Fibre release in sequential machine washes
PES-ss shed the most fibres in the first wash (6.3 × 106 kg−1), followed by the technical t-shirts PES-ts1 (3.1 × 106 kg−1), PES-ts2 (1.4 × 106 kg−1) and PA-ts (5.2 × 105 kg−1). The fleece textiles shed 1.9 × 105 kg−1 (PES-fnap) and 1.8 × 105 kg−1 (PES-fap). The lowest emissions in the first wash were from PAN-je (1.0 × 105 kg−1). As described in Table 2, the PES-ss fabric consists of two surface layers which are (1) woven fabric from 1.1dtex continuous filaments yarns (with elastane) and (2) a “fleece”-type fabric with piles generated by ripping of textured 0.8dtex filaments. Fibre release from the woven fabric is expected only from the seam and cut edges, whereas the breaking of individual piles from the fleece surface (or remains from the manufacturing) is an additional source of released fibres.
The emissions of all textiles decreased in the sequential washes, with emission values in the fifth wash falling between 1.9 × 104 and 1.9 × 105 kg−1. Figure 3 shows the normalised emission values for the synthetic textiles tested. The initial emission values decreased in the following washes and the value was less than 20 % in the third wash except for PAN-je that had the normalised emission value between 0.22 and 0.47 in the second to fifth washes. The decreasing trends in the sequential washings have been earlier reported by Cesa et al. (2020), Belzagui et al. (2019), De Falco et al. (2019), Zambrano et al. (2019), Carney Almroth et al. (2018), Pirc et al. (2016) and Napper and Thompson (2016).
The length distributions of fibres released in the 5th washes are shown in Fig. 4. The lengths varied greatly, from the shortest fibre length of 30 μm to the longest one of 14,000 μm. PES-fnap had the highest mean fibre length (3500 μm), followed by PES-fap (1400 μm). The mean fibre lengths of PES-ss and the technical t-shirts varied between 360 and 550 μm. The means were somewhat higher than the corresponding medians, with the exception of PAN-je which had the mean fibre length of 1000 μm but a clearly smaller median (360 μm). The high mean of PAN-je is accounted for a few exceptionally long acryl fibres (the longest 14,000 μm). In terms of length distribution range, technical sports t-shirts and PES-ss had a narrower range than the other textiles tested. It was noted that PAN-je had the most variation not only in length, but also in width. These variations in fibre lengths between the textiles can most likely be explained by their textile and yarn characteristics.
The fibre length from PAN-je ranged between 60 and 14,000 μm, and it had a mean length of 1000 μm. PAN-je is a knitted product, manufactured from bean-shaped staple fibres that typically have a length between 60 and 100 mm and a diameter of 18–28 μm. The fibres are pulled out of the fabric construction and broken during washing due to the mechanical stress.
The fibres that were released from the technical t-shirts PES-ts and PA-ts manufactured from continuous filaments were mostly short, with a mean length of 480 μm for PES-ts1, 550 μm for PES-ts2 and 500 μm for PA-ts. The fibre lengths for the t-shirts ranged between 30 and 1500 μm. The origin of these fibres was most likely both the cut edges of the fabric and the seams that contained damaged fibres produced during manufacturing when a needle pierced the fabric.
The fibre lengths from the polar fleece textiles might be explained by the looped pile heights of the fabric which is about half the length of the fibres forming the looped piles. The height of the piles on the one-sided fleece of the PES-ss fabric was about 1000 μm, and the released fibres had a mean length of 360 μm. The double-sided fleece PES-fap had piles with heights of 1000 μm and 2000 μm, with the released fibres having a mean length of 1400 μm. The pile heights of the double-sided fleece PES-fnap were 900 μm and 800 μm, but the measured mean fibre length was unexpectedly high (3500 μm). These long fibres could be explained by a poor embedment of the piles in the fabric which enables longer fibres to be pulled out.
Previous studies have focused either the complete textile products or textile samples cut from synthetic fabrics. Pirc et al. (2016) studied the washing of a double-sided fleece blanket with a pile height of 1000 μm. They reported higher fibre lengths for the released fibres when compared with the present study: a mean fibre length of 5300 μm and a wide fibre length distribution with lengths ranging from 300 to 25,000 μm which refers to slightly longer fibres than in the present study. Hernandez et al. (2017) studied the fibre length distributions of two textiles samples cut from knitted (single jersey and interlock) polyester fabrics during five simulated home sequential washings. In comparison with the present study, they used two different polyester fabrics knitted from yarns made from staple fibres. The cut edges were folded and sealed with a thread. Their fibre size distributions in fifth washing were similar to those of technical t-shirts in the present study.
The pore size of the filter used for the filtration of the washing effluent in the present study (0.7 μm) was close to those (0.2 to 0.7 μm) used by Corami et al. (2020) and Hernandez et al. (2017) but smaller than those (20 to 200 μm) used by Cesa et al. (2020), Belzagui et al. (2019), Kelly et al. (2019), De Falco et al. (2018), Hartline et al. (2016), Napper and Thompson (2016) and Pirc et al. (2016). In terms of fibre number per fabric mass, the smaller the filter pore size was, the higher the fibre emission values were reported. The highest fibre emissions were from the present study, Corami et al. (2020) and De Falco et al. (2018). All the mentioned studies had emission values reaching millions of fibres per textile mass (kg).
Fibre release in sequential tumble dryings
As with the sequential washing, fibre emissions showed a decreasing trend also in sequential drying (Fig. 5), with some of the fibre emissions reaching a plateau within the five sequential dryings. The first tumble-drying released the most fibres in all the textiles: the highest emissions were from PES-fnap (1700 mg/kg), PES-fap (690 mg/kg) and PES-ss (340 mg/kg). All these three textiles were fleece or contained fleece that is the fabric with raised, looped piles. They were followed by the fluffy knitted PAN-je (140 mg/kg). The fabrics releasing the least fibres in the first drying were the PA-ts (10 mg/kg) and PES-ts (22 mg/kg, with PES-ts1 and PES-ts2 textiles tumble dried together). The double-sided fleece textiles PES-fnap and PES-fap continued to release considerably more fibres than other textiles, and without stabilizing their emissions, throughout the sequential dryings. The technical t-shirts PA-ts and PES-ts continued to release the least amount of fibres throughout the sequential dryings.
The high fibre emission values of PES-nfap, PES-fap and PES-ss during tumble drying (Fig. 5) result likely from the loosely knitted structure of the fabrics, i.e. that the fabrics have on their surface both raised fibre-ends and raised, looped piles. These kinds of loose fibres are susceptible to being broken off from the textile surface, for example due to mechanical stress from washing (Zambrano et al. 2019). In general, loosely knitted fibres are also easily entangled together to form pills on the textile surface (Hussain et al. 2008), for example during washing-drying cycles (Okubayashi and Bechtold 2005; Okubayashi et al. 2005). The formed pills can then be worn away from the textile surface due to mechanical stress. While fibre loss from pilling has been discussed in relation to machine wash (Napper and Thompson 2016), it should be also considered in relation to tumble drying.
PAN-je, manufactured from staple fibres, also has many raised fibre-ends (no piles) that should be susceptible to fibre loss during drying. This shows in the emission values of PAN-je in that they are placed between the other textiles. Finally, the technical t-shirts PA-ts and PES-ts, manufactured from continuous filaments, have “less hairy” and more firmly knitted textile surface without piles or multiple fibre-ends sticking out of the surface. This likely results in less fibre release than the other textile samples during tumble drying. The t-shirts only have fibre-ends sticking-out from their seams and from the cut edge of the fabric from which fibres might have been released during drying, as well as during washing.
For comparison of fibre emissions from washings and dryings, the fibre numbers from machine washes were converted to the fibre mass by multiplying the emission numbers by the fibre linear density (dtex values in Table 2) and the mean length of released fibres. The machine wash-to-tumble drying ratio of the fibres released from the fifth treatment is presented in Fig. 6. The ratios of polyester and polyamide technical t-shirts are higher than 1, which indicates the fibre release being larger in machine wash than in tumble drying. The other tested textiles had the ratio lower than 1, which refers the tumble drying to be the dominating treatment in fibre release.
Fibre release can be accounted for two different mechanisms: (1) the detachment of already loose fibres from the fabric surface, produced for example from the manufacturing process of the fabric, and (2) the breaking-off of fibres from the fabric itself. Tumble drying the textiles was accompanied by more mechanical stress than when washing the textiles in water. Therefore, the breaking-off of fibres from the fabric itself during tumble drying is expected to be higher than during washing. Some fibres released during washing were already loose on the fabric surface or weakened from previous treatments to break off. The t-shirts were more firmly knitted fabrics with fibre-ends sticking-out only in the seam and the cut edges of the fabric. Also, loose fibres might have been generated during the stitching of the seam. The loose fibres, broken off from continuous filaments by a needle, may have been released during washing with declining fibre numbers during the sequential wash-and-drying cycles.
Pirc et al. (2016) is the only study where fibre release during tumble drying has been investigated. They treated six 100% polyester fleece blankets in ten sequential wash-and-drying cycles. Like in the present study, they observed a decreasing trend in the fibre emissions. The fibre emissions in the first drying (washing done with liquid detergent) was 200 mg/kg which then decreased to 61 mg/kg in the fifth drying, and finally reaching 34 mg/kg in the tenth drying. For the first five dryings, they reported lower fibre emission values for fleece fabrics than the present study. The differences between these two studies are likely due to differences in fabric characteristics and drying conditions. The mesh size of the lint filter was larger (180 μm) in Pirc et al. (2016) than in the present study (60 μm) but also the drying program and the shape of the drying drum may have affected the fibre release. More research should be done on tumble drying to better understand the effects of different fabric characteristics and drying conditions on the fibre emissions during tumble drying.
It must be noted that the fibres released in the tumble drying will not be led straight into a wastewater treatment plant, unlike the fibres released in machine wash or washer dryers. Instead, the fibres are trapped in the tumble dryer’s lint filter that is cleaned by hand, with a vacuum cleaner or washing with water. Thus, it is up to the consumer whether the trapped fibres end up into the trash or sewage water. O’Brien et al. (2020) have shown that tumble drying releases microplastic fibres into indoor air, though the concentrations were low (1.6 ± 1.8 fibres/m3) in their study. It is worth mentioning that the residence time of large particles/fibres (much larger than 10 μm) is typically short in the air and therefore they deposit in the vicinity of their emission source.
Efficiency of fibre traps
The collection efficiencies of two commercial fibre traps are presented in Fig. 7. The Guppyfriend clearly reduces fibre emissions during washing, with 39% reductions observed with fibre numbers. It should be noted that the collection efficiency is likely higher for the loosely knitted textiles that release longer fibres than the firmly knitted textiles studied here. The Guppyfriend did not prevent the stains of blackcurrant juice and cream cheese from being removed from the textile. Thus, the Guppyfriend can be used to reduce fibre emissions without compromising cleaning efficiency.
The Cora Ball trapped 10% of the short polyester fibres studied here (Fig. 7). McIlwraith et al. (2019) has also investigated the collection efficiency of Cora Ball. They found that Cora Ball mitigated fibre emissions by 26% on a basis of fibre number. The difference between the two studies can be explained by the different length of studied fibre (longer than 100 μm in McIlwraith et al. 2019), since the trapping efficiency of Cora Ball increases with the size (length and lint) of fibres.
The overall effectiveness of a fibre trap to mitigate microplastic pollution is greatly impacted by their user-friendliness. Herweyers et al. (2020) examined the perceptions and attitudes of customers toward products that mitigate fibre emissions from domestic washings. Based on questionnaires (n = 411) and user observations with interviews (n = 8), they found that the effectiveness and durability of a product followed by its usability, with special focus on convenience, were the most important factors in convincing people to use the product. They concluded that to keep people using the product for a long time, the product should be simple-to-use and user friendly. With both Cora Ball and Guppyfriend, the consumer needs to both clean the traps and dispose the collected fibres by hand. However, compared with other commercial fibre traps like the external Lint LUV-R filter, Cora Ball and Guppyfriend do not need to be installed or similarly maintained. There is also no danger of blockage when a filter is not changed often enough. Overall, there is a lower threshold for a consumer to purchase a Cora Ball or a Guppyfriend than an external filter to combat fibre emissions during washing.