- Email: [email protected]
Impact of room fragrance products on indoor air quality
Atmospheric Environment xxx (2014) 1e11
Contents lists available at ScienceDirect
Atmospheric Environment journal homepage: www.elsevier.com/locate/atmosenv
Impact of room fragrance products on indoor air quality Erik Uhde*, Nicole Schulz Fraunhofer Wilhelm-Klauditz-Institut, Bienroder Weg 54 E, 38108 Braunschweig, Germany
h i g h l i g h t s Various room fragrance products were tested in emission test chambers. Many products were strong sources of odorless solvents. More than 100 different fragrance substances were identified and quantified. The long term emission behavior of some products was tested. Accidental spilling of fragrance liquids may cause extreme solvent concentrations.
a r t i c l e i n f o
a b s t r a c t
Article history: Received 9 April 2014 Received in revised form 6 November 2014 Accepted 10 November 2014 Available online xxx
Everyday life can no longer be imagined without fragrances and scented products. For the consumer, countless products exists which are solely or partly intended to give off a certain scent in sufficient concentrations to odorize a complete room. Sprays, diffusers and evaporators, scented candles and automatic devices for the distribution of fragrance liquids are typical examples of such products. If the consumer uses such products, his consent to the release of certain chemicals in his home can be implied, however, he may not know what kind of fragrance substances and solvents will be present in which concentrations. In this study, we determined the volatile emissions of a number of fragrance products in detail. Measurements were carried out under controlled conditions in test chambers. The products were tested in a passive (unused) and an active state, wherever applicable. Following a defined test protocol, the release of volatile organic compounds, ultrafine particles and NOx was monitored for each product. The potential for forming secondary organic aerosols under the influence of ozone was studied, and for a selection of products the long-term emission behavior was assessed. A remarkable variety of fragrance substances was found and more than 100 relevant compounds were identified and quantified. While it is the intended function of such products to release fragrance substances, also considerable amounts of non-odorous solvents and by-products were found to be released from several air fresheners. Emissions rates exceeding 2 mg/(unit*h) were measured for the five most common solvents. © 2014 Published by Elsevier Ltd.
Keywords: Fragrance product Room spray Candle Emission Solvent
1. Introduction Scenting of household products like cleaning agents or detergents with volatile organic substances, specifically terpenes and terpenoids, has been common for many decades. Nowadays, a variety of products is available for the consumer to also perfume the air in rooms and offices. As these products become more and more popular, questions in regard to a possible impairment of indoor air
* Corresponding author. E-mail address: [email protected] (E. Uhde).
quality arise (Nazaroff and Weschler, 2004; Steinemann, 2009). The topic is of specific interest because of the substantial time per day dwellers are exposed to the scent substances (and additives/solvents) in case of home application, and when the source strength of such products is considered. It also has to be taken into account that fragrance substances of other sources may already be present in indoor air due to use of household products (Coleman et al., 2008; Kang et al., 2012; Rossignol et al., 2013; Singer et al., 2006), and room scenting products may increase both the number and the concentrations of such compounds in the indoor environment. Terpenoid substances, aldehydes and lactones are common substances in scenting products, and a majority of the compounds
http://dx.doi.org/10.1016/j.atmosenv.2014.11.020 1352-2310/© 2014 Published by Elsevier Ltd.
Please cite this article in press as: Uhde, E., Schulz, N., Impact of room fragrance products on indoor air quality, Atmospheric Environment (2014), http://dx.doi.org/10.1016/j.atmosenv.2014.11.020
2
E. Uhde, N. Schulz / Atmospheric Environment xxx (2014) 1e11
used are reactive enough to be susceptible to indoor chemistry (Nazaroff and Weschler, 2004; Singer et al., 2006; Uhde and Salthammer, 2007). Certain conditions in the distribution devices (e.g., elevated temperature, or the storage of concentrated liquids under oxygen contact for long periods of time) will also increase the possibility of not only the fragrance substances, but also degradation products being present in the air. The impact of terpene release in indoor air has been studied for a long time. The mono-terpenes as well as their ozone-initiated degradation products have been a research topic for several decades. Early work (Weschler and Shields, 1997, 1999) indicated that ozone-terpene reactions can actually be observed in the indoor environment. Extensive studies were carried out on the physiological effects of terpenes and their decomposition products/intermediates (Clausen et al., 2001; Nørgaard et al., 2006; Wilkins et al., 2003; Wolkoff et al., 2012, 1999, 2013). Allergic properties of numerous fragrance substances are recognized today, but the knowledge about long-term inhalative exposure is still fragmentary. For cosmetics usage of these substances the dermal exposure pathway is considered most important, and data on the sensitizing potential exists, see e.g. Bhatia et al. (2008), Lalko et al. (2007) and Lapczynski et al. (2008). In some countries, such substances are regulated, or the declaration is legally enforced like in e.g. EU (2003). The effectiveness of such risk management efforts is discussed controversially (Klaschka, 2010). Air fresheners and room fragrances, however, are no cosmetic products and the market is therefore mostly unregulated. Unlike fragranced detergents, cleaning products or cosmetics (Huang et al., 2011; Jo et al., 2008; Nazaroff and Weschler, 2004), the air fresheners have no other (primary) function; their sole purpose is to odorize air in a building. Their use therefore represents an intended release of a e mostly unknown e substance mix in an indoor environment (VITO, 2008). In this study a variety of room fragrance products was tested under defined conditions. The released scent substances and solvents were identified, long term emission behavior was measured and possible room concentrations were calculated. The intention of this article is to provide a more detailed overview on substances currently on the market, and on the release characteristics of different types of air fresheners.
2. Methods 2.1. Studied products All tested products were bought in local stores or supplied by manufacturers. The products were commercial items sold in complete packaging including consumer information. Due to the huge variety of products and the countless number of product types, a representative testing is hardly possible. Therefore, products were chosen to give a broad overview over existing technologies and fragrance substances. For the selection of products the following points were considered: 1) Availability (Major brands were preferred to niche products) 2) Different scents (For example, lemon or vanilla scented products are available from most brands, therefore other scents were selected when possible) 3) Different techniques (A selection of passively evaporating products, spray products, electrically driven items and scented candles was considered) Whenever a product had several intensity settings, a mean setting was selected for the test. A short description of the tested products is given in Table 1. 2.1.1. Diffusers and evaporators ‘Diffuser type’ odorizers usually consist of a fragrance liquid/gel, a storage container and some means to control evaporation or diffusion to an evaporation surface. Typical examples are stick diffusers, where several (typically 6e12) wood sticks are placed in a bottle and slowly evaporate the rising fragrance liquid over a time frame of days to weeks. Other constructions with similar function, e.g. with a wooden lid, a membrane or perforated plastic part can be found as well. For small rooms or cars gel-/liquid-filled evaporators are available. Items of each group were tested. Some products require the consumer to fill the diffuser with the fragrance liquid, while others come pre-assembled, with the liquid reservoir already in place. Therefore, the possible impact of spilling was assessed separately.
Table 1 Overview of tested fragrance products. Product
Type
Scent type
Package size
Remarks
S1
Passive diffuser
‘Spring’
5.5 ml
S2 S3
Electric evaporator Wood-stick
‘Exotic mango-orange’ ‘Green tea’
8 ml 100 ml
S4 S5
Car diffuser Automatic spray
‘White’ ‘Cherry magnolia’
18 g 250 ml
S6 S7 S8 S9
Spray Potpourri bag Potpourri Wood-stick
‘Natural’ ‘Lavender’ ‘Vanilla’ ‘Cherry blossom’
50 ml 100 g 100 g 50 ml
S10
Wood top
‘Vanilla’
50 ml
S11 S12
Wood stick Jar þ wood balls
‘Vanilla’ ‘Vanilla’
50 ml 100 ml
Fragrance liquid cartridge activated by inserting, adjustable strength Removable fragrance liquid cartridge 10 sticks need to be inserted into the pre-filled bottle. Activation by removing a plastic strip Pressurized container, timer setting adjustable in 3 steps Pressurized container Potpourri stays in fabric bag Loose potpourri Fragrance liquid needs to be poured in glass container, 12 sticks are placed in separate drill holes of a wood lid. Fragrance liquid needs to be poured in glass container, wood lid with a wooden plug immersed in the liquid. 10 sticks need to be inserted into the pre-filled bottle. Pump spray needs to be applied to wooden balls in a glass dish
S14
Candle
‘Ice cream’
450 g
Please cite this article in press as: Uhde, E., Schulz, N., Impact of room fragrance products on indoor air quality, Atmospheric Environment (2014), http://dx.doi.org/10.1016/j.atmosenv.2014.11.020
E. Uhde, N. Schulz / Atmospheric Environment xxx (2014) 1e11
2.1.2. Spray odorizers Two spray odorizers were tested: an automatic spray mechanism, which scents the room in pre-set timing intervals, and a conventional room spray specifically advertised for use in schools.
3
detector at 360 nm (HPLC/VWD, Agilent Technologies, US) using a 10 cm ACE3 C18 column (Advanced Chromatography Technologies Ltd, GB). 2.6. Determination of nitrogen oxides (for candle tests)
2.1.3. Candles Scented candles/tea lights differed from the other products in two ways e after lighting, they are active heat sources, so the test chamber must be able to remove excess heat and moisture. The candle flame will also generate combustion products, which may interfere with on-line and off-line analysis techniques. Generally, candles can be substantial sources of ultrafine and fine particles in hin et al., 2008). Undisturbed burning the indoor environment (Ge is a prerequisite to reliable results (Sunderland et al., 2011), especially particle emissions and NOx-generation will be affected by unsteadily burning candles. To protect the candle flame from the air flow and allow undisturbed burning in the test chamber, the candles were shielded by a stainless-steel mesh cylinder (30 cm diameter) behind a wind shield in the middle of the test chamber. To find out if the candle base has an influence of emission, unscented tea lights made of pure paraffin, stearin, tallow and wax were tested under the same conditions. 2.2. Test chambers
NO and NO2 were measured by reduced pressure chemiluminescence (CLD) based online detector (Horiba, APNA-370) with a flow rate of 0.8 l/min and a time resolution of 60 s. This device uses the excited state of NO2* for the quantitation of both NO (directly) and NO2 (indirectly) by cross flow modulation. 2.7. Determination of ultrafine particles The concentrations of ultrafine particles during the chamber tests were measured using a TSI Model 3091 Fast Mobility Particle Sizer (FMPS), which covers a size range of 5.6e560 nm. The time resolution was 1 spectrum per second. 2.8. Generation and determination of ozone Ozone was generated from air using an UV-Ozone generator. The concentration in the chamber was determined by a Horiba APOA 370 ozone analyzer (ND-UV).
Products were tested in emission test chambers of two sizes, the smallest chambers used were 1 m3-glass chambers (for tea lights, candles, car diffuser), and the other items were tested in 3 m3stainless steel chambers. All chambers fulfilled the requirements of ISO 16000-9. The tests were carried out under standard conditions (23 C, 50% r.h.) with an air exchange rate of 0.5 h1. For special experiments (evaporation tests to assess the consequences of spillage, candle tests), higher air exchange rates were used.
2.9. Determination of emitting substances by PTR/MS
2.3. Determination of volatile organic compounds
2.10. Determination of TOC/CO2/CO
Volatile organic compounds were determined according to ISO16000-6 (ISO, 2011) by adsorption on Tenax® TA filled stainlesssteel tubes, thermal desorption at 290 C (TD 100, Markes International, UK) and subsequent GC/MS analysis (GC 7890, MS 5975C, Agilent Technologies, USA) using a J&W Scientific DB-5MS column (L ¼ 60 m; I.D. ¼ 0.25 mm; film ¼ 0.25 mm). Calibration was carried out with internal and external standards, whenever possible against pure reference compounds (some substances, e.g. piperonal, are regulated as drug precursors and are therefore not available). The air sampling volume was between 0.5 l and 4 l, depending on the expected chamber concentrations.
For general control of the experimental progress a photoacoustic detector (Innova 1412), equipped with a TOC filter was utilized. Filters for CO2 and CO were also installed and used to monitor the combustion of the candles.
2.4. Determination of benzene Especially during candle testing or secondary organic aerosol (SOA) generation experiments, reactive substances like NOx or ozone can cause benzene artifact formation on Tenax® TA-filled sampling tubes. Therefore, additional air samples were taken on Carbotrap®-filled sampling tubes. These tubes were thermally desorbed and benzene was quantified using GC/MS with a method identical to that of the VOC determination.
A proton-transfer reaction mass spectrometer (High-sensitivity quadrupole PTR-MS, Ionicon Analytik, Austria) was used to monitor the experiments in scan-mode with a time resolution of 180 s. Especially during ozone addition this data was used to follow the concentration increase/drop of newly formed/decomposed substances.
2.11. Test protocol Various fragrance products were tested in this study using a defined test protocol. 1) Blank values for all studied parameters were measured in the test chambers. 2) Each product was unpacked and put into the test chamber without activating/lighting it. (For purely passive emitting products that started to give off odor immediately after unpacking, the procedure continued with step 5.) 3) The emissions of the product were measured. 4) The product was activated/lit in the chamber. 5) The emissions were measured for several hours to days (depending on the operation/burning duration, see below). 6) For some of the products the measurement was continued for several weeks.
2.5. Determination of volatile aldehydes Aldehydes were analyzed according to ISO 16000-3 using silica gel cartridges (Supelco) that contained 2,4-dinitrophenylhydrazine (DNPH). The cartridges were solvent extracted with acetonitrile (ACN) and the target compounds were measured by high performance liquid chromatography coupled with a variable wavelength
S1eS6: The samples were tested in 3 m3-chambers according to ISO 16000-9. Each test was carried out with one product per chamber. All samples give off odor immediately after unpacking except the electric evaporator (S2; this product had to be activated by plugging it into a wall socket). During the tests chamber air was collected on sampling tubes filled with Tenax® TA after 0.5 h, 1 h,
Please cite this article in press as: Uhde, E., Schulz, N., Impact of room fragrance products on indoor air quality, Atmospheric Environment (2014), http://dx.doi.org/10.1016/j.atmosenv.2014.11.020
4
E. Uhde, N. Schulz / Atmospheric Environment xxx (2014) 1e11
Table 2 List of identified substances and their maximum observed emission rate one hour after application start. Samples in bold print showed the highest emission rates. Substances marked with * were not available as pure reference, quantification was done against toluene. Substances marked with are listed in EU (2003) as potential allergens. CAS-no.
Substance
Identified in sample
5989-27-5 140-11-4 18479-58-8 78-70-6 115-95-7 99-85-4 123-35-3 18172-67-3 470-82-6 105-53-3 105-54-4 628-44-4 5392-40-5 78-93-3 27939-60-2 80-56-8 123-68-2 87-44-5 5655-61-8 89-78-1 99-87-6 464-49-3 93-92-5 7785-53-7 105-87-3 513-86-0 141-12-8 10385-78-1 39255-32-8 2550-26-7 106-72-9 586-62-9 112-31-2 79-20-9 106-68-3 495-61-4 123-11-5 110-41-8 1191-16-8 123-38-6 106-30-9 469-61-4 31499-72-6 642-41-9 540-18-1 79-92-5 562-74-3 104-61-0 103-95-7 17699-14-8 502-61-4 142-19-8 12542-30-2 91-64-5 13877-91-3 85-91-6 106-25-2 124-13-0 112-44-7 60-12-8 106-23-0 120-57-0 3779-61-1 106-24-1 51115-64-1 110-93-0 124-19-6 123-92-2 111-87-5 498-15-7 77-54-3
Limonene Benzyl acetate Dihydromyrcenol Linalool Linalyl acetate gamma-Terpinene Myrcene beta-Pinene Eucalyptol Malonic acid diethyl ester Butyric acid ethyl ester 2-Methyl-2-octanol Citral*, 2-Butanone Trivertal alpha-Pinene Hexanoic acid 2-propenyl ester Caryophyllene Bornyl acetate Menthol p-Cymene Camphor alpha-Methylbenzyl acetate alpha-Terpineol Geranyl acetate 3-Hydroxy-2-butanone Neryl acetate Borneol Ethyl-2-methylpentanoate Benzylacetone 2,6-Dimethyl-5-heptenal Terpinolene Decanal Methyl acetate 3-Octanone beta-Bisabolene 4-Methoxy-benzaldehyde 2-Methyl-undecanal Prenyl acetate Propanal Heptanoic acid ethyl ester alpha-Cedrene alpha-Dihydro-ionon 2-Methylbutyl acetate Butyric acid pentyl ester Camphene Terpinen-4-ol gamma-Nonalactone Cyclamal alpha-Cubebene Farnesene Isomers Heptanoic acid 2-propenyl ester Dihydrodicyclopentadienyl acrylate Coumarine Ocimene Dimethylanthranilate Nerol Octanal Undecanal 2-Phenylethanol Citronellal Piperonal* beta-Ocimene trans-Geraniol 2-Methyl-butyl butyrate 2-Methyl-2-hepten-6-one Nonanal iso-Pentyl acetate n-Octanol 3-Carene Cedryl acetate
S1, S2, S3, S4, S5, S6, S7, S9, S10, S12 S1, S2, S3, S5, S9 S1, S2, S3, S4, S8, S9, S12 S1, S2, S3, S4, S5, S6, S7, S8, S9, S12 S1, S2, S3, S4, S6, S7, S8, S12 S1, S6 S1, S2, S3, S4, S5, S6, S7, S8, S12 S1, S3, S4, S5, S6, S7, S9, S10, S12 S1, S2, S3, S4, S7 S2 S1, S2, S10, S11 S1 S3, S4, S6 S2 S2, S3, S4, S5 S1, S2, S3, S4, S5, S6, S9, S12 S2 S6, S8, S12 S1, S2, S4, S12 S1, S4, S6 S1, S6 S1, S2, S6, S7 S2, S5, S9 S5, S6, S7, S8 S1, S2, S4, S6, S7, S8, S12 S10, S11 S1, S4, S6, S7, S12 S6, S7 S2, S5, S9 S1, S2 S1, S2, S5, S9 S1, S3, S6 S1, S4, S6 S10, S11 S1, S6 S6 S1, S2, S6, S7, S8, S10, S14 S1, S2 S2 S11 S2 S7 S8 S2 S2 S1, S3, S4, S6, S12 S6, S7 S8, S11, S12 S2, S5 S6 S6 S2 S1, S4 S1, S2, S6, S7, S8, S11 S1, S12 S6 S2, S5, S6 S1, S2, S4, S6 S6 S1, S3, S5, S9 S6 S1, S8, S11, S12 S12 S2, S3, S4, S5 S1 S1, S2 S1, S2, S5, S2 S6 S1, S4, S5, S7, S12 S8
Maximum emission rate after 1 h of operation (mg/(unit*h) 9132 3920 3155 2994 2711 2688 1679 1391 992 992 774 723 711 584 584 552 507 488 452 429 381 326 257 254 254 244 230 222 191 191 189 188 174 156 147 128 122 119 119 114 108 107 98 90 90 87 86 76 68 68 66 66 66 65 57 56 50 50 47 45 39 39 38 36 36 35 30 30 30 30 30
Please cite this article in press as: Uhde, E., Schulz, N., Impact of room fragrance products on indoor air quality, Atmospheric Environment (2014), http://dx.doi.org/10.1016/j.atmosenv.2014.11.020
E. Uhde, N. Schulz / Atmospheric Environment xxx (2014) 1e11
5
Table 2 (continued ) CAS-no.
Substance
Identified in sample
6753-98-6 20085-93-2 24851-98-7 127-51-5 2111-75-3 763-32-6 3691-12-1 100-51-6 3681-71-8 3681-71-8 121-33-5 14073-97-3 17909-77-2 464-49-3 100-52-7 97-53-0 50373-59-6 3691-11-0 4674-50-4 560-32-7 66-25-1 706-14-9 514-51-2 111-71-7 43052-87-5 110-62-3 103-95-7 121-32-4 2436-90-0 529-16-8 104-55-2 6874-35-7 473-13-2 13851-11-1 106-72-9 928-96-1 98-86-2 Solvents 64-17-5 88917-22-0 67-63-0 56539-66-3
alpha-Caryophyllene Seychellene Methyldihydrojasmonate alpha-Isomethyl-ionone Perilla aldehyde 3-Methyl-3-buten-1-ol alpha-Guaiene Benzylalcohol cis-3-Hexenyl acetate cis-3-Hexenyl acetate Vanilline Menthone Alpha-sinensal Camphor Benzaldehyde Eugenol Lavandulol acetate alpha-Bulnesene Nootkatone alpha-Patchoulene Hexanal gamma-Decalactone beta-Patchoulene Heptanal alpha-Damascone Pentanal Cyclamal Ethylvanilline beta-Citronellene Santene Cinnamaldehyde alpha-Citronellene alpha-Selinene Fenchyl acetate 2,6-Dimethyl-5-heptenal cis-3-Hexen-1-ol Acetophenone
S6 S8, S12 S3, S8, S9 S4, S5, S9 S6 S10, S11 S7, S8, S12 S1, S8 S3, S9 S9 S6, S7, S8 S1, S2, S4, S6 S6 S7 S1, S2, S5, S6, S8 S8 S6 S12 S6 S8, S12 S1, S4, S5, S6 S14 S12 S6 S1, S2, S4 S3 S1, S2, S5 S7, S8 S1, S4, S12 S12 S6 S1 S12 S1, S4 S5 S1, S6 S10
25265-71-8 541-02-6 13475-82-6 75-65-0 78-78-4 93-58-3 34590-94-8 108-10-1 623-84-7 84-66-2 106-97-8 109-66-0 71-36-3 29911-28-2 540-97-6 541-05-9 556-67-2 Degradation 141-97-9 64-19-7 4680-24-4 78-79-5 80-62-6
Ethanol Dipropylene glycol mono methyl ether acetate 2-Propanol 3-Methoxy-3-methyl-1-butanol iso-Alkanes Dipropylene glycol Isomers Decamethyl cyclopentasiloxane 2,2,4,6,6-Pentamethylheptane tert-Butanol iso-Pentane Benzoic acid methylester DPGMME Isomers Methylisobutylketone (MIBK) 1,2-Propandiol diacetate Diethyl phthalate (DEP) n-Butane n-Pentane n-Butanol Dipropylene glycol butyl ether isomers Dodecamethyl cyclohexasiloxane Hexamethyl cyclotrisiloxane Octamethyl cyclotetrasiloxane products and others Ethyl acetoacetate Acetic acid Limonen-1,2-epoxid Isoprene Methylmethacrylate
2 h, 4 h and 24 h. In case of the electric evaporator, additional air samples were taken in inactive state (before switching it on) after 2 h and 24 h. During the chamber experiments online determination of NOx, TOC, ozone, ultrafine particles, and selected VOC was carried out.
S3, S5, S6, S10, S11 S7, S8, S10, S11, S12 S1, S5, S6 S7, S8, S9, S10, S11, S12 S9, S10 S2, S8, S12 S7, S9, S10, S11, S12 S4 S2, S3 S1, S2, S5 S1, S2 S1, S2, S5, S8, S10, S11 S10, S11 S11 S1, S3 S5 S3, S5 S1, S6 S5 S10 S9, S10, S11, S12 S10 S11 S1, S2, S3, S5, S6, S7, S8, S10, S12 S6 S2, S10, S11 S1, S4
Maximum emission rate after 1 h of operation (mg/(unit*h) 29 29 28 27 27 27 26 26 25 25 23 20 20 19 18 18 17 16 15 14 14 12 9 9 9 8 8 7 6 5 5 3 3 3 3 3 1 35,532 12,337 5690 4763 3110 2529 647 480 191 168 137 131 115 86 71 68 32 15 11 9 9 3 359 183 9 5 5
S7eS8: These samples were tested in 1 m3-glass chambers according to ISO 16000-9. For the emission chamber test one potpourri was placed in a chamber. S7 was used with the textile bag it came in. The potpourri S8 was filled in a glass petri dish and placed in the chamber. Chamber air was collected on TENAX® TA
Please cite this article in press as: Uhde, E., Schulz, N., Impact of room fragrance products on indoor air quality, Atmospheric Environment (2014), http://dx.doi.org/10.1016/j.atmosenv.2014.11.020
6
E. Uhde, N. Schulz / Atmospheric Environment xxx (2014) 1e11
after 5 h, 24 h and 48 h. In addition, the same samples were retested under the same conditions after 14 d, 15 d and 16 d and after 29 d, 30 d and 31 d. S9eS12: The samples were tested in 1 m3-glass chambers according to ISO 16000-9. The chamber air was collected on TENAX® TA after 5 h, 24 h, 48 h, 72 h and 96 h. The same samples were retested under the same conditions after 14 d, 15 d and 16 d and after 29 d, 30 d and 31 d. In addition, ozone and ultrafine particle were determined during the entire testing time. S14: One candle was placed in a 1 m3-glass chamber. The chamber air was collected on TENAX® TA when the candle was still unlit (5 h, 24 h, and 48 h), after lighting (0.5 h, total burning duration was 4 h) and after extinguishing (0.5 h, 2 h, 24 h). During the entire experiment NOx, TOC, CO, CO2, ozone, and ultrafine particle were monitored online. 3. Results and discussion 3.1. Product information Most fragrance products had a declaration of ingredients printed on the packaging, the detail of that information, however, varied considerably: While all major fragrance substances were listed on some products, several more only showed two or three single substances, and some just listed “fragrance”. The odorless solvents, which can be released in high concentrations, were declared in only one case (ethanol). Many products carried danger symbols like “highly flammable” or “harmful” printed on either the covering box or the bottles/ containers. The information about hazards varied both in presentation as well as in depth: Some packages listed hazard information in detail, but in extreme small-print. Others supplied no or fragmentary information. Directions in regard to possible spilling/ breakage were never given. Disposal information was generally fragmentary. Only the pressurized dispensers and the electrically driven devices had specific guidance for the consumer. To assess the completeness of the substance declaration on the packaging, the results of the chamber tests were reviewed for some products and the declared and non-declared substances were counted. For product S1 only 2 fragrance substances were declared, 38 were found e this resulted in 6 mg/m3 declared substances and 5484 mg/m3 non-declared substances 1 h after test start. For S2 (7 declared/31 non-declared substances; 1042/5870 mg/m3), S3 (3 declared/24 non-declared substances; 505/695 mg/m3), S4 (2 declared/26 non-declared substances; 0/1183 mg/m3), S5 (3 declared/20 non-declared substances; 227/228 mg/m3) and S6 (1 declared/43 non-declared substances; 6088/10,512 mg/m3) similarly low numbers of declared substances were found. 3.2. Emission studies of sprays and diffusers Volatile organic compounds and aldehydes were measured continuously and discontinuously during the tests. The standardized test scheme ensured consistent sampling at certain times (e.g. 1 h or 24 h after activation/test start) to be able to directly compare concentration data for all products at these defined times. For comparison purposes these were converted to unit-specific emission rates. Table 2 shows the identified substances and the maximum observed emission rate for all products after one hour of operation. It must, however, be noted, that the emission behavior of different products varies greatly e a spray-type device may create very high chamber concentrations after few minutes of operation, while a diffuser-type device may reach maximum chamber concentrations after many hours or days.
3.2.1. Sprays The discontinuous usage of any spray device makes standardized testing difficult. The automatic spray unit S5 was set to medium intensity (one spray pulse every 28 min). The concentration in the 3 m3-chamber rose and stabilized after approximately 6 h. Ethanol was the main solvent in this product, with pentanes, 2propanol and acetone being present in low concentrations. Due to the quick evaporation of the solvents the TOC (measured by photo-acoustic monitor) increased rapidly (Fig. 1) and showed a saw-toothed shape caused by intermittent spray operation. The fragrance substance concentrations followed that course, albeit on a much lower level. The spray device also generated a short-lived ultrafine aerosol with each spray pulse. This multimodal aerosol is probably caused by evaporation of spray droplets; the main particle size range of this short-lived aerosol is between 30 and 100 nm. The device generated stable concentrations of all organic compounds over the entire test. When adding ozone after 24 h, the concentrations of several terpenes decreased and SOA formation started. The aerosol concentration in the test chamber increased suddenly, but did not exceed 20,000 particles per cubic centimeter. This aerosol was bimodal with count median diameter (CMD) around 10 and 40 nm, respectively. The test of the room spray S6 started after chamber blank determination with 5 short spray pulses into the chamber through a 6-cm-stainless steel flange, which was subsequently closed. Very high solvent concentrations (ethanol, 2-propanol) were measured in the chamber during the following hours. Limonene concentrations reached 12 mg/m3 after 0.5 h. Linalool, linalyl acetate, and beta-pinene decreased from 4, 3.5 and 2 mg/m3, respectively. The product generated a high aerosol concentration in the chamber, particle number concentrations exceeding 400,000 were reached. Due to the extremely high concentrations of reactive monoterpenes, the product can also form large numbers of SOA particles when ozone is added.
3.2.2. Diffusers Several diffusers/evaporators were strong sources of VOC. The fragrance compounds and especially the carrier liquids/solvents were released with high emission rates. In this regard the electric evaporator S2 was an exception because in deactivated state it only released very low amounts of fragrance substances. When switched on, a strong release of a solvent (dipropylene glycol) and the fragrance substances was observed (Fig. 2). Main emittents were benzyl acetate, malonic acid diethyl ester, linalool, linalyl acetate, 2hexenylpropanoate, and dihydromyrcenol. After ozone addition, both formaldehyde and acetaldehyde became detectable at 20 and 30 mg/m3, respectively. The passive diffuser S1 came with a replaceable fragrance liquid cartridge and was adjustable for the scent strength. After activation and adjustment to medium strength, the device was placed in a 3 m3-emission test chamber. The chamber concentrations (Fig. 3) increased quickly and reached levels above 1 mg/m3 for the two main fragrance substances (benzyl acetate and dihydromyrcenol). More than 60 substances could be identified and quantified in the chamber air, which made the tracing of selected substances by PTRMS difficult. Using (non-specific) ions, the effect of ozone dosage was detectable e on m/z 137 a decrease of signal was observed, while an increase on m/z 57 happened at the same time. Using GC/ MS quantification results, such behavior could also be detected for e.g. dihydromyrcenol, linalool or geranyl acetate. Other main fragrance substances were dimethyl octanol, dimethyl heptenal, alpha-citronellene, eucalyptol, limonene, methoxy benzaldehyde, benzylacetone and decanal.
Please cite this article in press as: Uhde, E., Schulz, N., Impact of room fragrance products on indoor air quality, Atmospheric Environment (2014), http://dx.doi.org/10.1016/j.atmosenv.2014.11.020
E. Uhde, N. Schulz / Atmospheric Environment xxx (2014) 1e11 Start of O3 addition
80 Activation of diffuser
600 Linalool 2-Phenylethanol Benzylacetate alpha-Terpineol
40
400
Particle number concentration
200
20
20000
10000
0
0
0 0
5
10
15
20
Particle number concentration (#/cm³)
TOC (mg/m³)
800 Fragrance substance concentration (μg/m³)
60
TOC ref. Propane
7
25
Time (hours) Fig. 1. Emissions from an automatic spray freshener. Concentration vs. time course over 24 h of normal operation, thereafter addition of ozone for SOA formation estimation.
30000
3.3. Emission studies of potpourri samples The two potpourri samples were mixtures of dried flowers, seedpods, leaves and other plant parts. Both were impregnated with a fruity/vanilla fragrance. The results of the loose potpourri (S8) were surprising because a solvent (dipropylene glycol, 1900 mg/ m3 after two days in a 1 m3-glass chamber with air exchange rate of 0.5 h1) was the main emitting compound. The major fragrance substances (caryophyllene 216 mg/m3, linalyl acetate 204 mg/m3, linalool 105 mg/m3, alpha-methylionon 93 mg/m3) reached moderate concentrations. After 36 days the product was almost odorless, but still showed a chamber concentration of dipropylene glycol
ozone addition Switched on
FMPS Dipropylene glycol Ethylheptanoat 2-Propenyl hexanoate Benzylacetat
35000
until the end of the experiment. For other substances, which are not shown here (piperonal, acetyl pyridine) the time course was very similar.
4000
25000 20000 15000 Start
10000
2000
5000
Concentration of fragrance substances (μg/m³)
Concentrations of ultrafine particles (#/cm³)
According to the manufacturers data sheets some air fresheners are supposed to work for several weeks. In two experiments the release of solvent substances and of some fragrance substances was monitored over the course of one month. To find out which influence the diffuser type has, the same fragrance liquid was filled into a wood-stick diffuser (12 sticks) and another diffuser consisting of a glass jar with a wooden lid (of which the lower part was immersed in the liquid). These very different diffuser types showed surprisingly similar emission behavior in regard to the major solvent substances (Fig. 4). Solvent concentrations peaked on day two and decreased to approximately one-tenth of the initial concentrations within ten days. The volatile fragrance substance ethyl butyrate showed almost identical behavior. The less volatile fragrance compounds of the mixture (e.g. methoxy benzaldehyde, vanilline, ethylvanilline and coumarine) increased in concentration over several days, peaked between day 3 and 5 and decreased slowly
0
0 0
10
20
30
40
50
60
Time (hours) Fig. 2. Emission from an electric (evaporation) air freshener. Concentration vs. time course over 24 h in switched-off-state and 30 h in operation. Ozone addition after 46 h.
Please cite this article in press as: Uhde, E., Schulz, N., Impact of room fragrance products on indoor air quality, Atmospheric Environment (2014), http://dx.doi.org/10.1016/j.atmosenv.2014.11.020
8
E. Uhde, N. Schulz / Atmospheric Environment xxx (2014) 1e11
1600
200000
6000 Dihydromyrcenol Linalool Benzylacetate n-Decanal Geranylacetate
1400 1200
5000
4000
Particles FMPS
1000
3000
PTR-MS: m/z 137 (Terpenes) m/z 57 (Terpenes)
800
2000
600 400
150000
100000
50000
1000
Particle number concentration (#/cm³)
1800
Activation of diffuser
2000
Start O3 addition
relative concentration (a.u.)
Fragrance substance concentration (μg/m³)
2200
200 0
0 0
5
10
15 Time (hours)
20
25
30
Fig. 3. Emissions from a passive diffuser (mean intensity setting). PTR-MS online measurements show decrease of terpene signal (m/z 137) after ozone addition, and formation of ultrafine aerosol.
around 300 mg/m3. The second potpourri sample (S7), which was intended to be used in its fabric bag, showed lower concentrations of solvents (main compound dipropylene glycol mono methyl ether acetate with 276 mg/m3 after two days) and higher fragrance substance concentrations. Here, linalyl acetate (628 mg/m3), linalool (400 mg/m3), alpha-cedrene (288 mg/m3), eucalyptol (84 mg/m3) and geranyl acetate (55 mg/m3) were the major emittents. After 36 days, only low concentrations of the solvent (27 mg/m3) and the fragrance substances linalyl acetate (107 mg/m3) and linalool (103 mg/m3) were detectable. Table 2 includes the identified substances and the maximum observed emission rate for the two potpourri samples after one hour of operation.
cartridge). When the consumer prepares/assembles the product, the fragrance liquid may be spilled. On a carpet, spilling 50 ml of the liquid may lead to a stain of 30e50 cm diameter (for a wood stick diffuser set, a 50 ml bottle can be regarded as best case, while sets with larger bottles (100 ml, 250 ml) are quite common). To estimate the consumer's exposure in case of such spilling, a small amount of two fragrance liquids was poured in glass petri dishes to create a defined surface (1.6*103 m2) and tested in 1 m3-glass chambers under standard conditions (23 C, 50% r.h.) at an air exchange rate of 2 h1. The resulting chamber concentrations of the main solvent components (3-methoxy-3-methyl-1-butanol, decamethyl cyclopentasiloxane and dipropylene glycol monomethylether acetate) were converted to emission rates and used to calculate possible concentrations in a hypothetic 30 m3-room with an air exchange rate of 0.5 h1. A stained/soaked area with a diameter of 34 cm (0.09 m2) was assumed. Fig. 5 shows the results of this calculation. Although still far from the saturation vapor concentration of the substances, the estimated concentrations in the model room are
3.4. Emission studies of accidental spilling of fragrance liquids A number of products consisted of a diffusing appliance (e.g. reed sticks and bottle, wood balls and jar, electric diffuser), and the fragrance liquid (provided either in a bottle or in a special 1300
W ood stick diffuser
Cham ber concentration (μg/m ³)
1200
Chamber concentration (μg/m³)
25000
20000
15000
Ethylbutyrate 4-Methoxy-benzaldehyde Vanilline Coumarin Ethylvanilline
1100 1000 900 800 700 600 500 400 300 200 100 0 0
5
10
15
20
25
30
35
Tim e (days)
10000
Wood sticks: 3-Methoxy-3-methyl-1-butanol Dipropylene glycol monomethyl ether acetate
5000
Wood lid: 3-Methoxy-3-methyl-1-butanol Dipropylene glycol monomethyl ether acetate
0 0
5
10
15
20
25
30
35
Time (days) Fig. 4. Comparison of two different diffuser types (wood top and wood sticks) with the same fragrance liquid in 1 m3-glass chambers (23 C, 50% r.h., air exchange rate 1 h1). Concentration vs. time data of major solvent substances is shown in the main diagram, fragrance substances in the inset diagram.
Please cite this article in press as: Uhde, E., Schulz, N., Impact of room fragrance products on indoor air quality, Atmospheric Environment (2014), http://dx.doi.org/10.1016/j.atmosenv.2014.11.020
E. Uhde, N. Schulz / Atmospheric Environment xxx (2014) 1e11
Calculated solvent concentrations (μg/m³)
80000
9
3-Methoxy-3-methyl-1-butanol Decamethyl cyclopentasiloxane Dipropylene glycol monomethyl ether acetate
70000 60000 50000 40000 30000 20000 10000 0 0
10
20
30
40
50
Time (hours) Fig. 5. Calculated reference room (30 m3) concentration (based on 1 m3 chamber tests) of three fragrance liquid solvents after spilling (free evaporation).
quite high. While a substantial decay can be observed for the most volatile compound, 3-methoxy-3-methyl-1-butanol, the other two solvent substances show very stable concentrations over the entire test period of two days. Accidental spilling of fragrance liquid may therefore expose consumers to high concentrations of odorless solvents over an extended period of time. The concentrations of some fragrance substances could also reach high concentrations under such conditions, although most of them were present in the products in much lower content.
3.5. Emission studies of candles 3.5.1. Test of unlit/lit candles A scented candle will already give off a weak odor in unlit state; therefore, emission of the major fragrance substance can be expected. The tested scented candle released 28 mg/m3 4-methoxybenzaldehyde, 13 mg/m3 gamma-nonalactone, <1 mg/m3 ethylvanilline and 2 mg/m3 gamma-decalactone in unlit state. One-half hour after lighting the chamber concentrations increased to 126, 57, 19 and 22 mg/m3, respectively. During four hours of burning, the chamber concentrations increased further, after extinguishing 356, 141, 30, and 43 mg/m3, respectively, were measured (Fig. 6). The heated candle body continued to release fragrance substances at an increased rate for several hours (chamber concentrations 321, 125, 31, 42 mg/m3 two hours after extinguishing). The identified substances and the maximum observed emission rate for the candle sample can be found in Table 2. The study of several VOCs (especially benzene, benzaldehyde, and benzophenone) from lit candles requires special care because nitrogen oxides generated by the flame can lead to substantial artifact formation with the sorbent Tenax® TA. Carbon based sorbents (graphitized carbon blacks) can be used to quantify such emissions properly. In general, the simultaneous occurrence of benzene with benzaldehyde, benzoic acid, acetophenone and 2,5diphenyl-2,5-cyclohexadiene-1,4-dione indicates Tenax® TA degradation processes and necessitates sampling with less sensitive sorbents (Clausen and Wolkoff, 1997). Unfortunately, carbon based sorbents are less suitable for fragrance substances, for some substances (e.g. linalool, linalyl acetate, dihydromyrcenol)
decomposition during thermal desorption was found for graphitized carbon blacks (Carbotrap®, Carbograph®), the recovery rate being below 50% for each sorbent. Using Carbograph®-packed sorbent tubes the benzene emission rates were below 1 mg/(unit*h) for all tested candles and tea lights. 3.5.2. The influence of the wax type Candles can be produced from various raw materials. Today, most candles currently on the market are made of paraffin, stearin, tallow, or wax. To find out if the base material has any general influence on the emissions of candles, unscented standard tea lights (diameter 38 mm, height 16 mm, aluminum cup) made of paraffin, stearin, tallow and wax are tested under identical conditions to find out if the based material does specifically contribute to the emissions. The non-lit tea lights did not release substances in concentrations above 1 mg/m3. As can be seen in Table 3, all tea lights were weak emission sources after lighting. The emitted substances were non-specific and probably impurities of the base material. Some of them can likely be attributed to Tenax® TA degradation (see 3.5.1). No specific odor could be detected during testing of the candles. 4. Conclusions The presented results show the great variety of substances that are released from room fragrance products. It also became evident during the study that these products can be substantial sources of air pollutants: Besides releasing scent substances in high concentrations, even higher concentrations could be found for some solvents. Considering the possible exposure time, which can reach weeks for the high volume diffusers, the use of such products in homes may impair indoor air quality notably. Especially because of the odorless solvents, which can easily reach mg/m3-levels in rooms, it is close to impossible for the consumer to predict the impact of such products. Even products that appear to be solventfree, like potpourris or the car diffuser, can release solvents in considerable amounts. Both passive and active diffusers can be strong sources, and there is some indication that the composition of the fragrance liquid is more important than the type of diffuser. Should the
Please cite this article in press as: Uhde, E., Schulz, N., Impact of room fragrance products on indoor air quality, Atmospheric Environment (2014), http://dx.doi.org/10.1016/j.atmosenv.2014.11.020
10
E. Uhde, N. Schulz / Atmospheric Environment xxx (2014) 1e11
1,4
1,0
300
0,8 200 0,6 0,4
100
0,2 0,0
Chamber concentration (μg/m³)
NOx concentration (ppm)
1,2
400
Burning time
NO NO2 4-Methoxy-benzaldehyde gamma-Nonalactone Ethylvanilline gamma-Decalactone
0 0
5
10
15
20
25
30
35
40
45
50
55
60
Time (hours) Fig. 6. Concentration of nitrous oxides and several fragrance substances before and during burning of a scented candle in a 1 m3-glass chamber (air exchange rate 1 h1).
References
Table 3 Emissions from different wax types during burning of tea lights. CAS number
Substance
Chamber concentrationsa in mg/m3 (0.5 h after lighting) Paraffin
Stearin
Plant wax
Tallow
50-00-0 75-07-0 64-17-5 67-64-1 109-66-0 64-19-7 71-36-3 142-82-5 108-88-3 111-65-9 111-84-2 100-52-7 100-47-0 98-86-2 541-02-6
Formaldehydeb Acetaldehyde Ethanol Acetone n-Pentane Acetic acid n-Butanol Heptane Toluene Octane Nonane Benzaldehyde Benzonitril Acetophenone Decamethyl cyclopentasiloxane Dodecamethyl cyclohexasiloxane Benzoic acid Heptadecane Octadecane Other iso-Alkanes Other Siloxanes
13 6 n.d. n.d. 23 n.d. 5 n.d. 5 n.d. n.d. 3 n.d. n.d. n.d.
30 7 6 16 n.d. n.d. 12 n.d. 22 n.d. n.d. 5 n.d. 3 4
25 7 8 13 n.d. 16 9 2 9 4 3 5 n.d. 3 n.d.
12 6 7 14 n.d. n.d. 8 n.d. 5 n.d. n.d. 6 1 3 n.d.
n.d.
3
n.d.
n.d.
2 n.d. n.d. n.d. n.d.
n.d. n.d. n.d. n.d. 5
n.d. 2 7 5 n.d.
2 n.d. n.d. n.d. n.d.
540-97-6 65-85-0 629-78-7 593-45-3
a b
One tea light in a 1 m3-glass chamber, air exchange rate 1 h1. Possible interference of NOx on Formaldehyde determination.
consumer decide to refill a certain diffuser with a different liquid (e.g. another brand which may contain more volatile solvents/fragrances), an involuntary exposure to very high concentrations of these substances may occur. As Nazaroff and Weschler (2004) pointed out, the risks of air fresheners can be clearly defined and assessed, the benefits, however, remain very subjective.
Acknowledgments Special thanks are due to Ms Doreen Markewitz and Ms Astrid Schwarz for GC/MS evaluation. Mr Christian Fauck's assistance with PTR-MS operation and Ms Jenny Bartsch's help in experimental work is greatly appreciated.
Bhatia, S.P., McGinty, D., Letizia, C.S., Api, A.M., 2008. Fragrance material review on myrtenol. Food Chem. Toxicol. 46, S237eS240. Clausen, P.A., Wilkins, C.K., Wolkoff, P., Nielsen, G.D., 2001. Chemical and biological evaluation of a reaction mixture of R-(þ)-limonene/ozone: formation of strong airway irritants. Environ. Int. 26, 511e522. Clausen, P.A., Wolkoff, P., 1997. Degradation products of Tenax TA formed during sampling and thermal desorption analysis: indicators of reactive species indoors. Atmos. Environ. 31, 715e725. Coleman, B.K., Lunden, M.M., Destaillats, H., Nazaroff, W.W., 2008. Secondary organic aerosol from ozone-initiated reactions with terpene-rich household products. Atmos. Environ. 42, 8234e8245. EU, 2003. In: Commission, E. (Ed.), Directive 2003/15/EC (7th Amendment to Directive 76/768/EEC, Annex III, Part I). hin, E., Ramalho, O., Kirchner, S., 2008. Size distribution and emission rate Ge measurement of fine and ultrafine particle from indoor human activities. Atmos. Environ. 42, 8341e8352. Huang, Y., Ho, S.S.H., Ho, K.F., Lee, S.C., Gao, Y., Cheng, Y., Chan, C.S., 2011. Characterization of biogenic volatile organic compounds (BVOCs) in cleaning reagents and air fresheners in Hong Kong. Atmos. Environ. 45, 6191e6196. ISO, 2011. 16000: Indoor Air e Part 6: Determination of Volatile Organic Compounds in Indoor and Test Chamber Air by Active Sampling on Tenax TA Sorbent, Thermal Desorption and Gas Chromatography Using MS/FID. Beuth Verlag, Berlin. Jo, W.-K., Lee, J.-H., Kim, M.-K., 2008. Head-space, small-chamber and in-vehicle tests for volatile organic compounds (VOCs) emitted from air fresheners for the Korean market. Chemosphere 70, 1827e1834. Kang, D.H., Choi, D.H., Won, D., Yang, W., Schleibinger, H., David, J., 2012. Household materials as emission sources of naphthalene in Canadian homes and their contribution to indoor air. Atmos. Environ. 50, 79e87. Klaschka, U., 2010. Risk management by labelling 26 fragrances?: Evaluation of Article 10(1) of the seventh Amendment (Guideline 2003/15/EC) of the Cosmetic Directive. Int. J. Hyg. Environ. Health 213, 308e320. Lalko, J., Lapczynski, A., McGinty, D., Bhatia, S., Letizia, C.S., Api, A.M., 2007. Fragrance material review on methyl ionone (mixture of isomers). Food Chem. Toxicol. 45, S300eS307. Lapczynski, A., Bhatia, S.P., Foxenberg, R.J., Letizia, C.S., Api, A.M., 2008. Fragrance material review on geraniol. Food Chem. Toxicol. 46, S160eS170. Nazaroff, W.W., Weschler, C.J., 2004. Cleaning products and air fresheners: exposure to primary and secondary air pollutants. Atmos. Environ. 38, 2841e2865. Nørgaard, A.W., Nøjgaard, J.K., Larsen, K., Sporring, S., Wilkins, C.K., Clausen, P.A., Wolkoff, P., 2006. Secondary limonene endo-ozonide: a major product from gas-phase ozonolysis of -()-limonene at ambient temperature. Atmos. Environ. 40, 3460e3466. ^me, A., D'Anna, B., Rossignol, S., Rio, C., Ustache, A., Fable, S., Nicolle, J., Me Nicolas, M., Leoz, E., Chiappini, L., 2013. The use of a housecleaning product in an indoor environment leading to oxygenated polar compounds and SOA formation: gas and particulate phase chemical characterization. Atmos. Environ. 75, 196e205. Singer, B.C., Coleman, B.K., Destaillats, H., Hodgson, A.T., Lunden, M.M., Weschler, C.J., Nazaroff, W.W., 2006. Indoor secondary pollutants from cleaning product and air freshener use in the presence of ozone. Atmos. Environ. 40, 6696e6710.
Please cite this article in press as: Uhde, E., Schulz, N., Impact of room fragrance products on indoor air quality, Atmospheric Environment (2014), http://dx.doi.org/10.1016/j.atmosenv.2014.11.020
E. Uhde, N. Schulz / Atmospheric Environment xxx (2014) 1e11 Steinemann, A.C., 2009. Fragranced consumer products and undisclosed ingredients. Environ. Impact Assess. Rev. 29, 32e38. Sunderland, P.B., Quintiere, J.G., Tabaka, G.A., Lian, D., Chiu, C.W., 2011. Analysis and measurement of candle flame shapes. Proc. Combust. Inst. 33, 2489e2496. Uhde, E., Salthammer, T., 2007. Impact of reaction products from building materials and furnishings on indoor air qualitydA review of recent advances in indoor chemistry. Atmos. Environ. 41, 3111e3128. VITO, 2008. Exposure and Risk Assessment of Air Fresheners. Study Conducted by the Flemish Institute for Technological Research NV on the Order of the Belgium Federal Public Service Health, Food Chain Safety and Environment, Report 2008/IMS/R222. Weschler, C.J., Shields, H.C., 1997. Potential reactions among indoor pollutants. Atmos. Environ. 31, 3487e3495. Weschler, C.J., Shields, H.C., 1999. Indoor ozone/terpene reactions as a source of indoor particles. Atmos. Environ. 33, 2301e2312.
11
Wilkins, C.K., Wolkoff, P., Clausen, P.A., Hammer, M., Nielsen, G.D., 2003. Upper airway irritation of terpene/ozone oxidation products (TOPS). Dependence on reaction time, relative humidity and initial ozone concentration. Toxicol. Lett. 143, 109e114. Wolkoff, P., Clausen, P.A., Larsen, S.T., Hammer, M., Nielsen, G.D., 2012. Airway effects of repeated exposures to ozone-initiated limonene oxidation products as model of indoor air mixtures. Toxicol. Lett. 209, 166e172. Wolkoff, P., Clausen, P.A., Wilkins, C.K., Hougaard, K.S., Nielsen, G.D., 1999. Formation of strong airway irritants in a model mixture of (þ)-a-pinene/ozone. Atmos. Environ. 33, 693e698. Wolkoff, P., Larsen, S.T., Hammer, M., Kofoed-Sørensen, V., Clausen, P.A., Nielsen, G.D., 2013. Human reference values for acute airway effects of five common ozone-initiated terpene reaction products in indoor air. Toxicol. Lett. 216, 54e64.
Please cite this article in press as: Uhde, E., Schulz, N., Impact of room fragrance products on indoor air quality, Atmospheric Environment (2014), http://dx.doi.org/10.1016/j.atmosenv.2014.11.020
Contents lists available at ScienceDirect
Atmospheric Environment journal homepage: www.elsevier.com/locate/atmosenv
Impact of room fragrance products on indoor air quality Erik Uhde*, Nicole Schulz Fraunhofer Wilhelm-Klauditz-Institut, Bienroder Weg 54 E, 38108 Braunschweig, Germany
h i g h l i g h t s Various room fragrance products were tested in emission test chambers. Many products were strong sources of odorless solvents. More than 100 different fragrance substances were identified and quantified. The long term emission behavior of some products was tested. Accidental spilling of fragrance liquids may cause extreme solvent concentrations.
a r t i c l e i n f o
a b s t r a c t
Article history: Received 9 April 2014 Received in revised form 6 November 2014 Accepted 10 November 2014 Available online xxx
Everyday life can no longer be imagined without fragrances and scented products. For the consumer, countless products exists which are solely or partly intended to give off a certain scent in sufficient concentrations to odorize a complete room. Sprays, diffusers and evaporators, scented candles and automatic devices for the distribution of fragrance liquids are typical examples of such products. If the consumer uses such products, his consent to the release of certain chemicals in his home can be implied, however, he may not know what kind of fragrance substances and solvents will be present in which concentrations. In this study, we determined the volatile emissions of a number of fragrance products in detail. Measurements were carried out under controlled conditions in test chambers. The products were tested in a passive (unused) and an active state, wherever applicable. Following a defined test protocol, the release of volatile organic compounds, ultrafine particles and NOx was monitored for each product. The potential for forming secondary organic aerosols under the influence of ozone was studied, and for a selection of products the long-term emission behavior was assessed. A remarkable variety of fragrance substances was found and more than 100 relevant compounds were identified and quantified. While it is the intended function of such products to release fragrance substances, also considerable amounts of non-odorous solvents and by-products were found to be released from several air fresheners. Emissions rates exceeding 2 mg/(unit*h) were measured for the five most common solvents. © 2014 Published by Elsevier Ltd.
Keywords: Fragrance product Room spray Candle Emission Solvent
1. Introduction Scenting of household products like cleaning agents or detergents with volatile organic substances, specifically terpenes and terpenoids, has been common for many decades. Nowadays, a variety of products is available for the consumer to also perfume the air in rooms and offices. As these products become more and more popular, questions in regard to a possible impairment of indoor air
* Corresponding author. E-mail address: [email protected] (E. Uhde).
quality arise (Nazaroff and Weschler, 2004; Steinemann, 2009). The topic is of specific interest because of the substantial time per day dwellers are exposed to the scent substances (and additives/solvents) in case of home application, and when the source strength of such products is considered. It also has to be taken into account that fragrance substances of other sources may already be present in indoor air due to use of household products (Coleman et al., 2008; Kang et al., 2012; Rossignol et al., 2013; Singer et al., 2006), and room scenting products may increase both the number and the concentrations of such compounds in the indoor environment. Terpenoid substances, aldehydes and lactones are common substances in scenting products, and a majority of the compounds
http://dx.doi.org/10.1016/j.atmosenv.2014.11.020 1352-2310/© 2014 Published by Elsevier Ltd.
Please cite this article in press as: Uhde, E., Schulz, N., Impact of room fragrance products on indoor air quality, Atmospheric Environment (2014), http://dx.doi.org/10.1016/j.atmosenv.2014.11.020
2
E. Uhde, N. Schulz / Atmospheric Environment xxx (2014) 1e11
used are reactive enough to be susceptible to indoor chemistry (Nazaroff and Weschler, 2004; Singer et al., 2006; Uhde and Salthammer, 2007). Certain conditions in the distribution devices (e.g., elevated temperature, or the storage of concentrated liquids under oxygen contact for long periods of time) will also increase the possibility of not only the fragrance substances, but also degradation products being present in the air. The impact of terpene release in indoor air has been studied for a long time. The mono-terpenes as well as their ozone-initiated degradation products have been a research topic for several decades. Early work (Weschler and Shields, 1997, 1999) indicated that ozone-terpene reactions can actually be observed in the indoor environment. Extensive studies were carried out on the physiological effects of terpenes and their decomposition products/intermediates (Clausen et al., 2001; Nørgaard et al., 2006; Wilkins et al., 2003; Wolkoff et al., 2012, 1999, 2013). Allergic properties of numerous fragrance substances are recognized today, but the knowledge about long-term inhalative exposure is still fragmentary. For cosmetics usage of these substances the dermal exposure pathway is considered most important, and data on the sensitizing potential exists, see e.g. Bhatia et al. (2008), Lalko et al. (2007) and Lapczynski et al. (2008). In some countries, such substances are regulated, or the declaration is legally enforced like in e.g. EU (2003). The effectiveness of such risk management efforts is discussed controversially (Klaschka, 2010). Air fresheners and room fragrances, however, are no cosmetic products and the market is therefore mostly unregulated. Unlike fragranced detergents, cleaning products or cosmetics (Huang et al., 2011; Jo et al., 2008; Nazaroff and Weschler, 2004), the air fresheners have no other (primary) function; their sole purpose is to odorize air in a building. Their use therefore represents an intended release of a e mostly unknown e substance mix in an indoor environment (VITO, 2008). In this study a variety of room fragrance products was tested under defined conditions. The released scent substances and solvents were identified, long term emission behavior was measured and possible room concentrations were calculated. The intention of this article is to provide a more detailed overview on substances currently on the market, and on the release characteristics of different types of air fresheners.
2. Methods 2.1. Studied products All tested products were bought in local stores or supplied by manufacturers. The products were commercial items sold in complete packaging including consumer information. Due to the huge variety of products and the countless number of product types, a representative testing is hardly possible. Therefore, products were chosen to give a broad overview over existing technologies and fragrance substances. For the selection of products the following points were considered: 1) Availability (Major brands were preferred to niche products) 2) Different scents (For example, lemon or vanilla scented products are available from most brands, therefore other scents were selected when possible) 3) Different techniques (A selection of passively evaporating products, spray products, electrically driven items and scented candles was considered) Whenever a product had several intensity settings, a mean setting was selected for the test. A short description of the tested products is given in Table 1. 2.1.1. Diffusers and evaporators ‘Diffuser type’ odorizers usually consist of a fragrance liquid/gel, a storage container and some means to control evaporation or diffusion to an evaporation surface. Typical examples are stick diffusers, where several (typically 6e12) wood sticks are placed in a bottle and slowly evaporate the rising fragrance liquid over a time frame of days to weeks. Other constructions with similar function, e.g. with a wooden lid, a membrane or perforated plastic part can be found as well. For small rooms or cars gel-/liquid-filled evaporators are available. Items of each group were tested. Some products require the consumer to fill the diffuser with the fragrance liquid, while others come pre-assembled, with the liquid reservoir already in place. Therefore, the possible impact of spilling was assessed separately.
Table 1 Overview of tested fragrance products. Product
Type
Scent type
Package size
Remarks
S1
Passive diffuser
‘Spring’
5.5 ml
S2 S3
Electric evaporator Wood-stick
‘Exotic mango-orange’ ‘Green tea’
8 ml 100 ml
S4 S5
Car diffuser Automatic spray
‘White’ ‘Cherry magnolia’
18 g 250 ml
S6 S7 S8 S9
Spray Potpourri bag Potpourri Wood-stick
‘Natural’ ‘Lavender’ ‘Vanilla’ ‘Cherry blossom’
50 ml 100 g 100 g 50 ml
S10
Wood top
‘Vanilla’
50 ml
S11 S12
Wood stick Jar þ wood balls
‘Vanilla’ ‘Vanilla’
50 ml 100 ml
Fragrance liquid cartridge activated by inserting, adjustable strength Removable fragrance liquid cartridge 10 sticks need to be inserted into the pre-filled bottle. Activation by removing a plastic strip Pressurized container, timer setting adjustable in 3 steps Pressurized container Potpourri stays in fabric bag Loose potpourri Fragrance liquid needs to be poured in glass container, 12 sticks are placed in separate drill holes of a wood lid. Fragrance liquid needs to be poured in glass container, wood lid with a wooden plug immersed in the liquid. 10 sticks need to be inserted into the pre-filled bottle. Pump spray needs to be applied to wooden balls in a glass dish
S14
Candle
‘Ice cream’
450 g
Please cite this article in press as: Uhde, E., Schulz, N., Impact of room fragrance products on indoor air quality, Atmospheric Environment (2014), http://dx.doi.org/10.1016/j.atmosenv.2014.11.020
E. Uhde, N. Schulz / Atmospheric Environment xxx (2014) 1e11
2.1.2. Spray odorizers Two spray odorizers were tested: an automatic spray mechanism, which scents the room in pre-set timing intervals, and a conventional room spray specifically advertised for use in schools.
3
detector at 360 nm (HPLC/VWD, Agilent Technologies, US) using a 10 cm ACE3 C18 column (Advanced Chromatography Technologies Ltd, GB). 2.6. Determination of nitrogen oxides (for candle tests)
2.1.3. Candles Scented candles/tea lights differed from the other products in two ways e after lighting, they are active heat sources, so the test chamber must be able to remove excess heat and moisture. The candle flame will also generate combustion products, which may interfere with on-line and off-line analysis techniques. Generally, candles can be substantial sources of ultrafine and fine particles in hin et al., 2008). Undisturbed burning the indoor environment (Ge is a prerequisite to reliable results (Sunderland et al., 2011), especially particle emissions and NOx-generation will be affected by unsteadily burning candles. To protect the candle flame from the air flow and allow undisturbed burning in the test chamber, the candles were shielded by a stainless-steel mesh cylinder (30 cm diameter) behind a wind shield in the middle of the test chamber. To find out if the candle base has an influence of emission, unscented tea lights made of pure paraffin, stearin, tallow and wax were tested under the same conditions. 2.2. Test chambers
NO and NO2 were measured by reduced pressure chemiluminescence (CLD) based online detector (Horiba, APNA-370) with a flow rate of 0.8 l/min and a time resolution of 60 s. This device uses the excited state of NO2* for the quantitation of both NO (directly) and NO2 (indirectly) by cross flow modulation. 2.7. Determination of ultrafine particles The concentrations of ultrafine particles during the chamber tests were measured using a TSI Model 3091 Fast Mobility Particle Sizer (FMPS), which covers a size range of 5.6e560 nm. The time resolution was 1 spectrum per second. 2.8. Generation and determination of ozone Ozone was generated from air using an UV-Ozone generator. The concentration in the chamber was determined by a Horiba APOA 370 ozone analyzer (ND-UV).
Products were tested in emission test chambers of two sizes, the smallest chambers used were 1 m3-glass chambers (for tea lights, candles, car diffuser), and the other items were tested in 3 m3stainless steel chambers. All chambers fulfilled the requirements of ISO 16000-9. The tests were carried out under standard conditions (23 C, 50% r.h.) with an air exchange rate of 0.5 h1. For special experiments (evaporation tests to assess the consequences of spillage, candle tests), higher air exchange rates were used.
2.9. Determination of emitting substances by PTR/MS
2.3. Determination of volatile organic compounds
2.10. Determination of TOC/CO2/CO
Volatile organic compounds were determined according to ISO16000-6 (ISO, 2011) by adsorption on Tenax® TA filled stainlesssteel tubes, thermal desorption at 290 C (TD 100, Markes International, UK) and subsequent GC/MS analysis (GC 7890, MS 5975C, Agilent Technologies, USA) using a J&W Scientific DB-5MS column (L ¼ 60 m; I.D. ¼ 0.25 mm; film ¼ 0.25 mm). Calibration was carried out with internal and external standards, whenever possible against pure reference compounds (some substances, e.g. piperonal, are regulated as drug precursors and are therefore not available). The air sampling volume was between 0.5 l and 4 l, depending on the expected chamber concentrations.
For general control of the experimental progress a photoacoustic detector (Innova 1412), equipped with a TOC filter was utilized. Filters for CO2 and CO were also installed and used to monitor the combustion of the candles.
2.4. Determination of benzene Especially during candle testing or secondary organic aerosol (SOA) generation experiments, reactive substances like NOx or ozone can cause benzene artifact formation on Tenax® TA-filled sampling tubes. Therefore, additional air samples were taken on Carbotrap®-filled sampling tubes. These tubes were thermally desorbed and benzene was quantified using GC/MS with a method identical to that of the VOC determination.
A proton-transfer reaction mass spectrometer (High-sensitivity quadrupole PTR-MS, Ionicon Analytik, Austria) was used to monitor the experiments in scan-mode with a time resolution of 180 s. Especially during ozone addition this data was used to follow the concentration increase/drop of newly formed/decomposed substances.
2.11. Test protocol Various fragrance products were tested in this study using a defined test protocol. 1) Blank values for all studied parameters were measured in the test chambers. 2) Each product was unpacked and put into the test chamber without activating/lighting it. (For purely passive emitting products that started to give off odor immediately after unpacking, the procedure continued with step 5.) 3) The emissions of the product were measured. 4) The product was activated/lit in the chamber. 5) The emissions were measured for several hours to days (depending on the operation/burning duration, see below). 6) For some of the products the measurement was continued for several weeks.
2.5. Determination of volatile aldehydes Aldehydes were analyzed according to ISO 16000-3 using silica gel cartridges (Supelco) that contained 2,4-dinitrophenylhydrazine (DNPH). The cartridges were solvent extracted with acetonitrile (ACN) and the target compounds were measured by high performance liquid chromatography coupled with a variable wavelength
S1eS6: The samples were tested in 3 m3-chambers according to ISO 16000-9. Each test was carried out with one product per chamber. All samples give off odor immediately after unpacking except the electric evaporator (S2; this product had to be activated by plugging it into a wall socket). During the tests chamber air was collected on sampling tubes filled with Tenax® TA after 0.5 h, 1 h,
Please cite this article in press as: Uhde, E., Schulz, N., Impact of room fragrance products on indoor air quality, Atmospheric Environment (2014), http://dx.doi.org/10.1016/j.atmosenv.2014.11.020
4
E. Uhde, N. Schulz / Atmospheric Environment xxx (2014) 1e11
Table 2 List of identified substances and their maximum observed emission rate one hour after application start. Samples in bold print showed the highest emission rates. Substances marked with * were not available as pure reference, quantification was done against toluene. Substances marked with are listed in EU (2003) as potential allergens. CAS-no.
Substance
Identified in sample
5989-27-5 140-11-4 18479-58-8 78-70-6 115-95-7 99-85-4 123-35-3 18172-67-3 470-82-6 105-53-3 105-54-4 628-44-4 5392-40-5 78-93-3 27939-60-2 80-56-8 123-68-2 87-44-5 5655-61-8 89-78-1 99-87-6 464-49-3 93-92-5 7785-53-7 105-87-3 513-86-0 141-12-8 10385-78-1 39255-32-8 2550-26-7 106-72-9 586-62-9 112-31-2 79-20-9 106-68-3 495-61-4 123-11-5 110-41-8 1191-16-8 123-38-6 106-30-9 469-61-4 31499-72-6 642-41-9 540-18-1 79-92-5 562-74-3 104-61-0 103-95-7 17699-14-8 502-61-4 142-19-8 12542-30-2 91-64-5 13877-91-3 85-91-6 106-25-2 124-13-0 112-44-7 60-12-8 106-23-0 120-57-0 3779-61-1 106-24-1 51115-64-1 110-93-0 124-19-6 123-92-2 111-87-5 498-15-7 77-54-3
Limonene Benzyl acetate Dihydromyrcenol Linalool Linalyl acetate gamma-Terpinene Myrcene beta-Pinene Eucalyptol Malonic acid diethyl ester Butyric acid ethyl ester 2-Methyl-2-octanol Citral*, 2-Butanone Trivertal alpha-Pinene Hexanoic acid 2-propenyl ester Caryophyllene Bornyl acetate Menthol p-Cymene Camphor alpha-Methylbenzyl acetate alpha-Terpineol Geranyl acetate 3-Hydroxy-2-butanone Neryl acetate Borneol Ethyl-2-methylpentanoate Benzylacetone 2,6-Dimethyl-5-heptenal Terpinolene Decanal Methyl acetate 3-Octanone beta-Bisabolene 4-Methoxy-benzaldehyde 2-Methyl-undecanal Prenyl acetate Propanal Heptanoic acid ethyl ester alpha-Cedrene alpha-Dihydro-ionon 2-Methylbutyl acetate Butyric acid pentyl ester Camphene Terpinen-4-ol gamma-Nonalactone Cyclamal alpha-Cubebene Farnesene Isomers Heptanoic acid 2-propenyl ester Dihydrodicyclopentadienyl acrylate Coumarine Ocimene Dimethylanthranilate Nerol Octanal Undecanal 2-Phenylethanol Citronellal Piperonal* beta-Ocimene trans-Geraniol 2-Methyl-butyl butyrate 2-Methyl-2-hepten-6-one Nonanal iso-Pentyl acetate n-Octanol 3-Carene Cedryl acetate
S1, S2, S3, S4, S5, S6, S7, S9, S10, S12 S1, S2, S3, S5, S9 S1, S2, S3, S4, S8, S9, S12 S1, S2, S3, S4, S5, S6, S7, S8, S9, S12 S1, S2, S3, S4, S6, S7, S8, S12 S1, S6 S1, S2, S3, S4, S5, S6, S7, S8, S12 S1, S3, S4, S5, S6, S7, S9, S10, S12 S1, S2, S3, S4, S7 S2 S1, S2, S10, S11 S1 S3, S4, S6 S2 S2, S3, S4, S5 S1, S2, S3, S4, S5, S6, S9, S12 S2 S6, S8, S12 S1, S2, S4, S12 S1, S4, S6 S1, S6 S1, S2, S6, S7 S2, S5, S9 S5, S6, S7, S8 S1, S2, S4, S6, S7, S8, S12 S10, S11 S1, S4, S6, S7, S12 S6, S7 S2, S5, S9 S1, S2 S1, S2, S5, S9 S1, S3, S6 S1, S4, S6 S10, S11 S1, S6 S6 S1, S2, S6, S7, S8, S10, S14 S1, S2 S2 S11 S2 S7 S8 S2 S2 S1, S3, S4, S6, S12 S6, S7 S8, S11, S12 S2, S5 S6 S6 S2 S1, S4 S1, S2, S6, S7, S8, S11 S1, S12 S6 S2, S5, S6 S1, S2, S4, S6 S6 S1, S3, S5, S9 S6 S1, S8, S11, S12 S12 S2, S3, S4, S5 S1 S1, S2 S1, S2, S5, S2 S6 S1, S4, S5, S7, S12 S8
Maximum emission rate after 1 h of operation (mg/(unit*h) 9132 3920 3155 2994 2711 2688 1679 1391 992 992 774 723 711 584 584 552 507 488 452 429 381 326 257 254 254 244 230 222 191 191 189 188 174 156 147 128 122 119 119 114 108 107 98 90 90 87 86 76 68 68 66 66 66 65 57 56 50 50 47 45 39 39 38 36 36 35 30 30 30 30 30
Please cite this article in press as: Uhde, E., Schulz, N., Impact of room fragrance products on indoor air quality, Atmospheric Environment (2014), http://dx.doi.org/10.1016/j.atmosenv.2014.11.020
E. Uhde, N. Schulz / Atmospheric Environment xxx (2014) 1e11
5
Table 2 (continued ) CAS-no.
Substance
Identified in sample
6753-98-6 20085-93-2 24851-98-7 127-51-5 2111-75-3 763-32-6 3691-12-1 100-51-6 3681-71-8 3681-71-8 121-33-5 14073-97-3 17909-77-2 464-49-3 100-52-7 97-53-0 50373-59-6 3691-11-0 4674-50-4 560-32-7 66-25-1 706-14-9 514-51-2 111-71-7 43052-87-5 110-62-3 103-95-7 121-32-4 2436-90-0 529-16-8 104-55-2 6874-35-7 473-13-2 13851-11-1 106-72-9 928-96-1 98-86-2 Solvents 64-17-5 88917-22-0 67-63-0 56539-66-3
alpha-Caryophyllene Seychellene Methyldihydrojasmonate alpha-Isomethyl-ionone Perilla aldehyde 3-Methyl-3-buten-1-ol alpha-Guaiene Benzylalcohol cis-3-Hexenyl acetate cis-3-Hexenyl acetate Vanilline Menthone Alpha-sinensal Camphor Benzaldehyde Eugenol Lavandulol acetate alpha-Bulnesene Nootkatone alpha-Patchoulene Hexanal gamma-Decalactone beta-Patchoulene Heptanal alpha-Damascone Pentanal Cyclamal Ethylvanilline beta-Citronellene Santene Cinnamaldehyde alpha-Citronellene alpha-Selinene Fenchyl acetate 2,6-Dimethyl-5-heptenal cis-3-Hexen-1-ol Acetophenone
S6 S8, S12 S3, S8, S9 S4, S5, S9 S6 S10, S11 S7, S8, S12 S1, S8 S3, S9 S9 S6, S7, S8 S1, S2, S4, S6 S6 S7 S1, S2, S5, S6, S8 S8 S6 S12 S6 S8, S12 S1, S4, S5, S6 S14 S12 S6 S1, S2, S4 S3 S1, S2, S5 S7, S8 S1, S4, S12 S12 S6 S1 S12 S1, S4 S5 S1, S6 S10
25265-71-8 541-02-6 13475-82-6 75-65-0 78-78-4 93-58-3 34590-94-8 108-10-1 623-84-7 84-66-2 106-97-8 109-66-0 71-36-3 29911-28-2 540-97-6 541-05-9 556-67-2 Degradation 141-97-9 64-19-7 4680-24-4 78-79-5 80-62-6
Ethanol Dipropylene glycol mono methyl ether acetate 2-Propanol 3-Methoxy-3-methyl-1-butanol iso-Alkanes Dipropylene glycol Isomers Decamethyl cyclopentasiloxane 2,2,4,6,6-Pentamethylheptane tert-Butanol iso-Pentane Benzoic acid methylester DPGMME Isomers Methylisobutylketone (MIBK) 1,2-Propandiol diacetate Diethyl phthalate (DEP) n-Butane n-Pentane n-Butanol Dipropylene glycol butyl ether isomers Dodecamethyl cyclohexasiloxane Hexamethyl cyclotrisiloxane Octamethyl cyclotetrasiloxane products and others Ethyl acetoacetate Acetic acid Limonen-1,2-epoxid Isoprene Methylmethacrylate
2 h, 4 h and 24 h. In case of the electric evaporator, additional air samples were taken in inactive state (before switching it on) after 2 h and 24 h. During the chamber experiments online determination of NOx, TOC, ozone, ultrafine particles, and selected VOC was carried out.
S3, S5, S6, S10, S11 S7, S8, S10, S11, S12 S1, S5, S6 S7, S8, S9, S10, S11, S12 S9, S10 S2, S8, S12 S7, S9, S10, S11, S12 S4 S2, S3 S1, S2, S5 S1, S2 S1, S2, S5, S8, S10, S11 S10, S11 S11 S1, S3 S5 S3, S5 S1, S6 S5 S10 S9, S10, S11, S12 S10 S11 S1, S2, S3, S5, S6, S7, S8, S10, S12 S6 S2, S10, S11 S1, S4
Maximum emission rate after 1 h of operation (mg/(unit*h) 29 29 28 27 27 27 26 26 25 25 23 20 20 19 18 18 17 16 15 14 14 12 9 9 9 8 8 7 6 5 5 3 3 3 3 3 1 35,532 12,337 5690 4763 3110 2529 647 480 191 168 137 131 115 86 71 68 32 15 11 9 9 3 359 183 9 5 5
S7eS8: These samples were tested in 1 m3-glass chambers according to ISO 16000-9. For the emission chamber test one potpourri was placed in a chamber. S7 was used with the textile bag it came in. The potpourri S8 was filled in a glass petri dish and placed in the chamber. Chamber air was collected on TENAX® TA
Please cite this article in press as: Uhde, E., Schulz, N., Impact of room fragrance products on indoor air quality, Atmospheric Environment (2014), http://dx.doi.org/10.1016/j.atmosenv.2014.11.020
6
E. Uhde, N. Schulz / Atmospheric Environment xxx (2014) 1e11
after 5 h, 24 h and 48 h. In addition, the same samples were retested under the same conditions after 14 d, 15 d and 16 d and after 29 d, 30 d and 31 d. S9eS12: The samples were tested in 1 m3-glass chambers according to ISO 16000-9. The chamber air was collected on TENAX® TA after 5 h, 24 h, 48 h, 72 h and 96 h. The same samples were retested under the same conditions after 14 d, 15 d and 16 d and after 29 d, 30 d and 31 d. In addition, ozone and ultrafine particle were determined during the entire testing time. S14: One candle was placed in a 1 m3-glass chamber. The chamber air was collected on TENAX® TA when the candle was still unlit (5 h, 24 h, and 48 h), after lighting (0.5 h, total burning duration was 4 h) and after extinguishing (0.5 h, 2 h, 24 h). During the entire experiment NOx, TOC, CO, CO2, ozone, and ultrafine particle were monitored online. 3. Results and discussion 3.1. Product information Most fragrance products had a declaration of ingredients printed on the packaging, the detail of that information, however, varied considerably: While all major fragrance substances were listed on some products, several more only showed two or three single substances, and some just listed “fragrance”. The odorless solvents, which can be released in high concentrations, were declared in only one case (ethanol). Many products carried danger symbols like “highly flammable” or “harmful” printed on either the covering box or the bottles/ containers. The information about hazards varied both in presentation as well as in depth: Some packages listed hazard information in detail, but in extreme small-print. Others supplied no or fragmentary information. Directions in regard to possible spilling/ breakage were never given. Disposal information was generally fragmentary. Only the pressurized dispensers and the electrically driven devices had specific guidance for the consumer. To assess the completeness of the substance declaration on the packaging, the results of the chamber tests were reviewed for some products and the declared and non-declared substances were counted. For product S1 only 2 fragrance substances were declared, 38 were found e this resulted in 6 mg/m3 declared substances and 5484 mg/m3 non-declared substances 1 h after test start. For S2 (7 declared/31 non-declared substances; 1042/5870 mg/m3), S3 (3 declared/24 non-declared substances; 505/695 mg/m3), S4 (2 declared/26 non-declared substances; 0/1183 mg/m3), S5 (3 declared/20 non-declared substances; 227/228 mg/m3) and S6 (1 declared/43 non-declared substances; 6088/10,512 mg/m3) similarly low numbers of declared substances were found. 3.2. Emission studies of sprays and diffusers Volatile organic compounds and aldehydes were measured continuously and discontinuously during the tests. The standardized test scheme ensured consistent sampling at certain times (e.g. 1 h or 24 h after activation/test start) to be able to directly compare concentration data for all products at these defined times. For comparison purposes these were converted to unit-specific emission rates. Table 2 shows the identified substances and the maximum observed emission rate for all products after one hour of operation. It must, however, be noted, that the emission behavior of different products varies greatly e a spray-type device may create very high chamber concentrations after few minutes of operation, while a diffuser-type device may reach maximum chamber concentrations after many hours or days.
3.2.1. Sprays The discontinuous usage of any spray device makes standardized testing difficult. The automatic spray unit S5 was set to medium intensity (one spray pulse every 28 min). The concentration in the 3 m3-chamber rose and stabilized after approximately 6 h. Ethanol was the main solvent in this product, with pentanes, 2propanol and acetone being present in low concentrations. Due to the quick evaporation of the solvents the TOC (measured by photo-acoustic monitor) increased rapidly (Fig. 1) and showed a saw-toothed shape caused by intermittent spray operation. The fragrance substance concentrations followed that course, albeit on a much lower level. The spray device also generated a short-lived ultrafine aerosol with each spray pulse. This multimodal aerosol is probably caused by evaporation of spray droplets; the main particle size range of this short-lived aerosol is between 30 and 100 nm. The device generated stable concentrations of all organic compounds over the entire test. When adding ozone after 24 h, the concentrations of several terpenes decreased and SOA formation started. The aerosol concentration in the test chamber increased suddenly, but did not exceed 20,000 particles per cubic centimeter. This aerosol was bimodal with count median diameter (CMD) around 10 and 40 nm, respectively. The test of the room spray S6 started after chamber blank determination with 5 short spray pulses into the chamber through a 6-cm-stainless steel flange, which was subsequently closed. Very high solvent concentrations (ethanol, 2-propanol) were measured in the chamber during the following hours. Limonene concentrations reached 12 mg/m3 after 0.5 h. Linalool, linalyl acetate, and beta-pinene decreased from 4, 3.5 and 2 mg/m3, respectively. The product generated a high aerosol concentration in the chamber, particle number concentrations exceeding 400,000 were reached. Due to the extremely high concentrations of reactive monoterpenes, the product can also form large numbers of SOA particles when ozone is added.
3.2.2. Diffusers Several diffusers/evaporators were strong sources of VOC. The fragrance compounds and especially the carrier liquids/solvents were released with high emission rates. In this regard the electric evaporator S2 was an exception because in deactivated state it only released very low amounts of fragrance substances. When switched on, a strong release of a solvent (dipropylene glycol) and the fragrance substances was observed (Fig. 2). Main emittents were benzyl acetate, malonic acid diethyl ester, linalool, linalyl acetate, 2hexenylpropanoate, and dihydromyrcenol. After ozone addition, both formaldehyde and acetaldehyde became detectable at 20 and 30 mg/m3, respectively. The passive diffuser S1 came with a replaceable fragrance liquid cartridge and was adjustable for the scent strength. After activation and adjustment to medium strength, the device was placed in a 3 m3-emission test chamber. The chamber concentrations (Fig. 3) increased quickly and reached levels above 1 mg/m3 for the two main fragrance substances (benzyl acetate and dihydromyrcenol). More than 60 substances could be identified and quantified in the chamber air, which made the tracing of selected substances by PTRMS difficult. Using (non-specific) ions, the effect of ozone dosage was detectable e on m/z 137 a decrease of signal was observed, while an increase on m/z 57 happened at the same time. Using GC/ MS quantification results, such behavior could also be detected for e.g. dihydromyrcenol, linalool or geranyl acetate. Other main fragrance substances were dimethyl octanol, dimethyl heptenal, alpha-citronellene, eucalyptol, limonene, methoxy benzaldehyde, benzylacetone and decanal.
Please cite this article in press as: Uhde, E., Schulz, N., Impact of room fragrance products on indoor air quality, Atmospheric Environment (2014), http://dx.doi.org/10.1016/j.atmosenv.2014.11.020
E. Uhde, N. Schulz / Atmospheric Environment xxx (2014) 1e11 Start of O3 addition
80 Activation of diffuser
600 Linalool 2-Phenylethanol Benzylacetate alpha-Terpineol
40
400
Particle number concentration
200
20
20000
10000
0
0
0 0
5
10
15
20
Particle number concentration (#/cm³)
TOC (mg/m³)
800 Fragrance substance concentration (μg/m³)
60
TOC ref. Propane
7
25
Time (hours) Fig. 1. Emissions from an automatic spray freshener. Concentration vs. time course over 24 h of normal operation, thereafter addition of ozone for SOA formation estimation.
30000
3.3. Emission studies of potpourri samples The two potpourri samples were mixtures of dried flowers, seedpods, leaves and other plant parts. Both were impregnated with a fruity/vanilla fragrance. The results of the loose potpourri (S8) were surprising because a solvent (dipropylene glycol, 1900 mg/ m3 after two days in a 1 m3-glass chamber with air exchange rate of 0.5 h1) was the main emitting compound. The major fragrance substances (caryophyllene 216 mg/m3, linalyl acetate 204 mg/m3, linalool 105 mg/m3, alpha-methylionon 93 mg/m3) reached moderate concentrations. After 36 days the product was almost odorless, but still showed a chamber concentration of dipropylene glycol
ozone addition Switched on
FMPS Dipropylene glycol Ethylheptanoat 2-Propenyl hexanoate Benzylacetat
35000
until the end of the experiment. For other substances, which are not shown here (piperonal, acetyl pyridine) the time course was very similar.
4000
25000 20000 15000 Start
10000
2000
5000
Concentration of fragrance substances (μg/m³)
Concentrations of ultrafine particles (#/cm³)
According to the manufacturers data sheets some air fresheners are supposed to work for several weeks. In two experiments the release of solvent substances and of some fragrance substances was monitored over the course of one month. To find out which influence the diffuser type has, the same fragrance liquid was filled into a wood-stick diffuser (12 sticks) and another diffuser consisting of a glass jar with a wooden lid (of which the lower part was immersed in the liquid). These very different diffuser types showed surprisingly similar emission behavior in regard to the major solvent substances (Fig. 4). Solvent concentrations peaked on day two and decreased to approximately one-tenth of the initial concentrations within ten days. The volatile fragrance substance ethyl butyrate showed almost identical behavior. The less volatile fragrance compounds of the mixture (e.g. methoxy benzaldehyde, vanilline, ethylvanilline and coumarine) increased in concentration over several days, peaked between day 3 and 5 and decreased slowly
0
0 0
10
20
30
40
50
60
Time (hours) Fig. 2. Emission from an electric (evaporation) air freshener. Concentration vs. time course over 24 h in switched-off-state and 30 h in operation. Ozone addition after 46 h.
Please cite this article in press as: Uhde, E., Schulz, N., Impact of room fragrance products on indoor air quality, Atmospheric Environment (2014), http://dx.doi.org/10.1016/j.atmosenv.2014.11.020
8
E. Uhde, N. Schulz / Atmospheric Environment xxx (2014) 1e11
1600
200000
6000 Dihydromyrcenol Linalool Benzylacetate n-Decanal Geranylacetate
1400 1200
5000
4000
Particles FMPS
1000
3000
PTR-MS: m/z 137 (Terpenes) m/z 57 (Terpenes)
800
2000
600 400
150000
100000
50000
1000
Particle number concentration (#/cm³)
1800
Activation of diffuser
2000
Start O3 addition
relative concentration (a.u.)
Fragrance substance concentration (μg/m³)
2200
200 0
0 0
5
10
15 Time (hours)
20
25
30
Fig. 3. Emissions from a passive diffuser (mean intensity setting). PTR-MS online measurements show decrease of terpene signal (m/z 137) after ozone addition, and formation of ultrafine aerosol.
around 300 mg/m3. The second potpourri sample (S7), which was intended to be used in its fabric bag, showed lower concentrations of solvents (main compound dipropylene glycol mono methyl ether acetate with 276 mg/m3 after two days) and higher fragrance substance concentrations. Here, linalyl acetate (628 mg/m3), linalool (400 mg/m3), alpha-cedrene (288 mg/m3), eucalyptol (84 mg/m3) and geranyl acetate (55 mg/m3) were the major emittents. After 36 days, only low concentrations of the solvent (27 mg/m3) and the fragrance substances linalyl acetate (107 mg/m3) and linalool (103 mg/m3) were detectable. Table 2 includes the identified substances and the maximum observed emission rate for the two potpourri samples after one hour of operation.
cartridge). When the consumer prepares/assembles the product, the fragrance liquid may be spilled. On a carpet, spilling 50 ml of the liquid may lead to a stain of 30e50 cm diameter (for a wood stick diffuser set, a 50 ml bottle can be regarded as best case, while sets with larger bottles (100 ml, 250 ml) are quite common). To estimate the consumer's exposure in case of such spilling, a small amount of two fragrance liquids was poured in glass petri dishes to create a defined surface (1.6*103 m2) and tested in 1 m3-glass chambers under standard conditions (23 C, 50% r.h.) at an air exchange rate of 2 h1. The resulting chamber concentrations of the main solvent components (3-methoxy-3-methyl-1-butanol, decamethyl cyclopentasiloxane and dipropylene glycol monomethylether acetate) were converted to emission rates and used to calculate possible concentrations in a hypothetic 30 m3-room with an air exchange rate of 0.5 h1. A stained/soaked area with a diameter of 34 cm (0.09 m2) was assumed. Fig. 5 shows the results of this calculation. Although still far from the saturation vapor concentration of the substances, the estimated concentrations in the model room are
3.4. Emission studies of accidental spilling of fragrance liquids A number of products consisted of a diffusing appliance (e.g. reed sticks and bottle, wood balls and jar, electric diffuser), and the fragrance liquid (provided either in a bottle or in a special 1300
W ood stick diffuser
Cham ber concentration (μg/m ³)
1200
Chamber concentration (μg/m³)
25000
20000
15000
Ethylbutyrate 4-Methoxy-benzaldehyde Vanilline Coumarin Ethylvanilline
1100 1000 900 800 700 600 500 400 300 200 100 0 0
5
10
15
20
25
30
35
Tim e (days)
10000
Wood sticks: 3-Methoxy-3-methyl-1-butanol Dipropylene glycol monomethyl ether acetate
5000
Wood lid: 3-Methoxy-3-methyl-1-butanol Dipropylene glycol monomethyl ether acetate
0 0
5
10
15
20
25
30
35
Time (days) Fig. 4. Comparison of two different diffuser types (wood top and wood sticks) with the same fragrance liquid in 1 m3-glass chambers (23 C, 50% r.h., air exchange rate 1 h1). Concentration vs. time data of major solvent substances is shown in the main diagram, fragrance substances in the inset diagram.
Please cite this article in press as: Uhde, E., Schulz, N., Impact of room fragrance products on indoor air quality, Atmospheric Environment (2014), http://dx.doi.org/10.1016/j.atmosenv.2014.11.020
E. Uhde, N. Schulz / Atmospheric Environment xxx (2014) 1e11
Calculated solvent concentrations (μg/m³)
80000
9
3-Methoxy-3-methyl-1-butanol Decamethyl cyclopentasiloxane Dipropylene glycol monomethyl ether acetate
70000 60000 50000 40000 30000 20000 10000 0 0
10
20
30
40
50
Time (hours) Fig. 5. Calculated reference room (30 m3) concentration (based on 1 m3 chamber tests) of three fragrance liquid solvents after spilling (free evaporation).
quite high. While a substantial decay can be observed for the most volatile compound, 3-methoxy-3-methyl-1-butanol, the other two solvent substances show very stable concentrations over the entire test period of two days. Accidental spilling of fragrance liquid may therefore expose consumers to high concentrations of odorless solvents over an extended period of time. The concentrations of some fragrance substances could also reach high concentrations under such conditions, although most of them were present in the products in much lower content.
3.5. Emission studies of candles 3.5.1. Test of unlit/lit candles A scented candle will already give off a weak odor in unlit state; therefore, emission of the major fragrance substance can be expected. The tested scented candle released 28 mg/m3 4-methoxybenzaldehyde, 13 mg/m3 gamma-nonalactone, <1 mg/m3 ethylvanilline and 2 mg/m3 gamma-decalactone in unlit state. One-half hour after lighting the chamber concentrations increased to 126, 57, 19 and 22 mg/m3, respectively. During four hours of burning, the chamber concentrations increased further, after extinguishing 356, 141, 30, and 43 mg/m3, respectively, were measured (Fig. 6). The heated candle body continued to release fragrance substances at an increased rate for several hours (chamber concentrations 321, 125, 31, 42 mg/m3 two hours after extinguishing). The identified substances and the maximum observed emission rate for the candle sample can be found in Table 2. The study of several VOCs (especially benzene, benzaldehyde, and benzophenone) from lit candles requires special care because nitrogen oxides generated by the flame can lead to substantial artifact formation with the sorbent Tenax® TA. Carbon based sorbents (graphitized carbon blacks) can be used to quantify such emissions properly. In general, the simultaneous occurrence of benzene with benzaldehyde, benzoic acid, acetophenone and 2,5diphenyl-2,5-cyclohexadiene-1,4-dione indicates Tenax® TA degradation processes and necessitates sampling with less sensitive sorbents (Clausen and Wolkoff, 1997). Unfortunately, carbon based sorbents are less suitable for fragrance substances, for some substances (e.g. linalool, linalyl acetate, dihydromyrcenol)
decomposition during thermal desorption was found for graphitized carbon blacks (Carbotrap®, Carbograph®), the recovery rate being below 50% for each sorbent. Using Carbograph®-packed sorbent tubes the benzene emission rates were below 1 mg/(unit*h) for all tested candles and tea lights. 3.5.2. The influence of the wax type Candles can be produced from various raw materials. Today, most candles currently on the market are made of paraffin, stearin, tallow, or wax. To find out if the base material has any general influence on the emissions of candles, unscented standard tea lights (diameter 38 mm, height 16 mm, aluminum cup) made of paraffin, stearin, tallow and wax are tested under identical conditions to find out if the based material does specifically contribute to the emissions. The non-lit tea lights did not release substances in concentrations above 1 mg/m3. As can be seen in Table 3, all tea lights were weak emission sources after lighting. The emitted substances were non-specific and probably impurities of the base material. Some of them can likely be attributed to Tenax® TA degradation (see 3.5.1). No specific odor could be detected during testing of the candles. 4. Conclusions The presented results show the great variety of substances that are released from room fragrance products. It also became evident during the study that these products can be substantial sources of air pollutants: Besides releasing scent substances in high concentrations, even higher concentrations could be found for some solvents. Considering the possible exposure time, which can reach weeks for the high volume diffusers, the use of such products in homes may impair indoor air quality notably. Especially because of the odorless solvents, which can easily reach mg/m3-levels in rooms, it is close to impossible for the consumer to predict the impact of such products. Even products that appear to be solventfree, like potpourris or the car diffuser, can release solvents in considerable amounts. Both passive and active diffusers can be strong sources, and there is some indication that the composition of the fragrance liquid is more important than the type of diffuser. Should the
Please cite this article in press as: Uhde, E., Schulz, N., Impact of room fragrance products on indoor air quality, Atmospheric Environment (2014), http://dx.doi.org/10.1016/j.atmosenv.2014.11.020
10
E. Uhde, N. Schulz / Atmospheric Environment xxx (2014) 1e11
1,4
1,0
300
0,8 200 0,6 0,4
100
0,2 0,0
Chamber concentration (μg/m³)
NOx concentration (ppm)
1,2
400
Burning time
NO NO2 4-Methoxy-benzaldehyde gamma-Nonalactone Ethylvanilline gamma-Decalactone
0 0
5
10
15
20
25
30
35
40
45
50
55
60
Time (hours) Fig. 6. Concentration of nitrous oxides and several fragrance substances before and during burning of a scented candle in a 1 m3-glass chamber (air exchange rate 1 h1).
References
Table 3 Emissions from different wax types during burning of tea lights. CAS number
Substance
Chamber concentrationsa in mg/m3 (0.5 h after lighting) Paraffin
Stearin
Plant wax
Tallow
50-00-0 75-07-0 64-17-5 67-64-1 109-66-0 64-19-7 71-36-3 142-82-5 108-88-3 111-65-9 111-84-2 100-52-7 100-47-0 98-86-2 541-02-6
Formaldehydeb Acetaldehyde Ethanol Acetone n-Pentane Acetic acid n-Butanol Heptane Toluene Octane Nonane Benzaldehyde Benzonitril Acetophenone Decamethyl cyclopentasiloxane Dodecamethyl cyclohexasiloxane Benzoic acid Heptadecane Octadecane Other iso-Alkanes Other Siloxanes
13 6 n.d. n.d. 23 n.d. 5 n.d. 5 n.d. n.d. 3 n.d. n.d. n.d.
30 7 6 16 n.d. n.d. 12 n.d. 22 n.d. n.d. 5 n.d. 3 4
25 7 8 13 n.d. 16 9 2 9 4 3 5 n.d. 3 n.d.
12 6 7 14 n.d. n.d. 8 n.d. 5 n.d. n.d. 6 1 3 n.d.
n.d.
3
n.d.
n.d.
2 n.d. n.d. n.d. n.d.
n.d. n.d. n.d. n.d. 5
n.d. 2 7 5 n.d.
2 n.d. n.d. n.d. n.d.
540-97-6 65-85-0 629-78-7 593-45-3
a b
One tea light in a 1 m3-glass chamber, air exchange rate 1 h1. Possible interference of NOx on Formaldehyde determination.
consumer decide to refill a certain diffuser with a different liquid (e.g. another brand which may contain more volatile solvents/fragrances), an involuntary exposure to very high concentrations of these substances may occur. As Nazaroff and Weschler (2004) pointed out, the risks of air fresheners can be clearly defined and assessed, the benefits, however, remain very subjective.
Acknowledgments Special thanks are due to Ms Doreen Markewitz and Ms Astrid Schwarz for GC/MS evaluation. Mr Christian Fauck's assistance with PTR-MS operation and Ms Jenny Bartsch's help in experimental work is greatly appreciated.
Bhatia, S.P., McGinty, D., Letizia, C.S., Api, A.M., 2008. Fragrance material review on myrtenol. Food Chem. Toxicol. 46, S237eS240. Clausen, P.A., Wilkins, C.K., Wolkoff, P., Nielsen, G.D., 2001. Chemical and biological evaluation of a reaction mixture of R-(þ)-limonene/ozone: formation of strong airway irritants. Environ. Int. 26, 511e522. Clausen, P.A., Wolkoff, P., 1997. Degradation products of Tenax TA formed during sampling and thermal desorption analysis: indicators of reactive species indoors. Atmos. Environ. 31, 715e725. Coleman, B.K., Lunden, M.M., Destaillats, H., Nazaroff, W.W., 2008. Secondary organic aerosol from ozone-initiated reactions with terpene-rich household products. Atmos. Environ. 42, 8234e8245. EU, 2003. In: Commission, E. (Ed.), Directive 2003/15/EC (7th Amendment to Directive 76/768/EEC, Annex III, Part I). hin, E., Ramalho, O., Kirchner, S., 2008. Size distribution and emission rate Ge measurement of fine and ultrafine particle from indoor human activities. Atmos. Environ. 42, 8341e8352. Huang, Y., Ho, S.S.H., Ho, K.F., Lee, S.C., Gao, Y., Cheng, Y., Chan, C.S., 2011. Characterization of biogenic volatile organic compounds (BVOCs) in cleaning reagents and air fresheners in Hong Kong. Atmos. Environ. 45, 6191e6196. ISO, 2011. 16000: Indoor Air e Part 6: Determination of Volatile Organic Compounds in Indoor and Test Chamber Air by Active Sampling on Tenax TA Sorbent, Thermal Desorption and Gas Chromatography Using MS/FID. Beuth Verlag, Berlin. Jo, W.-K., Lee, J.-H., Kim, M.-K., 2008. Head-space, small-chamber and in-vehicle tests for volatile organic compounds (VOCs) emitted from air fresheners for the Korean market. Chemosphere 70, 1827e1834. Kang, D.H., Choi, D.H., Won, D., Yang, W., Schleibinger, H., David, J., 2012. Household materials as emission sources of naphthalene in Canadian homes and their contribution to indoor air. Atmos. Environ. 50, 79e87. Klaschka, U., 2010. Risk management by labelling 26 fragrances?: Evaluation of Article 10(1) of the seventh Amendment (Guideline 2003/15/EC) of the Cosmetic Directive. Int. J. Hyg. Environ. Health 213, 308e320. Lalko, J., Lapczynski, A., McGinty, D., Bhatia, S., Letizia, C.S., Api, A.M., 2007. Fragrance material review on methyl ionone (mixture of isomers). Food Chem. Toxicol. 45, S300eS307. Lapczynski, A., Bhatia, S.P., Foxenberg, R.J., Letizia, C.S., Api, A.M., 2008. Fragrance material review on geraniol. Food Chem. Toxicol. 46, S160eS170. Nazaroff, W.W., Weschler, C.J., 2004. Cleaning products and air fresheners: exposure to primary and secondary air pollutants. Atmos. Environ. 38, 2841e2865. Nørgaard, A.W., Nøjgaard, J.K., Larsen, K., Sporring, S., Wilkins, C.K., Clausen, P.A., Wolkoff, P., 2006. Secondary limonene endo-ozonide: a major product from gas-phase ozonolysis of -()-limonene at ambient temperature. Atmos. Environ. 40, 3460e3466. ^me, A., D'Anna, B., Rossignol, S., Rio, C., Ustache, A., Fable, S., Nicolle, J., Me Nicolas, M., Leoz, E., Chiappini, L., 2013. The use of a housecleaning product in an indoor environment leading to oxygenated polar compounds and SOA formation: gas and particulate phase chemical characterization. Atmos. Environ. 75, 196e205. Singer, B.C., Coleman, B.K., Destaillats, H., Hodgson, A.T., Lunden, M.M., Weschler, C.J., Nazaroff, W.W., 2006. Indoor secondary pollutants from cleaning product and air freshener use in the presence of ozone. Atmos. Environ. 40, 6696e6710.
Please cite this article in press as: Uhde, E., Schulz, N., Impact of room fragrance products on indoor air quality, Atmospheric Environment (2014), http://dx.doi.org/10.1016/j.atmosenv.2014.11.020
E. Uhde, N. Schulz / Atmospheric Environment xxx (2014) 1e11 Steinemann, A.C., 2009. Fragranced consumer products and undisclosed ingredients. Environ. Impact Assess. Rev. 29, 32e38. Sunderland, P.B., Quintiere, J.G., Tabaka, G.A., Lian, D., Chiu, C.W., 2011. Analysis and measurement of candle flame shapes. Proc. Combust. Inst. 33, 2489e2496. Uhde, E., Salthammer, T., 2007. Impact of reaction products from building materials and furnishings on indoor air qualitydA review of recent advances in indoor chemistry. Atmos. Environ. 41, 3111e3128. VITO, 2008. Exposure and Risk Assessment of Air Fresheners. Study Conducted by the Flemish Institute for Technological Research NV on the Order of the Belgium Federal Public Service Health, Food Chain Safety and Environment, Report 2008/IMS/R222. Weschler, C.J., Shields, H.C., 1997. Potential reactions among indoor pollutants. Atmos. Environ. 31, 3487e3495. Weschler, C.J., Shields, H.C., 1999. Indoor ozone/terpene reactions as a source of indoor particles. Atmos. Environ. 33, 2301e2312.
11
Wilkins, C.K., Wolkoff, P., Clausen, P.A., Hammer, M., Nielsen, G.D., 2003. Upper airway irritation of terpene/ozone oxidation products (TOPS). Dependence on reaction time, relative humidity and initial ozone concentration. Toxicol. Lett. 143, 109e114. Wolkoff, P., Clausen, P.A., Larsen, S.T., Hammer, M., Nielsen, G.D., 2012. Airway effects of repeated exposures to ozone-initiated limonene oxidation products as model of indoor air mixtures. Toxicol. Lett. 209, 166e172. Wolkoff, P., Clausen, P.A., Wilkins, C.K., Hougaard, K.S., Nielsen, G.D., 1999. Formation of strong airway irritants in a model mixture of (þ)-a-pinene/ozone. Atmos. Environ. 33, 693e698. Wolkoff, P., Larsen, S.T., Hammer, M., Kofoed-Sørensen, V., Clausen, P.A., Nielsen, G.D., 2013. Human reference values for acute airway effects of five common ozone-initiated terpene reaction products in indoor air. Toxicol. Lett. 216, 54e64.
Please cite this article in press as: Uhde, E., Schulz, N., Impact of room fragrance products on indoor air quality, Atmospheric Environment (2014), http://dx.doi.org/10.1016/j.atmosenv.2014.11.020