Access the full text.
Sign up today, get DeepDyve free for 14 days.
Loading next page...
/lp/ou_press/indoor-air-quality-investigation-of-a-university-library-based-on-GF0rRZQqRd
- Publisher
- Oxford University Press
- Copyright
- © The Author(s) 2018. Published by Oxford University Press.
- ISSN
- 1748-1317
- eISSN
- 1748-1325
- DOI
- 10.1093/ijlct/cty007
- Publisher site
- See Article on Publisher Site
Abstract
This article presents field studies and questionnaire survey on the indoor air quality (IAQ) in library rooms in University of Science and Technology Beijing in April 2016, with no heating, to find out actual situation of IAQ in university library. Nine rooms equipped without centralized air-conditioning system were care- fully selected for the test. Results showed that each room had diverse indoor thermal environment and con- centrations of CO , PM2.5, formaldehyde and TVOC. The concentration of CO ranged from 575 to 2 2 2400 ppm, PM2.5 concentration was 40–70% of the outdoor, the highest concentration of formaldehyde −3 andTVOCwas ~0.042 and0.285mgm , respectively with half of the upper limit. The concentrations of CO and PM2.5 in the holding-reading rooms were higher than that in the rooms only holding books. The situation of formaldehyde and TVOC concentrations was just opposite. The correlations between satisfac- tion of IAQ and the concentrations of indoor air pollutants were not as a constant, but changing with differ- ent rooms. Based on gray system theory, satisfactions were affected seriously by CO and PM2.5. One improving plan with the theme of ‘ventilating rapidly along with purifying’ was determined based on a gen- eral mass balance equation. In theory, IAQ can be improved <5 min for the most unfavorable condition. Through analysis, requirements of fresh air and functionality of the space usages should be considered in the ventilation or central air-conditioning system design. Keywords: indoor air quality; university library; pollution evaluation; rapid ventilation; air purification Received 23 October 2017; revised 2 January 2018; editorial decision 10 January 2018; accepted 16 *Corresponding author: 18811345204@163.com February 2018 ......... ................. ................ ................. ................. ................ ................. ................. . ............... ................. ................. Typical indoor-generated pollutants, also found in libraries, are 1 INTRODUCTION volatile organic compounds (VOCs) such as organic acids includ- University students spend a lot of time in library. Due to the con- ing acetic and formic acid and aldehydes, including acetaldehyde struction history gets several dozen years even much more, many and formaldehyde [5]. What’s more, high levels of aldehyde in university libraries are usually equipped without central air- libraries were also reported, when aldehyde were measured in conditioning systems for heating, cooling and ventilation, there- various public places (railway station, airport, shopping center, fore, it might be much more difficult to provide students with libraries, underground partaking garage, etc.) in Srtasboung, comfort indoor thermal environment and indoor air quality [1]. France, the highest indoor level of formaldehyde were observed in Especially in transition season with windows and doors closed, public libraries [6]. The indoor formaldehyde levels in library 1 −3 IAQ will be deteriorated further because the fresh air only relies and 2 were 55.9 ± 4.9 and 33.7 ± 2.2 μmm , respectively. on air infiltration [2–4]. Therefore, the issue of indoor air pollu- Library 1 had many lines tracks filled with old books and journals tion in university library has drawn more and more concerns. [7]. The reason might be that a variety of VOCs are known to be International Journal of Low-Carbon Technologies 2018, 13, 148–160 © The Author(s) 2018. Published by Oxford University Press. This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/ by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact journals.permissions@oup.com doi:10.1093/ijlct/cty007 Advance Access Publication 24 February 2018 148 Downloaded from https://academic.oup.com/ijlct/article-abstract/13/2/148/4907958 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Indoor air quality investigation of a university library emitted from paper and other cellulose-based materials during and Technology Beijing (USTB) were selected which were donated degradation [8, 9]. as 1EE, 1EW, 1CL, 1CT, 1WS, 1WN, 3CT, 3WN, 3WS, respect- Formaldehyde and VOCs are regarded as highly toxic and car- ively. The selection of nine rooms followed the way of controlling cinogens sources that can cause respiratory illness [10, 11]. Most variables and considered factors of similar building scale, floor lay- VOCs are widely used in construction, furniture, textiles, carpentry out, HVAC systems, human behavior or function of rooms, the and chemical industry [12, 13]. High levels of formaldehyde and age of books and so on. For example, the influence of books at TVOC are risk factors to asthma and rhinitis, and may even lead various times on IAQ can be acquired by dividing the rooms into to skin, melanoma, lung and endocrine-related cancers [14–21]. two groups, the one is only to hold books (1EE, 1CL, 1CT, 3CT), Moreover, indoor CO ,formaldehyde and totalVOC (TVOC) are the other one is holding-reading integration which provides a per- risk factors of sick building syndrome (SBS) which have been regu- manent place for reading or studying (1EW, 1WS, 1WN, 3WS, larly reported worldwide [22–25]. 3WN). This four-storey library building is constructed in brick- What’s more, emissions of formaldehyde and TVOC are influ- concrete. The layout and total area of each floor are similar. enced by many environment factors [26–31], such as temperature, After several renovations, the library that built in 1952 covers humidity and air velocity. Numerous studies have reported that an area of seven acres and a half, construction area of 20 000 m , there are seasonal discrepancies of the IAQ. VOCs with high boil- includes more than 2000 seats for reading and learning. All the ing point will only be detected under high indoor temperature, for rooms are designed with double plastic steel windows. All of the example, more alkanes were detected in summer than in winter rooms were decorated at least 5 years ago. As for ventilation, no [32]. Formaldehyde concentrations in large departments reached mechanical ventilation system was installed in the library. The peak in summer for almost the whole year [33]. Indoor concentra- occupants relied on air infiltration in winter and transition sea- tion of particulate matter with aerodynamic diameter <2.5 μm son and natural ventilation by opening windows in summer. (PM2.5, go directly to the alveoli of the lungs) changed with out- When the exterior windows are closed, ventilation is purely by door concentration, indoor PM2.5 concentration was much higher air infiltration driven by indoor–outdoor temperature differ- in winter than it in summer [34]. ences and wind pressure. Beijing, the capital city of China located in the North China Beijing has a climate of cold winter and hot summer. The aver- Plain (NCP), the region with the heaviest air pollution [35, 36], age outdoor temperatures in December, January and February are attracts the most attention of the government and the public all below 0°C. There are centralized district heating systems in since it has a population of 20.7 million in 2012 and 4.5 million Beijing and they are generally operated from November 15 to foreign tourists. The annual average concentrations of PM2.5 March 15. In winter, occupants seldom open windows so as to −3 were observed ~87 μgm in 2013 [37], greatly exceeding the assure thermal comfort and save energy. Consequently, the indoor World Health Organization (WHO) guideline value of 10 μg temperature is almost constant during the entire heating season. −3 m [38]. Annual average concentrations of air pollutants in For the library rooms, the indoor temperature varies within a −3 Beijing were given by Guo et al.[39], the data were 86 μgm small range; in spring and autumn, no heating or cooling is used for PM2.5, 62 ppb for O , 35 ppb for NO , 9.5 ppb for SO and and the room temperature varies but in general within the comfort 3 2 2 1.1 ppm for CO, respectively. range; in summer, the outdoor temperature can reach 35°C during Although IAQ and the emission sources of air pollutants were daytime, so split-unit air conditioners will be run in the rooms. investigated many office buildings, residential buildings and public These temperature differences were ~20°Cinwinterand 10°C buildings, the most of previous studies are conducted only for in transition season. Thus, higher infiltration rate would be expected objective measurement while overlook the individual feelings in in winter. According to the equation regressed by Wallace et al. the environment. The aim of this research is to investigate IAQ in [41], the estimated infiltration rate of the library rooms in winter nine rooms of one university library in Beijing through onsite would be about two to three times that in transition season. The −1 measurements (temperature, relative humility, CO ,PM2.5,for- infiltration rate can hardly be larger than 0.5 h in urban area of maldehyde, TVOC) and questionnaire surveys. According to the northern China where the air tightness is rather good [40]. In finding [40], air infiltration rate in winter will be two to three the measurement conducted by Chen et al. [42] in northern times to it in transition seasons under the condition of doors and China, the average infiltration rate of the buildings built in the −1 windows closed because of the temperature difference between 1990s was 0.24 h . In conclusion, IAQ is much worse in trans- indoor and outdoor. So the research was conducted in April based ition season. This is why we decide to make this study in April, on the most disadvantageous situation. just ~2 weeks after the central heating stopped. We assumed that −1 each room’s natural infiltration change rate was 0.5 h accord- ing to the study [43]. During measurement, the nine rooms of 2 MATERIALS AND METHODS Heating, Ventilation, Windows options were the same, respect- ively ‘no heating’, ‘infiltration’ and ‘closed’. More details of nine 2.1 Site location and description rooms are listed in Table 1. This study was conducted in the city of Beijing(northern China, As shown in Figure 1, there are a lot of mountains in the 39°54′20′, N-116°25′29′E) during the sampling campaign for 5 northwest part of the Beijing so the city will not be influenced weekdays. Nine rooms of the library in the University of Science seriously by Siberian cold current. Thus, for wind condition, International Journal of Low-Carbon Technologies 2018, 13, 148–160 149 Downloaded from https://academic.oup.com/ijlct/article-abstract/13/2/148/4907958 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Y. Wu et al. Table 1. Library room characteristics overview during the measurement. 1EW 1EE 1CT 1CL 1WS 1WN 3CT 3WS 3WN Total area (m ) 160 80 350 350 650 650 350 650 650 Floor covering Marbles Marbles Marbles Marbles Synthetic rubber Synthetic rubber Marbles Synthetic rubber Synthetic rubber Furniture/material Bookshelf, chair, Bookshelf Bookshelf Bookshelf Bookshelf, chair, Bookshelf, chair, Bookshelf Bookshelf, chair, Bookshelf, chair, desk desk desk desk desk Type/function Hold books and Hold Hold Hold Hold books Hold books Hold Hold books Hold books and read books books books and read and read books and read read Age of books (years) <1 <11–60 1–60 1–20 1–20 1–60 1–20 1–20 The Chinese Standard for Technical specifications for moni- toring of IAQ (JGJ.T 167-2004) [44] was used as a reference for the decision of the sampling points. Combined with the using function and structure of rooms, three sampling points are set up in accordance with the diagonal way in each room. Finally, we take an average. The height of each sampling point was con- sistent with the human respiratory zone, i.e. between 1.2 and 1.5 m above the floor level. The measurement points avoided interference from ventilation vents, human activities and inter- ior walls. The distance between the wall and measurement points was >0.5 m. Outdoor measurements points were located ~30 m away from the entrance of the library. 2.3 Questionnaire survey Field measurement is easily to be mastered and has good repeat- ability, but there will be a difficulty for measuring accurately once Figure 1. Location of Beijing and location of USTB. pollutant kinds are pretty or the concentrations of some pollutants are much lower [45]. Besides, the main evaluation of IAQ is peo- ple and there are huge differences in the feeling of people. For this, dominate wind direction is northeast wind of the year. The Prof. Fanger put the concept ‘perceived air quality’ to evaluate windows of selected rooms are open toward west, so the influ- IAQ [46]. There is a leap in cognition that standards issued by ence of wind direction or speed outdoor is ignored. The library American Society of Heating, Refrigerating and Air-conditioning building is far away industry area so the industry pollution Engineers (ASHRAE) involved the principle both subjective and could be disregarded. Outline of the library in USTB is shown objective evaluation [47, 48]. in Figure 2. The questionnaire was adopted from a Sweden study on health-relevant exposure and modified for the actual situ- ation in China [49–51]. For the questionnaire survey, 270 2.2 Sampling and monitoring valid questionnaires in total were collected, and the num- Indoor and outdoor measurements were performed in each ber of questionnaires from nine rooms was almost equal. library room, reading and recording the data from equipment The questionnaire includes ~30 questions on basic infor- every 2 h at a time from 8:00 to 22:00 for 5 weekday. mation, indoor environment evaluation, symptoms and The physical and chemical parameters assessed were tem- perception when leave library rooms, which are summar- perature, relative humidity, CO , PM2.5, HCHO, TVOC. ized in Table 3. Temperature and relative humidity were continuously moni- There mainly focused on the two parts: indoor environment tored with data loggers, CO was monitored everyday with a evaluation and symptoms. The quantification on indoor envir- portable IAQ analyzer (TSI-IAQ-CALC). An optical light scat- onment used a scale model ranging from 1 to 5, representing tering spectrometer (BGPM-02) was used for measuring simul- the salespersons perception levels from ‘very bad’ to ‘very well’; taneously PM2.5 fractions. One portable detector with the while for the SBS perception degree, 1–5 represented ‘never’ to function of automatic calibration based on electrochemical sen- ‘very serious’, details were shown as follows. sor, was used for measuring HCHO and TVOC (LZY-204). All the testing equipment was calibrated as per the manufacturers’ Satisfaction on indoor environment Very bad Bad Neutral Well Very well recommended calibration procedures to ensure accuracy and e.g. Temperature □1 □2 □3 □4 □5 consistence of the measurements. Table 2 lists the equipment If get following symptom? Never → Very serious e.g. Fatigue □1 □2 □3 □4 □5 used in the test and theirs specifications. 150 International Journal of Low-Carbon Technologies 2018, 13, 148–160 Downloaded from https://academic.oup.com/ijlct/article-abstract/13/2/148/4907958 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Indoor air quality investigation of a university library Figure 2. Outline of the library in USTB. (a) Locations of the study rooms (the number represents the layer). (b) Pictures of two different categories of repre- sentative rooms (1WS: holding books and reading; 1CT: only holding books). Table 2. Detailed information of testing equipment. Parameters Instrument model Measuring principle Measuring range Accuracy Distinguishability Temperature WSZY-1 Temperature sensor 40–100°C ±0.5°C 0.1°C Relative humidity WSZY-1 Humidity sensor 0–100% ±3% 0.10% CO TSI-IAQ-CALC NDIR 0–5000 ppm ±3%/± 50 ppm 1 ppm −3 −3 PM2.5 BGPM-02 Optical light scattering 0–999 ug m ±5% 1 ug m −3 −3 Formaldehyde LZY-204 Electrochemical sensor 0~3.000 mg m ±2% 0.001 mg m −3 −3 TVOC LZY-204 Electrochemical sensor 0~9.999 mg m ±2% 0.001 mg m Table 3. Summary of the subjective questionnaire. Questions Basic information Gender, age, daily studying hours, illness history Evaluation of indoor environment Temperature, RH, indoor air quality, ventilation Symptoms Fatigue, dizziness, breathing difficulty, dust feeling, stuffy/pungent odor, efficiency decreasing, irritable feeling Perception when leave library rooms When leaving malls, the symptom change (more severe, stay same, better, almost disappear) The participants sat still indoors for at least 30 min, and except for ‘very bad’ and ‘bad’, others were all satisfied with the completed the questionnaire while seated. The questionnaire factor;in symptom,expectfor ‘never’,othersall have this kind of was compiled by the surveyor independently. In satisfaction, symptom. International Journal of Low-Carbon Technologies 2018, 13, 148–160 151 Downloaded from https://academic.oup.com/ijlct/article-abstract/13/2/148/4907958 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Y. Wu et al. In order to show the statistic more intuitively, the number of Let X ,iN ∈ be two sequences with the same length that is people involved in the questionnaire survey was taken as the defined as the sum of the distances between two consecutive denominator, the number of people satisfied with the factor of time moments, shown as follows: indoor environment or the people with the symptom was taken sX =( ∫ −x(1))dt as numerator. Thereby, the result expressed as a percentage was i ii required in following charts. n ss − = ∫ {[X −x (11 )] − [X −x ( )]}dt 12 11 2 2 Then 2.4 Evaluation of IAQ The evaluation of IAQ in buildings is complex because IAQ 1 +|ss | + | | X = 12 − involves a broad spectrum of substances and agents that vary 1 +|ss | + | | +|s −s | 12 1 2 over time and space. To address this complexity, IAQ indices is referred to as the absolute degree of grey incidence between are used to describe, classify and improve IAQ by providing X and X . For more details, please refer to the book by Liu easy-to-understand and comprehensive rankings of IAQ levels 1 2 et al.[56]. in buildings. There are many IAQ indices have been proposed all over the world [52]. We adopt the synthetical pollution index method set by Humphreys [53] to analyze and evaluate 2.6 Measures to improve IAQ the level of air quality monitored in library rooms, the following Based on the result of the Section 2.4, we take ventilation and equation gives the IAQ index (I): purification as the main idea to control or improve IAQ. The theory is based on conservation of mass. ⎛ ⎞ ⎛ ⎞ C C C 1 C 1 2 n i ⎜ ⎟ IM=… ⎜ AX ,,… ⎟· (1) ⎜ ⎟ S S Sn S ⎝ ⎠ 1 2 n ⎝ i ⎠ i=1 3 RESULTS AND DISCUSSION C and S are the concentration of measurement and stand- i i ard for the same air pollutant; n is the kind of measured air pol- In this section, China Indoor Air Standard (GB/T 18 883-2002) lutants. ‘C /S ’ is called sub-index. i i [57] was used as reference mode to evaluate the thermal envir- IAQ according to the calculating index (I) can be divided onment and IAQ, in which the range of temperature as well as into five levels [35], as shown in Table 4. relative humidity were 22–28°Cand 40–80%, respectively. Besides, the upper limits of CO , TVOC and formaldehyde con- −3 −3 centrations were 1000 ppm (24 h), 0.6 mg m (8 h) and 0.1 mg m 2.5 Correlation analysis (1 h), respectively. To determine which indoor air pollutant has the most seriously influence on people’s feelings, we determine the correlation 3.1 Comfort parameters (T,RH, CO ) between satisfaction of IAQ and the concentrations of pollu- 2 A summary of temperature and relative humidity is shown in tants based on the gray system theory. The gray system theory Table 5. In the nine test rooms, 3WN has the greatest tempera- is theorized by Deng [54] and developed by Liu [55]. It focuses ture fluctuating ranging from 18.0 to 30.0°C. 3CT and 1WS on the study of problems involving small samples and poor have the lowest 12.4% and highest 47.6% humidity, respectively information and deals with uncertain systems with partially and almost every room has excessively low humidity (<40%) known information through generating, excavating and extract- but no one excessively high humidity (higher than 80%). ing useful information from what is available. Grey incidence Figures 3 and 4 show the number of people and indoor CO analysis based on the gray system theory, provides a new meth- concentration respectively on a typical sampling day. od to analyze which factors have primary influence, and which According to the stable daily routine for most students, the have less influence, on the development of the system. In this number of people had been a gradual increase during 9:00– paper, the model of absolute degree of grey incidence is used 11:00 and reached peak at 11:00 in the rooms (1EW, 1WS, for analysis. The definition of absolute degree of grey incidence 1WN, 3WS, 3WN). With the arrival of the lunch break, the is as follows. number of people was falling rapidly during 11:00–13:00 and reached the lowest at 13:00. Then the number increased quickly Table 4. The level of indoor air quality. again during 13:00–15:00 and remained stable until the library Index Indoor air quality grade Evaluation closed. During the same time period, the number of people in ≤0.49 I Clean other rooms (1EE, 1CL, 1CT, 3CT) had been in a low level all 0.50–0.99 II No pollution the time. 1.00–1.49 III Light pollution As shown in Figure 4, indoor CO concentrations seem to 1.50–1.99 IV Moderate pollution be influenced by human occupancy. Indoor CO concentrations ≥2.00 V Heavy pollution were excessive (except 1CL and 1CT) by the Chinese IAQ 152 International Journal of Low-Carbon Technologies 2018, 13, 148–160 Downloaded from https://academic.oup.com/ijlct/article-abstract/13/2/148/4907958 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Indoor air quality investigation of a university library Table 5. Summary of the average 5-day temperature and relative humidity during the sampling campaign. Room 1EE 1EW 1CL 1CT 1WS 1WN 3WS 3WN 3CT Outdoor Temperature (°C) 21.4 20.3 19.7 15.4 20.2 19.9 20.4 18.0 16.4 9.8 ~27.0 ~27.4 ~25.0 ~21.4 ~29.5 ~28.7 ~31.2 ~30.0 ~21.7 ~23.4 RH (%) 22.4 22.4 29.7 18.2 19.4 22.4 18.7 19.3 12.4 31.5 ~25.3 ~37.4 ~45.2 ~20.7 ~47.6 ~46.0 ~36.2 ~41.5 ~17.7 ~56.1 Figure 3. The variation of people’s number in library rooms during open time. standard which the maximum value is 1000 ppm [57]. In most rooms, CO was beginning to build up when the pupils start occupying the library rooms and then started declining slightly during the lunch break. The same upward trend was followed during 13:00–21:00. It was noteworthy that higher CO concentration happened as the number of people declined during dinner. This might be explained for the fol- lowing reasons: one, CO was accumulated constantly with the doors and windows closed during measurement; the other, the emission of CO by the remainder was still greater than wind seepage. By a simple estimation, there is a balance between CO concentration produced and released based on −2 Figure 4. CO concentration variation in library rooms. (a) The room only the density of 0.1 per m . holding books. (b) The room holding-reading integration. 3.2 TVOC and formaldehyde concentration levels almost more than twice it in other rooms and lots of old books Figure 5 shows the concentration of indoor formaldehyde varies are hold in 1CL and 1CT. during the daily testingperiod. From thefigurewecan seethat As show in Figure 6, the change of TVOC concentration is indoor formaldehyde concentrations were all relatively stable and more gently. TVOC concentration gets peak during 11:00–13:00 had similar trends except for 1EW. This trend can be explained because of some factors such as people’s activities, temperature by formaldehyde gathering due to the sources that are irrelevant and relative humidity. In 1CT and 3CT, TVOC concentration forpeople. Thevolatiledatein1EW might be resultedbythe is much higher than other rooms; the 3CT indoor concentra- operation. In 3CT and 1CT, formaldehyde concentrations are the tion of TVOC is ~40% higher than 3WS indoor. The reason −3 highest, ~0.035–0.04 mg m .Among them,formaldehydecon- might be similar with the explanation of formaldehyde. centration in 3CT is ~60% higher than it in 3WS. The reason It is important that both the concentrations of formaldehyde might be that the placed density of books in 1CL and 1CT is and TVOC are under the levels of Chinese standard [57]. International Journal of Low-Carbon Technologies 2018, 13, 148–160 153 Downloaded from https://academic.oup.com/ijlct/article-abstract/13/2/148/4907958 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Y. Wu et al. Figure 7. PM2.5 concentration variation in library rooms. Figure 5. Formaldehyde concentration variation in library rooms. Figure 6. TVOC concentration variation in library rooms. 3.3 Particulate matter concentrations (PM2.5) Figure 7 shows that the concentration of indoor and outdoor PM2.5 distribution during one typical testing day. Significant strong correlation between indoor and outdoor PM2.5 concentra- tions is found. Outdoor concentration of PM2.5 is excessive ser- iously within the sampling periods, after the biggest growth of PM2.5 during 9:00–11:00, the growth begins slowing gradually and −3 gets peak around 15:00, ~140 ug m . However, the lowest indoor −3 concentration measured is only ~55 ug m at the same time. The concentrations of PM2.5 are the lowest and relatively stable in 1CT and 3CT; the highest are in 1EE, 1WS, 1WN, 3WN, 3WS and similar with outdoor changes in general. The levels indoors are all lower than found outdoors. As shown in Figure 8, taken together, Figure 8. The comparison of PM2.5 concentration indoor and outdoor. (a) indoor PM2.5 concentration is ~40–70% to it outdoor. For the dif- PM2.5 concentration in 3CT and outdoor. (b) PM2.5 concentration in 3WN ference of PM2.5 concentration in each indoor room, there might and outdoor. be other indoor sources such as people’sactivity[58, 59]. In order to facilitate the analysis, the data can be simplified into three categories according to statistical results above: 1EE 3.4 Feelings of people and 1EW; 1WS, 1WN, 3WS and 3WN; 1CL, 1CT and 3CT. The data of questionnaires are shown in Table 6 focused on The results are shown in Table 7. people’s satisfactions and symptoms. 154 International Journal of Low-Carbon Technologies 2018, 13, 148–160 Downloaded from https://academic.oup.com/ijlct/article-abstract/13/2/148/4907958 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Indoor air quality investigation of a university library Table 6. Data statistic of questionnaires. Satisfactions Symptoms Temperature Relative Ventilation Indoor air Dust Breathing Stuffy/pungent Fatigue Irritable Efficiency Dizziness (%) humidity (%) (%) quality (%) feeling (%) difficulty (%) odor (%) (%) feeling (%) decreasing (%) (%) 1EE 63 63 50 50 43 23 23 7 0 13 0 1EW 70 67 43 40 47 27 30 20 7 30 13 1WS 70 67 43 40 30 60 47 37 13 40 30 1WN 63 70 63 43 37 50 30 40 20 30 23 1CL 70 60 20 7 67 63 77 3 0 13 43 1CT 67 60 3 0 83 87 83 10 7 13 63 3WS 83 43 40 43 50 63 23 30 23 30 27 3WN 80 53 33 63 40 77 47 17 10 37 20 3CT 77 40 7 0 87 83 97 0 3 7 60 Table 7. Simplified data statistic of questionnaires. Satisfactions Symptoms Temperature Relative Ventilation Indoor air Dust Breathing Stuffy/pungent Fatigue Irritable Efficiency Dizziness (%) humidity (%) quality (%) feeling difficulty odor (%) (%) feeling decreasing (%) (%) (%) (%) (%) (%) 1EE, 1EW 67 65 47 45 45 25 27 14 4 22 7 1WS, 1WN, 3WS, 3WN 74 58 45 47 39 63 37 31 17 34 25 1CL, 1CT, 3CT 71 53 10 2 79 78 86 4 3 11 55 For people in different rooms, there may be a big difference books, the air quality levels are all ‘no pollution’. This is easy to on subjective feeling of the same parameter. A more intuitive understand: the more seriously exceed the standard, the higher comparison is shown in Figure 9a. The satisfactions of indoor sub-index (C /S ) and the synthetical pollution index (I) are. In i i temperature and relative humidity for people are almost the holding-reading rooms, CO concentrations were excessed 60–80%, similar and higher. It indicates that the satisfactions of very seriously, so the air quality levels in these rooms are ‘pollu- ventilation and air quality are much more different, even <10% tion’. It is inferred that when outdoor air quality is worse in 1CT, 1CL and 3CT. With doors and windows closed during (PM2.5 concentration is pretty high), the levels of IAQ will be the transition season, without air conditioning or ventilation ‘moderate pollution’ and even ‘heavy pollution’ because the system, the concentration of carbon dioxide is high, the ventila- sub-index of PM2.5 will be higher. The upper limit with indoor −3 tion rate is low, and the air is not fresh. PM2.5 concentration is 75 ug m [57], combining with the Figure 9b illustrates that people in the library rooms get conclusion ‘indoor PM2.5 concentration is ~40–70% to it out- noticeable dust feeling, breathing difficulty, stuffy/pungent odor door’ in Section 3.4, it can be concluded that PM2.5 and CO and dizziness symptoms. This might be because people in the are both the primary affecting factors of IAQ when outdoor −3 mall had long exposure time in the library rooms and automat- PM2.5 concentration is over 190 ug m . ically to maintain a stable position. SBS were associated with odors and environmental factors. Odors perception is an early predictor of SBS, and odors are caused by one or more VOCs. 3.6 The correlation between satisfaction of indoor Thus, the overall odors perceptions are influenced by many air quality and the concentrations of pollutants kinds of individual odors, for example, the degradation of We analyze the correlation by the modeling software of grey cellulose-based materials, mold of walls and emission from system theory, downloaded from the web ‘http://igss.nuaa.edu. body [60]. cn’ for free. According to the indoor air levels, 3WS and 3CT is taken as the representative of different type room respectively to determine the correlation. The satisfaction of IAQ by ques- 3.5 The level of IAQ tionnaire and the concentrations of indoor pollutants by field The air quality levels in library rooms are calculated based on measurement are shown in Tables 8 and 9. the data of field measurements, as shown in Figure 10. The After entering the data in Tables 8 and 9 into the software, levels of ‘light pollution’ and ‘moderate pollution’ occurred to we get the results. The correlations between satisfaction of IAQ the rooms, 1EW, 1WS, 1WN, 3WS, 3WN, with the holding- and the concentrations of pollutants arranged from the largest reading integration. In other library rooms that only hold to smallest, are in order, 0.88 (PM2.5), 0.61 (TVOC), 0.58 International Journal of Low-Carbon Technologies 2018, 13, 148–160 155 Downloaded from https://academic.oup.com/ijlct/article-abstract/13/2/148/4907958 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Y. Wu et al. 80% (a) 70% 60% 50% 40% 30% 20% 10% 0% tempreature relative humidity ventilation air quality 1EE, 1EW 1WS, 1WN, 3WS, 3WN 1CL, 1CT, 3CT Dust feeling (b) 100% 1EE, 1EW 80% Breathing Dizziness 1WS, 1WN, 3WS, 3WN 60% difficulty 1CL, 1CT, 3CT 40% 20% 0% Efficiency Stuffy/pungent decreasing odor Irritable Fatigue feeling Figure 9. Subjective satisfaction reported by people in the library booking rooms. (a) Satisfactions with indoor environment. (b) SBS symptoms indoor environment. between indoor air satisfaction and the concentrations of formal- 1EE 1EW dehyde and TVOC noteworthy, especially in the rooms only 1CL 1.7 1CT books. 1.6 1WS It can be seen that the primary indoor air pollutant is not 1.5 1WN 3CT the same in different room. In order to build a more satisfac- 1.4 3WN 1.3 tory indoor environment, the requirements of fresh air and 3WS no pollution 1.2 the functionality of the space usages should be considered in 1.1 the ventilation system or central air-conditioning system 1.0 design. 0.9 0.8 0.7 3.7 The effect of IAQ improvement in theory 0.6 0.5 From Section 3.5, we get the conclusion: when outdoor air quality is good, CO concentration is the primary factor affect- 9:00 11:00 13:00 15:00 17:00 19:00 21:00 ing IAQ; when outdoor air quality is poor (outdoor PM2.5 con- Time (h) −3 centration is over 190 ug m ), IAQ is affected by the Figure 10. The synthetical pollution index (I) of indoor air quality in library concentrations of PM2.5 and CO . According to research [61], rooms. the plan with the theme of ventilating rapidly in a short time to decrease CO concentration and purifying to eliminate PM2.5 (formaldehyde) and 0.57 (CO )in 3WS, 0.73 (CO ), 0.63 by ventilating is determined. While for mechanical ventilation, 2 2 (PM2.5), 0.58 (formaldehyde), 0.58 (TVOC) in 3CT. It is no temperature difference between indoor and outdoor air was doubt that IAQ are affected seriously by CO and PM2.5 accord- ignored based on the assumption that there was an electric ing the field measurement, however, the strong correlation heating or heat recovery unit at air inlet. 156 International Journal of Low-Carbon Technologies 2018, 13, 148–160 Downloaded from https://academic.oup.com/ijlct/article-abstract/13/2/148/4907958 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Comprehensive index Indoor air quality investigation of a university library Table 8. The data of satisfaction and measured concentrations in 3WS. Time 9:00 11:00 13:00 15:00 17:00 19:00 21:00 Satisfaction of indoor air quality (%) 95 77 72 56 61 69 55 CO (ppm) 715 1469 1414 1744 1880 2180 2422 −3 PM2.5 (ug m) 465369 808781 80 −3 Formaldehyde (mg m ) 0.025 0.027 0.026 0.021 0.027 0.028 0.026 −3 TVOC (mg m ) 0.178 0.192 0.193 0.198 0.189 0.197 0.189 Table 9. The data of satisfaction and measured concentrations in 3CT. Time 9:00 11:00 13:00 15:00 17:00 19:00 21:00 Satisfaction of indoor air quality (%) 3 5 6 5 4 4 2 CO (ppm) 633 873 892 950 1060 1084 1170 −3 PM2.5 (ug m ) 88 92 106 100 104 112 100 −3 Formaldehyde (mg m ) 0.037 0.041 0.039 0.042 0.039 0.038 0.038 −3 TVOC (mg m ) 0.262 0.284 0.27 0.278 0.281 0.274 0.272 The calculation of required fresh air increment for the con- trol of indoor CO concentration is based on a general mass balance equation as shown in the following equation. dC ′ nQ =( NC′ − C ′)+ (2) oi dτ V Where, C′ is the concentration of outdoor CO ; C′ is the con- o i −1 centration of indoor CO , ppm; is the time period, s ; N is −1 the rate of indoor ventilation, h ; n is the number of indoor people, per; Q is the releasing rate of CO for human body, ml/ (per s); V is the room volume, m . The volume fraction of outdoor CO is 0.03%, national stand- ard of indoor CO concentration is 1000 ppm and the correspond- ing volume fraction is 0.1%. The function is holding-reading in 3WS, CO concentration is at much higher level, so choose 3WS as an example to calculate. When indoor CO concentration is 1500 ppm, begin to ventilate mechanically to decrease indoor CO Figure 11. CO concentration variation in 3WS in mechanical ventilation. concentration to the national standard of 80% or 800 ppm. The −1 time interval of iterating calculation is 1 min, indoor ventilating and infiltration wind and air purifier, h , N is the 0.5 times −1 rate (N) is 10 times/h, indoor CO concentration can be reduced h [43], N is the 0 when the window is open; P is the penetra- to 781 ppm ~5 min. Because the CO concentration in 3WS is tion coefficient, P is 1 when the window is open, P is 0.8 [62] much higher, the calculating results can be applied in other when the window is closed; η is the filter efficiency of air puri- holding-reading rooms. The results of calculating are shown in fier for first time; K is the sedimentary rate of indoor particles, −1 Figure 11. is always 0.000025 s ; the secondary suspension has little In the rooms only holding books, CO concentration gets effect, therefore, ignored. beyond the national standard until 15:00 but excessive rarely, so Taking 1WS as a represent for it is one of the rooms that ventilating naturally can decrease obviously CO concentration. PM2.5 pollution is the most serious. The area is 650 m in 1WS For PM2.5, we established equation (3) based on a general and the processing size of the air purifier is 30–50 m ,so10 air mass balance equation: purifiers are set averagely. Beginning to calculate when PM2.5 −3 −1 concentration is 180 ug m . N is ~5 times h , η is 90%. During dC =− NC NC+ NPC− NC− N η C− KC (3) ventilating, N is 0, so equation (3) can be simplified into following no n i i o i i p ii i dτ equation. −3 where, C is the indoor PM2.5 concentration, ug m ; C is the i o −3 −1 dC outdoor PM2.5 concentration, ug m ; τ is the time period, s ; =− NC NC− N η C− KC (4) no n i p ii N , N , N are the air changing rate respectively for ventilation dτ n i p International Journal of Low-Carbon Technologies 2018, 13, 148–160 157 Downloaded from https://academic.oup.com/ijlct/article-abstract/13/2/148/4907958 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Y. Wu et al. � The correlations between satisfaction of IAQ and the concentra- tions of indoor air pollutants are not as a constant, but changing with different rooms. So the requirements of fresh air and the functionality of the space usages should be considered in the ventilation system or central air-conditioning system design. � One simple plan of improving IAQ with the theme of ‘venti- lating rapidly in a short time along with purifying’ is deter- mined based on a general mass balance equation. In theory, for the most unfavorable condition, CO concentration can be decreased from 1500 to 800 ppm <5 min by ventilating −1 mechanically with 10 times h ventilating rate, PM2.5 con- −3 centration can be decreased from 180 to 75 ug m ~2–3min by setting air purifiers averagely for area. In this research, only four types of air pollutants were measured namely CO , PM2.5, formaldehyde and TVOC. More detailed and comprehensive survey is recommended to include other Figure 12. PM2.5 concentration variation opening the purifier after ventilat- pollutants such as benzene, mould and Radon. The impacts of ing in 1WS. formaldehyde and TVOC on the health of people studying and working in library rooms should be highly considered. The −1 During calculating, N is 10 times h , indoor and outdoor measurement of a Long term and wide range is vital and should −3 −3 PM2.5 concentrations are 60 ug m and 270 ug m . be carried out in the future. If air purifiers are no working during ventilating, so N is 0. By −3 calculating, indoor PM2.5 concentration increases to 184.91 ug m after ventilating for 5 min. Indoor PM2.5 concentration decreases −3 ACKNOWLEDGMENTS to 56.72 ug m after air purifier working for 2 min. The calculat- ing results are as shown in Figure 12. This research was financially supported by National Key Tech- Synthesize the above results, the preliminary plan of improv- nologies R&D Program (2016YFB0601700), National Natural Sci- ing IAQ with the theme ‘ventilating rapidly in a short time and ence Foundation of China (51 076 011), Funding for Academic along with purifying’ is feasible. Collaboration between USTB and NTUT (TW201606), The Opening Funds of State Key Lab of Subtropical Building Science, South China University of Technology (2012KB08). 4 CONCLUSIONS This article presents the onsite measurement of nine library rooms REFERENCES without central HVAC systems in the University of Science and Technology Beijing during non-heating period (April in 2016). [1] Guy Robertson. Robertson on Library Security and Disaster Planning: [M]. The field measurements including indoor and outdoor air tem- Elsevier, 2016, 41–4. perature, relative humility and the concentrations of CO ,PM2.5, [2] Li N, Li J, Fan R et al. Probability of occupant operation of windows during formaldehyde, TVOC and questionnaire surveys are measured transition seasons in office buildings [J]. Renew Energy 2015;73:84–91. and analyzed. Results show that each library room has diverse [3] Mihucz VG, Záray G. The quality of air [J]. Compr Anal Chem 2016;73: IAQ. One improving plan is made based on the results. The main 45–71. findings are summarized as follows: [4] Shi S, Chen C, Zhao B. Air infiltration rate distributions of residences in Beijing [J]. Build Environ 2015;92:528–37. [5] Fenech A, Strlic M, Kralj Cigic I et al. Volatile aldehydes in libraries and � Various indoor air pollutants with varying concentrations archives [J]. Atmos Environ 2010;44:2067–73. were detected in the library rooms in USTB, in terms of the [6] Marchand C, Bulliot B, Lecalve S et al. Aldehyde measurements in indoor high CO concentration of 575–2400 ppm, the changing environments in Strasboung (France). Atmos Environ 2006;40:1336–45. PM2.5 concentration (40–70% of the outdoor), the highest [7] Kim J, Kim S, Lee K et al. Indoor aldehydes concentration and emission concentration of formaldehyde and TVOC is ~0.042 and rate of formaldehyde in libraries and private reading rooms. Atmos Environ 0.285 mg/m , respectively, with half of the upper limit. 2013;71:1–6. � CO , PM2.5, formaldehyde, TVOC concentrations are differ- [8] Strict M. Material demographics: on the smell of old books. Anal Chem ent in different library rooms. The concentrations of CO 2009;81:8617–22. and PM2.5 in the holding-reading rooms are higher than [9] Radiello. The Radial Diffusive Sampler, Fondazzione Salvatore Maugeri that in the rooms only holding books. The situation of for- IRCCS. Retrieved from: http://www.radiello.com/english/ald_en.htm on 1 maldehyde and TVOC concentrations is just opposite. September 2009. 158 International Journal of Low-Carbon Technologies 2018, 13, 148–160 Downloaded from https://academic.oup.com/ijlct/article-abstract/13/2/148/4907958 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Indoor air quality investigation of a university library [10] Bu Z, Zhang Y, Mmereki D et al. Indoor phthalate concentration in resi- [32] Lin CC, Yu KP, Zhao P et al. Evaluation of impact factors on VOC emis- dential apartments in Chongqing, China: implications for preschool chil- sions and concentrations from wooden flooring based on chamber tests [J]. dren’s exposure and risk assessment [J]. Atmos Environ 2016;127:34–45. Build Environ 2009;44:525–33. [11] Ashford NA, Caldart CC. Negotiated environmental and occupational [33] Cheng L, Li B, Cheng Q et al. Investigations of indoor air quality of large health and safety agreements in the United States: lessons for policy. department store buildings in China based on field measurements [J]. J Clean Prod 2001;2:99–120. Build Environ 2017;118:128–43. [12] Tang X, Bai Y, Duong A et al. Formaldehyde in China: production, consump- [34] Su Y, Tao J, Wang M et al. Surveillance for fine particles concentration in tion, exposure levels, and health effects [J]. Environ Int 2009;35:1210–24. air in Shijingshan district of Beijing [J]. Dis Surveill 2015;30:672–8. [13] Jiang C, Li S, Zhang P et al. Pollution level and seasonal variations of car- [35] Hu J. Spatial and temporal variability of PM20.5 and PM10 over the North bonyl compounds, aromatic hydrocarbons and TVOC in a furniture mall China Plain and the Yangtze River Delta, China [J]. Atmos Environ 2014; in Beijing, China [J]. Build Environ 2013;69:227–32. 95:598–609. [14] Boeglin ML, Wessels D, Henshel D. An investigation of the relationship [36] Zhang Y. Impact of synoptic weather patterns and inter-decadal climate between air emissions of volatile organic compounds and the incidence of variability on air quality in the North China Plain during 1980–2013 [J]. cancer in Indiana counties [J]. Environ Res 2006;2:242–54. Atmos Environ 2014;95:119–28. [15] Eigurenfernandez A, Shinyashiki M, Schmitz DA et al. Redox and electro- [37] Wang Y. Spatial and temporal variations of six criteria air pollutants in 31 philic properties of vapor-and particle-phase components of ambient aero- provincial capital cities in China during 2013–2014 [J]. Environ Int 2014;73: sols [J]. Environ Res 2010;110:207–12. 413–22. [16] Kelly FJ, Fussell JC. Air pollution and airway disease [J]. Clin Exp Allergy [38] WHO, 2006. Air quality guidelines for particulate matter, ozone, nitrogen 2011;41:207–12. dioxide and sulfur dioxide: global update 2005: summary of risk assessment. [17] Martins PC, Valente J, Papoila AL et al. Airways changes related to air pol- [39] Guo H, Wang Y, Zhang H. Characterization of criteria air pollutants in lution exposure in wheezing children [J]. Eur Respir J 2012;39:246–53. Beijing during 2014–2015 [J]. Environ Res 2017;154:334–44. [18] Baccioglu A, Kalpaklioglu AF. An unusual form of formaldehyde induced [40] Liang W, Yang C, Yang X. Long-term concentrations of volatile organic lung disease [J]. Allergologia et Immunopathologia 2007;35:110–2. compounds in a new apartment in Beijing, China [J]. Build Environ 2014; [19] Klinmalee A, Srimongkol K, Oanh NTK. Indoor air pollution levels in pub- 82:693–701. lic buildings in Thailand and exposure assessment [J]. Environ Monit [41] Wallace LA, Emmerich SJ, Howard-Reed C. Continuous measurements of Assess 2009;156:581–94. air change rates in an occupied house for 1 year: the effect of temperature, [20] Hulin M, Simoni M, Viegi G et al. Respiratory health and indoor air pollu- wind, fans, and windows [J]. Expo Anal Environ Epidemiol 2002;12: tants based on quantitative exposure assessments [J]. Eur Respir J 2012;40: 296–306. 1033–45. [42] Chen S, Levine MD, Li H et al. Measured air tightness performance of resi- [21] Bakian AV, Huber RS, Coon H et al. Acute air pollution exposure and risk dential buildings in North China and its influence on district space heating of suicide completion [J]. Am J Epidemiol 2015;181:295. energy use [J]. Energy Build 2012;51:157–64. [22] Tsai DH, Lin JS, Chan CC. Office workers’ sick building syndrome and indoor [43] Gu H, Cao Y, Yang J. Probe into indoor air change frequency of residences in carbon dioxide concentrations [J]. JOccup EnvironHyg 2012;9:345–51. the area hot in summer and cold in winter [J]. Housing Sci 2012;32:36–9. [23] Ng LC, Musser A, Persily AK et al. Indoor air quality analyses of commer- [44] Standard for technical specifications for monitoring of indoor air quality of cial reference buildings [J]. Build Environ 2012;58:179–87. the People’s Republic of China. JGJ/T 167-2004 (in Chinese). [24] Takigawa T, Saijo Y, Morimoto K et al. A longitudinal study of aldehydes [45] Zhu Y. Built Environment [M] 0.3. Tsinghua University, Beijing, 2010: and volatile organic compounds associated with subjective symptoms 141–2. related to sick building syndrome in new dwellings in Japan [J]. Sci Total [46] Fanger PO. Air quality not just air quality. ASHRAE J 1989;31:46–9. Environ 2012;417-418:61–7. [47] ASHERA. Public Review Draft 62-1989. Ventilation for acceptable indoor [25] Sahlberg B, Gunnbjörnsdottir M, Soon A et al. Airborne molds and bac- air quality. 1989. teria, microbial volatile organic compounds (MVOC), plasticizers and for- [48] ASHER. A Public Review Draft 62-1989 R. Ventilation for acceptable maldehyde in dwellings in three North European cities in relation to sick indoor air quality. 1996. building syndrome (SBS) [J]. Sci Total Environ 2012;444 C:433–40. [49] Pearce N, Weiland S, Keil U et al. Self-reported prevalence of asthma [26] Tao H, Fan Y, Li X et al. Investigation of formaldehyde and TVOC in symptoms in children in Australia, England, Germany and New Zealand: underground malls in Xi’an, China: concentrations, sources, and affecting an international comparison using the ISAAC protocol [J]. Eur Respir J factors [J]. Build Environ 2015;85:85–93. 1993;6:1455–61. [27] Gao J-M, Chen L, Zhang Y et al. Occurrence and pollution source of [50] Bornehag CG, Sundell J, Hagerhedengman L et al. ‘Dampness’ at home TVOC in underground stores in Chongqing [J]. Cent S Univ (Sci Technol) and its association with airway, nose, and skin symptoms among 10,851 2012;43:4554–8. preschool children in Sweden [J]. Indoor Air 2005;15:48–55. [28] An J-Y, Kim S, Kim H-J. Formaldehyde and TVOC emission behavior of [51] Wang H, Li B, Yu W et al. Early-life exposure to home dampness asso- laminate flooring by structure of laminate flooring and heating condition ciated with health effects among children in Chongqing, China [J]. Build [J]. J Hazard Mater 2011;187:44–51. Environ 2016;94:327–34. [29] Lim J, Kim S, Kim A et al. Behavior of VOCs and carbonyl compounds [52] Wei W, Ramalho O, Derbez M et al. Applicability and relevance of six emission from different types of wallpapers in Korea [J]. Environ Res indoor air quality indexes [J]. Build Environ 2016;109:42–9. Public Health 2014;11:4326–39. [53] Humphreys Michael A. Quantifying occupant comfort: are combined [30] Chi C, Chen W, Guo M et al. Law and features of TVOC and formalde- indices of the indoor environment practicable. Build Res Inf 2005;33: hyde pollution in urban indoor air [J]. Atmos Environ 2016;132:85–90. 317–25. [31] Kim S, Kim J-A, Kim H-J et al. Determination of formaldehyde and TVOC [54] Deng J. Control problems of grey systems [J]. Syst Control Lett 1982;1:9–18. emission factor from wood-based composites by small chamber method. [55] Liu S. Emergence and development of grey system theory and its forward Poly Test 2006;25:605–14. trends [J]. J Nanjing Univ Aeronaut Astronautics 2004;36:267–71. International Journal of Low-Carbon Technologies 2018, 13, 148–160 159 Downloaded from https://academic.oup.com/ijlct/article-abstract/13/2/148/4907958 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Y. Wu et al. [56] Liu S, Yang Y, Wu L. The Grey System Theory and its Application [M]. 7. [60] Yang S, Gao K, Yang X. Volatile organic compounds (VOCs) formation Science Press, 2014. due to interactions between ozone and skin-oiled clothing: measure- [57] GB/T 18883–2002. Indoor Air Quality Standard. Ministry of Health, Beijing ments by extraction-analysis-reaction method [J]. Build Environ 2016; (2000) [in Chinese]. 103:146–54. [58] Ramos CA, Reis JF, Almeida T et al. Estimating the inhaled dose of pollutants [61] Ma H, Shao X, Li X. Feasibility analysis on combination strategies of win- during indoor physical activity [J]. Sci Total Environ 2015;527–528:111–8. dow ventilation and air cleaners in residential buildings under haze wea- [59] Zhang M, Zhang S, Feng G et al. Indoor airborne particle sources and out- ther. J HV&AC 2016;46:18–23. door haze days effect in urban office areas in Guangzhou [J]. Environ Res [62] Chen C. Effect of outdoor inhaled particles on indoor air quality and its 2017;154:60–5. control. Ph.D. Thesis. Tsinghua University, Beijing, 2012. 160 International Journal of Low-Carbon Technologies 2018, 13, 148–160 Downloaded from https://academic.oup.com/ijlct/article-abstract/13/2/148/4907958 by Ed 'DeepDyve' Gillespie user on 21 June 2018
Journal
International Journal of Low-Carbon Technologies – Oxford University Press
Published: Feb 24, 2018