“What impacts does increasing airtightness have on mould, condensation and measures of indoor air quality?” Rapid Review Report V 2.0 Authors Leela Kempton, Daniel Daly, Mark Dewsbury
Sustainable Buildings Research Centre, University of Wollongong [email protected] [email protected]
Title What impacts does increasing airtightness have on mould, condensation and measures of indoor air quality?
ISBN
Date September 2020
Keywords Airtightness, Ventilation, Indoor air quality, mould, moisture, condensation
Publisher CRC LCL
Preferred citation Kempton, L, Daly, D, Dewsbury, M (2020), Rapid Review: What impacts does increasing airtightness have on mould, condensation and measures of indoor air quality? CRC Low Carbon Living, Sydney.
Rapid Review - Airtightness 2
Acknowledgements This research is funded by the CRC for Low Carbon Living Ltd supported by the Cooperative Research Centres program, an Australian Government initiative
Disclaimer Any opinions expressed in this document are those of the authors. They do not purport to reflect the opinions or views of the CRCLCL or its partners, agents or employees. The CRCLCL gives no warranty or assurance and makes no representation as to the accuracy or reliability of any information or advice contained in this document, or that it is suitable for any intended use. The CRCLCL, its partners, agents and employees, disclaim any and all liability for any errors or omissions or in respect of anything or the consequences of anything done or omitted to be done in reliance upon the whole or any part of this document.
Peer review statement The CRCLCL recognises the value of knowledge exchange and the importance of objective peer review. It is committed to encouraging and supporting its research teams in this regard. The author(s) confirm(s) that this document has been reviewed and approved by the project’s steering committee and by its program leader. These reviewers evaluated its: originality methodology rigour compliance with ethical guidelines conclusions against results conformity with the principles of the Australian Code for the Responsible Conduct of Research (NHMRC 2007), and provided constructive feedback which was considered and addressed by the author(s).
Abbreviations and acronyms ACH – Air changes per hour, how many times the air volume in a space is completely replaced in an hour.
ACH50 – Air changes per hour at an indoor – outdoor pressure difference of 50 Pa, used to present the results from blower door testing. CO – Carbon monoxide
CO2 – Carbon dioxide HCOH - Formaldehyde IAQ – indoor air quality MVHR – mechanical ventilation with heat recovery NCC – National Construction Code
NO2 – Nitrogen dioxide
PM2.5 – Particulate matter, very small particles with diameter of 2.5 micrometres (0.0025 mm) or smaller. VOC – volatile organic compounds WHO - World Health Organisation
Registration1 This protocol is archived in an Open Science Framework project folder https://osf.io/2m9wf/.
1 The archived protocol follows PRISMA-P format, as outlined in Moher D, Shamseer L, Clarke M, Ghersi D, Liberati A, Petticrew M, Shekelle P, Stewart LA, PRISMA-P Group Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement. Syst Rev. 2015;4:1
Rapid Review - Airtightness 3
Contents Acknowledgements...... 3 Disclaimer ...... 3 Peer review statement ...... 3 Abbreviations and acronyms ...... 3 Registration...... 3 List of tables ...... 5 List of figures ...... 5 Executive summary ...... 6 Objectives of this Rapid Review ...... 6 Key findings of this rapid review ...... 6 Methods, limitations and future research ...... 6 Summary and conclusions ...... 7 Introduction ...... 8 Rationale ...... 8 Objectives ...... 8 Method ...... 8 Eligibility criteria ...... 8 Information sources ...... 8 Search strategy ...... 8 Study records ...... 9 Outcomes and prioritisation ...... 10 Risk of bias of individual studies ...... 10 Deviations from protocol ...... 10 Data synthesis ...... 10 Meta-bias(es) ...... 10 Confidence in cumulative evidence ...... 10 Data items ...... 11 Results ...... 12 Overview of the included studies ...... 12 Qualitative summary of study findings ...... 21
Carbon dioxide (CO2) ...... 21
Particulate levels (PM2.5) ...... 21 Formaldehyde (HCHO) ...... 21 Volatile organic compounds (VOC) ...... 21
Nitrogen dioxide (NO2) ...... 22 Mould / moisture ...... 22 Other ...... 23 Summary ...... 23 Quality, risk of bias and confidence in cumulative evidence ...... 38
Rapid Review - Airtightness 4
Overview of the excluded studies ...... 38 Excluded studies related to school buildings ...... 39 Review limitations ...... 39 Summary and conclusions ...... 39 Resources, workload and timeline ...... 40 Included studies ...... 42 Excluded studies ...... 43 Other references ...... 44
List of tables
Table 1. List of the main study variables that were extracted and coded for the included studies, with relevant values .. 11 Table 2. Main characteristics of the included studies ...... 13 Table 3. Sample sizes, data measurements and findings of included studies ...... 16 Table 4: Summary of impact of increased airtightness on IAQ parameters based on the 21 reviewed studies...... 23
Table 5: Summary of results for CO2 ...... 24
Table 6: Summary of results for PM2.5 ...... 26 Table 7: Summary of results for formaldehyde ...... 29 Table 8: Summary of results for VOC ...... 31
Table 9: Summary of results for NO2 ...... 33 Table 10: Summary of results for mould/moisture ...... 35 Table 11: Summary of results for other IAQ factors ...... 37 Table 12: Excluded studies ...... 38 Table 13: Excluded studies related to airtightness in schools...... 39 Table 14: Workloads (in hours) of the team members for each main review stage ...... 41
List of figures Figure 1. Airtightness is measured using a blower door test...... 6 Figure 2.Green walls are designed to improve indoor air quality ...... 7 Figure 3. PRISMA flow diagram, providing an overview of the search and screening ...... 10 Figure 4. Air change rates of properties covered in each included study ...... 12 Figure 5: Review team members ...... 40 Figure 6: Review timeline...... 40
Rapid Review - Airtightness 5
home (for instance, new paint is a major source of volatile organic compounds (VOCs) in a home) or Executive summary from outdoor air entering the home (for example, small particulates from motor vehicles or bushfire Objectives of this Rapid Review smoke events). Mould has also been identified as a possible risk, as there is the potential for increased Leaky homes need more energy to keep warm or condensation if water vapour control is not cool. Sealing the gaps in a house, or increasing adequately considered. airtightness, is a common tip to reduce energy consumption, as well as make the house more comfortable to live in. However, people need a Key findings of this rapid review certain amount of fresh air in their homes to The main finding of this review was that if the maintain air quality, and concerns have been raised pollutant is from an external source, airtightness is that increased airtightness may have the likely to be good for IAQ, whereas if the pollutant is unintended consequence of reduced indoor air from an internal source, airtightness will likely quality (IAQ). worsen IAQ. Airtightness is only part of the story though. Fixing For locations with high levels of particulate matter air leaks in a building will reduce the air movement (PM2.5) or nitrogen dioxide (NO2), in the outside air, through cracks and gaps, known as infiltration. a relatively clear trend was identified that tighter However, there are other sources of fresh air in a buildings resulted in lower pollutant concentrations. building. If designed appropriately, natural Conversely, levels of VOC and carbon dioxide ventilation (which is air exchange through doors and (CO2) were higher in more airtight buildings, as windows) and mechanical ventilation (such as these pollutants are generated indoors, and have exhaust fans or ducted ventilation systems) can less ability to escape from an airtight building. ensure minimum fresh air levels are maintained in a However, further research is still needed to better building regardless of airtightness. This is captured understand this relationship. in the maxim “seal tight and ventilate right”. Most building regulations require mechanical ventilation This review also highlighted that the issue of mould to be installed once airtightness goes beyond a and mould risk is complex. Nine studies reported on certain level. mould, or mould related risk factors (e.g. relative humidity). Three studies used direct mould Increased stringency in airtightness requirements observation, and two of these found that increased has been foreshadowed as a possible inclusion in airtightness actually reduced the occurrence of future versions of the National Construction Code mould. This goes against commonly held views that 23 (NCC) so it is important to understand what, if increasing airtightness may increase mould risk. any, unintended consequences there might be. Our The study authors put this down to leaky homes review explored the recent literature to answer the having more issues with water leakage, though it question “What impacts does increasing could be an indication of confounding variables. airtightness have on mould, condensation and Two of the four studies that measured relative measures of indoor air quality?” humidity found an increase which can increase the IAQ is a catch-all term to indicate the potential risk of mould in airtight homes. The remaining two health risk in a building from a range of pollutants. studies found no relationship). The two simulation These pollutants can come from items within the studies also showed a similar increased risk of mould growth for more airtight homes.
Methods, limitations and future research In total, 21 studies were included in this review. These studies were published between 1998 and 2019, and covered 10 countries, with most studies coming from the United Kingdom (7) or North America (6). The studies cover a variety of climates, although the largest number of studies were in temperate climates similar to the cooler temperate climate of south-eastern Australia. The reviewed studies took place in a mix of existing Figure 1. Airtightness is measured using a blower buildings, buildings being upgraded and new builds. door test. Detached houses, apartments and semi-detached
2 ASBEC (2018), ‘Built to Perform An industry led pathway to a 3 ABCB (2020), ‘NCC 2022 Residential Energy Efficiency. zero carbon ready building code’, Australian Sustainable Built Frequently asked questions’, Australian Building Codes Board, p. Environment Council, Building Code Energy Performance 3, Q 13, accessed at https://www.abcb.gov.au/- Trajectory Project - Final Report July 2018 /media/Files/Resources/Education- Training/FAQ_NCC2022_Residential_Energy_Efficiency.pdf.
Rapid Review - Airtightness 6
properties were covered. To work out how airtight a Research focused on the threshold at which building was, most studies used a blower door test. mechanical ventilation is typically There was a range of tightness levels, from very implemented (usually 5 air changes per hour airtight to very leaky, but most studies were at 50Pa - ACH50) focussed on more airtight properties. Larger and more long-term studies are still The review had a number of important limitations. required to improved confidence in this area. An important question that could not be answered by this current review was what impact the Summary and conclusions occupant of the homes in these studies has on IAQ through their ventilation practices. Advocates of This rapid review identified 21 studies that airtight homes suggest appropriate ventilation will investigated the impact of increasing airtightness on mitigate any increased risk of mould or pollutants. indoor air quality. These studies covered a broad However this necessitates either mechanical range of locations, climates and building types. ventilation (which is not currently common in Indoor air quality parameters investigated were Australian homes) or engaged and educated CO2, PM2.5, formaldehyde, VOC, NO2, relative occupants. humidity, mould issues, carbon monoxide (CO) and radon. Another limitation in the studies was sample size. Most of the studies had a sample size of less than Based on the studies reviewed, there was limited 25 homes, which makes it hard to draw general evidence to identify direct correlations between conclusions. The length of time each home was increasing airtightness and indoor air quality in monitored also varied, meaning some studies took general. A negative correlation with CO2 and VOC place in both winter and summer, while others did concentration was found from the studies, with not, and short-term changes in outdoor pollutant concentrations increasing with a decrease in the air levels may have been important. This means the exchange rate. A positive correlation was found comparisons are in some cases less valid. between the air exchange rate and PM2.5 and NO2 concentrations in areas where there were high Three important pieces of future work would be: outdoor levels. In these cases, increasing To consider airtightness in the context of airtightness was found to reduce the infiltration of total ventilation rates for a home. This is outdoor contaminants. There were no direct particularly important for understanding correlations identified for mould issues, mould risk, which appears to have a complex formaldehyde, radon or CO, or for PM2.5 or NO2 in relationship with airtightness. areas with average outdoor levels.
Figure 2. Green walls are designed to improve indoor air quality
Rapid Review - Airtightness 7
Introduction Method
Rationale Eligibility criteria Increased airtightness is a commonly recommended The following study characteristics were used as strategy to increase the energy efficiency and thermal inclusion criteria for the review: comfort of residential buildings. Increased stringency in airtightness requirements has been foreshadowed as a 1. Studies published in peer-reviewed academic possible potential inclusion in future versions of the journals, or grey literature from reputable National Construction Code. Internationally, increased organisations. airtightness has been adopted into many building codes, 2. Studies published in English including requirements for mechanical ventilation when 3. Studies on residential buildings the air change rate is less than 5 ACH . Presently, 50 4. Studies that measured or modelled different mechanical ventilation is not widely used in Australia so this will require a shift in construction practices. infiltration or airtightness rates. 5. Studies that measure or model a specific The University of Wollongong (UOW) node of the impact IAQ – e.g. moisture levels, mould Cooperative Research Centre on Low Carbon Living growth, CO , PM , etc (CRC LCL) carried out residential building energy and 2 2.5 thermal performance simulations for ‘Built to Perform An 6. Full text available. industry led pathway to a zero carbon ready building code’. In this work, increasing airtightness of residential An iterative process was used during the actual review buildings was found to be the most cost-effective process to refine the above criteria to either scale up or measure available in terms of increasing the energy down the number of resulting potential studies in order to efficiency (NatHERS star rating) of residential buildings. achieve the desired number for analysis. Scaling was done by adding further restrictive inclusion criteria, as However, increased envelope airtightness can also described in the “Search Strategy” and “Deviations from potentially reduce Indoor Air Quality (IAQ), and increase Protocol” sections below. the risk of elevated moisture levels, condensation and mould if insufficient ventilation in the building is provided, or insufficient means are provided for water vapour Information sources transfer and control in the envelope itself. This can 1. The main (core) searches were conducted cause both surface and internal damage to the building using two multidisciplinary broad-coverage fabric, as well as negative impacts on the health of occupants. academic literature databases: Scopus and Web of Science. Objectives 2. Building specific research organisations (BRANZ, BRE). The main objective was to rapidly synthesize knowledge 3. Google Scholar was also used to identify from primary literature to answer the question: additional applicable studies. “What impacts does increasing airtightness have on 4. Snowballing (forward and backward reference mould, condensation and measures of indoor air screening) from the included studies using quality?” Scopus search engine and database was also It should be noted that while this is an extremely used. important question, it is also broad. Thus, given the limited resources available for the proposed Rapid Search strategy Review, the review is focused on some important sub- topics. To construct search strings, we used combinations of keywords and phrases related to airtightness, buildings What evidence is there of benefits and negative and indoor air quality parameters. Several search strings impacts from changing airtightness on mould and were tested in SCOPUS before a suitable search was condensation risk and indoor air quality? identified, including the following: How do these impacts vary with the climatic conditions considered? What measures (e.g. imposed ventilation) are Search string for SCOPUS (search date 29/01/2020): TITLE-ABS-KEY ( airtight* OR air-tight* ) AND TITLE- required to mitigate the impacts of increased ABS-KEY ( building OR hous* OR home OR haus airtightness? OR residential ) AND TITLE-ABS-KEY ( "indoor air Due to the limited timeframe, this review was not a quality" OR "indoor environment quality" OR moisture comprehensive review of all available evidence, rather OR mould ) AND DOCTYPE ( ar OR re ) [275 hits] the analysis was of a representative sample, identified via a rapid systematic review method (Tricco et al., 2015)
Rapid Review - Airtightness 8
Study records Search string for SCOPUS (search date 29/01/2020): Records from Scopus and Web of Science electronic TITLE-ABS-KEY ( airtight* OR air-tight* ) AND TITLE- databases were exported to Mendeley reference ABS-KEY ( building OR hous* OR home OR management software. One reviewer screened all de- residential OR haus ) AND TITLE-ABS-KEY ( duplicated records (titles, abstracts, keywords) to identify hygrothermal OR mould ) AND DOCTYPE ( ar OR re relevant studies using the eligibility criteria described in ) [60 hits] the previous section. Full papers were retrieved for 39 studies deemed potentially relevant. One reviewer screened all full Search string for SCOPUS (search date 29/01/2020): papers using the same criteria for the titles and TITLE-ABS-KEY ( airtight* OR air-tight* OR infiltration ) abstracts, with a second reviewer checking the review AND TITLE-ABS-KEY ( building OR hous* OR home and final list of included studies and their coding (i.e. OR haus OR residential ) AND TITLE-ABS-KEY ( data extraction) produced by another reviewer . moisture OR hygrothermal OR mould ) AND DOCTYPE ( ar OR re ) [345 hits] In addition, relevant studies produced by building specific research organisations such as BRANZ and BRE were reviewed and included in the scope. Google Search string for SCOPUS (search date 29/01/2020): Scholar was also used in a limited capacity, as all search ( TITLE-ABS-KEY ( airtight* OR air-tight* ) AND terms returned large numbers of articles. TITLE-ABS-KEY ( building OR hous* OR home OR Finally, one reviewer performed snowballing (i.e. forward residential OR haus ) AND TITLE-ABS-KEY ( and backward reference searches) for the studies moisture OR hygrothermal OR mould ) AND deemed suitable for inclusion using Scopus online DOCTYPE ( ar OR re ) [137 hits] database. The overview of the search and screening process is presented as a PRISMA flow diagram in Figure 3. . Search string for SCOPUS (search date 29/01/2020): TITLE-ABS-KEY ( airtight* OR air-tight* ) AND TITLE- The final search terms were: ABS-KEY ( building OR hous* OR home OR haus Search string for SCOPUS (search date 10/02/2020): OR residential ) AND TITLE-ABS-KEY ( moisture OR ( TITLE-ABS-KEY ( airtight* OR air-tight* ) AND hygrothermal OR mould ) AND TITLE-ABS-KEY ( TITLE-ABS-KEY ( building OR hous* OR home compar* ) AND DOCTYPE ( ar OR re ) [36 hits] OR haus OR residential ) AND TITLE-ABS-KEY (
moisture OR hygrothermal OR mould OR CO2 Search string for SCOPUS (search date 29/01/2020): OR PM2.5 OR VOC ) ) AND DOCTYPE ( ar OR TITLE-ABS-KEY ( airtight* OR air-tight* ) AND TITLE- re ) [224 hits] ABS-KEY ( building OR hous* OR home OR residential OR haus ) AND TITLE-ABS-KEY ( "indoor air quality" OR "indoor environment quality" ) AND Search string for Web of Science (search date DOCTYPE ( ar OR re ) [163 hits] 10/02/2020): TOPIC: (airtight* OR air-tight*) AND TOPIC: (building
OR hous* OR home OR haus OR residential) AND Search string for SCOPUS (search date 07/02/2020): TOPIC: (moisture OR hygrothermal OR mould OR TITLE-ABS-KEY ( airtight* OR air-tight* ) AND TITLE- CO2 OR PM2.5 OR VOC) Refined by: DOCUMENT ABS-KEY ( building OR hous* OR home OR haus TYPES: ( ARTICLE OR REVIEW ) Timespan: All OR residential ) AND TITLE-ABS-KEY ( "CO2" OR years. Indexes: SCI-EXPANDED, SSCI, A&HCI, "PM2.5" OR VOC) ) AND DOCTYPE ( ar OR re ) [94hits] CPCI-S, CPCI-SSH, ESCI, CCR-EXPANDED, IC. [174 hits]
Rapid Review - Airtightness 9
Figure 3. PRISMA flow diagram, providing an overview of the search and screening
Meta-bias(es) Outcomes and prioritisation Not applicable, due to the rapid and scoping nature of Not applicable, due to the rapid and scoping nature of this review. this review. Confidence in cumulative evidence Risk of bias of individual studies A discussion of the issues and confidence in the We did not perform a full quality assessment of the cumulative evidence is included in the results section. included primary studies. However, we coded the key study characteristics that could indicate risk of bias, such as study funding and declared conflict of interest.
Deviations from protocol After reviewing the studies available, it was decided to only include studies that established specific building airtightness using methods such as blower door tests. Many studies included infiltration rates or air exchange rates which vary with the environmental conditions, and therefore are difficult to compare to each other.
Data synthesis No quantitative assessment was performed. A qualitative summary is provided in the form of tables and narrative description of the patterns and gaps observed in the primary literature on the rapid review topic.
Rapid Review - Airtightness 10
Data items We recorded the study characteristics as shown in Table 1 below.
Table 1. List of the main study variables that were extracted and coded for the included studies, with relevant values
Study variable Description
First Author_year Key (ID) of the article is created by concatenating the last name of the first author and the year published
Year Year published
Exclude Y/N/?
Exclusion reason Statement of reason for exclusion
Main topic Main topic addressed in the article
Study aim Aim of study
Study method Field data, simulation, etc.
Study location Country in which compared buildings are located
Study Climate Climate conditions of study
Sample size Number of cases included in the study sample
Airtightness Method used to measure airtightness value used measurement method
Airtightness results Values of range of airtightness measurements
IAQ measurements E.g. moisture, mould, CO2, VOC, PM2.5 etc
IAQ monitoring IAQ parameter measurement duration and frequency frequency
Results for IAQ Specific results for each IAQ parameter measurements
Study funding Funding sources declared in the article
Conflict of interests Conflicts of interests declared in the article
Findings Findings and conclusions of study
Recommendations Highlighted recommendations relating to air tightness and ventilation
Reference Full publication reference information, including title of the article
Rapid Review - Airtightness 11
Airtightness of dwellings has been determined by a blower door test in the majority of the included studies Results (17). A small number of studies use a tracer gas technique to estimate the airtightness either as the Overview of the included studies primary method (2) or as a secondary method (4). The range of values reported for the airtightness of the In total, 21 studies were included in this study. These properties in each included study is given in Figure 4. studies were published between 1998 and 2019, and Most of the studies cover low to moderate air change covered 10 countries, with most studies coming from the rates (2-10 ACH ). United Kingdom (7) or North America (6). Table 2 50 provides an overview of these studies, including A variety of indoor air quality parameters were covered publication details, main topic, location, funding and by the included studies. These are listed below with the conflict of interests. Further details including the sample number of studies associated with each. The results of size, building construction considered, airtightness each of these is explored further for each individual measurement method and indoor air quality parameters parameter in the qualitative assessment. measured are given in Table 3. CO2 – 9 studies The studies covered a mixture of existing buildings (6), retrofitted buildings (6) and new builds (5). Two were PM2.5 – 12 studies simulation only studies. Two studies considered a Formaldehyde – 9 studies mixture of new builds and either existing buildings or retrofitted buildings. The building type considered in the VOC – 9 studies studies also varied, with detached houses (10), NO2 – 7 studies apartments (5) and semi-detached (2). Four (4) of the studies cover multiple building types in a single study. Moisture or mould – 9 studies The studies also cover a variety of climates (CC&ID, 2019), with the largest number in temperate locations Other (including CO, radon) – 7 studies (Koppen climate Cfb – 9 studies). Shorter term climatic characteristics were not reviewed in this study.
Broderick et al., 2017 Choi and Kang, 2017 Chu et al., 2018 Derbez et al., 2014 Dimitroulopoulou et al., 2005 Doll et al., 2016 Fernández-Agüera et al., 2019a Fernández-Agüera et al., 2019b Francisco et al., 2017 Hall et al., 2013* Hashemi and Khatami, 2015 Langer and Bekö, 2013
Lawton, 1998 Less et al., 2015 McKay et al., 2010 Milner et al., 2015 Offermann, 2009 Ridley et al., 2006 Shrestha et al., 2019 Szirtesi et al., 2018 Wang et al., 2016
0 10 20 30 40 50 *Note: Hall et al., (2013) is based on a Air Change Rate (ACH50) single room only
Figure 4. Air change rates of properties covered in each included study
Table 2. Main characteristics of the included studies
Reference Title Main topic Study Climate Koppen Study funding Conflict of location climate interest class Broderick et al., A pre and post evaluation of indoor air Assessment of the indoor environment Dublin, Temperate Cfb Irish Environmental Not declared 2017 quality, ventilation, and thermal comfort conditions of 15 semi-detached dwellings Ireland Protection Agency in retrofitted co-operative social housing before and after energy efficient upgrades.
Choi and Kang, Infiltration of ambient PM2.5 through Impact of building airtightness on PM2.5 Seoul, Cold Dwa National Research Not declared 2017 building envelope in apartment housing infiltration in Korean apartments Korea Foundation of Korea (NRF), units in Korea Ministry of Science, ICT & Future Planning Chu et al., 2017 Retrofitting existing buildings to control Understanding retrofit impacts on Shanghai, Subtropical Cfa No funding received for the None indoor PM2.5 concentration on smog controlling indoor PM2.5 concentrations, China research days: Initial experience of residential including improving air tightness, positive buildings in China pressure control (mechanical ventilation) and air purifiers. Derbez et al., Indoor air quality and comfort in seven Evaluated the IAQ and occupant comfort France Temperate Cfb French Observatory of Not declared 2014 newly built, energy-efficient houses in in seven low-energy, newly built houses in Indoor Air Quality (OQAI in France France. French) Dimitroulopoulou Ventilation, air tightness and indoor air Investigates airtightness, ventilation and England, Temperate Cfb Building Regulations Not declared et al., 2005 quality in new homes air quality in a sample of 37 recently built UK Division of ODPM homes to assess the effectiveness of the UK design standards. Doll et al., 2016 Weatherization impacts and baseline Looked at impacts on IEQ of North Humid Cfa / U.S. Department of Not declared indoor environmental quality in low weatherization upgraded in 69 homes. Carolina, subtropical / Cfb Housing and Urban income single-family homes USA Cool humid Development Fernández-Agüera Thermal comfort and indoor air quality in Impact of building airtightness on CO2 Madrid and Mediterranean Csa Spanish Ministry of None et al., 2019a low-income housing in Spain: The concentrations Seville, and Economy and influence of airtightness and occupant Spain Continental Competitiveness behaviour
Fernández-Agüera TVOCs and PM2.5 in naturally ventilated Investigate indoor air quality in 3 Seville, Mediterranean Csa No funding received for the None et al., 2019b homes: Three case studies in a mild apartments; focus on effect of ventilation Spain and research climate strategies (e.g. window opening) Continental Francisco et al., Ventilation, indoor air quality, and health Compares indoor air quality and health Illinois and Humid Dfa, Cfa U.S. Department of Not declared 2017 in homes undergoing weatherization impacts in houses pre and post Indiana, continental, Housing & Urban weatherisation for houses with natural USA humid Development ventilation and mechanical ventilation subtropical installed.
Reference Title Main topic Study Climate Koppen Study funding Conflict of location climate interest class Hall et al., 2013 Analysis of UK domestic building retrofit Retrofit refurbishment options for a test Nottingham, Temperate Cfb Engineering and Physical Not declared scenarios based on the E.ON Retrofit house, including changes to the UK Sciences Research Council Research House using energetic airtightness, were simulated using an (EPSRC) and E.ON (UK) hygrothermics simulation - Energy energetic hygrothermics approach to Ltd. efficiency, indoor air quality, occupant determine the relative effect on indoor air comfort, and mould growth potential psychrometric conditions and mould growth potential.
Hashemi and The effects of air permeability, Simulated study in EnergyPlus of CO2 and London Temperate Cfb Not declared None Khatami, 2015 background ventilation and lifestyle on condensation risk for differing air tightness Gatwick energy performance, indoor air quality rates and risk of condensation in domestic buildings Langer and Bekö, Indoor air quality in the Swedish housing Investigates indoor air quality Sweden Temperate Cfb Swedish Research Council Not 2013 stock and its dependence on building concentrations for a range of single family Formas declared` characteristics houses and apartments and compares regressions against building characteristics. Lawton, 1998 The influence of house characteristics in Comparison of characteristics of houses Canada - Cold Dfb Canada Mortgage and Not declared a Canadian community on based on the levels of biological growth Ontario continental Housing Corporation and microbiological contamination measured. Health Canada Less et al., 2015 Indoor air quality in 24 California Indoor air quality sampling performed in 24 California Mediterranean Csa/ California Energy Not declared residences designed as high- homes in California - compares 12 deep Csb Commission performance homes energy retrofit homes to 12 new homes McKay et al., 2010 Ventilation and Indoor Air Quality in Part Investigates dwelling compliance with Oxford, UK Temperate Cfb The Department for Not declared F 2006 Homes Approved Document Part F (ADF) 2006 Communities and Local standards, and evaluates if this standard is Government effective at providing adequate ventilation and good indoor air quality in new dwellings, thereby minimising the risks to health of the occupants. Milner et al., 2015 What should the ventilation objectives Simulation study to investigate impacts of London, Temperate Cfb European Union Seventh None be for retrofit energy efficiency low permeability on indoor air pollutants in England Framework Programme interventions of dwellings? UK Offermann, 2009 Ventilation and Indoor Air Quality in New A multi-season study of ventilation and California Mediterranean Csa/ California Energy Not declared Homes indoor air quality of 108 new single family Csb Commission detached homes in California. Ridley et al., 2006 Sensitivity of Humidity and Mould Combination of literature review, England, Temperate Cfb The Department for Not declared Growth to Occupier Behaviour in interpretation of available data sets and UK Communities and the Local Dwellings Designed to the New Air theoretical modelling to determine Government (DCLG) Tightness Requirements likelihood of problems related to mould growth in houses with low air leakage rates.
Rapid Review - Airtightness 14
Reference Title Main topic Study Climate Koppen Study funding Conflict of location climate interest class Shrestha et al., Impact of low-income home energy- Assessing the relationship between energy Colorado, Semi-arid/Dry Bsk Environment Protection None 2019 efficiency retrofits on building air efficiency retrofits specific to air sealing USA agency tightness and healthy home indicators and qualitative indicators of healthy homes Szirtesi et al., Airborne Particulate Matter: An Investigates particle concentration within 3 Hungary Continental Dfb European Union, European Not declared 2018 Investigation of Buildings with Passive buildings - 2 passive and 1 conventional Social Fund, and the House Technology in Hungary National Research, Development and Innovation Office - NKFIH Wang et al., 2016 Indoor-outdoor relationships of PM2.5 in Compares PM2.5 concentrations and Nanjing and Warm Cfa National Natural Science None four residential dwellings in winter in the indoor/outdoor concentration ratios for 4 Huzhou, Temperate Foundation of China and Yangtze River Delta, China dwellings with normal and high air China China Postdoctoral Science tightness. Foundation
Rapid Review - Airtightness 15
Table 3. Sample sizes, data measurements and findings of included studies
Reference Sample size Building type Construction Ventilation Airtightness Airtightness IAQ Overall study findings method results parameters (ACH50)
Broderick et al., 15 Semi- Retrofit Natural Blower door Pre retrofit - CO, CO2, Following the retrofit, there was a significant increase in 2017 detached ventilation with post retrofit, 10.6 to 11.5, PM2.5, TVOC, the concentration of a number of pollutants including background wall blower door Post retrofit - formaldehyde, PM2.5, formaldehyde and VOCs which may be vents or estimated 6.4 and 9.1 NO2, and attributable to reduced ventilation rates as a result of for pre BTEX improved building air tightness. There was a statistically retrofit significant increase in CO2 concentration in both building types post retrofit.
Choi and Kang, 11 Apartments Existing Natural Blower door 1.44 to 12.39 PM2.5 Building airtightness has a significant impact on reducing 2017 houses ventilation the infiltration rate of PM2.5 in locations where ambient PM2.5 is high.
Chu et al., 2017 1 Apartments Retrofit Natural Blower door Pre retrofit - PM2.5 Increasing the airtightness of buildings resulted in a ventilation for 9.5, post decrease in PM2.5 infiltration rates, although airtightness some and retrofit- 6.4 improvements along were not able to maintain healthy mechanical PM2.5 levels under all circumstances with high outdoor ventilation PM2.5 levels. A combination of airtightness with mechanical ventilation with filter and air purifier was required to maintain healthy air levels.
Derbez et al., 7 Detached New build Mechanical Blower door 0.29 - 3.33 TVOC, CO2, When compared to the IAQ of standard French houses 2014 ventilation with VOC, which have a lower airtightness and no mechanical heat recovery aldehydes, ventilation, the median concentrations of PM2.5, radon, (MVHR) CO, PM2.5, benzene and toluene were lower in the houses studied. radon The concentrations of formaldehyde were found to be similar, and some VOC concentrations were found to be higher. This study also noted that MVHR systems seem to be difficult to use, and high noise levels can be produced at the highest fan speed.
Dimitroulopoulou 37 Detached New build Natural Blower door 4.8 - 20.2 NO2, CO, The present study shows that ventilation rates can be et al., 2005 houses, ventilation, with (average PM10, HCHO, below design values and guidelines for air quality can be semi- extract fans or 12.9) total VOCs, exceeded in some homes. The main statistical finding detached passive stack and relative was that lower ventilation rates were associated with houses, ventilators in humidity higher TVOC concentrations. apartments bathrooms and kitchens
Rapid Review - Airtightness 16
Reference Sample size Building type Construction Ventilation Airtightness Airtightness IAQ Overall study findings method results parameters (ACH50)
Doll et al., 2016 69 Detached Retrofit Natural Blower door 3 – 32 CO, NO2, Findings from this study identify several trends for ventilation. Most (mean of 17 formaldehyde, changes in IAQ conditions as a result of weatherization, houses had pre-retrofit PM2.5, radon which may adversely affect occupant health and warrant forced air and 12 post- further investigation. Weak to moderate negative heating/ cooling. retrofit). correlations with 2-tailed significance were found Average between lower air change rates and higher CO2, PM2.5 change pre to and radon concentrations. However, the high levels of post- retrofit of non-compliance with IAQ guidelines for both pre and 4.8. post weatherization indicate there is a need for IAQ improvement independent of weatherization impacts. Findings also showed a relationship between improved air quality and higher mechanical ventilation rates which raises interesting questions about the role that each type of ventilation plays in maintaining good IAQ.
Fernández- 6 Apartments Existing Natural Blower door 3.2 to 9.4 CO2, relative An indirect correlation was observed between Agüera et al., houses ventilation humidity airtightness and CO2 concentration with less permeable 2019a envelopes experiencing higher indoor CO2 concentrations, particularly in winter. The correlations observed between airtightness and relative humidity also showed that the humidity generated by occupants was dissipated more effectively in flats with more permeable envelopes. Fernández- 3 Apartments Existing Natural Blower door 9.5, 14, 15.5 Relative There is a clear relationship observed between infiltration Agüera et al., houses ventilation humidity, CO2, and CO2 concentration values. However, there is no 2019b VOC and PM2.5 clear discernible pattern with either PM2.5 or TVOC concentrations. Both of these factors are influenced by dwelling occupancy and behavioural patterns, and the small sample size does not allow for distinct relationships to be seen.
Francisco et al., 81 houses Detached Retrofit 42 had Blower door 3.2 - 39.3 CO2, This study showed that following weatherisation, 2017 mechanical (mean of 18.8 formaldehyde, properties with mechanical ventilation installed had ventilation and 17.5 for TVOC, relative airflows twice as great as those with natural ventilation, installed; 39 had natural and humidity, and as a result had significantly lower indoor moisture natural mechanical radon balance, as well as improved air quality. ventilation ventilation respectively)
Rapid Review - Airtightness 17
Reference Sample size Building type Construction Ventilation Airtightness Airtightness IAQ Overall study findings method results parameters (ACH50)
Hall et al., 2013 1 Semi- Retrofit Natural Blower door 0.1-0.74 Mould growth This study has highlighted that in cases of low detached ventilation in test, or (based on a risk potential airtightness, or use of mechanical ventilation, indoor base case, assumed single room) modelled using relative humidity fluctuations are dominated by outdoor some with WUFI psychrometric conditions plus any additional effects from mechanical heating the incoming air. When the air change rate is ventilation with very low (<0.4), internal relative humidity fluctuation is heat recovery dominated by the internal loads, e.g. occupancy. (MVHR) Conditions with low air change rates and no mechanical ventilation are susceptible to high mould growth potential.
Hashemi and 1 (30 Bedroom in Simulated Natural Simulated 1.5, 4.5, 7.5, CO2; risk of The results of this study show that increasing Khatami, 2015 scenarios) detached ventilation with 10.5 and 15 condensation airtightness may escalate the risk of poor IAQ by house trickle vents on window reducing the rate of ventilation, resulting in unhealthy (hours) CO2 levels; although, this relation was not as evident for the risks of condensation. The latter requires more investigation. Langer and 157 houses Detached Existing 80% of Tracer gas - 1.5 to 50 Nitrogen Indoor air quality and air exchange rate was monitored in Bekö, 2013 and 148 houses and houses apartments and perfluorocar dioxide (NO2), 157 single-family houses and 148 apartments across the apartments apartments less than 50% of bon formaldehyde, Swedish housing stock. The significant positive detached VOC, relative correlation between the concentrations of NO2 and air houses have humidity exchange rates suggest that the major source of NO2 is mechanical outdoors. An opposite tendency was found for ventilation formaldehyde and TVOC, indicating that ventilation decreases the concentrations of formaldehyde and TVOC originating indoors. Lawton, 1998 59 Detached Existing Natural Tracer gas - Average of 19 Mould growth, The houses with high biological contamination levels houses ventilation, 3 SF6 for high relative had, on average, higher tested air leakage areas and houses with biological humidity, CO2, higher measured and predicted air change rates mechanical growth and VOCs indicating that natural ventilation and high leakage does ventilation 10.2 for low not control biological contamination. The expected biological inverse correlation was found between measured and predicted air change and concentrations of internally generated contaminants such as CO2 and VOCs (p<0.05) but the ranked measures of endotoxin, ergosterol dust mite antigen and colony-forming units on glycerol bases all showed a positive correlation with predicted air change (p<0.05 for endotoxins and CPUs p<0.10 for dust mite F).
Rapid Review - Airtightness 18
Reference Sample size Building type Construction Ventilation Airtightness Airtightness IAQ Overall study findings method results parameters (ACH50)
Less et al., 2015 24 Detached New build, Half with natural Blower door 0.4 to 10.3 CO2, PM2.5, This study compared the indoor air quality of high Retrofit ventilation, half (median 2.8) formaldehyde, performance homes. It was found that although the with mechanical NO2, relative mechanically vented homes were six times as airtight as ventilation humidity non-mechanically ventilated homes (medians of 1.1 and 6.1 ACH50 respectively), their use of mechanical ventilation systems and possibly window operation meant their median air exchange rates were almost the same (0.30 versus 0.32 hr-1). Pollutant levels including PM2.5, NO2 and formaldehyde were also similar in vented and unvented homes. Although specific correlations between these parameters and the airtightness of the homes was not provided, there was evidence that mechanical ventilation, particularly with filtration, improved the particle counts measured. McKay et al., 20 Apartments, New build Natural Blower door, 4.4 to 19.8 Relative Although there are limitations with the sample size, there 2010 terraced ventilation with tracer gas (tracer gas humidity, NO2, appears to be a correlation between air permeability and houses, trickle vents results as TVOC, pollutant levels, with high levels observed in the five semi- blower door formaldehyde most airtight buildings. However, 4 of these 5 building detached results were are flats. With these removed, there is still a correlation, and given in although it is less clear. The capacity of the ventilation detached permeability)4 systems in the houses measured did not meet the houses. recommended levels in UK regulations Approved Document F 2006, therefore it is difficult to make a conclusive judgement as to whether or not the current recommended natural ventilation provisions are sufficient for airtight homes.
Milner et al., 1 Detached Simulated Natural Simulated 1.2 -48 PM2.5, Radon This study has shown, based on a specific set of 2015 ventilation assumptions and average indoor pollutant values, that reducing permeability of dwellings to low levels without additional purpose-provided ventilation is likely to have net adverse effects on human health. The critical point of transition to dominant adverse effects for the assumptions used in this study is around 8.4 ACH50.
4 Note: Tracer gas measurements were taken and a generalised rule-of-thumb was used to multiply these values by 20 to convert to ACH50 (Johnston and Stafford, 2017). Blower door test results were provided, but as they were given in permeability (m3/m2/hr) and the building types varied, it was not possible to convert these values to ACH50 without individual building volume ratios.
Rapid Review - Airtightness 19
Reference Sample size Building type Construction Ventilation Airtightness Airtightness IAQ Overall study findings method results parameters (ACH50)
Offermann, 2009 104 Detached New build Most properties Blower door 2.8 to 8.4, CO2, PM2.5, New homes in California are built relatively tight, such were natural test median 4.8 Formaldehyde, that outdoor air exchange rates through the building ventilation, 8 VOC, NO2, CO envelope can be very low. Most of these new homes had had MVHR and levels of formaldehyde and VOC exceeding exposure 14 had guidelines, although there was no discernible intermittent relationship between airtightness and contaminant ducted outdoor levels. air system. Ridley et al., 105 (Warm Detached Existing Natural Blower door 5 - 35 had Mould Analysis of the ‘Warm Front’ database suggests that 2006 Front) houses ventilation Simulated ventilation occurrence, occupant density and air tightness by themselves cannot rates of 0.2, simulated risk be seen as the only predictors of mould problems. An 0.4, 0.6, 0.8 of mould average ventilation rate of 0.5 -1 ACH or greater is and 1 were needed to avoid mould growth. simulated Shrestha et al., 216 Detached, Retrofit 11 of the 216 Blower door 2 - 40 Visual The results suggest that mould observation was 2019 some semi- houses had Average 11.9 observance of positively correlated with the infiltration rate, which is detached mechanical presence of related to the airtightness. This is counter-intuitive to the houses ventilation mould, conventional notion that tight homes can trap humidity systems (5%) dampness, indoors for longer periods of time. One possible condensation explanation for leakier homes with observations of mould growth could be that the leakier homes had a greater probability of outdoor mould spores infiltrating into the indoor environment.
Szirtesi et al., 3 houses, 2 Detached Existing and Two houses Blower door 0.89, 1.6 and PM2.5 and PM8 This study highlighted several factors that bear influence 2018 passive new have mechanical 7.13 on the indoor air quality of passive houses, particularly design, 1 ventilation, one showing that the average PM concentration was higher conventional has natural in the two buildings with passive house technology than ventilation in the conventional one, with most of the excess PM concentration accounted for by coarse particles (particles with aerodynamic diameter > 2.5μm). These are thought to be from indoor sources that were trapped indoors by superior airtightness and also insufficient clearance.
Wang et al., 4 Apartments New build 2 properties had Blower door 0.3 and 3.8 PM2.5 - indoor This study investigated four apartments of two different 2016 natural and outdoor air tightness levels, each with or without an HVAC ventilation; 2 system under high ambient PM2.5 loads. Results showed properties had that enhanced air tightness of the building envelope centralised decreased the infiltration factor of ambient PM2.5 into the HVAC system home effectively. Although the use of a fresh makeup air conditioning system with filters for makeup air decreased this benefit. The use of high efficiency filters improved the infiltration rate of PM2.5 in the apartment with normal airtightness. They also showed that opening windows to increase ventilation in an area with high outdoor PM2.5 levels results in a significant increase in the indoor PM2.5 concentration that takes a long time to dissipate.
Rapid Review - Airtightness 20
internal PM2.5 concentration, but they also increased the effect of the internal source on internal concentrations Qualitative summary of study findings levels. Overall, the impact of decreasing air change rate The qualitative summary of the study findings has been resulted in a higher PM2.5 concentration. However, this is grouped according to the individual IAQ parameters that also dependent on the quality of the outdoor climate. were measured and assessed in the studies. Of the remaining 4 studies, only one showed a significant change in PM2.5 concentration following Carbon dioxide (CO2) increases in airtightness. This change was negative – i.e. increasing airtightness resulted in an increase in the Of the included studies, nine reported the CO2 levels monitored in the house. The specific results of these indoor PM2.5 concentration. The other 3 studies included studies are given in Table 5. did not show any statistical change in PM2.5. This is likely because of the variability in measurements as a result of Six out of the nine studies demonstrated a statistically indoor and outdoor sources, and the limited sample size significant negative impact on CO2, that is properties or timeframe used for sampling. with a low air change rate experienced higher CO2 levels. Each of the three studies which did not show an From the studies investigated, it can be assumed that in impact had short monitoring durations – from 24 hours environments with a high outdoor particulate level, (Offermann, 2009) to 1 week. All three of them also increasing airtightness can have a positive impact on the included properties in their sample group which had indoor environment by reducing infiltration of particles mechanical ventilation installed. One of these studies into the house. There is insufficient evidence to suggest (Francisco et al., 2017), did not show a link between that increasing airtightness in properties where outdoor environments have low particulate loads would have any airtightness and CO2 levels. However, it did demonstrate that installing mechanical ventilation in the houses detrimental impact on the indoor environment. correlated with a reduced CO2 concentration. Formaldehyde (HCHO) It is reasonable to conclude from the studies considered that there is a negative correlation between air change Of the included studies, nine monitored formaldehyde rate and the CO2 level experienced inside a house. levels in the house. The specific results of these studies are given in Table 7. Four of these showed evidence of a negative correlation between air change rate and Particulate levels (PM2.5) formaldehyde concentration. One of these (Offermann, Of the included studies, twelve monitored particulate 2009) did not display a direct correlation, although the levels in the house. The specific results of these studies high levels of formaldehyde measured in relatively are given in Table 6. One of these studies (Doll et al., airtight houses indicated a negative association. The 2016) investigated a range of particulate levels including other five showed no statistical impact or no distinct 0.3, 0.5, 1.0, 2.5, 5.0 and 10.0. One study correlation between the concentration of formaldehyde (Dimitroulopoulou et al., 2005) only looked at PM10. and airtightness.
Ratings of the location outdoor PM2.5 level have been Formaldehyde is often associated with new furnishings obtained from a 2019 Air Quality Report (IQAir, 2019). and building materials. Of the included studies, four Seven of the studies included were in areas with specifically looked at new build houses, three looked a moderate outdoor PM2.5 concentrations. The focus of retrofit houses and one looked at existing houses. The three of these studies was on reducing the PM2.5 remaining study involved a comparison between retrofit concentration indoors. The variable used for this and new build. Of the three studies showing a negative comparison was the infiltration factors or I/O correlation, there was one each considering new build, concentration ratios. All of these studies showed that retrofit and existing houses. As the majority of studies increasing airtightness had a positive impact, that is, involve new builds or retrofits, it is likely that any they resulted in decreased PM2.5 I/O ratios. However, differences in formaldehyde levels could be attributed to they highlighted the need for filtration in order to achieve new materials. However, for one of the studies showing healthy levels of IAQ. Of the remaining four studies from a significant negative correlation (Broderick et al., 2017) moderate air quality areas, one also showed a positive it was recorded that materials used did not contain impact on PM2.5 levels (Derbez et al., 2014), which formaldehyde. related to the use of mechanical ventilation. Two of the remaining studies did not provide enough information to There is insufficient evidence from the studies correlate against airtightness and the last one showed considered to conclude that there is a definitive no observed impact. relationship between airtightness and formaldehyde levels, although there is a potential for elevated risk of Five studies were located in areas which were identified higher formaldehyde levels with increased airtightness. as having good or lower than a World Health Therefore, the mitigation of contributing factors such as Organisation (WHO) target range for outdoor PM2.5 furnishings or formaldehyde containing materials should levels. One of these studies, Milner et al. (2015) involved be considered in more airtight dwellings. a simulation of airtightness levels and associated PM2.5 levels from both a high outdoor source and an internal Volatile organic compounds (VOC) source. They found that lower air change rates decreased the effect of the outdoor source on the Of the included studies, nine monitored volatile organic compound (VOC) levels in the houses. The specific
results of these studies are given in Table 8. Generally, vapour through the building envelope, which can reduce the VOC measurements are expressed in terms of the the risk of condensation occurring. Building regulations total VOC, or TVOC. However, one study (Offermann, for modern buildings in the UK, Europe and the USA 2009) presented results for individual contaminants. have required the use of vapour permeable membranes to manage moisture control since the 1990’s/early Five of the studies included showed evidence of a 2000’s. Therefore, any of the studies involving new build negative correlation between TVOC concentration and houses are likely to have been constructed using vapour air change rate. Of the other four studies, two did not permeable membranes. show any correlation with airtightness, one was indeterminate but showed a suggested negative Of the included studies, nine monitored variables related correlation (Offermann, 2009), and the other shows that to mould or moisture issues within the house. The more airtight dwellings had higher VOC concentrations, specific results of these studies are given in Table 10. but it was difficult to draw a distinct correlation due to Three of these studies looked specifically at observance other impacting factors (Fernández-Agüera et al., of mould issues, two looked at simulations of 2019b). condensation or mould risk and four measured the relative humidity to understand the impact on moisture In a number of the studies, houses with low air change within the house. rates were found to have VOC concentrations that were above the guidelines for safe exposure. In the studies that involved direct mould observations, two of the three included showed a positive correlation From the studies investigated, it can be assumed that between reduced air change rate and mould occurrence. there is a relationship between airtightness and VOC Although this is recognised as being counter-intuitive to concentration and that there is a high likelihood that low the notion that airtight homes trap humidity indoors air change rates can lead to unsafe levels of VOC (Shrestha et al., 2019) it has been suggested that this concentrations. could be explained by leakier homes also having a higher rate of water leakage (Lawton, 1998). This could Nitrogen dioxide (NO2) also be a result of limitations in the study or confounding Of the included studies, seven monitored the nitrogen factors that were not measured. All of the studies included considered properties that had mostly higher air dioxide (NO2) levels in the house. The specific results of these studies are given in Table 9. change rates, and as they involved existing houses. These houses were likely not constructed using vapour Ratings of the location outdoor NO2 level have been permeable membranes, although the construction obtained from data sourced for 2018 (EEA 2018, EPA materials are not explicitly specified. Both the studies 2018). Three of the studies were located in areas with a with a positive correlation involved a mixture of natural moderate rating of NO2, and four of the studies were and mechanical ventilation. Mechanical ventilation is located in areas with a good rating. Of the four studies in likely to be installed in properties with higher locations with good air quality rating, none of them airtightness, so this may confound the results. demonstrated any impact of airtightness on NO2 concentrations. In the studies that measured relative humidity, two of the four included showed a negative correlation, with Of the three studies located in areas with moderate increased relative humidity experienced at low outdoor NO2 levels, two of them demonstrated a positive airtightness levels, and the other two did not show any relationship between air change rate and NO2 impact, although one study had one house with a very concentrations. This is likely to be caused by reducing low air change rate and high humidity levels. the infiltration rate of NO2 from outside. Two studies involving simulations of mould or Most of the dwellings sampled had levels that were condensation risk potential both showed that there was a below the recommended thresholds. The one study that high risk of condensation or mould for low air change had levels above the guidelines (McKay et al., 2010) rates. One of the studies (Hashemi and Khatami, 2015) was linked to insufficient or underutilised extraction in showed that for high levels of internal moisture load, the kitchen. there was a high risk of mould even for higher air change rates. The other (Hall et al., 2013) showed how There does not appear to be any correlation between mechanical ventilation could reduce the risk of airtightness values and NO concentrations in areas 2 condensation occurrence. where there are low outdoor levels of NO2. However, it could be that in areas with high outdoor levels of NO2, Overall, it is difficult to draw any direct correlation particularly urban areas, increasing airtightness could between increased airtightness and mould occurrence. lead to lower NO2 concentrations indoors. Whilst simulations would suggest that increased airtightness would increase the risk of mould growth, and Mould / moisture measurements of relative humidity support this notion, studies of actual occurrences of mould tend to show the Mould issues within dwellings are strongly linked to opposite. This is confounded by differences in the air moisture and condensation issues. Water vapour levels change rate of the properties involved in the studies. within dwellings, which can cause condensation issues, Studies of actual mould occurrence involved houses with are affected not only by the airtightness of the building, higher air change rates than those involving simulations. but also the permeability of the building fabric. Vapour This may indicate that the impact of increased permeable membranes allow for the transfer of water
Rapid Review - Airtightness 22
airtightness on mould growth is not seen until the air Table 4: Summary of impact of increased airtightness on change rate is below a certain threshold. It also IAQ parameters based on the 21 reviewed studies. highlights the complexities and importance of other factors influencing the growth of mould. Other Increased Increased No clear There were seven studies which included additional airtightness airtightness relationship indoor air quality measurements. The specific results of resulted in resulted in between these studies are given in Table 11. The additional worsened IAQ improved IAQ airtightness and parameters included: IAQ.
- Carbon monoxide (CO)- three studies - Radon - five studies IAQ Parameter Impact For the studies investigating CO, there were no clear relationships found between airtightness and CO. All measurements taken were found to be in acceptable CO2 ranges.
Five studies investigated the influence of airtightness on PM2.5 – in areas with high outdoor radon concentrations. Although radon is not a major levels concern for the Australia context (ARPANSA, 2020), results are provided for completeness. One of these PM2.5 – in areas with average found no discernible correlation between airtightness outdoor pollutant levels and radon. Two of the studies identified a negative correlation between low air change rate and radon levels. The final study indicated a positive correlation Formaldehyde with houses with low airtightness displaying lower radon * concentrations, although the properties involved in this study also had mechanical ventilation which may have contributed to this result. VOC *
Summary NO2 – in areas with high outdoor In summary, the studies reviewed showed evidence that levels increasing airtightness has a negative correlation with CO2 and VOC concentration, and a positive impact on NO – in average areas PM2.5 and NO2 concentrations in areas where there are 2 high outdoor levels. For all other IAQ parameters investigated, there was insufficient evidence provided from the studies reviewed to identify any direct Mould/moisture correlations. Table 4 provides a visual summary of the impact of increasing airtightness on each of the IAQ parameters based on the studies reviewed. Radon
CO * Reviewed studies in Formaldehyde and VOC either identified no clear relationship, or a negative relationship. More studies are needed to better understand this relationship.
Rapid Review - Airtightness 23
Table 5: Summary of results for CO2
Reference Location Climate Sample Ventilation Airtightness Monitoring frequency Results Finding summary Impact size (ACH50) and duration
Hashemi and London, UK Temperate 1 (30 Natural 1.5, 4.5, 7.5, Simulated for 6 For air permeability of 1.5 ACH50, CO2 Lower air change Negative Khatami, 2015 scenarios) ventilation with 10.5 and 15 months (October to concentration levels were above 1000 rates were found to trickle vents March). Frequency of ppm, 1500 ppm and 5000 ppm for 85%, result in higher CO2 the times when CO2 69% and 10% of all occupied periods, concentrations. levels were higher respectively. For 4.5 ACH50 the results Unhealthy IAQ levels than 1000, 1500 and indicated considerable reduction to around were found to occur 5000 parts per million 31% during which CO2 was above 1500 when air change rate (ppm) are reported. ppm. However, CO2 remained above 1000 was less than 4.5 ppm in 54% of the occupied periods. IAQ ACH50. continued to improve for all other air permeability values.
Fernández-Agüera Seville, Mediterranean 3 Natural 9.5, 14, 15.5 Measured at 10 min Typical value (median) for daytime living There is a clear Negative et al., 2019b Spain and continental ventilation intervals over 12 rooms ranged between 853 ppm (CS1) and relationship between months in living room 882 ppm (CS2), while for bedrooms this infiltration and CO2 and bedroom varied between 1199 (CS2) and 2385 ppm concentration values, (CS1). stronger in bedrooms at night. The least airtight dwelling (CS3) remains below 1000 ppm more often (for a similar emission source of two people during sleep time) than in the two more airtight dwellings.
Fernández-Agüera Madrid and Mediterranean 6 Natural 3.2 to 9.4 Measured at 10-min Mean CO2 concentrations in the bedroom An indirect correlation Negative et al., 2019a Seville, and continental ventilation intervals for a full year in winter varied between 452 ppm and was observed Spain in bedroom and living 2848 ppm, and in the living room, varied between airtightness room between 102 ppm and 2076 ppm. Mean and CO2 concentrations in summer were less concentration, with variable, with an average value of 454 ppm higher airtightness for living room and 183 ppm for bedrooms. values resulting in The highest correlation between higher indoor CO2 airtightness and CO2 concentration was concentrations. found in living rooms in winter (0.56) when windows are kept closed.
Broderick et al., Dublin, Temperate 15 Natural Pre retrofit - Measured in living Post retrofit CO2 concentrations (24 h There was a Negative 2017 Ireland ventilation with 10.6 to 11.5, room/bedroom every average values) increased from 578.57 statistically significant background Post retrofit - 1 min for 24 h ppm and 571.37 ppm to 747.71 ppm and change in CO2 wall vents 6.4 and 9.1 676.5 respectively (p=0.014 and p=0.003) concentration in both building types post retrofit.
Lawton, 1998 Ontario, Cold continental 59 Natural Average of 19 Continuously The average concentrations of CO2 at night An inverse correlation Negative Canada ventilation, 3 for high monitored in a were higher than in the day by about 45% was found between houses with biological bedroom for 1 week (984 vs. 680 ppm). 40% of the houses had measured and mechanical growth and night period average concentrations over predicted air change ventilation 10.2 for low l000 ppm, with four over 1500 ppm and and concentrations of biological one over 3000 ppm. CO2 measurements CO2 (p<0.05). showed an inverse correlation to the measured air change rate.
Reference Location Climate Sample Ventilation Airtightness Monitoring frequency Results Finding summary Impact size (ACH50) and duration
Doll et al., 2016 North Humid 69 Natural 3 – 32 Measured every 5 min The rate of compliance to CO2 average There are indications Negative Carolina, subtropical / ventilation, with (mean of 17 for 2-3 day period guidelines (<1000 ppm) increased from of a moderate USA Cool humid extract fans pre-retrofit 84% pre-weatherization to 93% post- negative correlation bathrooms. and 12 post- weatherization, although the comparison of between CO2 and Most houses retrofit). pre to post values showed no statistically airtightness. had forced air Average significant difference following the heating/cooling. change of 4.8. weatherization. Overall a moderate negative correlation with 2-tailed significance (p=0.065) was found for lower CO2 with higher ACH50.
Derbez et al., 2014 France Temperate 7 MVHR 0.29 - 3.33 Measured every The median CO2 levels were always below This study did not No clear 10 min during each 1000 ppm and ranged between 351 and show any correlation relationship week of investigation 811 ppm. When compared to the national between CO2 levels after inhabitants median levels, half of the measured and airtightness. moved in houses were below the national median, and half were above it. There was no correlation seen between airtightness levels and CO2 concentrations.
Francisco et al., Illinois and Humid 81 42 had 3.2-39.3 Measured by passive The levels of carbon dioxide pre and post This study did not No clear 2017 Indiana, continental, houses mechanical (mean of 18.8 sampling over 1 week weatherisation for properties with natural show any correlation relationship USA humid ventilation and 17.5 for ventilation showed no significant difference between airtightness
subtropical installed, 39 natural as a result of increased airtightness. and CO2 levels, had natural ventilation and However, in the group with mechanical although it has been ventilation mechanical ventilation there was a statistically shown that ventilation significant decrease in the CO2 levels mechanical ventilation respectively) experienced, with the average level has a positive impact dropping from 936 ppm to 787 ppm on the CO2 levels (P<0.005). .
Offermann, 2009 California, Mediterranean 104 Most properties 2.8 to 8.4, Measured every 1 The median indoor concentration was 564 There was no No clear USA were natural median 4.8 minute over a 24 hour ppm, with a range of 334 ppm to 1,108 relationship observed relationship ventilation, 8 period ppm. These values are below the between airtightness
had MHRV and recommended guidelines. and CO2 levels. 14 had intermittent ducted outdoor air system.
Rapid Review - Airtightness 25
Table 6: Summary of results for PM2.5
Reference Location Outdoor Climate Sample Ventilation Airtightness Monitoring Results Finding summary Impact 1 PM2.5 size (ACH50) frequency and duration
Doll et al., 2016 North WHO Humid 69 Natural 3 – 32 Particle meter There was very poor compliance with the There was no evidence PM2.5 - no Carolina, Target subtropical / ventilation, with (mean of 17 counts (0.3, 0.5, hourly maximum value of PM2.5 and PM10 in of a relationship clear USA cool humid extract fans pre-retrofit 1.0, 2.5, 5.0 and all houses, both pre and post observed between relationship bathrooms. and 12 post- 10.0) logged weatherization, although there was higher airtightness and PM2.5 Most houses retrofit). every 5 min compliance with average limits for non- concentrations. had forced air Average over 2-3 days. smoking houses both pre and post. There However, there is Negative for heating/cooling change of Also, mass was no statistical difference between the indication of a low 0.3 µm . 4.8. concentration pre and post weatherization concentrations negative correlation every 5 min. measured. The percentage of post smoking between 0.3µm homes below reference values is lower for particles and small (0.3 µm) particles. A low negative airtightness. correlation with 2-tailed significance (p=0.1) was found for lower 0.3 µm particles with higher ACH50.
Fernández- Seville, WHO Mediterranean 3 Natural 9.5, 14, 15.5 Measured at 10 PM2.5 indoor concentrations present a wide A relationship between No clear Agüera et al., Spain Target and ventilation min intervals oscillation without a clear pattern. Annual PM2.5 concentrations relationship 2019b continental over 12 months average values are 16.09 µg/m3 (CS1), 7.1 and infiltration is not in living room µg/m3 (CS2), and 9.66 µg/m3 (CS3). The determinable from the and bedroom properties exceed the PM2.5 threshold of 10 data given. µg/m3 between 45% (CS1) and 15% (CS2) of the hours across the year. The most airtight building (CS1) has the highest average values, but this is not representative of the spread of the values and has likely been affected by the lack of mechanical extraction in the kitchen.
Broderick et al., Dublin, Good Temperate 15 Natural Pre retrofit - Measured in Concentrations of PM2.5 post retrofit (17.87 There was a significant Negative 3 3 2017 Ireland ventilation with 10.6 to 11.5 living µg/m and 15.53 µg/m ) increased (p=0.005 change in PM2.5 background Post retrofit room/bedroom and p 0.008, respectively), although the concentrations in both wall vents - average of every 1 min for values were still less than the WHO dwellings post retrofit. 6.4 and 9.1 24 h guideline of 25 µg/m3for indoor air.
Dimitroulopoulou England, Good Temperate 37 Natural 4.8-20.2 Particle In winter, PM10 concentrations ranged from There was no clear No clear 3 et al., 2005 UK ventilation, with (average sampling (PM10) 10.2 to 112.9 µg/m . The mean value (26.2) correlation seen relationship extract fans or 12.9) undertaken over was below the UK air quality strategy between air tightness 3 passive stack 24 hours objective (50 µg/m ). The two highest levels and PM10 ventilators in were recorded in two smokers’ homes. In concentrations bathrooms and summer concentrations ranged from 21.2 to kitchens 133.7 µg/m3. The UK air quality strategy objective (50 µg/m3) was exceeded in 14 out of 34 homes. There does not seem to be any evidence that PM10 levels are affected by ventilation parameters such as the use of trickle vents or extract fans, although correlation against air change rate is not directly considered.
1 Based on data from 2019 Air Quality report (IQAir, 2019)
Rapid Review - Airtightness 26
Reference Location Outdoor Climate Sample Ventilation Airtightness Monitoring Results Finding summary Impact 1 PM2.5 size (ACH50) frequency and duration
Milner et al., London, Good Temperate 1 Natural 1.2 – 48 Modelled based Result of the PM2.5 analysis showed that for Findings show that Negative 2015 England ventilation on outdoor both outdoor and indoor sources, reducing permeability of
source and permeability above 18 ACH50 showed no dwellings to low levels indoor sources. difference in the PM2.5 concentration. Below without additional this level, the PM2.5 from outdoor sources purpose-provided decreased with a curvilinear decrease, ventilation is likely to whilst the PM2.5 from indoor sources have net adverse increased with an approximately effects on PM2.5 levels. exponential decay form. The impact of trickle vents and exhaust fans compounded these results, having a positive impact on the indoor pollutant source results, but a negative impact on the outdoor pollutant source results.
Choi and Kang, Seoul, Moderate Cold 11 Natural 1.44 to Infiltration rate Infiltration factor (ratio of the indoor to Increasing building Positive 2017 Korea ventilation 12.39 of PM2.5 is outdoor steady state concentration) varied airtightness decreases measured using between 0.38 and 0.88. The infiltration the infiltration rate of a blower door factor was correlated to the airtightness PM2.5 in locations where 2 test with level, with an R value of 0.5243. ambient PM2.5 is high. pressure difference of 10 Pa to measure impact on steady state indoor PM2.5 levels.
Chu et al., 2018 Shanghai, Moderate Subtropical 1 Initial building Pre retrofit - Indoor and I/O ratio decreased post retrofit from 0.66 to Increasing the Positive, but China and first retrofit 9.5 post outdoor PM2.5 0.53. Indoor PM2.5 concentration could be airtightness of buildings filtration had natural retrofit- 6.4 concentration maintained at a healthy level below resulted in a decrease needed 3 ventilation. measured every 25 µg/m when outdoor PM2.5 concentration in PM2.5 infiltration rate. Mechanical 60 s over was at a low level, but not when the outdoor ventilation and multiple days. PM2.5 concentration was at a higher level. an air purifier I/O ratio used Including mechanical ventilation could were also as the further reduce the I/O ratio to 0.26 (low investigated. comparison volume) or 0.16 (high volume) and using an between indoor air purifier resulted in an I/O ratio of impacts for high around 0.33. A combination of increased air pollutant and tightness, mechanical ventilation and indoor low pollutant air purifier resulted in an I/O ratio of 0.05 days. which was low enough to control indoor PM2.5 concentration to healthy levels.
Derbez et al., France Moderate Temperate 7 MVHR 0.29 - 3.33 Measured from Weekly concentrations of PM2.5 varied When compared to Positive 3 2014 5 pm-8 pm on between 7 and 21 µg/m for summer and 13 national averages, the weekdays and and 28 µg/m3 for winter. These values were increased airtightness 24 hrs on lower than the national median and use of MVHR in weekend during concentration in standard French dwellings. these properties may each week of Higher concentrations in winter in some correlate to a lower investigation properties may be related to the use of PM2.5 concentration. wood for heating.
Rapid Review - Airtightness 27
Reference Location Outdoor Climate Sample Ventilation Airtightness Monitoring Results Finding summary Impact 1 PM2.5 size (ACH50) frequency and duration
Less et al., 2015 California, Moderate Mediterranean 24 Half with 0.4 to 10.3 1 min for 6 days An assessment of the impact of the There was no clear No clear USA natural (median 2.8) in kitchen - mechanical ventilation systems showed that correlation between relationship ventilation, half number of central recirculating system homes had particle count and with particles per m3 67% lower average particle count levels airtightness from the mechanical of air was than unfiltered homes, and filtered supply results, although ventilation measured, but ventilation homes had levels 38% lower mechanical ventilation not mass based than unfiltered homes. Examining the data with filtration is shown to therefore results provided showed a potential correlation have a significant were not between airtightness and particle count, impact on reducing discussed. which is markedly reduced in the cases of particle count. mechanical ventilation.
Offermann, 2009 California, Moderate Mediterranean 104 Most properties 2.8 to 8.4 Measured over For PM2.5, only one home, with an indoor There was no clear No clear USA were natural median 4.8 24 hour average concentration of 36 µg/m3, exceeded the relationship observed relationship ventilation, 8 period U.S. Environmental Protection Agency’s between airtightness
had MHRV and PM2.5 24‐hour ambient air quality standard and PM2.5 14 had ducted of 35 µg/m3 All other homes had acceptable concentrations. outdoor air levels. system.
Szirtesi et al., Hungary Moderate Continental 3 Two houses 0.89, 1.6 Sampled over In all three houses, measured PM2.5 There was no clear No clear 2018 (passive) have and 7.13 48 hours concentration levels were well below the correlation, but this relationship mechanical recommended guidelines. Average study would suggest ventilation, one concentrations were higher in the two that increasing has natural buildings with passive house technology airtightness may lead to ventilation than in the conventional one, with most a higher concentration excess PM concentration accounted for by of coarse particles, but coarse particles (>2.5μm) from indoor have no impact on fine sources that were trapped by superior particles. airtightness and also insufficient clearance. There was no difference in the concentration and composition of the fine fraction (PM2.5), these were not filtered by mechanical ventilation.
Wang et al., Nanjing Moderate Warm 4 2 - natural 0.3 - 3.8 Every 30 min for The mean indoor air concentrations ranged Enhanced air tightness Positive, but 2016 and temperate ventilation, 2 months in from 26 to 64.9 µg/m3, while outdoor of the building envelope filtration Huzhou, 2 - centralized winter concentrations ranged from 75.3 to decreased infiltration needed 3 China HVAC system 137.8 µg/m . For dwellings with normal factor of ambient PM2.5 airtightness, mean I/O results were 0.876 into the home, although and 0.463 (with high efficiency HVAC high efficiency filters filters). For high air tightness, results were were needed to achieve 0.197 and 0.308 (with HVAC with fresh air healthy PM2.5 levels. supply). Opening windows to increase ventilation in an area with high outdoor PM2.5 levels results in a significant increase in the indoor PM2.5 concentration that takes a long time to dissipate
Rapid Review - Airtightness 28
Table 7: Summary of results for formaldehyde
Reference Location Climate Build type Ventilation Airtightness Monitoring Results Finding summary Impact (ACH50) frequency and duration
Derbez et al., France Temperate New build MVHR 0.29 - 3.33 Passive Concentrations of formaldehyde varied This study does not show No clear 2014 sampling during between 13 and 41.4 μg/m3for winter and any correlation between relationship week of 12 and 39.7 μg/m3 for summer. airtightness and investigation Formaldehyde concentrations were formaldehyde levels. similar to those found in typical French homes.
Dimitroulopoulou England, Temperate New build Natural 4.8-20.2 Formaldehyde Three day mean results for HCHO were No evidence of a correlation No clear et al., 2005 UK ventilation, with (average 12.9) and aldehydes between 11 and 80 μg/m3, in winter and between formaldehyde and relationship extract fans or measured over 3 12-87 μg/m3 in summer, which are below airtightness observed. passive stack consecutive the WHO air quality guideline of ventilators in days, as well as 100 μg/m3. There were no statistical bathrooms and sampled over relationships observed between kitchens 30 min. ventilation parameters and formaldehyde concentrations.
McKay et al., Oxford, Temperate New build Natural 4.4 to 19.8 Measured over 3 Formaldehyde levels varied between 12 There is evidence of a Negative 2010 UK ventilation with consecutive and 63 μg/m3 (mean of 30 μg/m3). Based negative correlation trickle vents days. on assumptions of variations in ventilation between formaldehyde and rate, it would suggest that at most 3 airtightness for low air dwellings out of the 20 studied would change rates, although it is exceed the WHO criterion of 100 μg/m3 difficult to extrapolate the averaged over 30 min. These 3 dwellings results. all had low air permeability levels.
Offermann, 2009 California, Mediterranean New build Most properties 2.8 to 8.4 Measured over For formaldehyde, 98 percent of the Whilst there was no clear Negative USA were natural median 4.8 24 hour average homes exceeded the 2008 Chronic and 8‐ relationship observed ventilation, 8 had period hour Reference Exposure Levels for between airtightness and MHRV and 14 irritant effects of 9 µg/m3, 59% exceeded formaldehyde, the very high had intermittent the 2005 California Air Resources Board’s levels of formaldehyde ducted outdoor indoor air guideline for irritant effects of measured in homes that air system. 33 µg/m3 28% exceeded the 2008 ‘Acute were considered relatively Reference Exposure Levels’ for irritant airtight suggests that there effects of 55 µg/m3. The percentages of is a negative relationship. homes exceeding the ‘No Significant Risk’ levels concentration for formaldehyde and acetaldehyde were 100% and 93%, respectively. The primary source of the indoor concentrations of formaldehyde and acetaldehyde is believed to be composite wood products.
Rapid Review - Airtightness 29
Reference Location Climate Build type Ventilation Airtightness Monitoring Results Finding summary Impact (ACH50) frequency and duration
Less et al., 2015 California, Mediterranean New Half with natural 0.4 to 10.3 6 day passive Six-day formaldehyde concentrations There was no correlation No clear USA build, ventilation, half (median 2.8) samples ranged from 11.7-47 μg/m3, and 8.1-33.2 observed between relationship Retrofit with mechanical μg/m3 in the kitchen. This exceeded the formaldehyde and ventilation California EPA chronic reference airtightness from the results. exposure level (REL) of 9 μg/m3 in 23 of 24 homes, although not a single study home exceeded the California acute REL of 55 μg/m3. Although an assessment is not provided, examining the data does not reveal a correlation between airtightness and formaldehyde, and mechanical ventilation does not seem to have an impact on the results.
Broderick et al., Dublin, Temperate Retrofit Natural Pre retrofit: 10.6 to Measured in Formaldehyde concentrations increased There was a significant Negative 2017 Ireland ventilation with 11.5, living room every from average concentrations of 15.43 ppb change in formaldehyde background wall Post retrofit: 6.4 - 30 min for 24 before the retrofit to 24.27 ppb after the concentrations post retrofit. vents 9.1 hours energy retrofit (p=0.001). Post retrofit formaldehyde concentrations were below the WHO guideline of 80 ppb but were above the California EPA's chronic REL of 7.2 ppb. Two dwellings reported purchasing new furnishings for their dwellings after the retrofit, building materials used in this study did not contain formaldehyde.
Doll et al., 2016 North Humid Retrofit Natural 3 – 32 Measured every The rate of compliance to formaldehyde There was no evidence of a No clear Carolina, subtropical / ventilation, with (mean of 17 pre- 30 min over 2-3 average guidelines increased from 70% relationship observed relationship USA cool humid extract fans retrofit and 12 days pre-weatherization to 75% post- between airtightness and bathrooms. Most post- retrofit). weatherization, although the comparison formaldehyde houses had Average change of of pre to post values showed no concentrations. forced air 4.8. statistically significant difference following heating/cooling. the weatherization.
Francisco et al., Illinois Humid Retrofit 42 had 3.2-39.3 Measured by The levels of formaldehyde post There is insufficient data No clear 2017 and continental, mechanical (mean of 18.8 and passive weatherisation were seen to decrease provided to determine any relationship Indiana, humid ventilation 17.5 for natural sampling over 1 both for properties with natural ventilation correlation between USA subtropical installed, 39 had ventilation and week (from 31 ppb to 25 ppb, p<0.02) and for airtightness and natural mechanical properties with mechanical ventilation formaldehyde levels, ventilation ventilation (from 26 ppb to 21 ppb, p<0.05). although there was a minor respectively) decrease following retrofit (increasing airtightness).
Langer and Sweden Temperate Existing 80% of 1.5 to 50 Measured using The mean formaldehyde concentration This study suggests that Negative Bekö, 2013 houses apartments and passive was found to be 26 μg/m3 for single family lower air change rates are less than 50% of samplers for 24 houses and 14.3 μg/m3 for apartments. associated with higher detached hrs The formaldehyde concentration was formaldehyde houses have found to be statistically negatively concentrations. mechanical correlated with the air exchange rate ventilation (p<0.01).
Rapid Review - Airtightness 30
Table 8: Summary of results for VOC
Reference Location Climate Sample Ventilation Airtightness Monitoring Results Finding summary Impact size (ACH50) frequency and duration
Broderick et al., Dublin, Temperate 15 Natural Pre retrofit - Measured in living Concentrations of total volatile organic TVOC concentrations Negative 2017 Ireland ventilation with 10.6 to 11.5 room and bedroom compounds (TVOCs) increased post retrofit increased post retrofit, background wall Post retrofit - every 1 min for 24 from 365 to 482 ppb for CW and 375 to 464 with post retrofit values vents 6.4 and 9.1 hours ppb HB (p=0.009, 0.032 respectively). Post comparable to those retrofit concentrations in the current study reported for newly exceeded both the Finnish standard for occupied energy efficient excellent air quality (87 ppb, 8 h average) dwellings in France. and guideline values published by the UK Building Regulations.
Derbez et al., 2014 France Temperate 7 MVHR 0.29 - 3.33 Passive sampling Concentrations of TVOC measured varied This study does not show No clear during week of between 6 and 276 μg/m3 for winter. Due to any correlation between relationship investigation, and the different compounds measured, it was airtightness and TVOC TVOC every difficult to compare the VOC concentrations levels. 10 min to those found in typical French homes. Some measured values (benzene and toluene) were lower, whereas some were slightly higher and others exceeded typical values by more than 50%.
Dimitroulopoulou England, Temperate 37 Natural 4.8-20.2 Measured by TVOCs were measured between 48 and Lower ventilation rates Negative et al., 2005 UK ventilation, with (average 12.9) diffusive sampling 713 μg/m3 in winter and 36-444 μg/m3 in were associated with extract fans or over 2 weeks, and summer. Most were within guidelines but 3 higher TVOC passive stack directly over were above the target value of 300 μg/m3 concentrations, both in ventilators in 30 min. for intermediate indoor air quality in winter winter and summer. bathrooms and and 9 in summer. When the homes were kitchens grouped into 3 approximate groups of low, medium and high air exchange rates a statistically significant difference was found for living room TVOC. Peak level studies showed that the only house with TVOC levels within recommended guidelines had the highest ventilation rate (1.18 ACH50), whereas those with the highest TVOC values had low ventilations rates (below 0.5 ACH50).
Fernández-Agüera Seville, Mediterranean 3 Natural 9.5, 14, 15.5 10 min intervals Indoor average TVOC concentration is Globally, the more airtight No clear et al., 2019b Spain and ventilation over 12 months in usually below the 1200 ppb threshold with dwellings have higher relationship continental living room and uneven distribution: the 95th percentile TVOC concentrations, bedroom values are 1698 ppb (CS1), 420 ppb (CS2), although this is impacted and 672 ppb (CS3), with no regular by different emission behaviour. rates in individual dwellings, which makes it difficult to determine any relationship between airtightness and TVOC concentration.
Rapid Review - Airtightness 31
Reference Location Climate Sample Ventilation Airtightness Monitoring Results Finding summary Impact size (ACH50) frequency and duration
Francisco et al., Illinois and Humid 81 42 had 3.2-39.3 Measured by The levels of TVOCs pre and post There is no evidence of No clear 2017 Indiana, continental, houses mechanical (mean of 18.8 passive sampling weatherisation for properties with natural any correlation between relationship USA humid ventilation and 17.5 for over 1 week ventilation showed no difference as a result airtightness and TVOC subtropical installed, 39 had natural of increased airtightness. However, in the levels, although it has natural ventilation and group with mechanical ventilation there was been shown that ventilation mechanical a statistically significant decrease in the mechanical ventilation ventilation TVOC levels experienced, with the average has a positive impact on respectively) level dropping from 204 ppb to 142 ppm the TVOC levels. (p<0.05).
Langer and Bekö, Sweden Temperate 157 80% of 1.5 to 50 Measured using The mean TVOC concentration was found This study suggests that Negative 2013 single- apartments and passive samplers to be 306 μg/m3 for single family houses lower air change rates are family less than 50% of for 14 days and 174 μg/m3 for apartments. The TVOC associated with higher houses detached concentration was found to be statistically TVOC concentrations. and houses have negatively correlated with the air exchange 148 mechanical rate (p<0.01). apartm ventilation ents
Lawton, 1998 Ontario, Cold 59 Natural Average of 19 Passive The average concentrations of VOCs was An inverse correlation Negative Canada continental ventilation, 3 for high dosimeters were 0.313 μg/m3 in houses with high biological between air change rate houses with biological deployed in the contamination, and 0.243 μg/m3 in houses and VOC concentration. mechanical growth and living room for with low biological contamination. Specific ventilation 10.2 for low approximately one results are not presented but it is stated that biological week. an inverse correlation was found between measured and predicted air change and concentrations of internally generated contaminants such as VOCs (p<0.05).
McKay et al., 2010 Oxford, UK Temperate 20 Natural 4.4 to 19.8 TVOCs measured For the TVOC concentrations measured, 8 There is evidence of a Negative ventilation with over 7 days out of the 15 dwellings (53%) had levels that negative correlation trickle vents were higher than the guideline figure of between TVOC and 300 μg/m3. There is no clear indication of a airtightness. particular source for these high levels. When compared with the air permeability, there is a distinct correlation with houses with low permeability having the highest TVOC levels.
Offermann, 2009 California, Mediterranean 104 Most properties 2.8 to 8.4, Measured over 24 For each of the VOC compounds measured, There was no clear No clear USA were natural median 4.8 hour average the percentage of homes that exceeded the relationship observed relationship ventilation, 8 had period ‘No Significant Risk’ levels for concentration between airtightness and MHRV) and 14 indoors, varied from 8% to 63%. A total of VOC. However, as a had intermittent 20% of the homes had indoor benzene substantial percentage of ducted outdoor concentrations that exceeded the calculated new homes have indoor air system. indoor ‘Maximum Allowable Dose Levels’. concentrations exceeding Indoor sources such as mothballs, furniture guidelines for VOCs, this polish, cleaning chemicals and air suggests there may be a fresheners were identified as potential negative relationship. contributing sources to the high levels.
Rapid Review - Airtightness 32
Table 9: Summary of results for NO2
Reference Location Outdoor NO2 Climate Ventilation Airtightness Monitoring Results Finding summary Impact 2 levels (ACH50) frequency and duration
Broderick et al., Dublin, Ireland Good Temperate Natural ventilation Pre retrofit - Measured in There was no significant change in No impact observed. No clear 2017 with background 10.6 to 11.5 living room for 3 concentrations of NO2 post retrofit in relationship wall vents Post retrofit - weeks either group. 6.4 and 9.1
Doll et al., 2016 North Good Humid Natural ventilation, 3 – 32 Passive air The rate of compliance to NO2 There was no No clear Carolina, USA subtropical / with bathroom (mean of 17 sampling over average guidelines increased from evidence of a relationship cool humid extract fans. Most pre-retrofit and 2-3 days 91% pre-weatherization to 94% relationship observed houses had forced 12 post- post-weatherization, although the between airtightness air heating/cooling. retrofit). comparison of pre to post values and NO2 Average showed no statistically significant concentrations. change of 4.8.. difference following the weatherization.
Less et al., 2015 California, Good Mediterranean Half with natural 0.4 to 10.3 6 days of Median kitchen concentrations of It is not possible to No clear USA ventilation, half with (median 2.8) passive NO2 varied between 5.4 and 13.1 determine if there is a relationship mechanical samples ppb. Levels of NO2, NO and NOx correlation between
ventilation were 2 to 3 times higher in the 15 NO2 and airtightness homes using gas-cooking from the results. appliances compared to the 9 with electric appliances. Although an assessment is not provided, examining the data there does not appear to be a correlation between airtightness and NO2 and mechanical ventilation does not seem to have an impact on the results.
Offermann, 2009 California, Good Mediterranean Most properties 2.8 to 8.4, Measured over The median indoor concentration of There was no clear No clear 3 USA were natural median 4.8 a 24 hour NO2 was 3.1 µg/m with a range of relationship observed relationship ventilation, 8 had period - only 2.6 to 50 µg/m3. This is well below between airtightness
MHRV and 14 had measured in 31 the guideline for healthy air. and NO2 intermittent ducted homes. concentrations. outdoor air system.
2 3 3 Outdoor NO2 level rating determined from data in EEA (2018) and EPA (2018) – good is annual mean of <50µg/m , Moderate is 51-100µg/m
Rapid Review - Airtightness 33
Reference Location Outdoor NO2 Climate Ventilation Airtightness Monitoring Results Finding summary Impact 2 levels (ACH50) frequency and duration
Dimitroulopoulou England, UK Moderate Temperate Natural ventilation, 4.8-20.2 NO2 measured Results were between 1.9 and 28.5 There was a positive Positive et al., 2005 with extract fans or (average 12.9) by diffusion ppb in winter and 3.8 and 27.7 ppb correlation found passive stack tubes over 2 in summer. Six homes were above between air change ventilators in weeks and the WHO annual mean rates and NO2 levels bathrooms and logged to recommendation in summer, only 1 in the living room in kitchens determine peak in winter. Correlation was found summer. levels of a 7 day between NO2 levels and window period. opening, although this may be related to higher external NO2 levels. When the homes were grouped into 3 approximate groups of low, medium and high air exchange rates a statistically significant difference was found for living room NO2 in summer, with lower NO2 levels experienced in homes with lower exchange rates.
Langer and Bekö, Sweden Moderate Temperate 80% of apartments 1.5 to 50 Measured using The mean NO2 concentration was This study suggests Positive 3 2013 and less than 50% passive found to be 7 µg/m for single family that lower air change of detached houses samplers for 14 houses and 10.8 µg/m3 for rates are associated have mechanical days apartments. The likely cause of this with lower NO2 ventilation difference is differences in outdoor concentrations, which air concentration, as apartments are suggests the major more likely located in urban areas source of NO2 is with higher outdoor levels. The NO2 outdoors. concentration was found to be statistically positively correlated with the air exchange rate (p<0.01).
McKay et al., 2010 Oxford, UK Moderate Temperate Natural ventilation 4.4 to 19.8 Measured using NO2 concentrations recorded in the There was no No clear with trickle vents passive samples were between 7.9 and evidence of any relationship diffusion over 7 56.6 μg/m3 (mean 24, or 29 μg/m3 if correlation between days, accuracy only houses using gas for cooking the airtightness and of ±10%. considered). Four dwellings the NO2 exceeded ADF recommendations - 3 concentrations. did not have fans that were adjacent to the cooker with sufficient flowrate and the fourth did not use the extract ventilation when cooking.
Rapid Review - Airtightness 34
Table 10: Summary of results for mould/moisture
Reference Location Climate Ventilation Airtightness Monitoring frequency and Results Finding summary Impact (ACH50) duration
Dimitroulopoulou England, Temperate Natural 4.8-20.2 Condensation problems Humidity measured varied between 26.4 There was no clear relationship No clear et al., 2005 UK ventilation, (average monitored through user and 61.7% in winter and 41-63% in observed between airtightness and relationship with extract 12.9) survey; humidity also summer, below required levels for mould relative humidity.
fans or measured over the two growth. In the peak study phase, one house passive stack weeks. presented with relative humidity levels over ventilators in 70% for prolonged periods. This house had bathrooms the lowest ventilation rate (0.19 ACH) but and kitchens also had the lowest temperatures. Relative humidity was not correlated with the ventilation rate, although it was impacted by the use of trickle vents in summer and window opening in winter. No significant impact of ventilation on damp problems was found in the study.
Fernández-Agüera Madrid and Mediterranean Natural 3.2 to 9.4 Relative humidity was Mean relative humidity values varied The correlations observed between Negative et al., 2019a Seville, and ventilation logged at 10-min intervals between 40 and 78% in winter in the airtightness and relative humidity Spain continental for a full year bedroom and 37 to 76% in the living room. showed that the humidity generated by occupants was dissipated more effectively in flats with more permeable envelopes.
Francisco et al., Illinois and Humid 42 had 3.2 - 39.3 Relative humidity The moisture balance, or difference There is insufficient evidence to No clear 2017 Indiana, continental, mechanical (mean of measured continuously between indoor and outdoor vapour suggest any correlation between relationship USA humid ventilation 18.8 and over 6 months pressure was used to characterize the airtightness and humidity levels, subtropical installed, 39 17.5 for humidity in the house. Following although the installation of mechanical had natural natural weatherisation, both natural ventilation and ventilation has been shown to make a ventilation ventilation mechanical ventilation groups experienced statistically significant decrease in the and a decrease in the moisture balance, humidity levels. mechanical although only the mechanical ventilation ventilation group was statistically significant (p<0.001). respectively)
McKay et al., 2010 Oxford, UK Temperate Natural 4.4 to 19.8 Relative humidity and Weekly average relative humidity levels There is a negative correlation Negative ventilation (Tracer gas temperature were varied from 43 to 71%. Four homes between relative humidity and with trickle results as recorded at 5 minute recorded average weekly levels above 65%, airtightness, but there is insufficient vents blower door intervals over 7 days. and these all had low air change and data to provide a correlation to mould results were Humidity accuracy of ±2% ventilation rates. However, this is potentially occurrence. given in compounded by the fact that 3 of these 4 permeability) dwellings were flats with insufficient ventilation. There is a rough negative correlation between the relative humidity and the airtightness of the dwelling.
Rapid Review - Airtightness 35
Reference Location Climate Ventilation Airtightness Monitoring frequency and Results Finding summary Impact (ACH50) duration
Hall et al., 2013 Nottingham, Temperate Natural 0.1-0.74 Simulated – using the Of the scenarios that were tested, four had Results show that if the air change Negative UK ventilation in (based on a WUFI hygrothermal high levels of mould growth. These include rate is higher (either where ACH >0.4, base case, single room) simulation program to phase 1, with a higher air change rate and or when using MVHR), internal relative with some look at mould growth inefficient boiler, scenario #15, which had humidity fluctuations are dominated by cases potential over 10 years the lowest air change rate and no outdoor psychrometric conditions investigated mechanical ventilation and scenario #23 When ACH is very low (<<0.4), with (MVHR) which also had a low air change rate with no internal relative humidity fluctuation is mechanical ventilation. Each of these dominated by the internal loads and scenarios leads to high internal relative those without mechanical ventilation humidity. Additionally, 6 other scenarios were exposed to mould risk. have a significant mould growth potential.
Hashemi and London, UK Temperate Natural 1.5, 4.5, 7.5, Simulated - dry, moist and A high risk of condensation was found for all The risk of mould was more influenced Negative Khatami, 2015 ventilation 10.5 and 15 wet occupancy conditions, wet occupancy conditions regardless of the by the occupancy conditions than the with trickle with total number of hours airtightness. For moist conditions, the airtightness, although, low air change vents when there was a risk of situation considerably improved and for dry rates did result in higher risk of condensation on the conditions the risk was negligible for all condensation. window reported. airtightness values except for 1.5 ACH50
Lawton, 1998 Ontario, Cold Natural Average of Measurements of The average airtightness for houses with The results suggest that natural Positive Canada continental ventilation, 3 19 for high biological contaminants low biological contamination was 0.51, and ventilation, even in leaky houses, does houses with biological taken and used to group the average for the houses with high not control biological contamination, mechanical growth and the houses as either high biological contamination was 0.95. A and leaky houses may even be more ventilation 10.2 for low biological growth or low positive correlation was also seen with likely to have high levels of biological biological biological growth. measures of biological contamination contamination. These findings could including ranked measures of ergosterol, possibly be explained by houses with colony-forming units on glycerol base, dust high levels of air leakage also having mite antigen, and endotoxin. High biological high levels of water leakage or having contamination was not directly correlated hidden mould growth due to with visible mould growth in the houses condensation within envelope though. assemblies.
Ridley et al., 2006 England, Temperate Natural 5- 35 Occurrence of mould For observed mould occurrence, 43% of The results suggest that air tightness No clear UK ventilation Ventilation measured (including houses in study were found to have mould. by itself cannot be seen as the only relationship rates of 0.2, mould on window frames). No clear association could be found predictor of mould problems. 0.4, 0.6, 0.8 The simulated risk of between occurrence of mould and air Modelling suggests that lower and 1 were mould was developed for infiltration level. For simulations, the risk of ventilation rates correlate with higher simulated a range of ventilation mould at certain times or all year was found mould risk. An average ventilation rate rates (0.2-1.0 ACH) and to occur for moisture production of >6 of 0.5-1 ACH or greater is needed to moisture production rates kg/day for 0.2 ACH, >10 kg/day for 0.4 and avoid mould growth. (2-14 kg/day). >12 kg/hr for 0.6 ACH. No risk of mould for 0.8 or 1.0 ACH.
Shrestha et al., Colorado, Semi-arid/dry 11 of the 216 2-40, Visual observance of Mould growth was observed in 31.9% of the The results suggest that mould Positive 2019 USA houses had Average 11.9 mould growth, houses investigated, with window observation was positively correlated mechanical observance of window condensation observed on 10.2% and with the infiltration rate, which is ventilation condensation, dampness on walls/floor in 21.3%. Homes related to the airtightness. This is systems (5%) with visible mould in surveys had a counter-intuitive to the conventional statistically significant 12% higher median notion that tight homes can trap annual average infiltration rate, but there humidity indoors for longer periods of was no statistical difference for window time. condensation or dampness on walls/floor.
Rapid Review - Airtightness 36
Table 11: Summary of results for other IAQ factors
Reference Climate Climate Ventilation Airtightness Other measurement Results Finding summary Impact (ACH50) frequency and duration
Broderick et al., Dublin, Temperate Natural ventilation Pre retrofit - Radon – over 3 Three month-averaged radon levels decreased post It was not possible to No clear 2017 Ireland with background wall 10.6 to 11.5, months retrofit in both groups, from 56.42 Bq/m3 to 42.07 discern a relationship relationship vents Post retrofit - Bq/m3 , but with no discernible pattern; radon between airtightness and 6.4 and 9.1 concentrations post retrofit increased (by a maximum radon levels. of 40.9%) in eight dwellings and decreased (by a maximum of 50%) in seven.
Derbez et al., France Temperate MVHR 0.29 - 3.33 Radon - passive The radon concentrations ranged between 7 and 66 The increase airtightness Positive 2014 sampling during more Bq/m3. These values were lower than the national and use of MVHR in these than 2 months after median concentration in standard French dwellings, properties may correlate to inhabitants moved in which may be associated with the low airtightness of a lower radon concentration. these buildings.
Doll et al., 2016 North Humid Natural ventilation, 3 – 32 Radon - monitored Compliance rates with average radon guidelines was There are indications of Negative Carolina, subtropical / with extract fans (mean of 17 every 60 min for 2-3 above 80% for all properties. A moderate negative moderate negative USA cool humid bathrooms. Most pre-retrofit days correlation with 2-tailed significance (p=0.066) was correlation between radon houses had forced and 12 post- found for higher radon with lower ACH50. and airtightness. air heating/cooling. retrofit). Average change of 4.8.
Milner et al., London, Temperate Natural ventilation 1.2 - 48 Radon - modelled The radon results showed an approximately Findings show that reducing Negative 2015 England based on an indoor exponential decay form with concentrations highest at permeability of dwellings to source very low levels of permeability, initially falling rapidly low levels without additional with increasing permeability but settling out around 15 purpose-provided ventilation 3/ 2 m m .h (~18 ACH50). The impact of trickle vents and is likely to increase radon exhaust fans decreased the concentrations for lower levels. permeability.
Francisco et al., Illinois Humid 42 had mechanical 3.2-39.3 Carbon monoxide The levels of radon pre and post weatherisation for There is insufficient No clear 2017 and continental, ventilation installed, (mean of 18.8 (CO) measured by properties with natural ventilation showed no evidence to suggest any relationship Indiana, humid 39 had natural and 17.5 for passive sampling over difference as a result of increased airtightness. correlation between USA suptropical ventilation natural and 1 week. However, in the group with mechanical ventilation airtightness and radon mechanical Radon samples there was a minor decrease in the radon levels levels or CO levels. ventilation collected in basement experienced, that was not statistically significant respectively) and living areas (p<0.08). There was no change observed in the CO levels between pre and post weatherisation for natural ventilation or mechanical ventilation.
Dimitroulopoulou England, Temperate Natural ventilation, 4.8-20.2 CO - measured CO values measured were between 0.03 and There was no clear No clear et al., 2005 UK with extract fans or (average 12.9) through colorimetric 0.69 ppm in winter and 0.05 - 0.39 ppm in summer, relationship observed relationship passive stack diffusion tubes over which is below the WHO 8 hour average guidance between airtightness and ventilators in two weeks. value (8.6 ppm). No correlation was found between CO levels. bathrooms and ventilation and CO rates. kitchens
Offermann, 2009 California, Mediterranean Most properties were 2.8 to 8.4, CO measured in real The median maximum 1‐hour average indoor CO There was no clear No clear USA natural ventilation, 8 median 4.8 time at one minute concentration was 1.6 ppm, with a range of 0.4 ppm to relationship observed relationship had MHRV and 14 intervals over 24 hour 6.8 ppm. The median maximum 8‐hour average between airtightness and had intermittent period. outdoor concentration was 2.4 ppm, with a range of CO levels. ducted outdoor air 0.4 ppm to 4.9 ppm. None of the indoor or outdoor system. concentrations of carbon monoxide exceeded either the 9 ppm 8‐hour guideline or the 20 ppm 1‐hour guideline.
Rapid Review - Airtightness 37
Only looks at infiltration rate of Dai et al., 2018 Quality, risk of bias and confidence in bedroom cumulative evidence Fernández-Agüera No link to air quality measurements There are a number of items to consider when assessing et al., 2016 the quality of studies, risk of bias and confidence in Only considers generalised impacts Gens et al., 2014 cumulative evidence. of increased air tightness Did not measure airtightness of Firstly, the measurement of airtightness of the houses Holsteijn et al., 2016 can introduce variability when being used to compare homes building type and performance. Whilst the majority of the Howieson, Sharpe No specific comparison of results for studies reviewed used blower door tests, a small number and Farren, 2014 different airtightness levels. used tracer gas measurements, which were converted to Does not explicitly compare air Huang et al., 2018 ACH50 values for comparison. However, the applicability tightness impacts of ACH50 as a measure of the building background Langer et al., 2016 Airtightness not measured or used ventilation rate is still under question (Johnston and Less and Walker, Stafford, 2017). The impact of ventilation was Same study as Less et al., 2015 2014 considered for some studies which had mechanical ventilation, although the impact of occupant ventilation Less, 2012 Same study as Less et al., 2015 practices (e.g. window opening) was not fully explored in No comparison for different Liu et al., 2019 many of the studies, and may have influenced the airtightness levels results. Does not compare moisture levels McNeil et al., 2015 There was substantial variation in the timeframes using for different airtightness results. Does not measure airtightness of the studies reviewed – from as little as 24 hours through Meier et al., 2015 to a full year. Some of these variations were due to properties limitations of the sampling method (e.g. passive Does not consider total airtightness Nantka, 2005 sampling). As a result, some of the results presented but individual effects. may not accurately represent the air quality over a Nguyen et al., 2020 Not a peer reviewed article prolonged period of time. There is also inherent Only provides details on air variability in results from the impact of outdoor Noris et al., 2013 concentrations of contaminants. In some studies where exchange rate, not airtightness Only looks at air tightness this was accounted for by measuring the outdoor Ramos et al., 2015 concentration and considering indoor/outdoor ratios, in measurements the majority of cases outdoor measurements were not No specific results for the building air Sassi, 2017 undertaken. tightness or impact on IAQ Does not correlate air permeability Sample sizes of the studies reviewed also varied Sharpe et al., 2014a significantly, from studies looking at changes in a single against the indoor air quality house, through to the largest study which investigated Airtightness values were not Sharpe et al., 2014b 216 homes. The majority (57%) of the studies had a measured as part of the study. sample size of less than 25 houses. The lack of large Airtightness values were not Sharpe et al., 2015 scale studies causes limitations when attempting to measured as part of the study No specific results reported for the reach generalised conclusions. Shrestha et al., 2019 airtightness or impact on IAQ Shrubsole et al., No specific results for comparing the Overview of the excluded studies 2012 airtightness with PM2.5 values Humidity only correlated against Thirty six studies were excluded from this review at the Stephen et al., 1997 house leakage not airtightness full paper review stage. A summary of the excluded No direct correlation with indoor air Stephen, 1998 studies, including reasons for their exclusion is provided quality. in Table 12. Full reference details for these studies is No specific results for airtightness or Telfar Barnard, 2016 included in the reference list. indoor air quality Does not consider air quality impacts Timusk et al., 1991 Table 12: Excluded studies of changing air tightness Vardoulakis et al., No specific results for airtightness or Reference Exclusion Reason 2015 indoor air quality No consideration of infiltration Does not draw direct correlations Banfill et al., 2012 impact on indoor air quality Vinha et al., 2018 between airtightness and indoor air quality. Bomberg et al., No direct correlation with indoor air Focus is only on fire safety not on 2016 quality. Wade, 1994 indoor air quality Collins and No specific results for airtightness or Did not compare results for changing Willem et al., 2013 Dempsey, 2019 indoor air quality airtightness No impacts on air quality variables Does not provide specific air Crawley et al. 2018 Yoshino et al., 2004 measured tightness measurements Simulated data only, more focused Yuan et al., 2018 on the economic benefit
Excluded studies related to school buildings where further investigation may be warranted, and include: There were a number of studies that were identified as fitting the inclusion criteria, with the exception of the The impact of changes in building code requirement of residential buildings. Eight of these airtightness requirements on indoor air quality related specifically to schools. Table 13 includes a list of all studies that were found that investigated the Hygrothermal performance and moisture control relationship between airtightness and indoor air quality in within building envelopes schools. To consider airtightness in the context of total Table 13: Excluded studies related to airtightness in ventilation rates for a home. This is particularly schools important for understanding mould risk, which appears to have a complex relationship with airtightness. Reference Research focused on the threshold at which Almeida, R. M. S. F. et al. (2017) ‘Natural ventilation and mechanical ventilation is typically implemented indoor air quality in educational buildings: experimental (usually 5 air changes per hour at 50Pa - ACH50) assessment and improvement strategies’, Energy Efficiency. Springer, 10(4), pp. 839–854. Larger and more long-term studies are still required to improved confidence in this area. Blondeau, P. et al. (2005) ‘Relationship between outdoor The review has highlighted the complex relationship and indoor air quality in eight French schools’, Indoor Air, between airtightness and IAQ. To improve our 15(1), pp. 2–12. understanding of this issue, more research is needed that explores the relationship with explicit Fernández-Agüera, J. et al. (2019) ‘CO2 Concentration and consideration and reporting of total ventilation rates occupants’ symptoms in naturally ventilated schools in Mediterranean climate’, Buildings. Multidisciplinary Digital (i.e. including natural or mechanical ventilation), local Publishing Institute, 9(9), p. 197. ambient pollutant concentrations, and sources of indoor pollutants (for instance, any new sources of Gil-Baez, M. et al. (2017) ‘Natural ventilation systems in pollutants introduced during upgrades). 21st-century for near zero energy school buildings’, Energy. Elsevier, 137, pp. 1186–1200. Summary and conclusions Mijakowski, M. and Sowa, J. (2017) ‘An attempt to improve indoor environment by installing humidity-sensitive air inlets In summary, this rapid review identified 21 studies that in a naturally ventilated kindergarten building’, Building and investigated the impact of increasing airtightness on Environment. Elsevier, 111, pp. 180–191. indoor air quality. These studies covered a broad range of locations, climates and building types. Indoor air Moon, H. J., Sohn, J.-R. and Hwang, T. (2016) ‘Indoor Air quality parameters investigated were CO2, PM2.5, Quality Performance of Ventilation Systems in Classrooms’, formaldehyde, VOC, NO2, relative humidity, mould Journal of Asian Architecture and Building Engineering. issues, CO and radon. Taylor & Francis, 15(2), pp. 343–348. Based on the studies reviewed, there was limited Persson, J., Wang, T. and Hagberg, J. (2019) ‘Indoor air evidence to identify direct correlations between quality of newly built low-energy preschools–Are chemical increasing airtightness and indoor air quality in general. emissions reduced in houses with eco-labelled building A negative correlation with CO2 concentration was found materials?’ Indoor and Built Environment. Sage Publications from the studies, with concentrations increasing with a Sage UK: London, England, 28(4), pp. 506–519. decrease in the air exchange rate. Some evidence of a negative correlation was also found for VOC and formaldehyde, though a number of studies found no Review limitations clear relationship for these parameters. A positive correlation was found between the air exchange rate and The current rapid review was limited in scope by PM and NO concentrations in areas where there are definition. One of the main limitations of the study was 2.5 2 high outdoor levels. In these cases, increasing using the measure of airtightness as the comparison airtightness was found to reduce the infiltration of variable and considering only those studies with outdoor contaminants. measured or simulated airtightness values. The impact of airtightness on indoor air quality will be affected by For mould/moisture, the results were somewhat ventilation practices within the house (both mechanical confounding with evidence from the studies of a negative and occupant controlled natural ventilation). There are correlation on relative humidity, but a positive correlation some studies which investigate the impact of the with actual mould growth. This contradiction highlights ventilation and air exchange within houses which were the complexities surrounding mould growth. excluded as they did not measure airtightness explicitly.
There were also key aspects of the original question which were not able to be addressed due to the limited and scoping nature of the rapid review. These are areas
Rapid Review - Airtightness 39
Resources, workload and timeline
Figure 5: Review team members
Figure 6: Review timeline.
Table 14: Workloads (in hours) of the team members for each main review stage
Review Stage PC GK DD MD LK Total Comments
Team formation 5 15
Question refinement 2 2 5 2 5 16
Protocol preparation 7 7 14
Search and 16 2 7 screening
Data extraction 5 2 30 37
Synthesis / Report 10 5 35 54
Total 9 9 34 9 91 152
Rapid Review - Airtightness 41
Hashemi, A. and Khatami, N. (2015) ‘The effects of air References permeability, background ventilation and lifestyle on energy performance, indoor air quality and risk of Included studies condensation in domestic buildings’, Sustainability (Switzerland), 7(4), pp. 4022–4034. doi: Broderick, Á., Byrne, M., Armstrong, S., Sheahan, J. and 10.3390/su7044022. Coggins, A.M. (2017) ‘A pre and post evaluation of indoor air quality, ventilation, and thermal comfort in Langer, S. and Bekö, G. (2013) ‘Indoor air quality in the retrofitted co-operative social housing’, Building and Swedish housing stock and its dependence on building Environment, 122, pp. 126–133. doi: characteristics’, Building and Environment. Elsevier, 69, 10.1016/j.buildenv.2017.05.020. pp. 44–54. Choi, D. H. and Kang, D. H. (2017) ‘Infiltration of Lawton, M. D. (1998) ‘The influence of house ambient PM2.5 through building envelope in apartment characteristics in a Canadian community on housing units in Korea’, Aerosol and Air Quality microbiological contamination’, Indoor Air, 8(1), pp. 2– Research, 17(2), pp. 598–607. doi: 11. doi: 10.1111/j.1600-0668.1998.t01-3-00002.x. 10.4209/aaqr.2016.06.0287. Less, B., Mullen, N., Singer, B. and Walker, I. (2015) Chu, Y., Xu, P., Yang, Z. and Li, W. (2017) ‘Retrofitting ‘Indoor air quality in 24 California residences designed existing buildings to control indoor PM2.5 concentration as high-performance homes’, Science and Technology on smog days: Initial experience of residential buildings for the Built Environment. Taylor & Francis, 21(1), pp. in China’, Building Services Engineering Research and 14–24. Technology. SAGE Publications Ltd STM, 39(3), pp. McKay, S., Ross, D., Mawditt, I. and Kirk, S. (2010) 263–283. doi: 10.1177/0143624417728187. ‘Ventilation and indoor air quality in Part F 2006 Homes’, Derbez, M., Berthineau, B., Cochet, V., Lethrosne, M., Department for Communities and Local Government Pignon, C., Riberon, J. and Kirchner, S. (2014) ‘Indoor report BD, 2702. air quality and comfort in seven newly built, energy- Milner, J., Hamilton, I., Shrubsole, C., Das, P., Chalabi, efficient houses in France’, Building and Environment. Z., Davies, M. and Wilkinson, P. (2015) ‘What should the Elsevier, 72, pp. 173–187. ventilation objectives be for retrofit energy efficiency Dimitroulopoulou, C., Crump, D., Coward, S., Brown, B., interventions of dwellings?’, Building Services Squire, R., Mann, H., White, M., Pierce, B. and Ross, D. Engineering Research and Technology. Sage (2005) ‘Ventilation, air tightness and indoor air quality in Publications Sage UK: London, England, 36(2), pp. 221– new homes’, BR477, BRE Bookshop. 229. Doll, S. C., Davison, E. L. and Painting, B. R. (2016) Offermann, F. J. (2009) ‘Ventilation and indoor air quality ‘Weatherization impacts and baseline indoor in new homes’, PIER Collaborative Report. environmental quality in low income single-family Ridley, I. Pretlove, S., Ucci, M., Mumovic, D., Davies, M., homes’, Building and Environment. Elsevier, 107, pp. Oreszczyn, T., McCarthy, M. and Singh, J. (2006) 181–190. ‘Asthma/dust mite study-Final report: sensitivity of Fernández-Agüera, J., Domínguez-Amarillo, S., Alonso, humidity and mould growth to occupier behaviour in C. and Martín-Consuegra, F. (2019) ‘Thermal comfort dwellings designed to the new air tightness and indoor air quality in low-income housing in Spain: requirements’. Office of the Deputy Prime Minister, The influence of airtightness and occupant behaviour’, London. Energy and Buildings, 199, pp. 102–114. doi: Shrestha, P. M. Humphrey, J.L., Barton, K.E., Carlton, 10.1016/j.enbuild.2019.06.052. E.J., Adgate, J.L., Root, E.D. and Miller, S.L. (2019) Fernández-Agüera, J., Dominguez-Amarillo, S., ‘Impact of low-income home energy-efficiency retrofits Fornaciari, M. and Orlandi, F. (2019) ‘TVOCs and PM on building air tightness and healthy home indicators’, 2.5 in naturally ventilated homes: Three case studies in Sustainability (Switzerland), 11(9). doi: a mild climate’, Sustainability (Switzerland), 11(22). doi: 10.3390/su11092667. 10.3390/su11226225. Szirtesi, K., Angyal, A., Szoboszlai, Z., Furu, E., Török, Francisco, P. W., Jacobs, D. E., Targos, L., Dixon, S. L., Z., Igaz, T. and Kertész, Z. (2018) ‘Airborne Particulate Breysse, J., Rose, W. and Cali, S. (2017) ‘Ventilation, Matter: An Investigation of Buildings with Passive House indoor air quality, and health in homes undergoing Technology in Hungary’, Aerosol and Air Quality weatherization’, Indoor Air. John Wiley & Sons, Ltd, Research, 18, pp. 1282–1293. 27(2), pp. 463–477. doi: 10.1111/ina.12325. Wang, F. Meng, D., Li, X. and Tan, J. (2016) ‘Indoor- Hall, M. R., Casey, S.P., Loveday, D.L. and Gillott, M. outdoor relationships of PM2.5 in four residential (2013) ‘Analysis of UK domestic building retrofit dwellings in winter in the Yangtze River Delta, China’, scenarios based on the E.ON Retrofit Research House Environmental Pollution, 215, pp. 280–289. doi: using energetic hygrothermics simulation - Energy 10.1016/j.envpol.2016.05.023. efficiency, indoor air quality, occupant comfort, and mould growth potential’, Building and Environment, 70, pp. 48–59. doi: 10.1016/j.buildenv.2013.08.015.
Excluded studies Less, B. and Walker, I. (2014) ‘Indoor Air Quality and Ventilation in Residential Deep Energy Retrofits. Banfill, P., Simpson, S., Haines, V. and Mallaband, B. Lawrence Berkeley National Lab’. (LBNL), Berkeley, CA (2012) ‘Energy-led retrofitting of solid wall dwellings: (United States). Technical and user perspectives on airtightness’, Structural Survey, 30(3), pp. 267–279. doi: Liu, Z., Wu, D., He, B., Wang, Q., Yu, H., Ma, W. and 10.1108/02630801211241829. Jin, G. (2019) ‘Evaluating potentials of passive solar heating renovation for the energy poverty alleviation of Bomberg, M., Kisilewicz, T. and Nowak, K. (2016) ‘Is plateau areas in developing countries: A case study in there an optimum range of airtightness for a building?’, rural Qinghai-Tibet Plateau, China’, Solar Energy, 187, Journal of Building Physics, 39(5), pp. 395–421. doi: pp. 95–107. doi: 10.1177/1744259115603041. https://doi.org/10.1016/j.solener.2019.05.049. Collins, M. and Dempsey, S. (2019) ‘Residential energy McNeil, S., Plagmann, M., McDowall, P. and Basset, M. efficiency retrofits: potential unintended consequences’, (2015) ‘The role of ventilation in managing moisture Journal of Environmental Planning and Management, inside New Zealand homes’, SR341. BRANZ Ltd, 62(12), pp. 2010–2025. doi: Judgeford. 10.1080/09640568.2018.1509788. Meier, R., Eeftens, M., Phuleria, H.C., Ineichen, A., Crawley, J., Wingfield, J. and Elwell, C. (2018) ‘The Corradi, E., Davey, M., Fierz, M., Ducret-Stich, R.E., relationship between airtightness and ventilation in new Aguilera, I., Schindler, C. and Rochat, T. (2015) UK dwellings’, Building Services Engineering Research ‘Differences in indoor versus outdoor concentrations of and Technology. SAGE Publications Ltd STM, 40(3), pp. ultrafine particles, PM2.5, PM absorbance and NO2 in 274–289. doi: 10.1177/0143624418822199. Swiss homes’, Journal of exposure science & Dai, X., Liu, J., Yin, Y., Song, X. and Jia, S. (2018) environmental epidemiology, 25(5), pp. 499–505. doi: ‘Modeling and controlling indoor formaldehyde 10.1038/jes.2015.3. concentrations in apartments: on-site investigation in all Nantka, M. B. (2005) ‘Airtightness and natural climate zones of China’, Building and Environment. ventilation: A case study for dwellings in Poland’, Elsevier, 127, pp. 98–106. International Journal of Ventilation, 4(1), pp. 79–92. Fernández-Agüera, J. Domínguez-Amarillo, S., Sendra, Nguyen, M., Holmes, E. C. and Angenent, L. T. (2020) J.J. and Suárez, R. (2016) ‘An approach to modelling ‘The short-term effect of residential home energy retrofits envelope airtightness in multi-family social housing in on indoor air quality and microbial exposure: a case- Mediterranean Europe based on the situation in Spain’, control study’, bioRxiv, p. 2020.03.09.983452. doi: Energy and Buildings, 128, pp. 236–253. doi: 10.1101/2020.03.09.983452. 10.1016/j.enbuild.2016.06.074. Noris, F., Adamkiewicz, G., Delp, W.W., Hotchi, T., Gens, A., Hurley, J. F., Tuomisto, J. T. and Friedrich, R. Russell, M., Singer, B.C., Spears, M., Vermeer, K. and (2014) ‘Health impacts due to personal exposure to fine Fisk, W.J. (2013) ‘Indoor environmental quality benefits particles caused by insulation of residential buildings in of apartment energy retrofits’, Building and Environment. Europe’, Atmospheric Environment, 84, pp. 213–221. Elsevier, 68, pp. 170–178. doi: https://doi.org/10.1016/j.atmosenv.2013.11.054. Ramos, N. M. M., Howieson, S. G., Sharpe, T. and Farren, P. (2014) Almeida, R.M., Curado, A., ‘Building tight - Ventilating right? How are new air Pereira, P.F., Manuel, S. and Maia, J. (2015) tightness standards affecting indoor air quality in ‘Airtightness and ventilation in a mild climate country dwellings?’, Building Services Engineering Research rehabilitated social housing buildings – What users want and Technology, 35(5), pp. 475–487. doi: and what they get’, Building and Environment, 92, pp. 10.1177/0143624413510307. 97–110. doi: https://doi.org/10.1016/j.buildenv.2015.04.016. Huang, K., Song, J., Feng, G., Chang, Q., Jiang, B., Wang, J., Sun, W., Li, H., Wang, J. and Fang, X. (2018) Sassi, P. (2017) ‘Thermal comfort and indoor air quality ‘Indoor air quality analysis of residential buildings in in super-insulated housing with natural and northeast China based on field measurements and decentralized ventilation systems in the south of the UK’, longtime monitoring’, Building and Environment, 144, pp. Architectural Science Review, 60(3), pp. 167–179. doi: 171–183. doi: 10.1016/j.buildenv.2018.08.022. 10.1080/00038628.2017.1301371. Langer, S. Ramalho, O., Derbez, M., Ribéron, J., Sharpe, T, Porteous, C.D.A., Foster, J. and Shearer, D. Kirchner, S. and Mandin, C. (2016) ‘Indoor (2014a) ‘An assessment of environmental conditions in environmental quality in French dwellings and building bedrooms of contemporary low energy houses in characteristics’, Atmospheric Environment. Pergamon, Scotland’, Indoor and Built Environment. Sage 128, pp. 82–91. doi: Publications Sage UK: London, England, 23(3), pp. 393– 10.1016/J.ATMOSENV.2015.12.060. 416. Less, B. (2012) ‘Indoor Air Quality in 24 California Sharpe, T., McQuillan, J., Howieson, S., Farren, P. and Residences Designed as High Performance Green Tuohy, P. (2014b) ‘Research Project to Investigate Homes’. University of California, Berkeley. Occupier Influence on Indoor Air Quality in Dwellings’. The Scottish Government.
Rapid Review - Airtightness 43
Sharpe, T., Farren, P., Howieson, S., Tuohy, P. and Willem, H., Hult, E.L., Hotchi, T., Russell, M.L., McQuillan, J. (2015) ‘Occupant interactions and Maddalena, R.L. and Singer, B.C (2013) ‘Ventilation effectiveness of natural ventilation strategies in control of volatile organic compounds in new US homes: contemporary new housing in Scotland, UK’, results of a controlled field study in nine residential International journal of environmental research and units’. Report number LBNL6022E, Berkeley, CA, USA, public health. Multidisciplinary Digital Publishing Lawrence, Berkeley National Laboratory Institute, 12(7), pp. 8480–8497. Yoshino, H., Amano, K., Matsumoto, M., Netsu, K., Shrestha, P. M., Humphrey, J.L., Carlton, E.J., Adgate, Ikeda, K., Nozaki, A., Kakuta, K., Hojo, S. and Ishikawa, J.L., Barton, K.E., Root, E.D. and Miller, S.L. (2019) S. (2004) ‘Long-Termed Field Survey of Indoor Air ‘Impact of Outdoor Air Pollution on Indoor Air Quality in Quality and Health Hazards in Sick House’, Journal of Low-Income Homes during Wildfire Seasons’, Asian Architecture and Building Engineering, 3(2), pp. International Journal of Environmental Research and 297–303. doi: 10.3130/jaabe.3.297. Public Health. Multidisciplinary Digital Publishing Institute, 16(19), p. 3535. doi: 10.3390/ijerph16193535. Yuan, Y., Luo, Z., Liu, J., Wang, Y. and Lin, Y. (2018) ‘Health and economic benefits of building ventilation Shrubsole, C., Ridley, I., Biddulph, P., Milner, J., interventions for reducing indoor PM2.5 exposure from Vardoulakis, S., Ucci, M., Wilkinson, P., Chalabi, Z. and both indoor and outdoor origins in urban Beijing, China’, Davies, M. (2012) ‘Indoor PM2. 5 exposure in London’s Science of the Total Environment, 626, pp. 546–554. domestic stock: Modelling current and future exposures doi: 10.1016/j.scitotenv.2018.01.119. following energy efficient refurbishment’, Atmospheric Environment. Elsevier, 62, pp. 336–343. Stephen, F. R., McIntyre, D.A., Lane, A., Raw, G.J., Other references Wiech, C.R. and Frederick, J. (1997) ‘Ventilation and house air tightness: Effect on indoor temperature and ARPANSA (2020), ‘Radon Map of Australia’, Published humidity in Southampton, UK’, Building Services by Commonwealth of Australia as represented by the Engineering Research and Technology. SAGE Australian Radiation Protection and Nuclear Safety Publications Ltd STM, 18(3), pp. 141–147. doi: Agency (ARPANSA), 10.1177/014362449701800302. https://www.arpansa.gov.au/understanding- radiation/radiation-sources/more-radiation- Stephen, R. K. (1998) ‘Airtightness in UK dwellings: sources/radon-map, Accessed 15/05/2020 BRE’s test results and their significance’. CRC. ASBEC (2018), ‘Built to Perform An industry led pathway Telfar Barnard, L. (2016) ‘Mechanised residential to a zero carbon ready building code’, Australian ventilation systems: do health outcomes change Sustainable Built Environment Council, Building Code following installation’, Porirua, New Zealand: BRANZ. Energy Performance Trajectory Project - Final Report Timusk, J., Seskus, A. L. and Ary, N. (1991) ‘The Control July 2018 of Wind Cooling of Wood Frame Building Enclosures’, EEA (2019) ‘Air quality in Europe – 2018 Report’, Journal of Thermal Envelope and Building Science, European Environment Agency Report No 12/2018, 15(1), pp. 8–19. doi: 10.1177/109719639101500103. Available at: https://www.eea.europa.eu/publications/air- van Holsteijn, R. C. A, Li, W. L., Valk, H. J. J. and quality-in-europe-2018, Accessed 14/5/2020 Kornaat, W. (2016) ‘Improving the Energy and IAQ EPA (2018), ‘Air Quality Statistics Report’, Available at: Performance of Ventilation Systems in Dutch Dwellings’, https://www.epa.gov/outdoor-air-quality-data/air-quality- International Journal of Ventilation. Taylor & Francis, statistics-report, Accessed 14/5/2020 14(4), pp. 363–370. doi: 10.1080/14733315.2016.11684093. CC&ID (2019), ‘World map of the Koppen-Geiger climate classification updated’, Climate Change and Infectious Vardoulakis, S., Dimitroulopoulou, C., Thornes, J., Lai, Diseases Group, University of Vienna, Available at: K.M., Taylor, J., Myers, I., Heaviside, C., Mavrogianni, http://koeppen-geiger.vu-wien.ac.at/present.htm, A., Shrubsole, C., Chalabi, Z. and Davies, M. (2015) Accessed 9/9/2020 ‘Impact of climate change on the domestic indoor environment and associated health risks in the UK’, IQAir AirVisual, (2019), ‘2019 World Air Quality Report’, Environment International, 85, pp. 299–313. doi: IQAir AirVisual, Staad, Switzerland. Available at: 10.1016/j.envint.2015.09.010. https://www.airvisual.com/world-most-polluted-cities, Accessed 14/05/2020 Vinha, J., Salminen, M., Salminen, K., Kalamees, T., Kurnitski, J. and Kiviste, M. (2018) ‘Internal moisture Johnston, D. and Stafford, A. (2017), ‘Estimating the excess of residential buildings in Finland’, Journal of background ventilation rates in new-build UK dwellings– Building Physics. SAGE Publications Sage UK: London, Is n50/20 appropriate?’ Indoor and Built England, 42(3), pp. 239–258. Environment, 26(4), pp.502-513. Wade, C. A. and Building Research Association of New Tricco, A.C., Antony, J., Zarin, W., Strifler, L., Ghassemi, Zealand (1994) Report on the effect of passive M., Ivory, J., Perrier, L., Hutton, B., Moher, D. and ventilation on the rate of fire development in dwellings. Straus, S.E., (2015), ‘A scoping review of rapid review Porirua, N.Z. methods’. BMC medicine, 13(1), p.224
Rapid Review - Airtightness 44