable microorganisms contamination characteristics in the integrated HVAC systems of office buildings in Beijing. Although some international studies have pointed out that the
cr
HVAC systems for indoor environment were just like the lungs for human, which were closely related to
us
occupants’ health [1], [2] and [3], dust and culturable microorganisms pollution problems in HVAC systems in China have not drawn sufficient attention due to the lack of sufficient field measurement data. The aim of this
an
work is to investigate the dust loading and culturable microorganisms (including fungi and bacteria)
M
contamination characteristics of HVAC systems in 24 office buildings, and to provide basic data and reference
2. Method
Ac ce pt e
2.1 Study design
d
for improving healthy office buildings indoor environment.
The twenty-four HVAC systems of office buildings, located in downtown were selected as measurement objectives. These HVAC systems and air filters were cleaned two years before the current testing, and the filters had been in operation without clean for two years. The outdoor air quality, including PM10 and PM2.5 mass concentration, for last two years in Beijing (2013.7-2015.6) , had been shown in the following Fig.1. The general description of sampling sites and HVAC information (including year built, air filter category, area, HVAC locations，supply air volume，percentage of fresh air, and occupants number) were shown in Table.1. All the buildings were relatively close to main streets and located in commercial areas. The centralized air-conditioning systems with primary return air in use in all the office buildings. The sampling sites were determined in return air segment, fresh air segment, mixture air segment, cooling segment
Page 4 of 29
and supply air segment, as shown in Fig.2. Measurement parameters included temperature, RH, air velocity, dust loading, fungi/bacteria loading. Two parallel samplings were performed for each sampling point. All the HVAC systems were kept in the state of normal operation during the sampling process. The field measurement
ip t
actions were conducted in summer (July 6-September 15) when the weather was allowable, without rain and the wind speed was below 8.0 m/s
cr
2.2 Measuring methods
us
A TSI multi-channel portable tester, model 8392A, was used to measure the temperature, RH and air velocity. The dust weight was quantified by an electronic balance, model 9367S. The test range and precision of
an
these four parameters were shown in Table.2.
M
The dust loading was tested by gravimetrical techniques according to the Chinese Hygienic Standard (WS394-2012) [24]. A sterile non-woven fabric was employed to wipe the deposition dust with an area equal to
d
50 cm2. Before the sampling, the sterile no-woven fabric was put into the sealing bags in advance. All sealing
Ac ce pt e
bags with no-woven fabric were together dried for 4 hours in a silica drier, and then the initial weight of them were obtained by the calculation of the average values [24]. After non-woven fabric wiped up the dust in sampling location, it would be the put back to the sealing bag. The final average weight could also be obtained by twice weighting. The weight of dust would be the difference of the initial and the final values. Therefore, the dust loading could be calculated by the following equation (1):
Ldsut =
ΔM S
(1)
where, Ldsut (g/cm2) is the dust loading, ΔM (g) is the difference of initial and final weight and S (cm2) is the wiping area of the inner surface. The fungi and bacteria loading were tested by the wiping method. Firstly, the sterile non-woven fabric was used to wipe an area equal to 50 cm2. After non-woven fabric wiped up the dust in the sampling location, it
Page 5 of 29
would be put back into the sealing bag. It was imperative to get the sealing bags back to the lab within six hours in order to keep the microorganism viability [22]. The dust sample on the non-woven fabric was dissolved in 100mL of 0.01% Tween 80 by hand-shaking, and then 1ml solution was incubated on culture petri dish. The
ip t
analysis for each dilution was performed three times and the average number of colonies formed was recorded. The bacteria petri dishes were incubated at 36℃for 48 hours on Tryptic Soy Agar (TSA), then the colonies
cr
were counted. The fungi petri dishes were incubated at 28℃ for 5 days on the Sabourand’s agar (SDA), and
us
then the number of colonies were enumerated [23] and [25]. Blank samples were required in setting up each sampling process. The test was proved to be effective only when no colony occurred on blank sample. The
N fungi /bacteria ×100
(2)
S
M
L fungi / bacteria =
an
bacteria and fungi loading could be calculated by the following equation (2).
where, L fungi / bacteria (CFU/cm2) is the fungi/bacteria loading, N fungi /bacteria (CFU) is the number of colony forming
Ac ce pt e
d
unit fungi or bacteria, S (cm2) is the wiping area of inner surface. Because the dust sample on the non-woven fabric was dissolved in 100mL of 0.01% Tween 80, only 1ml solution was incubated on a petri dish, therefore the results should be multiplied by 100.The fungi/bacteria number per gram dust could be calculated by the following equation [26].
ϕ fungi / bacteria =
L fungi / bacteria Ldust
(3)
where, ϕ fungi / bacteria (CFU/g) is the fungi/bacteria loading, and L fungi / bacteria and Ldust had been defined in the previous chapter. 2.3 Microorganism Identification Method The bacteria and fungi from the experimental incubation process were identified by the morphology method, which was employed in previous studies [18] and [23]. The petri dishes were examined by optical
Page 6 of 29
methods by an experienced analyst to identify and quantify the fungi growth on the petri dishes. Identification was conducted according to morphological features of the fungi and bacteria. An optical microscope (XSP-63B) and a fluorescent optical microscope (PF6200) were employed in this process. The fungi spores were observed
ip t
by electron microscope in case it was needed. 2.4 Statistical Analysis Methods
cr
The mean, median, 25th percentile，75th percentile, and geometric standard deviation (GSD) were used to
us
describe the data’s characteristics. The Pearson product-moment correlation coefficient was used to explore the strength of the linear relationship between each two sets of measuring variables in our study.
an
3. Results and Discussion
M
The field measurement results in the 24 HVAC systems of office buildings are summarized in Table.3. All the parameters including temperature, RH, air velocity, dust loading, fungi loading, bacteria loading, fungi
d
number per gram dust, and bacteria number per gram dust in different sites ,such as return air segment, fresh air
Ac ce pt e
segment, mixture air segment, cooling segment and supply air segment were described in terms of both mean and geometric standard deviation. The temperature in these five sampling sites (R, F, M, C and S) was in range of 21.4-28.5, 30.2-35.6, 26.1-30.2, 16.2-18.9, and 18.1-21.8℃, respectively. The RH in these five segments was in the range of 52-68%, 71-82%, 59-73%, 89-94%, and 83-89%, respectively. The velocity in these five segments was in the range of 5.8～8.1, 2.2～4.5 2.8～3.8, 1.4～2.9, and 5.1～7.8 m/s, respectively. 3.1. Dust loading
The mean values and geometric average deviation (GSD) of the dust loading in the five different segments are shown in Table.3.The maximum dust loading occurred in fresh air segments, being followed, sequentially, by mixture air segments, return air segments, cooling segments, and supply air segments. The mean value of dust loading was 33.37g/m2 in the fresh air segments and was about six times higher than that of the supply air
Page 7 of 29
segments. The average dust loading in these five segments was in the range of 12.45-22.43, 19.34-48.25, 14.89-32.14, 4.89-12.98, and 3.25-8.89 g/m2, respectively. The data distribution of dust loading (including min, max, medium, 25th percent, and 75th percent) in these five segments is shown in Fig.3.The threshold value of
ip t
dust loading in the HVAC system is less than 20 g/m2 according to the Chinese standard[24]. The exceeding percentages in the return air segments, fresh air segments, and mixture air segment were 43.75%, 93.75% and
cr
87.5% respectively. All samplings for the dust loading in cooling and supply air segments were less than the
us
threshold value. According to the HVAC’s system recorded operating time and to the dust loading, the dust loading rates could be calculated. The dust loading rates in the five sampling sites (R, F, M, C, and S) was in
an
the range of 0.519-0.935, 0.806-2.011, 0.621-1.339, 0.204-0.541, and 0.135-0.371 g/m2·month. As reported by
M
Waring et al., the median predicted residential and commercial loading rates were 0.0051 and 1.00 g/m2·month for the supply duct loading rates, and 0.262 g/m2·month for the commercial return ducts[27]. However, the
d
current result is distinctly different from the predicted results, in terms of loading rates for the return segments.
Ac ce pt e
The discrepancies of the specific value would be mainly likely to be due to the fact that the difference for the indoor sources exist, depending on outdoor particle distributions, indoor sources, HVAC operation strategy, filtration, and so on [28] and [29].The dust loading in the return air segments was significantly higher than that of the supply air segments, and this indicates that there was some indoor dust pollution sources, as identified in the many other many studies[3], [6], [28] [29] and [30]. More attention should be paid on the return air segments and some purified equipments should be installed, if necessary. The dust loading in the fresh air segments was the highest, and implication would be that the fresh air might be the important dust pollution source and a higher efficiency purifier should be employed at the fresh air inlet. 3.2. Culturable fungi/bacteria loading The culturable fungi/bacteria loadings, in terms of box-plot, in different segments are shown in Fig.4 and 5.
Page 8 of 29
The threshold value of culturable fungi/bacteria loading in the HVAC system is less than 100 CFU/cm2, according to Chinese standard [24].The maximum value of fungi/bacteria loading also occurred in fresh air segments, being followed, sequentially, by mixture air segments, return air segments, cooling segments, and
ip t
supply air segments. This was similar to the distribution of the dust loading. The culturable fungi loading in these five segments were in the range of 67-138, 112-221, 89-191, 46-162, and 32-148 CFU/cm2, and the
cr
corresponding exceeding percentages was 43.75%, 100%, 87.5%, 50%, and 25%, respectively. The mean value
us
of culturable fungi loading was 170 CFU/cm2 in the fresh air segments and was about 2.11 times higher than that of supply air segment. The culturable bacteria loading in these five segments were in the range of 67-138,
an
112-232, 89-192, 46-162, and 31-148 CFU/cm2, and the corresponding exceeding percentages were 68.75%,
M
100%, 100%, 6.25%, and 0%, respectively. The mean maximum of culturable fungi loading was 171 CFU/cm2 in fresh air segment and was about 2.67 times higher than that of the supply air segments. Meng et al.’s results
d
indicated that the exceeding percentages of fungi/ bacteria loading in the eight points on the HVAC duct
Ac ce pt e
surface before cleaning was 100%, and that values seemed much higher than our study [31].The culturable fungi loadings in the HVAC system was from 31-192 CFU/cm2 in this study, which was in the range of 1-261 CFU/cm2 reported by Lappalainen et al.[32]. The culturable bacteria loading in this study, ranging from 32-221 CFU/cm2, was similar to the Li et al.’s findings, and that was in the range of 24 to 204 CFU/cm2 [33]. All the values of dust loading were below the threshold value in cooling segments and supply air segments, while a few values of culturable fungi loading exceeded the threshold, this could be shown by comparing Fig.3 and 4.This indicated that the culturable fungi loading should draw more attention even though the dust loading met requirements. 3.3. Culturable fungi/bacteria number per gram dust The culturable fungi/bacteria number per gram dust in different segments is shown in Fig.6 and 7. The
Page 9 of 29
largest culturable fungi number per gram dust occurred in the cooling segments, being followed, sequentially, by the supply air segments, fresh air segments, mixture air segments and return air segments, which was different from the sequence of dust loading and culturable fungi/bacteria loading. The culturable fungi loading
ip t
in these five segments was in the range of 53815-61525, 80144-88082, 59771-76540, 94070-124807, and 98462-114735 CFU/gdust, respectively. The mean value of the culturable fungi loading was 109525 CFU/gdust in
cr
the cooling segments and was about 1.88 times higher than that of the return air segments. The largest
us
culturable bacteria number per gram dust also appeared in fresh air segment, followed, sequentially, by the mixture air segments, supply air segments, return air segments, and cooling air segments. The culturable
an
bacteria number per gram dust in these five segments was in the range of 81141-89960, 92037-103834,
M
90665-99876, 63395-79353, and 70769-93793 CFU/gdust. The mean value of culturable bacteria loading was 97302 CFU/gdust in th fresh air segments and was about 1.37 times higher than that of the cooling segments. The
d
culturable fungi/bacteria number per gram dust was from 105-106, lying in the range of 104-106 fungi number
Ac ce pt e
and 105-107 bacteria number per gram dust reported by Noris et al.[17]. Pasanen et al. investigated the fungal spore content of dust accumulated in the air ducts of 24 mechanically ventilated single-family homes, and the results showed that culturable spore concentrations in the dust varied from 104 -107 CFU/gdust [34], which was much broader than our results. Another experiment investigation conducted by Li et al. assumed that the culturable fungi/bacteria number per gram dust was in the range of 25000-60000CFU/gdust, which was slightly lower than our results [21]. Although the values of culturable fungi/bacteria number per gram dust in our study were slightly different from the related above references, it greatly exceeded the threshold value (50000 CFU/g for culturable bacteria and culturable fungi), as regulated in Chinese standard [24]. Some specific measurements, such as disinfecting and cleaning, should be taken to reduce the contamination of the HVAC systems in office buildings in subsequent management and operation.
Page 10 of 29
3.4. Culturable fungi/bacteria species identification and analysis The predominant culturable fungi/bacteria species from the field measurements were identified by the morphology method in the lab. The morphological characteristics for six species of fungi are shown in Fig.8
ip t
and the morphological characteristics for four species of bacteria are shown in Fig.9. The statistics of the percentages and occurrences for the different culturable fungi and bacteria in these five segments (R, F, M, C
cr
and S) are listed in Table.4 and Table.5. The predominant culturable fungi were Penicillium, Aspergillus,
us
Cladosporium, and Alternaria due to the higher occurrences. Similar findings have been noted by Li et al.[35].There was a little discrepancy when comparing to Lu et al.’s results, which was that the predominant
an
species were Cladosporium and Penicillium in AHU systems[23]. The predominant culturable bacteria were
Pastuszka et al.’s observation [36].
d
3.5. Correlation analysis
M
Micrococcus, Bacillus, Staphylococcus, and Pseudomonas. Similar interactions have been described in the
Ac ce pt e
The correlation between each two measuring parameters has been explored in this section, as shown in Table.6. There was a significant positive correlation between dust loading and culturable fungi/bacteria loading (p<0.05), which indicated that the culturable microorganism in the HVAC system is partially determined by the accumulated dust. Similar findings, which discussed the correlation between fungi and PM2.5 in the atmosphere environment has been noted by Degobbi et al.[37]. In light of the results by Li et al., the deposition dust in the supply air ducts presented a positive correlation between the microorganism loadings, which was consistent with our findings [21]. There was a significant positive correlation between culturable fungi number per gram dust and RH (p<0.05), however, not a significant correlation between fungi loading and RH. There was a significant positive correlation between culturable bacteria number per gram dust and temperature (p<0.05), but not a significant correlation between bacteria loading and temperature. These results could be explained as
Page 11 of 29
follows: the effect of temperature and RH on microorganism growth could be revealed only after the factor of dust loading was excluded. RH and temperature contributed to the microorganism growth has been stated by previous studies [35], [36], [38], [39] and [40]. However, few conflicting studies have presented that there are
ip t
no significant correlations between RH, temperature and microorganism concentration [19]. There is no significant correlation between air velocity and culturable microorganism, and these similar results have been
cr
addressed by Forthomme et al. [13]. The results of this investigation suggested that culturable fungi were more
us
sensitive to RH and the culturable bacteria were more sensitive to temperature in the HVAC’s operating environment. Therefore, dust accumulation, high humidity, and/or temperature should be properly controlled in
an
the HVAC systems to prevent the growth of microorganism on the premise of meeting the requirements of
3.6. Limitation
M
operating.
d
So far, this study is still preliminary and needs to be confirmed by future work. There were two limitations
Ac ce pt e
listed in the following. Firstly, every species of fungi and bacteria might grow/die in any given environmental and nutrient conditions. Some species will dominate the microbial community after a very short period of time. Spores may be unable to grow in a culture because they are nonviable, or they are viable, but they are injured, or because the nutrients or temperature conditions are not appropriate [41]. These phenomena can be influenced by sample storage, transport, and age factors. Secondly, only the culturable fractions of the microbial population were measured in our study. According to the findings by Tovila et al., only a small part of microbial population indoors were culturable and a much greater fraction of the microbial community could be detected by molecular biological techniques [42] and [43]. 4. Conclusions The dust loading and culturable microorganism distribution characteristics in the HVAC systems of office
Page 12 of 29
building were investigated and the following conclusions could be drawn from this study: (1) The distribution characteristics of dust and culturable fungi/bacteria loading was similar; namely, all the maximum values occurred in the fresh air segments, being followed, sequentially, by mixture air segments,
ip t
return air segments, cooling segments and supply air segments. Such a trend indicates that the fresh air might be source of dust pollution and higher efficiency purifier should be employed at the inlet of fresh air.
cr
(2) The fungi/bacteria loading might exceed the threshold when the dust loading was below the threshold in
attention even though the dust loading met requirements.
us
the cooling and supply air segments, which indicated that the culturable fungi loading should draw more
an
(3) The predominant culturable fungi was Penicillium, Aspergillus, Cladosporium, and Alternaria and the
M
predominant culturable bacteria was Micrococcus, Bacillus, Staphylococcus, and Pseudomonas, respectively, in the HVAC systems of the office buildings in Beijing.
d
(4) There was a significant positive correlations between dust and fungi loading, dust and bacteria loading,
Ac ce pt e
fungi number per gram dust and RH, bacteria number per gram dust and the temperature (p<0.05), but no significant correlation between air velocity and culturable microorganism. The culturable fungi was more sensitive to RH, and the culturable bacteria was more sensitive to the temperature in the HVAC’s operating environment, after excluding the influence of dust loading. On basis of these research results, the potential specific solutions could be obtained as follows. (1)
The dust accumulation and/or high humidity and/or temperature should be properly controlled in
HVAC systems to prevent the growth of culturable fungi and bacteria. (2) Regular cleaning and disinfection work for the HVAC systems should be conducted in avoid of air contamination occurrence. (3) Appropriate air filter category selection in office building should take the outdoor air quality into
Page 13 of 29
consideration, which could be the most important step to achieve good IAQ in HVAC design. Nevertheless, these conclusions were obtained on the basis of a relative size sample of 24 HVAC systems. This investigation was just aimed at the office buildings in Beijing. More detailed and in depth studies are
ip t
required in future works to verify these results with a larger sample population. Future studies are also needed to address the tests of airborne fungi/bacteria spores and develop of cleaning and disinfection strategies in the
cr
HVAC systems in China.
us
Acknowledgement
This study is supported by the Fundamental Research Funds for the Central Universities, No. 2015MS108.
an
References
M
[1] B.F. Yu, Z.B. Hu, M. Liu, H.L. Yang, Q.X. Kong, Y.H. Liu, Review of research on air-conditioning systems and indoor air quality control for human health, International journal of refrigeration 32 (1) (2009)
d
3-20.
Ac ce pt e
[2] P.M. Bluyssen, C. Cox, O. Seppänen, E.D. Oliveira Fernandes, Why, when and how do HVAC-systems pollute the indoor environment and what to do about it? The European AIRLESS project, Building and Environment 38 (2) (2003) 209-225.
[3] F. Bitter, K. Fitzner, Odour emissions from an HVAC-system, Energy and buildings 34 (8) (2002) 809-816.
[4] O. Seppänen, Ventilation strategies for good indoor air quality and energy efficiency, International Journal of Ventilation 6 (4) (2008) 297-306. [5] M. Ongwandee, R. Moonrinta, S. Panyametheekul, C. Tangbanluekal, G. Morrison, Investigation of volatile organic compounds in office buildings in Bangkok, Thailand: Concentrations, sources, and occupant symptoms, Building and Environment 46(7) (2011) 1512-1522.
Page 14 of 29
[6] Z. Bakó‐Biró, P. Wargocki, C.J. Weschler, P.O.Fanger, Effects of pollution from personal computers on perceived air quality, SBS symptoms and productivity in offices, Indoor air 14 (3) (2004) 178-187. [7] J.H. Choi, V. Loftness, A. Aziz, Post-occupancy evaluation of 20 office buildings as basis for future IEQ
ip t
standards and guidelines, Energy and buildings 46 (2012) 167-175. [8] S. Gupta, M. Khare, R. Goyal, Sick building syndrome—A case study in a multistory centrally
cr
air-conditioned building in the Delhi City, Building and Environment 42 (8) (2007) 2797-2809.
us
[9] J. Sundell, T. Lindvall, B. Stenberg, Associations between type of ventilation and air flow rates in office buildings and the risk of SBS-symptoms among occupants, Environment International 20 (2) (1994) 239-251.
an
[10] R.L. Gorny, T. Reponen, S.A. Grinshpun, K. Willeke, Source strength of fungal spore aerosolization from
M
moldy building material, Atmospheric Environment 35 (2001) 4853–4862. [11] S.K. Sivasubramani, R.T. Niemeier, T. Reponen, S.A. Grinshpun, Assessment of the aerosolization
d
potential for fungal spores in moldy homes, Indoor Air 14 (2004) 405–412.
Ac ce pt e
[12] J.P. Zhang, A.G. Li, D. Li, Modeling deposition of particles in typical horizontal ventilation duct flows, Energy Conversion and Management 49 (12) (2008) 3672-3683. [13] F. Noris, J.A. Siegel, K.A. Kinney, Evaluation of HVAC filters as a sampling mechanism for indoor microbial communities, Atmospheric environment 45 (2) (2011) 338-346. [14] R. Gao, A.G. L, Modeling deposition of particles in vertical square ventilation duct flows, Building and Environment 46 (1) (2011) 245-252.
[15] J. Wu, B. Zhao, Effect of ventilation duct as a particle filter, Building and environment 42 (7) (2007) 2523-2529. [16] J. Chang, K.K. Foarde, D.W. VanOsdell, Assessment of fungal (Penicillium chrysogenum) growth on three HVAC duct materials, Environment international 22 (4) (1996) 425-431.
Page 15 of 29
[17] F. Noris, J.A. Siegel, K.A. Kinney, Biological and Metal Contaminants in HVAC Filter Dust, ASHRAE Transactions 115 (2) (2009) 484-491. [18] A. Forthomme, A. Joubert, Y. Andrès, X. Simon, P. Duquenne, Microbial aerosol filtration: Growth and
ip t
release of a bacteria–fungi consortium collected by fibrous filters in different operating conditions, Journal of Aerosol Science 72 (2014) 32-46.
cr
[19] S.C. Wilson, R.N. Palmatier, L.A. Andriychuk, J.M. Martin, C.A. Jumper , H.W. Holder, D.C. Straus,
us
Mold contamination and air handling units, Journal of occupational and environmental hygiene 4 (7) (2007) 483-491.
an
[20] P. Pasanen, A. Nevalainen, J. Ruuskanen, P. Kalliokoski, The composition and location of dust settled in
M
supply air ducts, In 13th AIVC Conference.
[21] A.G. Li, Z.J. Liu, X.B. X.B. Zhu, Y. Liu, Q.Q. Wang, The effect of air-conditioning parameters and
d
deposition dust on microbial growth in supply air ducts, Energy and Buildings 42 (4) (2010) 449-454.
Ac ce pt e
[22] C.F. Chen, B. Zhao, X.D. Yang, Investigation and review of microbial pollution in air conditioning systems of public buildings, HV &AC 39 (2) (2009) 50–55. [23] Z. Lu, W.Z. Lu, J.L. Zhang, D.X Sun, Microorganisms and particles in AHU systems: Measurement and analysis, Building and Environment 44(4) 2009 694-698. [24] Ministry of Health of the People's Republic of China, WS 394-2012, Hygienic specification of central air conditioning ventilation system in public buildings, Beijing, 2012. [25] H.J. Oh, I.S. Nam, H. Yun, J. Kim, J. Yang, J.R. Sohn, Characterization of indoor air quality and efficiency of air purifier in childcare centers, Korea, Building and Environment 82 (2014) 203-214. [26] K. Ashley, G.T. Applegate, T.J. Wise, J.E. Fernback, M.J. Goldcamp, Evaluation of a standardized micro-vacuum sampling method for collection of surface dust, Journal of occupational and environmental
Page 16 of 29
hygiene 4(3) (2007) 215-223. [27] M.S. Waring, J.A. Siegel, Particle loading rates for HVAC filters, heat exchangers, and ducts, Indoor Air 18 (2008) 209-224.
ip t
[28] P. Wargocki, Z. Bako-Biro, G. Clausen, P.O. Fanger, Air quality in a simulated office environment as a result of reducing pollution sources and increasing ventilation, Energy and buildings 34(8) (2002) 775-783.
cr
[29] J.F. Montgomery, S.I. Green, S.N. Rogak, K. Bartlett, Predicting the energy use and operation cost of
us
HVAC air filters, Energy and Buildings 47 (2012) 643-650.
[30] I.Sarbu, C. Sebarchievici, Aspects of indoor environmental quality assessment in buildings, Energy and
an
Buildings 60 (2013) 410-419.
M
[31] C. Meng, Q.Wang, Y Song, Y. Cao, N. Zhao, Y. Shi, Experimental study on both cleaning effect and motion performance of the duct-cleaning robot, Sustainable Cities and Society 14 (2015) 64-69.
d
[32] S. Lappalainen, E. Kähkönen, P. Loikkanen E. Palomäki, O. Lindroos, K. Reijula, Evaluation of priorities
981-986.
Ac ce pt e
for repairing in moisture-damaged school buildings in Finland, Building and Environment 36 (8) (2001)
[33] A.G. Li, L.Z. Yao, J.J. Hou, Q.Q. Wang, M. Li, Field test and analysis of microbial concentration in central air condition systems, HV &AC 40 (3) (2010) 120–125. [34] A.L. Pasanen, L. Kujanpää, P. Pasanen, P. Kalliokoski, G. Blomquist, Culturable and Total Fungi in Dust Accumulated in Air Ducts in Single‐Family Houses, Indoor Air 7(2) (1997) 121-127. [35] A.G. Li, Z.J. Liu, Y. Liu, X. Xue, Y. Pu, Experimental study on microorganism ecological distribution and contamination mechanism in supply air ducts, Energy and Buildings 47 (2012) 497-505. [36] J.S. Pastuszka, U. Kyaw Tha Paw, D.O. Lis, A. Wlazło, K. Ulfig, Bacterial and fungal aerosol in indoor environment in Upper Silesia, Poland, Atmospheric Environment 34 (22) (2000) 3833-3842.
Page 17 of 29
[37] C. Degobbi, F.D. Lopes, R. Carvalho-Oliveira, J.E. Munoz, P.H. Saldiva, Correlation of fungi and endotoxin with PM2.5 and meteorological parameters in atmosphere of Sao Paulo, Brazil, Atmospheric environment 45(13) (2011) 2277-2283
prevalence child care centers, Atmospheric Environment 43 (15) 2009 2391-2400.
ip t
[38] M.S. Zuraimi, L. Fang, T.K. Tan, F.T. Chew, K.W. Tham, Airborne fungi in low and high allergic
cr
[39] C. He, H. Salonen, X. Ling, L. Crilley, N. Jayasundara, H.C. Cheung , L. Morawska, The impact of flood
us
and post-flood cleaning on airborne microbiological and particle contamination in residential houses, Environment international 69 (2014) 9-17.
an
[40] Z.J. Liu, A.G. Li, Z.P. Hu, H.F. Sun, Study on the potential relationships between indoor culturable fungi,
M
particle load and children respiratory health in Xi’an, China, Building and Environment 80 (2014) 105-114. [41] Z. Fang, Z. Ouyang, H. Zheng, X. Wang, Concentration and size distribution of culturable airborne
Ac ce pt e
325-334.
d
microorganisms in outdoor environments in Beijing, China, Aerosol Science and Technology 42 (5) (2008)
[42] M. Toivola, S. Alm, T. Reponen, S. Kolari, A. Nevalainen, Personal exposures and microenvironmental concentrations of particles and bioaerosols, Journal of Environmental Monitoring 4 (1) (2002) 166-174. [43] Y. Zheng , Y. Wen , A.M. George,
A. Steinbach, E. Freeman, B. Philips, N. Giannoukakis, L.M.
Weiland, E.S. Gawalt, and W.S. Meng. A Peptide-Based Material Platform for Displaying Antibodies to Engage T Cells. Biomaterials 32 (2011) 249-257.
Fig.1 The information of outdoor air quality in Beijing in the period of 2013.7-2015.6 Fig.2 The schematic diagram of sampling location in the HVAC system Fig.3 The dust loading distribution on the inner surface of the different segments
Page 18 of 29
Fig.4 The culturable fungi loading distribution on the inner surface of the different segments Fig.5 The culturable bacteria loading distribution on the inner surface of the different segments Fig.6 The culturable fungi number per gram deposition dust in the different segments Fig.7 The culturable
ip t
bacteria number per gram deposition dust in the different segments Fig.8 The morphological characteristics for 6 species of the culturable fungi identified Fig.9 The morphological
Ac ce pt e
d
M
an
us
cr
characteristics for 4 species of the culturable bacteria identified
Fig.1 The information of outdoor air quality in Beijing in the period of 2013.7-2015.6
Fig.2 The schematic diagram of sampling location in the HVAC system
Page 19 of 29
ip t cr us an
Ac ce pt e
d
M
Fig.3 The dust loading distribution on the inner surface of the different segments
Fig.4 The culturable fungi loading distribution on the inner surface of the different segments
Page 20 of 29
ip t cr us an
Ac ce pt e
d
M
Fig.5 The culturable bacteria loading distribution on the inner surface of the different segments
Fig.6 The culturable fungi number per gram deposition dust in the different segments
Page 21 of 29
ip t cr us an M
Ac ce pt e
d
Fig.7 The culturable bacteria number per gram deposition dust in the different segments
Cladosporium
Penicillium)
Aspergillus
Alternaria
Mucor
Trichoderma
Fig.8 The morphological characteristics for 6 species of the culturable fungi identified
Page 22 of 29
ip t cr us an M d Ac ce pt e
Micrococcus
Bacillus
Staphylococcus
Pseudomonas
Fig.9The morphological characteristics for 4 species of the culturable bacteria identified
Page 23 of 29
Table.1 General Description of the sampling sites and HVAC information
Roof
4
2
Ground
1 2 3 4 9
1 3 2 6 1
Roof Underground Ground Roof Roof
2
Ground
3 3
Underground Roof
1
10
2009
2010
11 12
14 15 16 17 18 19 20 21 22 23 24
2011
Ground
Ac ce pt e
13
2012
6 1 1 7 2 1 6 1 2 7 1
Underground Underground Roof Underground Roof Ground Underground Underground Roof Underground Underground
Percentage of Fresh air 40.01% 46.18%
24588
30.43%
404
35.58%
625
18858 22018 22267 12089 18848
48.97% 43.73% 41.19% 33.75% 19.89%
308 321 462 204 122
32069
30.01%
321
22458 13587
24.85% 34.67%
186 157
35505
28.47%
337
27522 10852 11205 8568 33255 24217 28199 12073 37545 26153 28949
32.92% 23.19% 21.15% 25.56% 33.67% 26.51% 30.32% 44.23% 41.15% 37.28% 34.63%
302 85 79 186 425 321 285 178 515 352 401
35127
M
3
Supply air volume(m3/h) 22251 23428
ip t
Roof Roof
Area (m2) 4515 8148 1108 8 1555 0 5852 9630 9702 4488 5928 1614 6 6292 3933 1542 5 8442 2014 2291 4650 9350 5778 6555 4628 8241 6336 7218
cr
2008
HVAC Location
us
3
MERV Values 1 2
an
Year Built
d
Buildings No. 1 2
Occupants 305 382
Table .2 The test range and precision for these four parameters.
Page 24 of 29
Parameters
Manufacturer
Range
Precision
Temperature
TSI
RH
TSI
0~95%
±1%
Air velocity
TSI
0~50m/s
±0.1m/s
Dust weight
UPGRN
0-3000g
±0.01
±0.1
us
cr
ip t
-17.8~93.3
R
F
M
C
S
M
Mean±GSD.
an
Table.3 The statistics of the environment parameters, dust loading, culturable fungi and bacteria in the 24 HAVC systems
Ac ce pt e
d
Temperature( ) 25.6±2.2 33.1±1.5 28.8±1.5 17.8±0.8 19.7±1.0 RH（%） 59±5 77±3 66±6 91±2 86±2 Air velocity（m/s） 7.0±0.7 3.5±0.7 3.4±0.3 2.0±0.4 6.7±0.7 2 Dust loading（g/m ） 16.79±2.59 30.79±7.69 27.47±7.14 9.35±2.83 6.28±1.98 2 Fungi loading(CFU/cm ) 91±22 170±33 141±34 102±31 85±34 110±15 170±32 144±26 78±12 64±13 Bacteria loading(CFU/cm2) Fungi number per gram dust (CFU/g) 58208±3159 84736±2483 68036±5612 109525±10321 103762±6427 Bacteria number per gram dust(CFU/g) 85280±2522 97302±3375 94185±2722 71173±4515 82819±7433 R-Return air segment, F-Fresh air segment, M-Mixture air segment, C-Cooling segment, S-Supply air segment.
Page 25 of 29
cr
ip t
Table.4 The statistics of the percentages and occurrences for the different species of culturable fungi Penicillium Aspergillus Cladosporium Alternaria Mucor Trichoderma Others Location Per Occ Per Occ Per Occ Per Occ Per Occ Per Occ Per Oc R 30.1 100 11.9 85 37.2 100 5.6 88 3.5 58 1.1 53 10.6 96 F 26.4 100 7.6 86 41.2 100 12.8 92 5.1 65 3.7 58 3.2 98 M 28.1 98 12.4 84 39.6 100 10.9 89 4.8 71 2.9 48 1.3 92 C 39.1 100 24.6 95 22.6 92 5.2 75 2.1 42 1.9 39 4.5 89 S 39.8 100 25.3 98 23.5 94 4.9 82 2.8 32 2.3 42 1.4 99
Ac ce pt e
d
M
an
us
Annotation: Per-Percentage, Occ-Occurrence, R-Return air segment, F-Fresh air segment, M-Mixture air segment, C-Cooling segment, S-Supply air segment
26
Page 26 of 29
R
Micrococcus Per Occ 28.9 95
Bacillus Per Occ 29.6 100
Staphylococcus Per Occ 17.6 92
Pseudomonas Per Occ 11.6 85
Others Per Occ 12.3 88
F M C
21.2 24.8 19.8
26.3 27.8 19.9
23.2 18.6 24.1
16.4 15.5 19.8
12.9 13.3 16.4
95 98 94
18.2 90 17.2 88 25.5 70 17.8 75 21.3 Table.5 The statistics of the percentages and occurrences for the different species of culturable bacteria
100
78 86 74
80 76 72
ip t
92 74 93
us
d
M
an
Annotation: Per-Percentage, Occ-Occurrence, R-Return air segment, F-Fresh air segment, M-Mixture air segment, C-Cooling segment, S-Supply air segment
Ac ce pt e
S
100 90 92
cr
Location
27
Page 27 of 29
Table.6 The pearson correlation among measuring parameters in the different segments of 24 HVAC systems
0.0568 0.2761 0.2125 0.2754 0.2179
0.3658*
1 -0.12 58 0.08 91 0.10 23 0.53 58* 0.10 83 0.23 12
1 -0.18 97 0.082 9 -0.06 58 0.125 8 -0.11 24
1
0.5478 6*
1
0.1253
0.2581
ip t
1
0.4896 *
0.2858
-0.127 8
1
0.1256
0.1291
0.1891
0.1025
1
Ac ce pt e
Air velocity Dust loading Fungi loading Fungi number Bacteria loading Bacteria number
Bacteri a numbe r per
cr
-0.1654
Bacteri a loading
us
RH
Fungi numbe r per gram
an
1
Dust Fungi loading loading
M
Temperat ure
RH
Air veloc ity
d
Var
Temper ature
Annotation: Italicised values indicate a significant (p < 0.05, 2-tailed) association between the paired variables
Contamination distribution characteristics of dust loading were investigated. Distribution characteristics of culturable fungi and bacteria were investigated. The predominant fungi was Penicillium, Aspergillus, Cladosporium and Alternaria. The predominant bacteria was Micrococcus, Bacillus, Staphylococcus and Pseudomonas. There was a significant positive correlation between dust and fungi loading.
28
Page 28 of 29
29
Page 29 of 29
d
Ac ce pt e us
an
M
cr
ip t
S0378-7788(15)30085-2 http://dx.doi.org/doi:10.1016/j.enbuild.2015.06.056 ENB 5962
To appear in:
ENB
Received date: Revised date: Accepted date:
4-4-2015 19-6-2015 20-6-2015
Please cite this article as: Z. Liu, Z. Zhu, Y. Zhu, W. Xu, H. Li, Investigation of dust loading and culturable microorganisms of HVAC systems in 24 office buildings in Beijing, Energy and Buildings (2015), http://dx.doi.org/10.1016/j.enbuild.2015.06.056 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Investigation of dust loading and culturable microorganisms of HVAC systems in 24 office buildings in Beijing
ip t
Zhijian Liua, Zunqiang Zhua, Yexuan Zhua, Wei Xub, Hao Lic a
Department of Power Engineering, North China Electric Power University, Baoding, Hebei 071003, PR
China Institute of Building Environment and Energy, China Academy of
us
b
cr
China
Building Research, Beijing 100013, P.R. China c
an
College of Chemistry, Sichuan University, Chengdu, Sichuan 610064, PR China
M
The corresponding author Email: [email protected] Fax number: +86 0312 7522443;
d
Phone: +86 0312 7522242
Ac ce pt e
Abstract: To investigate the dust loading and culturable microorganisms contamination characteristics of HVAC systems in 24 office buildings, a series of field tests, which included temperature, RH, air velocity, dust loading, culturable fungi/bacteria loading, were conducted in the following: return air, fresh air, mixture air, cooling, and supply air segments. On basis of these measuring results, the culturable fungi/bacteria number per gram dust was calculated and the predominant culturable fungi/ bacteria species was identified. Statistical analysis showed that dust loading, culturable fungi/bacteria loading, culturable fungi/bacteria number per gram dust were in the range of 3.25~48.25g/cm2, 32~221CFU/cm2, 46~232CFU/cm2, 53815~124807CFU/gdust, and 63395~10383CFU/gdust respectively. The maximum value for dust loading and culturable fungi /bacteria loading appeared in fresh air segments. The predominant culturable fungi were Penicillium, Aspergillus, Cladosporium and Alternaria and the predominant culturable bacteria were Micrococcus, Bacillus,
Page 1 of 29
Staphylococcus, and Pseudomonas. There were significant positive correlations between dust and culturable fungi loading, dust and culturable bacteria loading, culturable fungi number per gram dust and RH, bacteria number per gram dust and temperature (p<0.05). Results of these field measurements indicated that dust
ip t
accumulation and/or high humidity and/or temperature should be properly controlled in HVAC systems to prevent the growth of culturable fungi and bacteria.
cr
Keywords: HVAC system; Dust loading; Culturable fungi; Culturable bacteria; RH; Temperature
us
1. Introduction
The interaction between dust and bio-aerosol retained in HVAC and indoor air quality has gained more
an
attention due to their possible relationship to irritation, health outcomes, and odors [1], [2] and [3]. The HVAC
M
systems was usually used to provide fresh and to clean air and remove the pollutants caused by the activity of occupants, the equipment of furniture and the materials of decoration[4]. Through normal indoor office activity
d
and occupation, a great deal of contaminants and air pollutants, such as dander, dust, and chemicals could be
Ac ce pt e
generated [5]. These contaminants could pull into the HVAC system and re-circulated 5 to 7 times per day, on average [6]. This re-circulation could cause a build-up of contaminants in HVAC system. When the fresh air could not be treated and purified appropriately, the HVAC system could become more seriously polluted and act as the source of indoor air pollution, which could further pose a severe threat to the office worker’s health [7] As deposition dust and culturable microorganism are suggested as the important pollution parameters in HVAC system, which might be closely associated to officers’ SBS and BRI [8] and [9], it is therefore very significant to investigate the characterization of dust loading and culturable microorganism in HVAC systems of office buildings. For culturable microorganism and dust in the HVAC systems, there has been a lot of research performed in several aspects, such as, the aerosolization of the fungi [10] and [11], the movement and dynamics
Page 2 of 29
characteristic of the dust in ducts work [12], the microorganism growth on the surface of filter [13]. Gao et al. compared the modeling deposition of dust in vertical square HVAC duct flows and concluded that the gravity of particles does not directly change the dimensionless deposition velocity in vertical flows [14]. Wu et al.
ip t
discussed the effect of ventilation duct as a particle filter and concluded that the filtration of the duct might not be ignored as the obvious loss by deposition when particle diameter is larger than 5μm [15]. Chang et.al
cr
evaluated the susceptibility of three types of ventilation duct materials (fibrous glass ductboard, galvanized
us
steel, and insulated flexible duct) to fungal growth [16]. Noris et.al discussed microbial contaminants and metals captured on HVAC filters in one commercial building and eight residential [17]. Forthomme et al.
an
examined the conditions (such as air relative, the airflow after microbial contamination of filter and the nature
M
of fibrous) leading to microbial development on to fibrous filters and to microbial release downstream of filters[18].Stephen noted that Cladosporium sp. fungi was commonly recovered in terms of growth sites and
d
deposited spores, and that were found mainly in the blower wheel fan blades, the ductwork, and the cooling coil
Ac ce pt e
fin in air handling units, and concluded that there was no relationship between mold growth and operating condition[19]. Pasanen investigated the accumulated dust on the air duct of 24 houses which had mechanical ventilation and found that Cladosporium, Penicillium, Aspergillus, and Yeasts made up more than 90% of fungi[20]; Li et.al investigated the effect of air-conditioning parameters (including temperature, relative humidity, and air velocity) and the deposition dust on microbial growth in supply air ducts[21]. Chen et al. from Tsinghua University indicated that 82% HVAC systems were moderately and seriously polluted [22]. Lu et.al investigate microorganisms and particles in AHU systems, and the results found that the microbial growth on surfaces of duct and equipment was noticed , and that it may be transferred into indoor air, which could lead to adverse health effect[23]. The previous researchers typically paid close attention to the dust deposition mechanism or microorganism
Page 3 of 29
growth phenomenon in only one segment, such as duct, filter, fan, and coil, to date and to our best knowledge, but nearly no related studies concerning the dust loading and culturable microorganism pollution characteristics in the integrated HVAC systems were conducted. Indeed, almost no comprehensive research has been
ip t
conducted on the dust loading and culturable microorganisms contamination characteristics in the integrated HVAC systems of office buildings in Beijing. Although some international studies have pointed out that the
cr
HVAC systems for indoor environment were just like the lungs for human, which were closely related to
us
occupants’ health [1], [2] and [3], dust and culturable microorganisms pollution problems in HVAC systems in China have not drawn sufficient attention due to the lack of sufficient field measurement data. The aim of this
an
work is to investigate the dust loading and culturable microorganisms (including fungi and bacteria)
M
contamination characteristics of HVAC systems in 24 office buildings, and to provide basic data and reference
2. Method
Ac ce pt e
2.1 Study design
d
for improving healthy office buildings indoor environment.
The twenty-four HVAC systems of office buildings, located in downtown were selected as measurement objectives. These HVAC systems and air filters were cleaned two years before the current testing, and the filters had been in operation without clean for two years. The outdoor air quality, including PM10 and PM2.5 mass concentration, for last two years in Beijing (2013.7-2015.6) , had been shown in the following Fig.1. The general description of sampling sites and HVAC information (including year built, air filter category, area, HVAC locations，supply air volume，percentage of fresh air, and occupants number) were shown in Table.1. All the buildings were relatively close to main streets and located in commercial areas. The centralized air-conditioning systems with primary return air in use in all the office buildings. The sampling sites were determined in return air segment, fresh air segment, mixture air segment, cooling segment
Page 4 of 29
and supply air segment, as shown in Fig.2. Measurement parameters included temperature, RH, air velocity, dust loading, fungi/bacteria loading. Two parallel samplings were performed for each sampling point. All the HVAC systems were kept in the state of normal operation during the sampling process. The field measurement
ip t
actions were conducted in summer (July 6-September 15) when the weather was allowable, without rain and the wind speed was below 8.0 m/s
cr
2.2 Measuring methods
us
A TSI multi-channel portable tester, model 8392A, was used to measure the temperature, RH and air velocity. The dust weight was quantified by an electronic balance, model 9367S. The test range and precision of
an
these four parameters were shown in Table.2.
M
The dust loading was tested by gravimetrical techniques according to the Chinese Hygienic Standard (WS394-2012) [24]. A sterile non-woven fabric was employed to wipe the deposition dust with an area equal to
d
50 cm2. Before the sampling, the sterile no-woven fabric was put into the sealing bags in advance. All sealing
Ac ce pt e
bags with no-woven fabric were together dried for 4 hours in a silica drier, and then the initial weight of them were obtained by the calculation of the average values [24]. After non-woven fabric wiped up the dust in sampling location, it would be the put back to the sealing bag. The final average weight could also be obtained by twice weighting. The weight of dust would be the difference of the initial and the final values. Therefore, the dust loading could be calculated by the following equation (1):
Ldsut =
ΔM S
(1)
where, Ldsut (g/cm2) is the dust loading, ΔM (g) is the difference of initial and final weight and S (cm2) is the wiping area of the inner surface. The fungi and bacteria loading were tested by the wiping method. Firstly, the sterile non-woven fabric was used to wipe an area equal to 50 cm2. After non-woven fabric wiped up the dust in the sampling location, it
Page 5 of 29
would be put back into the sealing bag. It was imperative to get the sealing bags back to the lab within six hours in order to keep the microorganism viability [22]. The dust sample on the non-woven fabric was dissolved in 100mL of 0.01% Tween 80 by hand-shaking, and then 1ml solution was incubated on culture petri dish. The
ip t
analysis for each dilution was performed three times and the average number of colonies formed was recorded. The bacteria petri dishes were incubated at 36℃for 48 hours on Tryptic Soy Agar (TSA), then the colonies
cr
were counted. The fungi petri dishes were incubated at 28℃ for 5 days on the Sabourand’s agar (SDA), and
us
then the number of colonies were enumerated [23] and [25]. Blank samples were required in setting up each sampling process. The test was proved to be effective only when no colony occurred on blank sample. The
N fungi /bacteria ×100
(2)
S
M
L fungi / bacteria =
an
bacteria and fungi loading could be calculated by the following equation (2).
where, L fungi / bacteria (CFU/cm2) is the fungi/bacteria loading, N fungi /bacteria (CFU) is the number of colony forming
Ac ce pt e
d
unit fungi or bacteria, S (cm2) is the wiping area of inner surface. Because the dust sample on the non-woven fabric was dissolved in 100mL of 0.01% Tween 80, only 1ml solution was incubated on a petri dish, therefore the results should be multiplied by 100.The fungi/bacteria number per gram dust could be calculated by the following equation [26].
ϕ fungi / bacteria =
L fungi / bacteria Ldust
(3)
where, ϕ fungi / bacteria (CFU/g) is the fungi/bacteria loading, and L fungi / bacteria and Ldust had been defined in the previous chapter. 2.3 Microorganism Identification Method The bacteria and fungi from the experimental incubation process were identified by the morphology method, which was employed in previous studies [18] and [23]. The petri dishes were examined by optical
Page 6 of 29
methods by an experienced analyst to identify and quantify the fungi growth on the petri dishes. Identification was conducted according to morphological features of the fungi and bacteria. An optical microscope (XSP-63B) and a fluorescent optical microscope (PF6200) were employed in this process. The fungi spores were observed
ip t
by electron microscope in case it was needed. 2.4 Statistical Analysis Methods
cr
The mean, median, 25th percentile，75th percentile, and geometric standard deviation (GSD) were used to
us
describe the data’s characteristics. The Pearson product-moment correlation coefficient was used to explore the strength of the linear relationship between each two sets of measuring variables in our study.
an
3. Results and Discussion
M
The field measurement results in the 24 HVAC systems of office buildings are summarized in Table.3. All the parameters including temperature, RH, air velocity, dust loading, fungi loading, bacteria loading, fungi
d
number per gram dust, and bacteria number per gram dust in different sites ,such as return air segment, fresh air
Ac ce pt e
segment, mixture air segment, cooling segment and supply air segment were described in terms of both mean and geometric standard deviation. The temperature in these five sampling sites (R, F, M, C and S) was in range of 21.4-28.5, 30.2-35.6, 26.1-30.2, 16.2-18.9, and 18.1-21.8℃, respectively. The RH in these five segments was in the range of 52-68%, 71-82%, 59-73%, 89-94%, and 83-89%, respectively. The velocity in these five segments was in the range of 5.8～8.1, 2.2～4.5 2.8～3.8, 1.4～2.9, and 5.1～7.8 m/s, respectively. 3.1. Dust loading
The mean values and geometric average deviation (GSD) of the dust loading in the five different segments are shown in Table.3.The maximum dust loading occurred in fresh air segments, being followed, sequentially, by mixture air segments, return air segments, cooling segments, and supply air segments. The mean value of dust loading was 33.37g/m2 in the fresh air segments and was about six times higher than that of the supply air
Page 7 of 29
segments. The average dust loading in these five segments was in the range of 12.45-22.43, 19.34-48.25, 14.89-32.14, 4.89-12.98, and 3.25-8.89 g/m2, respectively. The data distribution of dust loading (including min, max, medium, 25th percent, and 75th percent) in these five segments is shown in Fig.3.The threshold value of
ip t
dust loading in the HVAC system is less than 20 g/m2 according to the Chinese standard[24]. The exceeding percentages in the return air segments, fresh air segments, and mixture air segment were 43.75%, 93.75% and
cr
87.5% respectively. All samplings for the dust loading in cooling and supply air segments were less than the
us
threshold value. According to the HVAC’s system recorded operating time and to the dust loading, the dust loading rates could be calculated. The dust loading rates in the five sampling sites (R, F, M, C, and S) was in
an
the range of 0.519-0.935, 0.806-2.011, 0.621-1.339, 0.204-0.541, and 0.135-0.371 g/m2·month. As reported by
M
Waring et al., the median predicted residential and commercial loading rates were 0.0051 and 1.00 g/m2·month for the supply duct loading rates, and 0.262 g/m2·month for the commercial return ducts[27]. However, the
d
current result is distinctly different from the predicted results, in terms of loading rates for the return segments.
Ac ce pt e
The discrepancies of the specific value would be mainly likely to be due to the fact that the difference for the indoor sources exist, depending on outdoor particle distributions, indoor sources, HVAC operation strategy, filtration, and so on [28] and [29].The dust loading in the return air segments was significantly higher than that of the supply air segments, and this indicates that there was some indoor dust pollution sources, as identified in the many other many studies[3], [6], [28] [29] and [30]. More attention should be paid on the return air segments and some purified equipments should be installed, if necessary. The dust loading in the fresh air segments was the highest, and implication would be that the fresh air might be the important dust pollution source and a higher efficiency purifier should be employed at the fresh air inlet. 3.2. Culturable fungi/bacteria loading The culturable fungi/bacteria loadings, in terms of box-plot, in different segments are shown in Fig.4 and 5.
Page 8 of 29
The threshold value of culturable fungi/bacteria loading in the HVAC system is less than 100 CFU/cm2, according to Chinese standard [24].The maximum value of fungi/bacteria loading also occurred in fresh air segments, being followed, sequentially, by mixture air segments, return air segments, cooling segments, and
ip t
supply air segments. This was similar to the distribution of the dust loading. The culturable fungi loading in these five segments were in the range of 67-138, 112-221, 89-191, 46-162, and 32-148 CFU/cm2, and the
cr
corresponding exceeding percentages was 43.75%, 100%, 87.5%, 50%, and 25%, respectively. The mean value
us
of culturable fungi loading was 170 CFU/cm2 in the fresh air segments and was about 2.11 times higher than that of supply air segment. The culturable bacteria loading in these five segments were in the range of 67-138,
an
112-232, 89-192, 46-162, and 31-148 CFU/cm2, and the corresponding exceeding percentages were 68.75%,
M
100%, 100%, 6.25%, and 0%, respectively. The mean maximum of culturable fungi loading was 171 CFU/cm2 in fresh air segment and was about 2.67 times higher than that of the supply air segments. Meng et al.’s results
d
indicated that the exceeding percentages of fungi/ bacteria loading in the eight points on the HVAC duct
Ac ce pt e
surface before cleaning was 100%, and that values seemed much higher than our study [31].The culturable fungi loadings in the HVAC system was from 31-192 CFU/cm2 in this study, which was in the range of 1-261 CFU/cm2 reported by Lappalainen et al.[32]. The culturable bacteria loading in this study, ranging from 32-221 CFU/cm2, was similar to the Li et al.’s findings, and that was in the range of 24 to 204 CFU/cm2 [33]. All the values of dust loading were below the threshold value in cooling segments and supply air segments, while a few values of culturable fungi loading exceeded the threshold, this could be shown by comparing Fig.3 and 4.This indicated that the culturable fungi loading should draw more attention even though the dust loading met requirements. 3.3. Culturable fungi/bacteria number per gram dust The culturable fungi/bacteria number per gram dust in different segments is shown in Fig.6 and 7. The
Page 9 of 29
largest culturable fungi number per gram dust occurred in the cooling segments, being followed, sequentially, by the supply air segments, fresh air segments, mixture air segments and return air segments, which was different from the sequence of dust loading and culturable fungi/bacteria loading. The culturable fungi loading
ip t
in these five segments was in the range of 53815-61525, 80144-88082, 59771-76540, 94070-124807, and 98462-114735 CFU/gdust, respectively. The mean value of the culturable fungi loading was 109525 CFU/gdust in
cr
the cooling segments and was about 1.88 times higher than that of the return air segments. The largest
us
culturable bacteria number per gram dust also appeared in fresh air segment, followed, sequentially, by the mixture air segments, supply air segments, return air segments, and cooling air segments. The culturable
an
bacteria number per gram dust in these five segments was in the range of 81141-89960, 92037-103834,
M
90665-99876, 63395-79353, and 70769-93793 CFU/gdust. The mean value of culturable bacteria loading was 97302 CFU/gdust in th fresh air segments and was about 1.37 times higher than that of the cooling segments. The
d
culturable fungi/bacteria number per gram dust was from 105-106, lying in the range of 104-106 fungi number
Ac ce pt e
and 105-107 bacteria number per gram dust reported by Noris et al.[17]. Pasanen et al. investigated the fungal spore content of dust accumulated in the air ducts of 24 mechanically ventilated single-family homes, and the results showed that culturable spore concentrations in the dust varied from 104 -107 CFU/gdust [34], which was much broader than our results. Another experiment investigation conducted by Li et al. assumed that the culturable fungi/bacteria number per gram dust was in the range of 25000-60000CFU/gdust, which was slightly lower than our results [21]. Although the values of culturable fungi/bacteria number per gram dust in our study were slightly different from the related above references, it greatly exceeded the threshold value (50000 CFU/g for culturable bacteria and culturable fungi), as regulated in Chinese standard [24]. Some specific measurements, such as disinfecting and cleaning, should be taken to reduce the contamination of the HVAC systems in office buildings in subsequent management and operation.
Page 10 of 29
3.4. Culturable fungi/bacteria species identification and analysis The predominant culturable fungi/bacteria species from the field measurements were identified by the morphology method in the lab. The morphological characteristics for six species of fungi are shown in Fig.8
ip t
and the morphological characteristics for four species of bacteria are shown in Fig.9. The statistics of the percentages and occurrences for the different culturable fungi and bacteria in these five segments (R, F, M, C
cr
and S) are listed in Table.4 and Table.5. The predominant culturable fungi were Penicillium, Aspergillus,
us
Cladosporium, and Alternaria due to the higher occurrences. Similar findings have been noted by Li et al.[35].There was a little discrepancy when comparing to Lu et al.’s results, which was that the predominant
an
species were Cladosporium and Penicillium in AHU systems[23]. The predominant culturable bacteria were
Pastuszka et al.’s observation [36].
d
3.5. Correlation analysis
M
Micrococcus, Bacillus, Staphylococcus, and Pseudomonas. Similar interactions have been described in the
Ac ce pt e
The correlation between each two measuring parameters has been explored in this section, as shown in Table.6. There was a significant positive correlation between dust loading and culturable fungi/bacteria loading (p<0.05), which indicated that the culturable microorganism in the HVAC system is partially determined by the accumulated dust. Similar findings, which discussed the correlation between fungi and PM2.5 in the atmosphere environment has been noted by Degobbi et al.[37]. In light of the results by Li et al., the deposition dust in the supply air ducts presented a positive correlation between the microorganism loadings, which was consistent with our findings [21]. There was a significant positive correlation between culturable fungi number per gram dust and RH (p<0.05), however, not a significant correlation between fungi loading and RH. There was a significant positive correlation between culturable bacteria number per gram dust and temperature (p<0.05), but not a significant correlation between bacteria loading and temperature. These results could be explained as
Page 11 of 29
follows: the effect of temperature and RH on microorganism growth could be revealed only after the factor of dust loading was excluded. RH and temperature contributed to the microorganism growth has been stated by previous studies [35], [36], [38], [39] and [40]. However, few conflicting studies have presented that there are
ip t
no significant correlations between RH, temperature and microorganism concentration [19]. There is no significant correlation between air velocity and culturable microorganism, and these similar results have been
cr
addressed by Forthomme et al. [13]. The results of this investigation suggested that culturable fungi were more
us
sensitive to RH and the culturable bacteria were more sensitive to temperature in the HVAC’s operating environment. Therefore, dust accumulation, high humidity, and/or temperature should be properly controlled in
an
the HVAC systems to prevent the growth of microorganism on the premise of meeting the requirements of
3.6. Limitation
M
operating.
d
So far, this study is still preliminary and needs to be confirmed by future work. There were two limitations
Ac ce pt e
listed in the following. Firstly, every species of fungi and bacteria might grow/die in any given environmental and nutrient conditions. Some species will dominate the microbial community after a very short period of time. Spores may be unable to grow in a culture because they are nonviable, or they are viable, but they are injured, or because the nutrients or temperature conditions are not appropriate [41]. These phenomena can be influenced by sample storage, transport, and age factors. Secondly, only the culturable fractions of the microbial population were measured in our study. According to the findings by Tovila et al., only a small part of microbial population indoors were culturable and a much greater fraction of the microbial community could be detected by molecular biological techniques [42] and [43]. 4. Conclusions The dust loading and culturable microorganism distribution characteristics in the HVAC systems of office
Page 12 of 29
building were investigated and the following conclusions could be drawn from this study: (1) The distribution characteristics of dust and culturable fungi/bacteria loading was similar; namely, all the maximum values occurred in the fresh air segments, being followed, sequentially, by mixture air segments,
ip t
return air segments, cooling segments and supply air segments. Such a trend indicates that the fresh air might be source of dust pollution and higher efficiency purifier should be employed at the inlet of fresh air.
cr
(2) The fungi/bacteria loading might exceed the threshold when the dust loading was below the threshold in
attention even though the dust loading met requirements.
us
the cooling and supply air segments, which indicated that the culturable fungi loading should draw more
an
(3) The predominant culturable fungi was Penicillium, Aspergillus, Cladosporium, and Alternaria and the
M
predominant culturable bacteria was Micrococcus, Bacillus, Staphylococcus, and Pseudomonas, respectively, in the HVAC systems of the office buildings in Beijing.
d
(4) There was a significant positive correlations between dust and fungi loading, dust and bacteria loading,
Ac ce pt e
fungi number per gram dust and RH, bacteria number per gram dust and the temperature (p<0.05), but no significant correlation between air velocity and culturable microorganism. The culturable fungi was more sensitive to RH, and the culturable bacteria was more sensitive to the temperature in the HVAC’s operating environment, after excluding the influence of dust loading. On basis of these research results, the potential specific solutions could be obtained as follows. (1)
The dust accumulation and/or high humidity and/or temperature should be properly controlled in
HVAC systems to prevent the growth of culturable fungi and bacteria. (2) Regular cleaning and disinfection work for the HVAC systems should be conducted in avoid of air contamination occurrence. (3) Appropriate air filter category selection in office building should take the outdoor air quality into
Page 13 of 29
consideration, which could be the most important step to achieve good IAQ in HVAC design. Nevertheless, these conclusions were obtained on the basis of a relative size sample of 24 HVAC systems. This investigation was just aimed at the office buildings in Beijing. More detailed and in depth studies are
ip t
required in future works to verify these results with a larger sample population. Future studies are also needed to address the tests of airborne fungi/bacteria spores and develop of cleaning and disinfection strategies in the
cr
HVAC systems in China.
us
Acknowledgement
This study is supported by the Fundamental Research Funds for the Central Universities, No. 2015MS108.
an
References
M
[1] B.F. Yu, Z.B. Hu, M. Liu, H.L. Yang, Q.X. Kong, Y.H. Liu, Review of research on air-conditioning systems and indoor air quality control for human health, International journal of refrigeration 32 (1) (2009)
d
3-20.
Ac ce pt e
[2] P.M. Bluyssen, C. Cox, O. Seppänen, E.D. Oliveira Fernandes, Why, when and how do HVAC-systems pollute the indoor environment and what to do about it? The European AIRLESS project, Building and Environment 38 (2) (2003) 209-225.
[3] F. Bitter, K. Fitzner, Odour emissions from an HVAC-system, Energy and buildings 34 (8) (2002) 809-816.
[4] O. Seppänen, Ventilation strategies for good indoor air quality and energy efficiency, International Journal of Ventilation 6 (4) (2008) 297-306. [5] M. Ongwandee, R. Moonrinta, S. Panyametheekul, C. Tangbanluekal, G. Morrison, Investigation of volatile organic compounds in office buildings in Bangkok, Thailand: Concentrations, sources, and occupant symptoms, Building and Environment 46(7) (2011) 1512-1522.
Page 14 of 29
[6] Z. Bakó‐Biró, P. Wargocki, C.J. Weschler, P.O.Fanger, Effects of pollution from personal computers on perceived air quality, SBS symptoms and productivity in offices, Indoor air 14 (3) (2004) 178-187. [7] J.H. Choi, V. Loftness, A. Aziz, Post-occupancy evaluation of 20 office buildings as basis for future IEQ
ip t
standards and guidelines, Energy and buildings 46 (2012) 167-175. [8] S. Gupta, M. Khare, R. Goyal, Sick building syndrome—A case study in a multistory centrally
cr
air-conditioned building in the Delhi City, Building and Environment 42 (8) (2007) 2797-2809.
us
[9] J. Sundell, T. Lindvall, B. Stenberg, Associations between type of ventilation and air flow rates in office buildings and the risk of SBS-symptoms among occupants, Environment International 20 (2) (1994) 239-251.
an
[10] R.L. Gorny, T. Reponen, S.A. Grinshpun, K. Willeke, Source strength of fungal spore aerosolization from
M
moldy building material, Atmospheric Environment 35 (2001) 4853–4862. [11] S.K. Sivasubramani, R.T. Niemeier, T. Reponen, S.A. Grinshpun, Assessment of the aerosolization
d
potential for fungal spores in moldy homes, Indoor Air 14 (2004) 405–412.
Ac ce pt e
[12] J.P. Zhang, A.G. Li, D. Li, Modeling deposition of particles in typical horizontal ventilation duct flows, Energy Conversion and Management 49 (12) (2008) 3672-3683. [13] F. Noris, J.A. Siegel, K.A. Kinney, Evaluation of HVAC filters as a sampling mechanism for indoor microbial communities, Atmospheric environment 45 (2) (2011) 338-346. [14] R. Gao, A.G. L, Modeling deposition of particles in vertical square ventilation duct flows, Building and Environment 46 (1) (2011) 245-252.
[15] J. Wu, B. Zhao, Effect of ventilation duct as a particle filter, Building and environment 42 (7) (2007) 2523-2529. [16] J. Chang, K.K. Foarde, D.W. VanOsdell, Assessment of fungal (Penicillium chrysogenum) growth on three HVAC duct materials, Environment international 22 (4) (1996) 425-431.
Page 15 of 29
[17] F. Noris, J.A. Siegel, K.A. Kinney, Biological and Metal Contaminants in HVAC Filter Dust, ASHRAE Transactions 115 (2) (2009) 484-491. [18] A. Forthomme, A. Joubert, Y. Andrès, X. Simon, P. Duquenne, Microbial aerosol filtration: Growth and
ip t
release of a bacteria–fungi consortium collected by fibrous filters in different operating conditions, Journal of Aerosol Science 72 (2014) 32-46.
cr
[19] S.C. Wilson, R.N. Palmatier, L.A. Andriychuk, J.M. Martin, C.A. Jumper , H.W. Holder, D.C. Straus,
us
Mold contamination and air handling units, Journal of occupational and environmental hygiene 4 (7) (2007) 483-491.
an
[20] P. Pasanen, A. Nevalainen, J. Ruuskanen, P. Kalliokoski, The composition and location of dust settled in
M
supply air ducts, In 13th AIVC Conference.
[21] A.G. Li, Z.J. Liu, X.B. X.B. Zhu, Y. Liu, Q.Q. Wang, The effect of air-conditioning parameters and
d
deposition dust on microbial growth in supply air ducts, Energy and Buildings 42 (4) (2010) 449-454.
Ac ce pt e
[22] C.F. Chen, B. Zhao, X.D. Yang, Investigation and review of microbial pollution in air conditioning systems of public buildings, HV &AC 39 (2) (2009) 50–55. [23] Z. Lu, W.Z. Lu, J.L. Zhang, D.X Sun, Microorganisms and particles in AHU systems: Measurement and analysis, Building and Environment 44(4) 2009 694-698. [24] Ministry of Health of the People's Republic of China, WS 394-2012, Hygienic specification of central air conditioning ventilation system in public buildings, Beijing, 2012. [25] H.J. Oh, I.S. Nam, H. Yun, J. Kim, J. Yang, J.R. Sohn, Characterization of indoor air quality and efficiency of air purifier in childcare centers, Korea, Building and Environment 82 (2014) 203-214. [26] K. Ashley, G.T. Applegate, T.J. Wise, J.E. Fernback, M.J. Goldcamp, Evaluation of a standardized micro-vacuum sampling method for collection of surface dust, Journal of occupational and environmental
Page 16 of 29
hygiene 4(3) (2007) 215-223. [27] M.S. Waring, J.A. Siegel, Particle loading rates for HVAC filters, heat exchangers, and ducts, Indoor Air 18 (2008) 209-224.
ip t
[28] P. Wargocki, Z. Bako-Biro, G. Clausen, P.O. Fanger, Air quality in a simulated office environment as a result of reducing pollution sources and increasing ventilation, Energy and buildings 34(8) (2002) 775-783.
cr
[29] J.F. Montgomery, S.I. Green, S.N. Rogak, K. Bartlett, Predicting the energy use and operation cost of
us
HVAC air filters, Energy and Buildings 47 (2012) 643-650.
[30] I.Sarbu, C. Sebarchievici, Aspects of indoor environmental quality assessment in buildings, Energy and
an
Buildings 60 (2013) 410-419.
M
[31] C. Meng, Q.Wang, Y Song, Y. Cao, N. Zhao, Y. Shi, Experimental study on both cleaning effect and motion performance of the duct-cleaning robot, Sustainable Cities and Society 14 (2015) 64-69.
d
[32] S. Lappalainen, E. Kähkönen, P. Loikkanen E. Palomäki, O. Lindroos, K. Reijula, Evaluation of priorities
981-986.
Ac ce pt e
for repairing in moisture-damaged school buildings in Finland, Building and Environment 36 (8) (2001)
[33] A.G. Li, L.Z. Yao, J.J. Hou, Q.Q. Wang, M. Li, Field test and analysis of microbial concentration in central air condition systems, HV &AC 40 (3) (2010) 120–125. [34] A.L. Pasanen, L. Kujanpää, P. Pasanen, P. Kalliokoski, G. Blomquist, Culturable and Total Fungi in Dust Accumulated in Air Ducts in Single‐Family Houses, Indoor Air 7(2) (1997) 121-127. [35] A.G. Li, Z.J. Liu, Y. Liu, X. Xue, Y. Pu, Experimental study on microorganism ecological distribution and contamination mechanism in supply air ducts, Energy and Buildings 47 (2012) 497-505. [36] J.S. Pastuszka, U. Kyaw Tha Paw, D.O. Lis, A. Wlazło, K. Ulfig, Bacterial and fungal aerosol in indoor environment in Upper Silesia, Poland, Atmospheric Environment 34 (22) (2000) 3833-3842.
Page 17 of 29
[37] C. Degobbi, F.D. Lopes, R. Carvalho-Oliveira, J.E. Munoz, P.H. Saldiva, Correlation of fungi and endotoxin with PM2.5 and meteorological parameters in atmosphere of Sao Paulo, Brazil, Atmospheric environment 45(13) (2011) 2277-2283
prevalence child care centers, Atmospheric Environment 43 (15) 2009 2391-2400.
ip t
[38] M.S. Zuraimi, L. Fang, T.K. Tan, F.T. Chew, K.W. Tham, Airborne fungi in low and high allergic
cr
[39] C. He, H. Salonen, X. Ling, L. Crilley, N. Jayasundara, H.C. Cheung , L. Morawska, The impact of flood
us
and post-flood cleaning on airborne microbiological and particle contamination in residential houses, Environment international 69 (2014) 9-17.
an
[40] Z.J. Liu, A.G. Li, Z.P. Hu, H.F. Sun, Study on the potential relationships between indoor culturable fungi,
M
particle load and children respiratory health in Xi’an, China, Building and Environment 80 (2014) 105-114. [41] Z. Fang, Z. Ouyang, H. Zheng, X. Wang, Concentration and size distribution of culturable airborne
Ac ce pt e
325-334.
d
microorganisms in outdoor environments in Beijing, China, Aerosol Science and Technology 42 (5) (2008)
[42] M. Toivola, S. Alm, T. Reponen, S. Kolari, A. Nevalainen, Personal exposures and microenvironmental concentrations of particles and bioaerosols, Journal of Environmental Monitoring 4 (1) (2002) 166-174. [43] Y. Zheng , Y. Wen , A.M. George,
A. Steinbach, E. Freeman, B. Philips, N. Giannoukakis, L.M.
Weiland, E.S. Gawalt, and W.S. Meng. A Peptide-Based Material Platform for Displaying Antibodies to Engage T Cells. Biomaterials 32 (2011) 249-257.
Fig.1 The information of outdoor air quality in Beijing in the period of 2013.7-2015.6 Fig.2 The schematic diagram of sampling location in the HVAC system Fig.3 The dust loading distribution on the inner surface of the different segments
Page 18 of 29
Fig.4 The culturable fungi loading distribution on the inner surface of the different segments Fig.5 The culturable bacteria loading distribution on the inner surface of the different segments Fig.6 The culturable fungi number per gram deposition dust in the different segments Fig.7 The culturable
ip t
bacteria number per gram deposition dust in the different segments Fig.8 The morphological characteristics for 6 species of the culturable fungi identified Fig.9 The morphological
Ac ce pt e
d
M
an
us
cr
characteristics for 4 species of the culturable bacteria identified
Fig.1 The information of outdoor air quality in Beijing in the period of 2013.7-2015.6
Fig.2 The schematic diagram of sampling location in the HVAC system
Page 19 of 29
ip t cr us an
Ac ce pt e
d
M
Fig.3 The dust loading distribution on the inner surface of the different segments
Fig.4 The culturable fungi loading distribution on the inner surface of the different segments
Page 20 of 29
ip t cr us an
Ac ce pt e
d
M
Fig.5 The culturable bacteria loading distribution on the inner surface of the different segments
Fig.6 The culturable fungi number per gram deposition dust in the different segments
Page 21 of 29
ip t cr us an M
Ac ce pt e
d
Fig.7 The culturable bacteria number per gram deposition dust in the different segments
Cladosporium
Penicillium)
Aspergillus
Alternaria
Mucor
Trichoderma
Fig.8 The morphological characteristics for 6 species of the culturable fungi identified
Page 22 of 29
ip t cr us an M d Ac ce pt e
Micrococcus
Bacillus
Staphylococcus
Pseudomonas
Fig.9The morphological characteristics for 4 species of the culturable bacteria identified
Page 23 of 29
Table.1 General Description of the sampling sites and HVAC information
Roof
4
2
Ground
1 2 3 4 9
1 3 2 6 1
Roof Underground Ground Roof Roof
2
Ground
3 3
Underground Roof
1
10
2009
2010
11 12
14 15 16 17 18 19 20 21 22 23 24
2011
Ground
Ac ce pt e
13
2012
6 1 1 7 2 1 6 1 2 7 1
Underground Underground Roof Underground Roof Ground Underground Underground Roof Underground Underground
Percentage of Fresh air 40.01% 46.18%
24588
30.43%
404
35.58%
625
18858 22018 22267 12089 18848
48.97% 43.73% 41.19% 33.75% 19.89%
308 321 462 204 122
32069
30.01%
321
22458 13587
24.85% 34.67%
186 157
35505
28.47%
337
27522 10852 11205 8568 33255 24217 28199 12073 37545 26153 28949
32.92% 23.19% 21.15% 25.56% 33.67% 26.51% 30.32% 44.23% 41.15% 37.28% 34.63%
302 85 79 186 425 321 285 178 515 352 401
35127
M
3
Supply air volume(m3/h) 22251 23428
ip t
Roof Roof
Area (m2) 4515 8148 1108 8 1555 0 5852 9630 9702 4488 5928 1614 6 6292 3933 1542 5 8442 2014 2291 4650 9350 5778 6555 4628 8241 6336 7218
cr
2008
HVAC Location
us
3
MERV Values 1 2
an
Year Built
d
Buildings No. 1 2
Occupants 305 382
Table .2 The test range and precision for these four parameters.
Page 24 of 29
Parameters
Manufacturer
Range
Precision
Temperature
TSI
RH
TSI
0~95%
±1%
Air velocity
TSI
0~50m/s
±0.1m/s
Dust weight
UPGRN
0-3000g
±0.01
±0.1
us
cr
ip t
-17.8~93.3
R
F
M
C
S
M
Mean±GSD.
an
Table.3 The statistics of the environment parameters, dust loading, culturable fungi and bacteria in the 24 HAVC systems
Ac ce pt e
d
Temperature( ) 25.6±2.2 33.1±1.5 28.8±1.5 17.8±0.8 19.7±1.0 RH（%） 59±5 77±3 66±6 91±2 86±2 Air velocity（m/s） 7.0±0.7 3.5±0.7 3.4±0.3 2.0±0.4 6.7±0.7 2 Dust loading（g/m ） 16.79±2.59 30.79±7.69 27.47±7.14 9.35±2.83 6.28±1.98 2 Fungi loading(CFU/cm ) 91±22 170±33 141±34 102±31 85±34 110±15 170±32 144±26 78±12 64±13 Bacteria loading(CFU/cm2) Fungi number per gram dust (CFU/g) 58208±3159 84736±2483 68036±5612 109525±10321 103762±6427 Bacteria number per gram dust(CFU/g) 85280±2522 97302±3375 94185±2722 71173±4515 82819±7433 R-Return air segment, F-Fresh air segment, M-Mixture air segment, C-Cooling segment, S-Supply air segment.
Page 25 of 29
cr
ip t
Table.4 The statistics of the percentages and occurrences for the different species of culturable fungi Penicillium Aspergillus Cladosporium Alternaria Mucor Trichoderma Others Location Per Occ Per Occ Per Occ Per Occ Per Occ Per Occ Per Oc R 30.1 100 11.9 85 37.2 100 5.6 88 3.5 58 1.1 53 10.6 96 F 26.4 100 7.6 86 41.2 100 12.8 92 5.1 65 3.7 58 3.2 98 M 28.1 98 12.4 84 39.6 100 10.9 89 4.8 71 2.9 48 1.3 92 C 39.1 100 24.6 95 22.6 92 5.2 75 2.1 42 1.9 39 4.5 89 S 39.8 100 25.3 98 23.5 94 4.9 82 2.8 32 2.3 42 1.4 99
Ac ce pt e
d
M
an
us
Annotation: Per-Percentage, Occ-Occurrence, R-Return air segment, F-Fresh air segment, M-Mixture air segment, C-Cooling segment, S-Supply air segment
26
Page 26 of 29
R
Micrococcus Per Occ 28.9 95
Bacillus Per Occ 29.6 100
Staphylococcus Per Occ 17.6 92
Pseudomonas Per Occ 11.6 85
Others Per Occ 12.3 88
F M C
21.2 24.8 19.8
26.3 27.8 19.9
23.2 18.6 24.1
16.4 15.5 19.8
12.9 13.3 16.4
95 98 94
18.2 90 17.2 88 25.5 70 17.8 75 21.3 Table.5 The statistics of the percentages and occurrences for the different species of culturable bacteria
100
78 86 74
80 76 72
ip t
92 74 93
us
d
M
an
Annotation: Per-Percentage, Occ-Occurrence, R-Return air segment, F-Fresh air segment, M-Mixture air segment, C-Cooling segment, S-Supply air segment
Ac ce pt e
S
100 90 92
cr
Location
27
Page 27 of 29
Table.6 The pearson correlation among measuring parameters in the different segments of 24 HVAC systems
0.0568 0.2761 0.2125 0.2754 0.2179
0.3658*
1 -0.12 58 0.08 91 0.10 23 0.53 58* 0.10 83 0.23 12
1 -0.18 97 0.082 9 -0.06 58 0.125 8 -0.11 24
1
0.5478 6*
1
0.1253
0.2581
ip t
1
0.4896 *
0.2858
-0.127 8
1
0.1256
0.1291
0.1891
0.1025
1
Ac ce pt e
Air velocity Dust loading Fungi loading Fungi number Bacteria loading Bacteria number
Bacteri a numbe r per
cr
-0.1654
Bacteri a loading
us
RH
Fungi numbe r per gram
an
1
Dust Fungi loading loading
M
Temperat ure
RH
Air veloc ity
d
Var
Temper ature
Annotation: Italicised values indicate a significant (p < 0.05, 2-tailed) association between the paired variables
Contamination distribution characteristics of dust loading were investigated. Distribution characteristics of culturable fungi and bacteria were investigated. The predominant fungi was Penicillium, Aspergillus, Cladosporium and Alternaria. The predominant bacteria was Micrococcus, Bacillus, Staphylococcus and Pseudomonas. There was a significant positive correlation between dust and fungi loading.
28
Page 28 of 29
29
Page 29 of 29
d
Ac ce pt e us
an
M
cr
ip t