Sage Journals: Discover world-class research

Abstract

The development of efficient fiber-based air filter media with increased dust storage capacity and reduced pressure gradient is a key to the advancement of filtration systems in residential ventilation. However, the manufacturing of homogeneous deep-air filter media remains challenging due to the complexity of fiber formation and the predominantly unregulated deposition of fibers in nonwoven structures. This study aims to address these challenges by utilizing continuous crimped islands-in-the-sea fibers for highly porous and efficient filter textile applications, which can be integrated into ventilation system components such as heat exchangers. To define the required filter parameters for the production of highly porous and efficient filter textiles, a mathematical model to simulate the filter characteristics of separation efficiency and pressure drop was derived based on relevant filter theories and evaluated using computational fluid dynamics simulations combined with discrete element method and current literature. The microfibers, essential for the project’s objectives, were produced using the islands-in-the-sea melt spinning technique for polyamide 12, incorporating a water-soluble polyvinyl alcohol as the sea component. The resulting microfibers had diameters of 803 ± 83 nm and exhibited a technical strength of 61.25 ± 1.83 cN/tex. The microfibers were subsequently crimped by two texturing processes: bulked continuous filament and false twist texturing. The force–elongation curves of the textured yarns were analyzed to assess the fiber crimp structure and its suitability for air filtration media. The mathematical model describing fiber crimp and curvature was validated by using scanning electron microscopy. The findings indicate that the ratio of the single islands-in-the-sea fiber bending radius to the microfiber diameter is a critical parameter for producing highly porous structures. It was observed that the false twist texturing process resulted in a threefold crimping contraction of 32.4 ± 4.0%, in comparison with the bulked continuous filament process. Furthermore, it is necessary to analyze both crimp contraction and fiber curvature in order to evaluate the potential of crimped yarns in the development of highly porous and efficient filter textiles. The proposed approach offers promising applications in air filtration, enhancing system efficiency and longevity.

Keywords

Islands-in-the-sea melt spinning PA12 microfibers crimped microfibers porosity crimp measurement DTY texturing filter textile applications

The transition from fossil fuels to renewable energy sources is an essential part of mitigating climate change and ensuring sustainable energy use. In Germany, the world’s third-largest economy, the building sector (responsible for approximately 40% of CO₂ emissions) plays a crucial role. Consequently, reducing heat losses is needed for enhancing energy efficiency and promoting climate protection.^1,2 The implementation of controlled ventilation systems with heat recovery has been demonstrated to reduce avoidable heat losses by up to 90%,³ while concurrently enhancing indoor air quality by continuously removing pollutants. However, despite their evident benefits, the adoption of such systems is hindered by concerns regarding installation costs, space requirements, and noise emissions.

The role of air filtration in ventilation systems is of particular significance, given its effect on both system performance and energy efficiency. The demand for energy-efficient and separation-efficient air filter media has been increasing steadily, a trend that was already evident prior to the recent coronavirus pandemic. Conventional air filters primarily use fiber nonwovens, but their irregular fiber deposition results in inhomogeneous structures that limit filtration efficiency and increase pressure drop. Therefore, there has been a significant amount of research and publication in the field of air filtration on fiber air filters with ultrafine fiber diameters, increased dust holding capacity, and low pressure drop.^4–13 In addition, the geometry of the air filter media is usually optimized to enhance energy efficiency, reduce pressure drop, and improve dust storage capacity. Increasing the surface area reduces local flow velocities, thereby reducing pressure loss and allowing deeper particle penetration. As a result, pleated fiber air filters are widely used in a variety of applications, including air filtration in pharmaceutical and microelectronics clean rooms, gas turbine intake systems, and residential ventilation systems.^14–16

In recent years, there has been a wide variety of materials and manufacturing methods developed to improve filtration efficiency, including staple fiber, melt-blown fiber, and electrospun nanofiber filtration media.^4–6,17 While electrospinning provides defined ultrafine fibers and high surface area, it suffers from scalability limitations and high production costs.^7,13 In contrast, melt blown technology, which is widely used for air filtration applications, offers a cost-effective approach. However, this method results in nonuniform fiber distribution and limited fiber crimping, which reduces porosity and dust storage capacity.^8,9

In order to overcome these challenges, an alternative approach using crimped islands-in-the-sea microfibers is discussed in this study. In addition to the previously mentioned techniques, islands-in-the-sea spinning allows the production of fibers in the lower nanofiber range down to 100 nm,^18,19 but with an aligned structure and a constant fiber diameter distribution. The formation of highly porous and homogeneous textile structures through defined crimping of these fibers can increase dust storage capacity, reduce pressure drop, and improve the overall filtration characteristics of textile filters.

Despite the significant research on fiber characterization and filtration technologies, there is a lack of studies investigating the use of crimped islands-in-the-sea microfibers for air filtration applications. While previous studies have primarily focused on fiber morphology or mechanical properties, the potential of crimped islands-in-the-sea microfibers to form efficient filter textiles remains unexplored. In addition, the effect of fiber crimping on crucial filtration parameters such as separation efficiency, pressure drop, and dust storage capacity remains to be examined to a limited extent.¹²

Addressing these gaps, the present study aims to contribute to the development and characterization of crimped islands-in-the-sea microfibers for applications as highly porous filter textiles, with a particular focus on integration into air-to-air plate heat exchangers in controlled ventilation systems. This integration offers a dual advantage: enhancing air filtration performance while significantly reducing the space required for heat exchangers and filters. Consequently, ventilation units become more compact, quieter, and efficient. However, integrating a fiber filter media into a heat exchanger introduces additional specific challenges: it is required to maintain high filtration efficiency^20,21 while keeping the pressure drop across the filter media to a minimum to avoid significant increases in power consumption and noise emissions.²⁰ Whereas conventional filter media allow surface area expansion to reduce pressure drop, fiber filter materials integrated into heat exchangers must conform to predefined geometric constraints. Therefore, the air filtration media must meet specific requirements determined by the air filtration properties and the geometric dimensions of the heat exchanger, such as the available free flow cross sections, as well as the aerodynamic properties within the heat exchanger while ensuring the thermodynamic requirements of heat transfer.

The potential applications of these advanced filtration textiles extend beyond residential ventilation systems. The application of similar filtration principles can be used in industrial air purification, automotive cabin filters, medical filtration (such as respiratory masks), and heating, ventilation, and air conditioning (HVAC) systems in commercial buildings. Further validation of these applications through performance comparisons and experimental testing would serve to strengthen the practical significance of this research. By addressing existing scientific gaps, this study contributes to the advancement of energy-efficient air filtration technologies, supporting both environmental sustainability and improved indoor air quality.

Experimental details

Materials

The sea component used is Mowiflex C600 (T_m 165°C), a water-soluble polyvinyl alcohol (PVA) from Kuraray Europe GmbH, Hattersheim am Main, Germany. The islands component consists of either VESTAMID L1600 or VESTAMID L1700, both polyamide 12 (PA12) grades from Evonik Operations GmbH, Marl, Germany with a T_m of 178°C. The selection of this material combination is based on an overlapping processing window in the range of 190–220°C. With regard to the maximum allowed processing temperature of 220°C for PVA, other technical polyamides could not be used.

Methods

Islands-in-the-sea melt spinning

The islands-in-the-sea microfiber yarns were developed in a pilot-scale study on a melt spinning plant for technical multifilament yarns from Fourné Maschinenbau GmbH, Alfter-Impekoven, Germany. These yarns were spun using a 24/48-hole islands-in-the-sea spinneret with 37 islands each. Prior to spinning, PVA was dried in a vacuum oven at 60°C and PA12 in a dry air dryer at 70°C for at least 12 h. Spinning conditions were 160–220°C within the extrusion system, 185–200°C spinneret temperature, and 300–900 m/min winding speed.

Fiber characterization

Tensile properties of the yarns were determined with a Statimat 4U from Textechno, Mönchengladbach, Germany according to DIN EN ISO 2062. A sample number of 20 was used for each trial, with a sample length of 250 mm, a testing speed of 250 mm/min, and a pretension of 0.5 cN/tex. Scanning electron microscopy (SEM) was performed on an EVO 10 from Carl Zeiss Microscopy Deutschland GmbH, Oberkochen, Germany.

Microfiber texturing process

In order to create a durable and permanent three-dimensional crimp structure, the bulked-continuous-filament (BCF) process²² and the false twist texturing process^22–24 were used. To ensure the protection of the fine microfibers, the entire islands-in-the-sea yarn was textured. Subsequently, the PVA sea component was dissolved. The BCF texturing tests were performed on a Oerlikon BCF S8 plant at Oerlikon Neumag, Neumünster, Germany. The DTY texturing lab tests were performed on a false twist laboratory plant at Institute for Textile Technology (ITA) of RWTH Aachen University, Aachen, Germany.

The yarn/process parameters used within the above-mentioned test procedures are listed in Table 1.

Table 1.

Overall texturing parameters

Texturing parameters	BCF-TexturingLab tests	DTY-TexturingLab test
Number of filaments	96	24
Number of isles	3552	888
Titer	1305 dtex	320 \| 310 \| 235 dtex
Pre-Draw ratio	2.2	2.2
Number of trial samples	26^a	9^a
Texturing temperature	120–170°C	240–270°C
Draw ratio	1.38	1.13 \| 1.23 \| 1.38 \| 1.60
Texturing nozzle	1.25 mm	–
Type of lamella chamber^b	3.0/4.5 mm \| LoK	–
Outlet pipe	7 \| 8 \| 9 mm	–
d/y ratio	–	1.77

Six representative examples shown in this study for BCF and DTY Lab tests.

Lamella chamber with an extension from 3.0 to 4.5 mm or LoK (lamella chamber with holes).

Within the BCF process, the achieved crimp contraction can be influenced primarily by the texturing temperature and by the geometries of the lamella chamber and the subsequent plug chamber.^22,25,26 For this reason, both the texturing temperature and the geometry of the texturing unit were varied within the tests. The DTY process was employed to investigate the influence of varying degrees of stretching and texturing temperatures during the texturization process. The published literature indicates that higher degrees of stretching and texturing temperatures lead to an increase in crimp contraction.^23,27 In addition to the above-mentioned process-related variables, the potential influence of a higher molecular weight of the PA12 islands polymer is of interest. It has been observed that a higher molecular weight and the associated higher viscosity of the material enhance the shrinkage tendency of a yarn.^28,29 Therefore, it is necessary to investigate the higher crimp contraction induced by a possible higher shrinkage of the yarn. This will provide insight into the potential for further optimization of the islands-in-the-sea yarn for texturing and the production of high-volume filter material.

Figure 1 presents the resulting experimental matrix for both texturing trials. Tables 2 and 3 present the process parameter variations of the selected crimped samples evaluated in this study.

Figure 1.

Experimental matrix of the BCF (a) and DTY (b) texturing trials.

Table 2.

Selected BCF samples

Process parameter	Sample
Process parameter	V04	V08	V13	V16/2	V18/2	V19/2
Texturing temperature (°C)	140	130	150	160	150	140
Lamella chamber (mm)	3/4.5	LoK	3/4.5	3/4.5	3/4.5	3/4.5
Outlet pipe diameter (mm)	7	9	8	8	8	8

Table 3.

Selected DTY samples

Process parameter	Sample
Process parameter	V16	V17	V18	V20	V23	V24
PA12 grade	L1600	L1600	L1600	L1700	L1700	L1700
Texturing temperature (°C)	240	240	270	240	240	240
Draw ratio	1.23	1.38	1.60	1.38	1.23	1.38

Crimp characterization

In order to determine and analyze the crimp characteristics and crimp contraction of texturized yarns, measurements according to the standards DIN 53840-1 for DTY-Yarns and DIN 53840-2 for BCF-Yarns were carried out on a Texturmat ME+ from Textechno, Mönchengladbach, Germany for five samples of texturized yarns for each trial.

In addition to these measurements, an alternative measurement method using a fiber tensile test was also applied to ensure a uniform comparison between the two texturing processes based on the same measurement method (Figure 2). This method involves the evaluation of the force–elongation curve of crimped fibers or yarns and the determination of the so-called uncrimping length $Δ L$ and the uncrimping work $w_{crimp}$ (area between the curve and the extension of the linear elastic region up to the point of intersection with the x-axis). The uncrimping length is measured at a specific titer-related force value of 2.0 cN/tex, consistent with the methods outlined in DIN 53840. This allows for direct comparison with DIN measurements, since uncrimping length can be correlated to crimp contraction $C C$ .

Figure 2.

Force–elongation curve of a crimped yarn (a) and self-constructed fiber tensile testing machine (b).

By analyzing the entire force–elongation curve (Figure 2(a)), the uncrimping length $Δ L$ (at 2.0 cN/tex) and uncrimping work $w_{crimp}$ ^30–32 were determined. The determination of the theoretical curvature and the crimp geometry of the textured yarn is enabled by the uncrimping work. This determination is made using the theoretical approaches found in Geitel.³⁰ This analysis provides insights into the number of arcs per length and the dimensions of the arcs, including arc radius, arc length, and arc height.^30,32 To carry out these specific measurements, it is necessary to use a testing device that is capable of allowing a change in fiber/yarn length by a margin of >500 mm. Therefore, a self-constructed testing device (Figure 2(b)) was assembled that allows the measurement of the entire force–elongation curve of crimped fibers and yarns without the need for pretensioning. The test parameters were set to a clamping length of 250 mm without pretension and a testing speed of 1 mm/s, with a sampling rate of 100 ms, and a sample number of 10 for each trial.

Crimp angles were measured by evaluating 10 arcs using SEM images performed on an EVO 10 from Carl Zeiss Microscopy Deutschland GmbH, Oberkochen, Germany.

CFD-DEM simulation

Computational fluid dynamics (CFD) is a valuable tool for addressing complex fluid mechanics problems. In the context of analyzing the filtration characteristics of fiber filters, CFD in combination with the discrete-element method (DEM) is used to calculate the separation efficiency and pressure drop^33–35 by simulating the air flow, particle movements, and particle–fiber–fluid interactions using, for example, the lattice-Boltzmann approach.^36–40 In the present study the simulation software Direct Numerical Simulation laboratory (DNSlab) from IT for Engineering (it4e) GmbH, Kaiserslautern, Germany was used to validate the mathematical calculation model described in the supplementary material derived from relevant filter theories. A total of 50 simulations were performed using the software’s embedded actions of FluidFlowFD and FilterEfficiency for the purpose of validation.

Results and discussion

CFD-DEM simulations of filter media performance for validation of the derived mathematical models

The examined models demonstrated an average deviation in the area of the separation efficiency of 12.3 ± 3.3%. A total of 50 distinct simulations were conducted, with fiber diameters ranging from 0.7 to 3.8 µm and porosities ranging from 90.0% to 99.98%. Given the necessity for high computational resources and accurate fiber resolution, the flow lengths employed in this comparative analysis were maintained in a range of 0.1 to 0.25 mm. Figure 3(a)–(c) provides an illustration of the CFD-DEM simulation of a 3D fiber filter structure.

Figure 3.

CFD-DEM Filter simulation of a 3D fiber filter structure with $d_{F}$ = 2.0 µm, $ε$ = 95%, and $L$ = 0.1 mm (a), particle trajectories (b), and particle deposition (c). Comparison of different porosities with porosity of 99.0% (d), 99.7% (e) and 99.9% (f) with a layer thickness of 160 µm. Comparison of pressure drop gradient (g).

It is noted that the mathematical model used deviates significantly from the simulations in terms of pressure drop, with an average deviation of 40.0 ± 7.9%. This discrepancy can be primarily explained by the deviations in the area of very high porosities, which also represent limits for the simulation calculations. This is caused by the high-resolution accuracy of the individual fibers and the large domain size required to find a sufficient number of fibers in the computational volume under consideration and to obtain a simulation result that is independent of the model. This phenomenon is shown in Figure 3(d)–(f). These figures offer a visual comparison of porosities generated in DNSlab. As the porosity increases, the number of fibers present in the volume under consideration with a layer thickness of 160 µm and an edge length of 100 µm decreases significantly. Therefore, the required computational model must be significantly larger to obtain reliable results. The number of cells required to resolve a model with sufficient accuracy rapidly increases to several hundred million (up to over a billion). The resulting increase in computing power exceeds the capabilities of conventional desktop systems,⁴¹ necessitating the adoption of advanced computing infrastructures to ensure the reliability and efficacy of the modeling process. Consequently, analytical methodologies maintain their relevance in the calculation of theoretical filter characteristics.

As a result, more recent approaches to calculating the pressure drop of fiber filter media can be found in the current literature. Chaudhuri et al.⁴² developed an empirical correlation on the basis of numerical flow simulation for calculating the pressure gradient in fiber filter media. The correlation found by Chaudhuri et al.⁴² was validated using experimentally determined data from Abishek et al.⁴³ and Ehrhard et al.⁴⁴ Figure 3(g) presents a comparative analysis of experimental data from Ehrhard et al.⁴⁴ with the correlation found in Chaudhuri et al.,⁴² the empirical correlation established by Davies⁴⁵ and the semiempirical correlation developed by Förster,⁴⁶ the work on which the used simulations is based.

Figure 3(g) demonstrates that the semiempirical model developed by Förster and implemented in the present work can predict the pressure drop of fiber filter media with sufficient accuracy, even when compared with current studies and empirical correlations based on numerical flow simulations. Therefore, the results of the comparative analyses presented show that the models applied can be used to approximate the resulting collection efficiency and pressure drop of a fiber filter media.

Theoretical calculations of the required filtration characteristics of a fiber air filter media based on the derived mathematical models

The configuration of heat exchangers for residential buildings generally consists of a substantial number of parallel plates, forming flow channels with heights ranging from 2.0 to 3.0 mm. In advanced enthalpy exchangers, this distance can be reduced to approximately 0.5 mm, enabling moisture transfer via vapor pressure gradients and preventing condensate formation. The flow velocities within these microchannels range from 0.5 to 1.0 m/s, with hygienic air exchange rates between 120 and 140 m³/h. Consequently, the requested characteristics of the filter media are identified as follows:

minimum separation efficiency according to DIN EN 779, 35%;

average flow velocity, 0.75 m/s;

maximum initial pressure drop (obtain the initial pressure loss of a coarse filter media), ≪50 Pa;

minimum heat exchanger plate distance, 0.5 mm.

According to these required characteristics, the filtration properties can essentially be defined by the parameters fiber diameter, porosity/packing density, and flow length of the filter. It is obvious that the separation rate increases with decreasing fiber diameter and increasing packing density. However, a high packing density also leads to high flow resistance, resulting in a high pressure drop and a low dust holding capacity.

To meet the specific performance characteristics of a fine filter medium while achieving an initial pressure drop similar to that of a coarse filter, the fiber diameter, porosity, and flow length of the filter material can be varied. In consideration of the mathematical models derived in the present study, a parametric study was performed, which resulted in a parameter field within which the required filtration characteristics can be fulfilled. As illustrated in Figure 4(a) and (b), the resulting pressure drop (a) and flow length (b) fields are shown for the necessary combination of air filter parameters, fiber diameter, and packing density to achieve the minimum separation efficiency of 35% at an average inflow velocity of 0.75 m/s.

Figure 4.

3D plot of the pressure drop field (a) and flow length field (b).

By superimposing the resulting fields, a range of combinations of fiber diameter and packing density can be determined to produce the required air filter material, taking into account useful flow lengths and initial pressure drops. This approach allows for the direct elimination of combinations of packing density and fiber diameter that are either extremely low or would result in excessively high flow lengths, which are therefore not feasible from a process technology perspective.

The optimization study, which was based on a parametric analysis, resulted in the following range for filter parameters:

packing density ( $α$ ), 0.02 $\leq α \leq$ 0.06%;

porosity ( $ε = 1 - α$ ), 99.94 $\leq ε \leq$ 99.98%;

fiber diameter ( $d_{F}$ ), 0.5 $\leq d_{F} \leq$ 1.2 µm;

with resulting flow lengths and initial pressure drops of:

flow length ( $L$ ), 1.0 $\leq L \leq$ 15.0 mm;

pressure drop ( $Δ p$ ), 28 $\leq Δ p \leq$ 40 Pa.

Accordingly, a filter media with average fiber diameters of <1.0 μm and a porosity of >99.9% must be produced in order to meet the described requirements and be able to be integrated into the enthalpy exchanger. Therefore, the grammage of nonwovens intended for integration into the flow channels of enthalpy exchangers must be maintained below 0.1 g/m², ensuring the required porosity and requested filter characteristics. In contrast, conventional meltblown nonwovens typically range from 3 to 100 g/m².^11,17 On the other hand, electro- or (electro-)centrifugal spinning¹⁰ or newer approaches, such as electrospun electret nano-wool nonwovens,¹¹ have been shown to achieve grammage of 0.3 g/m² in the scientific field. However, these methods have been observed to exhibit a lack of porosity, with a maximum porosity of 98.7%.¹¹

Microfiber production

In this research work, an islands-in-the-sea multifilament yarn with individual PA12 fiber diameters of 803 ± 83 nm (see Figure 5(d) and (e)) was developed, which corresponds to the calculated minimum diameter of 1.0 µm.

Figure 5.

Mechanical properties of a technical microfiber yarn with 99.15 ± 0.2 dtex and 7104 filaments (a) and (b). SEM image of the developed islands-in-the-sea yarn (25% islands) (c). SEM images of PA12 microfiber yarn (d) and (e).

Depending on the spinning parameters the pure microfiber yarn (after continuous washing off the sea component in water) can reach a technical strength of 61.25 ± 1.83 cN/tex (Figure 5(a)), with an elongation of 41.2 ± 4.1% (Figure 5(b)), thus fulfilling most criteria for further textile processing.

The force–elongation curve presented in Figure 5(a) corresponds to the classic curve observed in a ductile material, exhibiting a rapid increase in linear elastic range up to the yield point, followed by a gradual rise in the plastic range up to the maximum tensile force and the point of complete failure of the specimen. The yarn reached a titer per filament of 0.014 dpf, which corresponds to a theoretical single fiber diameter of 1.33 µm for PA12 fibers. In order to achieve microfiber diameters of less than 1.0 µm, as illustrated in Figure 5(d) and (e), it was necessary to switch from low-viscosity PA12 grade L1600 to medium-viscosity PA12 grade L1700. This was done in order to maintain the required pressure build-up in the spinneret for a stable spinning process, despite a reduction of polymer mass throughput within the extrusion. The optimized and used yarns for texturing can be found in the section “Microfiber texturing process.”

Evaluation of texturing processes for microfiber crimping

The quantitative parameters of the crimp contraction, measured in accordance with DIN 53840, were evaluated and compared. The BCF texturing process was found to achieve significantly better results in terms of maximum crimp contraction (Figure 6).

Figure 6.

Crimp contraction according to DIN 53840-1/2 (DTY/BCF).

An analysis of variance (ANOVA) of the test series demonstrated that these outcomes are statistically significant with p = 0.0007, which is consequently well below the significance criterion of p < 0.05. In all test series, the BCF process was observed to achieve higher crimp contractions than was possible with DTY texturing. However, the measured results are questionable when considered in the context of the theoretically determined maximum crimp values of 36.3% for BCF and 69.5% for DTY using the described equations in the supplementary material, as well as the external appearance of the crimped yarns in Figure 7 (higher crimp contraction should also lead to a bulkier yarn).

Figure 7.

Comparison of BCF (a) and DTY (b) yarn.

The substantial discrepancy between the expected outcomes and the observed results suggests that the measurement procedure outlined in the DIN standard may not be a suitable method for evaluating texturized islands-in-the-sea yarns.

In the context of DIN measurement, the determination of the stretched length using a specific force of 2.0 cN/tex incorporates the nominal titer of the whole yarn. At the same time, it does not consider the islands-to-sea ratio, which influences the overall tenacity and maybe also the crimp stability. However, if the limit for the elongation of individual fibers of the yarn is surpassed when the yarn is loaded with 2.0 cN/tex, this may result in an underestimated crimp. As a consequence, the subsequently measured crimped length is no longer suitable for adequately describing the properties of the original yarn. While the initial examination suggested that the evaluation of the crimp contraction was a sufficient method for establishing a relationship between the increase in volume of the yarn and the porosity of the filter material, it was subsequently observed that the crimp contraction determined in accordance with the DIN standard merely represents a length ratio of crimped and stretched length. Consequently, the crimp contraction is no longer a sufficient criterion with regard to the increase in volume of the yarn or the potential porosity of the filter material. Therefore, it is essential to consider the crimp geometry in order to achieve a comprehensive understanding of the relationship between the volume of the yarn and the porosity of the filter material.

Due to the challenges associated with interpreting the texturizing results based on the measurement of the crimp characteristics in accordance with the DIN standard, an alternative measurement method was selected. As discussed in the section “Crimp characterization,” the crimp characteristics of textured yarns can be determined from the force–elongation curve using a fiber tensile testing device. This method provides a more precise result regarding the suitability of the texturized yarns for use as highly porous structures. The tensile testing of fibers can reveal significant disparities between BCF and DTY textured yarns, as evidenced by the divergence in force–elongation curves (Figures 8 and 9).

Figure 8.

Force–elongation curve crimped islands-the-sea yarn by BCF.

Figure 9.

Force–elongation curve crimped islands-the-sea yarn by DTY.

In this case, the DTY process results in significantly larger uncrimping lengths (Figure 9), which in turn leads to higher crimp contractions (Figure 10). The ANOVA test analysis of these results also shows a strong statistical significance with p = 0.0003, which is again well below the significance criterion of p < 0.05. The DTY tests V20, V23, and V24 were performed using L1700 with a higher molecular weight, and it was observed that the higher tendency to shrink is also reflected in a higher crimp contraction, which leads to a statistical significance of p = 0.03. This correlation was not revealed by the previously conducted tests in accordance with the DIN standard. The ANOVA of these tests led to p = 0.34, which was not statistically significant, and the 0-hypothesis could not be rejected (Figure 6). The analysis of the force–elongation curve of the textured yarns can also be used to illustrate the influence of a higher draw ratio during the DTY texturing tests. It has been observed that an elevated draw ratio tends to result in a greater degree of crimp contraction, a phenomenon that is particularly predominant in combination with a higher molecular weight. This observation is substantiated by the results obtained for V24 and V20, as shown in Figure 10. Nevertheless, it is important to note that further investigations are necessary to fully validate this relationship, as the ANOVA was unable to reject the 0-hypothesis based on the available measurements. The BCF texturing tests indicate that an increased texturing temperature leads to higher values in the crimp contraction, with p = 0.048, which is slightly under the significance criterion. Further studies are needed to investigate this in more depth. Considering the BCF results, the measurements also show the comparability to the measurement results according to DIN 53840-2. This could be attributed to the lower degree of crimping, which reduces the effect of crimp stability. Therefore, these results provide further validation of the measurement method. However, the BCF method generally results in significantly lower crimp contractions, with an order of magnitude factor of three lower than the DTY process.

Figure 10.

Crimp contraction: fiber tensile testing method.

As stated previously, crimp contraction alone is an insufficient indicator of textured yarn suitability for highly porous textile structures. This is due to the fact that knowledge of crimp contraction alone is limited in determining the specific fiber structure that has been achieved. The force–elongation curve of a crimped yarn can be used to ascertain the requisite uncrimping work and, consequently, the curvature. In addition, further geometric relationships of a theoretically ideal crimp can be used to derive the number of arcs and the geometric dimensions of the individual ideal arcs (crimp angle, arc length, arc height, arc radius) of the crimped yarn.³⁰ The theoretical curvature evaluation presented in Figure 11 illustrates the differences in the crimping structures of the two texturing processes. In BCF, a multitude of planar bends are generated. This structural element leads to a negligible increase in yarn volume, thereby limiting the potential for achieving the requisite high porosity in the filter material. In contrast, the DTY process results in the formation of a structure comprising arcs with a high degree of curvature, while exhibiting a relatively low arc number. The overarched curvatures achieved by the high crimping angles enable a substantial expansion in yarn volume, thereby enhancing the porosity of potential filter media. The optimal crimp configuration for the implementation of a crimped yarn as a highly porous filter medium is characterized by a high number of arcs with large crimp angles and reduced arc dimensions (arc radius, arc length, and arc height).

Figure 11.

Theoretical arc dimensions.

The crimp geometry applied by BCF is constrained to the dimensions of the yarn itself. As outlined in the supplementary material, the theoretical considerations indicate that the crimp angle by BCF can reach a maximum value of π in the case of an ideal composition of semicircular arcs.^30,32 In this case, the fiber surfaces are in contact, and further compression of the fibers is not possible. The arc dimensions are defined by the dimensions of the fiber. The use of islands-in-the-sea yarns during the texturing process restricts the maximum curvature of the embedded microfibers to the overall dimension of a single islands-in-the-sea fiber. Subsequent to the removal of the water-soluble PVA matrix, the curvature applied results in only a slight undulation of the microfibers, attributable to the substantially reduced microfiber diameters. As a result, the high porosity necessary for the filtration process cannot be achieved. The ratio of bending radius to microfiber diameter is of crucial importance (Figure 12). This limitation of the maximum angle of curvature is avoided by the crimping structure as helix curls in the DTY process, so higher and overarched crimp angles can be achieved.

Figure 12.

SEM images of BCF crimped islands-in-the-sea yarn (a) and related crimped PA12 microfibers (b) illustrating the crimp angles and visualized 2D crimp structure (c) of BCF V16-2 trial.

As illustrated in Figure 12 the SEM images present the crimp angles of the crimping structure achieved by BCF texturing. The observed values of 75 ± 13° are consistent with the theoretical calculations shown in Figure 11. To better understand the differences of the texturing processes Figure 13 provides a visual representation of the twisting of a multifilament yarn resulting from DTY texturing (sample number DTY V24), as well as the associated crimp structure, as observed by SEM. Figure 13(a) clearly illustrates the helical curls that have been created in the yarn. Figure 13(b) illustrates the textured microfibers after dissolution of the PVA component. A graphical evaluation determined a crimp angle of 128 ± 12°. The crimp angle shown in Figure 13 is approximately 137°, which is also in accordance with the calculated theoretical values for the crimp angle of the DTY yarns illustrated in Figure 11.

Figure 13.

DTY textured islands-in-the-sea yarn (a), textured PA12 microfibers illustrating the crimp angles (b), and visualized 2D crimp structure (c) of DTY V24 trial.

Through an analysis of the maximum achieved volume of the texturized islands-in-the-sea yarn, as shown in Figure 11 and supported by theoretical approaches outlined by Geitel,³⁰ it was calculated that the texturing results of the DTY V20 and DTY V24 tests could reach packing densities of approximately 0.3%.

A comparison with measured packing densities from conventional filter media (coarse to fine filter class) indicates that the densities of the crimped islands-in-the-sea yarns are already approximately two to three times lower than those of conventional nonwoven filter media, which have packing densities ranging from 0.8% to 8.5%. These values are consistent with literature data on packing densities of staple fiber nonwovens and glass fiber papers, which range from 0.5% to 6.0%.⁴

Conclusion and outlook

The present study demonstrated that texturing an islands-in-the-sea yarn induces a permanent crimp in the microfibers embedded within the lost matrix. With a tensile strength of up to 60 cN/tex, the resulting microfibers are suitable for further textile processing. In comparison with commercial filter media, the nonwashed crimped islands-in-the-sea yarn exhibited a two to three times lower packing density, indicating its potential for highly porous textile structures. Maintaining this bulkiness in the crimped microfibers after washing out the sea component would further reduce the packing density. Subsequent studies will focus on this process.

Among the investigated processes, the DTY texturing process showed the highest potential for optimizing the production of highly porous filter materials. Within this study, approximately 50% of the theoretical maximum crimp contraction (69.5%) was achieved. Statistical analysis confirmed the significance of the DTY results, particularly for higher-molecular-weight PA12. Further optimization of the polymeric yarn components and the DTY process could enhance microfiber crimping, thereby increasing yarn porosity to support the development of highly porous filter textiles. CFD-DEM simulations validated theoretical predictions with minor deviations due to computational limitations. However, evaluating crimp contraction alone is insufficient to determine the suitability of a textured yarn for filter applications. This study demonstrates that a comprehensive evaluation requires a force–elongation analysis and an uncrimping work evaluation in combination with crimp contraction. Future research should focus on improving the DTY texturing process to achieve an optimal crimp structure for porous filter materials. In addition, the development of a scalable manufacturing process for textured fiber materials remains challenging, particularly for their integration into heat exchangers. Addressing these challenges through material and process optimization will be crucial for the successful application of highly porous filter textiles.

Footnotes

Acknowledgments

We thank Dr.-Ing. Kirsten Prehn from Oerlikon Neumag and Mathias Ortega, M.Sc. from ITA for performing the texturing tests.

Declaration of conflicting interests

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

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The presented results were all generated during the HaLo-Filter project. This research was funded by Federal Ministry for Economic Affairs and Climate Action (BMWK) via Project Management Jülich (PTJ) as project management organization (grant number 03ET1655A/B), on the basis of a resolution of the German Bundestag. The APC was funded by the State and University Library Bremen (SuUB).

ORCID iDs

Lukas Herrmann

Lars Bostan

Supplementary material

Supplemental material for this article is available online.

References

Federal Ministry for the Environment, Nature Conservation, Building and Nuclear Safetyet al. Climate Action Plan 2050 – Principles and goals of the German government’s climate policy. https://www.bmuv.de/publikation/climate-action-plan-2050-de

Rheude

Röder

Estimating the use of materials and their GHG emissions in the German building sector. Clean Environ Syst 2022; 7: 100095.

Köck

Köntopp

eBulletin: List für Wohnungslüftungsgeräte mit und ohne Wärmerückgewinnung, www.tzwl.de (2023, accessed 5 January 2024).

Gail

Gommel

Reinraumtechnik. Berlin: Springer, 2018.

Kumar

(eds). Handbook of Fibrous Materials. Weinheim: Wiley-VCH, 2020.

Hutten

IM.

Handbook of Nonwoven Filter Media. 1st edn. Oxford: Butterworth-Heinemann, 2007.

Zhou

Liu

Zhang

, et al. Electrospun nanofiber membranes for air filtration: A review. Nanomaterials (Basel) 2022; 12. DOI: 10.3390/nano12071077

Drabek

Zatloukal

Meltblown technology for production of polymeric microfibers/nanofibers: A review. Phys Fluids 2019; 31. DOI: 10.1063/1.5116336

Hassan

Yeom

Wilkie

, et al. Fabrication of nanofiber meltblown membranes and their filtration properties. J Membrane Sci 2013; 427: 336–344.

10.

Desch

Reiter

Nanofaser-Spinnanlage. Tech Text 2010; 222. https://www.dienes.net/nanofaser-spinnanlage/ (2010, accessed 05 2025)

11.

Cao

Yin

, et al. Ultrafine, self-crimp, and electret nano-wool for low-resistance and high-efficiency protective filter media against PM0.3. J Colloid Interface Sci 2020; 578: 565–573.

12.

Zaarour

Zhu

Huang

, et al. A mini review on the generation of crimped ultrathin fibers via electrospinning: Materials, strategies, and applications. Polymers for Advanced Techs 2020; 31: 1449–1462.

13.

Russo

Castro-Muñoz

Santoro

, et al. A review on electrospun membranes for potential air filtration application. Journal of Environmental Chemical Engineering 2022; 10: 108452.

14.

Kim

Y-L

Kwon

M-S

Lee

M-H.

Optimized pleat geometry at specific pleat height in pleated filters for air purification. Indoor Air 2022; 32: e13135.

15.

Fraunhofer-Institut für Techno- und Wirtschaftsmathematik ITWM. Computer-aided optimization of filter pleats, www.itwm.fraunhofer.de (2018, accessed 6 November 2024).

16.

Chen

D-R

Pui

DYH

Liu

BYH.

Optimization of Pleated Filter Designs Using a Finite-Element Numerical Model. Aerosol Science and Technology 1995; 23: 579–590.

17.

Gries

Veit

Wulfhorst

Textile Technology. 2nd edn. München: Hanser, 2019.

18.

Hills Inc. Hills Nano-Technology, https://www.hillsinc.net/assets/pdfs/nano-fibers.pdf (accessed 20 December 2023).

19.

Teijin Frontier Co., Ltd. Ultra-fine Polyester Nanofibers Nanofront Nonwoven Sheet, https://www2.teijin-frontier.com/static/img_corp/exhibition_old_en/techtextile2019/pdf/03_nanofibersheet.pdf (accessed 20 December 2023).

20.

Sparks

Chase

Filters and Filtration Handbook. 6th edn. London: Butterworth-Heinemann/Elsevier, 2015.

21.

Association of German Engineers 6022/1. Ventilation and indoor-air quality - Hygiene requirements for ventilation and air-conditioning systems and units. VDI 6022 Part 1; 01.2018; ICS 91.140.30; Berlin: Beuth Verlag GmbH.

22.

Hearle

JWS

Hollick

Wilson

DK.

Yarn Texturing Technology. Boca Raton, FL: CRC Press/Woodhead, 2001.

23.

Atkinson

False Twist Textured Yarns: Principles, processes and applications. Amsterdam: Elsevier/Woodhead, 2012.

24.

Ortega

Saynisch

Yurtseven

B-M

, et al. A review on false-twist texturing. Fibers 2024; 12(4). DOI: 10.3390/fib12040036

25.

Fourné

Synthetic fibers: Machines and equipment, manufacture, properties. In Handbook for Plant Engineering, Machine Design, and Operation. München: Carl Hanser Verlag, 1999.

26.

Hearle

JW.

The 20th-century revolution in textile machines and processes. Part 2: Textured yarns and other technologies. Ind Archaeol Rev 2014; 36: 32–47.

27.

Veit

Fibers: History, Production, Properties, Market. Berlin: Springer, 2023.

28.

Trilinskaya

Egorov

Dromenko

BP.

Molecular weight and processing conditions of polyformaldehyde as factors in the shrinkage of the yarn. Fibre Chem 1979; 11: 38–40.

29.

Fujiyama

Kimura

Effect of molecular parameters on the shrinkage of injection-molded polypropylene. J Appl Polym Sci 1978; 22: 1225–1241.

30.

Geitel

Die Bogenzahl, eine notwendige Größe zur Beschreibung der Kräuselung?

Faserforsch Textiltech 1969: 259–266.

31.

Bauer-Kurz

Oxenham

Shiffler

DA.

The mechanism of crimp removal in synthetic staple fibers. Part I: Crimp geometry and the load–extension curve. Text Res J 2004; 74(4): 343–350.

32.

Bauer-Kurz

Fiber Crimp and Crimp Stability in Nonwoven Fabric Processes. Dissertation, North Carolina State University, 2000.

33.

Huang

Wang

Zhao

Numerical study of pressure drop and diffusional collection efficiency of several typical noncircular fibers in filtration. Powder Technol 2016; 292: 232–241.

34.

Gervais

P-C

Bémer

Bourrous

, et al. Airflow and particle transport simulations for predicting permeability and aerosol filtration efficiency in fibrous media. Chem Eng Sci 2017; 165: 154–164.

35.

Berry

Beckman

Cho

A comprehensive review of particle loading models of fibrous air filters. J Aerosol Sci 2023; 167: 106078.

36.

Yue

Zhang

Zhai

Numerical simulation of the filtration process in fibrous filters using CFD-DEM method. J Aerosol Sci 2016; 101: 174–187.

37.

Qian

Huang

, et al. CFD–DEM simulation of the filtration performance for fibrous media based on the mimic structure. Comput Chem Eng 2014; 71: 478–488.

38.

Wang

Zhao

Wang

, et al. Simulation of filtration process for multi-fiber filter using the lattice-Boltzmann two-phase flow model. J Aerosol Sci 2013; 66: 164–178.

39.

Przekop

Moskal

Gradoń

Lattice-Boltzmann approach for description of the structure of deposited particulate matter in fibrous filters. J Aerosol Sci 2003; 34(2): 133–147.

40.

Rao

Faghri

Computer modeling of aerosol filtration by fibrous filters. Aerosol Sci Technol 1988; 8: 133–156.

41.

Abishek

King

Schuler

, et al. Representative domain size for the simulation of coalescence filtration in nonwoven and foam media. Sep Purific Technol 2018; 207: 344–352.

42.

Chaudhuri

Boettcher

Ehrhard

Pressure drop in fibrous filter – Representative domain size and effect of fibre orientation. Chem Eng Sci 2021; 246: 116865.

43.

Abishek

King

Mead-Hunter

, et al. Generation and validation of virtual nonwoven, foam and knitted filter (separator/coalescer) geometries for CFD simulations. Sep Purific Technol 2017; 188: 493–507.

44.

Ehrhard

Mölter-Siemens

, van de Schoot

, et al. Abschlussbericht SEEDT – Steigerung der Energieeffizienz der Drucklufttechnik: Analyse von Zweiphasen-Prozessen in Filtrationsmedien. IGF–Project Nr. 19109 N, 2018.

45.

Davies

CN.

The separation of airborne dust and particles. In Proceedings of the Institution of Mechanical Engineers, 1953.

46.

Förster

Ein neues Zellenmodell zur Bestimmung von Abscheidegrad und Druckverlust der in der Klimatechnik verwendeten Filtermedien. Dissertation, GHS Essen, 1999.

Crimped islands-in-the-sea microfibers for highly porous filter textile applications

Abstract

Keywords

Experimental details

Materials

Methods

Islands-in-the-sea melt spinning

Fiber characterization

Microfiber texturing process

Crimp characterization

CFD-DEM simulation

Results and discussion

CFD-DEM simulations of filter media performance for validation of the derived mathematical models

Theoretical calculations of the required filtration characteristics of a fiber air filter media based on the derived mathematical models

Microfiber production

Evaluation of texturing processes for microfiber crimping

Conclusion and outlook

Footnotes

Acknowledgments

Declaration of conflicting interests

Funding

ORCID iDs

Supplementary material

References