The Fire Behaviour of Fabrics Containing Dried Emollient Residues

Abstract

A significant number of UK fire fatalities have been reported to involve textiles contaminated with emollients. In the following study, the flammability of a variety of fabrics containing 14 different emollients, including paraffin-free creams, was evaluated. This is the first time the impact of the presence of such a large range of emollients has been examined. Horizontal burn tests were conducted on emollient-contaminated fabrics. Significantly earlier ignition time were noted upon heating for all emollient-contaminated fabrics (p < 0.001) when compared to the behaviour of blank fabrics were noted using a vertical burn test. The mean time to ignition for 100% cotton fabric (151 ± 2 g/m²) was reduced from 71.5 to 14.4 s and for 52%/48% polyester/cotton fabric (103 ± 2 g/m²) from 328 to 12.9 s by the presence of emollients. Horizontal burn tests with a direct flame on 100% cotton fabric (114 ± 1 g/m²) displayed an accelerated mean flame speed from 0.0032 to 0.0048 ms⁻¹ and an increased maximum flame height of 56.6 to 175.4 mm for emollient-contaminated fabrics. These findings demonstrate the fire risk of fabrics contaminated with a dried emollient. Their potential to ignite quickly and to propagate a fire may strongly decrease the reaction time of an impacted individual. Therefore, it is important that this risk and appropriate safety advice be continually highlighted and communicated not only in the UK but worldwide.

1. Introduction

According to the Medicines and Healthcare products Regulatory Agency (MHRA), there were more than 50 reported deaths attributed to fires involving clothing or bedding materials containing residues of skin care products between 2010 and 2018 in the UK [1]. This number is widely accepted as being inaccurate. The lack of awareness by fire and rescue services (FRS), healthcare professionals and the public has resulted in the underreporting of fire deaths and injuries linked to emollients [2,3,4]. Many of these cases have noted the contamination of fabrics with emollients as a potential contributing factor and reporting the victims as elderly or less mobile and the source of ignition detailed as open flames and bar heaters but typically cigarettes or lighters [5,6]. There have also been several coroners’ reports in the UK recommending that more information should be made available to patients and healthcare professionals on the increased risk of fire when emollients are dried onto bedding and clothing [7,8,9].

Emollient is a general term used for oil-based personal care products such as lotions, creams and ointments that are used to treat various skin conditions such as eczema and psoriasis. The most common excipients include the following: emulsifiers that form a stable emulsion; occlusives that act as a barrier to keep the skin hydrated; permeation enhancers to increase skin permeability and humectants to attract and retain skin moisture and prevent drying of the formulation, as well as preservatives, thickeners and gelling agents [10,11,12]. Emollient formulations differ in oil–water ratios, type of moisturiser and occlusives, depending on the therapeutic purpose, application site and skin type. Lotions are described as less greasy, with a lower viscosity and greater water content compared to creams and ointments. Generally, lotions and creams are reported to have up to 50% oil–water content, where ointments contain 80% or above [13].

Approximately 24% of the UK population visit their GP with skin-related problems [10], and the World Health Organization [14] has reported that, globally, skin health issues affect an estimated 1.8 billion people at some point in their lifetime. In 2023 alone, 26.3 million items were prescribed for skin-related ailments in England [15]. The growing global market of skin care products is estimated to be worth 1.75 billion dollars [16], with an increase in the number of plant-based or paraffin-free products. A range of these products have been evaluated [17].

Emollients are applied to the skin directly as needed and may transfer to clothing and bedding [18]. This may result in build-up on clothing and bedding. Emollients dried onto fabric are capable of acting as fuel, with the fabric acting as a wick [19,20]. Similar behaviour has been noted with contaminated workwear, such as overalls [21].

The fire risk associated with emollient-contaminated fabrics has not been extensively reported [19,20]. Previous limited vertical flammability testing of fabrics contaminated with emollients, using an indirect flame for ignition (i.e., high heat source), has highlighted the potential fire risk. It identified that when an emollient was present, there was a significant decrease in time to ignition for 100% cotton sheeting from 68 s to 6 s, which may impact an individual’s reaction time [19]. Therefore, a number of 100% cotton and polyester/cotton blended fabrics samples containing 14 emollients were evaluated. These included two paraffin-based lotions, three non-paraffin creams, seven paraffin-based creams and two paraffin-based ointments (Appendix A 

Table A1. Emollient types used (1.25 mL single application) to contaminate fabrics for all flammability tests and ingredients detailed by the manufacturers.

Emollient Type	Manufacturer	Ingredients
Lotion 1	Reckitt Benckiser 1	10% white soft paraffin, 4% light liquid paraffin, aqua, petrolatum, isopropyl palmitate, paraffinium liquidum, glyceryl stearate, ceteareth-20, lanolin, phenoxyethanol, ethylparaben, methylparaben, hydroxylethylcellulose, carbomer, ethylhexylglycerin, sodium hydroxide and BHT
Lotion 2	Johnson and Johnson 2	Aqua, glycerin, distearyldimonium chloride, isopropyl palmitate, paraffinum liquidum, cetyl alcohol, dimethicone, avena sativa kernel flour, paraffin, cera microcristallina, sodium chloride, BHT and benzyl alcohol
Non-paraffin cream 1	Fontus Health 1	Avena sativa kernel flour 1%, purified water, apricot kernel oil, glycerin, sucrose stearate, cetearyl alcohol, glyceryl stearate SE, dimethicone, phenoxyethanol, vitamin F, ethyl ester, ethylhexylglycerin, xanthan gum, disodium EDTA and vitamin E
Non-paraffin cream 2	Fenton Pharmaceuticals 3	Aqua, glyceryl stearate SE, glycerin, lanolin alcohol, benzyl alcohol, cetearyl alcohol, oleic acid, triethanolamine, stearic acid, p-Chloro-m-Cresol, parfum, linalool, limonene, citronellol, coumarin, geraniol, anise alcohol and benzyl cinnamate
Non-paraffin cream 3	Naturally thinking 4	Aqua, capric triglyceride oil, vitis vinifera seed oil, glyceryl stearate S/E, glyceryl stearate, Peg 100 stearate, glycerine, cetyl alcohol, stearic acid, glyceryl stearate, vitamin E, phenoxyethanol and ethylhexylglycerin
Cream 4	Almirall Ltd 2	Urea 5% and lauromacrogols 3%, dimethicone, phenyl dimethicone, liquid paraffin, cetylpalmitate, stearic palmitic acid, octyldodecanol, glycerol 85%, polysorbate, carbomer, trometamol, benzyl alcohol and purified water
Cream 5	Bayer UK 2	15% white soft paraffin, 6% liquid paraffin, cetostearyl alcohol, macrogol cetostearyl ether, chlorocresol, sodium dihydrogen phosphate, sodium hydroxide, phosphoric acid and purified water
Cream 6	Reckitt Benckiser 1	14.5% white soft paraffin, 12.6% light liquid paraffin, lanolin, cetyl alcohol, methyl parahydroxybenzoate, propyl parahydroxybenzoate, empilan GMS, sodium cetostearyl sulphate, carbomer, sodium hydroxide, citric acid monohydrate and purified water
Cream 7	Thorton and Ross 1	15% white soft paraffin, 6% light liquid paraffin, benzyl alcohol, cetostearyl alcohol, potassium sorbate, macrogol 1000 monostearate, glycerol, citric acid monohydrate, povidone and purified water
Cream 8	Pinewood Healthcare 1	15% white soft paraffin, 6% liquid paraffin, cetostearyl alcohol, sodium lauryl sulfate, phenoxyethanol and purified water
Cream 9	Tosara Pharma Limited 2	Zinc oxide, lanolin, benzyl alcohol, benzyl benzoate, benzyl cinnamate, sodium benzoate, purified water, paraffin wax, microcrystalline wax, heavy liquid wax, synthetic beeswax, sorbitan sesquioleate, propylene glycol, antioxidant, linayl acetate and lavender
Cream 10	Dermal Laboratories 2	15% liquid paraffin, isopropyl myristate, glycerol, carbomer, sorbitan laurate, trolamine, phenoxyethanol and purified water
Ointment 1	Bayer UK 2	95% white soft paraffin and 5% liquid paraffin
Ointment 2	Zuche Pharma 2	50% white soft paraffin and 50% liquid paraffin
Purchased: 1 My Pharmacy, Great Harwood, UK. 2 Chemist direct, Leeds, UK. 3 Powermed Plus Ltd, Bloxham, UK. 4 Naturally thinking, London, UK.

). Vertical flammability tests using an indirect flame (heating) ignition source were carried out and, for the first time, horizontal flammability tests, using a direct flame ignition method, have been conducted. The most commonly prescribed emollients as detailed in the formularies and prescribing guidelines of integrated care systems [17] and health boards in the UK were used. The selection of fabric types was based on commonly used bedding in the UK and polyester blended types that are often used in care settings, such as those that are BS7175 compliant [22].

The inclusion of horizontal burn tests in this evaluation adds a further dimension to the understanding of the risks involved when fabrics are impregnated with emollients. This is particularly important when considering that many of the fatality cases often involve the ignition of bedding fabrics and horizontal flame propagation as the primary method of fire spread [7,8]. The initial increase and then decrease in flame speed and height, with the increasing moisture content of textiles [23], and the effect of inclination angles on flame spread [24], as well as the effects of the presence of flame retardants on vertical and horizontal flame spread [25], have been reported. However, the involvement of horizontal flame spread in emollient-impregnated fabrics has not been explored. No comparison with vertical fire spread has been attempted.

The fire behaviour of fabrics changes in the presence of an emollient, yet this risk has not been communicated to the community [6]. The National Patient Safety Agency (NPSA) released guidance in 2007 [26], which was followed in 2016 [27] and 2018 [1] with more up-to-date advice from the MHRA based on new findings and reports of fire deaths. Additionally, the MHRA requested new warning labels during license updates on emollient products. The results presented herein are intended to raise the understanding and awareness of fire and rescue services, healthcare professionals and emollient users and their families/carers to these risks. This testing evaluates the impact on burn behaviour of different cotton and polyester/cotton blended fabrics when contaminated with dried-on emollients. This was assessed using vertical flammability testing, measuring the time to ignition (once the fabric is exposed to a high heat source as an indirect flame), flame time and glowing time, and horizontal testing, measuring flame height and speed.

2. Materials and Methods

The term “emollient” has been adopted to describe the skin care products used and includes lotions, creams, non-paraffin creams and ointments (Appendix A Table A1). The fabrics chosen were based on fabrics used domestically and were purchased from common UK stores. The same cotton pillowcases (114 ± 1 g/m²) were used in both vertical and horizontal flammability tests. However, owing to the discontinuation and availability of some of the fabrics, different polyester/cotton blends were used in vertical and horizontal flammability testing. The purpose of testing was not to compare the fire behaviour of the fabrics on their own, but to compare the individual fabric burning behaviour to when they are contaminated with an emollient.

2.1. Vertical Flammability Test Method

The vertical flammability test was based on a previously reported method [19], adapted from those published by the US Federal Aviation Administration (FAA) [28]. In place of a Bunsen burner, a micro-Bunsen burner (Humboldt, purchased from Cole-Parmer, St Neots, UK.) was used with an adjustable valve to allow for a fixed flame height and a 7 ± 0.4 cm gap between the tip of the flame and the lower edge of the test textile strip (Figure 1) for indirect ignition testing (n = 5). Another modification included two additional side arms on the stainless steel specimen holders (fabricated by Mackays Metals, Cambridge, UK) to help secure the fabric strips between them. The FAA standard [28] as with other textile flammability testing standards utilises a direct flame on the lower edge of the fabric (FAA-16mm) and is designed to test flame-retardent materials. The previous reported method was modified by utilising an indirect flame, with the aim of testing more flammable cotton-based fabrics, as used domestically and reported in fire fatalities and serious incidents. This allowed for the measurement of time to ignition to be carried out, which was deemed to be too quick using a direct flame, and statistical comparisons of the data were carried out [19].

The specimen holder was used for all tests to sandwich the textile strips (Figure 1 and Figure 2). For the vertical flammability test, it was suspended above the burner using retort stands, (Fisher Scientific, Loughborough, UK.) with only the bottom edge being exposed to the flame (Figure 2). A bespoke heatproof tile (Eurocell Plc, Loughborough, UK) with a drilled circular indent (0.5 cm depth) of the same diameter as the base of the micro-Bunsen burner was used for the consistent placement of the burner under the specimen holder. Time to ignition was measured, which was the time from when the burner was placed under the lower edge of the textile to the point at which an independent flame (fire point) was held. Flame and glowing combustion times were also measured, i.e., the time the flame was present on the surface of the textile and the time after the flame had been extinguished and ‘glowing’ was seen on the remaining fabric. The textiles used were common fabrics used for bed linen: 200-thread-count 100% combed cotton pillowcases (114 ± 1 g/m²) manufactured by Noah’s Linen, 52% polyester/48% cotton pillowcases (103 ± 2 g/m²) purchased from Tesco PLC and 400-thread-count 100% Egyptian cotton pillowcases (151 ± 3 g/m²) manufactured by Linen Zone LTD. A range of 14 emollients were utilised as contaminants, varying from paraffin-free emollients up to 100% paraffin content, on each fabric, with all tests repeated 5 times (n = 5).

2.2. Horizontal Flammability Test Method

The method, as detailed by Nazaré and Horrocks [29], includes the specimen holder (Figure 1) at an angle of inclination of 0° orientation so that the front face of the specimen is flat and facing upwards. Direct ignition was carried out utilising a butane/air lighter (Swedish Match Lighters B.V, Assen, Netherlands.) which was held to the edge of the fabric specimen until an independent flame was established. The flame spread (based on the edge of the fabric to the top part of the frame length = 305 mm) and maximum flame height were calculated from footage of each test, utilising Kinovea software (version 0.9.5, Joan Charmant & contributors). All horizontal tests were repeated 3 times (n = 3).

The textiles tested for the horizontal flammability tests were 100% cotton (137 ± 2 g/m²) and 65% polyester/35% cotton (70 ± 2 g/m²) fabrics, which were purchased from Cheap Fabrics, and 200-thread-count pillowcases (114 ± 1 g/m²) manufactured by Noah’s Linen.

2.3. Textile Strip Preparation

The textiles were cut into 80 mm by 320 mm strips and weighed. Emollients were sampled from their tubs/pump containers/tubes using a spoon-ended spatula (1.25 mL when level). The emollient was then applied by dabbing the spatula along the bottom edge of the strips (0–30 mm), and any remaining emollient on the spatula was also removed using the bottom edge of the fabric to ensure that all of the emollient was transferred. A fingerprint roller (WA products LTD, Burnam on Crouch, UK) was repeatedly rolled up the fabrics back and forth (10 times), ensuring that the emollient was spread across the entire fabric sample. Due to differences in emollient packaging and viscosity, this method of application was deemed most suitable and consistent. The strips were then reweighed and dried for 24 h and then reweighed to calculate the % mass loss. Both the vertical and horizontal flammability tests were carried out in a fume cabinet (W: 165 cm, D: 60 cm, H: 50 cm) initially, with no extraction until the flame began to spread, to reduce any flame movement before and during ignition. The extraction had a 0.44 ms⁻¹ flow rate.

2.4. Data Analysis

The aim of this research is to compare blank fabric burning behaviour to a fabric that has been contaminated with a dried-on emollient. Statistical analysis was carried out using ANOVA, post hoc Tukey and Sidak tests (IBM® SPSS® v29) for the vertical flammability test results (n = 5) and the Sidak test for the horizontal flammability tests (n = 3), owing to the smaller sample size [30]. With the lower variability and consistent horizontal test results for cotton fabric, it was deemed appropriate to carry out three tests that gave enough statistical power to detect the significant effects of the emollients. Comparisons between the variance of the means were regarded as significantly different when p < 0.05. The masses were recorded before drying to calculate the % mass loss after 24 h.

3. Results

3.1. Vertical Flammability Tests

Table 1. Vertical flammability test results (mean (n = 5) and standard deviation) of 100% cotton (114 ± 1 g.m⁻²) when contaminated with the 14 dried-on emollients (%RSD = % relative standard deviation).

Textile Test and Emollient	Time to Ignition ± Stdev/s	%RSD	Flame time ± Stdev/s	%RSD	Glowing time ± Stdev/s	%RSD
Blank 100% cotton	62.6 ± 4.0	6.4	26.4 ± 3.7	14.1	70.3 ± 38.2	54.3
Lotion 1	8.0 ± 4.7	58.3	13.4 ± 0.8	5.8	20.3 ± 2.7	13.1
Lotion 2	8.9 ± 4.1	46.2	11.5 ± 1.4	11.9	41.7 ± 10.4	24.9
Non-paraffin cream 1	5.4 ± 2.8	50.6	19.0 ± 8.6	45.2	43.2 ± 12.1	27.9
Non-paraffin cream 2	8.4 ± 2.1	25.4	13.7 ± 2.3	16.6	40.8 ± 4.5	11.1
Non-paraffin cream 3	4.4 ± 2.8	62.8	16.2 ± 2.3	14.4	51.9 ± 10.2	19.7
Cream 4	6.7 ± 3.3	49.6	18.5 ± 2.9	15.8	41.1 ± 14.5	35.3
Cream 5	12.9 ± 4.9	38.2	12.7 ± 0.5	4.0	23.3 ± 7.0	30.2
Cream 6	13.4 ± 2.2	16.6	11.7 ± 0.8	7.1	20.0 ± 2.3	11.5
Cream 7	5.1 ± 4.0	78.1	15.1 ± 3.6	24.0	41.7 ± 2.2	5.3
Cream 8	12.3 ± 5.2	42.3	17.9 ± 5.4	30.0	41.3 ± 10.8	26.0
Cream 9	4.7 ± 1.7	36.9	19.4 ± 6.3	32.2	65.4 ± 24.4	37.2
Cream 10	6.3 ± 3.7	59.1	16.9 ± 1.0	6.1	33.2 ± 7.3	22.1
Ointment 1	9.9 ± 2.5	25.7	19.6 ± 1.9	9.8	28.7 ± 3.9	13.7
Ointment 2	5.2 ± 3.8	74.0	25.0 ± 4.8	19.4	42.3 ± 10.6	25.0

,

Table 2. Vertical flammability test results (mean (n = 5) and standard deviation) of 100% cotton (151 ± 2 g/m²) when contaminated with the 14 dried-on emollients (%RSD = % relative standard deviation).

Textile Test and Emollient	Time to Ignition ± Stdev/s	%RSD	Flame Time ± Stdev/s	%RSD	Glowing Time ± Stdev/s	%RSD
Blank 100% cotton	71.5 ± 35.5	49.6	21.5 ± 1.4	6.7	31.2 ± 3.5	11.3
Lotion 1	14.0 ± 6.6	47.2	21.2 ± 3.0	14.3	34.8 ± 5.2	15.0
Lotion 2 1	24.0 ± 5.3	21.9	22.6 ± 3.8	16.8	29.8 ± 10.9	36.4
Non-paraffin cream 1 1	23.5 ± 5.1	21.8	23.0 ± 2.5	10.7	38.6 ± 6.6	17.0
Non-paraffin cream 2	13.5 ± 6.4	47.2	23.0 ± 7.2	31.2	39.0 ± 5.6	14.4
Non-paraffin cream 3	10.2 ± 6.3	61.9	21.5 ± 2.3	10.5	33.1 ± 3.2	9.6
Cream 4	14.5 ± 9.0	61.5	19.5 ± 1.4	7.1	33.2 ± 3.3	9.6
Cream 5	6.5 ± 2.2	33.3	15.1 ± 1.1	7.2	25.6 ± 2.5	10.0
Cream 6	14.7 ± 4.3	29.4	19.5 ± 5.9	30.4	40.4 ± 12.7	31.4
Cream 7	17.4 ± 8.7	50.2	21.2 ± 3.4	16.1	38.7 ± 7.8	20.1
Cream 8	12.7 ± 9.8	77.0	20.1 ± 2.6	13.0	25.3 ± 3.5	13.7
Cream 9	8.9 ± 4.2	47.2	23.2 ± 1.4	6.1	53.4 ± 6.7	12.5
Cream 10	13.0 ± 1.6	12.6	21.2 ± 1.8	8.4	34.9 ± 8.9	25.5
Ointment 1	10.9 ± 2.5	23.1	19.6 ± 1.7	8.7	32.7 ± 3.7	11.2
Ointment 2	18.1 ± 12.4	68.5	28.7 ± 5.9	20.4	42.5 ± 6.8	16.1

¹ n = 4.

and

Table 3. Vertical flammability test results (mean (n = 5) and standard deviation) of 52% polyester/48% cotton (102 ± 2 g/m²) when contaminated with the 14 dried-on emollients (%RSD = % relative standard deviation).

Textile Test and Emollient	Time to Ignition ± Stdev/s	%RSD	Flame Time ± Stdev/s	%RSD	Glowing Time ± Stdev/s	%RSD
Blank 52/48% poly/cotton	328 ± 71.8	21.9	22.6 ± 2.0	8.8	48.4 ± 32.2	66.4
Lotion 1	16.7 ± 7.3	43.7	15.6 ±1.3	8.1	12.5 ± 3.9	31.1
Lotion 2	13.4 ± 5.2	38.7	15.2 ± 1.7	11.4	64.4 ± 30.7	47.6
Non-paraffin cream 1	12.6 ± 8.8	69.9	17.0 ± 2.8	16.6	47.2 ± 18.8	39.9
Non-paraffin cream 2	24.6 ± 18.3	74.6	17.2 ± 4.5	26.1	48.4 ± 21.5	44.5
Non-paraffin cream 3	15.4 ± 8.5	55.4	19.0 ± 2.6	13.9	82.6 ± 35.7	43.2
Cream 4 1	6.7 ± 2.2	33.4	20.2 ± 5.3	26.3	44.4 ± 15.6	35.1
Cream 5	10.9 ± 5.2	48.0	15.3 ± 1.5	9.5	6.9 ± 3.0	43.5
Cream 6	2.9 ± 1.2	41.5	16.4 ± 3.6	21.7	98.5 ± 34.1	34.6
Cream 7 1	11.3 ± 7.9	69.5	16.1 ± 1.7	10.3	77. 0 ± 20.5	26.7
Cream 8	9.8 ± 7.9	79.9	18. 8 ± 4.1	22.0	57.9 ± 33.9	58.5
Cream 9	11.1 ± 2.8	25.2	18.5 ± 0.9	4.9	76.9 ± 21.8	28.3
Cream 10	11.9 ± 8.3	69.8	19. 5 ± 5.2	26.7	58.1 ± 35.2	60.7
Ointment 1	19.2 ± 3.0	15.4	17.7 ± 1.2	7.0	11.1 ± 2.8	25.8
Ointment 2 1	13.4 ± 9.4	69.9	41.9 ± 12.4	29.6	59.3 ± 36.4	61.4

¹ n = 4.

include the results of the vertical flammability tests using an indirect flame, with the time to ignition, flame time and glowing time recorded for the three fabrics with each of the 14 emollients.

Table 4. Vertical flammability test results of the total burn time (s) and overall (mean and standard deviation) mass loss upon drying of the emollient-contaminated fabrics.

Textile Test/ Emollient Type	Mean Mass Loss/%	Mean Vertical Total Burn Time (s)
		114 ± 1 g/m2 Cotton	151 ± 2 g/m2 Cotton	102 ± 2 g/m2 Poly/Cotton
Blank fabric		159.3	124.2	457.6
Lotion 1	71.7 ± 2.1	41.7	70.0	44.8
Lotion 2	69.8 ± 1.9	62.1.	76.4	93.0
Non-paraffin cream 1	67.0 ± 1.2	67.6	85.1	76.8
Non-paraffin cream 2	71.3 ± 0.5	62.9	75.5	90.2
Non-paraffin cream 3	69.2 ± 0.6	72.5	64.8	117.0
Cream 4	64.1 ± 1.0	66.3	67.2	71.3
Cream 5	66.7 ± 0.9	48.9	47.2	33.1
Cream 6	58.9 ± 1.8	45.1	74.6	117.8
Cream 7	63.4 ± 0.9	61.9	77.3	104.4
Cream 8	64.7 ± 1.0	71.5	58.1	86.5
Cream 9	36.3 ± 3.1	89.5	85.5	106.5
Cream 10	60.9 ± 1.5	56.4	69.1	89.5
Ointment 1	1.4 ± 1.2	58.2	63.2	48.0
Ointment 2	1.0 ± 0.5	72.5	89.3	114.6

shows the overall burn time (sum of the time to ignition and flame and glowing times) and the 24 h % mass loss on drying of the textile strips with an emollient. The mean measurements were based on five repeats, unless an outlier was calculated and removed, using Dixon’s Q-test.

Figure 3 shows the plots of the time to ignition for all three fabrics (when contaminated with the 14 emollients) and includes the blank/control fabric. Figure 2 shows a typical image of the 100% cotton blank (114 ±1 g/m²) compared to when a dried-on emollient is present on ignition. Figure 4a is a combined plot of the flame time and Figure 4b shows the glowing time. All figures include error bars of the standard deviation.

3.1.1. Time to Ignition

Figure 3a and 3b shows reductions in ignition time for both 100% cotton fabrics tested (114 ± 1 g/m² and 151 ± 2 g/m²) when contaminated with a dried-on emollient. Using ANOVA post hoc Tukey and Sidak tests, the times to ignition for both 100% cotton blank controls, compared to when all 14 emollients are present, were significantly lower (all p < 0.001).

Figure 3c shows that there is also a large reduction in the ignition time of the blank 52% polyester/48% cotton fabric to when an emollient is present. Similarly, the same statistical comparison of the blank fabric to when all 14 emollients were present was significantly lower (all p ≤ 0.001) in terms of the time to ignition, despite the variability in the data (Table 3).

3.1.2. Flame and Glowing Time

Figure 4a shows that the mean flame time is quicker when an emollient is present compared to both the blank controls of 100% cotton (114 ± 1 g/m²) and 52% polyester/48% cotton. For the 100% cotton (151 ± 2 g/m²), all were similar in flame time compared to the blank control, except for tests with ointment 2, but there was high variability in the test results. Statistical comparisons using both post hoc Tukey and Sidak tests for 114 ± 1 g/m² 100% cotton found that for eight tests—lotions 1 and 2, non-paraffin creams 2 and 3 and creams 5, 6, 7 and 10—the flame time was significantly quicker (p < 0.05). The remaining six tests—including the non-paraffin cream 1, creams 4, 8 and 9 and ointments 1 and 2—had a similar flame time to the blank (p > 0.05). Comparison between the blank 151 ± 2 g/m², 100% cotton and 52% polyester/48% cotton fabric and the test with all 14 emollients present on the fabric showed that all were similar in flame time (p > 0.05).

There was significant variability in glowing time data for tests using the 52% polyester/48% cotton fabric (Table 1, Table 2 and Table 3); however, Figure 4b does show that the mean glow time is quicker in the presence of an emollient when compared to both the 114 ± 1 g/m² 100% cotton and 52% polyester/48% cotton blank controls. The 151 ± 2 g/m² 100% cotton tests were less variable than those for the other two fabrics, and with an emollient present, the glowing time appeared to be longer or similar. Statistical analysis comparing both post-hoc tests of the blank 100% cotton fabric (114 ± 1 g/m²) to when an emollient was present showed that in five tests—lotion 1, creams 5, 6 and 10 and ointment 1—the flame times were significantly quicker (p < 0.05). The other tests were not significantly different (p > 0.05) in glowing time when compared to the blank. For 151 ± 2 g/m² 100% cotton fabric, 13 tests were not significantly different (p > 0.05); however, one test in the presence of cream 9 (p ≤ 0.001) was significantly longer, with the longest glowing time of 53.4 ± 6.7 s. Using both post hoc test comparisons of the 52% polyester/48% cotton fabric blank to those contaminated with an emollient showed that all tests were found to not be significantly different (p > 0.05) from the blank (%RSD of 66%), probably owing to the large variability in the data.

3.2. Horizontal Flammability Tests

The horizontal flammability tests were carried out, the maximum flame height was measured, and the flame spread was calculated (Figure 5 and Figure 6). The mean results were based on three tests (

Table 5. Horizontal flammability test results of the flame speed (mean (n = 3) and standard deviation) of 2 × 100% cotton fabrics (137 ± 2 gm² and 114 ± 1 g/m²) and 65% polyester/35% cotton (70 ± 2 g/m²) when contaminated with the 14 dried-on emollients (%RSD = % relative standard deviation).

Textile Test and Emollient	Flame Speed/ms−1 (Cotton) 1	%RSD	Flame Speed/ms−1 (Cotton) 2	%RSD	Flame Speed/ms−1 (Poly/Cotton) 3	%RSD
Blank fabric	0.0019 ± 0.0003	15.8	0.0032 ± 0.00005	1.6	0.0060 ± 0.0004	6.7
Lotion 1	0.0022 ± 0.0002	9.1	0.0042 ± 0.0004	9.5	0.0066 ± 0.0002	3.0
Lotion 2	0.0036 ± 0.0005	13.9	0.0061 ± 0.0005	8.2	0.0095 ± 0.0003	31.4
Non-paraffin cream 1	0.0025 ± 0.0001	4.0	0.0043 ± 0.0002	4.7	0.0066 ± 0.0004	6.1
Non-paraffin cream 2	0.0030 ± 0.0003	10.0	0.0047 ± 0.0002	4.3	0.0080 ± 0.0001	11.3
Non-paraffin cream 3	0.0032 ± 0.0003	9.4	0.0051 ± 0.0002	3.9	0.0088 ± 0.0017	19.3
Cream 4	0.0037 ± 0.0002	5.4	0.0044 ± 0.0002	4.5	0.0056 ± 0.0005	8.9
Cream 5	0.0030 ± 0.0002	6.7	0.0049 ± 0.0003	6.1	0.0057 ± 0.0006	10.5
Cream 6	0.0037 ± 0.0005	13.5	0.0048 ± 0.0003	6.3	0.0064 ± 0.0020	31.3
Cream 7	0.0031 ± 0.0001	3.2	0.0054 ± 0.0002	3.7	0.0062 ± 0.0002	3.2
Cream 8	0.0038 ± 0.0002	5.3	0.0050 ± 0.0002	4.0	0.0064 ± 0.0004	5.6
Cream 9	0.0031 ± 0.0002	6.5	0.0043 ± 0.0002	4.7	0.0061 ± 0.0005	8.1
Cream 10	0.0035 ± 0.0005	14.3	0.0060 ± 0.0004	6.7	0.0099 ± 0.0008	8.1
Ointment 1	0.0027 ± 0.0002	7.4	0.0035 ± 0.0004	11.4	0.0044 ± 0.0005	11.4
Ointment 2	0.0034 ± 0.0002	5.9	0.0050 ± 0.0003	6.0	0.0042 ± 0.0001	2.4

¹ 137 ± 2 g/m², ² 114 ± 1 g/m², ³ 70 ± 2 g/m².

and

Table 6. Horizontal flammability test results of the flame height (mean (n = 3) and standard deviation) of 100% cotton (137 ± 2 g/m² and 114 ± 1 g/m²) and 65% polyester/35% cotton (70 ± 2 g/m²) fabrics when contaminated with the 14 dried-on emollients (%RSD = % relative standard deviation).

Textile Test and Emollient	Flame Height/mm (Cotton) 1	%RSD	Flame Height/mm (Cotton) 2	%RSD	Flame Height/mm (Poly/Cotton) 3	%RSD
Blank fabric	56.1 ± 2.8	5.0	56.6 ± 17.6	31.1	77.0 ± 25.6	33.2
Lotion 1	121.3 ± 18.3	15.0	140.6 ± 4.9	3.5	176.1 ± 20.8	11.8
Lotion 2	139.9 ± 3.8	2.7	181.9 ± 37.3	20.5	228.2 ± 26.1	11.4
Non-paraffin cream 1	145.6 ± 27.0	18.5	128.6 ± 26.5	20.6	228.7 ± 26.7	9.0
Non-paraffin cream 2	160.6 ± 27.7	17.2	153.9 ± 14.5	9.4	216.9 ± 102.4	47.4
Non-paraffin cream 3	120.7 ± 4.4	3.6	166.3 ± 16.3	9.8	272.8 ± 16.3	14,2
Cream 4	183.6 ± 19.2	10.5	144.3 ± 17.7	12.3	269.3 ± 79.2	29.4
Cream 5	180.1 ± 20.5	11.4	171.4 ± 9.5	5.5	254.0 ± 42.7	16.8
Cream 6	189.7 ± 61.9	32.6	172.8 ± 23.1	13.4	279.0 ± 66.7	23.9
Cream 7	178.1 ± 16.3	9.2	179.6 ± 1.7	1.0	231.1 ± 19.3	8.4
Cream 8	150.8 ± 10.1	6.7	189.0 ± 25.1	13.3	309.2 ± 10.6	3.4
Cream 9	165.7 ± 9.8	5.9	195.4 ± 10.0	5.1	249.4 ± 30.9	12.4
Cream 10	186.0 ± 38.9	20.9	210.3 ± 5.1	2.4	315.1 ± 9.6	3.1
Ointment 1	192.5 ± 25.7	13.4	206.9 ± 5.4	2.6	280.5 ± 23.7	8.5
Ointment 2	216.1 ± 0.6	0.3	215.0 ± 0.6	0.3	289.1 ± 25.1	8.7

¹ 137 ± 2 g/m², ² 114 ± 1 g/m², ³ 70 ± 2 g/m².

) with no outliers recorded and with good repeatability compared to the vertical tests.

3.2.1. Horizontal Flame Speed

Comparisons between the flame speed of the blank 100% cotton fabric (137 ± 2 g/m²) and those of fabrics contaminated with the 14 emollients were performed using ANOVA post hoc Sidak tests. All tests were significantly quicker (p ≤ 0.001) in flame speed compared to the control fabric (p < 0.05), apart from lotion 1, non-paraffin cream 1 and ointment 1, which were not significantly different to the blank (p > 0.05). All comparisons between the blank 100% cotton fabric’s (114 ± 1 g/m²) flame speed and those with an emollient present, other than ointment 1 (p > 0.05), resulted in a significantly quicker horizontal flame speed (p < 0.01). Comparisons between the blank 65% polyester/35% cotton fabric and the tests with an emollient present showed that only cream 10 was significantly faster in flame speed.

3.2.2. Horizontal Flame Height

When using ANOVA post hoc Sidak tests to compare the maximum flame heights of the blank 100% cotton (137 ± 2 g/m²) fabric to tests with an emollient present, all were significantly higher in flame height (p <0.05), apart from lotion 1 and non-paraffin cream 3 (p > 0.05). Comparing the flame heights of the blank 100% cotton (114 ± 1 g/m²) to the fabric impregnated with emollients, all were found to be significantly different (p < 0.05) and larger. The maximum flame heights were also significantly higher when all emollients were present on the 65% polyester/35% cotton fabric when compared to the blank (p < 0.05).

4. Discussion

4.1. Vertical Flammability Tests

All of the measurements of time to ignition of the three fabrics tested were significantly quicker when contaminated with the 14 dried-on emollients, demonstrating the serious fire risk of emollients, including those described as paraffin-free products, when dried onto fabrics. The paraffin-free emollients contain oils such as nut kernel and grape seed oils (Vitis vinifera), which typically contain fatty acids such as linoleic, oleic and palmitic acids and [31], or wool oil (lanolin), which contains long-chain fatty acids [13]. All of these oils, although replacing paraffin-based ingredients, still have the potential to be flammable. The burning behaviour of the emollients is supported by the fabric acting as a wick, which increases the surface area to produce a flammable vapour/air mixture [32]. There are noted differences in the burning behaviour between the control fabrics, with the high-density 151 ± 2 g/m² 100% cotton fabric having a slower time to ignition, probably owing to the reduced air flow [33], and they are discussed further in the horizontal flammability test results. The polyester/cotton blended fabric compared to the 100% cotton fabrics has a much longer time to ignition, which is to be expected for a synthetic cotton blend, which is less flammable than 100% cotton and has inherent flame-retardant properties [34]. The blended fabrics in this research are used in the UK health service and care homes as BS7175-certified fire-resistant bedding (source 5 and 7) [22]. Once ignited, such blended fabrics burn intensely owing to the scaffolding effect [35], whereby the cotton (cellulose) char supports the melting and eventually burning of the polyester. Therefore, it is likely that this char layer would be irregular in distribution on the surface of the fabric strips and could partially explain why there is large variation in the test results of flame glowing time.

The test results of flame time generally suggest that even with the additional fuel contaminant on the fabrics, the flame time is generally very similar to the blank fabric. This indicates that the flame on the contaminated fabrics burns with greater intensity.

The large variation in the ointment test results could in part be attributed to the challenges of spreading and applying the more viscous ointment evenly on the fabric samples. As an example, some differences are shown in Figure 4a, but the lotions and creams on the 100% cotton fabric (114 ± 1 g/m²) generally result in quicker flame times than the ointments. The ointments contain a larger percentage of light paraffin and white soft paraffin and therefore contain more and higher molecular weight saturated hydrocarbons, which would take more time to form a volatile mixture during burning.

The % mass loss after drying was higher (Table 4) with lotions and non-paraffin-based products, resulting in more dried fuel on the textile strips. However, the results show no relationship with the amount of water loss or differences in the dried mass of fuel in the vertical flammability test results. Although the glowing times and total burn time for tests with cream 9 were higher compared to the other creams tested, and this cream had smaller loss upon drying compared to the other creams (Table 4), this was the only product that contained zinc oxide (Appendix A Table A1), which is often added to emollients as a sunscreen due to its high refractive index [36] and to reduce infections and inflammation owing to its antibacterial characteristics [37]. Zinc oxide also possesses flame-retardant properties and is used as a microparticle in fabrics [38,39] and therefore could have facilitated the longer total burn and glowing times seen, but no effect was observed on the time to ignition.

4.2. Horizontal Flammability Tests

Figure 5 shows that there is a general upward trend of increasing flame speed for the lotions and paraffin-free products compared to the blank fabrics. A ‘drop’ in speed occurs in the results when a cream is present but then increases in velocity with the increasing paraffin content of the creams; however, this trend reverses with the ointments. Despite the higher water loss after drying for fabrics contaminated with lotions and paraffin-free products compared to the creams and ointments, when examining the effect of water loss on flame speed, no trend was found. When comparing horizontal flame speed between the different emollients, generally, the ointments had a slower horizontal flame speed. This could be explained by their higher liquid paraffin and white soft paraffin content, and containing more high-molecular-weight compounds that would take longer to heat and form volatiles at the flame front. The results show that although the volatiles produced would rise vertically and less heat transfer occurs from the flame [32] during the horizontal flammability tests, there is still an increase in the flammable mixture that increases the horizontal flame speed and flame height of the fabrics with an emollient present. The flame heights do follow an overall upward trend (Figure 6). These increases could also be explained by the increasing paraffinic content (Appendix A Table A1) and mass loss on drying (Table 4) which subsequently causes differences in the amount of dry fuel left. With lotions losing between 70 to 74%, non-paraffin creams losing 66 to 74%, creams losing 37 to 71% and ointments losing 0.2 to 3% of their mass, this therefore results in different amounts of fuel available between the lotions, non-paraffin creams, paraffin-based creams and, in particular, the ointments when tested. The methods used are designed to mimic a recommended volume applied to the skin by a patient [17,18] and are therefore repeatable within the tests themselves. This is noted as a limitation of the contaminated textile strip preparation, and it could be suggested that by knowing the mass loss of each emollient, this could be reflected in the volume applied to the textile, resulting in comparable dried amounts of each emollient being applied, although, as discussed above, this does not seem to have any effect on the time to ignition of the vertical flammability test results.

It is worth noting the difference in flame speed between the two cotton fabrics, where a slower flame speed and lower flame height were found for the denser 100% cotton fabric (Table 5 and Table 6), even when an emollient was present. The denser fabric would have less air circulation and permeation [29,30] and a slight reduction in the fuel and air mixture in the stoichiometric limits/ratios needed to support a flame, resulting in a slower burn time and smaller flame height. The significant reduction in ignition times when an emollient was present is largely explained by this additional fuel. The difference in the flammability of the fabrics chosen is affected by both their limiting oxygen index (LOI) and irregularities in the direction of warp and weft, in addition to damage during manufacturing and differences in surface properties such as neps [29]. Therefore, future research on the irregularities of the fabrics and the potential effect on the flammability of contaminated fabrics could be of merit and might aid in the reduction of variability in test results.

5. Conclusions

It is important to emphasise that these products are important medicines in the treatment of skin conditions and disease and the intention is not to deter their use but to instead improve communication of the risks involved.

The impact of emollient-contaminated fabrics on fire behaviour in both vertical and horizontal orientations has been demonstrated. This is the first time that such comprehensive testing (14 common emollients) has been conducted using vertical and horizontal testing as a direct method of comparison. The results demonstrate an increase in the observable fire development on 100% cotton and 52/48% polyester/cotton blend fabrics in both horizontal and vertical orientation in the presence of dried-on emollients. The results also show a significant reduction in the time to ignition of emollient-contaminated fabrics compared to blanks for all 14 emollients evaluated. The flame times of the emollient-contaminated fabrics were also generally lower. Similar or slightly shorter glowing times were observed for all fabrics, except for the denser cotton fabric, where a slightly longer time was observed. Horizontal flammability tests for cotton showed increases in both flame height and flame speed, with flame heights up to four times greater with the addition of a dried-on emollient. The paraffin content and the volume of emollient applied did not impact the results.

These factors may strongly decrease the reaction time of an impacted individual, resulting in more severe injury or increased fatality rates. The fire retardancy of materials is affected when contaminated with emollients and therefore does not reduce the risk. Further testing is required in relation to examining more flame-retardant materials.

It is imperative to note that this study involves one application of emollient onto the fabric surface. NHS guidance recommends emollients be applied three or four times a day, which results in build-up on clothing and increases transfer to other surfaces. The risk of multiple applications needs to be evaluated.

The message to those who use emollients should be to let the emollient dry on the skin before applying dressings, putting on clothing or going to bed, avoid open flame ignition sources and regularly change clothing to prevent build-up. Healthcare providers and fire and rescue professionals should be advised to document incidents through reporting systems such as the MHRA Yellow Card Scheme in the UK to improve recording and hence awareness.