Development of Thermally Insulating Nonwovens from Milkweed Fibers Using an Air-Laid Spike Process

Abstract

Milkweed (MW) fiber is a natural fiber that provides tremendous thermal insulation properties due to its lightweight hollow structure. This study aimed to investigate the effect of milkweed fiber as a thermal fiber in nonwovens. Milkweed fibers were blended with a low-melt fiber consisting of a polyethylene terephthalate core, a polyolefin sheath (LM 2.2), and polylactic acid (PLA) fiber. Nonwovens with different fiber contents were manufactured using an air-laid Spike process to determine their effect on thermal and mechanical properties. Then, the nonwovens were compared with Thinsulate^® and Primaloft^®, two commercially synthetic insulation products. Structural properties, including mass per unit area, thickness, and porosity and thermal properties were studied. Furthermore, compression and short-term compression recovery were also evaluated. The results revealed that milkweed-based nonwovens that contained 50 wt% or 70 wt% of milkweed presented a lower thermal conductivity than synthetic nonwovens. Milkweed nonwovens of the same thickness provided identical thermal resistance as Thinsulate^® and Primaloft. Sample 3, composed of 50 wt% MW, 20 wt% LM 2.2, and 30 wt% PLA, demonstrated the same thermal insulation as Thinsulate^® with a weight three times lighter. Milkweed nonwovens presented higher moisture regain values than Thinsulate^® and Primaloft^®, without affecting thermal conductivity.

1. Introduction

Textile and clothing industries are recognized as significant sources of global pollution, ranking among the sectors with the highest intensive consumption of raw materials [1]. Several estimates indicate that the clothing industry is responsible for nearly 10% of global carbon emissions and 20% of the world’s water usage. Moreover, this sector is a major contributor to the generation of non-biodegradable wastes in landfills, with an estimated annual production of 92 million tons [2,3]. Additionally, the clothing industry is the largest consumer of synthetic fibers, notably polyester (PET) and polyethylene (PE) [4]. Specifically, the industry’s consumption figures for PET are substantial, reaching 57 million tons in 2020 and escalating to 61 million tons in 2021, representing approximately 54% of the fiber’s annual production [5]. The widespread adoption of these synthetic fibers for clothing carries significant environmental implications, primarily stemming from the intensive exploitation of non-renewable resources and the substantial energy demands required for their manufacturing process.

Despite its environmental drawbacks, PET remains crucial for manufacturing thermal insulation filling for winter jackets and gear. PET filling materials offer a low density and excellent thermal insulation properties, essential for mitigating thermal stress and ensuring heat comfort during the cold season. Furthermore, PET filling materials provide three-dimensional stability and good compressibility, both essential for winter jackets and gear. Additionally, these materials serve as a cost-effective alternative to animal fibers such as sheep wool and down feathers, alleviating concerns about their ethical use. However, to curb the use of non-sustainable materials, the clothing industry is beginning to embrace the adoption of renewable and environmentally friendly filling materials. This includes nonwovens made from natural fibers, recycled plastics, or biopolymers [6,7]. The utilization of natural fibers presents numerous advantages, including low cost, non-toxicity, biodegradability, and renewability. However, unlike their manmade counterparts, natural fibers come with intrinsic limitations such as dissimilar shapes and morphology, hydrophilic nature, and relatively weak mechanical properties [8,9]. The combination of natural fibers and manmade polymers to produce nonwovens is a well-known strategy to improve mechanical properties and mitigate the impact of the inherent defects of natural fibers on the final material’s properties. Among manmade polymers, polylactic acid (PLA) stands out as a sustainable biopolymer for producing eco-friendly materials. This semicrystalline and biodegradable thermoplastic has demonstrated enhanced dimensional and mechanical stability in nonwovens reinforced with natural fiber such as flax, jute, and kapok [10]. Furthermore, to respect the sustainability principles, the natural fibers used for thermal insulation in clothing should be indigenous to the production region.

Asclepias Syriaca, or common milkweed (MW), is a native plant to North America that grows easily across the regions in most types of soil without the need for pesticides and fertilizers [11]. Despite its abundance and unique characteristics, the utilization of the floss or fiber from common milkweed in clothing has not received the attention it deserves. Milkweed fiber extends from the seeds as a solid cylinder composed of a thin cell wall and a large lumen representing 70% of the total volume [12]. This distinctive hollow structure confers exceptional attributes to milkweed fibers compared to other natural fibers, notably, lightweight and a large air volume. These features offer exceptional acoustic and thermal insulation capabilities due to the low thermal conductivity of the air entrapped in the lumens [13,14]. However, milkweed exhibits relatively low mechanical properties and significant variations in morphology [8,15,16,17]. Additionally, unlike most natural fibers, milkweed fiber is protected by a hydrophobic wax layer, which provides affinity to olefins and polymeric resins [18,19,20].

In light of the above considerations, several research studies have explored the use of milkweed and other natural hollow fibers in textile materials for thermal insulation, using carding to obtain either fiber webs or slivers for spinning [14,21,22,23,24,25]. A. Gharehaghaji and S. Davoodi [21] investigated the fiber damage and loss rate of Estabragh, a type of natural hollow fiber, during carding. Using scanning electron microscopy (SEM), the authors observed that several Estabragh fibers sustained considerable damage after passing through the various carding cylinders. For a nonwoven containing 90% Estabragh fibers and 10% PET fibers, they achieved a fiber loss rate of 50%. Similarly, Karthik and Murugan [22] studied the cohesion between cotton and hollow seed fibers from trellis-vine, Pergularia daemia B. Their study revealed that the carding of blends generated a significant fiber loss in contrast to yarns made entirely of cotton [22]. They noted that both fiber loss and yarn quality deterioration were directly proportional to the hollow fiber content, irrespective of the spinning technique used. The lack of cohesion between cotton and hollow fiber was attributed to the brittleness and smooth surface of the latter [22]. Sakthivel, Mukhopadhyay, and Palanisamy reported similar results in their work with yarns made of cotton and Mudar, a different variety of hollow natural fiber. They also pointed out the brittleness and smooth surface of Mudar as factors contributing to the lack of fiber cohesion [26]. The thermal insulation capacity of milkweed fibers was evaluated by Parmar et al., who showed that adding milkweed fiber to cotton or polyester fabric leads to a better thermal insulation [27]. By the same token, Crews et al. studied the thermal insulation capacity of milkweed fiber and assessed the insulation efficiency of winter jackets filled with loose-fill milkweed alone, milkweed combined with down feathers, and a nonwoven made by the meltblown process and composed of 60% milkweed and 40% polyolefin [28]. They measured the thermal insulation capacity of each type of insulation filling before and after washing to determine the effect on thermal insulation properties. The results showed that a jacket filled with 100% loose milkweed had a notably low thermal insulation capacity. Washing also strongly affected the fibers and their insulating efficiency. In contrast, a milkweed/down 50:50 blend provided an efficiency comparable to pure down, even after washing. The nonwoven exhibited superior insulation capacity to Thinsulate^®, a commercially available synthetic insulation filling while retaining a smooth texture even after cleaning.

Whereas milkweed fiber has demonstrated its excellent thermal insulation capacity, its suitability for other functional properties of winter jackets must be investigated. Comfort, compressibility for convenient storage, resilience to quickly regain their shape, durability, and resistance to moisture are important characteristics to consider. It is also crucial to explore alternative spinning processes to minimize fiber loss, prevent fiber damage, and improve the structural and mechanical properties of textiles made from milkweed while preserving their insulation capacity. With the ultimate objective to develop an alternative to the use of synthetic fibers for thermal insulation, various blends of milkweed, PET, and PLA were produced and characterized. The manufacturing process used was air-laid technology because of its high-performance regarding fiber loss and damage. It must be noted that to the best of the authors’ knowledge, this is one of the first times that the air-laid Spike process was used for such a production. The mass per unit area, thermal conductivity, thermal insulation, compression, short-term compression recovery, and moisture regain of the nonwovens were then evaluated. Thermal conductivity at a relative humidity of 90% was also measured. Furthermore, the thermal insulation properties of the milkweed-based nonwovens were compared against Thinsulate^® and Primaloft^®, two commercially available synthetic insulating nonwovens commonly used in winter jackets.

2. Materials and Methods

2.1. Material

Milkweed fiber was provided by Coop Monark (Quebec, QC, Canada). The density, length, diameter, and wall thickness were equal to 0.30 g/cm³, 25 ± 3 mm, 22 ± 6 μm, and 1.50 ± 0.23 μm, respectively, [13]. Figure 1 shows a milkweed fiber.

A low-melt fiber (LM 2.2) consisting of a PET core and a polyolefin sheath in a 50:50 ratio was purchased from Indorama Ventures (Bobinger, Germany). PLA fiber with a fineness of 1.5 dtex was purchased from Ingéo NatureWorks (Blair, NE, USA).

Table 1. Properties of the fibers.

Fiber	Fineness (dtex)	Fiber Length (mm)	Fiber Diameter (μm)	Density (g/cm3)
MW	0.84–2.2	25 ± 3	22 ± 6	0.30 [12]
LM 2.2	1.98–2.42 *	12 ± 0.02	16 ± 3	1.43
PLA	1.5 *	51 ± 0.4	12.2 ± 1	1.28

* Datasheets from companies. LM 2.2 and PLA were chosen due to their similar range of fineness values.

reports the properties of the fibers.

2.2. Sample Preparation

In our previous work, preliminary tests were carried out to make nonwovens from milkweed and PLA fibers in a weight ratio of 75:25 using a SPIKE air-laid process with the objective of producing carpeting for automotive applications. In the present work, the objective was to prepare and characterize different blends of milkweed, LM 2.2, and PLA by varying the process parameters for the purpose of thermal insulation of winter jackets. Two types of thermal insulation fillings were prepared: a thick nonwoven (TNW) with a weight per unit area between 110 and 170 g/m² and a thinner nonwoven, called web, with a weight per unit area between 40 and 60 g/m². The fiber contents were varied to evaluate their impact on the functional properties of the nonwovens.

Firstly, the fibers were introduced into the air-laid system to produce the nonwovens. Subsequently, the nonwovens were thermally bonded at 150 °C by melting the polymeric fiber network. This consolidation step was applied to strengthen the nonwoven structure and promote its three-dimensional stability. Figure 2 depicts a schematic design illustrating the different stages of the SPIKE air-laid process. The nonwovens were processed at the Centre Européen des non-tissés (CENT) in Tourcoing, France.

Figure 3a shows the thick nonwoven (TNW) fabric produced using the air-laid process. Figure 3b illustrates the different webs.

Table 2. Specification of nonwoven sample production.

Samples	MW (wt%)	LM 2.2 (wt%)	PLA (wt%)	Type of Product
1	30	10	60	Web
2	70	10	20	Web
3	50	20	30	Web
4	50	10	40	TNW
5	50	20	30	TNW
6	50	40	10	TNW
7	50	20	30	Web

illustrates the composition of different nonwovens, including milkweed, LM 2.2, and PLA, along with their respective product types.

Two commercially available thermal insulation nonwovens made of 100% PET, Thinsulate^® and Primaloft^®, were chosen for comparison purposes.

2.3. Characterization of Raw Fibers

Fineness

The fineness of milkweed fiber (T) is calculated from the diameter (d) and the density (ρ) of fibers as expressed in Equation (1):

$$\mathrm{T}={\mathrm{d}}^2\times \mathrm{\rho }\times {\displaystyle \frac{\mathrm{\pi }}{4}}$$

(1)

In total, 30 fibers were used to calculate the fineness.

2.4. Characterization of the Nonwovens

2.4.1. Physical Properties

The mass per unit area ($\mathrm{M}\mathrm{a})$ in g/m² was measured by ISO 9073-1 [29], using at least 10 specimens per condition. Thickness ($\mathrm{T}\mathrm{H})$ was measured according to ISO 9073-2 [30], using a resistance of 0.1 kPa. At least ten measurements were obtained per specimen. The density ($\mathrm{\rho }$) was calculated for each nonwoven using Equation (2):

$$\mathrm{\rho }={\displaystyle \frac{\mathrm{M}\mathrm{a}}{\mathrm{T}\mathrm{H}}}$$

(2)

2.4.2. Thermal Resistance Properties

Thermal conductivity, $\mathrm{\lambda },$ was measured using a heat-flow meter FOX 314 (TA instruments, New Castle, DE, USA) according to ASTM C518 [31] (two-plates method). The nonwovens were placed between a hot and a cold plate simulating the human body temperature and external environment, respectively. Measurements of thermal conductivity were converted into values of thermal insulation of clothing (I_T) in clo, as defined by the ASHRAE Standard [32]. The “clo” unit represents the amount of insulation required for a person at rest to maintain thermal equilibrium at 21 °C. I_T is calculated using Equation (3).

$${\mathrm{I}}_{\mathrm{T}}={\displaystyle \frac{\mathrm{T}\mathrm{H}}{\mathrm{\lambda }\times \mathrm{K}}}={\displaystyle \frac{\left({\mathrm{T}}_{\mathrm{h}}-{\mathrm{T}}_c\right)\times \mathrm{S}}{\mathrm{Q}\times \mathrm{K}}}$$

(3)

where $\mathrm{\lambda }={\displaystyle \frac{-\mathrm{H}\times \mathrm{T}\mathrm{H}}{\left({\mathrm{T}}_{\mathrm{h}}-{\mathrm{T}}_{\mathrm{c}}\right)\times \mathrm{S}}}$ and “$\mathrm{T}\mathrm{H}$” represent the thickness of the nonwoven in meters, “Q” the input power in Watts, T_h and T_c the temperatures of the hot and cold plates, “S” the area of the nonwoven in square meters, and “K” a constant (0.155 m².°C/clo W) [33]. The values of T_h and T_c were set at 37 °C and 21 °C, respectively. For each value, at least five measurements were taken. The thermal resistance R was calculated from Equation (4):

$$\mathrm{R}={\displaystyle \frac{\mathrm{T}\mathrm{H}}{\mathrm{\lambda }}}$$

(4)

2.4.3. Compression and Short-Term Compression Recovery Properties

Compression and short-term compression recovery were measured according to ASTM D6571-22 [34], following the four steps outlined in Figure 4. For each value, at least eight measurements were taken.

Step 1: The average initial height (A) of the nonwoven was measured without weight.

Step 2: A weight of 7.26 kg was placed on the nonwoven and left for 10 min. The average height under compression (B) was measured.

Step 3: The weight was removed and the nonwoven was allowed to recover for an additional 10 min. The height of the nonwoven (E) was then measured.

Step 4: The nonwoven was compressed again for 24 h. Subsequently, the weight was removed and the nonwoven was allowed to recover for 1 h. The final height, (J), was then measured.

Figure 4 shows the different steps to measure the compression and short-term compression recovery of the nonwovens according to ASTM D6571-22.

The compression (C) was calculated using Equation (5).

$$\mathrm{C}={\displaystyle \frac{\mathrm{A}-\mathrm{B}}{\mathrm{A}}}\times 100$$

(5)

The short-term compression recovery (N) was determined using Equation (6).

$$\mathrm{N}={\displaystyle \frac{\mathrm{J}}{\mathrm{E}}}\times 100$$

(6)

2.4.4. Porosity

The porosity (ε) was calculated as the ratio of pore volume to total volume, using Equation (7). In this equation, ${\mathrm{V}}_{\mathrm{p}\mathrm{o}\mathrm{r}\mathrm{e}\mathrm{s}}$ represents the pore volume, ${\mathrm{V}}_{\mathrm{n}\mathrm{o}\mathrm{n}\mathrm{w}\mathrm{o}\mathrm{v}\mathrm{e}\mathrm{n}}$ the total volume of the nonwoven, ${\mathrm{V}}_{\mathrm{f}\mathrm{i}\mathrm{b}\mathrm{e}\mathrm{r}\mathrm{s}}$ the fiber volume, $\mathrm{S}$ the surface of the nonwoven, ${\mathrm{m}}_{\mathrm{f}\mathrm{i}\mathrm{b}\mathrm{e}\mathrm{r}\mathrm{s}}$ the mass of the fiber, $\mathrm{M}\mathrm{a}$ the mass per unit area, ${\mathrm{\rho }}_{\mathrm{f}\mathrm{i}\mathrm{b}\mathrm{e}\mathrm{r}\mathrm{s}}$ the density of the fiber, and $\mathrm{T}\mathrm{H}$ the thickness of the nonwoven.

$$\mathrm{\epsilon }={\displaystyle \frac{{\mathrm{V}}_{\mathrm{p}\mathrm{o}\mathrm{r}\mathrm{e}\mathrm{s}}}{{\mathrm{V}}_{\mathrm{n}\mathrm{o}\mathrm{n}\mathrm{w}\mathrm{o}\mathrm{v}\mathrm{e}\mathrm{n}}}}={\displaystyle \frac{{\mathrm{V}}_{\mathrm{n}\mathrm{o}\mathrm{n}\mathrm{w}\mathrm{o}\mathrm{v}\mathrm{e}\mathrm{n}}-{\mathrm{V}}_{\mathrm{f}\mathrm{i}\mathrm{b}\mathrm{e}\mathrm{r}\mathrm{s}}}{{\mathrm{V}}_{\mathrm{n}\mathrm{o}\mathrm{n}\mathrm{w}\mathrm{o}\mathrm{v}\mathrm{e}\mathrm{n}}}}=1-{\displaystyle \frac{{\mathrm{V}}_{\mathrm{f}\mathrm{i}\mathrm{b}\mathrm{e}\mathrm{r}\mathrm{s}}}{{\mathrm{V}}_{\mathrm{n}\mathrm{o}\mathrm{n}\mathrm{w}\mathrm{o}\mathrm{v}\mathrm{e}\mathrm{n}}}}=1-{\displaystyle \frac{{\mathrm{V}}_{\mathrm{f}\mathrm{i}\mathrm{b}\mathrm{e}\mathrm{r}\mathrm{s}}}{\mathrm{S}\times \mathrm{T}\mathrm{H}}},$$

(7)

Given that the mass of pores ($m_{pores}$) is negligible relative to the mass of the nonwoven ($m_{nonwoven}$), it can be inferred that $m_{nonwoven}=m_{fibers}$.

$$\mathrm{\epsilon }=1-{\displaystyle \frac{{\mathrm{m}}_{\mathrm{f}\mathrm{i}\mathrm{b}\mathrm{e}\mathrm{r}\mathrm{s}}}{{\mathrm{\rho }}_{\mathrm{f}\mathrm{i}\mathrm{b}\mathrm{e}\mathrm{r}\mathrm{s}}\times \mathrm{S}\times \mathrm{T}\mathrm{H}}}=(1-\mathrm{M}\mathrm{a}\times {\displaystyle \frac{1}{{\mathrm{\rho }}_{\mathrm{f}\mathrm{i}\mathrm{b}\mathrm{e}\mathrm{r}\mathrm{s}}\times \mathrm{T}\mathrm{H}}})$$

(8)

2.4.5. Moisture Regain (MR)

The moisture regain was quantified using Equation (9) in accordance with ASTM D2654-22 [35] and D1776/D1776M-20 [36]. The samples were stored at a relative humidity of 90% at 21 °C until the mass variation was less than 0.01% between two successive weightings.

$$\mathrm{M}\mathrm{R}=\left({\displaystyle \frac{{\mathrm{M}}_{\mathrm{w}\mathrm{e}\mathrm{t}}-{\mathrm{M}}_{\mathrm{d}\mathrm{r}\mathrm{y}}}{{\mathrm{M}}_{\mathrm{d}\mathrm{r}\mathrm{y}}}}\right)\times 100$$

(9)

where $\mathrm{M}\mathrm{R}$ is the percentage of moisture regain, ${\mathrm{M}}_{\mathrm{w}\mathrm{e}\mathrm{t}}$ the mass of the wet sample, and ${\mathrm{M}}_{\mathrm{d}\mathrm{r}\mathrm{y}}$ the mass of the dry sample.

3. Results and Discussion

3.1. Properties of Nonwovens

Table 3. Specification of nonwoven samples.

Samples	MW (wt%)	Mean Ma (g/m2)	Standard Deviation Ma (g/m2)	Mean TH (mm)	Standard Deviation TH (mm)	Density (kg/m3)
1	30	55.34	1.47	5.05	0.16	10.96
2	70	60.75	1.61	4.80	0.35	12.66
3	50	60.51	1.56	5.12	0.24	11.82
4	50	161.45	1.54	8.40	0.46	19.22
5	50	113.20	1.98	6.65	0.24	17.02
6	50	170.15	2.22	10.45	0.60	16.28
7	50	56.90	1.07	4.55	0.60	12.51
Thinsulate®	0	166.84	0.60	5.02	0.39	33.24
Primaloft®	0	86.86	0.30	5.08	0.12	17.10

shows the milkweed content, thickness, mass per unit area, and density of the nonwovens.

3.2. Thermal Conductivity

Table 4. Values of thermal conductivity (λ), thermal resistance, specific thermal resistance, and thermal insulation.

Samples	MW (wt%)	Average λ (mW/m·K)	Standard Deviation λ (mW/m·K)	Thermal Resistance R (m2·K)/W	Specific Thermal Resistance Rspecific (m2·K)/(W·kg/m3)	Thermal Insulation (clo)
1	30	35.32	0.04	0.143	0.013	0.922
2	70	33.13	0.03	0.145	0.011	0.935
3	50	33.96	0.09	0.151	0.013	0.972
4	50	32.81	0.02	0.256	0.014	1.651
5	50	33.44	0.01	0.199	0.012	1.283
6	50	33.65	0.02	0.311	0.019	2.003
7	50	32.55	0.07	0.140	0.011	0.902
Thinsulate®	0	33.31	0.09	0.151	0.005	0.972
Primaloft®	0	34.14	0.02	0.149	0.009	0.960

presents the thermal conductivity, thermal resistance, specific thermal resistance, and thermal insulation of the nonwovens.

It can be observed that all the nonwovens present thermal conductivity values lower than Thinsulate^® and Primaloft^® except for sample 1. This outcome is due to its lower milkweed content (30%). Samples 2 to 7 show better insulation properties. However, increasing the milkweed content from 50 wt% to 70 wt% does not improve the thermal conductivity. Indeed, as seen in Table 4, sample 5, which contains 50% milkweed, has a thermal conductivity of 33.44 mW/m·K, whereas sample 2 has a very similar thermal conductivity (33.13 mW/m·K) with 70% milkweed.

Milkweed nonwovens demonstrate superior thermal conductivity compared to those made from natural and synthetic fibers. Indeed, many researchers have studied the thermal conductivity values of nonwoven fibers made of natural and synthetic fibers [37,38,39,40,41]. For example, for multilayered polypropylene (PP) nonwoven fabrics, researchers found thermal conductivity values ranging from 31.1 to 38.1 mW/m·K for densities ranging from 95.4 to 178.6 kg/m³ [41]. In contrast, for air-laid nonwovens made of kenaf/PE/PP in a weight ratio of 35:35:30, they found thermal conductivity values ranging from 34.9 to 36.6 mW/m·K for a density ranging from 57.7 to 70.4 kg/m³ [39].

3.2.1. Thermal Resistance

Samples 1 to 3, Thinsulate^®, and Primaloft^® have a thickness of approximately 5 mm, and all demonstrate a thermal resistance value of 0.14–0.15 (m²·K)/W; however, samples 1 to 3 are made of milkweed fibers, as shown in Table 3 and Table 4. Milkweed can limit heat transfer similarly to Thinsulate^® and Primaloft^®, which are composed of polyester fibers. Furthermore, as shown in Table 4, Sample 6 exhibits a better thermal insulation-to-density ratio compared to the other samples, with a specific thermal resistance value of 0.019 (m²·K)/(W·kg/m³).

3.2.2. Thermal Insulation

Figure 5 compares the thermal insulation of clothing for commercial and milkweed nonwovens as a function of thickness. It may be observed that the thermal insulation is correlated to the thickness, as described by Equation (10). Accordingly, a specific thermal insulation value may be obtained by only adjusting the thickness of the MW/LM 2.2/PLA nonwovens.

$${\mathrm{I}}_{\mathrm{T}}=(0.195\times \mathrm{TH})-0.013$$

(10)

A similar relationship was reported in two studies conducted by Crews et al. and Kaufmman et al. [28,42].

Finally, in addition to the insulation they provide, milkweed nonwoven fabrics are also lighter than Thinsulate^® and Primaloft^®. For example, the mass per unit area of sample 3 is 63.7% lighter than Thinsulate with the same clo value (0.972). Consequently, adding milkweed fiber to nonwoven materials would strongly reduce the weight of the clothing. Therefore, replacing a percentage of PET with MW and PLA fibers does not result in a loss of insulation, as shown in Figure 5 and Table 2, Table 3 and Table 4.

3.3. Compression and Short-Term Compression Recovery

Table 5. Compression (C) and short-term compression recovery (N) of nonwovens.

Samples	C (%)	N (%)
1	54.6 ± 2.1	84.6 ± 4.2
2	47.5 ± 2.6	93.1± 2.1
3	52.6 ± 2.9	84.0 ± 3.4
4	49.0 ± 4.1	88.4 ± 3.6
5	48.4 ± 3.7	84.6 ± 3.1
6	58.7 ± 3.6	87.3 ± 2.8
7	48.8 ± 4.1	90.6 ± 2.5
Thinsulate®	59.5 ± 2.1	95.1± 2.4
Primaloft®	79.9 ± 1.9	99.8 ± 1.8

reports the different values of compression (C) and short-term compression recovery (N) of nonwovens obtained according to ASTM D6571-22 [34], whereas Figure 6 displays the compression values based on milkweed content.

Thinsulate^® and Primaloft^® show compression values of 59.5 ± 2.1% and 79.9 ± 1.9, respectively, and are more compressible than the MW/LM 2.2/PLA nonwovens, whatever the fiber content is. The values for milkweed-containing samples are similar, except for sample 6, which presents a better compression (58.7 ± 3.6%). One explanation for this result is the high LM 2.2 fiber content (40%). This result is further supported by investigating samples 4 and 6, which have the same percentage of milkweed fibers (50 wt%) but differ in the synthetic fiber ratio, with 10 wt% LM 2.2 fibers and 40 wt% PLA fibers for sample 4, and 40 wt% LM 2.2 fibers and 10 wt% PLA fibers for sample 6. Many research papers have reported compression values from 16% to 55% for polyester needle punched nonwoven fabric [43,44,45]. Furthermore, all milkweed-based nonwovens demonstrate good short-term compression recovery, with a minimum recovery value of 84%, which is higher than those found by Crews et al., who reported a value of 69% for a filling material only composed of milkweed fibers. Thinsulate^® and Primaloft^® exhibit a higher short-term compression recovery (95.1 ± 2.4 and 99.8 ± 1.8, respectively) [28].

3.4. Moisture Regain

Table 6. Moisture regains, porosity, and thermal conductivity before and after saturation.

Samples	MW (wt%)	Moisture Regain (%)	Porosity (%)	Average λ (mW/m·K) Before Saturation	Standard Deviation λ (mW/m·K)	Average λ After Water Saturation at a RH of 90% (mW/m·K)	Standard Deviation λ After Water Saturation at a RH of 90% (mW/m·K)
1	30	8.41 ± 0.51	98.95 ± 0.10	35.32	0.04	35.69	0.06
2	70	33.35 ± 0.18	98.16 ± 0.12	33.13	0.03	33.38	0.04
3	50	19.80 ± 0.47	98.63 ± 0.18	33.96	0.09	34.02	0.03
4	50	18.94 ± 1.59	97.76 ± 0.21	32.81	0.02	32.83	0.05
5	50	21.21 ± 2.33	98.05 ± 0.11	33.44	0.01	33.57	0.03
6	50	23.43 ± 0.36	98.19 ± 0.10	33.65	0.02	33.69	0.04
7	50	20.29 ± 0.36	98.59 ± 0.12	32.55	0.07	32.59	0.05
Thinsulate®	0	0.99 ± 0.10	97.58 ± 0.13	33.31	0.09	33.67	0.07
Primaloft®	0	3.32 ± 0.20	98.84 ± 0.10	34.14	0.02	34.17	0.04

and Figure 7 present the moisture regain, thermal conductivity before and after saturation at 90% RH, and the porosity of MW/LM 2.2/PLA nonwovens, Thinsulate^®, and Primaloft samples.

Milkweed-based nonwovens present higher moisture regain than Thinsulate^® and Primaloft^®. For example, sample 6, which contains 50 wt% milkweed, presents a moisture regain of 23.43 ± 0.36, whereas Thinsulate^® has a moisture regain of 0.99 ± 0.10. The moisture regain is proportional to the milkweed content. Authors have estimated the milkweed moisture regain at 11% [26], whereas L.M 2.2 and PLA fibers present a moisture regain of less than 2% [46]. Comparing the thermal conductivity before and after saturation, none of the milkweed, Thinsulate^®, or Primaloft^® samples show a thermal insulation loss. Since the samples are highly porous (≥98%), the amount of material that can absorb water is very low.

3.5. Air-Laid Efficiency

The nonwovens processed with air-laid Spike exhibit an overall yield rate of 90.3 ± 2%. The milkweed fibers were observed to be the only ones that exhibited a fiber loss, attributed to their volatile nature. There was about 10% of milkweed fiber loss. In comparison, Gharehaghaji and Davoodi [23] achieved a fiber loss rate of 50% when producing nonwovens made up of 90% Estabragh (a type of natural hollow fiber similar to milkweed) and 10% PET, using a roller carding machine. The large difference between the yield to the Air-laid process and carding is due to the type of chamber used: closed in the first one, with basically no fiber loss, and opened in the second, with big losses. Therefore, the air-laid Spike process may achieve approximately twice the efficiency of roller carding for processing nonwovens when using natural hollow fibers.

4. Conclusions

In this paper, milkweed nonwovens were produced using an air-laid process, characterized, and compared to two commercial products. The main points that can be drawn from this study are as follows:

Nonwovens that contain 50 and 70 wt% milkweed present a lower thermal conductivity than Thinsulate^® and Primaloft^®. For the same thermal insulation value (0.912), milkweed-containing nonwoven is 63.7% lighter than Thinsulate^®, which represents a very high gain of weight.

The compression decreased with the addition of milkweed. However, a nonwoven composed of 50% milkweed and 40% LM 2.2 has the same compression as Thinsulate^®.

Even if the moisture regain of MW/LM 2.2/PLA nonwovens at a relative humidity of 90% increased with the fiber content, this does not affect the thermal conductivity.

Because of its outstanding lightness and thermal insulation capacity, milkweed is a promising eco-friendly alternative to conventional synthetic materials. Therefore, developing products based on milkweed and bio-sourced polymer fibers appears to be a promising solution but further investigations must be conducted on the low surface functionalization and thermal properties of the fibers to optimize their potential.