Investigation of the Use of Mycelial Filler with Different Cultivation Times for the Filtration of Particulate Airflow

Abstract

Balanced and sustainable development has challenges in utilizing the best efficiency technologies, using new types of materials with reduced environmental impact, including composite types, reusable materials, and easily recyclable consumer adaptable products. One understudied biologically produced material is mycelium, the scientifically studied and improved cultivation of which produces an environmentally friendly material with unique properties with a wide range of applications. In this work, filtration fillers from mycelia of different cultivation periods and their abilities to filter airflow from solid particles were experimentally studied. Numerical modeling studied the interaction and trapping of particles in the flow with the surface of mycelium filters. The results of the research revealed a high airflow filtration efficiency of more than 91%, as well as differences and advantages in the properties and structure of mycelia of different growth periods, and the need for further study of this biomaterial.

1. Introduction

Mycelia, and fungi in general, could be a good alternative to common man-made materials and are regarded as biodegradable and a part of the circular/sustainable economy [1]. Mycelium materials can have different properties and have become promising as replacements for petrochemically produced polymeric materials or myco-leather for animal-based leather [2]. The fungal mycelium consists of hyphae, which are tubular filaments, usually with a diameter between 1 and 30 μm, with a wall reinforced with chitin. The diameter of hyphae depends on the species and growth environment. The length of hyphae can range from several microns to several meters. A mycelium itself is a biopolymer network, properties of which depend on individual filament behavior [3]. The properties of mycelial-developed composites (including their porosity) can also vary depending on their substrates [4,5,6,7]. Modifications of pure mycelium material have allowed for an even wider adaptation of their properties, adding, for example, hydrophobicity and elasticity [8,9], as well as thermodynamic and physical mechanical properties of biocomposites based on mycelium. A review is given by Girometta et al. [1], who mention that mycelium bio-composites demonstrate the same accuracy in the results of research as synthetic materials or monocomponent natural materials in relation to mechanical and thermodynamic parameters. Nevertheless, the mechanical characteristics of composites of mycelia can vary significantly and can be controlled by mycelium architecture, the composition of the cell wall, compositional components, and growth kinetics, which, in turn, are affected by inherent and exogenous factors [3]. Composites of mycelia and bio-stratum can be problematic due to their low mechanical properties, high water absorption, lack of information about the assessment of the life cycle, and the absence of standard production methods, as well as standardized methods for testing the properties of the material [4]. Elsacker et al. [10] especially mentioned that slower growth of mycelium makes the substrate prone to pollutants. Based on Elsacker et al. [11] and Shakir et al. [12], in the production of mycelian composites, mechanical properties depend on the types of fiber, on the growing types of substrates, and the duration of growth. The mechanical characteristics and growth of composites based on mycelia for future biomediation and the production of food in the production cycle are given by Houette et al. [13]. They mention that composite material density has good correlation with mechanical performance.

The functions of materials of mycelium-based biocomposites were explored in a review presented by Yang et al. [2]. They mentioned that biocomposites of mycelia are rich in chitin, which gives strength to the cell walls. Analysis of the mechanics of the mycelium material has been mentioned. The mechanics and morphology of fungal mycelia were analyzed by Islam et al. [5]. They mentioned that, when stretched, the reaction of the material is a linear elastic one. An analysis of mycelium-based particulate composite mechanics is given by Islam et al. [6]. They mentioned that composite properties are largely insensitive to the size of the particles for this volumetric fraction of the filler. Physical–mechanical and morphological properties of biocomposites of mycelia with natural particles for reinforcement were investigated by Gou et al. [7]. They mentioned that biocomposites of mycelium cultivated with P. ostreatus had the biggest compression strength, considering marine fungi and O. radicata. Mycelium gradient porous structures, taking into account an analysis of the quantitative structure of their mechanical properties, were determined by Olivero et al. [8]. They mentioned that a change of approximately 30 kPa in the mechanical module was observed at the 30 mm vertical height of mycelium fabric, and the oldest growth network had the highest elastic modulus. The mechanical possibilities of mycelium materials were investigated by Lelivelt [9]. He mentioned that wood, flax, sawdust, and hemp as substrates can give the best mechanical results. A study on the mechanical properties of a latex–mycelium composite was given by He et al. [14]. They mentioned that the mycelium was associated with latex, forming a double network, and an increased certain composite strength was expected.

Production factors affecting the mechanical, water, and moisture properties of composites based on mycelium were studied by Appels et al. [15]; they mentioned that the tensile strength and elasticity of heat-pressed materials were higher than corresponding non-pressed or cold-pressed materials.

In this work, we studied the effect of substrate composition and preparation on the growth of G. lucidum and P. ostreatus. According to our strategy, mycelial composition and morphology are two key factors that allow the final composites associated with mycelium to have significant properties. Barley effectively stimulates the growth of both G. lucidum and P. ostreatus under substrate conditions that are nonoptimized, and carbohydrate-rich feedings further promoted the growth of more mycelium than lipid-rich feedings such as those from bamboo and wood chips [16].

The search for and analysis of possible materials for air filtration used in face masks have been studied in various papers, such as Konda et al. [17]. The main focus is also on the use of widely available fabrics and the most effective degree of filtration.

The application of mycelium material is very wide [18,19,20]. We would like to mention that mycelium as a material was not used in air filtration in this study, as we encounter problems when we consider the possible implementation of this. We also want to emphasize that biological materials are becoming more popular at present, and the improved use of mycelial fillers as a more stable alternative to traditional inorganic surfaces of filters is becoming more profitable.

Scope of the study

The integration of new materials in a nonstandard application may show certain beneficial characteristics of the object, which represent advantages and may be an alternative to conventional types, in this case, filter media, which have obvious drawbacks in the broad sense of environmental impact assessment and technological performance. In this context, this paper analyzes mycelium aged between 3 and 5 weeks as a filter medium for filtering of polluted air at a velocity of 1 m/s or less and compares the numerical efficiency of the collection of fine particles compared to a synthetic filter.

Problem formulation

Air quality is essential for healthy human activity. Without the right conditions, there is a very high risk of both mild damage to external organs and severe and sometimes irreversible damage or chronic diseases.

The most common air pollutant is particulate matter, which can include harmful mixtures of organic and inorganic substances, adding to their adverse effects, in other words, synergistic effects. The smallest particles, especially those smaller than 2 µm, are more likely to reach the more distant respiratory tract and are therefore dangerous.

Efficient air purification filters are therefore essential in industrial environments, in manufacturing workplaces, including mobile workplaces with car and truck drivers and special vehicles [21], in offices, or in residential environments. However, application conditions, depending on the source of pollution, pollution parameters, airflow rate, etc., significantly limit the choice of suitable technologies.

The widespread use of filtering surfaces for particulate retention is currently confronted with material selection problems related to their environmental performance during production, including energy consumption, as well as to post-treatment of waste.

In view of the above, there is a strong demand for alternatives to conventional filtration materials that offer the best performance, are sustainable, are environmentally friendly, and require low raw material and energy consumption throughout their life cycle.

2. Materials and Methods

Fomes fomentarius (L.) Fr. (Polyporaceae) mycelium was used for the filters. This fungus is found on deciduous trees, especially birches (Betula spp.), on dead wood and on live trees. It is one of the most common polyporoid fungi in Lithuania [22]. Fomes fomentarius is commonly used as a material for the cultivation of mycelial products. Mycelium was obtained from fresh F. fomentarius fruiting bodies collected from dead wood of Betula trees. Mycelial culture and the mycelium for the filters were generally grown following the guidelines of Pohl et al. [23]. A pure culture was isolated from the fruiting body of Fomes fomentarius on agarized malt extract nutrient medium. The cultured mycelium was inoculated into rye grains, autoclaved at 121 °C for two hours. After the mycelium had completely overgrown the rye, the inoculated grains were mixed with a substrate of hemp shives with additives. The mixture was wrapped in fiberglass mesh. Cultivation was carried out in polypropylene bags with a 0.2 µm filter. Cultivation was carried out at 3, 4, and 5 weeks of age, respectively, at room temperature.

Distilled water (ISO 3696 Grade II) [24] for substrate preparation was prepared with an Adrona E30 distiller, the substrate was sterilized with a Raypa AES-28-DRY autoclave, and a Telstar Aeolus V laminar flow cabinet was used to inoculate the substrate with mycelium.

For the comparative data, a conventional material made of 100% polyester synthetic fibers by thermal bonding was used to filter the gas flow. The canvas is formed by a progressive method; the density of the material structure increases in the direction of air movement. This optimizes the distribution of suspended particles in the thickness of the roll filter, increasing the dust capturing capacity and service life. To prevent nonwoven fabric particles from entering the clean room area, the surface of the material on the outlet side of the airflow is calendared. The filter surface corresponds to type M5, according to EN 779.

The self-determined surface parameters of the mycelial filters and the declared performance of the synthetic filter are given in

Table 1. Physical parameters of the applied filtering surfaces.

Filter Base	Material	Density Porous/Nonporous, kg/m3	Open Porosity	Additional Features
Mycelium, 3 weeks	mycelium + fiberglass base	98.4/196.9	0.48	Visual transparency 40%
Mycelium, 4 weeks		49.4/98.7	0.49	Visual transparency 25%
Mycelium, 5 weeks		56.9/81.2	0.31	Visual transparency 10%
Synthetic material—comparative filter	polyester, 100%	25	0.88	Pressure drop limit 450 Pa; for PM10, not less than 50% efficiency

. The density was calculated by the mass and volume of the samples prepared as part of the filtration area. The porous density state was determined by calculating the mass-to-volume ratio of the uncompressed sample of material and the nonporous state was determined by calculating the mass-to-volume ratio of the compressed sample up to 10 N, respectively. The open porosity was calculated as the result of a difference of 1 with the ratio of the nonporous to the porous structure.

Two simultaneous multi-function Testo 440 and 440 dP meters connected with thermocouples Testo were used to determine the dynamic parameters of the gas flow. The airflow velocity has an accuracy of ±0.03 m/s + 4% of the detected value, a detection range of 0–50 m/s (for multi-function meters) and 0.01–30 m/s (for thermocouples), and a resolution of 0.01 m/s. Airflow temperature accuracy is ±0.5 °C, detection range is −20–+70 °C, and resolution is 0.1 °C. The accuracy of relative humidity of the airflow is ±3.0%RH (10 to 35%RH) and ±2.0%RH (35 to 65%RH), with a detection range of 5–95% and resolution of 0.1%RH. The accuracy of the static pressure of the airflow is within ±3 hPa, with a detection range of 700–1100 hPa and a resolution of 0.1 hPa. Particle concentration was analyzed using a Fluke 985 six-channel particle counter. The device was equipped with an isokinetic sampling probe (Figure 1a, position 4.1). The detection mode was numerical particle concentration. Six particle sizes, 0.3, 0.5, 1.0, 3.0, 5.0, and 10.0 (µm) were detected. The sample flow rate was 2.83 L/min. The sample flow rate was constant for all values of the airflow rate through the test sample. However, in order to maintain isokinetic conditions, tips were used to maintain a constant flow rate for sample collection.

Particle concentrations were determined by using Palas RGB 1000, site No. 3 (Figure 1). Particle sampling before purification was carried out at site No. 4 and after purification at site No. 5. Palas Welas Digital 3000 laser particle analyzer and detectors were used. The detection principle was optical light scattering. The detection range was 0.2 to 10 µm in 128 size channels and the concentration range was 1 particle/cm³ up to 10⁶ particles/cm³.

For the experimental studies, several stands were designed to investigate the aerodynamic parameters. After initial testing, one optimal case was selected to continue the filtration performance studies. The basic layout of the bench is shown in Figure 1.

Each study consisted of at least five replicate experiments to achieve a steady-state value under identical meteorological conditions. The experimental layout of the stand with control points is shown in Figure 1. Three adjustable inlet fans were used separately for different airflow ranges. The first fan (hereinafter referred to as air source 1) was recuperation system airflow generator Domekt S 650 F C5, with productivity of 642 m³/h, nominal static pressure of 210 Pa, nominal rotation speed of 1200 rot/min, and power of 63 W. The second fan (hereinafter referred to as air source 2) was System Air CE 200-4, with productivity of 1087 m³/h, nominal static pressure of 280 Pa, nominal rotation speed of 1480 rot/min, and power of 230 W. The third fan (hereinafter referred to air source 3) was Dospel Comfort WK250, with productivity of 1600 m³/h, nominal static pressure of 590 Pa, nominal rotation speed of 2660 rot/min, and power of 210 W.

In the first phase of the study, the rate of gas flow was studied on a complete three-part apparatus in a module. The airflow generated by the installed tubular fan was investigated and various capacities of the equipment were tested. The system was equipped with an additional section to balance the flow and deliver it to the inlet spigot. This was particularly important when investigating changes in velocity before and after the cleaning equipment, as different impact principles were used for solid particle deposition. Thus, depending on the size of the creation device and the volume of the purified airflow, it was important to choose the best gas supply rate from the source to the input of the device, under the condition that the theoretical speeds of pollution deposition were not exceeded. The difficulty was that, without such a study, all the purification stages, with different optimum operating intervals, will not give maximum results. The hypotheses proposed in the preliminary study were based on the theoretical validation of the operating principles of each cleaning phase, as confirmed by the study. Further studies can be extended by conducting cleaning tests under realistic conditions or by modelling the dust gas flow.

The purification process is effective when the flow is uniform and isokinetic, without significant pulsation. In order to investigate this process, a case study has been carried out in the previous and subsequent sections to match a gas flow. A flow rate study was carried out over the entire cross-section of the test site. Particulate filtration efficiency studies were carried out using glass particles. This type (glass particles) is chosen because of its inertness to the effects of both filters. In addition, this type of particles has a clear dispersive composition and is easy to apply in the dosing system, which is very important for initial experiments. Of course, further studies should be carried out on the different types of dust generated in the areas where such filters are installed in order to analyze in detail the properties of the mycelial filler for filtration.

The physical characteristics obtained from the analysis of 4 replicates and the light microscope photography are presented in

Table 2. Examples of the different filtering surfaces used in the tests—open, transverse, and embedded test cartridge.

Filtration Filler
Surface view, Visibility test, Mounted filler in the filter flange connection, Macro view with division length of 50 µm
Mycelium, 3-week-old cultivation
Mycelium, 4-week-old cultivation
Mycelium, 5-week-old cultivation
Synthetic polyester filter material

and

Table 3. Physical characterization and microscopic appearance of glass particles used in this study.

Parameter	Glass PM
Skeletal density of the sample, kg/m3	680–700
Density of the saturated sample, kg/m3	2300–2350
Total porosity	0.0016–0.0018
Overall density, kg/m3	2450–2500
Bulk density, kg/m3	1500–1550
10th percentile of particle diameter, μm	2.6–2.7
50th percentile of particle diameter, μm	9.1–9.3
90th percentile of particle diameter, μm	18.5–18.7
Mean diameter, μm	9.8–10.1
Microscopic view, magnification ×20, division length is 10 µm

.

Each study consisted of at least five replicate experiments to a steady-state value under identical meteorological conditions. The Pearson index was used to comparatively assess the correlations between the different gas flow parameters in different cases.

The aim of these experiments was to obtain the first full range of possible aerodynamic conditions and to identify any discrete and optimal cases for a uniform airflow distribution and a potentially better airflow and mycelium model for PM saturation in the installation.

3. Results

The research may involve several things:

(a)

To see the penetration of the particle into the mycelium;

(b)

To see how quickly the particle adheres to the surface of the mycelium;

(c)

To understand the forces and magnitude of the interaction between the particle and the mycelium;

(d)

How fast the particle is moving and the velocities of the particle during the interaction;

(e)

Providing an example of a theoretical model to analyze the motion of a particle during its interaction with a mycelium material. In addition, being able to analyze the adhesion process.

3.1. Results of Numerical Experiment

This paper presents a numerical experiment that examines the interaction of a pollutant particle with the surface of the mycelium (Figure 1). All values of initial parameters for the particle and mycelium surface are presented (

Table 4. Initial data considering previous research on mycelium as a filter material [24].

Objects	Initial Parameters	Values	References, Source
Glass particle (pollution)	Diameter, $d_i$	$1 \mathsf{\mu }\mathrm{m}$	-
	Radius, $R_i$	$0.5 \mathsf{\mu }\mathrm{m}$	-
	Initial interaction distance, $h_S$	$0.4 \mathrm{n}\mathrm{m}$	Chlebnikovas and Jasevičius [25]
	Initial normal velocity, ${\upsilon }_0^N$	0.7 m/s	-
	mass, $m_i$	≈$1.293 \mathrm{p}\mathrm{g}$ (picograms)	-
	Density, ${\mathrm{\rho }}_i$	≈$2376 \mathrm{k}\mathrm{g}/{\mathrm{m}}^3$	Chlebnikovas and Jasevičius [25,26]
	Young’s modulus, $E_i$	$80.1 \mathrm{G}\mathrm{P}\mathrm{a}$	Chlebnikovas and Jasevičius [25,26]
	Poisson’s ratio, ${\mathsf{\nu }}_i$	0.27	Chlebnikovas and Jasevičius [25,26]
Mycelium	Density, ${\rho }_j$	680 $\mathrm{g}/{\mathrm{c}\mathrm{m}}^3$	Vaišis et al. and Appels et al. [24,27]
	Young’s modulus, $E_j$	$550 \mathrm{M}\mathrm{P}\mathrm{a}$	Vaišis et al. and Appels et al. [24,27]
	Poisson ratio, ${\nu }_j$	$0.3$	Vaišis et al. and Appels et al. [24,27]
Simulation	Time step	$0.1 \mathrm{p}\mathrm{s}$ (picoseconds)	-

).

The force–displacement is presented in Figure 2a and the force, displacement, and velocity are presented in Figure 2b, Figure 2c, and Figure 2d, respectively.

After performing the numerical experiment, it is easy to see that the pollutant particle easily adhered to the mycelium surface. This was not only due to the acting force of adhesion but also to the softness of the mycelium material. By examining the force–displacement dependence, it was observed that the particle oscillations during the adhesion process decreased until the particle stopped, stuck to the surface of the mycelium at the corresponding displacement/penetration. Regarding the force, displacement, and velocity histories, it was observed that, after a time interval of 0.3 microseconds, the particle oscillations could be considered negligible and the particle could be considered stuck to the surface.

Softness of the mycelium makes it easy for particles to stick to the surface of the mycelium, so an air cleaning filter made from mycelium material has a promising future. However, with this behavior, where particles stick too easily, it was assumed that such a filter can quickly become clogged. Therefore, the mycelium as a filter could be used in rooms with low levels of contamination.

3.2. Results of Physical Experiment

In a first step, the aerodynamic parameters were determined in a system with a low airflow source 1 (Figure 1b) using a synthetic filter. The results obtained are presented in

Table 5. Aerodynamic performance of a synthetic material filter using an air source 1 with a small volume airflow.

Fan Lever, Hz (% of Nominal Flow Rate)	I (15%)	II (25%)	III (50%)	IV (75%)	V (85%)	VI (100%)
Air velocity after all ducts, point 9 (Figure 1a), m/s	0.50–0.51–0.56	0.54–0.56–0.57	0.57–0.59–0.61	0.58–0.63–0.64	0.60–0.64–0.66	0.61–0.64–0.66
Resistance/Static pressure upstream of filter, Pa	231–233	240–242	260–262	275–277	288–290	295–297

and

Table 6. Aerodynamic performance of a synthetic material filter using air source 2 low-efficiency air source.

Air Source Frequency, Hz (Nominal 50 Hz)	Air Flow Rate into the Duct Zone, m/s			Aerodynamic Resistance/Static Pressure before the Filter, Pa
	Just after the Fan, in an Open Duct, Point 6 (Figure 1a) Min-Avg1-Avg2-Max	Just after the Filter, Zone B, Point 7, Open Duct (Figure 1a)	after All Ducts, Zone C, Open Duct (Figure 1a, Point 9), Min-Vid.1-Vid.2-Max
50	11.7–12.5–12.7–12.8	0.73	0.60–0.63–0.65–0.66	288–290
45	9.2–9.7–10.2–10.07	0.42	0.38–0.39–0.41–0.42	175–177
40	8.63–9.27–9.88–10.07	0.42	0.38–0.39–0.41–0.42	175–177

. In the measurement area immediately upstream of the fan, with the duct open, the results of the minimum of the two average values and the maximum are presented. The data at point 9 (Figure 1a) are presented in a similar way.

Studies on the velocity and aerodynamic drag were conducted for a small-volume airflow source (air source 1) (axial fan) used in indoor heat–cool recovery systems, using a synthetic material filter. The results obtained are presented in Table 5.

Due to the significant aerodynamic losses when using a synthetic filter, a more efficient, powerful, and high-speed flow source (air source 2) (axial fan) was selected. A more detailed aerodynamic performance test was conducted with air source 2 operating only at the three highest frequencies, including the nominal frequency. The results obtained are summarized in Table 6.

These results show that the synthetic filter created a large pressure drop that reduced the air velocity by more than 19 times.

The structure of the mycelial filler is quite heterogeneous. To verify this, an aerodynamic test was carried out with air source 2 at nominal frequency. The tests were conducted using a 3-week-old filter (

Table 7. A 4-week-old mycelium filter structure non-uniformity test based on air velocity just after filter cartridge.

Measuring Zone	Point B1	Point B2	Point B3	Point B4	Point B5
Just after the filter, near the filter surface, Zone B	0.17	0.28	0.18	0.26	0.19
Just after the filter, open opening, Zone C	0.17–0.20	0.18–0.26	0.22–0.24	0.17–0.21	0.15–0.19

). With all ducts open at Zone C, point 10, the air velocity across the entire cross-section was 0.11–0.14 m/s and the aerodynamic drag was 304 Pa. The tests were repeated with air source 3 using 3-week-old (

Table 8. Variation in aerodynamic parameters in the bench areas with air source 3 and a 3-week-old mycelium filter.

Air Source Frequency/Flow Amount/Load (%)		10.02/270/17 19.98/540/33 25.00/670/42	30.06/800/50	35.04/930/58	40.02/1070/67	45.0/1200/75	50.0/1330/83	55.0/1470/92	60/1600/100
Just after the filter, open opening, Zone C		≤0.03 *	0.06	0.09	0.16	0.24	0.27	0.33	0.45
Just after the filter, near the filter surface, Zone B	B1	≤0.03 *	0.18–0.19	0.22	0.31	0.41	0.47	0.52	0.68–0.69
	B2		0.11	0.17	0.23	0.31	0.43	0.42	0.45
	B3		0.13	0.16	0.23	0.32	0.42	0.44	0.48
	B4		0.14	0.19	0.23	0.32	0.42	0.43	0.48
	B5		0.07	0.12	0.18–0.19	0.21	0.28	0.31	0.47
After all ducts, Zone C, open duct (Figure 1a, Point 9)		≤0.03 */≤0.03 *–0.04/0.05–0.08	0.14	0.20	0.24	0.30–0.31	0.32–0.35	0.40–0.41	0.44–0.46
Aerodynamic resistance, Pa		-/96/147	204	274	353	438	533	635	743

* a minimal detection value.

) and 4-week-old (

Table 9. Variation in aerodynamic parameters in the bench areas with air source 3 and the 4-week-old mycelium filter.

Air Source Frequency/Flow Amount/Load (%)		10.02/270/17 19.98/540/33 25.00/670/42	30.06/800/50	35.04/930/58	40.02/1070/67	45.0/1200/75	50.0/1330/83	55.0/1470/92	60/1600/100
Just after the filter, open opening, Zone C		≤0.03 *	0.05–0.08	0.11	0.16–0.18	0.20–0.21	0.25–0.26	0.30	0.35
Just after the filter, near the filter surface, Zone B	B1	≤0.03 *	0.24	0.32	0.38	0.39	0.47	0.64	0.24
	B2		0.17	0.20	0.23	0.28	0.32	0.39–0.44	0.17
	B3		0.17	0.19	0.23	0.31	0.31	0.34–0.35	0.17
	B4		0.16	0.18	0.23	0.31	0.31	0.46	0.16
	B5		0.16	0.18	0.23	0.28	0.37	0.36	0.16
After all ducts, Zone C, open duct (Figure 1a, Point 9)		≤0.03 *		0.17	0.21	0.24	0.27	0.32	0.35
Aerodynamic resistance, Pa		-	170	260	338	425	523	625	735

* a minimal detection value.

) mycelium filters.

The porosity of different mycelial filters can be determined experimentally by analyzing the change in the velocity of gas flow leaving the filter near its surface (Figure 3), depending on the location of the points in Figure 1a. The lowest velocities were observed in the case of the 5-week-old mycelium filter and the nonuniformity was only observed at point B3, where the velocities increased by a factor of 2. For the 4-week-old mycelium filter, the velocities varied in the range 0.18–0.43 m/s across the entire cross-section as the gas flow passed through. A slightly larger increase in velocity was observed at point B3 and the largest decrease in velocity was observed at point B4. Vertically (points B1 and B2), the difference in gas velocity values was less than 15%. The 3-week-old mycelium filter had an opposite change in the gas velocity values compared to the 4-week-old mycelium filter. However, the variation in the gas velocity along the axis was the most erratic of all the cases considered, ranging from 0.11 m/s to 0.58 m/s. These results demonstrate once again that the permeability of the filters due to the presence of natural pores is highly dependent on the type of mycelium. Although the variation in gas flow velocity values was quite large, there was a tendency for the shorter cultured samples (mycelia from 3 and 4 weeks of age) to be more permeable and for the latter samples to exhibit properties characteristic of an isotropic material.

Due to the lower resistance of the synthetic fiber filter and the use of air source 3 at 60 Hz, i.e., 100% load, the following airflow velocities were obtained: after the filter, open aperture (Figure 1a, point 9)—2.3 m/s; average value at points B1-B5 immediately after the filter, near the filter surface, zone B—3.3 m/s; after the ducts after the filter, open aperture (Figure 1a, point 10)—3.4 m/s. The aerodynamic drag was 596 Pa.

Due to the extremely high aerodynamic drag and in view of the results obtained with the relatively low density samples of the 3- and 4-week-old filters, the 5-week-old mycelial filter tests were not carried out for practical reasons. However, the use of 5-week-old mycelium as a filter is feasible in excessive high-pressure systems, which may be the subject of a future study.

The results of studies on the cleaning efficiency of the glass-particle-contaminated air stream, with the calculation of the cumulative numerical concentration of particles, using a filter made of synthetic material filter with a nominal air velocity of 2.7 m/s are shown in Figure 4.

The distribution of efficiency dependence on the particle fraction sufficiently varied, so nine independent sequential tests were performed. The airflow rate after the filter was 120 m³/h. In most cases, the efficiency of 1 μm, 2 μm, and 5 μm particles was in the range of 40–60%. For the smallest particle size of 0.3 μm, the filter retained particles with an efficiency of, at most, 10% and, for 0.5 μm, with an efficiency of about 20%. Particles of the largest size were not captured before or after the filter in some cases but, when present, the efficiency was at least 60%.

Higher filtration efficiencies for 1 µm and 2 µm particles compared to 5 µm particles occurred in two cases. In the first case, since the numerical concentration of particles was measured, the number of 5 µm particles was very low, at a maximum of 5000 particles per liter, so even an insignificant increase in particles after the filter dramatically reduced the efficiency. In the second case, it was assumed that particles were more likely to collide with each other when passing through sufficiently small pores in the mycelium, resulting in agglomeration, and that a decrease in the number of particles per liter after the filter results in an increase in the number of particles per liter of 5 µm.

After reducing the flow rate in the filter section to 35 m³/h, the tests were repeated and the efficiency of the synthetic filter was determined as a function of particle fraction. An increase in efficiency on average by 25–30% for particles of 2–5 microns was determined, but the efficiency of the fine fractions did not change or even decreased by 20%. Furthermore, at these speeds, the efficiency of 0.3-1 µm particles was not constant and, among tests, varied by up to ±10% (Figure 5).

Analysis of the multichannel particle fraction tests before and after the filter revealed a more detailed numerical concentration distribution, as well as the dependence of efficiency from ultrafine 0.2 µm to coarse 10 µm particle fractions. At an airflow rate of 50.2 m³/h, the highest concentration was found for particles of 0.2–0.5 µm—more than 20 thousand particles/liter each. The capture efficiency of the synthetic filter increased steadily for particles of up to 10 µm, while the capture efficiency for particles of 0.25 µm, 0.5 µm, 1.1 µm, and 4 µm varied slightly. At the same time, the average concentration of particles after the filter was not more than 32 thousand particles/liter, and particle sizes larger than 0.8 μm were estimated to be less than 1 thousand particles/liter. Particles larger than 5 μm were completely captured (Figure 6).

As mentioned above, the use of mycelium as a filter material in research is new. Therefore, the comparative trapping tests were replicated under as similar conditions as possible with mycelium of different cultivation times, namely 3, 4, and 5 weeks. As it was already found in the initial phase of the airflow dynamics study that the resistance of the 5-week-old mycelium was several times higher than that of the 3-week-old and the 4-week-old mycelium, it was decided not to include the oldest mycelium filter type in the particle trapping trials.

The cyclical changes in efficiency obtained from the study could be due to the fact that the flow velocities in the experimental bench were quite high compared to the size of the filter cartridge. The larger surface area of the filter object results in a more even distribution of particles. Even with a synthetic filter produced in the factory, small densities or, in contrast, porosity of the aggregate are possible, which, in a small area, has a significant influence on the results, especially when comparing before and after filter concentrations. The final result may also be influenced by the dispersive distribution of particles, including the dosing apparatus.

The distribution of particle concentrations before and after the 3-week-old mycelium filter and the dependence of the cleaning efficiency on the particle fraction at the maximum throughput of air source 3 of 48.9 m³/h is shown in Figure 7.

As already established in the aerodynamic tests of the mycelium, this structure was not homogeneous; therefore, both the distribution of airflow through the pores and the capture capacity could be nonuniform. Under relatively similar test conditions, with particle concentrations of about 10–25 thousand particles/liter, the capture efficiency did not increase progressively depending on the particle fraction. Peaks (upward spikes) in efficiency were observed at particle sizes of 0.2 μm, 0.35–0.6 μm, and 3.5–10 μm, with values exceeding 90%. After the filter, the particle number concentration was within 6000 particles/liter.

The porous structure of the mycelium has the greatest influence on its ability to collect particles. As observed in this work, mycelium aggregate becomes denser and “skeleton-like” with increasing cultivation time. For this reason, it was hypothesized that the structure has some particle skipping windows that fall within some particle size ranges, leading to a decrease in the efficiency of particle collection for particles of this size.

A more detailed analysis is also needed in the future to investigate how the position of these skip intervals changes as the airflow changes.

By halving the airflow source rpm at an airflow rate of 22.3 m³/h, the capture efficiency studies were repeated. The lower dilution resulted in an increase in particle concentration by a factor of approximately 2. Under these conditions, all fractions from 0.2 µm to 10 µm showed an increase in efficiency. The highest increase in efficiency was for the 0.6 µm particles, exceeding 97%. There were a few zones where the efficiency fluctuated slightly, probably due to the irregularities of the mycelium structure, i.e., a 3–5% reduction in particles of sizes between 2.2 and 5 µm. The highest quantitative concentration of particles after the filter is distributed only in the 0.2–0.4 µm particle zone, the other particles being less than 1 thousand/liter (Figure 8).

The growth characteristics of mycelium are very specific. For example, the density of the earliest mycelium at 3 weeks of age is higher than that at 4 weeks of age but then the density increases again. As shown by the aerodynamic resistance index, the 4-week-old mycelium is the least dense, i.e., its structure is more porous. Similarly, particle trapping studies were conducted to obtain comparative data between the two types of mycelium. In the first case, the results of a filter from a 4-week-old mycelium filter are presented for a nominal airflow source velocity at an airflow of 41.5 m³/h (Figure 9).

The results show that the capture efficiency of ultrafine fractions also decreases in the range of 0.22–0.3 μm. However, later on, with small jumps, up to the particle size of 0.9 µm, the capture efficiency increases rapidly. The greatest drop in efficiency is found at the point where the particle sizes are 0.41 µm and 0.84 µm. Starting at the particle size of 0.47 µm, the capture efficiency exceeds 90%. Particles larger than 1.1 µm are completely captured, with only slight variations at particle sizes of 7.2 µm, 8.4 µm, and 9.7 µm, where the efficiency decreases by 2–4%. In this study, the highest numerical concentrations were for 0.26 µm, 0.31 µm, and 0.35 µm particles with a capture rate of more than 27,000 particles/liter. Concentrations greater than 5000 particles/liter were found for particle sizes between 0.21 µm and 0.84 µm.

In order to obtain a complete picture of the distribution of the results, a study was carried out where the airflow was reduced to 23.5 m³/h using a 4-week-old mycelium filter (Figure 10).

The results confirmed the trend observed in this work, i.e., that the 4-week-old mycelium has better structural properties than its counterparts of different ages, in this case, 3 weeks. At low velocities, possibly due to lower turbulence, even the capture of fine particles increased on average by 7–12%, while the unevenness of the capture of all fractions decreased. With decreasing flow velocity, the concentration of particles in the 0.26–0.35 µm fraction increased up to 70,000 particles/liter, while the average number of particles in the 0.26–0.35 µm fraction decreased.

In an experimental study investigating the cleaning efficiency of an air stream contaminated with glass particles, the airflow was 100,000 m³. In order to evaluate the loading of the synthetic filter sample, the condition of the filter cartridge was visually assessed. No cracks or other damage was observed and particle accumulation was uneven (Figure 11).

In general, filter beds are characterized by three basic parameters: efficiency, flow resistance, and dust absorption. These parameters are closely related and should be assessed together. However, in this phase of the study, the main objective was to investigate the efficiency per se and the feasibility of using mycelium as a filtering material.

This natural test has shown that the mycelium material had no outstanding negative properties compared to the synthetic material in terms of dynamic resistance to gas flow and a relatively long filtration period without high adhesion. In this study, the mycelium was grown on a glass fiber construction mesh, and this modification can be adapted in the future, depending on the gas flow rate, the properties of the mycelium, and the duration of the growth, as the main purpose of the mesh was structural rigidity.

4. Discussion

Filtration structures designed to capture a wide range of particles must have a high hydraulic load to resist airflow and to block out the particles of pollutants that are present on the one hand. On the other hand, the filter filler material must be sufficiently porous and have a dense structure, acting as a particle trap, in order to ensure a high cleaning efficiency of air stream at different pollutant concentrations.

In this study, an initial stage of aerodynamic and efficiency tests with mycelium of different growth times was performed. Alternatively, this material can be an excellent replacement for already known filtering materials. Due to its unique nature, the mycelium can be adapted to specific conditions during cultivation, that is, technologically engineered to grow by adjusting to the conditions of the environment in which it is incubated.

The results of this study showed that the density of mycelium needs special attention if it is to be used as a filtering material. Due to its complex pore structure and the specificity of its shell, it has fairly high resistance to airflow, which may limit its use due to high energy consumption. However, once the required flow parameters for the cleaned air have been determined, the cross-section of the filter should be taken into account and the volume of the air to be cleaned by a factor of many times, provided that the flow resistance did not damage the surface structure of the mycelium. The unique internal structure of mycelium, with its numerous pores and fibrous tissues, provides additional opportunities for forming barriers to trap pollutants.

The main objective of this work was to investigate the efficacy of mycelium as a filtering material per se and its feasibility. Therefore, studies on instantaneous concentrations have been carried out at this point but not on the continuous filtration process, including dust absorption. We would also like to add that a short lifetime test was performed with an airflow rate of 100,000 m³, where no significant increase in filter resistance was observed. Thus, in future studies, we will definitely extend the range of the airflow rate to determine the durability/clogging and the lifetime of the filter and, at the same time, analyze the pressure drop, i.e., aerodynamic resistance of mycelial filler.

The mycelium cultivation process will be automated as much as possible to speed up cultivation and increase the yield of the final product, while human error will be eliminated as much as possible to reduce the risk of contamination. In addition, energy consumption will be significantly reduced during the process, as the system will operate in semi-autonomous mode and, according to Sheridan’s classification, the action will be performed automatically, with the human being obliged to be informed afterwards or only informed afterwards at his request. In addition, compared to current conceptual models, the cost of producing the mycelium filler will be reduced and the material will be much more environmentally friendly.

This topic should be further developed by considering the possible types of mycelial structure and the age of mycelial growth. It is very likely that the results already achieved in this work could be improved by using alternative types of mycelium. The quality characteristics of the mycelial filters will be the same as those of the currently used filter fillers, i.e., throughput, filtration efficiency, and filter uptime without replacement. However, less stringent standards may be required to integrate such filters. However, the advantages of a mycelium filter in terms to balanced development will be unrivalled in the marketplace. In order to obtain characteristics close to the desired product, it is necessary to control the following mycelium growth parameters: humidity, temperature, sufficient fresh air supply, and, for some species, the necessary amount of light. Optimizing all these parameters will help to control the mycelial growth process in order to obtain the optimum product characteristics.

It is possible that the initial treatment of the structure will help to improve the technical parameters and broaden the scope of application of this material in the role of a filter and, subsequently, adapt it for finer cleaning and/or adapt it for the treatment of other pollutants.

Key finding

The time required for mycelial growth changed the internal structure of the resulting tissue, including porosity and density, resulting in reduced permeability and a static pressure drop as the gas flow moves through this type of filler.

A numerical experiment to assess the soft mycelial properties and the adhesion effect has shown that the contaminant particles readily adhere to its surface.

Over a growth period of 3 to 5 weeks, the mycelium became denser and harder and its heterogeneity increased.

The higher porosity mycelium trapped particulate matter with an efficiency of at least 91.7% at an optimum gas flow of 23.5 m³/h.

A filter with mycelial filler should be able to mechanically withstand the dynamic resistance of the gas flow without damage at a relative material load per unit of gas flow of up to 3 cm²/m³.

5. Conclusions

This work has shown that mycelial filler, as a more sustainable alternative to conventional inorganic filter surfaces, can be used to effectively trap particulate matter, thus filtering the gas flow. In a relatively short time (3–5 weeks), cultivation of mycelium without the need for very stringent conditions can provide a mycelial filler with the required properties on demand and a cartridge with a mycelium filter can be purposefully installed to clean the gas stream.

Using a 3-week-old mycelium filler at a fan-generated volumetric airflow rate of 1600 m³/h, due to an aerodynamic drag of 743 Pa, the airflow velocity just after the filter filler was 0.45 m/s. The longer cultivation time (4 weeks) for the filler resulted in a mycelium density approximately two times lower. In this case, the aerodynamic resistance of such a filter was 735 Pa at a flow velocity after the filter of 0.35 m/s at the same generated volumetric flow rate. At a volumetric flow rate of 22–24 m³/h through the filter, glass particles between 0.505 µm and 10 µm were captured by the mycelium filter with an efficiency of more than 95%. The highest efficiency was found with the 4-week-old mycelium but the alternative, the 3-week-old mycelium filler, also had a relatively high filtration efficiency.

Studies have shown that special attention should be paid to the choice of the type of mycelium and duration of cultivation, as these factors can significantly alter the performance of mycelium as a filter medium. The advantage of mycelium as a natural material is that it does not require complex handling in production, and the decomposition of this organic material produces neutral or even positive products that are returned to the nutrient chain. Adapting and/or modifying the cultivation process can produce the desired product size, improve the quality of the material, and, with additional treatments such as impregnation, significantly expand the application possibilities, but further scientific evaluation is needed to substantiate these additional perspectives of mycelium.