Effects of ambient temperature step changes on the heat storage and release in thermal protective clothing

Abstract

Ambient temperature steps between typical non-work and hot work environments will lead to clothing heat storage and release, which is crucial for human health and thermal comfort. In this study, the influence of five types of ambient temperature steps (from 0ºC/5ºC/10ºC/15ºC/20ºC to 40ºC to 0ºC/5ºC/10ºC/15ºC/20ºC) on the heat storage and release in clothing was investigated using a thermal manikin. A moving and temperature-controlled refrigerated incubator was constructed to realize the temperature step conditions. Results showed that ambient temperature step magnitude was positively related to heat storage and release. Some 80% of the amount of heat storage would be completed in around 10 min after the temperature steps. Increasing the inner clothing layer weight and specific heat capacity for local clothing at the chest, thigh and calf improved their heat storage against the large temperature step conditions. As for the abdomen, its heat storage was the largest. However, the clothing layer configuration impacted its heat release, so it was necessary to guide the heat transfer to the skin. The heat storage in the outer layer was increased by 2% to 39% due to adding the inner layer of the garment. The stored heat discharged completely until the temperature step magnitude exceeded 32.5°C. A multiple linear regression formula was proposed to obtain the clothing average heat storage by considering ambient temperature step magnitude and clothing insulation. The results of this study could contribute to the optimization of thermal protective clothing and improvement of the research on human thermal comfort.

Keywords

Temperature step heat storage heat release clothing insulation thermal manikin

Heat-exposed workers not only suffer from long-term hot environments but also face sudden ambient temperature changes due to the shift to non-work environments. For example, firefighters always encounter temperature alteration during the assignment process, especially in winter. Furthermore, the use of some cooling interventions, such as air conditioners and fans, after high-temperature work may also create a sudden temperature change. Temperature step change is described as a process from one steady environment to another steady environment.¹ The large temperature step magnitude between two environments affects human health and thermal comfort.^1–6 As the only barrier between the skin and the environment, thermal protective clothing can store and discharge heat, benefitting human thermal regulation.⁷ The more thermal energy is stored in clothing and the longer the heat storage time is, the fewer and slower the heat transfer to the human body.⁸ Therefore, it is fundamental to study the effect of ambient temperature steps on clothing heat storage and release to improve the design of clothing thermal management and human thermal comfort.

Many methods for measuring the heat storage and release of clothing materials have been provided. From the perspective of fabric, Wan and Fan⁹ reported a guarded hot plate system to study the heat storage and release of fabrics. Later, Safavi et al.¹⁰ proposed a test procedure to simulate a wearer moving from an indoor environment to an outdoor subzero climate using the hot plate system. However, there was only one test chamber in their studies, which makes the repeatability of dynamic conditions difficult. Kim et al.¹¹ addressed this problem and provided a ‘human–clothing–environment’ simulator, including a hot vertical plate and two close climate chambers that were individually controlled. The heat storage and transfer in phase change material¹² and a multilayer fabric system¹³ were studied by the ‘human–clothing–environment’ simulator. It suggested that ambient temperature step magnitude is an essential factor influencing both the temperature and humidity of fabric microclimate. There was an average elevation of 0.3°C in the fabric microclimate temperature with every 1°C increase in the environment.¹³ In addition, a 6°C difference in ambient temperature compensated for a 45% difference in ambient relative humidity (RH).¹³ However, the study of the thermal performance of fabric cannot replace that of clothing completely, given the discrepancy in clothing structures or designs that affect the heat transfer.¹⁴

From the perspective of overall clothing, the thermal manikin is the most realistic device that is widely used to assess heat transfer between the human body and the environment.¹⁵ The two modes of the thermal manikin, the constant temperature mode and constant heat flux mode, are usually employed in transient conditions.^16,17 However, only the constant heat flux mode can be used in high-temperature environments since the general limitation of the thermal manikin is that it is not equipped with an active cooling system to simulate heat gains.¹⁵ Lee et al.¹⁸ used a thermal manikin to measure the temperature changes of the clothing microclimate under a dynamic environmental temperature from 0°C to 35°C. However, the limitation of this study is that the speed of shifting environmental temperature is uncontrollable, which causes every test time to be too long, around 11 h. Furthermore, this condition is more like temperature ramps instead of temperature steps between two different steady environments. Kim and Kim¹⁹ detected the heat storage of local clothing using light emission apparatus, but this temperature step condition is non-uniform. Therefore, the key point of the overall clothing study is to construct ambient temperature step change conditions based on the thermal manikin system.

Furthermore, the temperature changes of the clothing microclimate or skin were usually used to evaluate the thermal properties of clothing,^17,20 but these indices did not directly quantify the clothing heat storage and release. He et al.⁸ proposed a quantitative assessment of the thermal stored energy in fabric systems to discuss the effects of material properties.^21,22 The influence of the fabric system weight and thickness were investigated, but the contributions of multi-factors on the heat storage were not analyzed. Su et al.²³ applied this method to calculate the heat storage and release in phase change material under hot surface contact conditions. However, these studies mainly focus on fabric systems with huge temperature differences under thermal radiation or hot surface contact conditions, which is quite a lot larger than actual ambient temperature steps. In addition, the air gap between the fabric and skin increased the heat storage in the fabric system from 0.8% to 54.5%⁸ and extended heat transfer time to the skin.²⁴ Hence, the air gaps under actual wearing situations should be considered for heat transfer and storage in clothing. To examine the relationship between thermal resistance and heat storage, a positive linear relationship was proposed with an adjusted R² value of 0.97.⁸ Nevertheless, the thermal resistance mentioned in the study is not the experimental result with traditional measurements, such as ASTM F 1291: 2004 or ISO 15831: 2004, but rather that calculated under thermal radiation conditions. It may overestimate the actual thermal resistance. Therefore, it is necessary to quantify the heat storage in clothing to study the effects of ambient temperature steps with considerations of material properties and clothing thermal properties.

This work aims to conduct manikin experiments to study the effects of ambient temperature steps between typical outdoor and hot work environments on the heat storage and release in clothing. Five types of temperature step conditions (from 0ºC/5ºC/10ºC/15ºC/20ºC to 40ºC to 0ºC/5ºC/10ºC/15ºC/20ºC) and two thermal protective clothing ensembles were investigated for this test. The research findings will allow the possibility of optimizing the clothing thermal management and increasing the knowledge of human thermal comfort under transient conditions.

Materials and methods

Materials

Clothing ensembles

The thermal protective clothing was in line with GA 10-2014, which is commonly comprised of an outer shell (OS), a moisture barrier (MB) and a thermal liner (TL). Two ensembles of thermal protective clothing (175/92A) were used to investigate the difference in clothing layer configuration; E1 consisted of one single OS jacket and cotton underwear. E2 consisted of three layers of clothing (OS + MB + TL) and the same underwear.

Fabrics

The basic properties of every single layer in thermal protective clothing and underwear are listed in Table 1. The mass per unit area was tested on an electronic scale tester, which conformed with standard ASTM D3776-96. In accordance with standard ASTM D1777-96, the thickness of the test specimen was measured under a pressure of 1 kPa. The air permeability of the fabric was measured under a pressure drop of 100 Pa using a Frazier Air Permeability Tester, according to ISO 9237. Fabric specific heat was supported by the measurement of He et al.⁸

Table 1.
Basic properties of fabric in each single layer

Fabric type Fabric content Weight (g/m²) Thickness (mm) Density (kg/m³) Air permeability (mm/s) Specific heat at 20°C (J/(kg°C))

OS 100% X-Fiper (meta-aramid) 214.9 0.54 396.6 76.2 1300

MB X-Fiper (meta-aramid)/PTFE 108.2 0.52 208.5 0 1080

TL 100% X-Fiper (meta-aramid)/flame-resistant viscose 283.1 2.18 129.7 493.9 1110

Underwear 100% Cotton 194.0 0.71 273.2 552.9 1225

OS: outer shell; MB: moisture barrier; TL: thermal liner; PTFE: polytetrafluoroethylene.

Thermal insulation properties of clothing

The thermal insulation of these two ensembles was measured by a 34-segment Newton thermal manikin (Measurement Technology Northwest, Seattle, WA, USA), which was enclosed in a climate chamber during the tests (Figure 1). Measurements followed ISO15831-2004. A constant skin temperature mode (34 ± 0.2°C) was employed. The air temperature was controlled at 20 ± 0.5°C. The RH was 50 ± 5% and the wind speed was 0.4 ± 0.08 m/s. The test clothes were in the climate chamber for at least 12 h. Each ensemble was tested at least three times, and the variability was supposed to be lower than 4%.

Figure 1.
Clothing thermal insulation measurement by the thermal manikin: (a) underwear and (b) E2.

The measured thermal insulation of the two clothing ensembles is listed in Table 2. The total thermal insulation of the single-layer clothing E1 was 1.62 clo, lower than that in multilayer clothing E2 (2.28 clo). The largest and lowest local insulation were respectively at the abdomen and calf.

Table 2.
Total thermal insulation of clothing

Clothing ensemble Total thermal insulation (clo)

Chest Abdomen Forearm Thigh Calf Overall

E1 1.60 ± 3.5% 3.56 ± 5.6% 1.25 ± 2.4% 1.78 ± 2.4% 1.14 ± 0.9% 1.62 ± 2.1%

E2 1.65 ± 4.1% 6.77 ± 6.5% 1.79 ± 1.5% 2.65 ± 4.0% 1.60 ± 0.7% 2.28 ± 2.5%

Note: 1 clo = 0.155 m² · °C/W.

Experimental conditions and facilities

Experimental conditions

Two independent environments were undertaken to achieve ambient temperature step conditions. The hot work environment was simulated by the climate chamber (length: 5 m, width: 5 m, height: 2.5 m), in which the air temperature was set at 40°C. The non-work environment was simulated by a moving refrigerated incubator (length: 0.9 m, width: 0.6 m, height: 2 m), which was placed in the chamber. The air temperature in the refrigerated incubator was set at 0ºC/5ºC/10ºC/15ºC/20ºC. By this means, five temperature step conditions (0ºC/5ºC/10ºC/15ºC/20ºC to 40ºC to 0ºC/5ºC/10ºC/15ºC/20ºC) were conducted. Each condition included three phases, in which phase 1 meant the first time in the non-work environment, phase 2 meant a shift to the hot work environment and phase 3 meant going back to the non-work environment.

The air temperature in the chamber and incubator were measured by a thermistor sensor during the test. The thermistor sensor with an accuracy of ±0.1°C (Betatherm/MTNW, Ireland) provided by the manikin system was placed 130 cm above the ground according to ISO 15831: 2004 and 10 cm in front of the manikin according to the research indicating that the thermal boundary layer of the thermal manikin is within 4 cm.^25,26 Table 3 summarizes the measured air temperature. It can be seen that the air temperature of the incubator was a little higher than the set value, especially for 0°C. Hence, the measured values were used for the following analysis. RH in the chamber was kept around 50 ± 5%, and the air velocity was controlled to lower than 0.4 m/s.

Table 3.
The measured ambient temperatures in the refrigerated incubator and the chamber

Condition Phase 1 (°C) Phase 2 (°C) Phase 3 (°C)

0°C–40°C–0°C 3.2 (0.4) 39.7 (0.2) 3.1 (0.9)

5°C–40°C–5°C 6.9 (0.4) 39.4 (0.2) 6.0 (0.3)

10°C–40°C–10°C 11.7 (0.1) 39.5 (0.2) 11.5 (0.3)

15°C–40°C–15°C 16.6 (0.2) 39.4 (0.3) 16.9 (0.2)

20°C–40°C–20°C 20.6 (0.2) 39.4 (0.3) 20.1 (0.4)

Note: average value (standard deviation); 0°C–40°C–0°C means ambient temperature step from 0°C to 40°C to 0°C, and the same format for other conditions.

Refrigerated incubator

A moving and temperature-controlled refrigerated incubator was constructed to realize temperature step conditions. The refrigerated incubator covered the manikin at the beginning. Ambient temperature up step conditions could be actualized by opening and moving out of the incubator and exposing the manikin to the climate chamber. Correspondingly, pushing the incubator back to cover the manikin again could achieve temperature down step conditions.

The refrigerated incubator comprised an insulation can and refrigeration equipment, as shown in Figure 2. The plate of the insulation can adopted a sandwich structure according to ISO 5155, that is, polyurethane insulating material (thermal conductivity <0.022 W/(m · k)) with 50 mm thickness in the middle and a stainless plate with 1.2 mm on both sides. The double doors and the hole in them were designed to pass through the support structure of the manikin. There were six wheels at the bottom of the insulation can. The refrigeration equipment parts, such as the compression and condenser, were placed on the right-hand side of the insulation can.

Figure 2.
Schematic diagram of the refrigerated incubator.

A circulation system was connected by the refrigeration components through pipelines. Tetrafluoroethane (R-134a) refrigerant was used in the study, and the capacity of the incubator was 836 L. According to the required refrigeration capacity, the design saturated suction temperature and the design saturated discharge temperature were selected.^27,28 As a result, the compressor rated power was 735W. A copper tube of 12 mm diameter was used for the evaporator, and the total length of the tube was 7.41 m. A thermistor sensor with an accuracy of ±0.1°C (Heraeus, Germany) was set at 0.94 m from the bottom plate, half of the manikin height (1.68 m). In addition, a ventilation tube of 0.16 m diameter with aluminum material was placed in the refrigerated incubator to enhance the homogeneity of ambient temperature in the incubator. Two fans (rotational speed: 4000 rpm) were arranged at the bottom of both sides of the ventilation tube and there were six holes on the top of the tube, which realized the up and down wind circulation in the refrigerated incubator. A wind speed controller connecting to the fans kept the wind speed at the air outlets under 0.4 m/s.

Calculation and measurement of heat storage and release

Calculation of heat storage and release

The accumulative heat storage in clothing was calculated by the variation of clothing surface temperatures,⁸ written as
$Q_{c l, j} (t) = \sum_{i = 1}^{n} ρ_{i} c_{i} L_{i} \frac{Δ T_{out, i} (t) + Δ T_{i n, i} (t)}{2}$
(1)
where Q_cl, j (t) is the accumulative heat storage at local parts of clothing, kJ/m²; ρ is the density of the fabric, kg/m³; c is the specific heat capacity of the fabric, J/(g°C); L is the thickness of the fabric, mm, shown in Table 1; ΔT_out,i (t) and ΔT_in_,_i (t) are temperature variations of the outer and inner surfaces, respectively, of each clothing layer in the duration of time t, °C; the subscript i refers to the clothing layer; n is the total number of layers in a clothing ensemble; the subscript j refers to the local clothing part. Hence, the clothing surface temperatures at each clothing layer should be obtained.

The mean heat storage (Q_cl (t)) was calculated according to the weighted formula
$Q_{c l} (t)= \sum_{j = 1}^{m} \frac{A_{j}}{A_{tot}} Q_{c l, j} (t)$
(2)
where A_j is the local body surface area, m²; A_tot is the total body surface area, m²; the subscript j refers to the local clothing part; m is the total number of local parts, which here is 5.

The heat storage rate can be calculated
$q_{c l, j} (t) = \sum_{i = 1}^{n} ρ_{i} c_{i} L_{i} \frac{Δ T_{out, i} (t) + Δ T_{i n, i} (t)}{2 Δ t}$
(3)
where q_cl, j (t) is the heat storage rate at the local parts of the clothing, kW/m²; Δt refers to the sampling time interval, s.

The average heat storage rate (q_cl(t)) can be calculated
$q_{c l} (t) = \sum_{j = 1}^{m} \frac{A_{i}}{A_{tot}} q_{c l, j} (t)$
(4)

Measurement of clothing surface temperature

To determine the heat storage, clothing surface temperatures were measured. Figure 3 illustrates the measuring points (red dots) at five local parts: the chest, forearm, abdomen, thigh and calf. It was worth noting that the clothing layer configuration at the abdomen and forearm is the overlapping multilayer and double-layer structure, respectively. K-type thermocouples (Omega Engineering, USA; accuracy: ±0.5°C) with a diameter of 0.127 mm were used to measure the clothing surface temperatures. According to the study of attachment methods, a thin aluminum tape fixed these sensors to the data collection system.²⁹

Figure 3.
Locations of thermocouple sensors at five parts of clothing E1 and E2. OS1: outer shell layer of the upper garment; MB1: moisture barrier layer of the upper garment; TL1: thermal liner layer of the upper garment; OS2: outer shell layer of the pants; MB2: moisture barrier layer of the pants; TL2: thermal liner layer of the pants. (Color online only.)

Figure 4 describes the fixing approach of each sensor. It used a thread (red line) to pass through multilayers of clothing systems to identify the inner and outer measurement points. The measurement point was fixed with tapes and the sensor line was sewn with a thread (yellow line) every 30 cm.

Figure 4.
Fixing approach of sensors. OS: outer shell; MB: moisture barrier; TL: thermal liner. (Color online only.)

Simulation of interlayer temperature in multilayer clothing

The interlayer temperatures in multilayer clothing E2 were also needed for calculating heat storage according to Equation (1). The one-dimensional heat conduction model along the direction of fabric thickness was conducted to obtain the temperature distribution in each local part of the clothing by means of COMSOL Multiphysics 5.4a (COMSOL, Inc., USA). The boundary and initial conditions used the measured clothing inner and outer surface temperatures, recorded in real-time.

To verify the simulation data, two additional temperature sensors were added on the outer surface of the MB layer of clothing E2 during the ambient temperature step of 5°C–40°C–5°C. The standard deviation (SD) of the experimental data and the root mean square deviation (RMSD) and the bias of the simulation for the outer surface temperature of the MB layer at different clothing parts are shown in Table 4. The definition of RMSD and bias are explained in Equations (5) and (6). The RMSD serves to evaluate the model’s goodness of fit, and the bias describes the model’s accuracy.¹⁵ The fit is considered acceptable when the RMSD is smaller than the SD of the experimental data. The bias should be equal to or close to zero or not beyond the SD of the given data set to ensure unbiased model prediction. The RMSD of this simulation was lower than the SD, and the bias was close to zero. Therefore, this simulation method could be used for other ambient temperature step conditions to obtain the interlayer temperature in clothing E2
$RMSD = \sqrt{\frac{{\sum (x_{measured} - x_{predicted})}^{2}}{n}}$
(5)

$bias = \frac{\sum (x_{measured} - x_{predicted})}{n}$
(6)
where x_measured is the measured value, x_predicted is the predicted value and n is the total number of data points.

Table 4.
Results of experiment and simulation for outer surface temperature of the moisture barrier layer at 5°C–40°C–5°C (unit: °C)

E2_chest E2_forearm E2_abdomen_coat E2_abdomen_pants E2_thigh E2_calf

SD 1.12 1.51 1.28 0.64 1.02 1.43

RMSD 0.78 1.01 0.89 0.35 0.86 0.84

Bias 0.02 0.19 0.77 0.10 0.20 0.17

Experimental procedure

Figure 5 shows the experimental schedule. Firstly, 12 thermocouples were attached to the test clothing, and it was dressed on the thermal manikin. Then, the clothed manikin was put into the refrigerated incubator, the chamber and incubator were turned on and the constant skin temperature (34°C) mode was set for the manikin to ensure the same initial value of the skin temperature. It took around 60 min for preparation until the ambient temperature in the chamber and incubator reached stability. After the manikin skin temperature reached 34°C and fluctuated at 0.2°C within 20 min, the test started and the constant heat flux (0.058 kW/m²) mode was used during the whole test to simulate the rest state of the human body according to ISO 8996: 2004. The clothed manikin was kept in the refrigerated incubator for 30 min (phase 1), and then moved out of the incubator and the dressed manikin was moved into the hot chamber for 60 min (phase 2). Finally, the incubator was returned to cover the clothed manikin for 60 min (phase 3). During the experiment, five local skin temperatures of the clothed manikin were recorded every 10s, while the clothing layer temperatures of five corresponding parts were recorded by the data collection system every 10 s. Each condition was tested three times and the average value was obtained.

Figure 5.
Experimental protocol.

Statistical analysis

The SPSS version 22.0 statistical package program (IBM, Armonk, NY, USA) was used for statistical analysis, in which paired-sample t-tests were used to verify whether there were significant differences. To reveal the impact of temperature step magnitude, one-way variance analysis was used to analyze the normally distributed data, as well as the post-hoc test. Multivariate analysis of variance and linear regression analysis were applied to determine the multi-factors contributing to clothing heat storage. The level of significance was set at p < 0.05.

Results and discussion

Process of heat storage and release

The average accumulated heat storage and its rate of clothing E1 and E2 under five types of ambient temperature step conditions are illustrated in Figure 6. Clothing absorbed heat from the environment in temperature up steps (30th–90th min), whereas it discharged heat after temperature down steps (90th–150th min). Since the change of heat storage or release was different earlier and later, the entire process was divided into four periods.
Rapid growth period. This period occurred in the first few minutes after temperature up steps. The clothing absorbed heat so that the average accumulated heat storage increased rapidly. The peak values of the heat storage rate (E1: 0.107 kW/m², E2: 0.114 kW/m²) were reached in a short time (E1: 0.5 min, E2: 1.7 min), as shown in Figures 6(b) and (d). After that, the heat storage rate slumped but was still greater than zero, so the heat storage continued to rise. The rapid growth period can be determined by the heat storage rate. This period was completed when the heat storage rate dropped to that level before the temperature up steps. The average duration of the rapid growth period of clothing E1 and E2 was 7.6 and 11.8 min, respectively. The proportion of heat storage in rapid growth periods to that of the total 60 min was nearly 81% for clothing E1 and 85% for clothing E2.

Slow growth period. The rest of the exposures to the 40°C environment in the second period were around 52.4 min for clothing E1 and 48.2 min for clothing E2, respectively. Average accumulated heat storage grew slowly in this period, inconsistent with the previous study that showed no more extended stored heat after the rapid growth period.⁸ This discrepancy was attributed to the manikin system's constant heat flux of 0.058 kW/m² ignored by the previous study. In addition, the accumulated heat storage significantly (p < 0.001) increased with the enlargement of the ambient temperature step magnitude. For clothing E1, the accumulated heat storage under the five conditions was almost distributed uniformly, whereas there was a grouping phenomenon for clothing E2. The reason may be related to the temperature distribution of multilayer clothing under transient conditions.³⁰

Rapid decline period. The third period was observed after temperature down steps. The heat release decreased greatly, and the peak values of the heat release rate (E1: 0.046 kW/m², E2: 0.066 kW/m²) were reached quickly (E1: 0.8 min, E2: 2.3 min). The peak value of the heat release rate was lower than that of the storage rate. This was caused by the experimental process such that the stability of air temperature in the refrigerated incubator was affected when returning the incubator to cover the manikin. However, it took more time (E1: 8.2 min, E2: 13.4 min) to go back to the level of the heat release rate before temperature down steps. The proportion of heat release in rapid decline periods to that of the total 60 min was nearly 79% for clothing E1 and 84% for clothing E2.

Slow decline period. The fourth period, the rest of the low-temperature environment exposure, was around 51.8 min for clothing E1 and 46.6 min for clothing E2. The heat release decreased slowly, and the final values at the 150th min were greater than the initial values at the 30th min. This indicated that the heat storage during temperature up steps did not discharge entirely in temperature down steps. However, there was an exception for clothing E1 with the 0°C–40°C–0°C condition. This result suggested that both the ambient temperature step and clothing type affected the heat release.

Figure 6.
The average accumulated heat storage in clothing E1 (a) and E2 (c) and the heat storage rate of clothing E1 (b) and E2 (d) under five types of ambient temperature step conditions.

Effect of ambient temperature step magnitude on heat storage

To explore the influence mechanism of ambient temperature step magnitude on heat storage, the accumulated heat storage in each clothing layer at five local parts within 1 h after temperature up steps (the accumulated heat storage at the 90th min minus the initial one at the 30th min) was calculated, as shown in Figure 7.

Figure 7.
The accumulated heat storage at five local parts of clothing under the ambient temperature step: (a) average; (b) chest; (c) abdomen; (d) forearm; (e) thigh and (f) calf.

The heat storage was augmented with the increase of ambient temperature step magnitude. A simple linear relationship between heat storage and ambient temperature step magnitude is analyzed in Table 5. As for the local parts of clothing, the abdominal linear coefficient was the highest, followed by the forearm, calf, thigh and chest. Since the overlapping multilayer structure was at the abdomen and the double-layer structure was at the forearm, the heat storages at the abdomen and forearm were significantly higher than in other parts. The heat storage at the calf was also high due to the lowest initial temperature at the 30th min, which indicated it released some thermal energy before temperature up steps.

Table 5.
Simple linear relationships between ambient temperature step magnitude and heat storage

Average
Chest
Abdomen
Forearm
Thigh
Calf

E1 E2 E1 E2 E1 E2 E1 E2 E1 E2 E1 E2

Coefficient 0.24 0.47 0.15 0.36 0.34 0.62 0.33 0.56 0.18 0.34 0.21 0.49

Constant 1.52 5.67 0.86 4.20 2.49 10.49 2.25 5.03 1.10 4.38 0.83 3.86

Adjusted R² 0.98 0.97 0.98 0.95 0.99 0.97 0.93 0.93 0.99 0.92 0.99 0.95

The heat storage in different layers of clothing E2 is shown in Figures 7(b)–(f). Generally, the heat storage in the OS layer was the largest, followed by the TL layer and the MB layer, which was in agreement with the previous study.⁸ At the forearm, the heat storage in the OS layer was the largest, accounting for 56–58%. The reason was related to the double-layer structure of the outer layer. There was no significant difference between the heat storage in the OS layer and the TL layer at the chest, abdomen and thigh, accounting for 40–44%. A large amount of heat storage in the OS layer was due to the direct heat exchange with the environment and the large specific heat capacity.²² A large amount of heat storage in the TL layer was due to its weight and thickness,⁸ especially as the thickness of the TL layer was four times that of the OS layer, as shown in Table 1. However, the heat storage in the TL layer at the calf part of the clothing was slightly higher than that in the OS layer at 5%. This may be the result of multiple factors that are analyzed later. In addition, the heat storages in the MB layer at all local parts were the smallest, accounting for 12–16%, because its weight and specific heat capacity were small.

The location of each layer mattered for heat storage as well, according to Figure 7(c): E2_OS1 > E2_OS2, E2_MB1 > E2_MB2, E2_TL1 > E2_TL2. The amount of heat storage mainly depended on the clothing temperature changes. According to the temperature distribution of unsteady conduction heat transfer, the temperature change becomes smaller along the direction of fabric thickness.³⁰ Therefore, the closer to the environment for the same material layer, the greater the heat storage. Moreover, the heat storage in the same material but farther away from the environment reduced from 30% to 50% as the ambient temperature step magnitude increased from 19.2°C to 36.3°C. Specifically, the heat storage in the OS, MB and TL layers farther away from the environment decreased from 30% to 40% when the ambient temperature step magnitude varied from 19.2°C to 32.9°C, while the decrease percentage was TL 50% > MB 49% > OS 43% when the ambient temperature step magnitude altered from 32.9°C to 36.3°C.

Furthermore, comparing E1 (one OS layer clothing) and E2_OS1, it was found that different clothing configurations caused a discrepancy in the heat storage, even with the same location and material. In Figures 7(b) and (f), the heat storage of E2_OS1 was increased 2% to 39% higher than that of E1. The reason was that the inner clothing layers of E2 had thermal resistance, which weakened the heat transfer and increased the heat storage in the outer clothing layer. This result was consistent with the fabric study such that the heat storage in an individual layer highly depended on the material properties of this particular layer and the adjacent fabric layers.⁸ Consequently, we cannot measure the heat storage performance of single-layer material alone but should investigate the overall clothing. In addition, there was no significant difference (p > 0.05) in heat storage between the E1 and OS layer of E2 at the forearm and thigh part of clothing, according to Figures 7(d) and (e). The reason may be related to the air gap between the clothing and skin. The research showed that the air gap improved the heat storage for each clothing layer.²⁴ The larger air gap in clothing E1 may increase the heat storage and offset the advantage of multilayers of clothing E2.

From the above qualitative analysis, the ambient temperature step magnitude and the weight, specific heat capacity, thickness and location of the clothing layer were essential for heat storage. Multiple linear regression of clothing heat storage at five local parts was carried out, respectively. The standard regression coefficients are shown in Table 6. The standard regression coefficients can explain the contribution of factors. As for the chest, thigh and calf with the same configuration of each layer, the influence of weight was greater than that of specific heat capacity. Particularly for the calf, the effect of weight was four times that of the specific heat capacity, which explained the previous phenomenon that the heat storage in the TL layer with a larger weight was greater than that in the OS layer. However, for the forearm with a double-layer outer structure, the influence of specific heat capacity was greater. The thickness of the clothing layer at the abdominal part had the greatest effects, followed by the layer location and specific heat capacity. These results showed that the clothing layer weight and specific heat capacity were primary factors. Based on the lightweight requirements of thermal protective clothing, it is crucial to improve the specific heat capacity of each clothing layer. In addition, the strategy to improve heat storage for local clothing parts should be different. For example, increasing the thickness of each clothing layer should be an effective way to improve the heat storage level at the abdominal part, while it was not helpful for other parts.

Table 6.
Standard multiple linear regression coefficients of the heat storage at five clothing local parts

Local parts Standard linear regression coefficient (beta)
Adjusted R²

ΔT_a Weight Specific heat capacity Thickness Layer location

Chest 0.401 0.721 0.389 / / 0.964

Abdomen 0.288 / 0.514 0.816 −0.587 0.901

Forearm 0.287 0.307 0.815 / / 0.957

Thigh 0.380 0.747 0.358 / / 0.954

Calf 0.426 0.824 0.205 / / 0.961

Note: ΔT_a is the ambient temperature step magnitude; the layer location is determined by the order of each layer; the order is 1, 2, 3, 4, 5, 6 from the outer layer to the inner layer.

On the other hand, clothing insulation is a comprehensive index reflecting the effects of the specific heat capacity, weight, thickness of materials and air gap as well. The previous study found a good linear relationship between clothing heat storage and thermal insulation.⁸ This work proposed a linear relationship by combining the impacts of the ambient temperature step magnitude and clothing insulation. Equation (7) is a multiple linear regression formula to obtain the clothing average heat storage
$Q_{c l} = a \times I_{t} + b \times Δ T_{a} + c$
(7)
where I_t is the clothing total insulation, clo, and ΔT_a is the ambient temperature step magnitude, °C. The coefficients a, b and c were 0.348, 15.860 and –27.168 (regression analysis, R² = 0.98), respectively. However, there is still a lot of work to do for the validation, such as increasing the variety of test clothing in the future.

As we can see, the ambient temperature step magnitude significantly affected the amount of heat storage in clothing. The heat storage at the abdomen was the largest, which was mainly affected by the thickness, clothing layer location and specific heat capacity. For the chest, thigh and calf parts with low heat storage, additional protective measures should be provided, such as increasing the clothing layer weight and specific heat capacity. The heat storage in the outer layer was increased by 2% to 39% due to the thermal resistance of the inner layer. For the convenience of engineering calculation and practical application, a linear regression formula of the clothing average heat storage about ambient temperature step magnitude and clothing insulation was proposed. These results will help us better understand the heat exchange between the human body and the environment.

Relationship between heat storage and release

To establish the relationship between heat storage and release, the change of heat storage within 1 h after temperature up steps and the change of heat release within 1 h after temperature down steps (the accumulated heat storage at the 90th min minus the final one at the 150th min) were calculated.

Figure 8 displays a robust linear relationship between heat storage and release (E1: R² = 0.99, E2: R² = 0.98). The standard linear coefficients of clothing E1 and E2 were 0.992 and 0.977, respectively. This revealed that the heat storage was a little higher than the heat release, which was in line with the heat storage at the 150th min being higher than that at the 30th min, as shown in Figures 6(a) and (c). The results can be explained by a constant heat flux that heated the clothed manikin system during the whole temperature up and down step conditions, although it was consistent with the actual situation. Furthermore, the linear coefficients in both clothing E1 and E2 were above 1, which meant that the change rate of heat release was elevated higher than that of the heat storage.

Figure 8.
The relationship between heat storage and release under ambient temperature steps.

The difference between heat storage in temperature up steps and heat release in temperature down steps was calculated. Figure 9 illustrates these differences under five temperature step conditions. These differences became smaller with the ambient temperature step magnitude being enhanced. For single-layer clothing E1, there was no significant difference (p > 0.05) between heat storage and release when the temperature step magnitude was greater than 27.5°C. This result suggested that the heat stored in single-layer clothing can discharge completely. As for multilayer clothing E2, there was no significant difference (p > 0.05) between heat storage and release when the ambient temperature step magnitude was more than 32.5°C. Therefore, the threshold value for single-layer and multilayer clothing varied because the heat storage in multilayer clothing was larger, so it required a greater ambient temperature step magnitude for complete release within the same time.

Figure 9.
The difference between heat storage and release during ambient temperature steps: (a) E1 and (b) E2.

Figure 9 also reveals a discrepancy among the five clothing parts. The greatest difference between heat storage and release was at the abdominal parts, especially for multilayer clothing E2, in which the heat storage was 2.2–3.3 kJ/m² higher than the average, as shown in Figure 9(b). In addition, the discrepancy dropped rapidly at the chest, abdomen, thigh and forearm when the ambient temperature step magnitude increased to 32.5°C. As for the calf parts, there were no significant differences during any of the temperature step changes (p > 0.05).

Table 7 compares the differences between heat storage and release among the three layers of multilayer clothing E2. The largest difference between the heat storage and release was in the TL layer, followed by the OS and MB layers.

Table 7.
The difference between heat storage and release among three clothing layers of E2, kJ/m²

Clothing layer 0°C–40°C–0°C
5°C–40°C–5°C
10°C–40°C–10°C
15°C–40°C–15°C
20°C–40°C–20°C

S R Δ S R Δ S R Δ S R Δ S R Δ

Chest_OS 7.5 7.8 −0.3 6.5 6.5 0 6.2 5.3 0.9 5 4.1 0.9 4.6 3.8 0.8

Chest_MB 2.9 2.8 0.1 2.5 2.4 0.1 2.4 2 0.4 2 1.5 0.5 1.8 1.3 0.5

Chest_TL 7 6.6 0.4 6.2 5.5 0.7 6.2 4.6 1.6 5.1 3.5 1.6 4.6 2.8 1.8

Abdomen_OS1 8.8 8.6 0.2 8 8.1 −0.1 7.5 7 0.5 6 5.6 0.4 5.2 4.7 0.5

Abdomen_MB1 3.5 3.4 0.1 3.2 3.2 0 3 2.7 0.3 2.4 2.1 0.3 2.1 1.8 0.3

Abdomen_TL1 9.4 8.9 0.5 8.6 8.4 0.2 8 7.2 0.8 6.4 5.5 0.9 5.6 4.6 1

Abdomen_OS2 5 4.1 0.9 4.8 3.9 0.9 4.4 3 1.4 4.1 2.5 1.6 3.6 2 1.6

Abdomen_MB2 1.8 1.3 0.5 1.7 1.2 0.5 1.7 0.9 0.8 1.6 0.7 0.9 1.5 0.5 1

Abdomen_TL2 4.6 2.9 1.7 4.6 2.7 1.9 4.6 2 2.6 4.3 1.6 2.7 4.1 1.1 3

Forearm_OS 15 14.9 0.1 12.9 13.2 −0.3 12.4 11 1.4 10 8.2 1.8 9 6.8 2.2

Forearm_MB 3 3 0 2.6 2.6 0 2.5 2.2 0.3 2 1.5 0.5 1.9 1.3 0.6

Forearm_TL 8.2 7.9 0.3 6.8 6.6 0.2 6.8 5.7 1.1 5.5 4.1 1.4 5.2 3.5 1.7

Thigh_OS 7.3 7.5 −0.2 6.7 6.4 0.3 5.9 5.1 0.8 4.6 3.7 0.9 4.7 3.7 1

Thigh_MB 2.7 2.7 0 2.5 2.3 0.2 2.3 1.8 0.5 1.8 1.3 0.5 1.9 1.2 0.7

Thigh_TL 6.9 6.5 0.4 6.4 5.7 0.7 6 4.5 1.5 4.8 3.1 1.7 5.1 2.9 2.2

Calf_OS 8.8 9.2 −0.4 7.5 8 −0.5 7 6.9 0.1 5.8 5.7 0.1 5.4 5 0.4

Calf_MB 3.7 3.8 −0.1 3.1 3.3 −0.2 2.9 2.9 0 2.4 2.3 0.1 2.2 2.1 0.1

Calf_TL 10 10.3 −0.3 8.5 9 −0.5 7.8 7.8 0 6.5 6.3 0.2 6.1 5.5 0.6

S: heat storage, R: heat release, Δ: the difference between heat storage and release, kJ/m²; OS: outer shell; MB: moisture barrier; TL: thermal liner.

Furthermore, Table 8 shows the standard multiple linear regression coefficients of the difference between heat storage and release at five clothing parts. The ambient temperature step magnitude had the greatest impact at the chest, forearm, thigh and calf, since the standard linear regression coefficient was more than 0.8. Nevertheless, the influence degree of ambient temperature step magnitude decreased at the abdominal clothing, and clothing layer location became the greatest. There was a positive correlation between the clothing layer location and the difference; that is, the closer to the skin, the less the heat discharge.

Table 8.
Standard multiple linear regression coefficients of the difference between heat storage and release at five local parts of clothing

Local parts Standard linear regression coefficient (beta)
Adjusted R²

ΔT_a Layer location Heat storage

Chest −0.801 0.531 0.421 0.816

Abdomen −0.420 0.935 0.376 0.799

Forearm −0.893 0.483 0.415 0.720

Thigh −0.855 0.483 0.415 0.869

Calf −0.842 / / 0.710

The above analysis confirmed that the heat release in clothing depended on the magnitude of temperature down steps, the clothing layer location and the heat storage in clothing before temperature down steps. The heat storage can be discharged completely when the ambient temperature step magnitude was greater than 32.5°C. The clothing layer location dramatically affected the heat release at the abdominal part for the multilayer configuration. To mitigate the effects of large temperature down steps on human health and thermal comfort,^31,32 it is necessary to guide the heat transfer to the skin by increasing the inner layer heat storage. Utilization of phase change material in the inner layer of clothing provides protection for heat-exposure workers against an unavoidable ambient temperature step situation.^23,33,34

Conclusions

This report describes the results of a limited laboratory study designed to provide a feasible basis for investigating heat storage and release in thermal protective clothing under five types of ambient temperature step conditions. Ambient temperature step magnitude is critical for the amount of heat storage, heat storage rate and storage duration. The following main conclusions can be obtained. (1) Some 80% of heat storage or release would be completed in around 10 min after ambient temperature steps. (2) Heat storage at the chest, thigh and calf parts of clothing should be enhanced by increasing the inner clothing layer weight and specific heat capacity. As for the abdominal part of the clothing with the largest heat storage, it was necessary to guide the heat transfer to the skin when it discharged heat. The effective way was to elevate the inner clothing layer thickness and specific heat capacity at the abdomen. (3) Heat storage in the outer layer was the largest and can be increased by 2% to 39% due to adding the inner layer of the garment. (4) Clothing heat release was determined by the heat storage during temperature up steps, the magnitude in temperature down steps and the clothing layer location. The heat storage discharged completely until the ambient temperature step magnitude exceeded 32.5°C. (5) A multiple linear regression formula was proposed to obtain the clothing average heat storage by considering the ambient temperature step magnitude and total clothing insulation. The results of this study contribute to the optimization of thermal protective clothing and the improvement of the research in human thermal comfort.

Fabric type	Fabric content	Weight (g/m²)	Thickness (mm)	Density (kg/m³)	Air permeability (mm/s)	Specific heat at 20°C (J/(kg°C))
OS	100% X-Fiper (meta-aramid)	214.9	0.54	396.6	76.2	1300
MB	X-Fiper (meta-aramid)/PTFE	108.2	0.52	208.5	0	1080
TL	100% X-Fiper (meta-aramid)/flame-resistant viscose	283.1	2.18	129.7	493.9	1110
Underwear	100% Cotton	194.0	0.71	273.2	552.9	1225

Clothing ensemble	Total thermal insulation (clo)
E1	1.60 ± 3.5%	3.56 ± 5.6%	1.25 ± 2.4%	1.78 ± 2.4%	1.14 ± 0.9%	1.62 ± 2.1%
E2	1.65 ± 4.1%	6.77 ± 6.5%	1.79 ± 1.5%	2.65 ± 4.0%	1.60 ± 0.7%	2.28 ± 2.5%

Condition	Phase 1 (°C)	Phase 2 (°C)	Phase 3 (°C)
0°C–40°C–0°C	3.2 (0.4)	39.7 (0.2)	3.1 (0.9)
5°C–40°C–5°C	6.9 (0.4)	39.4 (0.2)	6.0 (0.3)
10°C–40°C–10°C	11.7 (0.1)	39.5 (0.2)	11.5 (0.3)
15°C–40°C–15°C	16.6 (0.2)	39.4 (0.3)	16.9 (0.2)
20°C–40°C–20°C	20.6 (0.2)	39.4 (0.3)	20.1 (0.4)

	E2_chest	E2_forearm	E2_abdomen_coat	E2_abdomen_pants	E2_thigh	E2_calf
SD	1.12	1.51	1.28	0.64	1.02	1.43
RMSD	0.78	1.01	0.89	0.35	0.86	0.84
Bias	0.02	0.19	0.77	0.10	0.20	0.17

	Average	Chest	Abdomen	Forearm	Thigh	Calf
Coefficient	0.24	0.47	0.15	0.36	0.34	0.62	0.33	0.56	0.18	0.34	0.21	0.49
Constant	1.52	5.67	0.86	4.20	2.49	10.49	2.25	5.03	1.10	4.38	0.83	3.86
Adjusted R²	0.98	0.97	0.98	0.95	0.99	0.97	0.93	0.93	0.99	0.92	0.99	0.95

Local parts	Standard linear regression coefficient (beta)	Adjusted R²
Chest	0.401	0.721	0.389	/	/	0.964
Abdomen	0.288	/	0.514	0.816	−0.587	0.901
Forearm	0.287	0.307	0.815	/	/	0.957
Thigh	0.380	0.747	0.358	/	/	0.954
Calf	0.426	0.824	0.205	/	/	0.961

Clothing layer	0°C–40°C–0°C	5°C–40°C–5°C	10°C–40°C–10°C	15°C–40°C–15°C	20°C–40°C–20°C
Chest_OS	7.5	7.8	−0.3	6.5	6.5	0	6.2	5.3	0.9	5	4.1	0.9	4.6	3.8	0.8
Chest_MB	2.9	2.8	0.1	2.5	2.4	0.1	2.4	2	0.4	2	1.5	0.5	1.8	1.3	0.5
Chest_TL	7	6.6	0.4	6.2	5.5	0.7	6.2	4.6	1.6	5.1	3.5	1.6	4.6	2.8	1.8
Abdomen_OS1	8.8	8.6	0.2	8	8.1	−0.1	7.5	7	0.5	6	5.6	0.4	5.2	4.7	0.5
Abdomen_MB1	3.5	3.4	0.1	3.2	3.2	0	3	2.7	0.3	2.4	2.1	0.3	2.1	1.8	0.3
Abdomen_TL1	9.4	8.9	0.5	8.6	8.4	0.2	8	7.2	0.8	6.4	5.5	0.9	5.6	4.6	1
Abdomen_OS2	5	4.1	0.9	4.8	3.9	0.9	4.4	3	1.4	4.1	2.5	1.6	3.6	2	1.6
Abdomen_MB2	1.8	1.3	0.5	1.7	1.2	0.5	1.7	0.9	0.8	1.6	0.7	0.9	1.5	0.5	1
Abdomen_TL2	4.6	2.9	1.7	4.6	2.7	1.9	4.6	2	2.6	4.3	1.6	2.7	4.1	1.1	3
Forearm_OS	15	14.9	0.1	12.9	13.2	−0.3	12.4	11	1.4	10	8.2	1.8	9	6.8	2.2
Forearm_MB	3	3	0	2.6	2.6	0	2.5	2.2	0.3	2	1.5	0.5	1.9	1.3	0.6
Forearm_TL	8.2	7.9	0.3	6.8	6.6	0.2	6.8	5.7	1.1	5.5	4.1	1.4	5.2	3.5	1.7
Thigh_OS	7.3	7.5	−0.2	6.7	6.4	0.3	5.9	5.1	0.8	4.6	3.7	0.9	4.7	3.7	1
Thigh_MB	2.7	2.7	0	2.5	2.3	0.2	2.3	1.8	0.5	1.8	1.3	0.5	1.9	1.2	0.7
Thigh_TL	6.9	6.5	0.4	6.4	5.7	0.7	6	4.5	1.5	4.8	3.1	1.7	5.1	2.9	2.2
Calf_OS	8.8	9.2	−0.4	7.5	8	−0.5	7	6.9	0.1	5.8	5.7	0.1	5.4	5	0.4
Calf_MB	3.7	3.8	−0.1	3.1	3.3	−0.2	2.9	2.9	0	2.4	2.3	0.1	2.2	2.1	0.1
Calf_TL	10	10.3	−0.3	8.5	9	−0.5	7.8	7.8	0	6.5	6.3	0.2	6.1	5.5	0.6

Local parts	Standard linear regression coefficient (beta)	Adjusted R²
Chest	−0.801	0.531	0.421	0.816
Abdomen	−0.420	0.935	0.376	0.799
Forearm	−0.893	0.483	0.415	0.720
Thigh	−0.855	0.483	0.415	0.869
Calf	−0.842	/	/	0.710

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed the receipt of the following financial support for the research, authorship and/or publication of this article: The authors would like to acknowledge the financial support from the Fundamental Research Funds for the Central Universities and Graduate Student Innovation Fund of Donghua University (Grant No. CUSF-DH-D-2021059); the Fundamental Research Funds for the Central Universities (Grant No. 2232022G-08); and International Cooperation Fund of Science and Technology Commission of Shanghai Municipality (Grant No. 21130750100).

ORCID iDs

Qianqian Huang

Meng Deng

Jun Li

References

Gagge

Stolwijk

JAJ

Hardy

JD.

Comfort and thermal sensations and associated physiological responses at various ambient temperatures. Environ Res 1967; 1: 1–20.

de Dear

Ring

Fanger

PO.

Thermal sensations resulting from sudden ambient temperature changes. Indoor Air 1993; 3: 181–192.

Chen

Hwang

Chang

, et al. Effects of temperature steps on human skin physiology and thermal sensation response. Build Environ 2011; 46: 2387–2397.

Xiong

Lian

Zhou

, et al. Effects of temperature steps on human health and thermal comfort. Build Environ 2015; 94: 144–154.

Xiong

Lian

Zhang

, et al. Correlation between health discomforts and temperature steps in winter of China. Build Environ 2017; 114: 387–396.

Zhang

Ding

Acceptable temperature steps for transitional spaces in the hot-humid area of China. Build Environ 2017; 121: 190–199.

Havenith

Interaction of clothing and thermoregulation. Exogen Dermatol 2002; 1: 221–230.

Yan

Lichuan

, et al. Quantitative assessment of the thermal stored energy in protective clothing under low-level radiant heat exposure. Text Res J 2018; 88: 2867–2879.

Wan

Fan

A new method for measuring the thermal regulatory properties of phase change material (PCM) fabrics. Meas Sci Technol 2009; 20: 1–6.

10.

Safavi

Amani-Tehran

Latifi

Evaluation of dynamic thermal behavior of fibrous layers in presence of phase change material microcapsules. Thermochim Acta 2014; 594: e16–e23.

11.

Kim

Yoo

Kim

Development of a human–clothing–environment simulator for dynamic heat and moisture transfer properties of fabrics. Fiber Polym 2003; 4: 215–221.

12.

Yoo

Lim

Kim

Thermal characteristics of the garment air-layers by PCM concentration changes.

J Kor Soc Clothing Text 2008; 32: 991–998.

13.

Kim

Assessment of the heat and moisture transfer in a multilayer protective fabric system under various ambient conditions. Text Res J 2015; 85: 227–237.

14.

Peng

, et al . Review of clothing for thermal management with advanced materials. Cellulose 2019; 26(11): 6415–6448.

15.

Psikuta

Allegrini

Koelblen

, et al. Thermal manikins controlled by human thermoregulation models for energy efficiency and thermal comfort research – a review. Renew Sustain Energ Rev 2017; 78: 1315–1330.

16.

Oliveira

AVM

Gaspar

Quintela

DA.

Measurements of clothing insulation with a thermal manikin operating under the thermal comfort regulation mode: comparative analysis of the calculation methods. Eur J Appl Physiol 2008; 104: 679–688.

17.

Gao

Kuklane

Wang

, et al. Personal cooling with phase change materials to improve thermal comfort from a heat wave perspective. Indoor Air 2012; 22: 523–530.

18.

Lee

Han

Lee

, et al. Thermal properties of combat uniforms treated with microencapsulated octadecane and change in clothing microclimate via thermal manikin. J Text Inst 2018; 109: 585–595.

19.

Kim

SJ.

Heat storage and release characteristics and mechanical properties of combat-uniform fabrics for emotional clothing. Autex Res J 2019; 19: 165–172.

20.

Shim

Park

Shim

HS.

Effect of ceramics on the physical and thermo-physiological performance of warm-up suit . Text Res J 2009; 79: 1557–1564.

21.

Analyzing the transmitted and stored energy through multilayer protective fabric systems with various heat exposure time. Text Res J 2016; 86: 235–244.

22.

Modeling the transmitted and stored energy in multilayer protective clothing under low-level radiant exposure. Appl Therm Eng 2016; 93: 1295–1303.

23.

Zhu

Tian

, et al. Intelligent bidirectional thermal regulation of phase change material incorporated in thermal protective clothing. Appl Therm Eng 2020; 174: 115340.

24.

Song

Cao

Gholamreza

Analyzing stored thermal energy and thermal protective performance of clothing. Text Res J 2011; 81: 1124–1138.

25.

Voelker

Alsaad

Simulating the human body’s microclimate using automatic coupling of CFD and an advanced thermoregulation model. Indoor Air 2018; 28: 415–425.

26.

Psikuta

, et al. Influence of human body geometry, posture and the surrounding environment on body heat loss based on a validated numerical model. Build Environ 2019; 166: 106340.

27.

Onyeocha

Nwaigwe

Ogueke

, et al. Design and construction of an integrated tetrafluoroethane (R134a) refrigerator-waste heat recovery dryer for fabric drying in tropical regions. Heliyon 2020; 6: e04838.

28.

Whitman

, Johnson B, Tomczyk J. Refrigeration and air conditioning technology. Delmar Cengage Learning, Stanford, Connecticut, 2012.

29.

Psikuta

Niedermann

Rossi

RM.

Effect of ambient temperature and attachment method on surface temperature measurements. Int J Biometeorol 2014; 58: 877–885.

30.

Shen

Yokoyama

Sukigara

Modeling of heterogeneous heat transfer in fabrics. Text Res J 2018; 88: 1164–1172.

31.

Xiong

Lian

Zhou

, et al. Potential indicators for the effect of temperature steps on human health and thermal comfort. Energ Build 2016; 113: 87–98.

32.

Yang

Lewington

, et al. Outdoor temperature, blood pressure, and cardiovascular disease mortality among 23 000 individuals with diagnosed cardiovascular diseases from China. Eur Heart J 2015; 36: 1178–1185.

33.

Fonseca

Mayor

Campos

JBLM.

Guidelines for the specification of a PCM layer in firefighting protective clothing ensembles. Appl Therm Eng 2018; 133: 81–96.

34.

Shaid

Wang

Padhye

The thermal protection and comfort properties of aerogel and PCM-coated fabric for firefighter garment. J Ind Text 2016; 45: 611–625.

Local parts	Standard linear regression coefficient (beta)					Adjusted R²
Local parts	ΔT_a	Weight	Specific heat capacity	Thickness	Layer location	Adjusted R²
Chest	0.401	0.721	0.389	/	/	0.964
Abdomen	0.288	/	0.514	0.816	−0.587	0.901
Forearm	0.287	0.307	0.815	/	/	0.957
Thigh	0.380	0.747	0.358	/	/	0.954
Calf	0.426	0.824	0.205	/	/	0.961