Study on the Performance of Cage Braided Tube with PEG/CNT Composite Coatings for Heated Tobacco Product Filters

Abstract

Heated tobacco products (HTPs) frequently induce user discomfort due to high mainstream smoke temperatures. To address this challenge and improve the inhalation experience, this experiment designed and prepared a cage-shaped braided tube as the cooling section of the filter for HTPs. The thermal, cooling, suction resistance, and smoke composition properties of the filter were tested and analyzed. Thermal analysis (DSC/TG) revealed a 116.53 J/g increase in endothermic enthalpy for PEG-impregnated samples, accompanied by maintained thermal stability (decomposition temperature ≈ approximately 350 °C). The 0.8 wt% Carbon Nanotube (CNT) composite achieved exceptional thermal conductivity (0.597 W/m·K), representing a 521% improvement over untreated controls. The braided tube optimal performance (3 mm inner diameter, 30% PEG/0.8% CNT) reduced the highest smoke temperature to 47.8 °C while maintaining acceptable suction resistance (68.5 Pa, 56.4% reduction vs. commercial IQOS filters). GC-MS analysis confirmed negligible alterations in smoke composition (p > 0.05). This innovation offers an effective thermal management solution that does not compromise sensory experiences.

1. Introduction

HTPs are a new type of tobacco product that utilize heating elements to heat tobacco leaves or reconstituted tobacco. They generate smoke through low-temperature pyrolysis of tobacco for consumers to inhale, resulting in a significant reduction in harmful components compared to traditional cigarette smoke [1,2]. However, due to the short length of the heated tobacco product section and filter, the heating temperature is as high as 200–400 °C [3]. The longer heating time means that the temperature of the smoke inhaled from heated non-combustible cigarettes is even higher than that from traditional cigarettes, with the highest temperature of the smoke at the filter tip reaching 70 °C. Excessively high smoke temperatures can cause smokers to experience a burning sensation in their mouth, resulting in a different smoking experience compared to traditional cigarettes. Existing studies on smoke cooling methods can generally be divided into two categories: one involves adding cooling materials, and the other involves designing cooling structures [4]. Common phase-transition cooling materials employed in cigarette filters include polylactic acid [5,6] and polyethylene glycol. Polylactic acid, as a phase-transition material, undergoes a phase transition that absorbs heat; however, its enthalpy change is small, resulting in limited cooling capacity [7]. Additionally, during smoking, the polylactic acid film melts and sticks together, which increases the suction resistance. Polyethylene glycol (PEG) is a phase-transition material with excellent heat storage properties [8], a broad phase-transition temperature range of −55 °C to 100 °C, and a significantly high latent heat of fusion [9,10]. It is frequently blended with other substances and incorporated into filters to achieve a cooling effect while preserving aroma [11]. However, this substance exhibits leakage problems during the melting phase transition and poor thermal conductivity, which restricts the application of PEG. Carbon nanotubes (CNTs) possess a unique structure and high thermal and electrical conductivity [12,13]. Currently, numerous studies have been conducted to improve the thermal conductivity of materials by adding carbon nanotubes [14]. For example, J. Lee et al. [15] prepared MX/O-CNT/PEG composite materials, which showed a 580% enhancement in thermal conductivity compared to pure PEG materials. Carbon nanotubes are combined with phase-transition materials in multiple fields, such as batteries and wall construction, to enhance the thermal performance of materials [16,17]. However, carbon nanotubes exhibit hydrophobic properties and are difficult to disperse, limiting their potential [18]. A single cooling method is difficult to achieve favorable cooling results; so, this study combines two cooling approaches to achieve flue gas cooling.

IQOS cigarettes are representative products of heated tobacco products [19,20] and boast a significant market share. Their smoking devices and cartridges have a structure as shown in Figure 1 [21] below, using polylactic acid pleated sheets as the cooling section to achieve both rapid reduction in smoke temperature and low resistance to draw while minimizing particulate retention in the HTP cooling section. This study innovatively designed and fabricated a novel cage-structured braided tube cooling section featuring low filtration, lowered draw resistance, and enhanced cooling performance. The research outcomes provide a theoretical foundation and innovative methodology for the parameter design and industrial-scale manufacturing of HTP filter components, thereby addressing the longstanding technical challenge in the tobacco industry of balancing cooling efficiency with smoking comfort in filter development. This advancement represents a significant breakthrough in overcoming the performance trade-offs between thermal regulation and sensory perception in heated tobacco product filtration systems. Acetate fiber is a widely used material for cigarette filters in the tobacco market [22,23,24]. This study employed Cellulose acetate yarn (CA), polyvinyl alcohol yarn (PVA), and polyester yarn (PET) as raw materials, using braiding, soaking, and impregnation processes to design and prepare braided tube filters with a cage-like structure. The braided architecture features several critical design elements: (1) perforated tube walls creating tortuous flow channels that extend the smoke path, (2) a hollow core structure with an optimized wall thickness-to-diameter ratio for balanced mechanical and thermal performance, and (3) an integrated phase-change cooling system using polyethylene glycol (PEG)/carbon nanotube (CNT) composites. Polyethylene glycol has hydrophilic groups and is highly water-soluble. To make carbon nanotubes more soluble, this study selected modified multi-walled carboxylated carbon nanotubes. PEG and CNT were dissolved in water to prepare a PEG/CNT mixed solution in a specific ratio. The PEG/CNT mixture was then loaded onto the braided tube via the impregnation method, achieving a composite cooling effect combining cooling structures and cooling materials. Additionally, the carbon nanotubes restrict the flow of polyethylene glycol, preventing leakage during phase transition while increasing the impregnation rate [25].

This study seeks to address the challenges of high draw resistance and excessive smoke temperature in heated tobacco product (HTP) filters by developing a novel cage-structured braided tube via braiding, soaking, and impregnation processes. The thermal properties, surface morphology, smoke cooling efficiency, absorption resistance, and smoke composition of the braided tubes were systematically characterized and analyzed. Special emphasis was placed on investigating the influence of polyethylene glycol/carbon nanotube (PEG/CNT) content on the cooling performance of the braided tube cooling section. These findings offer critical guidance for the development of advanced cooling sections in heated tobacco product filters. By optimizing material selection and structural design, this study offers new strategies for the industrial development of green and efficient filters, which is of great importance for promoting the industrialization and upgrading of HTPs, representing a significant advancement in HTP technology.

2. Materials and Methods

2.1. Materials

The materials used are as follows: Cellulose acetate yarn (CA) (denier 1100, Nantong Acetate Fibre Co., Ltd., Nantong, China); Water-soluble polyvinyl alcohol yarn (PVA) (denier 1120, dissolution temperature 90 °C, Shandong Rongfeng Textile Co., Ltd., Shandong, China); Polyester yarn (PET) (denier 900D, melting point 110 °C, Fujian Jinhaosheng Textile Technology Co., Ltd., Fujian, China); Multi-walled carboxylated carbon nanotube powder (tube diameter 3–15 nm, tube length 15–30 nm, specific surface area 260–270 m²/g); and Polyethylene glycol powder (mass fraction 6000, Shandong Ruisheng Pharmaceutical Excipients Co., Ltd., Shandong, China).

2.2. Preparation of PEG/CNT Mixture Solutions

Polyethylene glycol (PEG) exhibits excellent phase-change endothermic properties, while carbon nanotubes (CNTs) possess superior thermal conductivity. The relatively short length of the cooling section limits the residence time of smoke during passage, thereby constraining the heat absorption efficiency of polyethylene glycol (PEG). To address this limitation, CNTs are incorporated to enhance thermal conductivity, which facilitates more efficient heat transfer and enables PEG to absorb greater amounts of thermal energy from the smoke stream. This combination of phase-change material and thermally conductive material ultimately achieves a more effective reduction in smoke temperature. In coating applications, the mass fraction of PEG in PEG/CNT mixed solutions must balance two critical factors: CNT dispersibility and coating viscosity. The CNTs must effectively enhance the thermal conductivity of the composite coating while maintaining a non-agglomerated state. We hypothesize that an optimal PEG/CNT mass ratio exists that simultaneously ensures excellent CNT dispersion while coatings maintain superior thermal conductivity and heat absorption capacity, thereby enabling rapid smoke temperature reduction. Prepare PEG/CNT mixture solutions with different ratios (30 wt% PEG/0.2 wt% CNT, 30 wt% PEG/0.4 wt% CNT, 30 wt% PEG/0.6 wt% CNT, 30 wt% PEG/0.8 wt% CNT). PEG/CNT mixture solutions were carried out using ultrasonic treatment and magnetic stirring. First, different amounts of polyethylene glycol and carbon nanotube powder were dispersed in an aqueous solution in a beaker, and the two solutions were physically blended to form 50 g of PEG/CNT mixed solution. The solution was subjected to rigorous homogenization via alternating magnetic stirring (500 rpm, 1 h) and ultrasonic treatment (1 h), followed by additional stirring (30 min) to ensure complete dispersion. The solution was then allowed to stand, as shown in Figure 2, maintaining homogeneity without observable phase separation after 24 h of quiescent storage. This optimized preparation ensured uniform distribution of CNTs within the PEG matrix while preventing particle aggregation.

2.3. Preparation of the Cooling Section of the Braided Tube

Using cellulose acetate yarn as the base material, polyvinyl alcohol yarn and polyester yarn were co-spun, and a cage-shaped braided tube for heated tobacco product filters was prepared via a braiding, soaking, and impregnation process. CA/PVA/PET braided tubes were prepared, using CA yarn as the base material. During preparation, PVA yarn and PET yarn were co-braided with CA yarn, followed by heating and immersion to dissolve the PVA yarn in water and cause thermal shrinkage of the PET yarn. The residual concentrated solution gel layer from the dissolved PVA yarn and the thermally deformed PET yarn imparts shape retention and rigidity to the braided tube, resulting in HTP filters with excellent hardness and roundness.

The braiding process employs a 32-spindle braiding machine to interlace Cellulose acetate yarn and polyester yarn. The polyester yarn is heat-shrunk to shape the Cellulose acetate yarn. To ensure the braiding structure and uniformity of the subsequent braiding tube pores, a 1:1 braiding ratio is ultimately selected, meaning that the polyvinyl alcohol yarn and CA/PET composite yarns ratio is 16:16. The resulting fabric structure features each yarn strand overlapping with another, which is in turn covered by yet another strand, ensuring structural integrity and pore uniformity throughout the braided matrix. The braided circular tube is shown in Figure 3, and the surface structure of the fabric is illustrated in Figure 4. The braiding machine operates at a speed of 600 rpm. During the braiding process, 7 mm PU tubes are inserted to define the diameter of the braided fabric. This setup facilitates the preparation of subsequent filter samples, yielding braided tubes with consistent outer diameters of 6.8–7.0 mm.

The inner diameter of the CA/PVA/PET composite braided tubes was precisely controlled using nylon tube cores of varying diameters (2.5 mm, 3.0 mm, and 3.5 mm). To achieve optimal structural modification, the thermal processing protocol was carefully optimized as follows: (1) Immersion in 100 °C water (exceeding the 90 °C dissolution temperature of PVA yarn) for 30 min in a 3 L water bath ensured dissolution of the PVA component while creating the desired perforated morphology; (2) Subsequent immersion in PEG/CNT composite solution for 60 min facilitated homogeneous distribution of phase-change materials throughout the braided matrix; (3) Final oven drying at 80 °C for 30 min stabilized the composite structure. Figure 5 shows the braided tubes soaked in a mixed solution of 30% PEG and different mass fractions of CNT.

The prepared braided tubes were precision-cut to a length of 17.5 mm to ensure optimal compatibility with standard cigarette dimensions. For systematic characterization, samples were designated using a standardized nomenclature, reflecting their key parameters; for instance, braided tubes with an inner diameter of 2.5 mm (outer diameter 6.8–7.0 mm) treated with 30% PEG/0.2% CNT solution are identified as “2.5–30% PEG/0.2% CNT”. The complete fabrication process is illustrated in Figure 6. The braided tube filter was connected to the cigarette, and the internal structure of the physical cigarette sample is shown in Figure 7.

2.4. Characterization

Experimental Instruments and Test Conditions

The thermal conductivity, thermogravimetric properties, and phase-change enthalpy of braided tubes were tested and analyzed using a thermal constant analyzer and a comprehensive thermal analysis instrument. Thermal conductivity measurements were performed with a thermal constant analyzer (equipped with a Model 7531 probe) under controlled conditions (25 °C, 50% relative humidity), with a 0.002 W heating input. Thermogravimetric analysis was conducted using a comprehensive thermal analyzer under a nitrogen atmosphere (flow rate: 50 mL/min) at a heating rate of 10 °C/min, with the experiment conducted at room temperature. Differential scanning calorimetry (DSC) measurements were carried out in a nitrogen environment, with the temperature program ranging from 25 °C to 400 °C at 10 °C/min to determine phase-change enthalpy characteristics. This comprehensive thermal characterization ensured accurate quantification of the key performance parameters essential for evaluating the cooling efficiency and thermal stability of the composite filter materials.

The microstructure of the braided tubes was comprehensively characterized using advanced imaging techniques. Micro-computed tomography (Micro-CT) scanning provided three-dimensional visualization of both the overall architecture and detailed pore network within the tube walls. Scanning electron microscopy (SEM) revealed the morphology and distribution of residual components on individual yarn fibers, enabling a thorough evaluation of the structural integrity and material distribution throughout the braided matrix.

Cigarette smoke is an aerosol with an extremely complex composition. Selecting a suitable cooling filter that neither reduces the smoke concentration nor compromises the smoke volume is of great significance for enhancing the sensory quality of HTPs [26]. Gas chromatography offers advantages including high separation efficiency, excellent selectivity, high sensitivity, minimal sample demand, fast analysis speed, and wide applicability, making it one of the primary analytical methods for identifying harmful components in cigarette smoke [27,28,29]. The chemical composition of mainstream smoke was quantitatively analyzed using an Agilent 6890 A gas chromatography system coupled with Cambridge filter pad collection. Conventional indicators of mainstream flue gas (total particulate matter, nicotine, moisture, tar, and glycerol) were tested in accordance with national standards GB/T 19609-2024, GB/T 23355-2009, and GB/T 23203.1-2013. Additionally, the glycerol content in the total particulate matter of cigarette flue gas was measured. Analytical conditions—Injection port temperature: 250 °C; split ratio: 10:1; chromatographic column: DB170/1 30 m × 0.53 mm × 0.5 μm; column flow rate: 3 mL/min; oven temperature: 120 °C for 3 min, then heated at 20 °C/min to 220 °C and held for 6 min; detector temperature: 230 °C, H₂ 35 mL/min, air 350 mL/min.

Using the heat-not-burn (HNB) smoke comprehensive tester, the inlet smoke temperature and draw resistance of the cooling section of different sample filters were tested separately, using the Canadian deep-draw method. Measurements were conducted in a controlled environment (22 ± 1 °C, 60 ± 2% relative humidity) with the following parameters: 55 mL puff volume, 2 s duration, 30 s interval, and 10 replicate puffs per sample. The temperature sensor is connected to a computer, which reads and records data every 0.5 s. Each sample was tested three times in parallel experiments.

3. Results and Discussion

3.1. Thermal Properties and Morphological Characteristics of Braided Tubes

The thermal properties of the composite braided tubes were systematically investigated through thermogravimetric analysis (TG) and differential scanning calorimetry (DSC). As revealed by the DSC thermograms in Figure 8, PEG-impregnated samples exhibited distinct endothermic peaks corresponding to phase transitions. As shown in

Table 1. The enthalpy of absorption for each sample.

Samples	Heat Absorption Enthalpy (J/g)
Unsoaked PEG solution braided tube	1.27
Soak braided tubes in a 10 wt% PEG solution	45.74
Soak braided tubes in a 20 wt% PEG solution	77.79
Soak braided tubes in a 30 wt% PEG solution	92.85
Soak braided tubes in a 40 wt% PEG solution	117.80

, ΔH values increased from 44.47 J/g to 116.53 J/g as the PEG concentration rose from 10 wt% to 40 wt%. As a typical phase-change material (PCM), polyethylene glycol (PEG) exhibits significant endothermic behavior during its phase transition. The enthalpy of absorption of the braided tube shows a direct proportional relationship with the mass fraction of PEG. This linear correlation indicates that the thermal energy absorption capacity of the cooling section can be regulated by controlling the PEG content, where higher PEG concentrations lead to proportionally enhanced cooling performance. When the PEG mass fraction in the solution exceeds 30%, the increase in the enthalpy of absorption gradually decreases, and the viscosity of the solution increases significantly, affecting the subsequent dissolution of CNTs. Therefore, we established 30 wt% as the optimal PEG concentration for composite preparation.

CNTs have been well established for applications in cigarette filters, with no evidence of them reacting with smoke to produce toxic substances. Numerous studies and patents have validated their effectiveness in reducing harmful components in cigarette smoke [30,31]. Figure 9 shows the thermogravimetric (TG) curves of five types of braided tubes loaded with PEG/CNT mixtures of different mass ratios. The thermal decomposition temperature of the braided tubes not impregnated with the PEG/CNT mixture solution is around 325 °C. After impregnation with the PEG/CNT mixture solution, all samples exhibit a thermal decomposition temperature of approximately 350 °C. The observed variations in the thermogravimetric (TG) curves can be attributed to two synergistic mechanisms: first, the substantial thermal energy absorption by polyethylene glycol (PEG) during its solid–liquid phase transition effectively slows the rate of temperature rise of the braided tube matrix; second, the three-dimensional thermally conductive network formed by carbon nanotubes (CNTs) significantly enhances the composite’s thermal diffusivity and improves its overall thermal stability. The heating temperature for HTPs ranges from 300 °C to 400 °C, and the temperature of the smoke when it reaches the cooling section is approximately 110 °C to 120 °C. Within this temperature range, the material of the braided tube in the cooling section of the filter does not undergo thermal decomposition and generate additional chemical components, thereby demonstrating excellent thermal stability and chemical safety in the cooling section of braided tubes.

Thermal conductivity, also known as the thermal conductivity coefficient, is a key parameter for characterizing the heat transfer capability of fiber materials. Fiber materials with higher thermal conductivity coefficients result in fabrics with poorer thermal insulation and heat retention properties but superior rapid-cooling performance. This study tested CA/PVA/PET braided fabrics and CA/PVA/PET braided fabrics impregnated with four different mass fractions of PEG/CNT solutions (30 wt% PEG/0.2 wt% CNT, 30 wt% PEG/0.4 wt% CNT, 30 wt% PE G/0.6 wt% CNT, 30 wt% PEG/0.8 wt% CNT). The thermal transport properties of the fibrous composites were evaluated through conductivity measurements. As shown in

Table 2. Thermal conductivity of braided fabrics.

Sample Names	Thermal Conductivity (W/m·K)	Rate of Increase
30%PEG/0%CNT braided fabric	0.096	-
30%PEG/0.2%CNT braided fabric	0.146	52.1%
30%PEG/0.4%CNT braided fabric	0.213	122%
30%PEG/0.6%CNT braided fabric	0.414	331%
30%PEG/0.8%CNT braided fabric	0.597	521%

Note: “-” is a blank comparison.

, incorporation of carbon nanotubes (CNTs) into the CA/PVA/PET braided matrix significantly enhanced the thermal conductivity (λ), following a distinct concentration-dependent trend:

The pristine braided fabric exhibited a baseline thermal conductivity of 0.096 W/m·K. Introduction of 0.2–0.8 wt% CNTs generated progressive improvement:

0.2 wt% CNT: λ = 0.146 W/m·K (+52.1%)
0.4 wt% CNT: λ = 0.213 W/m·K (+122%)
0.6 wt% CNT: λ = 0.414 W/m·K (+331%)
0.8 wt% CNT: λ = 0.597 W/m·K (+521%)

(1)

This indicates that both the thermal conductivity and heat storage performance of the braided fabric are enhanced with the addition of carbon nanotubes, and the thermal conductivity of the braided tubes is directly proportional to the mass fraction of carbon nanotubes.

Microstructural characterization of the braided tubes was performed using scanning electron microscopy (SEM) and micro-computed tomography (Micro-CT), with representative images shown in Figure 10 and Figure 11. The three-dimensional CT reconstruction revealed an intricate network architecture, featuring highly curled yarns with complex deformation patterns and minimal surface fuzz, creating tortuous flow paths that enhance convective heat transfer. The braided structure exhibits tight inter-yarn connections, with uniformly distributed through-wall pores. SEM analysis of PEG/CNT-impregnated samples (30 wt% PEG/0.8 wt% CNT) showed partial retention of PVA residues in fiber interstices (Figure 9a,b), attributed to incomplete dissolution from densely packed fiber bundles during hydrothermal treatment. This residual PVA forms a beneficial adhesive layer upon drying, improving structural rigidity without compromising porosity. Figure 9c presents the characteristic tubular morphology of the MWCNTs-COOH, while Figure 9d confirms successful CNT deposition on fiber surfaces after solution processing, with nanotubes preferentially aligned along fiber axes to maximize thermal conduction pathways. Combined, these imaging results demonstrate effective microstructural control, achieving both optimized cooling geometry and enhanced thermal transport properties.

3.2. Pressure Drop Performance of the Cooling Section of the Braided Tube

The pressure drop in a cigarette reflects the ease or difficulty of smoking it and is generally characterized by suction resistance. Cigarette suction resistance is a key factor affecting the sensory quality of cigarettes and is one of the key indicators emphasized in cigarette design [32,33]. The cooling section of the braided tube is characterized by a perforated wall structure, which forms intricate smoke channels. These channels extend the smoke flow path, thereby enabling effective smoke cooling. Furthermore, the hollow interior of the tube contributes to reducing suction resistance. To eliminate the influence of factors such as uneven tobacco filling on the overall suction resistance of the cigarette, the suction resistance of the cooling section is measured separately.

The results of the suction resistance tests for samples from each cooling section after smoking are shown in Figure 12. As observed in the figure, the average suction resistance of the braided tube cooling section after suction is 68.5 Pa, and the diameter of the core material exhibits a negative correlation with the suction resistance. The incorporation of PEG/CNT has no significant impact on the suction resistance of the braided tube. A significance analysis was conducted on the suction resistance test results of all samples. Different letters in the samples indicate statistically significant differences between groups (p < 0.05). The analysis results are shown in

Table 3. Significance analysis of the suction resistance of each sample.

Sample Names	Suction Resistance			Significance
2.5–0%PEG0%CNT	91.0	92.0	89.0	c
2.5–30%PEG/0.2%CNT	93.5	95.2	91.8	ab
2.5–30%PEG/0.4%CNT	94.0	92.0	96.0	ab
2.5–30%PEG/0.6%CNT	92.4	94.0	90.8	bc
2.5–30%PEG/0.8%CNT	95.0	95.9	94.1	a
3–0%PEG/0%CNT	80.5	80.0	81.0	def
3–30%PEG/0.2%CNT	80.0	80.6	79.4	ef
3–30%PEG/0.4%CNT	82.0	81.0	83.0	de
3–30%PEG/0.6%CNT	83.0	81.0	85.0	d
3–30%PEG/0.8%CNT	83.1	82.1	83.9	d
3.5–0%PEG/0%CNT	78.0	80.0	76.0	fg
3.5–30%PEG/0.2%CNT	80.0	81.0	79.0	ef
3.5–30%PEG/0.4%CNT	76.0	74.5	77.5	g
3.5–30%PEG/0.6%CNT	78.0	76.0	80.1	fg
3.5–30%PEG/0.8%CNT	82.1	82.0	83.0	de

Note: Different letters in the samples indicate statistically significant differences between groups (p < 0.05).

. From the significance analysis results, it can be seen that the core material diameter has a significant effect on the suction resistance of the cooling section of the braided tube, while the change in suction resistance after immersing the braided tube in the PEG/CNT mixed solution is small. In samples of the cooling section of the braided tube with the same core material diameter, the addition of CNT in the mixed solution has no significant effect on the suction resistance of the samples. The suction resistance of the IQOS heated tobacco stick is 157 Pa. Therefore, the suction resistance of the cooling section of the braided tube is 56.4% lower than that of the IQOS heated tobacco stick, resulting in a more pleasant smoking experience—a critical advantage for the development of next-generation heated tobacco products.

3.3. Cooling Performance of the Cooling Section of the Braided Tube

When high-temperature smoke flows through the braided tube structure, thermal energy is initially transferred to the tube walls via conduction. The design featuring curved yarns and a porous morphology significantly prolongs the aerosol flow path, thereby enhancing heat dissipation. Concurrently, the incorporated carbon nanotubes establish an interconnected high-thermal-conductivity network that facilitates rapid heat redistribution from localized hot spots to the entire surface area. Meanwhile, the PEG on the tube wall undergoes phase transition to absorb heat, achieving simultaneous heat dissipation and absorption, which enables the smoke to cool rapidly. The temperature of the smoke entering the mouth after passing through the cooling section and filter section is defined as the inlet temperature, which is the parameter measured by the HNB smoke comprehensive tester. Figure 13 shows the inlet end temperature of each puff for test cigarettes prepared using different cooling section materials for the filter. As shown in Figure 13, the highest inlet temperature of the smoke typically occurs during the second puff. Since the smoke temperature decreases with each subsequent puff after reaching its peak, this study focuses primarily on the cooling effect of the braided tube cooling section on the highest smoke temperature. During the inhalation process of HTPs, factors such as uneven distribution of tobacco flakes in the tobacco segments of sample cigarettes and variations in the pore distribution of the braided tube walls may affect the flue gas temperature, resulting in certain measurement errors, even for cigarettes with identical braided tube cooling section parameters. As shown in the flue gas temperature test data below, the maximum flue gas temperature of the braided tube cooling section without immersion in the solution ranges from 55 °C to 60 °C. After immersion in the mixed solution, the cooling performance of the braided tube cooling section is further enhanced. As the proportion of CNTs in the mixed solution increases, the flue gas cooling effect gradually improves. This phenomenon can be attributed to the exceptional thermal conductivity of carbon nanotubes (CNTs), which enables highly efficient heat transfer. The CNTs form a thermal bridge network that rapidly redistributes localized heat from high-temperature smoke regions across the entire surface of the braided tube structure. This network effect significantly expands the effective heat dissipation area while simultaneously compensating for the low thermal conductivity of the base material through bidirectional thermal transport—both longitudinally along the tube axis and transversely across its cross-section.

When the inner diameter of the braided tube is 3 mm and it is immersed in a 30% PEG/0.8% CNT solution, the maximum flue gas temperature can be reduced to 47.8 °C. The maximum flue gas temperature of the IQOS heated tobacco stick can only reach a minimum of 54.3 °C.

Thermal profiling using calibrated K-type thermocouples (measurement plane shown in Figure 14) quantified the cooling section’s exceptional heat dissipation capacity: it reduces the smoke temperature from 112 °C at the cooling section inlet to 47.8 °C at the outlet, representing a 57.3% reduction (ΔT = 64.2 °C).

This study selected braided tubes treated with a 30% PEG/0.8% CNT solution—which demonstrated optimal cooling performance in preliminary experiments—as the research subject to investigate the influence of different core diameters on the smoke cooling efficiency. To ensure the reliability and reproducibility of the experimental results, five parallel samples were prepared for each of the three core diameter variations under simulated real-world usage conditions.

We used one-way ANOVA to compare the significance of the experimental results. The maximum flue gas temperature during puffing for each sample is presented in

Table 4. Maximum flue gas temperature during puffing for each sample.

Samples	Temp (°C)
2.5–30%PEG/0.8%CNT	52	51.3	52.1	51.6	52.9
3–30%PEG/0.8%CNT	47.8	47.5	48.1	47.7	48.7
3.5–30%PEG/0.8%CNT	49.2	49.6	48.9	49.5	49.9

. A homogeneity of variance test yielded a p-value > 0.05, indicating that the data met the assumption of homogeneity of variance; this validated the use of ANOVA and subsequent post hoc multiple comparisons. Key findings from the three standard ANOVA output tables (descriptive statistics, ANOVA table, and post hoc multiple comparisons table) were consolidated into

Table 5. Variance analysis of core material diameter’s impact on maximum flue gas temperature.

	2.5 mm	3 mm	3.5 mm	F	p	LSD
	(n = 5)	(n = 5)	(n = 5)
Temp (°C)	51.98 ± 0.61	47.96 ± 0.47	49.72 ± 0.38	84.86 **	<0.01	1 > 2,3; 3 > 2
	(M ± SD)	(M ± SD)	(M ± SD)

Note: ** p < 0.01; 1–2.5 mm, 2–3 mm, 3–3.5 mm.

for comprehensive presentation of the analytical results.

The results presented in Table 5 demonstrate statistically significant differences in the maximum smoke temperature among different core diameters (F = 84.86, p < 0.01). Comparative analysis of the mean values reveals that the 3 mm core diameter exhibits a lower maximum flue gas temperature compared to both the 2.5 mm and the 3.5 mm cores. Post hoc multiple comparison tests further indicate that the 2.5 mm core shows a significantly higher maximum smoke temperature than both the 3 mm and the 3.5 mm cores, while the 3.5 mm core maintains a significantly higher temperature than the 3 mm core. Collectively, these findings suggest that, under identical PEG and CNT content conditions, the braided tube section with a 3 mm core diameter achieves optimal cooling performance.

3.4. Cigarette Smoke Composition

Cambridge filter pads were used to fractionate mainstream smoke into particulate and gaseous phases for comprehensive chemical characterization. The particulate phase is captured on the filter surface, and the gaseous phase permeates through the filter. This study employs gas chromatography to quantify key aerosol constituents.

Table 6. Flue gas component release data.

Sample Name	TPM	Water	Nicotine	Tar	Glycerine
2.5–30%PEG/0.2%CNT	27.90	13.58	1.11	13.21	4.07
2.5–30%/0.4%CNT	28.60	13.43	1.10	14.07	3.88
2.5–30%PEG/0.6%CNT	28.40	14.15	1.07	13.18	3.33
2.5–30%PEG/0.8%CNT	27.80	13.51	0.98	13.31	3.51
3.0–30%PEG/0.2%CNT	29.50	14.44	1.12	13.95	3.16
3.0–30%PEG/0.4%CNT	31.00	15.23	1.20	14.57	3.51
3.0–30%PEG/0.6%CNT	25.80	12.33	0.83	12.64	2.69
3.0–30%PEG/0.8%CNT	30.90	15.03	1.11	14.76	4.43
3.5–30%PEG/0.2%CNT	30.50	15.70	1.16	13.64	3.59
3.5–30%PEG/0.4%CNT	28.90	15.18	1.01	12.71	2.69
3.5–30%PEG/0.6%CNT	30.30	14.72	1.17	14.42	3.54
3.5–30%PEG/0.8%CNT	29.40	14.87	1.11	13.42	4.27
PLA sheet	32.30	16.05	1.24	14.96	2.97

presents the data on the release of smoke components:

To evaluate the impact of structural parameters on performance, samples were categorized into three test groups, based on the inner diameter of the braided tube cooling section (2.5 mm, 3.0 mm, and 3.5 mm), and the data were subjected to differences to assess significant variations in key performance metrics across groups. The test results are shown in

Table 7. Differential testing of different parts in various components.

Ingredients	Hollow Diameter of Cooling Section of Braided Tube			F	Significance
	2.5 mm	3 mm	3.5 mm
	M ± SD	M ± SD	M ± SD
TPM	28.175 ± 0.386	29.3 ± 2.432	29.775 ± 0.754	1.222	0.339
Water	13.668 ± 0.327	14.258 ± 1.328	15.118 ± 0.433	3.1	0.095
Nicotine	1.065 ± 0.059	1.065 ± 0.162	1.113 ± 0.073	0.258	0.778
Tar	13.443 ± 0.422	13.98 ± 0.958	13.548 ± 0.704	0.612	0.563
Glycerine	3.698 ± 0.338	3.448 ± 0.736	3.523 ± 0.647	0.184	0.835

:

As shown in Table 7, the significance values for each flue gas component in the braided tube samples prepared from different types of core materials are greater than 0.05, which does not reach statistical significance. An overall analysis of the data reveals that there is no significant difference in the flue gas components between the samples. These results suggest that neither the raw materials used in the braided tube cooling section nor the incorporation of PEG/CNT have a significant impact on smoke components. The thermal-modulating braided tube structure achieves its cooling function without altering the fundamental smoke chemistry that determines sensory characteristics.

4. Conclusions

This study aims to address the challenges of high suction resistance and excessive smoke temperature in heated tobacco product (HTP) filters. CA/PVA/PET braided tubes were fabricated, via a multi-step process involving braiding, soaking, and impregnation processes, as the cooling section of HTP filters. The smoke passes through the perforated sections of the braided tubes, and the axial and radial flow paths within the tubes extend the smoke flow, thereby enhancing heat dissipation. The thermal, suction resistance, cooling, and smoke composition properties of the filters were tested and analyzed. Key findings are as follows:

• The endothermic capacity of the braided tubes showed a positive correlation with the polyethylene glycol (PEG) concentration. At a PEG impregnation level of 30 wt%, the composite achieved an optimal enthalpy of 92.85 J/g—a viscosity range that simultaneously facilitated effective carbon nanotube (CNT) dispersion. The synergistic effect of 30 wt% PEG and 0.8 wt% CNT resulted in a remarkable thermal conductivity of 0.597 W/m·K, corresponding to a 521% enhancement over the untreated baseline. Thermogravimetric analysis (TG) confirmed exceptional thermal stability across all PEG/CNT composites, with decomposition temperatures consistently maintained at 350 °C; therefore, the material of the braided tube at the cooling section of the filter rod will not undergo thermal decomposition and generate additional chemical components, and has good thermal stability.
• The average suction resistance of the braided tube cooling section after smoking is 68.5 Pa, and the diameter of the core material is negatively correlated with the suction resistance. The incorporation of PEG/CNT has no significant effect on the suction resistance of the braided tube. The suction resistance of the IQOS heated tobacco stick is 157 Pa. As a result, the suction resistance of the cooling section of the braided tube was reduced by 56.4% compared to the suction resistance of IQOS heated tobacco sticks, thereby improving the smoking experience.
• The cooling performance of CA/PVA/PET braided tubes was systematically evaluated, revealing that untreated samples exhibit maximum flue gas temperatures ranging from 55 to 60 °C. Impregnation with PEG/CNT composite solutions significantly enhanced the cooling efficiency, with the temperature reduction effect exhibiting a positive correlation with the CNT concentration. Optimal cooling performance was achieved using 3 mm diameter tubes treated with a 30 wt% PEG/0.8 wt% CNT solution, which effectively lowered the flue gas temperature to 47.8 °C through the combined mechanisms of PEG’s phase-change heat absorption properties and the enhanced thermal conductivity of CNTs.
• GC-MS analysis of smoke components revealed no statistically significant differences (p > 0.05 for all measured constituents) among CA/PVA/PET braided tubes fabricated with different core materials. An overall analysis of the data further confirmed the absence of significant variations in smoke components between samples. These results confirm that integrating braided tube structures into the cooling section maintains smoke composition integrity, thereby preserving the essential sensory characteristics of the smoking experience.

The application of braided tube cooling sections in heated tobacco product filters exhibits significant potential for industrial applicability, offering an optimal balance between raw material costs and safety requirements during manufacturing. The braided tube can function independently as a cooling section while remaining compatible with existing heated tobacco product commercial filter systems. Although the “Regulations on the Administration of Tobacco Products in Domestic Duty-Free Markets” policy prohibits the sale of HTPs in duty-free markets, the policy imposes no restrictions on the research and development of innovative filter materials, thereby creating a viable pathway for technological advancement in this field. The findings of this study provide both theoretical foundations and innovative paradigms for the parameter design and industrial-scale manufacturing of HTP filters. Through the development and publication of multiple patented technologies [34,35,36], this research has driven the industry toward overcoming the longstanding technical challenge of simultaneously achieving optimal cooling efficiency and smoking comfort in filter design—a critical advancement that reconciles previously incompatible performance requirements. Future research directions should focus on three aspects of improvement to enhance experimental reproducibility: (1) expanding sample sizes to improve statistical power, (2) optimizing tobacco segment homogeneity through advanced manufacturing controls, and (3) implementing more precise porosity characterization techniques for braided tube structures. Additionally, comprehensive long-term stability testing should be conducted to rigorously validate the safety profile of these innovative filter designs. These approaches will significantly enhance the reliability of research findings and their translational potential for industrial applications.