Anti-paste Bag Performance of Water-repellent Filter Media in High-humidity Environments Considering Ultra-low Emissions

This study performed surface treatment of Flumex, polyimide, and membrane-covered aramid (chosen as test substrates) by the C8-1833 water-repellent agent. Five key parameters (C8-1833 concentration, impregnation time, rolling residue rate, baking temperature, and baking time) were selected to investigate their effects on the filter media’s water-repellent grade and air permeability variation rate. Through an orthogonal optimization test, the optimal water-repellent process parameters were determined. Air and dust filter media test beds were employed to assess the anti-paste bag performance of water-repellent and membrane-covered filter media under high-humidity conditions. The results indicate that the water-repellent grade of the filter media was unaffected by the baking time during complete drying. At water-repellent agent concentrations of no less than 30 g L –1 , impregnation times exceeding 3 s, baking temperatures exceeding 240 ° C, and rolling residue rates below 46%, both filter materials achieved a water-repellent grade 8. Under optimized processing conditions, the water-repellent grade of the filter media reached 8, and the contact angle was maximized, significantly enhancing the hydrophobicity. Both water-repellent and membrane-covered filter media delayed wet dust adhesion. However, in excessively high humidity of flue gas, sustained filtration performance of the filter media might not be achievable. In bag sticking or rapid condensation situations, the water-repellent filter material should be initially dried using high-temperature flue gas and subsequently restores its filtration performance through thorough dust cleaning.


INTRODUCTION
Significant particulate matter emissions inevitably accompany iron and steel industry production.To mitigate this problem, industrial dust removal via so-called LT + BF systems (combining dry dust removal with a bag filter) has been industrially implemented (Yang et al., 2022).During filtration, high-humidity clogging of the filter media results in bag plugging, affecting flue gas's normal filtration process.This phenomenon may cause various hazards, including reduced filtration efficiency, increased energy consumption, decreased equipment lifespan, escalated downtime and maintenance costs, and higher environmental pollution risks.To bolster the filter bag's suitability for the LT + BF system within high-humidity environments, ensuring sustained efficacy and reliability of ultra-low emissions transformation in the LT system's primary flue gas, this study attempted to refine the filter material's surface characteristics.High-temperature-resistant and water-repellent filter materials were developed through a water-repellent finishing process.Surface modification of solid materials stood as a pivotal realm, where practical application hinged largely on surface performance.

MATERIALS AND METHODS
Flumex (FMS) and polyimide (P84) were chosen as lucrative materials for testing waterproof process parameters due to their favorable characteristics, such as a continuous usage temperature ≥ 250°C, absence of static electricity accumulation, and strong mechanical properties (Kaneda et al., 1986).To investigate the anti-clogging bag performance of membrane materials in high-humidity filtration scenarios, a comparative analysis was conducted between the membrane-covered aramid and two variants of waterproof filter materials produced using the above optimal process parameters.The subsequent high-humidity condensation environment filtration and anti-clogging bag tests were carried out.The initial fundamental performance parameters of the three kinds of filter materials are presented in Table S1.

Selection of Water-repellent Agents
The thermal stability of fluorine-containing waterproofing agents such as TF-4105, AG-7500, TF-4116G, FG-910, and C8-1833 was evaluated using infrared spectroscopy.Considering agent economy and market share factors, C8-1833 (perfluoroalkyl acrylate ethyl) (Wang et al., 2010a) was ultimately chosen as the waterproofing agent for the high-temperature-resistant filtration material.The thermal stability of C8-1833 was tested, and the infrared spectra at room temperature and 250°C are illustrated in Fig. S1.
To ensure uniform soaking of the filter media with the C8-1833 water-repellent agent, the filmforming behavior of C8-1833 at various temperatures was examined.This investigation aimed to comprehend how the process temperature affects the flow characteristics and film-forming ability of C8-1833, consequently guiding the selection of the production process temperature for water-repellent treatment of the filter media.

Selection of Process Parameters for Water-repellent Finishing
The process of rendering filter materials water-repellent is termed impregnation, rolling, and baking within the realm of textile expertise.Typically, singed and calendared felt materials are soaked in a water-repellent agent.The procedural workflow predominantly encompasses impregnation, drying, heat setting, and packaging stages (Gao, 2007), outlined in Fig. S2.In the experimentation, we attempted to adhere to the manufacturer's recommendation of pre-drying the filter material at approximately 100°C for an extended duration (approximately 30 min).During the actual production process, the immersion period of the filtering material is relatively short, resulting in less liquid absorption.Therefore, this experiment employs a high-temperature one-step drying method.
The parameters selected for the study on water-repellent finishing (Cao and Cui, 2015) include impregnation temperature, impregnation time, finishing agent concentration, baking time, and baking temperature.In actual operating conditions, it is common to directly immerse filter media at room temperature.Therefore, the parameter of impregnation temperature is replaced with the rolling residue rate (proportion of liquid weight on fabric to fabric weight).The pivotal process parameters for filter materials' water-repellent finishing include finishing agent concentration, impregnation time, rolling residue rate, baking temperature, and baking time.The initial process parameters were established as follows: C8-1833 concentration at 30 g L -1 , impregnation time of 3 s, baking temperature at 250°C, baking time lasting 480 s, and rolling residue rate of 43%.When investigating the impact of individual factors, two values were selected separately from above and below the initial value for each factor.At the same time, the other parameters remained constant at their initial settings.An orthogonal test was designed for the corresponding experimental study, incorporating five factors and four levels.The orthogonal experimental design table is provided in the supplementary materials.

Air Filter Media Test Simulates High-humidity Condensation Filtration Conditions
The test bench and test principles for evaluating the filtration performance of air filter materials, in compliance with the GB/T 14295 air filtration test standard, are illustrated in Fig. S3(a).The air filtration test bed employed KCl aerosol as the test particles, with a particle size ranging from 0.25 to 5 µm.The particle concentration at the filter material inlet was adjusted by regulating the flow rate of the aerosol generator.The filtration velocity could also be adjusted by controlling the fan's rotational speed.The test bed was equipped with a pressure detection device to measure the filtration resistance of the filter material.Moreover, an aerosol concentration meter was employed to assess the filtration efficiency of the filter material.
The KCl solution concentration was increased to 20% to enhance the test effectiveness.A humidifier was installed in the air inlet pipe of the tested filter material to maintain continuous air humidification.The relative humidity inside the pipe was monitored using a relative humidity meter.Humidification continued until the relative humidity reached 100%, visually confirmed by a significant amount of fog formation and dew condensation within the pipe.These conditions signified the attainment of the initial test preparation conditions.
As per the GB/T 14295 air filtration test standard, the filtration velocity for filter media was regulated between 1.0 and 2.5 m s -1 .This range exceeded the filtration velocity specified for dust removal filter media (0.5-1.0 m min -1 ).Due to limitations in adjusting the lower limit of air volume within the test apparatus, the test air volume was fixed at 250 m 3 h -1 .The effective filtration area of the filter media measured 15 × 25 cm 2 , resulting in a filtration velocity of approximately 1.85 m s -1 .
The test procedure involved the following steps: once the air humidity within the pipeline aligned with the initial test preparation conditions, the filter material was installed vertically with the dust-facing surface positioned outward.The fan and aerosol generator were activated while the humidifier continued to operate, ensuring continuous humidification.The filtration efficiency and resistance of the filter material were evaluated continuously over an hour.The wetting status of the dust-facing and rear dust surfaces was monitored and recorded every 6 min.
Following the test, the filter material was removed and weighed, and its weight gain was measured before and after the test.This weight gain comparison allowed for assessing the water absorption characteristics of different filter materials.

Anti-paste Bag Test for Wet Dust Removal Filter Media
In this experiment, high-humidity environmental filtration was conducted using KCl aerosols.However, this simulation differed from actual dust removal processes and did not allow for the repeated filtration-clearing-filtration cycles that occurred in actual processes.Therefore, standard dust was selected for wet dust filtration on the test bench to make the bag-plugging phenomenon more pronounced.By comparing the relationship between the filtration resistance and dust load at different air velocities, the adhesion of dust on the surface of the filter media after multiple cycles of dust emission was observed.The static filtration test bench designed for assessing dust removal filter materials is presented in Fig. S3(b).The dust removal filter material was placed horizontally on the test bench with its dust-facing surface upward.The filtration wind speed of the filter material was regulated via a rotameter on the test bench, while the dust concentration at the filter material's front end was adjusted using a dust generator.This test involved chosen three sets of filtration wind speeds: 0.5, 0.75, and 1.0 m min -1 .The dust generator maintained a dust concentration of 10 mg m -3 , with a gram of dust being generated per instance and every five grams constituting a dust generation cycle.
The test procedure can be described as follows: the surface of the filter material was moistened with a fine spray, and then the complete filter material was laid horizontally on the test bench, ensuring that the dust-facing side was upward.As the dust generator emits, watering can be positioned 20 cm away from the test bench collector to apply water onto the dust.Each gram of dust should be humidified with water spray twice.Record the filter resistance of the filter material after each dust generation under different wind speeds.After five filtration cycles at a wind speed of 1 m min -1 , the filter material was extracted and dried in an oven set at 100°C.Then, the filter material was removed, and the dust adhered to its surface was delicately flicked.Using an earwashing ball, perform two back-flushes on the filter material from the back dust surface.The weight gain of the filter material was measured before and after the test, and the extent of dust adhesion on the filter material surface was documented.

Thermal Stability Test and Film-forming Characteristic Test of C8-1833 Water-repellent Agent
Observing Fig. S1, it can be deduced that at 250°C, the wavelength position (characteristic absorption peak) of infrared light transmittance of the C8-1833 agent remained consistent with that at room temperature.This implies that heating to 250°C did not alter the chemical structure of C8-1833, demonstrating its robust tolerance and stability at this temperature.
The film-forming characteristic test procedure for the water-repellent agent was as follows: 5 mL of C8-1833 water-repellent emulsion was placed into a 100 mL beaker within a constant temperature electric blast drying oven; then, it was dried for 10 min at different temperatures, monitoring the film-forming attributes of the emulsion at these varied temperatures.The film-forming behavior of the C8-1833 agent at different temperatures is outlined in Table S2.
From Table S2, it is seen that the C8-1833 agent fluidity diminished as the temperature rose.Above 130°C, the fluidity of the water-repellent agent started to weaken.Consequently, under normal conditions, C8-1833 exhibited effective dispersion, offering favorable flow and infiltration conditions for water-repellent treatment of filter media, requiring no other emulsifying additives.

Effects of Different Factors on the Water-repellent Grade and Air Permeability Variation Rate of Filter Media
During high-temperature baking, the water-repellent agent generates a low surface tension membrane on the treated fiber surface (Li et al., 2022).This leads to reduced fabric surface tension compared to water, subsequently altering the water-repellent characteristics of the filter material surface.After water-repellent treatment, the filter material weight increases by 0.5-1.0%,resulting in smaller pore sizes, decreased air permeability, and enhanced filtration efficiency (Johnston, 1999;Bälz et al., 2016).
Hence, alongside assessing the water-repellent grade of filter media, this study experimentally evaluated the alteration in air permeability of filter media following distinct finishing processes.Notably, the related studies on the waterproof finishing of filter materials mostly focused on testing a single filter material.In contrast, we subjected both filter materials under study to single-factor and orthogonal tests.The optimal process parameter range for water repellency was established via the comparative analysis of test results, providing a practical reference.

Finishing agent concentration effect
As shown in Fig. 1(a), the water-repellent grade of both filter media achieved the highest rating (grade 8) at a C8-1833 agent concentration of 30 g L -1 .This implied the ability of the C8-1833 waterrepellent agent to confer effective water-repellency to filter media even at lower concentrations.Analyzing Fig. 1(b), it becomes evident that the variation rate in air permeability for the two filter materials escalated with increasing concentration.This observation underscored that an elevated C8-1833 concentration led to the filter materials' progressive air permeability drop.Moreover, FMS's air permeability variation rate surpassed that of P84 as the C8-1833 concentration rose.Judging from the coefficient of variation (CV) of the air permeability variation rate between the two filter media, the air permeability of FMS was more sensitive to changes in the C8-1833 concentration than that of P84.
As the concentration increased, the CV value of FMS continuously rose, peaking at 20 g L -1 (9.86%) and remaining below 6.5% for concentrations ≥ 30 g L -1 .However, P84's air permeability variation was stable, with FMS's CV values exceeding those of P84 as the water-repellent agent concentration grew.The test results suggested a close correlation between the water repellency of the filter media and the organic fluorine content on the fiber surface.Elevated finishing agent concentrations resulted in more effective water-repellent agent components on the fiber surface.Enhanced film-forming performance of the finishing agent on the fiber surface led to higher water-repellent grades and more pronounced pore size reductions (Wang et al., 2010b;Balu et al., 2008).Hydrophobic treatment enhanced the filter material's ability to capture fine particles (Shim et al., 2023).

Impregnation time effect
As depicted in Fig. 1(c), an impregnation time of 3 s yielded the highest water-repellent grade (grade 8) for both filter materials.Corroborated by the production process diagram in Fig. S2, a 3 s impregnation time was an essential requirement for the uninterrupted production of filter materials.This underscored the stringent need for continuous production while affirming the exceptional waterproof performance of the chosen C8-1833 water-repellent finishing agent.
According to Fig. 1(d), the overall trend involved an increasing variation rate of air permeability for FMS, whereas that of P84 slightly fluctuated by approximately 14%.With extended impregnation time, the FMS air permeability variation rate consistently exceeded that of P84.This implied that the interaction between the surface fibers of FMS and the finishing agent was more thorough, resulting in enhanced film-forming effects, reduced porosity, and more pronounced changes in air permeability.
The performed analysis of CV variation revealed a continuously increased dispersion of FMS's air permeability variation rate at 1-3 s, followed by a subsequent drop.This phenomenon can be attributed to prolonged impregnation periods leading to heightened water-repellent factors on the filter media surface.This subsequently promoted the creation a more uniformly distributed water-repellent film and a consistent surface pore size.Consequently, the air permeability variation rate fluctuated less, and the dispersion degree decreased.Conversely, the variation rate of air permeability of P84 showed comparatively lower variability and higher stability.

Baking temperature effect
As illustrated in Fig. 1(e), the baking temperature profoundly impacted the finishing outcome.At 230°C, the water-repellent grade of both filter materials amounted to grade 7, implying an insufficient cross-linking between the C8-1833 finishing agent and the fiber at lower baking temperatures, resulting in inadequate film-forming properties.As the baking temperature grews, the interaction between the finishing agent and fiber became more active, culminating in the filter materials' highest water-repellent grade (grade 8).The concentration of the water-repellent agent and the baking temperature were found to be the key most significant factors controlling the hydrophobic rating of the filter material (Cusola et al., 2013).According to Fig. 1(f), an overall declining trend characterized the variation rate of air permeability for FMS with increasing baking temperatures.This trend can be attributed to elevated baking temperatures fostering earlier cross-linking of organic fluorine on the filter fiber surface, resulting in film formation.This, in turn, boosted the evaporation potential of residual moisture within the fibers, amplifying the disruptive impact on organic fluorine films and ultimately enhancing the air permeability of filter media.
Nonetheless, the variation rate of air permeability for P84 remained close to 14% overall, while FMS consistently surpassed P84 as the baking temperature increased.Assessing the CV value change revealed that the variation rate of air permeability for FMS was less influenced by baking temperature, resulting in diminished dispersion and overall stability.

Baking time effect
As seen in Fig. 1(g), both filter materials achieved their highest water-repellent grade (grade 8) at 420 s.This duration aligned with the time needed for drying the filter materials, showcasing that the water-repellent grades remained relatively unaffected by baking time when the filter materials were fully dried.As illustrated in Fig. 1(h), the fluctuation range of the air permeability variation rate for FMS was marginally broader than that of P84 (about 14%).The FMS air permeability variation rate consistently exceeded that of P84 as the baking time increased.This was because the chosen FMS had a CV value of 8.54 for the basis weight, while P84's CV value was 1.57.Due to the increased baking time, the FMS mass distribution per unit area became more uneven, forming larger localized pores within the filter material.As a result, the CV value of the air permeability variation rate for FMS remained higher than that of P84.
The CV variation analysis revealed that the dispersion of the air permeability variation rate for FMS continually increased between 420-450 s and then gradually dropped after 450 s.The dispersion reached its minimum of about 4% at 480 s, and the overall air permeability variation rate remained relatively steady between 480-540 s.Meanwhile, that of P84 remained systematically stable, slightly fluctuating within the 2-4% range.At varying baking times, the CV value for FMS's air permeability variation rate consistently surpassed that of P84.

Rolling residue rate effect
The rolling residue rate holds pivotal significance in water-repellent finishing.An excessively low rolling residue rate can produce poor film-forming characteristics of the fiber surface's finishing agent, leading to inadequate water-repellent effects.Conversely, an excessive rolling residue rate can impede the drying process of nonwovens, subsequently affecting subsequent baking and setting efficacy.Moreover, a surplus rolling residue rate can lead to excessive liquid on the fiber surface, causing the formed film to become overly thick and subsequently influencing the nonwovens' air permeability.
As depicted in Fig. 1(i), when the rolling residue rate reached 49%, the water-repellent level for FMS was grade 5, while P84 achieved grade 7. When the rolling residue rate was below 46%, both filter materials attained a water-repellent grade of 8.This disparity emerged because a 49% rolling residue rate corresponded to heightened water content within the filter material.During the filter material drying process, excessive water vapor disrupted the organic fluorine film on the fiber surface as it evaporated, subsequently impacting the water-repellent efficacy of the filter material (Li et al., 2005).
According to Fig. 1(j), an overall reduction in the variation rate of FMS air permeability correlated with a diminishing rolling residue rate, dampening the downward trend.This change was due to lower rolling residue rates, resulting in fewer effective finishing agent components retained within the filter media.Consequently, the film formed on the fiber surface became thinner, causing a smaller reduction in pore size for the filter media and a less pronounced decrease in air permeability (Bader et al., 2019;Lee and Koplik, 2001).The variation rate of air permeability for P84 remained close to 15%, being consistently below that of FMS as the rolling residue rate decreased.This was similar to the situation described in Section 3.2.4,where the uneven mass distribution of FMS material per unit area led to a non-uniform distribution of pores within the filter media after the rolling process.As the pressure increased, this phenomenon became even more pronounced.Analyzing the CV value alteration revealed that with a reduction in the rolling residual rate, the dispersion degree of the FMS air permeability variation rate continuously dropped, while that of P84 remained stable within the range of 1-3.5%.

Orthogonal experimental design and analysis method
In the orthogonal test, the following five key factors were selected: concentration of the C8-1833 finishing agent, impregnation time, rolling residue rate, baking temperature, and baking time.An orthogonal table L16 (45) with five factors and four levels, as presented in Table S3, was employed.Although many test results in the orthogonal test demonstrated the attainment of the highest water-repellent grade (grade 8), contact angle measurements were also required to quantitatively assess the hydrophobic performance of the filter media (Murase and Fujibayashi, 1997).
The orthogonal test employed the water-repellent grade, variation rate of air permeability, and contact angle as evaluation criteria.Given that water-repellent finishing led to reduced pore size and decreased air permeability in filter media, the variation rate of air permeability was represented by its absolute value.
The effect of each factor on the test results was assessed based on range analysis.A larger range signified a more substantial impact of that particular factor on the test results, designating it as a primary influencing factor.Subsequently, the optimal combination of the above process parameters was determined.The range calculation method for the orthogonal test can be expressed as follows: R = max(k1, k2, ..., ki) -min(k1, k2, ..., ki) (1) where Ni is the number of repetitions of each factor i level; Ki is the sum of the values of each factor i level; ki is the average value of each factor i level; and R is the extreme difference, which is the maximum value of i ki values minus the minimum value.

Orthogonal test analysis of FMS and P84 water-repellent filter media
The range analyses of the orthogonal test for water-repellent finishing of FMS and P84 are described in Tables S5-S7.
It can be seen from Table S5 that since the water-repellent grades of each group of tests of the two filter materials almost reached grade 8, the orthogonal test failed to reflect the effects of different process parameters on the water-repellent grades of filter materials, and contact angle test was required, as mentioned in Section 3.3.1 (Kwok and Neumann, 1999;Zaman et al., 2022).
It can be seen from Table S6 that for FMS, the effects of various process parameters of waterrepellent finishing on the contact angle can be ranked in the decreasing order as A > B > E > D > C. The optimum combination of process parameters in the contact angle test was A4B4C3D4E4.For P84, the above ranking was as follows: A > B > C > D > E, differing from that of FMS only by switched positions of C and E factors.Thus, the water-repellent performance of filter media is mostly controlled by low surface tension of the agent.The optimum combination of process parameters for the contact angle test of P84 was A4B4C4D4E2.
It can be seen from Table S7 that for FMS, the water-repellent finishing process parameters can be ranked by their effect on the air permeability of filter media as C > A > D > E > B. This implies that the thermal shrinkage of fibers after water-repellent finishing and the thickness of organic fluorine film most significantly influenced the air permeability of FMS.A4B4C3D3E4 was the optimum combination of water-repellent process parameters in the air permeability test.Thus, the water repellency of filter media was the primary index, and the variation rate of air permeability was the secondary index.In summary, the optimum combination of process parameters of FMS waterrepellent finishing was A4B4C3D4E4, corresponding to the C8-1833 concentration of 40 g L -1 , an impregnation time of 5 s, a baking temperature of 250°C, a baking time of 540 s, and a rolling residue rate of 49%.
For P84, the ranking of the process parameters by their effects on the air permeability was as follows: E > B > A > D > C. Thus, the pore size of P84 after water-repellent finishing was the most influenced by the film-forming characteristics of the finishing solution.In summary, the optimum combination of process parameters of P84 water-repellent finishing was A4B4C4D4E1, i.e., the C8-1833 concentration of 40 g L -1 , an impregnation time of 5 s, a baking temperature of 260°C, a baking time of 540 s, and a rolling residue rate of 40%.
It is easy to see that both combinations (A4B4C3D4E4 for FMS and A4B4C4D4E1 for P84) are nearly identical, except for C3-C4 and E1-E4 variations.These results can be integrated as follows: the recommended C8-1833 concentration is 35-40 g L -1 , the impregnation time is 5 s, the baking temperature is 250-260°C, the baking time is 510-540 s, and the rolling residue rate is 40-43%.

Stability Test of the Water Repellency of Filter Media
After heating the filter material prepared by the previous orthogonal test at 250°C for 24 h, the contact angles of the filter material were measured, and the thermal stability of the water repellency of the filter material was tested.The test results are listed in Table 1.
It can be seen from Table 1 that the contact angles of the two filter materials exceeded 90° after water-repellent treatment, only slightly fluctuating before and after heating.This strongly indicates that the water-repellent grades of the two filter materials can be maintained at 250°C.Unlike simple coating processes, the film-forming process of the water-repellent agent on the fiber surface involved a curing process of chemical cross-linking reactions between organic fluorine and fiber molecules.This imparted low surface energy characteristics, providing considerable thermal stability and chemical resistance (Kaldybayeva et al., 2023).
Fig. 2 shows that the infrared characteristic absorption peak positions of the two filter materials shifted after water-repellent finishing, implying that water-repellent finishing was rather a chemical than physical process.The water-repellent agent reacted with fiber molecular functional groups, cross-linked to form a film, and solidified on the fiber surface to become a whole (Lee et al., 2005;Gao et al., 2021).Therefore, the water-repellent finishing process used in this test ensured the durability and stability of the water-repellent performance of filter media in practical engineering.

Adaptability Test of Water-repellent Filter Media in High-humidity Environments
Under high-humidity filtration conditions, the test outcomes for five different types of filter media are presented in Fig. 3 and Tables 2 and 3(1).The results reveal that FMS filter material lacking water-repellent finishing is unsuitable for high-humidity gas filtration.Over time, continuous exposure of this filter material to high-humidity environments leads to the complete saturation of its fibers by moisture within the flue gas.Consequently, an environment conducive to accelerated gelatinization of intercepted dust forms inside the filter material.This irreversible gelatinization leads to clogging within the filter material.Even employing drying and robust dust removal methods cannot restore the filtration function of the filter material, rendering it inoperative and necessitating disposal.
After water-repellent finishing, the filtration efficiency of FMS demonstrates greater stability, generally remaining above 80%.Furthermore, the growth rate of filtration efficiency and resistance of water-repellent filter media is significantly lower than that of untreated filter media over time.After testing, the filtration resistance of the filter media increased by 32.53%.Water-repellent finishing induces the cross-linking of organic fluorine on the fiber surface, resulting in a smaller pore size, thereby elevating the initial filtration efficiency and resistance of FMS filter media post-finishing.In contrast, untreated FMS filter media's weight gain predominantly stems from water.After testing, the moisture content of untreated FMS filter media reached 33.58%, causing an 81.55% increase in resistance.
Conversely, water-repellent FMS filter media experienced a weight gain of 4.58% before and after testing, with the increase stemming from KCl.Following water-repellent finishing, FMS filter media demonstrate a commendable resistance against wetting in high-humidity environments.
The experimental results of the P84 filter material echo those of the FMS filter media.When P84 filter media lacks water-repellent finishing and undergoes high-humidity filtration, the risk of bag sticking increases.Continuous filtration for an hour led to 24.43% water absorption and weight gain in the filter media, resulting in an 86.78% increase in filtration resistance.Conversely, water-repellent P84 filter media experienced a 2.64% weight gain and a 21.60% increase in filtration resistance following continuous high-humidity filtration for an hour.
Regarding membrane-coated aramid filter media, the filtration efficiency consistently exceeds 92% under the experimental high-humidity conditions.Following air filter standards, membrane-coated   filter media can achieve high-efficiency filtration levels due to the presence of the membrane.
Although the initial filtration resistance of membrane-coated aramid filter media is higher than that of conventional media due to the existence of the membrane, the coated filter media upholds a high and stable filtration efficiency.Under continuous high-humidity filtration for an hour, the membrane surface of the coated filter material became slightly wet after 36 min (compared to FMS and P84, which were not treated with water repellency, slightly wet at 6 min).This indicates that the membrane material possesses some water repellency, alleviating the impact of condensationinduced wetting to a certain extent.Nonetheless, complete avoidance of wetting phenomena is not achieved.This observation is mirrored by the results, which indicate an 8.39% increase in weight and a 16.87% increase in filtration resistance for the coated aramid filter material post-testing.

Performance Test of Anti-paste Bag for Wet Dust Filtration
The test results depicted in Fig. 4 proved that the filtration resistance of the five kinds of filter materials escalated with the increased dust load and filtration wind speed.Moreover, the filtration resistance of water-repellent filter materials was marginally higher than that of standard filter materials.This test scenario simulated only slight bag sticking conditions and did not encompass filter material dust cleaning operations.Additionally, due to the horizontal installation of the dustfacing surface of the filter material, it lacked the natural drainage conditions offered by vertical installation.Consequently, the filtration resistance results in Fig. 4 solely reflected the resistance of the two filter materials transitioning from a clean state to a dust-carrying one.These results cannot serve as a standard for evaluating the long-term filtration resistance of filter materials.
As detailed in Table 3(2-3), upon drying and cleaning after five cycles of dust generation, the weights of ordinary FMS filter materials and water-repellent filter materials increased by 87.72% and 11.09%, respectively.Compared to the filtration resistance before dust generation, the two filter materials significantly increased after dust cleaning, augmenting by 3.22 times and 1.55 times, respectively.For ordinary P84 and water-repellent filter media, the weight increased by 111.73% and 13.18%, respectively.Compared to the baseline before dust generation, the filtration resistance post-dust removal increased by five times and 1.56 times, respectively.After dust removal, the weight of the membrane-coated aramid filter media increased by 14.53%, and the filtration resistance was 1.42 times that before dust generation.After drying and dust removal of the FMS and P84 filter media, the dust on the surface of the standard filter media was visibly gelatinized, fragmented, and agglomerated, as illustrated in Fig. S4(a) and Fig. S4(c).Conversely, the dust on the surface of water-repellent filter media maintained a normal adhesion state after drying and dust removal, as shown in Fig. S4(b) and Fig. S4(d).For coated aramid filter material, the viscous dust on the surface was essentially removed by drying and dust removal, revealing some fine patches on the membrane material.These fine dust patches could eliminated through robust dust removal techniques.
The test results strongly indicate that water-repellent filter material delayed the adhesion of wet dust.However, when the flue gas humidity was excessively high, leading to severe condensation, the "gelatinization" of dust lost its discrete particle characteristics.In such cases, the water-repellent filter material's ability to maintain filtration performance over extended periods diminished, potentially affecting the system's normal operation.Consequently, in practical working conditions, when the bag sticking phenomenon was not severe, or dew condensation occurred swiftly, the water-repellent filter material could be dried using high-temperature dry flue gas, followed by a robust dust cleaning procedure to restore its filtration function.
The experimental results on membrane-covered aramid filter media showed that the membrane material possessed a certain anti-adhesion effect on wet dust.Dust remained in a dry state on the filter media surface, and the membrane material alleviated dust infiltration caused by water droplets on the dust-facing surface to a certain extent.
Based on the findings described in Sections 3.5 and 3.6, the dust-facing surface, airflow direction, and water droplet fall direction of the filter material were obtained and plotted in Fig. S3(c).On the dust removal filter table, where the filter material was positioned horizontally with the dust-facing surface oriented upwards, water droplets (sprayed on the filter material surface before testing) remained on the dust-facing surface.Despite these water droplets being dispersed discretely like beads on the filter material surface, they still created conditions conducive to dust "gelatinization".In contrast, the water-repellent filter material exhibited no wetting phenomenon on the air test bench from start to finish.The filter material surface remained dry, effectively preventing the bag from sticking.Therefore, while designing the air distribution for a bag filter, it is advisable to adopt the side air inlet mode (traditional dust hopper air inlet) and enhance dust filter insulation.If needed, heating devices should be employed to avert flue gas condensation occurrences.

CONCLUSIONS
The results obtained made it possible to draw the following conclusions: (1) Through an orthogonal optimization test, the optimal water-repellent process parameters should be determined.For the particular FMS and PM84 filter materials and the C8-1833 waterrepellent agent adopted in this study, the optimal process parameters for water-repellent finishing of the primary flue gas were as follows: the C8-1833 concentration of 35-40 g L -1 , an impregnation time of 5 s, a baking temperature of 250-260°C, a baking time of 510-540 s, and a rolling residue rate of 40-43%.
(2) The performed experimental study revealed that the film-forming process of the waterrepellent agent on the fiber surface is a curing process of the chemical cross-linking reaction between organic fluorine and fiber molecules, not a simple coating process on the fiber surface, so it has considerable thermal stability and chemical stability.(3) In the LT + BF system, the fabric structure of the filter material with water-repellent finishing and membrane covering on the dust-facing surface should have the optimal anti-paste bag performance and high filtration efficiency.To avoid the phenomenon of bag sticking, the following principles of the operation management and main design of the dust collector should be followed: First, to avoid the condensation of flue gas, it is necessary to turn on the heating device of the dust collector, ensuring that the flue gas temperature is 25-30°C above the dew point temperature.Second, when there is a slight bag sticking, it should be taken to dry the filter bag (if necessary, it can be washed with water before drying) and then use strong dust removal to restore the filtering function of the filter material.In the design of airflow organization of bag filter in the LT + BF system, it is suggested to adopt side air inlet to avoid creating favorable airflow conditions for bag sticking when condensation occurs.

Fig. 2 .
Fig. 2. Infrared spectra of two kinds of filter media before and after water-repellent finishing.(a) Infrared spectra of FMS before and after water-repellent finishing; (b) Infrared spectra of P84 before and after water-repellent finishing.

Fig. 3 .
Fig. 3. Filtration performance of three types of filter media in high-humidity environments.

Table 2 .
Wetting conditions of different filter materials in high-humidity filtration during one hour.a : 0 indicates no wetting, 1 indicates slight wetting, 2 indicates increased wetting, and 3 indicates complete wetting.* indicates water-repellent treatment.

Fig. 4 .
Fig. 4. Relationship between filtration resistance and dust load under different wind speeds.

Table 1 .
Contact angle test results of two kinds of filter media before and after heating at 250°C.

Table 3 .
Mass variations before and after an hour in high-humidity filtration environments and changes in mass and filtration resistance before and after five dust cycles.