How Flushable Wet Wipes Are Causing Sewer Blockages – and An Approach to Prevent That

Scientic publications and newsfeeds recently focused on ushable wet wipes and their role in sewage system blockages. It is stated that although products are marked as ushable, they do not disintegrate after being disposed of via the toilet. In this work it is shown that wetlaid hydroentangled wet wipes lose their initially good dispersive properties during their storage in wet condition. This behaviour is found for both, wet wipes from industrial production and wipes produced on pilot facilities. It is demonstrated that the deterioration of the wipe’s ability to disintegrate during wet storage is linked to the type of cellulosic bres used. Only wipes made from a combination of band-shaped viscose bres and unbleached softwood pulp bres were preserving good disintegration during wet storage. The results are also suggesting to add tests after dened times of wet storage when assessing the ushability of wet wipes.


Introduction
Wet wipes, with their broad range of applications, are part of the modern life and can be found in many households in a form like it is depicted in Fig. 1 (left). Especially the demand on personal hygiene products has increased and is forecasted to rise by 8.0 % p.a. in the next years (Mango, 2018).
Convenience and hygiene require the easy disposal of these wipes, preferably right after use. This creates a market for biodegradable and ushable products. Both terms must be discussed separately as wipes that are marked ushable are not necessarily biodegradable. Therefore, the term "truly ushable" was introduced (Phillip Mango, 2004) for wipes that, after successfully being disposed via sewage, are also able to degrade in nature.
Biodegradability thereby is de ned as a breakdown mechanism that creates simple substances in a biological way (Blackburn, 2005;Polman et al., 2021). In the list of bres used in nonwovens (Mango, 2018) two groups of polymers, cellulose and PLA (polylactic acid), ful l this criterion (Okada, 2002), where the biodegradability of PLA in a marine environment is in discussion recently (Narancic et al., 2018). Over 50% of the raw materials used in wet wipes are cellulosic natural biopolymers such as regenerated bres (Rayon and Lyocell), wood pulp and cotton (Mango, 2018), that are all able to biodegrade in an aquatic environment (Zambrano et al., 2019). For a product to be fully biodegradable it must entirely consist of these materials, however synthetic bres can be found in wet wipes labelled as ushable (Ó Briain et al., 2020; Pantoja Munoz et al., 2018). Toilet paper as an example is a material which solely consists of cellulosic pulp and disintegrates well after disposal (Eren and Karadagli, 2012).
Wet wipes produced as a blend of viscose bres and wood pulp show su cient strength for proper usage of the wipe (Zhang et al., 2019a). They consist only of cellulosic material (Zhang et al., 2019a) and therefore are biodegradable (Soukupova et al., 2007). In these wipes the long viscose bres form the loadcarrying structure providing wet strength, the pulp bres attached to the body are responsible for liquid absorption and dispersive properties (Zhang et al., 2017). The production process of these wipes, a combination of wetlaid forming of the fabric and subsequent hydroentanglement with ,high-pressure water jets (Mao and Russell, 2006), requires no chemical agents as additives or binders. For disposal it has to be possible to ush the wet wipes down the sewage system. According to the guidelines of the nonwoven industry association of America (INDA) and Europe (EDANA) ushability is tested in a row of tests, where particularly the slosh box disintegration test is considered as suitable to determine the dispersibility of wipes (INDA and EDANA, 2018). As they are not containing any type of binders nonwovens from pure cellulosic bres seem to be the most promising material to obtain true ushability, also according to slosh box tests (Atasagun and Bhat, 2018;Phillip Mango, 2004).
Flushability and labelling wet wipes as ushable have been in public discussion recently (Campbell, 2018; Hassan, 2019; Kary, 2019). One reason for this public interest could be the newsfeeds that reported major blockages in the New York (Flegenheimer, 2015) and London (Taylor, 2017) sewer system. Fatbergs, as these massive blockages are called in the media, consist of undispersed wet wipes and fatty deposits (Taylor, 2017), compare Fig. 1 (right). These deposits form in the sewer system from disposed fats, oils and greases (He et al., 2011;Kusum et al., 2020). Field measurements in Berlin (Mitchell et al., 2020;Thamsen et al., 2017) and Tokyo (Okamoto, 2018) as well suggest that blockages are caused by undispersed wet wipes disposed via the toilet (Thamsen et al., 2017). Next to wrongly disposed nonushable wipes (Karadagli et al., 2021) these investigations also found so-called ushable wipes that did not disperse properly ( . In all these studies the wipes dispersed well but were tested without prior wet storage, the dry webs were put in the dispersion test, which is also common practice in industrial testing. In the actual end use however, the wet wipes are stored wet in their packages for weeks and months before disposal. In this work we will show that dispersibility of ushable wet wipes, measured with the slosh box disintegration test, can decrease drastically over wet storage time. For this decrease in dispersibility of the wet wipes during wet storage, we coined the term 'ageing effect'. We will show that without wet storage the wipes are showing excellent dispersibility, however they are losing these properties within 24 hours of wet storage. This loss of dispersibility is demonstrated for wipes from industrial production as well as for wipes produced on pilot scale. We will also demonstrate that by selecting suitable bres to produce the wet wipe, it is possible to obtain wipes with little to no ageing effect, thus proving that the widespread wetlaid/hydroentanglement process is suited to manufacture biodegradable and truly ushable wet wipes. Still, stable dispersibility over wet storage was only found for one set of bres not currently used for commercial products as far as we know, indicating that typical commercially available wet wipes are deteriorating in their dispersibility properties during wet storage in the consumer package.

Raw Materials
The bres used for the wet wipes are a blend of viscose bres with chemical pulp bres. The viscose bres were produced in an industrial scale viscose bre line (Cook, 1984) at Kelheim Fibres (grade A and B). Viscose bre A is a at viscose bre with a rectangular cross section, a bre length of 10 mm and a linear mass density (called titer in textile engineering) of 2.4 dtex (2.4 g per 10.000 m bre). The second bre (viscose bre B) has a roundish, irregular-shaped cross section and a bre length of 8 mm, it is a ner bre with a titer of 0.9 dtex. The bres consist of cellulose and have a surface nishing with ethoxylated fatty acid according to viscose bre industrial standards (Wilkes, 2001). The chemical pulps are a bleached softwood kraft pulp (SW-BK) and an unbleached softwood kraft pulp (SW-UBK), both commercially available grades from industrial production. The speci cations of these pulps are listed in Table 1. The Kappa number, measured according to ISO 302:2004, is indicating the lignin content of the pulps, nearly all the lignin has been removed from the bleached pulp grade. All used materials are of cellulosic origin and therefore biodegradable. The blends of these viscose bres with the bleached pulp are representing standard recipes for ushable wet wipes produced for consumer markets. The unbleached softwood kraft pulp was only used in the laboratory wipes and represents a prototype for possible future applications.

Wet wipe production -Pilot scale
The pilot-scale method has been adapted to mimic the industrial production process described below.
The production of the tested wipes includes 4 steps. In a rst step an inclined wire wetlaid former with a web width of 290 mm was used to form a bre web where the bres were primarily aligned in machine direction. Web speed is 4 m/min. Similar to paper production the nonwoven was dried with air at 160°C and a web draw of 0.1m/min. These two steps provided a bulky and weak non-woven fabric. For the proper usage of these fabrics as personal care wet wipes hydroentanglement was used to create the required tensile strength. For this procedure the fabric was passing 3 bars of spray nozzles applying water jets to the fabric, created by pressures of 5, 60 and 70 bars. The water jet nozzles were mounted at 40 holes per inch, each hole 0.1 mm in diameter. Web speed is 5 m/min. After hydroentanglement the web was dried on-line at 130°C. A full list of the produced reel materials can be found in Table 2. The targeted grammage of the nonwovens was 65 g/m².

Wet wipe production -Industrial scale
In order to investigate the behaviour of the actual consumer product also an industrial production facility was used to produce non-woven fabrics. Therefore, one of worldwide largest producer for hydroentangled wetlaid nonwovens provided us with the nonwovens listed in Table 3 manufactured in a standard process for consumer wet wipes. Both components, viscose bres and pulp, were the same products like in the pilot-scale produced fabrics, but from different production lots. The machinery to produce the fabrics was similar to the pilot-scale method but web forming and the hydroentanglement process are carried out inline. Therefore, these fabrics were only dried once, after the hydroentanglement. For the dispersibility measurement single wipes, 125 x 175 mm, were cut from the reel fabric. The grammage for all wipes was 65 g/m², giving a single wipe the weight of 1.4 g. Orientation effects were eliminated as the wipes were always cut in the same direction. For consumer products the wipes were stored in a lotion aimed to improve skin comfort and scent of the product. For storage we used both, deionized water and lotion. The liquid was applied to the cut wipes in a mass ratio of 2:1 (liquid : ambient condition dry mass of wet wipe) and stored in a closed plastic bag. For the measurements with the lotion the web PW1 -VB BSK was used. The lotion consists of 95.9 w% of water and chemicals, similar to a widely used patent (Marsh, 2016). Apart from water the lotion contains 1.5% thickener (Propylenglycol), emulgators (0.8% castor oil and 0.45% silicone polyether), 0.5% buffer (citric acid) as well as stabilizers, other emulgators and preservation agents. To avoid bacterial and fungal decay the bags with the wet wipes were stored in a refrigerator at 4°C. For each measurement point three wipes were stored in one plastic bag. For the test a plastic container (435 x 335 x 270 mm) was lled with 2 litres of tap water. One wet wipe was placed on the water, as shown in Fig. 2a. An engine was tilting the box back and forth with 26 rpm for 30 minutes. The maximum tilting angle in both directions is 14.5° (Fig. 2b). The wet wipe was supposed to disintegrate due to the gentle agitation. Subsequently the state of disintegration was evaluated by collecting the larger non-disintegrated parts of the wipe on a 12.5 mm hole sieve. Therefore the 12.5 mm sieve was stacked on a 200 µm sieve, Fig. 2c, and the suspension was poured within 20 seconds from a height of 100 mm above the upper sieve. The distance between the two sieves was 27 mm. The bre material withheld on the sieves was collected and dried at 105°C for 4 hours. Then the mass of the two material fractions from the coarse (m 12.5mm [g]) and ne (m 200µm [g]) sieve was determined. Dispersibility was calculated via Eq. 1, a higher dispersibility value indicates a better dispersive behaviour.

Dispersibility measurement -Slosh box disintegration test
The used slosh box tester consisted of three liquid containers next to each other (Fig. 2a). Thus, for each measurement point three individual sheets were sloshed, one in each box. After storage the wet wipes were directly put into the slosh box. The wipes were not dried prior to slushing like for other tests (Atasağun and Bhat, 2019;Tipper, 2016), as this is not representing actual consumer usage.

Research Hypotheses
Wet wipes are facing the problem that they are contributing to sewer blockages which is documented publicly (Drinkwater and Moy, 2017; Hassan, 2019) and scienti cally (Mitchell et al., 2020). This is possible as so-called ushable wet wipes do not disperse when being purchased by consumers Both the good dispersibility of never wetted nonwovens and the missing dispersibility of consumer-sold wet wipes, whether they are marked ushable or not, were carried out by different work groups using diverse materials. The ndings in these publications were always the same therefore, we add two more statements to our working hypothesis: 2. The production process of the precursor material does only impact the level of dispersibility but not the time dependent behaviour.
3. The type of applied liquid is not in uencing any time dependent change in dispersibility. All of these effects that occur in wet wipes can also be caused by deionized water.
4. We will furthermore show that choosing appropriate bers as a raw material for the wet wipe production, namely band-shaped viscose bres and unbleached softwood pulp bres, the good dispersibility of the wipes can be preserved during shelf storage.

Results And Discussion
Commercial wet wipes are usually treated with lotions improving the customer experience in terms of pleasant smell and skin comfort. The rst trial was aimed at comparing if there is a difference in dispersibility loss over time between wet wipes stored in lotion and wet wipes stored in water, as the lotions are consisting mostly of water (Marsh, 2016). Figure 3 represents the difference between a wipe stored in lotion and the same wipe stored in deionized water. Both liquids lead to a pronounced decrease in dispersibility over the wet storage time. It is interesting that water, the liquid used to disperse the wipes, is at the same time the substance causing the solidi cation of the wipes. While lotion is initially slowing the ageing, over longer storage time also the dispersibility for the wet wipes in lotion drops to very low values. This is in full agreement with our hypotheses stating that (1) the wet storage is liable for the deteriorated dispersibility and (3) that the liquid type is not relevant for this effect.
This solidi cation during the wet storage we call the ageing effect of dispersibility. According to the INDA/EDANA guideline for wet wipes a minimum dispersibility of 60% is required (INDA and EDANA, 2018). The area below the dashed line at 60% dispersibility in the Figs. 3 to 6 indicates insu cient disintegration after being ushed. After two weeks of storage the wet wipes investigated in Fig. 3 hardly disintegrate at all in the slosh box test, their dispersibility is below 10%. This is only an effect of the wet storage, the initial dispersibility of the dry fabrics is well above 70%.
The wet storage time in our tests is even lower than the time commercial wet wipes are stored in their package before being sold to the consumer. The results from Fig. 3   was tested and none are reaching satisfactory dispersibility. In our investigations, Fig. 4, only one viscosepulp combination (PW 2) was able to overcome the loss in dispersibility. It was the wipes produced from unbleached softwood kraft pulp and band-shaped viscose bres, validating our fourth research hypothesis. Figure 4 also demonstrates that the type of viscose bre is playing a role for dispersibility. Utilizing the same pulp, viscose bre A (PW1 -VB BSK), which has a at cross section, provides better dispersibility than viscose bre B (PW3 -VB BSK) with a circular cross section. This has also been observed earlier (Zhang et al., 2019b), there the improved dispersibility had been attributed to the better access of shear forces to the at viscose bres during disintegration.
Similar to the ndings shown in Fig. 4 tests were carried out for wipes from industrial scale production, manufactured by a commercial producer of nonwovens for wet wipes. Figure 5 depicts the results for wipes using the same viscose bres and a similar bleached pulp as in the pilot-scale trials. IW1 VA BSK for example represents a wipe with at viscose bres (Fibre A). Again, the wipes show good dispersibility when tested dry but they undergo a loss in dispersibility over the wet storage time. This demonstrates that the reduction of dispersibility during wet storage occurs for both, wipes produced at industrial scale as well as pilot scale. A direct comparison is shown in Fig. 6. Although there are differences in the starting level, both, industrial and pilot-scale production wet wipes, show a similar rate of decline in dispersibility which is supporting the second statement of our working hypothesis. Considering that there are differences in machinery and parameters between pilot-and industrial scale production it is reasonable to conclude that the loss in dispersibility is rooted in the bres used, and not in the production process.

Conclusions
It has been shown consistently that storage of wet wipes in water or water-based lotion, as it is the case for commercial products, is severely reducing their ability to disintegrate properly after usage. After 24 hours, latest after 168 hours, the dispersive properties for most of the tested hydroentangled wetlaid wipes are strongly reduced. Dispersibility of wipes tested without wet storage, represented as 0-hour dispersibilities in Fig. 4 to Fig. 6, are in good accordance to recent publications (Zhang et al., 2019a, 2019b, 2018; Zhang and Jin, 2018), con rming that initially the wet wipes are disintegrating well. The problem clearly is the storage of the wipes in water or lotion. It is not very intuitive that storage of a wet wipe fabric in water (or a water-based liquid) reduces its ability to be dispersed in water, yet this is exactly what is happening. Possible mechanisms causing this ageing effect could be related to long term swelling processes (Fang et al., 2013;Jawaid et al., 2011;Tajvidi et al., 2006), which are common to cellulosic bres, swelling mediated interdiffusion between the bre surfaces or mechanical deformation of the softened bre networks during wet storage.
The reason that this effect has not been described, is probably due to the fact that this behaviour is quite unexpected and thus has not been examined so far. Nevertheless, it has profound environmental implications. In the supply chain, the time between producing a wet wipe (i.e. putting it in wet storage in the consumer package) and the sale in stores is by far longer than 168 hours. Thus, an even more severe loss in dispersibility may take place for wet wipes in commercial end use. Monitoring dispersibility of wet wipes after shelf storage will help to reduce sewer blockage and ensure proper disintegration and faster biodegradability of the wipes.
As one key consequence of the work presented here, we are suggesting to adapt the standard testing procedures for wet wipe ushability in such a way that also the reduction of dispersibility over time due to the ageing effect is covered. Such a procedure needs to include the test of wipes e.g. after 2, 4 and 12 weeks of wet storage, in order to correctly re ect the end use of the commercial product.
The ageing effect has been observed equivalently for industrial-and pilot-scale production wet wipes, the decrease in dispersibility is similar. For pilot-scale produced wipes, we demonstrated that using a combination of unbleached softwood pulp and band shaped viscose bers, wet wipes with good dispersibility properties over wet storage can be made, see Fig. 4. This shows that using appropriate bres, truly ushable and biodegradable wet wipes from purely lignocellulosic material can be produced.