Core-sheath Electrospinning of Shea Butter and Cellulose Acetate to Enhance Heat Transfer in Protective Clothing

4 Protective clothing for health workers requires heat transfer in hot and humid 5 environments. To study the thermal conduction of phase-change materials and protect them from 6 leakage, we selected skin-friendly shea-butter due to its suitable melting temperature, and the 7 electrospinning processibility of biocompatible cellulose acetate. The shea-butter as a phase- 8 change material was encapsulated in electrospun cellulose acetate fibres within a core/sheath 9 structure, which was stabilised by two concentric Taylor cones during coaxial electrospinning. 10 Transmission and scanning electron microscopy revealed a blood-in-tube vessel-like morphology. 11 Next, differential scanning calorimetry and thermogravimetric analyses confirmed the heat 12 capacity of shea-butter (latent heat of fusion: 42.73 J/g; thermal conductivity: 1.407 W/ m∙K) . The 13 flow rate of the core was proportional to the heat capacity of the shea-butter/cellulose acetate 14 fibres. This was consistent with the finding that the electrospun fibres of the highest-ratio shea- 15 butter (16.19%) had the highest thermal conductivity (0.421 J/ g∙K). The shea-butter:cellulose 16 acetate ratio was approximately 15:80. The efficacy of heat transfer for the core/sheath fibres in 17 human clothing was assessed by measuring skin temperatures at 13 sites in six males aged 25 to 35 18 under two conditions: wearing a mask and hood with attached cellulose acetate fibres in the 19 presence and absence of shea- butter. The mean difference in skin temperatures (0.5 ℃) between 20 the two conditions was significant. Coaxial electrospinning of shea-butter/cellulose acetate fibres 21 is therefore promising for protective clothing with efficient heat-transfer in the use of a large area.


Introduction 42
Protective clothing for COVID-19 medical workers in hot-humid environments 43 Demand is growing for smart clothing and textiles that can protect wearers working in 44 hot, humid environments by means of an integrated cooling mechanism. Due to the coronavirus 45 disease 2019 (COVID-19) pandemic, protective clothing has become essential for both patients 46 and medical personnel. Because the COVID-19 virus can be spread through inhalation of droplets 47 and aerosols, healthcare workers are required to cover their respiratory organs. Unfortunately, 48 wearing coveralls, medical gowns, N95 and filtering face masks, goggles, gloves, and powered air-49 purifying respirators for extended periods can result in heat strain and thermal discomfort (Park, 50 2020). For prolonged wear, virus-protective clothing should provide thermal comfort as well as 51 intrinsic protective functions.

53
Electrospinning of cellulose acetate for protective clothing 54 Textiles and other materials used in protective clothing are required to provide high levels 55 of functionality, which can cause thermal discomfort, leading in turn to a need for clothing that can 56 offer enhanced heat-transfer abilities. Although cellulose is one of the most abundant and skin-57 friendly of biomaterials, it is relatively vulnerable to cell propagation and cannot be easily 58 fabricated by electrospinning because it is insoluble in moderate acids and solvents (Khalf et al. 59 2015). Among cellulose derivatives, cellulose acetate (CA) is biodegradable and biocompatible 60 and can be chemically processed with various solvents to fabricate electrospun fibres with a wide 61 variety of applications, including separation membranes, tissue engineering, sensors, and catalysts 62 (Aboamera et al. 2018;Liebert 2010;Wang et al. 2020;Zhang et al. 2018). Because 63 electrospinning produces fibres with a high volume to specific-surface-area ratio, electrospun CA 64 can intensify cell attachment, corresponding to an increase in pore number and sizes (Khalf et al.

74
To enhance thermal comfort through thermal conductivity (k) mechanisms, phase-change 75 materials (PCM) can be applied to the fabrication of protective face masks. The ideal PCM 76 requires a melting point (Tm) range appropriate for its application; sufficient latent heat of fusion 77 (∆ ); sufficient specific heat (Cp) capacity; a small change in phase volume; the absence of 78 supercooling; chemical stability; high resistance to flammability, explosion, and toxicity; and high 79 resistance to corrosion of the sheath materials. Such a material must also be easily processed and 80 available at acceptable costs and quantities, according to Abhat 1983, Shchukina et al. 2018

88
The n-alkane chains of paraffin wax affect the phase transition between an isotropic liquid and a 89 well-ordered crystal. In this "metastable rotator phase", the number of carbon atoms (20 to 40) is  (2016) reported that diesters from fatty acids produced ∆ of 230 to 260 J•g -1 , which were higher 95 than ∆ of paraffins (150 to 200 J•g -1 ), with Tm of between 39 ℃ and 77 ℃ produced by reacting 96 dialcohols with methyl esters.

97
A fatty-acid methyl ester from lipids is non-toxic, environmentally friendly, renewable, 98 bactericidal, and safe for human skin, making it suitable for PCMs. Some fatty acids of PCMs 99 exhibit phase transition near skin temperatures, ranging from 28 ℃ to 35 ℃ (Sharma et al. 2015).

100
One such fatty acid, shea butter, has a Tm of 32 ℃ to 45 ℃, which is higher than that of coconut 101 oil (21 ℃ to 25 ℃). This can be attributed to the longer chains of the main components of shea 102 butter: stearic acid (41.8%) and oleic acid (46.5%) (

127
(2011) added 9 wt% polyethylene oxide (PEO) to an electrospun nanocomposite of cellulose 128 acetate to enhance its elastic modulus, elongation at break, and tensile strength by 253%, 54%, and 129 446%, respectively. The low k of polymeric sheaths can be supplemented by the addition of 130 inorganic materials such as expanded graphite, carbon nanotubes, GO, ZnO, or silica particles.

131
Adding ZnO nanoparticles of 20 wt% of PEO maximised those values by 31%, 12%, and 47%, 132 respectively, leading to a decrease in phase-change temperature by between −9 °C and 1 °C.

133
The design of protective textiles and clothing in hot, humid environments requires 134 wearing trials from the design stage. In such environments, body temperature and sweat are unable 135 to circulate out of the textiles and clothing, which activates heat dissipation by the human body to 136 improve thermal comfort through evaporation, convection, conduction, and radiation. Between the 137 skin and the textiles, the human body constantly transfers heat and moisture to keep a comfortable 138 state by controlling body and skin temperature through vaso-venodilation-constriction of arterial 139 flows to hands and feet, and evaporation of perspiration (Caldwell et al. 2014). Bedek et al. (2011) 140 reported that, when skin temperature was predicted in simulations, the temperature was affected 141 primarily by the k of underwear, and skin humidity was related to evaporation resistance,

172
Preparation and electrospinning 173 This study used a solvent system of acetone for electrospinning solutions: acetone with 174 DMF at five different acetone:DMF (v/v) ratios (4:1, 2:1, 1:1, 1:2, 1:4). In the mixed solvents, 175 electrospinning solutions of cellulose acetate were attempted at five concentrations between 17.5 176 wt% and 27.5 wt% at intervals of 2.5 wt%. As shown in Fig. 1, the CA powder was mixed 177 physically with additives equivalent to 5% of the acetate weight. The mixture of prepared powder 178 was then poured into an Erlenmeyer flask and dissolved in the mixed solvents at 55 ℃ to 60 ℃ 179 while strongly stirred by a magnetic stirrer. Following completion of the dissolution reaction, the 180 prepared solution was injected into a 10 cc syringe, and air bubbles in the syringe were removed.

181
The electrospinning device (ESR200, eS-robot, NanoNC Ltd., Korea) comprised two 182 pumps, a robot that moves the pumps along the x and y axes, and a rotary drum collector. This

218
First, 10 mg of each sample was placed in a Tzero aluminium pan. Next, the samples were 219 stabilised at 25°C in an isothermal state for 1 min. They were then heated to 50°C at a rate of 220 10°C/min, isothermalised for 1 min, and then cooled to −10 °C at 20 °C/min, after which they 221 stayed in an isothermal state for 1 min. For the second heating-cooling cycle, they were reheated 222 and recooled to between −10 °C and 50 °C, and each stage was accompanied by isothermalising 223 for 1 min. For TGA measurements, Q50 and Discovery (TA Instruments) devices were used. A 224 platinum pan was weighed without any sample and after each sample was contained, and the 225 samples were then heated from 25 °C to between 600 °C and 800 °C at a rate of 10 °C/min in a 226 nitrogen atmosphere.

228
Application to facial mask hoods through 3D scanning and 3D printing

295
Electrospun CA fibres of 25 wt% at a ratio of A:D 1:1 had a tree-leaf shape and a 296 diameter of 2.39 to 3.04 μm. In Fig. 2 (a)

310
In coaxial electrospinning using the dual nozzle, we investigated the morphological 311 characteristics of the encapsulation of shea butter (ShB) inside the CA sheath. As shown in Fig. 3 312 (a), two concentric Taylor cones of each core and sheath solution were observed when the 313 solutions were ejected from the tip of the dual nozzle. Due to Plateau-Rayleigh instability, the 314 droplet of the spinning solution, originally spherical in shape due to surface tension, was stretched 315 to an ellipse by jet stretching. Based on the TEM images in Fig. 3 (b) and (d), a cylindrical 316 morphology with small beads characterised the core-sheath fibre, and ShB was inserted into the 9 CA fibres. However, when the flow rate of sheath did not correspond to that of core, some ShB 318 leaked from the sheath, as shown in the left side the SEM image of Fig 3 (c). This may be 319 attributable to the fact that the excessively high voltage applied to the dual nozzle split the ejected 320 droplets.

321
Next, we investigated the effects of the core-sheath flow rate on the encapsulation of 322 PCM in CA during core-sheath electrospinning. To determine the appropriate ratio for the sheath 323 and the core, the inflow rate of the sheath was set from 1.0 mL/h to 8.0 mL/h, and the inflow rate 324 of the core was set to 0.5 mL/h and 1.0 mL/h. In coaxial electrospinning, the viscosity of the core 325 must be lower than that of the sheath, and the flow rate of the sheath must be sufficient to cover 326 the core. The inflow rate set by the viscosity ratio between the core and the sheath is therefore

357
(2019) pointed out that ShB was appropriate for thermal energy storage but inappropriate for a PCM because its Tm (4.3 ℃ to 15.8 ℃) was broader and its ∆ during freezing (29.9 to 41.6 J•g -359 1 ) was lower than those of palm kernel oil and Allanblackia oil. Their result of the Tm was quite 360 different from that of shown in Table 2 (Lawer-Yolar et al. 2019;Canale et al. 2005), because the 361 ShB in their study comprised the different content of a lower stearic acid (least 20%) and a higher 362 mono-unsaturated oleic acid (up to 60%). The feasibility of stearate in ShB for PCMs was 363 confirmed in studies of methyl palmitate:methyl stearate at a 4:1 ratio, the synthesis of stearic acid 364 into a porous carbonised-maize straw matrix by vacuum impregnating, and a eutectic mixture of 365 stearic acid and benzamide s of 65.9 ℃ and ∆ of 200.15 J•g -1 due to added graphite that boosted 366 k (Feldman et al. 1993;Suppes et al. 2003;Wen et al. 2021;Ma et al. 2019). We therefore 367 adopted ShB despite its large volume change during phase changes that required encapsulation of 368 its sheaths as a confined container.

369
When the results of Table 4 were scrutinised, ShB's absolute value of ∆ was found to 370 be 42.73 J•g -1 , which is a quarter of a typical ∆ of PCMs (150 to 260 J•g -1 ) in architectural 371 applications yet is also similar to that of a fatty-acid mixture (40 to 100 J•g -1 ), which was suitable

414
To measure the mass ratio of ShB and CA, TGA and DTG thermograms were made in a 415 range of 300 ℃ to 500 ℃ as shown in Fig. 5 (a) and Table 5. ShB was thermally degraded from  (Table 2). However, the ratio of palmitic acid (C16:0, 3.3% to 3.9%) and 419 TiO2 and ZnO additives in ShB was nearly imperceptible and they were excluded from the TGA 420 results presented in this study. The major component of stearic acid (C18:0) in ShB could be 421 distinguished from other fatty acids as well as the CA fibres. The ratio of sheath to core was 422 approximately 77.5-80.0 to 16.2-13.2. We therefore inferred that the B-1 sample had the largest 423 amount of ShB core and the least ShB was contained in the C-2 or C-1 sample. The ratio obtained 424 from the TGA results was in agreement with the DSC results that show the highest Cp were found 425 in B-1T (0.421 J•g -1 ·K -1 ), and the lowest ones in C-1 (0.111 J•g -1 ·K -1 ). Fig. 5 (b) and (c) represent 426 the ratio of C18:1 and C18:2 at 300 ℃ and that of C18:0 at 400 ℃ on the left side of the enlarged 427 thermograms, consistent with Table 5.

429
Human-wearing tests through paired differences in skin temperature in contact with electrospun 430 fibres with or without PCMs

431
Not only was the heat-transfer rate at the fibre level analysed in this study, we also 432 attempted to verify the difference between the heat-transfer capacity of dual-structured PCM-CA 433 electrospun fibres and that of single-structure CA fibres applied to the human body. To increase 434 the k of PCM, the skin and the sample had to be in close contact. We therefore needed to design a 12 new hood and protective mask as the existing ones were not suitable for the thermal conduction of 436 PCMs. Specifically, existing D-level protective clothing could not be used because PCM 437 performance is realised only when it comes into direct contact with the skin. Fig. 6 (a)

451
Next, after walking at 5 km/h on the treadmill for 20 min, they sat again and rested for 15 minutes.

452
With the collected data of skin temperature and sweat, we found a significant difference in the 453 effect of PCM cooling performance by comparing the skin temperature at the 13 sites, including 454 the inner ear, with a paired t-test using SPSS.   Second, DSC results revealed that the heat capacity of electrospun PCM-CA fibres was a 531 half to a quarter of that of ShB, with a ∆ of 42.73 J•g -1 , a Cp of 1.90 J•g -1 •K -1 and k of 1.407 532 W•m•K -1 . The sheath flow rate was inversely proportional to the heat capacity of the PCM-CA 533 fibres, as measured by ∆ , Cp, and k. This can be attributed to the thickness of the sheath wall, 534 which corresponded to the flow rate of the sheath. In contrast to the flow rate of the sheath, the 535 relationship between the flow rate of the core and the heat capacity was proportional. In TGA 536 results, the thermal degradation peak of ShB was 461.14 ℃ (due to its major component of steric 537 acid) and that of CA was 389.61 ℃ a result of mixing with the palmitic acid. The ratio of CA to 538 ShB was inferred to be as high as 78% to 80% and as low as 16% to 13%. The electrospun PCM-539 CA fibres with the highest ratio of ShB as shown in TGA coincided with the fibres with the largest 540 k of 0.421 J•g -1 •K -1 in the DSC results.

541
Finally, in the human-wearing assessment, wearing a hood with PCM-CA fibres (the 542 experimental condition) decreased mean skin temperature in five of six subjects by 0.5 ℃, 543 compared to attaching only CA fibres (the control). During the walking exercise, the difference in 544 the two conditions gradually rose to 0.25 ℃. At 70-75 min during the second sitting-rest after 545 walking, the average mean skin temperatures sharply dropped, resulting in the biggest gap between 546 the two conditions, as supported by the paired t-test (p < 0.05). In conclusion, the effectiveness of 547 PCM-CA fibres was validated as it delayed the increase of skin temperatures.

548
This study successfully encapsulated PCM by coaxially electrospinning a core-sheath 549 structure. It was also confirmed the potential of the technique for thermoregulating protective 550 clothing. We found it is necessary for the core to increase latent heat as well as for the sheath to TGA results for shea-butter and PCM-CA bers in a range of (a) 300 to 500 , (b) 300 to 400 and (c) 380 to 480  This is a list of supplementary les associated with this preprint. Click to download.