Enhancing thermophysiological comfort in defense clothing for hot climate: A comprehensive evaluation and introduction of passive cooling vests as a solution

Abstract

Protective clothing systems pose significant heat stress challenges in hot climate, limiting natural cooling through sweat evaporation. In such conditions, passive cooling systems could offer lightweight, silent, and energy-efficient solutions. However, their effectiveness and their physiological and psychological impact are influenced by specific operational conditions, including the type of clothing system used. This study assessed the thermophysiological effects of three passive cooling vests - Phase Change Material (PCM), Waterfill (WF), and Watersoak (WS) - worn under Belgian Defense clothing system with ballistic protection during wear trials at 30°C, 48% RH, 4 km/h, 120 min. The PCM vest with a transition temperature of 29°C provided the most effective cooling, significantly reducing torso microclimate temperature for 60 min. Over the total test duration, the torso microclimate temperature was lowered by an average of 1.5°C, accompanied by reduced cardiovascular strain, sweating, and enhanced thermal and wetness perception. Thus, the PCM can be a better solution than active cooling for operations up to 2 hours. Water-based cooling vests delivered mild cooling, with significant effect at 30 min, but temporary duration. They still showed reduced microclimate temperature by 0.5°C at 60 min, but became a thermal burden thereafter, confirming their limited performance under restrictive clothing. The results also showed that subjective experience did not fully align with objective data, confirming the importance of individual perception in comfort-related studies.

Keywords

thermophysiological comfort protective clothing defense clothing ballistic wear personal cooling systems cooling vest

Introduction

Protective clothing systems such as ballistic wear that typically covers 22%–33% of body surface area, restrict natural cooling, leading to discomfort, heat-stress risks and reduced operational performance.^1,2 A recent survey of 38 Belgian defense soldiers found that 92% were dissatisfied with the thermal comfort of their clothing systems in hot climate, emphasizing the need for effective cooling solutions.³ Comfort, defined as the subjective satisfaction with the surrounding environment, consists of thermal, ergonomic, tactile, physiological and psychological aspects.^4,5 Human wear trials are crucial in comfort evaluation to validate the real physiological response.^5–8 The air in the skin–clothing microclimate offers valuable insights into thermal comfort, reflecting changes in temperature and relative humidity between the skin and clothing layers.^6,9 Thermal comfort depends on the total heat and mass transfer from the body to the environment, given with the heat balance equation (Eq. 1):

M - W - (C + R + E_{resp} + E_{skin} + K) = S (W / m^{2})

(1)

where M is the metabolic rate, W the mechanical work by the body, C convection, R radiation, E_resp respiratory evaporation, E_skin sweat evaporation, K conduction, and S is the body heat storage which should be zero to avoid thermal strain.^2,6 When sweat evaporation (E_skin) is limited, enhancing alternative heat transfer mechanisms can help reduce thermal strain. Personal cooling systems can facilitate these alternative heat transfer pathways. There are passive or active systems, classified into six distinct categories: phase change materials (PCM), liquid-circulating, air ventilation, evaporative, thermoelectric and radiative cooling.^1,4,10 Passive cooling does not require power, while active systems rely on external energy but are less dependent of external conditions.^1,2,10,11 Liquid-circulating systems achieve active cooling by cold liquid circulation in tubes stitched on a garment, with portable or stationary liquid storage. They have reduced skin temperatures up to 6°C and work well under protective clothing. However, they add bulk, need external power and careful maintenance.^4,5,8 Under ballistic wear, a liquid circulating vest reduced skin temperature by 3°C and improved cooling duration compared to air ventilation and PCM vests (T = 42.2°C, 20% RH, sun radiation, 120 min).² Another liquid cooling vest under stab-resistance armor reduced mean skin temperature by 1.2°C for 40 min in an ambient of 30°C.⁷ Active air-ventilation systems circulate air through tubes, blowers, fans using external power. In 8-h manikin tests in hot conditions, they showed the longest cooling duration compared to other cooling vests, but were connected to a compressed air source which limits mobility in real use.¹² In wear trials they were shown as less effective than PCMs and the fan-introduced air can even cause infections.^1,4,5,11 Phase-change materials (PCMs) provide passive cooling thanks to the ability to store large amounts of latent heat at their phase-change temperature point.^1,5,13 They work at different relative humidities (RH) and have shown comparable cooling power to active systems without the need of energy consumption. PCMs include gels, paraffin, organic waxes and even ice which can be packed in a cooling vest, pockets, or microencapsulated in fabrics and fibers.⁵ PCM vests can reduce skin temperature up to 3–4°C for up to 120 min, however, they are impermeable, bulky and need frequent reactivation.^2,10 Higher PCM quantity, lower melting temperatures, larger body coverage, and a gradient over 6°C can increase cooling power, while higher phase-change temperatures can prolong cooling and improve comfort.^5,10 Composite PCMs, microencapsulation, optimized shell materials, addition of carbon and metal particles can enhance thermal conductivity, durability and overall cooling performance.^14,15 While PCM fabric microencapsulation allows for sweat transport, the smaller PCM quantity cannot provide as powerful cooling as PCM packs. PCM-coated Kevlar in soft body armor provided only mild, temporary cooling (less than 30 min), while also reducing ballistic efficiency.¹⁰ Passive evaporation systems rely only on the clothing design and fabric air and moisture transfer, without external power. Sweat evaporation has a large heat transport capacity, but requires strong thermal and humidity gradients.¹² Beyond sweat, soaking textiles in cold water or liquids like menthol and ethanol will further enhance cooling.⁵ Evaporative cooling vests such as water-soaked systems can provide milder but long cooling of up to 3.5 h at ambient T = 35°C, as shown with manikin tests.¹² However, they are not recommended for humid conditions and are less studied under ballistic wear. Radiative cooling garments work by reflecting solar rays or/and promoting human skin’s mid-infrared emissivity, thanks to structural properties at the clothing surface or functionalization with particles like titanium dioxide. They can be effective in intense sun radiation.⁵ Thermoelectric cooling fabrics rely on the Peltier effect under electric current to cool one fabric side while heating the opposite one.¹⁶ However, they are still under research and their cooling effectiveness may not be suitable under ballistics due to the contact with the heated thermoelectric side. There are also hybrid cooling systems that combine different cooling mechanisms. For example, Song and Wang⁸ tested an air-ventilation-PCM system (T = 36°C, RH = 60%, 90 min exercise), distributed in the upper and lower body. It significantly reduced skin temperature (by 0.72 ± 1.2°C), core temperature, HR, and subjective skin wetness.

Previous studies emphasize the need for more empirical research on human responses to different cooling solutions, in specific scenarios.⁵ For instance, the cooling vest placement within clothing, different PCM melting temperatures, different ambient conditions can largely impact cooling performance. Commercial evaporative vests are not recommended under restrictive clothing, but their impact in ballistic wear remains understudied.¹⁷ This study evaluates three passive cooling vests – PCM, Waterfill (WF), and Watersoak (WS) – as potential sustainable cooling solutions for the Belgian military clothing system for hot climate under ballistic wear. The aim is to evaluate the thermophysiological impact of each vest based on wear trials, identifying when each vest becomes a thermal burden. Microclimate temperature and humidity in the torso area are monitored and compared to subjective thermal sensation, sweating wetness, and comfort. This study uses “microclimate” to describe temperature and RH under clothing, emphasizing precise sensor placement, as skin and microclimate values can differ and are often not clearly distinguished.

Materials and methods

Study participants

A within-subject experimental approach was used to select participants with similar anthropometric data, fitness levels, and work experience in the Belgian Defense.¹⁸ A sample size of minimum eight was determined to achieve adequate statistical power (G*Power, ANOVA repeated measures, within-subjects, d = 1.59, α = 0.05, power = 0.95).^6,18,19 Due to limitations described later in Limitations, this sample size was not reached so the focus in this study is on observed trends in subjects. Three subjects were voluntarily recruited (Table 1), fully briefed on the protocol, provided verbal consent for their data usage, and were reassured of anonymity.

Table 1.

Subject details.

Subject ID	Gender	Age (years)	Body height (cm)	Body weight (kg)	BMI (kg/m²)	Body surface area (m²)	Acclimatization^a	Fitness level	Health
K	Male	41	177	83.1	26.52	0.07	Normal climate	Active	Good (smokes)
P	Male	35	172	70.0	23.66	0.06	Cold climate	Very active	Excellent
G	Male	35	179	79.0	24.66	0.07	Warm climate	Very active	Excellent
Mean	Male	37 ± 2	176 ± 2.94	77.3 ± 5.4	24.95 ± 1.19	0.07 ± 0.00

^aAcclimatization stands for the usual training environment of the subject (warm, normal, cold climate), 10 weeks in a year.

Materials

Figure 1 illustrates the layering in the torso area, while the full clothing system for hot climate used by Belgian Defense is described in Table 2. Fabric properties of clothing elements below the ballistic vest are given in Table 3. Most data was obtained from Belgian Defense and the rest measured following standards mentioned in Table 3. Thermal (R_ct) and evaporative resistances (R_et) were measured with Integrated Sweating Guarded Hotplate (ISGHP, Measurement Technology Northwest) following ISO 11092.²⁰ Moisture management indices, wetting time (WT), absorption rate (AR), accumulative one-way transport capability (R_top-bottom) and overall moisture management capacity (OMMC), were obtained with moisture management tester (MMT, SDL Atlas^® Hong Kong), following AATCC 195-2011.^21,22 The baselayer wicking property comes from the structural gradient between the inner and outer fabric side (Table 3). Clothing was washed and dried before each test following standard military procedures.

Figure 1.

Clothing layering for each wear trial.

Table 2.

Full clothing system for hot climate (clothing elements and sizes).

Clothing elements	Size per subject
Clothing elements	K	G	P
1. Combat helmet	Not given	Not given	Not given
2. Baselayer (wicking T-shirt)	L	M	M
3. Combat shirt DES	L	M	S
4. QRV (quick release vest)	L/Long	L/Long	S/Long
5. Ballistic inserts	L/Long	L/Long	S/Long
5. Daypack with hydration pack	1.80% BW^a	3.24% BW^a	3.88 % BW^a
6. Protective gloves	11	9	8
7. Combat belt	Not given	Not given	Not given
8. Combat pants DES	M/Short	M/Long	M/Short
9. Combat socks	Not given	Not given	Not given
10. Shoes (combat boots/own)	Not given	Not given	Not given

^aDaypack load is given as percentage of bodyweight (% BW) per subject: 100 × (daypack weight ÷ body weight).

Table 3.

Fabric properties of clothing elements covering the torso.

	Baselayer (wicking T-shirt, athletic body fit)	DES knitted fabric (combat shirt - torso area)	DES light fabric (combat shirt - arm area, combat pants)
Composition	60% MOD/40% VIS	65% MOD/30% VIS/50% EL	65% CO/35% PES
Structure	Knitted (Tricot jersey, single/double)	Knitted	Woven (Twill 2/1 Z)
Functionalization	Fire protection	Fire protection	Anti-insect (permethrin)
Weight (g/m²)ISO 3801 (1977)	180.00	277.36	220.00
Thickness (mm)ISO 5084 (1996)	0.71 ± 0.01	0.44 ± 0.02	0.45 ± 0.00
R_ct (Pa m²/W)^aISGHP	0.025 ± 0.000	0.020 ± 0.000	0.004 ± 0.000
R_et (Pa m²/W)^aISGHP	2.646 ± 0.111	4.328 ± 0.300	4.720 ± 0.030
WT_top (s)^aMMT	3.94 ± 0.25 (Grade 4)	7.35 ± 0.89 (Grade 3)	Not measured
WT_bottom (s)^aMMT	3.38 ± 0.30 (Grade 4)	9.91 ± 0.78 (Grade 3)	Not measured
AR_top (%/s)^aMMT	44.89 ± 2.07 (Grade 4)	63.86 ± 6.49 (Grade 4)	Not measured
AR_bottom (%/s)^aMMT	55.76 ± 2.59 (Grade 4)	106.31 ± 8.26 (Grade 5)	Not measured
R_top-bottom (%)^aMMT	184.08 ± 8.35 (Grade 3)	160.36 ± 14.94 (Grade 3)	Not measured
OMMC (0-1 scale)^aMMT	0.57 ± 0.02 (Grade 3)	0.54 ± 0.00 (Grade 3)	Not measured

Note. Grade (1-5): Moisture management performance rating from 1 to 5 (low to high), following AATCC 195-2011. Top and bottom represent fabric sides facing liquid drop (inner side) or MMT bottom sensors (outer side), respectively.

^aR_ct and R_et: Thermal and evaporative resistances, measured with ISWHP (Integrated Sweating Guarded Hot Plate). MMT (Moisture Management Tester) indices: WT (Wetting time for moisture spreading), AR (Moisture absorption rate), R_top-bottom (One-way moisture transport index, from top to bottom), OMMC (Overall moisture management capacity, graded from 0 = poor to 1 = excellent).

Three different passive cooling vests (CV) were evaluated: PCM, Waterfill (WF) and Watersoak (WS). Each CV was placed as a second layer (Figure 1), below the combat shirt and ballistic (QRV) vest. The vests were conditioned in the testing room for 20 min prior to each test. PCM CV was made of fabric-free, 100% biobased PCM (CrodaTherm^TM, melting T ≥ 29°C, crystallization T ≤ 26°C), with dry weight of 1251 ± 1 g and thickness of 12.8 mm. It was packed into impermeable plates with adjustable stripes for fit.²³ It was refrigerated at 7°C the day before each test. The PCM vest was selected considering a higher melting temperature (to prevent cold injuries or premature activation at the ambient T), textile-free encapsulation (to minimize bulkiness), material sustainability, safety, and availability in EU. WF CV consisted of an outer layer (40% Nylon jersey/60% PU), inner nylon fabric and a wicking PES fleece core, with a wet weight measured before each test of 712 ± 34 g and a thickness of 2.4 mm. It was available in size S/M, with a waistline adjustment and multiple ventilation hole openings (Figure 1) at its front and back. It was activated by filling it with water (500 g, 15°C). WS CV consisted of 100% PES outer and inner fabric layers (the inner is PU coated) and a superabsorbent core, was available in size S/M and had a waistline adjustment. Its wet weight and thickness before each trial were 742.5 ± 101.5 g and 1.6 mm, accordingly. The vest was activated by soaking it in water (500 g, 15°C) for 2 min.

Protocol description

There is no standard for testing cooling vests, so a custom wear trial protocol was developed consulting the guidelines by Lei et al.¹⁸ ISO 9886:2004²⁴ and ASTM F2371.²⁵ Four tests per subject were done: three with each CV and a reference test (Ref.) without a CV. Each subject and CV were tested once per day, at the same time. The test could be stopped if: core temperature exceeded 38.5°C; HR reached 90% of the age-predicted maximum²⁶ calculated with equation (2); headache, dizziness, vomiting, nausea, or any irritability occurred; or if the subject wanted to stop.¹⁴

{HR}_{\max} (bpm) = \frac{90}{100} (220 - a g e)

(2)

The test conditions per subject were: T = 29.83 ± 0.27°C; RH = 48.14% ± 2.56%; no wind and sun radiation; treadmill speed = 4 km/h (incline angle of 4.8°); test duration = 120 min; average water intake, 120 min (from hydration pack) = 849 ± 302 g. Ambient T and RH were constantly monitored every 30 min using a thermometer and hygrometer. An electric heater was used to reach the set temperature value, which was kept constant by manually reducing heater output if needed. To eliminate external solar radiation, the windows were covered with black curtains. The airflow was not monitored but was considered minimal by keeping the windows and door closed. The subjects were acclimatized 20–30 min in the testing room, while the sensors were mounted. Each subject was given a hydration pack (3 L capacity) in the daypack, allowing them to drink freely throughout the trial. Water intake was measured by weighing the daypack before and after each trial.

Near-skin microclimate temperature and relative humidity (T_mc and RH_mc), heart rate (HR) and breathing rate (BR), subjective parameters and weight changes before and after the test were measured. The collected data was analyzed using SPSS and Excel, primarily reported as mean values per subject and across subjects, along with the standard deviation (SD). Near-skin microclimate refers to the space between the skin and baselayer, where sensors were placed to simultaneously monitor T_mc and RH_mc. Its parameters directly impact the thermal and moisture transport from the skin.^4,6 Four DS1923 iButton data loggers were used to collect data at 30 s intervals, via Exactlog Software. They were attached to the skin with a textile adhesive tape, ensuring the iButton opening which connects to the internal sensors faced the microclimate between the skin and baselayer. A small hole was made on the adhesive tape to avoid obstructing the iButton opening (Figure 2(a)). The sensors were placed at the front and back of torso (Figure 2(b)), based on the procedure by Eijsvogels et al.²⁷ It is important to distinguish between skin and microclimate temperature, since the microclimate can change with clothing movement, while skin temperature reflects surface conditions.⁹ From the collected data, the average T_mc difference between each CV and the reference per subject, ΔT_mc was calculated using equations (3) and (4):

Δ T_{mc, 60 \min} (° C) = CV Mean T_{mc, 60 \min} - Ref . Mean T_{mc, 60 \min}

(3)

Δ T_{mc, 120 \min} (° C) = CV Mean T_{mc, 120 \min} - Ref . Mean T_{mc, 120 \min}

(4)

Figure 2.

(a) Securing the iButton sensor to the skin. (b) Two front and two back torso locations of iButton sensors.

Heart Rate (HR, bpm) and Breathing Rate (BR, rpm) were continuously monitored (Equivital EQ02 SEM, Hidalgo, UK) at a 15 s sampling rate. The recorder was attached to a chest-worn belt placed on the subjects. Core temperature was monitored in real-time via Bluetooth, through a core sensor strapped around the upper arm (CORE Body Temperature Monitor).²⁸ It was only used as safety parameter, so the data is not included in the current analysis due to reasons explained in Limitations.

Weight changes per subject and clothing were measured before and after each test. Clothing weight, W_CL (g) which included T-Shirt, Combat Shirt, Combat Pants, and gloves (no socks and equipment) was measured with ±0.001 g precision. Weight of fully clothed Subject, W_fully _clothed (g), including full clothing system and CV, was measured with ±1 g precision. Weight of CV (g) was also measured with ±1 g precision. The following parameters were calculated from the collected data, considering the water drank during the tests per subject (daypack weight difference):

(1) Clothing Liquid Absorption, ΔW_CL (g) - the difference in the weight of the clothing before and after test resulting from the sweat and water transfer to the clothing (ΔW_CL = W_CL,120min – W_CL,0min).

(2) Sweat Evaporation, ΔW_{fully clothed} (g) - the difference in weight of fully clothed subject before and after the test (ΔW_{fully clothed} = W_{fully clothed, 120min} – W_{fully clothed, 0min}).

(3) Total liquid (sweat, water) loss, ΔW_L (g), including both the sweat absorbed by the clothing and the evaporated sweat (ΔW_L = ΔW_{fully clothed} + ΔW_CL).

Subjective data was gathered using a questionnaire constructed based on ISO 10551:2019.²⁹ Five-point scales (Figure 3) were used to evaluate: temperature (warmth), sweating wetness sensation^30,31, overall comfort³², and perceived weight. In the evaluative scales, the dissatisfaction intensity increases from 0 (absence of negative effect) to 4 (highest dissatisfaction). A bipolar scale is used for overall comfort acceptance, with −2 indicating highest discomfort and +2 the highest comfort. Intermediate ratings in increments of 0.5 could be given to express in-between feelings. Subjects were verbally asked and answered questions during the test at 30 min intervals. Notes were taken during each test to capture additional comments. From the collected data, mean values for each subjective aspect were calculated.

Figure 3.

Parameters and scales for subjective comfort evaluation.

Analytical approach

The data was analyzed using IMB SPSS Statistics 29.0.1.0. The results are mainly presented as means ± standard deviations. Although the study included only three subjects, the four iButton sensors continuously measuring microclimate parameters per subject enabled for multiple repeated observations per vest condition for both T_mc and RH_mc. A one-way repeated measures ANOVA was used to assess the effect of vest type on T_mc and RH_mc, using data from the four iButton sensors for the three subjects (n = 12). The analysis was performed separately for different time points (30, 60, 90, and 120 min), followed by Bonferroni-adjusted post hoc comparisons. The results are reported with p-values and partial η² (effect size: 0.01 indicates a small, 0.06 a moderate effect and 0.14 or higher a large effect).³³ Reported means (M) and standard errors (SE) are model-adjusted. To further validate the findings and increase statistical power, an additional one-way ANOVA was conducted using data from a single sensor location (front upper chest), with six measurements collected per interval (25–30 min and 55–60 min) for the three subjects (n = 18, per vest per parameter). In addition, individual response trends were analyzed through graphic representations of objective T_mc and RH_mc parameters and subjective thermal and wetness sensation ratings over time, to assess whether vest-related differences outweighed inter-subject variation.

Results

Microclimate temperature and relative humidity

The graphs in Figure 4 show the T_mc slopes of one subject, including range-based error bars. Table 4 shows mean T_mc values from the front and back of torso at different time points, while Table 5 shows the cooling effect of each CV compared to Ref: the mean ΔT_mc. Over the entire test duration (0–120 min), PCM provided an average T_mc reduction of 1.5°C compared to Ref., while both water-based CVs reduced it only around 0.2°C. At the start, T_mc in all test conditions was slightly higher than the ambient air. CVs immediately reduced T_mc by 1.5 to 2°C compared to Ref., with PCM showing the largest average reduction followed closely by WF and WS. This can be attributed to the PCM refrigeration at 7°C before the test as well as the initially cool water (15°C) in the water-based vests. Water-based CVs showed a much faster T_mc increase than PCM, with a plateau at 35 to 45 min. At 60 min, the PCM showed the biggest cooling effect (1.89°C), while water-based CVs had only mild cooling (≤0.38°C). The Ref. T_mc maintained the same level from 60 to 120 min (37.01 ± 0.19°C). At 120 min, only PCM provided mild cooling at the front of torso by 0.46°C, while WF and WS increased T_mc by around 0.5°C at the back compared to Ref. In general, the back T_mc appears to be slightly higher than the front in the second half of the tests, possibly due to the daypack.

Figure 4.

Mean microclimate temperature (T_mc) from 0 to 120 min for one subject taken from: (a) two sensors at the front, and (b) two sensors at the back of torso.

Table 4.

Mean microclimate temperature (T_mc) from the front and back of torso at different time points.

	Mean T_mc (°C)
	t = 0 min			t = 60 min			t = 120 min			From 0 to 120 min
	Front + Back	Front	Back	Front + Back	Front	Back	Front + Back	Front	Back	Front + Back
Ref.	34.35 ± 0.25	34.4 ± 0.48	34.31 ± 0.03	37.04 ± 0.39	36.86 ± 0.38	37.22 ± 0.40	37.01 ± 0.19	37.09 ± 0.25	36.94 ± 0.13	36.73 ± 0.06
PCM	32.35 ± 1.11	32.72 ± 1.15	31.99 ± 1.06	35.15 ± 0.70	35.2 ± 1.06	35.10 ± 0.34	36.76 ± 0.66	36.63 ± 0.75	36.89 ± 0.57	35.17 ± 0.53
WF	32.49 ± 0.64	32.87 ± 0.61	32.12 ± 0.68	36.71 ± 0.54	36.32 ± 0.69	37.11 ± 0.40	37.23 ± 0.39	37.09 ± 0.25	37.36 ± 0.53	36.39 ± 0.17
WS	32.89 ± 0.66	33.6 ± 0.61	32.17 ± 0.71	36.66 ± 0.75	36.5 ± 0.62	36.83 ± 0.89	37.26 ± 0.43	36.96 ± 0.37	37.57 ± 0.50	36.58 ± 0.15

Note. Front and Back represent average T_mc from the two sensors at the front or back of torso, respectively. Front + Back represent average T_mc from 0 to 120 min. Values shown are average from all subjects.

Table 5.

Cooling effect of Cooling Vests, based on microclimate temperature difference with reference (ΔT_mc) at 60 and 120 min.

	Mean ΔT_mc (°C) at t = 60 min			Mean ΔT_mc (°C) at t = 120 min
	Front + Back	Front	Back	Front + Back	Front	Back
Ref.	0.00	0.00	0.00	0.00	0.00	0.00
PCM	−1.89	−1.65	−2.12	−0.11	−0.46	−0.05
WF	−0.33	−0.54	−0.11	0.21	0.00	0.41
WS	−0.38	−0.35	−0.39	0.25	−0.12	0.63

ΔT_{mc, 60 min} = CV Mean T_{mc, 60min} – Ref. Mean T_{mc, 60 min}

ΔT_{mc, 120 min} = CV Mean T_{mc, 120min} – Ref. Mean T_{mc, 120 min}

Figure 5 shows the RH_mc change in time for different CVs and the Ref. Table 6 shows the mean RH_mc for each CV and Ref. at the front and back at different time points. Over 0–120 min, RH_mc was in the order of PCM, Ref., WF and WS, from lowest to highest, but there were only small differences compared to Ref. (2%–3%). Across the different testing conditions, RH_mc at the back of torso was 5%–10% higher than the front. At 0 min, RH_mc was highest for WS (79.47 ± 6.00%), followed by Ref., WF, and PCM. WS had higher RH_mc from the start due to soaked water, while the initially lower RH_mc of PCM and WF compared to Ref. indicates a slightly delayed sweat response. RH_mc slopes from all test conditions plateau between 30 and 40 min, similar to T_mc slopes (Figure 4), except for the PCM where T_mc increase seems to be slower than RH_mc. At 60 min, water-based CVs show high RH_mc (90%) which is similar to Ref., while PCM maintains a slightly lower level by 3%–5%. At 120 min, mean RH_mc range from 90%–93% for all test conditions, showing similar saturation of the microenvironment.

Figure 5.

Mean microclimate relative humidity (RH_mc) values from 0 to 120 min for one subject taken from: (a) two sensors at the front, and (b) two sensors at the back of torso.

Table 6.

Mean microclimate relative humidity (RH_mc) values from the front and back of torso across different time points.

	Mean RHmc (%)
	t = 0 min			t = 60 min			t = 120 min			From 0 to 120 min
	Front + Back	Front	Back	Front + Back	Front	Back	Front + Back	Front	Back	Front + Back
Ref.	60.04 ± 4.60	56.51 ± 5.63	63.18 ± 3.99	90.41 ± 0.73	89.68 ± 1.32	90.56 ± 0.91	91.22 ± 0.99	91.11 ± 1.03	90.56 ± 1.71	85.85 ± 0.20
PCM	57.26 ± 5.40	51.37 ± 4.17	63.14 ± 7.39	87.20 ± 1.86	87.47 ± 3.72	88.22 ± 2.56	90.92 ± 0.07	90.56 ± 1.27	90.40 ± 1.75	83.41 ± 0.03
WF	58.15 ± 6.36	52.8 ± 5.59	63.8 ± 7.33	91.44 ± 1.40	90.64 ± 2.17	92.16 ± 0.71	93.09 ± 0.10	92.82 ± 0.09	93.63 ± 0.58	87.38 ± 1.96
WS	79.47 ± 6.00	73.79 ± 7.34	85.46 ± 4.88	90.89 ± 0.23	89.71 ± 1.03	91.83 ± 3.61	92.04 ± 1.19	90.82 ± 2.09	92.88 ± 0.65	88.67 ± 1.06

Note. Front and Back represent average RH_mc from the two sensors placed on the front or back of torso, respectively. Front + Back = average RH_mc from 0 to 120 min. Values shown are an average from all subjects.

Heart and breathing rate

Heart rate (HR) reflects cardiovascular strain, while breathing rate (BR) indicates respiratory effort.^2,20 A lower HR does not necessarily indicate reduced thermal load and should be interpret alongside other parameters. It is directly influenced by cardiac output, stroke volume and indirectly by vasoregulation in response to thermal load or other stimuli.³⁴ Lower HR is often linked to relaxation or sweat evaporation that naturally lowers blood circulation. However, heat stress can trigger reflex vasodilation, lowering HR by reducing vascular resistance.^34,35 In contrast, local skin cooling can cause vasoconstriction which raises HR. Table 7 shows the mean HR and BR per CV over the whole wear trial. Ref. HR (107.20 ± 26.21 bpm) was reduced by both PCM and WS, while WF showed a similar HR to Ref. PCM showed the lowest HR and BR values (90.51 ± 13.23 bpm; 21.42 ± 6.20 rpm). While PCM reduced BR, both WF and WS slightly elevated it, indicating greater respiratory stress.

Table 7.

Mean Heart Rate and Breathing Rate over the wear trials.

	Ref.	PCM	WF	WS
Total mean HR (bpm)	107.20 ± 26.21	90.51 ± 13.23	109.08 ± 11.48	97.35 ± 2.99
Total mean BR (rpm)	26.25 ± 0.10	21.42 ± 6.20	29.50 ± 0.61	28.61 ± 0.70

Note. Values shown are average from all subjects, from 0 to 120 min.

Sweating and weight changes

Figure 6 shows the mean values for Total Body Liquid Loss (ΔW_L), Sweat Evaporation (ΔW_{fully clothed}) and Clothing Liquid Absorption (ΔW_CL) for each CV. Their calculation was explained in Protocol Description. The mean weight changes per CV after each test were recorded: +3 g for the WF and −141 g for the WS. PCM did not change its weight due to the impermeability. Total liquid loss, sweat evaporation, and clothing absorption increased consistently after the tests in the following order: PCM, WF, Ref., WS. During the Ref. test, subjects lost around 1819.5 ± 95.5 g of water through sweating, which equals to 1 L per hour. PCM CV shows the lowest total liquid loss: around 398.5 g lower than Ref., with reduced clothing absorption by ∼ 131.5 g and sweat evaporation by ∼177.0 g. WF reduced total liquid loss and sweat evaporation by ∼126 g, but had similar liquid absorption in clothing as the Ref. WS showed the highest liquid loss: higher evaporation by ∼71.5 g and clothing absorption by ∼ 127 g compared to Ref.

Figure 6.

Total Mean Liquid Loss, Sweat Evaporation and Clothing Liquid Absorption per cooling vest after 120 min. Values are average from all subjects.

Subjective data

The mean subjective ratings per cooling vest are presented in Figure 7 and Table 8. Warmth was reduced by around 1 point when subjects were wearing CVs, with PCM having the best results. All CVs reduced sweating wetness sensation, the PCM and WS showing the biggest reduction of 1 point. Overall comfort was slightly better with the WS compared to Ref., while the PCM slightly reduced it (still, by less than 0.5 points compared to Ref). Subjects commented that PCM felt uncomfortable at the start, not in terms of thermal comfort but due to the sharp edges of the solid PCM. Weight was perceived as neutral across all different CVs, despite the actual difference in weight between CVs.

Figure 7.

Mean Subjective Ratings per cooling vest. Values are average for all subjects, from 0 to 120 min.

Table 8.

Mean subjective ratings during the wear trials per cooling vest.

	Subjective parameters
CV type	Temperature (warmth)	Sweating wetness	Overall comfort	Clothing weight
Ref.	Very warm (2.90 ± 1.43)	Very wet (2.60 ± 1.39)	Neutral (0.70 ± 0.41)	Neutral (0.00 ± 0.00)
PCM	Warm (1.50 ± 0.37)	Wet (1.56 ± 0.37)	Neutral (0.23 ± 0.39)	Neutral (0.00 ± 0.00)
WF	Very warm (2.27 ± 0.57)	Wet (2.04 ± 0.41)	Neutral (0.56 ± 0.49)	Neutral (0.17 ± 0.24)
WS	Warm (1.83 ± 0.29)	Wet (1.67 ± 0.41)	Neutral (1.37 ± 1.57)	Neutral (0.20 ± 0.28)

Note. Values are average from all subjects, recorded from 0 to 120 min at 30 min intervals. The scales to describe each parameter are given in Protocol Description.

Statistical analysis of microclimate parameters

The ANOVA analysis from the four-sensors (n = 12) indicated significant differences in the first hour of testing. At 30 min (p < 0.001, partial η² = 0.60): PCM showed significantly lower T_mc (M = 34.50, SE = 0.39) than Ref. (M = 36.54, SE = 0.27, p = 0.002), WF (M = 35.99, SE = 0.26, p = 0.008), and WS (M = 36.00, SE = 0.31, p = 0.006). WF was also lower than Ref. with p = 0.038. At 60 min (p < 0.001, partial η² = 0.51): PCM (M = 35.15, SE = 0.35) remained significantly lower than Ref. (M = 37.04, SE = 0.27, p = 0.003), WF (M = 36.71, SE = 0.31, p = 0.010), and WS (M = 36.66, SE = 0.34, p = 0.017). No other CV showed significant reduction compared to Ref. At 90 and 120 min, no significant T_mc differences were found. RH_mc results at 30 min showed significant differences (p = 0.024, partial η² = 0.246) between PCM (M = 82.72, SE = 3.61) and WS (M = 89.36, SE = 4.46, p = 0.006), but not with Ref. At 60 min (p = 0.001, partial η² = 0.326): PCM (M = 87.20, SE = 1.09) was significantly lower than WF (M = 91.44, SE = 0.80, p = 0.013), but not compared to Ref. At 90 and 120 min, no significant differences were found.

The results from single sensor location (front upper chest) and six time points per interval (n = 18) further strengthened the findings. T_mc results from the 25–30 min interval showed significant differences: p < 0.001, partial η² = 0.897. PCM (M = 34.52, SE = 0.18) was significantly lower than Ref. (M = 37.11, SE = 0.07, p = 0.001), WF (M = 36.60, SE = 0.09, p = 0.001), and WS (M = 36.71, SE = 0.48, p = 0.001); WF and WS were also lower than Ref. (p = 0.001). For the 55–60 min interval (p < 0.001, partial η² = 0.786): PCM (M = 35.49, SE = 0.25) and WF (M = 37.22, SE = 0.06) were both significantly lower than Ref. (M = 37.60, SE = 0.08, p < 0.001). PCM was also lower than WF and WS (M = 37.21, SE = 0.20, p < 0.001). RH_mc at 25–30 min showed significant differences (p < 0.001, partial η² = 0.666). PCM (M = 77.87, SE = 0.79) was lower than Ref. (M = 80.66, SE = 0.94, p = 0.027); PCM, Ref., and WF (M = 78.83, SE = 1.53) were all significantly lower than WS (M = 87.99, SE = 1.42, p < 0.001). At 55-60 min (p < 0.001, η² = 0.401), PCM (M = 87.75, SE = 1.06) was lower than Ref. (M = 91.97, SE = 0.18, p = 0.001) and WF (M = 92.43, SE = 0.12), while WF was significantly higher than Ref. (p = 0.043).

Cooling vest effect and individual response

Figures 8 and 9 compare microclimate and subjective data per subject and cooling vest to see whether individual differences are outweighed by the impact of the vest. PCM has consistently lowest T_mc values across subjects. Two subjects show similar T_mc trends across vests, while P displays a distinct response for WS (higher T_mc than PCM and WF at 30 and 60 min). Subjective thermal responses do not consistently follow T_mc trends. CVs reduce warmth compared to Ref., except for one subject (K) where only PCM is perceived as less warm. Generally, PCM yields lower warmth ratings (except for P at 30 min). RH_mc values show a similar trend across subjects at 60 min, with PCM having lowest %. Subjective sweating wetness follows a similar trend for two subjects across CVs, except for P (where PCM and WF have identical ratings and WS has higher ratings). This comparison shows that PCM has consistently lowest T_mc and RH_mc values across subjects, while for WF and WS there is greater inter-subject variability. The relative differences between cooling vests in the objective data remain consistent across subjects. There is greater variability in subjective perception between subjects, and they do not fully match the objective microclimate data. The lower daypack load (% BW) for subject K (Table 2) appears to contribute to lower objective T_mc and RH_mc values compared to the other subjects that had similar body weight percentages. However, this effect is not reflected in the subjective responses.

Figure 8.

Mean microclimate temperature (T_mc) and thermal warmth ratings at 30 and 60 min for three subjects (G, P, K) across the cooling vest conditions. Lines represent individual values with standard deviation.

Figure 9.

Mean microclimate relative humidity (RH_mc) and thermal warmth ratings at 30 and 60 min for three subjects (G, P, K) across the cooling vest conditions. Lines represent individual values with standard deviation.

Discussion

This study provides novel insights into the thermophysiological effects of three passive cooling vests (PCM, WF, WS) worn beneath restrictive military clothing. Both advantages and limitations of these cooling systems are identified by integrating objective and subjective assessments. The study identifies when the passive cooling of each vest shifts from thermal relief to a thermal burden. While PCM consistently lowers microclimate temperature and humidity, subjective thermal and sweating wetness perceptions vary, indicating that personal comfort may not always match objective performance.

The reference clothing was identical across all CV conditions. Combat Shirt sleeves likely facilitated some evaporative heat transport from the arms, whereas evaporation the torso area was restricted by the backpack and ballistic vest. The wicking T-shirt exhibits a good one-way moisture transfer and a greater evaporative heat transport capacity than the combat shirt over it (Table 3). It likely contributed to a moisture gradient between the skin and combat shirt until moisture saturation was reached at around 30 min. Generally, the body’s ability to cool through dry heat transfer (conduction, convection, radiation) declines in an ambient T > 35°C, after which only sweat evaporation is effective.³⁶ The microclimate T_mc values show that this critical temperature was reached within the first 10 min for Ref., at 20–30 min for water-based cooling vests, and at 60 min for the PCM vest. The introduction of CVs increases total clothing weight, thermal and evaporative resistances. For the vests studied, this has a negative impact only once their cooling effect is exhausted. Once T_mc and RH_mc plateau, cooling vests become a burden and the PCM delays this by prolonging cooling by dry heat transfer.

Evaporative cooling vests performance

Both water-based vests provided immediate cooling from the initially cool water, but this quickly diminished as the absorbed heat raised the water temperature. After 60 min, both vests became a thermal burden (T_mc values exceeding the Ref.) leading to moisture buildup, thermal and respiratory stress. Their rapid loss of cooling effect under restrictive layers supports manufacturer warnings proving their limited practicality in such settings. In contrast, without restrictive layers over them, evaporative CVs showed cooling of up to 3.5 h in thermal manikin tests.¹² Ciuha et al.¹² discussed that WS and WF vests both rely on surface evaporation but restrict sweat evaporation. They showed that WF provided significantly less cooling than WS. In our study, WF showed significant T_mc reduction compared to Ref. at 30 min, confirmed by both single and multiple-sensor ANOVA analyses. However, at 60 min it retained significantly more humidity than PCM and Ref., which might have compromised comfort. Based on our study and previous findings by Ciuha et al., the WF vest shows low cooling power due to its design, regardless of the presence of restrictive layers. The minimal weight change of the WF vest (Sweating and weight changes), lower evaporation, and comparable absorption to Ref. suggest inhibited sweat and water transport that caused greater sweat accumulation in the baselayer below it. WS initially increased RH_mc by around 20% due to the soaked water. WS showed significant T_mc reduction compared to Ref. only from the ANOVA results at the 25–30 min interval, though with less significance than WF. It showed the highest weight loss and evaporation compared to other CVs and Ref., which is contrary to other studies on water-soaked clothing. Zhang et al.¹⁷ reported that a water-soaked shirt under an air-ventilation jacket reduced total water loss, skin temperature, HR and sweat rate. Ciuha et al.⁶ found that WS vests did not enhance evaporation even without additional layers. Still, in our case WS raised T_mc higher than in the previously mentioned study, indicating reduced performance when air flow is restricted over the vest. The increased total evaporation could be due to soaked water diffusing into areas not covered by the ballistic vest. Compensatory evaporation from uncovered areas like face and neck might have also occurred.³⁷ The reduced sweating wetness sensation in WS, despite the high objective RH_mc and liquid absorption, may be related to the water-skin contact leading to a more pleasant sensation. This further supports the findings from Fillingeri et al.³⁰ that surface wetness sensation can differ from actual wetness, depending on thermal or tactile cues. Additionally, wetness in warm environments can be perceived as more pleasant due to sensory adaptation and psychological expectations.³⁸

PCM cooling vest performance

The PCM vest provided the longest cooling duration, slowing the T_mc increase, reducing cardiovascular strain (HR and BR), sweat rate, total liquid loss and improving subjective thermal and sweating wetness. Here, the reduced HR suggests lower cardiovascular strain since it is supported by the observations from the other parameters. The PCM reduced T_mc on average by 1.5°C compared to Ref. over the 2 hours of testing. The slightly reduced RH_mc and lowest clothing absorption which indicate reduced sweat rate likely contributed to the lower sweating wetness perception. The PCM cooling effect diminishes more quickly after 60–70 min as observed by the more rapid T_mc rise above this point. As the microclimate temperature rose well above the PCM’s phase transition temperature (29°C), the increased sensible heat absorption contributed to a faster rise in T_mc. One way ANOVA confirmed the consistency of PCM in reducing T_mc significantly lower than all other vests and Ref. at 30 and 60 min, while RH_mc differences are less consistent. Single-sensor analysis confirmed significantly lower RH_mc for PCM compared to Ref. and WF, and generally lower RH_mc for PCM than WS in the first hour of testing. Despite the added weight compared to Ref. and the other vests, this vest was not perceived as heavier. Despite better thermal and wetness perception, the PCM vest was not rated as more comfortable, mainly due to the solid-state rigidity and sharp edges at the start, suggesting further design improvements.

Direct comparisons with other studies are challenging due to differences in clothing, measurement methods, and test conditions, though some parallels can be drawn. In Ciuha et al.⁶ passive and active cooling vests were tested at 35°C, 40% RH, 4.5 km/h for 2.5 h with lighter clothing and no restrictive outerwear. PCM showed the highest cooling efficiency, with a more pronounced effect than in our study (mean T_mc 24.8 ± 2.1°C vs 35.17 ± 0.53°C). This suggests ballistic wear may reduce PCM performance, although differences in exposure time or the unknown PCM melting temperature could also play a role. RH_mc increased under PCM (85% ± 8%) after 150 min, similar to the values observed in our study. Further research should compare cooling vests under identical conditions with and without ballistic wear to verify these effects.

Practical considerations of PCM cooling vest

The PCM vest could be a better solution for shorter operations under ballistic wear compared to active technologies, providing similar cooling effect along with added benefits of silent operation, low power consumption, easy maintenance. A liquid circulating vest tested under ballistic wear in a similar ambient (30°C, 40% RH), but shorter duration and higher walking pace (50 min, 6.5 km/h), reduced skin temperature by 1.2°C over 50 min⁷ This is slightly lower if compared with the PCM’s near-skin T_mc reduction in our study (2°C below Ref. at 60 min). As previously observed, the current PCM vest is not suitable for applications in hot climate exceeding 2 h. The observed discomfort from sharp edges in its solid state should be further addressed through different design or using encapsulation materials with better skin comfort. The studied vest is good in preventing cold-related discomfort due to its higher phase-change point that allows storage at higher ambient temperatures. In field applications, the current PCM with a phase change temperature of 29°C could be reactivated in the field using a portable fridge for around 30 min or during cooler nighttime conditions. The vest can be easily maintained by wiping it with alcohol. However, quick reactivation remains a logistical limitation if a portable fridge cannot be used. In further optimization, enhanced passive cooling can be achieved by covering a bigger body surface with PCM packs, but cooling extremities should be avoided to prevent impairing fine motor skills.¹¹ Adding passive elements like drying agents beneath the vest to absorb sweat appears unpromising - a study on a PCM vest showed that while microclimate humidity was reduced in heated cylinder tests, microclimate temperature increased by 0.6°C due to higher thermal resistance and heat release during moisture absorption.³⁹ Although insulation over PCM packs can reduce heat absorption from a hot environment and may be beneficial during lower-intensity activities,⁴⁰ under the current test conditions where skin temperature exceeds ambient T, this approach is unlikely to be effective. Durability after repeated use is an important practical consideration. PCM performance can degrade over multiple thermal cycles leading to changes in melting point, latent heat capacity, thermal conductivity, leakage, depending on the PCM composition, encapsulation, system design, and cycling conditions.⁴¹

Study limitations and future work

The current study shows meaningful trends in responses to different passive cooling vests even with the small sample size. Moreover, other related studies on cooling vests have recruited a similar number (one to four subjects).^42–46 However, this small sample size limits generalizability of the findings. Due to time and budget constraints, it was impossible to recruit more subjects or include thermal manikin testing. A larger sample size of at least eight subjects (minimum sample size calculated in Study Participants) would increase the statistical power, reduce variability, standard error and allow for more complex analyses. Guidelines for thermal comfort studies advice to include three to six more subjects beyond the minimum calculation to account for participants dropout or individual variation.¹⁸ Hydration status was not carefully controlled - it was assessed only through camelback weight changes without accounting for sweat, respiratory, or metabolic losses, electrolyte balance, or prior fluid intake. Sweat measurements lacked precision, and clothing absorption was estimated only from main garments, excluding items like the backpack, QRV, helmet, socks, and shoes. Future studies should use more accurate methods, such as urine specific gravity, bioelectrical impedance, or precise sweat loss measurements. Core temperature, a key marker of heat strain, could not be analyzed due to Bluetooth data storage failure during the trials. As it directly reflects internal thermal load, improved core temperature monitoring in future studies is necessary. Also, measuring both near-skin microclimate and direct skin temperature could provide clearer insights into temperature gradients and individual responses and improve comparability with other studies. A broader challenge in personal cooling research is the lack of standardized testing methods, which can improve consistency and research progress. Further evaluation of the PCM cooling vest in additional settings (sun radiation, low RH, different clothing) could yield different insights. It would be interesting to study the effect of outer and inner clothing layers on PCM performance and PCMs with melting temperature higher than 29°C. Assessing long-term stability, cooling performance and wear resistance after repeated will provide clearer evidence of longevity. For example, using methods such as an accelerated thermal cycling technique, heat transfer experiments by contact (with hotplate) and radiant heat, or extended field trials such as the study by Chan et al.^47–49

Conclusions

This study evaluated the performance of three passive cooling vests (PCM, WF, and WS) under Belgian Defense clothing with ballistic wear in hot climate. Wear trials determined the point where cooling vests shift from providing thermal relief to becoming a thermal burden. Once microclimate temperature and relative humidity plateau, vests become a burden and the PCM delays this by prolonging cooling by dry heat transfer. The PCM vest with phase transition point at 29°C was the most effective cooling solution, lowering T_mc by 1.5°C over 120 min compared to the reference, consistently across subjects. It also reduced sweat, cardiovascular strain, perceived warmth and wetness. The vest can be a better solution than active cooling for operations up to 2 hours, allowing silent use, low energy consumption and easy maintenance. However, its performance declines after 1 hour and after 2 hours of use it likely becomes a burden. This is supported by the significant differences in T_mc and RH_mc found in PCM compared to the other CVs and Ref., at 30 and 60 min. No significant differences were found between CVs at 90 or 120 min for either T_mc or RH_mc, suggesting the cooling effects are most distinct in the first hour of use. Design improvements are needed for better comfort under ballistic wear and long-term durability of the PCM needs to be further investigated. Water-based vests (WS and WF) showed limited effectiveness under ballistic wear with only temporary cooling. They showed significantly lower T_mc than Ref. at 30 min, with the WF vest demonstrating stronger statistical evidence. Both vests showed T_mc reduction up to 0.5°C at 60 min followed by a rapid performance drop after 1 hour beyond which they become a thermal burden. Regardless of the presence of an outer layer, both vests limit natural sweat evaporation, but the additional airflow restriction from ballistic wear accelerates their loss of cooling effectiveness. Subjective comfort did not always align with objective cooling performance, especially sweating wetness, highlighting the importance of individual perception during evaluation of cooling systems. These findings provide direction for improving thermophysiological comfort of restrictive clothing in hot climate and support further research and design refinement of wearable passive cooling systems.

Footnotes

Acknowledgments

The authors extend their gratitude to the Ministry of Defense for funding this research. They also sincerely thank the test subjects for their voluntary participation, collaboration, and valuable feedback. Appreciation to Inuteq for generously providing the cooling vests used in the study.

ORCID iDs

Magdalena Georgievska

Benny Malengier

Informed consent

Informed consent was obtained from all individual participants included in the study. The authors confirm that all participants provided explicit consent for the publication of anonymized data derived from this study.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was funded by the Ministry of Defense, grant number FED.DIV.2020.0031.01.

Declaration of conflicting interests

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

Data Availability Statement

The datasets generated and analysed during the current study are available from the corresponding author on reasonable request.*

References

Goforth

Lisman

Deuster

. The physiological impact of body armor cooling devices in hot environments: a systematic review. Mil Med 2014; 179(7): 724–734.

Elson

. Evaluation of personal cooling systems and simulation of their effects on human subjects using basic and advanced virtual environments. Doctoral Thesis, Kansas State University, Manhattan, KS, 2016.

Georgievska

Malengier

Odhiambo

, et al. User-centric evaluation of thermophysiological comfort and challenges with defense garments used in hot climate operations. AuxDefense Book of Abstracts 2024. University of Minho, 2024.

Tokizawa

Son

Oka

, et al. Effectiveness of a field-type liquid cooling vest for reducing heat strain while wearing protective clothing. Ind Health 2020; 58(1): 63–71.

Zhou

Lou

Fan

. Quantitative comparison of personal cooling garments in performance and design: a review. Processes 2023; 11(10): 2976.

Ciuha

Valenčič

Ioannou

, et al. Efficacy of cooling vests based on different heat-extraction concepts: the HEAT-SHIELD project. J Therm Biol 2023; 112: 103442.

Yuan

Wei

, et al. Effects of a liquid cooling vest on physiological and perceptual responses while wearing stab-resistant body armor in a hot environment. Int J Occup Saf Ergon 2022; 28(2): 1025–1032.

Song

Wang

. The hybrid personal cooling system (PCS) could effectively reduce the heat strain while exercising in a hot and moderate humid environment. Ergonomics (Lond) 2016; 59(8): 1009–1018.

Ferri

Cravello

Dotti

, et al.

Thermal comfort physiological evaluation: which sensors for the most accurate assessment?

Proceedings of the 19th World Textile Conference - Autex 2019, Belgium, 2019. https://openjournals.ugent.be/autex/article/id/63811/

10.

Hunt

Tan

, et al. The incorporation of phase change material into soft armour inserts: achieving a level of cooling without compromising ballistic protection. Adv Sci Technol 2017; 100: 38–42.

11.

Maley

Minett

Bach

AJE

, et al. Internal and external cooling methods and their effect on body temperature, thermal perception and dexterity. PLoS One 2018; 13(1): e0191416.

12.

Ciuha

Valenčič

Mekjavic

. Cooling efficiency of vests with different cooling concepts over 8-hour trials. Ergonomics (Lond) 2021; 64(5): 625–639.

13.

Calautit

Wang

, et al. A review of the applications of phase change materials in cooling, heating and power generation in different temperature ranges. Appl Energy 2018; 220: 242–273.

14.

Sinaga

Darkwa

Omer

, et al. The microencapsulation, thermal enhancement, and applications of medium and high-melting temperature phase change materials: a review. Int J Energy Res 2022; 46(8): 10259–10300.

15.

Wang

Yan

Pan

. Optimization strategies of microencapsulated phase change materials for thermal energy storage. J Energy Storage 2023; 68: 107844.

16.

Elmoughni

Atalay

Ozlem

, et al. Thermoelectric clothing for body heat harvesting and personal cooling: design and fabrication of a textile-integrated flexible and vertical device. Energy Tech 2022; 10(10): 2200528.

17.

Tokizawa

. Effects of wetted inner clothing on thermal strain in young and older males while wearing ventilation garments. Front Physiol 2023; 14: 1122504.

18.

Lei

Lan

Wang

. Indoor thermal comfort research using human participants: guidelines and a checklist for experimental design. J Therm Biol 2023; 113: 103506.

19.

Leonidas

Mantzios

Tsoutsoubi

, et al. Gkiata. Occupational heat stress: multi-country observations and interventions. Int J Environ Res Public Health 2021; 18(12): 6303.

20.

ISO 9920:2007 . Textiles - physiological effects - measurement of thermal and water-vapour resistance under steady-state conditions (sweating guarded-hotplate test).

21.

Atlas SDL . MMT: moisture management tester. Textile testing products. https://sdlatlas.com/products/mmt-moisture-management-tester (accessed 8 April 2025).

22.

AATCC TM195 . Liquid moisture management properties of textile fabrics.

23.

Croda Industrial Specialties . CrodaTherm™ phase change materials overview. https://www.crodaindustrialspecialties.com/en-gb (accessed 8 April 2025).

24.

EN-ISO 9886: 2004 . Ergonomics — evaluation of thermal strain by physiological measurements.

25.

ASTM F2371-16 . Standard test method for measuring the heat removal rate of personal cooling systems using a sweating heated manikin.

26.

Tanaka

Monahan

Seals

. Age-predicted maximal heart rate revisited. J Am Coll Cardiol 2001; 37(1): 153–156.

27.

Eijsvogels

Bongers

Veltmeijer

, et al. Cooling during exercise in temperate conditions: impact on performance and thermoregulation. Int J Sports Med 2014; 35(10): 840–846.

28.

CORE body temperature sensor. https://corebodytemp.com/ (accessed 8 April 2025).

29.

ISO 10551:2019 . Ergonomics of the physical environment — subjective judgement scales for assessing physical environments.

30.

Filingeri

Fournet

Hodder

, et al. Tactile cues significantly modulate the perception of sweat-induced skin wetness independently of the level of physical skin wetness. J Neurophysiol 2015; 113(10): 3462–3473.

31.

Skenderi

Cubrić

Mijović

. The comparison of human perception to the measured sweat transfer rate. Coll Antropol 2011; 35(2): 433–437.

32.

Othman

Zaki

Ahmad

, et al. In-situ measurement of pedestrian outdoor thermal comfort in universities campus of Malaysia. KnE Soc Sci 2019; 3(21): 608–621.

33.

Lan

Lian

. Application of statistical power analysis – how to determine the right sample size in human health, comfort and productivity research. Build Environ 2010; 45(5): 1202–1213.

34.

Chou

Coyle

. Cardiovascular responses to hot skin at rest and during exercise. Temp Multidiscip Biomed J 2023; 10(3): 326–357.

35.

Siddiqui

. Effects of vasodilation and arterial resistance on cardiac output. Clin Exp Cardiol 2011; 02(11): 1–6.

36.

Baldwin

Benmarhnia

Ebi

, et al. Humidity’s role in heat-related health outcomes: a heated debate. Environ Health Perspect 2023; 131(5): 055001.

37.

Havenith

Richards

Wang

, et al. Apparent latent heat of evaporation from clothing: attenuation and “heat pipe” effects. J Appl Physiol 2008; 104(1): 142–149.

38.

Filingeri

Fournet

Hodder

, et al. Body mapping of cutaneous wetness perception across the human torso during thermo-neutral and warm environmental exposures. J Appl Physiol 2014; 117(8): 887–897.

39.

Itani

Ghaddar

Ghali

. Innovative PCM-desiccant packet to provide dry microclimate and improve performance of cooling vest in hot environment. Energy Convers Manag 2017; 140: 218–227.

40.

Udayraj

Song

, et al. Performance enhancement of hybrid personal cooling clothing in a hot environment: PCM cooling energy management with additional insulation. Ergonomics (Lond) 2019; 62(7): 928–939.

41.

Sharma

. PCM thermal storage Technology and its military applications: a review. Int J Res Appl Sci Eng Technol 2021; 9(12): 542–565.

42.

Biermann

. Improved thermal control body armor. Natl Inst Justice Rep 2003: 76.

43.

Amitrano

Coccia

Ricciardi

, et al. Design and validation of an E-textile-based wearable sock for remote gait and postural assessment. Sensors (Peterb, NH) 2020; 20(22): 6691.

44.

Chinevere

Cadarette

Goodman

, et al. Efficacy of body ventilation system for reducing strain in warm and hot climates. Eur J Appl Physiol 2008; 103(3): 307–314.

45.

Cadarette

Cheuvront

Kolka

, et al. Intermittent microclimate cooling during exercise-heat stress in US army chemical protective clothing. Ergonomics (Lond) 2006; 49(2): 209–219.

46.

Kim

. Heat release and wear comfort characteristics of the ceramic imbedded fabrics for cold weather protective clothing. J Ind Text 2022; 52.

47.

Katish

Allen

Squires

, et al. Thermal stability of organic Phase Change Materials (PCMs) by accelerated thermal cycling technique. Thermochim Acta 2024; 737: 179771.

48.

Renard

Machnowski

Puszkarz

. Assessment of thermal performance of phase-change material-based multilayer protective clothing exposed to contact and radiant heat. Appl Sci 2023; 13: 9447.

49.

Chan

APC

Zhang

Wang

, et al. A field study of the effectiveness and practicality of a novel hybrid personal cooling vest worn during rest in Hong Kong construction industry. J Therm Biol 2017; 70: 21–27.