1 Human body odour: source and textiles

Natural fibres are sought for manufacture to many types of products, including apparel worn next to the skin, notwithstanding the dominance of synthetic fibres/filaments in these and other applications. An issue of increasing interest with this product group is understanding and better managing the development, retention, and removal of volatile organic compounds (VOCs) present in worn/used fabrics. The principal reasons for this are environmental, the personal environment of individuals (i.e. to be odour-free) and broad environmental issues (e.g. sustainability in production and use of fibre-based products, reduced use of energy and water in cleaning). This review of recent published papers and related grey literature integrating different elements of the topic is thus timely.

The origin of human body odour has been explained in several publications (e.g. [1,2,3]) physiological secretions (sweat, sebum) and epithelial cells shed from the skin surface along with microbiological flora of the skin, are absorbed and/or fixed by adjacent fabrics; interactions among these leading to development of VOCs many of which are adsorbed/released over time. The presence of VOCs on or in fibres/textiles and patterns of adsorption and release can be determined instrumentally (proton transfer reaction-mass spectrometry (PTR-MS), solid-phase micro-extraction followed by gas chromatography–mass spectrometry (SPME-GC/MS), gas chromatography–mass spectrometry/olfactometry (GC–MS/O)/GC–O), Time-resolved (TR) spectrometry), as well as sensorially (trained human panels, consumer groups). Complementary information is provided by each, instrumental methods allowing identification of the presence of chemical constituents, and humans determining the minimum detection levels, ranking of intensity, and acceptability or not.

Differences attributable to fibre type and fabric structure, and malodour and apparel cleaning practices are considered. Although minimising development of malodour in fibres/textiles by application of antimicrobial treatments, and masking by application of fragrance or oils are both possible, these topics have been excluded.

2 Differences attributable to fibre type and fabric structure

Published and un-published research on adsorption and release of odour volatiles has focussed on fibres and fabrics common for next-to-skin products (e.g. cotton, wool, polyester, polyamide (nylon)). Table 1 provides key features of garments, fabrics, and yarns; Table 1a, where the source of the VOCs was the human body, and Table 1b, where a selection of VOCs from those known to be present in human body odour was made by the investigator, some as a matrix and some as single VOCs. (Note that hundreds of volatiles have been identified as constituents of human body odour and relevant to the human skin [4].) What is clear from the table is that interest in odour and textiles has continued for more than 2 decades, with differences in the fibre/fabric model used (e.g. garments such as socks and t shirts worn, swatches of fabrics of known properties stitched to the underarm of a t shirt and worn, standard fabrics (e.g. [5, 6]) disaggregated to yarn bundles and exposed to volatiles under laboratory conditions). These differences contribute to complexity in comparing results from the various investigations. Use of standard fabrics more recently is facilitating comparison of findings and replication of investigations. Differences in volatiles selected, periods of incubation, temperature, and in methods of detection are also evident.

Table 1 Differences in odour intensity attributable to fibre type
Full size table

Viscose, part of the group of cellulosic-based man-made fibres [7] (around 8% of man-made fibres in 2017 [8]) is expected to increase for apparel and other applications given the present trend [9] and that its source is renewable. Preliminary findings of our current work on viscose suggests, in general, viscose and cotton exhibit similar responses to VOCs, but that there are individual VOCs where viscose has a higher adsorption with the same release (i.e. phenol), or where viscose has a higher release when showing the same adsorption capability as cotton (i.e. 1-octanol) (Richter et al. 2019, personal communication). Blends of different fibre types are also being investigated. A recent report on blending wool with polyester [10] showed an optimum blend which would minimise undesirable odour typical of 100% polyester (20% wool, 80% polyester), thus suggesting the approach could be useful in optimisation. No research on VOCs adsorbed by or emitted from fabrics made from recycled materials has been identified notwithstanding interest in this and the ‘new’ forms of fibres/textiles.

Table 1a shows the natural fibres wool and cotton, have been commonly compared with synthetic fibres polyester and polyamide, and although varied conditions under which the comparisons were made, the order of intensity of odour (least to most) is remarkably consistent: natural fibres, especially wool, exhibited the least intense odour. Table 1b, where investigators selected VOCs (Table 1b), fibre-related results are similar. Thus, there is evidence supporting wool fabrics being considered low odiferous: the approximate decreasing order of odour intensity being wool < cotton < viscose < linen < polyester/polyamide. The mechanisms involved are not fully understood partly because of confounding factors such as fabric structure (affecting absorption of moisture and drying behaviour; thermal properties) and the way in which elements of the skin microbiome (i.e. bacteria, archaea, fungi) may be hosted and fixed to fabrics. This highlights the importance of fabric structure.

Table 2 summarises the few investigations in which yarns and fabric structures have been the focus. This limited interest in fabric structure is not surprising given the challenge of securing test fabrics varying only in fibre content: use of standard fabrics is a useful option. However, given several fabric structural and performance properties are linked to odour (e.g. thickness, mass per unit area, sett, moisture absorption, thermal resistance), omission of fabric structural description in some investigations inhibits clarity in understanding (e.g. [3]).

Table 2 Differences in odour intensity attributable to fabric structure
Full size table

3 Malodour and apparel cleaning practices

3.1 Efficacy of cleaning, and trends in reduced wash temperature and water volume

Table 3 provides detail on a range of specifications for domestic clothes washers used in Western parts of Europe, in the Americas, and in Australia and New Zealand. The table shows differences in requirements across different geographic regions, based on two groups of standards IEC/EN—horizontal-axis washing machines (Europe) and ANSI/AHAM—vertical-axis washing machines (the Americas). Differences exist in water temperature (e.g. in western Europe, the water temperature for machine washing is typically 60 °C +) and in determination of performance. For example, the standard for cleaning performance of domestic washing machines sold in Australasia requires compliance with change to soiling/stain only, determined instrumentally [22].

Table 3 Specifications of principal standards for domestic clothes washers [22,23,24]
Full size table

Any in-progress changes in specification, design and manufacture of machines for domestic washing have direct relevance to the efficacy of the cleaning process, including whether or not removal of odour volatiles (and microorganisms) forms part of what is measured. Manufacturers of washing machines continue efforts to design products which perform with reduced energy and water (i.e. lower water temperature, less water), and need to take account of these additional performance measures.

The extent to which cleaning practices are effective in removing volatiles has been explored, related to both product type and fibre. One study on worn socks (74% cotton, 19% polyester, 5% nylon, 2% elastane) and t shirts (100% polyester) were washed at 20 °C, with effective reduction in VOC concentrations on socks when both dry and damp, but malodour was reduced by 25–98% with the t shirt. The authors intimated some reduction was related to evaporation of these volatiles, and that there was a potential link with bacteria known to be present in the axillary region [25]. With socks (polyester) as a model and two VOCs (dimethyl disulfide, dimethyl trisulfide), Denawaka et al. showed washing at a higher temperature (50 °C) was more effective at removing these two VOCs than washing at a lower temperature (20 °C). Use of softeners in washing to enhance handle and reduce static electricity of textiles is common, however, softeners have been shown to increase odour on polyester garments over several cycles of use, washing, and airing [26]. The presence and removal of sebum from human contact with textiles has recently also been investigated with promising results of a new test method [27].

3.2 Duration of use/wear and perception of need to wash an item

The frequency of cleaning and perception of the need to clean are also of interest (e.g. [28,29,30]). There are at least two reasons; first, to reduce use of both energy and water, and second, to reduce fibre loss from the product during cleaning (resulting in reduced fibre released to waterways and an extension to product life). The frequency of washing clothing/textiles after use is determined largely by product category (i.e. outerwear/jackets, knitwear, next-to-skin, underwear). In this context, understanding effects of cleaning and the need for cleaning, life cycle analyses (LCA) and life cycle impact assessment (LCIA) have focussed on fibre types.

Cotton, polyester, nylon, acrylic, and elastane have been compared in terms of production to base material (i.e. extraction of raw material, processing to a textile (knitting, weaving, finishing) but excluding manufacture and use of consumer products, and some aspects of discarded textiles [31]. Ranking of fibres from that investigation from least to most environmental impact was acrylic < polyester < elastane < nylon < cotton. A more recent comparison of environmental impacts of different fibres highlighted the connection between the fibre content of a product and use and care, indicating distortionary effects of omitting product lifespan, quality, and function [32]. Consumers from several countries (e.g. Norway, UK, Netherlands, Sweden, Finland) reportedly use fibre type and properties common in product groups in deciding on the need to wash. The lifespan of selected garments across these countries was estimated as (years (range)): socks 2.6 (1.8–3.6), underwear 3.1 (2.4–4.4), jeans 3.5 (2.5–4.3), t shirts 4.6 (3.3–6.8), sweaters 6.0 (3.7–10.8), jackets/blazers 6.8 (4.0–11.5) [32]. Also considering differences in end use, an Australian-based investigation of environmental effects of the textile supply chain used the LCA of three items of apparel differing in fibre content (cotton knit shirt, polyester knit shirt, wool sweater) and considered climate change as the impact category with all phases of manufacture and use (washing, drying, ironing; end of life disposal) included. In terms of use, the order of impact least to most was wool < cotton < polyester, largely attributable to the type of garment [33]. Change in behaviour of end users (e.g. water and energy in washing and drying), changed transport (air, sea freight), and inclusion of recycling in the LCA model, along with avoidance of selected products, were additions to the model to better estimate the environmental impact of the textile supply chain from extraction to end of life.

4 Conclusions and recommendations

This review highlights key issues in recent research on odour and next-to-skin textiles. There are four issues:

  1. 1.

    In terms of fibre-based differences, the behaviour of common natural fibres such as wool and cotton in relation to odour is reasonably well established. However, more investigation is needed on a wider range of natural fibres, on man-made cellulosics (e.g. viscose), and on up-cycled/recycled sources. Establishing blends by selecting fibres known to minimise odour seems feasible using instrumental and sensory methods to determine the presence, quality, and intensity of odour.

  2. 2.

    Effects of fabric structure on patterns of adsorption and release of odour volatiles continues to be poorly understood. This warrants further investigation using standard fabrics (and yarns), as this facilitates replication of experimental work, aids understanding by the international scientific community, and provides underpinning for legitimate product marketing.

  3. 3.

    Both fibre type and fabric structure influence product end function, with function characterised by different exposures, duration of use, frequency of cleaning, and decisions in relation to disposal/re-use. Odour volatiles are inextricably linked to each of these, and need to be understood if fibre-based claims of environmental impact are to be defended.

  4. 4.

    Given the international trend to conserve water and energy leading to domestic washing at a lower temperature and with less water, evidence of effects of these changes on efficacy of cleaning will be required. Further, the criteria for determining this efficacy warrants inclusion of evidence of VOC reduction.