“Green” nano-filters: fine nanofibers of natural protein for high efficiency filtration of particulate pollutants and toxic gases

Abstract

Particulate and chemical pollutants are ubiquitous in polluted air. However, current air filters using traditional polymers can only remove particles from the polluted air. To efficiently filter both particulates and chemical pollutants, development of multi-functional air filter materials with environmental friendness is critically needed. In this study, gelatin is employed as an example to study the potential of natural proteins as high-performance air-filtering material. Based on an optimized composition of a “green” solvent, uniform gelatin nanofiber mats were fabricated via an electrospinning approach. For the first time, it is found that the resulting nanofabrics possess extremely high removal efficiencies for both particle matter (with a broad range of size from 0.3 μm to 10 μm) and various toxic chemicals (e.g. HCHO and CO). Moreover, these high efficiencies are realized by the protein nanofabrics with a much lower areal density (3.43 g m⁻²) when compared with that of commercial air filters (e.g. 164 g m⁻² for high efficiency particulate air filter (HEPA)). This study reveals that nanofabrics of natural proteins hold great potential for application in “green” and multi-functional air filtering materials.

---

Introduction

Air pollution has been of great concern due to the huge emission of particulate and chemical pollutions. The release of chemicals, particulates and biological materials into air can lead to various diseases or discomfort to humans and other living organisms, alongside other serious impacts on the environment. The unknown combination of particles and chemical pollutants makes the polluted air even more harmful. Particle Matter (PM) is usually categorized into two groups, PM_2.5 and PM_10–2.5 which denote particles with aerodynamic diameters smaller than 2.5 μm and between 2.5–10 μm, respectively.^1,2 According to the 2009 and 2012 World Bank report, more than 60% of Americans live in air quality levels that are potentially detrimental to health.^3,4 Recent studies have reported a more serious PM pollution problem in developing countries.⁵ A high degree of air pollution was responsible for numerous premature deaths. PM_2.5 particles are the critical particulate pollution to be filtered due to their ability to penetrate into human lungs and bronchi.^1,2,5–10 Indoor air quality has become an increasing issue as well. More and more buildings incorporate air filtration protection in their heating, ventilation, and air conditioning systems, but a significant amount of energy is required to maintain the air exchange process due to a high air-resistance (pressure drop) of the air filters. Therefore, air filters with high-efficiency of removing particles and chemicals simultaneously are critically needed.

Understanding of the composition of polluted-air is critical for the development of air filtering materials. In general, the composition of pollutants in polluted-air is extremely complicated due to the complexity of the sources of pollution. PM particles can be produced from variety of sources, such as fuel combustion in vehicles, industrial factory plants, cigarette smoke, dust, etc. These PM particles behave distinctly due to their diverse chemical composition. Most PM_2.5 particles are composed of organic compounds such as carbon derived matters (e.g. carbon dioxide and carbon monoxide), inorganic compounds (e.g. sulfur dioxide (SO₂²⁻), sulfate (SO₄²⁻), silicon dioxide (SiO₂), and nitrate (NO₃⁻), etc.), and biological threats (e.g. bacteria and viruses).^11–13 These particles are very stable in air and have lifetimes between hours to weeks due to their very small sizes. They can scatter visible light and reduce visibility because of the similarity between their particle size and visible light wavelengths.¹⁴ In addition to PM particles, polluted air includes a wide variety of chemical gases such as carbon monoxide (CO), nitrogen dioxide (NO₂), methane (CH₄), benzene, dioxin, ozone, etc. A large number of chemicals in polluted air are classified as volatile organic compounds (VOCs) which are primarily emitted by petrochemical and allied industries. VOCs can undergo different kinds of photochemical reactions in the atmosphere and cause various environmental hazards. In gas phase carcinogenic or otherwise toxic VOCs present a danger to humans.^15,16 Since the polluted air is usually composed of pollutants with complicated compositions and physicochemical properties, multi-functional air filtering materials that are able to generate various types of interactions with the pollutions are of great interest for air-filtering applications.

Air filters are the most common used device to remove pollutants from the air. They have been widely used in different areas, e.g. automotive industries, residential, general commercial, and even hospitals, general surgeries and so on. The filtration function is mainly realized via physical and PM size-based capturing mechanisms. There are four primary mechanisms for filtration based on the size of the pollutant particles. Sieving is one of the most important mechanisms and is only effective for particles with sizes larger than the pore size of the filter. For particles with sizes smaller than the pore size of the filter, inertial impaction, interception, and diffusion are the dominant mechanisms for filtration. In specific, interception occurs when small particles flow with the air stream and come into contact with the fiber surface. The attractive interactions between the small particles and fibers play a critical role for this mechanism. The diffusion mechanism is effective for even smaller particles with aerodynamic size smaller than 100 nm. For these particles, movement is dominated by Brownian motion and capturing occurs via random collision.¹⁷ Traditionally, air filters are made of porous films, such as non-woven fibrous mats with randomly oriented micron-size fibers. These types of air-filtering materials have several disadvantages as explained below. First, the fibers are made of chemically synthesized or petroleum based materials, such as polypropylene and fiberglass. These conventional materials provide very limited chemical functionality, resulting in insufficient interactions with pollutants. Secondly, disposing of used air filters made of these materials can cause further environmental pollution as most of them are not environmentally friendly. Finally, microfiber-based air filtering materials possess limited surface area, which further deteriorate the filtration performance.

To address the above issues related to conventional air-filter materials, nanofiber mats have been of great interest recently.¹⁸ Nanofiber mats possess several advantages as explained below. Firstly, nanofibers will tend to absorb substance from the environment due to a high surface energy, which enhances the interactions between fibers and pollutants. Secondly, nanofibers can significantly increase the surface area of filter materials. In other words, nanofibers provide more active sites for trapping pollutants. As a result, nanofiber mats can realize a high filtering efficiency for PM while possess low pressure drop or air resistance, which is critical for their practical application.^19,20 Therefore, nanofabrics of polymers rich in functional groups represent a promising solution for high-performance air-filtering materials. In particular, biomaterials, such as natural proteins, are promising candidates as high-performance air filtering materials.²¹ It is well-known that proteins are rich in functional groups, that is, the R-groups on the amino acids. These functional groups make proteins an ideal material for air filtering application. For example, chitosan has been mixed with poly(ethylene oxide) and fabricated into nanofibers as a filter material. The cationic nature of chitosan was used to achieve more than 70% removal of heavy metal ions.²² Other biomaterials were also studied as air filter materials,²³ however, they were usually mixed with conventional polymers to fabricate nanofibers. As a result, the potential of pure protein nanofabrics as high-performance air filtering materials has never been studied based on the author's knowledge. Of course there are still challenging issues for scaling—up the process using metric tons of natural protein materials.

In this study, the potential of pure protein nanofabrics for air filtering application is investigated. It is believed that the combination of nanomaterials with natural proteins can lead to a powerful nanofabric with the ability to trap various kinds of pollutants, including particulate and toxic gas. In particular, gelatin is employed as an example for that. Gelatin protein is derived from thermal denaturation of collagen, the most abundant protein in human and animal bodies.²⁴ Fabrication of gelatin nanofibers has been proved very successful and they are usually reported as scaffolds for food, energy, pharmaceutical, environmental, and medical applications,^25–27 except as air filtering material. Here, to study gelatin nanofabrics for air filtration purposes, the fabrication of gelatin nanofibers is further improved. Firstly, instead of using toxic solvents (e.g. 2,2,2-trifluoroethanol (TFE) or 1,1,1,3,3,3-hexaflouro-2-propanol (HFIP)) which are usually used for the electrospinning of gelatin,²⁸ a non-toxic solvent (mixture of acetic acid and water) is employed. Secondly, the diameter of the gelatin nanofibers is further reduced to be around 70 nm, which is smaller than the typical values (ca. 100 nm) for gelatin nanofibers.^29–31

Experimental

Materials and methods

Raw materials and solution preparation. Gelatin powder (type A) produced from porcine skin was supplied from Sigma-Aldrich (MO, USA). Acetic acid (99.9% purity) was purchased from J.T.Baker® (PA, USA). Gelatin was dissolved in mixed solvent (volume ratio, acetic acid : DI water = 80 : 20) with a concentration of 18 wt% at 65 °C. The mixed solvent was used to achieve a good electrospinning of the gelatin solution. With that ratio between water and acetic acid, it was found that a homogenous yellow solution and stable electrospinning of the solution can be achieved.

Preparation of protein filter nanofabrics. Protein nanofabrics were prepared by electrospinning techniques. The gelatin solution was loaded in a syringe (Monojet™ Kendall) with a 21-gauge blunt-tip needle. An operating voltage of 18–20 kV was employed for the electrospinning and was controlled by a high voltage power source (ES50P-5W, Gamma High Voltage Research). A mono-inject syringe pump (KD Scientific, KDS-100) was utilized to pump the gelatin solution. Commercial aluminum mesh with wire diameter of 0.011 inch and mesh size of 18 × 16 was grounded to collect the gelatin fibers. The distance between needle and sample collector was fixed to be 10 cm and an average flow rate of 0.6 ml h⁻¹ was utilized.^31,32

Polluted-air samples preparation and air-filtering testing. Cigarette smoke was used as the source of pollution to prepare polluted-air samples. It has been estimated that cigarette smoke includes PM particles with a broad range of sizes (0.01 to 10 μm), and more than 7000 different chemicals, hundreds of which are toxic. The most dangerous chemicals of interest for filtration are carcinogenic, such as formaldehyde (HCHO), carbon monoxide (CO), ammonia (NH₃), hydrogen cyanide (HCN) and toxic metal ions (chromium (Cr³⁺, Cr⁶⁺), lead (Pb²⁺), cadmium (Cd²⁺)).³³ Since the original polluted-air samples were so concentrated with PM and chemicals, they were diluted in a gas bag to a hazardous level which can be measured by the analyzer. The diluted polluted-air with detectable level of pollutions was used as the final polluted-air sample for air-filtering testing. Before the air-filtering testing, the initial concentrations of PM with different particle sizes (0.3–10 μm) and toxic chemicals (HCHO and CO) in the air samples were measured by a particle counter (CEM, DT-9881). To perform the air-filtering testing, the pressure difference of both sides of air filter was controlled and measured by a manometer (UEi, EM201-B) with a standard air flow velocity of 5 cm s⁻¹ to investigate the air flow resistance of the air filter material.³⁴ In all the measurements, a circular filter sample with diameter of 37 mm was placed in a home-made sample holder. The filtered air was collected by another clean gas bag which was vacuumed in advance (the digital image of experimental setup is shown in Fig. S1, ESI†). When the air-filtering testing ended at different filtering time, the concentrations of the PM and toxic chemicals inside the clean gas bag with filtered air were measured and recorded. Via the eqn (1), one can determine the removing efficiency η_p.where C_p is the concentration of the pollution in the polluted-air sample before air-filtering testing, and C_c is the concentration of the pollution in the filtered air sample.

Characterizations. To study how the particle pollutants were removed by the protein nanofabrics, SEM (FEI SEM Quanta 200F) was employed to investigate the morphology of the protein nanofabrics before and after air filtering testing. The samples were sputter-coated with 10 nm gold nanolayer in thickness using Technics Hummer V sputter coater. In order to study the possible interactions between the protein nanofabrics and pollutants, FTIR (Nicolet, Thermo Scientific) absorption spectra was employed. To distinguish the interactions between nanofibers and pollutants from the interactions inside the fabric or polluted-air themselves, the FTIR spectrum of three kinds of samples were recorded and compared. These samples include polluted-air, clean protein nanofabrics before and after filtration. All the measurement was repeated at least 3 times and good repeatability was found for these samples.

Statistical analysis. Data were analysed using one-way analysis of variance (ANOVA) multiple comparison method. The confidence interval was set to 95%. The differences among the data with a p-value < 0.05 were reflected to be statistically significant (see ESI†).

Results and discussion

Gelatin nanofabrics/nanofiber mats

The target of this work is to study the potential of gelatin nanofabrics as high-performance air filtering material with two levels of air filtering functions: (1) removing particles, such as dust, pollen, with particle sizes in a broad range from 0.1 to 100 μm; (2) removing toxic or obnoxious gases, such as formaldehyde and carbon monoxide in tobacco smoke. It is known that gelatin molecules possess a broad range of functional groups in their multi-level structures. The characteristics of these chemical structures provide capability for interaction with multiple species of polar molecules, which lead to a great potential to capture many chemicals. Specifically, gelatin consists of glycine (21.4%), proline (12.4%), hydroxyproline (11.9%), and glutamic acid (10.0%) in its amino acid profile.^35,36 The amino acids bring gelatin various functional groups (such as carboxylic and hydroxyl, charged groups, and many other polar/nonpolar functional groups). These functional groups can act as active sites generating numerous interactions with pollutants, including hydrogen bonding, ionic bonding, and charge–charge interactions and so on.^35,36 Combined with electrospinning technique, gelatin nanofibers can be fabricated (Fig. 1). Electrospinning is an effective method for making uniform nanofibers with high aspect ratio, and high pore interconnectivity with size ranging from micron to nanometer scale.

Fig. 1 Schematics of gelatin solution preparation and nanofibers fabrication via electrospinning, followed by the schematics of a single gelatin nanofiber with a functional surface due to various functional groups available in its structure. Porous structure and fiber diameter of nanofabrics contribute to particulate filtration while the rich functional groups on the fiber surface provides toxic chemical filtration property.

First of all, an appropriate solvent, in particular a non-toxic solvent, needs to be selected for preparing the gelatin solution effective for making nanofibers via electrospinning. Many studies have been reported on the fabrication of gelatin fibers by using toxic solvents (TFE, HFIP),^29,30 which yielded average fiber diameter ranges from 100–600 nm. In this study, gelatin nanofabrics were fabricated by employing aqueous acetic acid (AA) as a “green” solvent. In order to achieve efficient molecular dissolution of gelatin, good electrospinability and, as the result, uniform filter mat with nanoscale fiber diameter, the solvent composition must be adjusted. More importantly, the uniformity of the nanofibers in the mat along with smaller fiber diameters can result in high surface area for capturing more pollutants, which enables the filter to achieve high filtration efficiencies. Therefore, the effect of AA-to-water ratio on the resulting nanofiber diameters and their distribution was studied. The mixture solvent with optimized composition was determined for spinning out uniform gelatin nanofiber mats. The microstructures of gelatin nanofibers prepared in different ratios of the solvent and the fiber morphology of a commercial HEPA filter were compared using scanning electron microscopy (SEM) images (Fig. 2). It can be seen that by increasing the ratio of AA to water in the solvent from 60 : 40 to 80 : 20, the size of nanofibers was reduced from ca. 470 to ca. 70 nm. Moreover, the nanofiber uniformity was improved significantly: the relative deviation of the nanofibers reduced from 53% to 25% (see fiber diameter distributions and inserted table showing the statistic results in Fig. 2 and S2 for the pore size distribution of nanofibers in ESI†). These results show that by using the mixed solvent with the AA-to-water ratio of 80 : 20, uniform gelatin filter mats with the nanofiber diameter of 70 nm were successfully fabricated, which is smaller than the reported studies showing the fiber diameter of few hundred nanometers,^31,37 reduction in the nanofiber diameter can significantly improve the air filtration capabilities for both particulate and toxic chemicals due to their high active surface areas. In general, nanofibers with high surface-to-volume ratio and surface activity will have high particle removal efficiency via interception, diffusion, and other mechanisms while retaining very low resistance to air flow which results in low pressure drop.^38–40 At the same time, due to the high density of functional groups along the gelatin nanofibers, the nanofiber mats are expected to possess multiple filtering functions: toxic chemical and particulate filtration. The following air filtering performance studies were focused on the gelatin filter nanofibers produced using the optimal solvent.

Fig. 2 SEM images of gelatin nanofabrics prepared in different ratios of AA-to-water; effect of solvent ratio on morphology and fiber diameters of gelatin nanofabrics compared with that of commercial HEPA filter; percentages in the table are relative standard deviation.

Particulate filtration

As shown in Fig. 3a, for the gelatin nanofabrics, the removal efficiency is dependent on the PM size as well as the areal density of the nanofabrics. The areal density of the nanofabrics was controlled by controlling the volume of the solution that was electrospun on the substrate. All of the gelatin nanofabric mats possessed a thickness within the range of 8–20 μm. It can be found that with a high areal density (e.g. greater than ca. 3.43 g m⁻²), the gelatin nanofabrics showed almost similar removal efficiency of above 99.20% for PM with sizes from 0.3 to 10 μm (see Table S1† for number concentration of particulate pollutants before and after filtration). In particular, the removal efficiency for the most penetrating particle size (MPPS) of 0.3 μm particles was significantly improved from 77.10 to 99.32% by increasing the areal density of the nanofabrics from 2.25 to 3.43 g m⁻². PM particles with size around 0.3 μm (PM_0.3) are known as the hardest to capture and a high-performance air filter should give rise to an efficiency above 95% for PM_0.3.³⁴ The results indicate that the areal density of the nanofabrics is critical for the removal efficiency of small particles, but not large particles. This result can be explained as the difference in the mechanisms for filtering big particles (sieving) and small particles (smaller than the pore size). Specifically, large particles were removed by size effects, while small particles were trapped via the strong interactions between particles and nanofibers. The removal efficiency for PM_2.5 and PM_10–2.5 via nanofabrics with different areal density is shown in Fig. 3b. In particular, the areal density of 3.43 g m⁻² (thickness = 16 μm) is close to an optimal (minimum) value to achieve the highest removal efficiency of both PM_2.5 (99.51 ± 0.23%) and PM_10–2.5 (99.63 ± 0.11%). For the filters with areal densities higher than 3.43 g m⁻², the removal efficiency of both PM_2.5 and PM_10–2.5 was almost constant.

Fig. 3 Particulate removal efficiency of gelatin filter nanofabrics. (a) Particulate removal efficiency of gelatin air filters with different areal density for various PM particle sizes. (b) PM_2.5 and PM_10–2.5 removal efficiency comparison between gelatin air filters with different areal density and commercial HEPA filter.

The gelatin nanofabrics are promising air filtering materials with removing efficiency for PM_2.5 higher than 95%, the standard suggested for high-efficiency air filters.³⁴ More significantly, the gelatin nanofabrics can achieve the high efficiency for PM by a much lower area density (ca. 3.43 g m⁻²) as compared with the most successful commercial one (HEPA, 164 g m⁻²). The high removal efficiency for PM is likely contributed by the combination of “nano-size” effects and the surface properties of gelatin nanofibers. The diversity of functional groups on the nanofiber surface can provide strong interactions with the pollutants, which is critical for the removing of PM with sizes much smaller than the pore size of the nanofabrics and of toxic gases as will be discussed later.

Also, the morphology of gelatin nanofibers were studied via SEM and the results are shown by Fig. 4a–f. In specific, Fig. 4a and d (also see Fig. S3a–c, ESI†) are the SEM images of the nanofabrics before air filtration testing, that is, the morphology of as-spun pure gelatin nanofabrics. Fig. 4b and d–f (also see Fig. S3d–f, ESI†) are the SEM images of the nanofabrics after air filtration testing. The digital photos inserted in the SEM images are the nano-filter before and after filtration. The obvious color change from white to yellowish indicates that the nano-filter has absorbed lots of pollutants. Further, the SEM images confirm this point as one can observe lots of particles have been trapped on the surface of gelatin nanofibers. The notable change in color should also be related to the absorption of some toxic gases, which will be discussed later.

Fig. 4 (a, d) SEM images of gelatin filter with different magnifications along with photograph of gelatin air filter before filtration, (b, d–f) SEM images of gelatin filter with different magnifications along with photograph of gelatin air filter after filtration.

Toxic gases filtration

Removing of toxic chemicals via air filters with high efficiency is challenging since they are small molecules with sizes much smaller than that of particles. Conventional ways to remove toxic gases are using absorptive particles with high specific surface area, such as activated carbon. For the gelatin nanofabrics, two kinds of toxic chemicals, formaldehyde (HCHO) and carbon monoxide (CO) which can be detected by the analyzer, were chosen as examples to test the chemical removal capability. Fig. 5a shows the chemical removal efficiency of formaldehyde for gelatin nanofabrics with different areal densities compared with that of commercial HEPA filter (see Table S2† for concentration of particulate formaldehyde and carbon monoxide before and after filtration).

Fig. 5 Chemical removal efficiency of gelatin filter nanofabrics. (a) Formaldehyde (HCHO) removal efficiency comparison between gelatin air filters with different areal density and that of commercial HEPA filter. (b) Carbon monoxide (CO) removal efficiency comparison between gelatin air filters with different areal density and that of commercial HEPA filter.

It can be seen that the chemical capturing efficiency of formaldehyde increases from 65.0% to 83.0% by increasing the areal density of the filter from 2.25 to 3.80 g m⁻². Moreover, the removal efficiency of carbon monoxide as shown in Fig. 5b increased from 62.3% to 76.1% for gelatin nanofabrics with the same change in areal density. In comparison, for the commercial HEPA filter, one of the most successful air filter used today, the chemical removal efficiencies of formaldehyde and carbon monoxide molecules are less than 5 and 3%, respectively. The high chemical removal efficiency indicates that the combination of nanofabrics with functional polymers is the key to remove toxic chemicals which cannot be removed by size-based mechanisms. Therefore, protein nanofibers provide a promising solution for multi-functional air filtering materials.

Pressure drop and figure of merit (quality factor)

In addition to the particulate and chemical removal efficiency, air flow resistance (pressure drop) is another critical parameter describing the performance of an air filter. High pressure drop downstream of an air filter will consume a large amount of energy due to the pumping required to provide a sufficient air flow, which makes the air filter energy consuming. The suggested range for pressure drop (ΔP) set by DOE is less than 1.3 in H₂O (approximately 320 Pa).³⁴ The effect of areal density on pressure drop of the gelatin filters was investigated using standard 5 cm s⁻¹ air face velocity. Fig. 6a shows that the air flow resistance of the filters increases with the areal density. Quantitative analysis demonstrated that the pressure drop of the gelatin nanofabrics with the lowest areal density (2.25 g m⁻²) was ca. 143 Pa and increased to approximately 201 Pa for the filter with the highest particulate/chemical efficiency (3.43 g m⁻²). These pressure drop values meet the requirement of a high-efficiency filter. The pressure drop along with removal efficiency allow us to determine the optimal areal density of the gelatin nanofabrics for achieving a good balance between high removal efficiency of pollutants (PM and toxic chemicals) and pressure drop, which is critical for practical applications. For the gelatin nanofabrics when the areal density is higher than ca. 3.43 g m⁻², such as 3.80 and 7.67 g m⁻², the removal efficiency is improved by less than 0.5% (see Fig. 3b), while the pressure drop increases by more than 74% (see Fig. 6a). Thus, 3.43 g m⁻² should be close to the optimal areal density for the gelatin nanofabrics.

Fig. 6 Pressure drop and overall air filter performance evaluation of gelatin filter material. (a) Dependence of pressure drop (air flow resistance) on the areal density for the gelatin nanofabrics. (b) Quality factor comparison with commercial air filters and transparent PAN air filter,¹⁸ and gelatin air filter.

To comprehensively describe the filtration performance, the pressure drop and removal efficiency is combined into one parameter, the quality factor (QF), also known as figure of merit (FOM), which can be calculated using eqn (2).^41–44

where ƞ_p is the removal efficiency and ΔP is the corresponding pressure drop. QF is a representative of the ratio between removal efficiency and the air flow pressure drop. This quantitative factor indicates that a good air filter should provide a high removal efficiency and a low pressure drop; hence, a higher QF means a better filtration performance of an air filtering material. Fig. 6b shows the comparison of figure of merit of the gelatin filter fabrics with commercial air filters and poly(acrylonitrile) (PAN-85) nano-filters as reported recently.¹⁸ It can be found that the gelatin nanofabrics with 3.43 g m⁻² areal density possess the highest quality factor (0.026 Pa⁻¹) among them. The quality factor for gelatin nanofabrics was calculated at 30 minutes of filtration testing and 5 cm s⁻¹. It is noted that the figure of merit of an air filter (including the removal efficiency and pressure drop) is never constant and will change with time of using in practice.

Stages of filtration process

For air filtering materials, the analysis of the pollutant absorption process is critical for understanding the long-term filtration performance. For the multi-functional gelatin air filtering material, the pollutant absorption process was tracked quantitatively via calculation of the pollutant weight-percentage gain and pollutant weight absorption rate over time using eqn (3) and (4):where W_p is the pollutant weight percentage gain, W_t is the weight of the air filter after filtration of time t, W_f is the weight of the pure air filter before filtration test, W_{p_rate} is the pollutant weight absorption rate, and ΔW_pollutant is the absolute weight of the pollutants absorbed between each time interval. Results are compared with those of the commercial HEPA filter in Fig. 7.

Fig. 7 (a) Pollutant weight gain percentage (W_p) changes of the gelatin air filter and commercial HEPA filter over the filtration time, and (b) pollutant absorption rate (W_{p_rate}) of the gelatin air filter and commercial HEPA filter over the filtration time.

At the early stages of filtration (first 30 minutes to 1 hour), particles migrate and merge to form bigger, spherical aggregates.¹⁸ Moreover, the particle weight percentage gain reaches 53% of the nanofabrics' weight and sharply increases after 1 hour of testing to 106% while the HEPA filter only showed an increase from 1% to 1.3% due to its very high areal density (shown in Fig. 7a). With the increase of filtering time (after 2 hours), more particles were trapped by the nanofibers. The accumulation of particles is also coupled with complex deformation/transformation processes due to possible physicochemical interactions. This process has also been reported in the recent study on transparent PAN air filters.¹⁸ The phenomenon results in a decrease of clear filter area. In this stage, the particle weight percentage gain for gelatin nanofabrics increased moderately to 130%, while the absorption rate of pollutants decreased significantly (shown in Fig. 7b), which can prove the hypothesis mentioned above. By further increasing the filtration time, the particle weight percentage gain of the filter reached a plateau region (gelatin nanofabrics and commercial HEPA filter reached to 150% and 2.3% pollutant weight percentage gain, respectively) and the pollutant absorption rate was decreased even more significantly for both filters. However, the pollutant absorption rate of gelatin nanofabrics was higher than that of commercial HEPA filter due the functional and active surface of gelatin nanofabrics which enable it to absorb more particles and chemicals within a shorter time period. The pollutant absorption phenomenon is critical for practical applications as it is related to the long-term performance or the life-time of the air filter material (also see Fig. S4 and S5, ESI† for SEM images showing the evolution of surface morphology of gelatin nanofabrics during filtration time and thermal stability of gelatin nanofibers before and after filtration).

Filtration mechanism analysis

To further analyse the performance of the gelatin nanofabrics, the filtration mechanisms were studied based on examining the surface chemistry of pollutants and gelatin nanofabrics before and after filtration. As mentioned above, numerous functional groups exist in gelatin structure. These functional groups can strongly interact with various pollutants in the air and enable the filter to remove the pollutants (toxic chemicals and solid particles) via an interaction-based mechanism besides existing size-based primary mechanisms. Fig. 8a depicts a simplified gelatin molecule (filter) along with PM, and formaldehyde as examples of solid and gaseous pollutants. It can be seen that the aldehyde group can undergo addition reactions with amine groups in gelatin (provided by amino acids such as lysine) forming aldimine linkages. This reaction elicits a change in filter color from white to a yellow color. PM particles and other pollutants with different compositions can interact with the gelatin nanofabrics through hydrogen bonding, charge–charge interactions etc. Fig. 8b shows an SEM image of gelatin nanofilter that captured numerous pollutants via a combination of interaction-based and size-based mechanisms. This schematic indicates a new interaction-based mechanism of filtration for gelatin nanofabrics besides the primary four size-based mechanisms. To further understand the possible interactions between the gelatin nanofabrics and the pollutants (solid particles and toxic chemicals), Fourier transform infrared spectroscopy (FTIR) and dielectric constant measurement (see Fig. S6, ESI†) were used to identify the functional groups from the pure gelatin nanofabrics, polluted air and nanofabrics with trapped pollutants. The FTIR spectra of the cigarette smoke is shown in Fig. 8c. The specific peaks of functional groups in the cigarette smoke are around 3649, 2360, 1653, 1558, 1506, and 1456 cm⁻¹ which indicate the existence of O–H, C–H (aldehyde), C=O, and C–O (last three peaks) groups, respectively. All of these groups in polluted-air sample may interact with the groups on the surface of gelatin nanofibers. The comparison of the FTIR spectra for the gelatin nanofibers before and after filtration is shown in Fig. 8d (see Fig. S7, ESI†). It can be found that there is no new peak formed after filtration. However, there is a significant change in the intensity at specific groups and interactions, including hydroxyl, carboxyl, and amine functional groups (see Fig. 8e). These results can be explained as that the types of interactions between the pollutants in the smoke and gelatin fibers are covered by those inherently existing in gelatin nanofabrics. Therefore, the dramatic increase in the peak intensity of these functional groups after filtration testing should be the result of the interactions between gelatin nanofabrics and the pollutants, such as hydrogen bonding, ionic bonding, and charge–charge interactions, etc.

Fig. 8 (a) Simplified representation of functional-capturing filtration mechanism of gelatin nanofabrics via interactions between gelatin molecules and pollutants. (b) SEM image of gelatin nanofabrics after filtration (c) FTIR characterization of cigarette smoke PM particles demonstrating existing functional groups. (d) FTIR characterization of gelatin filter before and after filtration showing the active functional groups and PM-filter interactions. (e) Functional group peak intensity comparison between gelatin fibers before and after filtration indicating strengthening of corresponding bonds due to PM-filter interactions.

Conclusions

In summary, the gelatin protein was studied as an example to demonstrate the potential of natural proteins to serve as environmentally friendly and high-performance air-filtering materials. Uniform gelatin nanofiber mats with very small diameters were fabricated by employing a “green” solvent with optimized composition. It has been found that the gelatin nanofiber mats with a controlled uniformity and small fiber diameters possess extremely high particulate removal efficiencies of more than 99.3% and 99.6% for PM_0.3 and PM_2.5, respectively. These results indicate that the gelatin nanofibers with a much lower areal density (e.g. 3.43 g m⁻²) can efficiently remove a broad range of PM particles similar to one of the most efficient particulate air filters, HEPA with areal density of 164 g m⁻². More significantly, the combination of the inherent surface chemistry of gelatin nanofibers (i.e., various functional groups on the fiber surface) and nanofiber technology enables gelatin protein nanofibers to have high interaction capability with toxic chemicals present in the air. Particularly, the gelatin nanofabrics possess excellent efficiency of absorbing toxic chemicals (e.g. ca. 80% for HCHO; 76% for CO), which has never been realized in any air filters with a single material composition. The mechanisms responsible for such simultaneous high capturing capabilities of particulate and toxic chemical were analyzed. It is believed that the interaction-based filtration mechanism besides the existing size-based primary mechanisms result in these functions. This study indicates that protein nanofabrics are promising “green” air-filtering materials for next generation air filtration systems.

Acknowledgements

The authors appreciate the School of Biological Sciences Franceschi Microscopy & Imaging Center (FMIC), Washington State University for providing the Field emission electron microscope. Also the authors would like to thank Mr Allen W. Eyler, and Ms Tian Liu for their helpful discussion.

Notes and references

1. R. M. Harrison and J. Yin, Sci. Total Environ., 2000, 249, 85–101.
2. R. D. Brook, S. Rajagopalan, C. A. Pope, J. R. Brook, A. Bhatnagar, A. V. Diez-Roux, F. Holguin, Y. Hong, R. V. Luepker, M. a. Mittleman, A. Peters, D. Siscovick, S. C. Smith, L. Whitsel and J. D. Kaufman, Circulation, 2010, 121, 2331–2378.
3. J. L. Repace, R.-T. Jiang, V. Acevedo-Bolton, K.-C. Cheng, N. E. Klepeis, W. R. Ott and L. M. Hildemann, Environ. Res., 2011, 111, 473–484.
4. L. Li, G.-Z. Lin, H.-Z. Liu, Y. Guo, C.-Q. Ou and P.-Y. Chen, Environ. Pollut., 2015, 205, 153–160.
5. R. Betha, S. N. Behera and R. Balasubramanian, Environ. Sci. Technol., 2014, 48, 4327–4335.
6. A. Nel, Science, 2005, 308, 804–806.
7. R. Zhang, J. Jing, J. Tao, S. C. Hsu, G. Wang, J. Cao, C. S. L. Lee, L. Zhu, Z. Chen, Y. Zhao and Z. Shen, Atmos. Chem. Phys., 2013, 13, 7053–7074.
8. C. A. Pope III and D. W. Dockery, J. Air Waste Manage. Assoc., 2006, 56, 709–742.
9. D. E. Horton, C. B. Skinner, D. Singh and N. S. Diffenbaugh, Nat. Clim. Change, 2014, 4, 698.
10. S. Wu, F. Deng, H. Wei, J. Huang, X. Wang, Y. Hao, C. Zheng, Y. Qin, H. Lv, M. Shima and X. Guo, Environ. Sci. Technol., 2014, 48, 3438–3448.
11. H. Wang, Y. Zhuang, Y. Wang, Y. Sun, H. Yuan, G. Zhuang and Z. Hao, J. Environ. Sci., 2008, 20, 1323–1327.
12. X. Han and L. P. Naeher, Environ. Int., 2006, 32, 106–120.
13. M. Matti Maricq, J. Aerosol Sci., 2007, 38, 1079–1118.
14. D. Y. H. Pui, S. C. Chen and Z. Zuo, Particuology, 2014, 13, 1–26.
15. Z. Zhou, K. L. Dionisio, T. G. Verissimo, A. S. Kerr, B. Coull, R. E. Arku, P. Koutrakis, J. D. Spengler, A. F. Hughes, J. Vallarino, S. Agyei-Mensah and M. Ezzati, Environ. Res. Lett., 2013, 8, 44025.
16. F. I. Khan and A. Kr. Ghoshal, J. Loss Prev. Process Ind., 2000, 13, 527–545.
17. D. Sinclair, J. Aerosol Sci., 1976, 7, 175–179.
18. C. Liu, P. Hsu, H. Lee, M. Ye, G. Zheng, N. Liu, W. Li and Y. Cui, Nat. Commun., 2015, 6, 6205.
19. J. Li, F. Gao, L. Q. Liu and Z. Zhang, eXPRESS Polym. Lett., 2013, 7, 683–689.
20. S. Sundarrajan, K. Luck, S. Huat and S. Ramakrishna, Procedia Eng., 2014, 75, 159–163.
21. H. Souzandeh, K. S. Johnson, Y. Wang, K. Bhamidipaty and W.-H. Zhong, ACS Appl. Mater. Interfaces, 2016, 8, 20023–20031.
22. K. Desai, K. Kit, J. Li, P. Michael Davidson, S. Zivanovic and H. Meyer, Polymer, 2009, 50, 3661–3669.
23. G. Lang, S. Jokisch and T. Scheibel, J. Visualized Exp., 2013, 75, 50492.
24. G. a. Di Lullo, S. M. Sweeney, J. Körkkö, L. Ala-Kokko and J. D. San Antonio, J. Biol. Chem., 2002, 277, 4223–4231.
25. K. S. Rho, L. Jeong, G. Lee, B. M. Seo, Y. J. Park, S. D. Hong, S. Roh, J. J. Cho, W. H. Park and B. M. Min, Biomaterials, 2006, 27, 1452–1461.
26. S. Panzavolta, M. Gioffrè, M. L. Focarete, C. Gualandi, L. Foroni and A. Bigi, Acta Biomater., 2011, 7, 1702–1709.
27. L. Ghasemi-mobarakeh, M. P. Prabhakaran and M. Morshed, Biomaterials, 2008, 29, 4532–4539.
28. Z.-M. Huang, Y. Zhang, S. Ramakrishna and C. Lim, Polymer, 2004, 45, 5361–5368.
29. S. Gu, Z. Wang, J. Ren and C. Zhang, Mater. Sci. Eng., C, 2009, 29, 1822–1828.
30. H. W. Kim, H. S. Yu and H. H. Lee, J. Biomed. Mater. Res., Part A, 2008, 87, 25–32.
31. C. Seok, D. Hyun, K. Don, K. Hoon, I. Chul and Y. Hwan, Polymer, 2005, 46, 5094–5102.
32. N. Okutan, P. Terzi and F. Altay, Food Hydrocolloids, 2014, 39, 19–26.
33. C. Wright, TrAC, Trends Anal. Chem., 2015, 66, 118–127.
34. Specification for HEPA Filters used by DOE Contractors, DOE Technical Standard DOE-STD-3020-2005, U.S. Department of Energy, Washington, D.C., 2005.
35. J. E. Eastoe, Biochem. J., 1955, 61, 589–600.
36. H. Dakin, J. Biol. Chem., 1920, 44, 499–529.
37. S. Wang and G. Zhao, Mater. Lett., 2012, 79, 14–17.
38. K. W. Lee and B. Y. H. Liu, Aerosol Sci. Technol., 1982, 1, 147–161.
39. A. Podgórski, A. Bałazy and L. Gradoń, Chem. Eng. Sci., 2006, 61, 6804–6815.
40. B. Maze, H. Vahedi Tafreshi, Q. Wang and B. Pourdeyhimi, J. Aerosol Sci., 2007, 38, 550–571.
41. R. C. Brown, Air Filtration, Pergamon Press, Oxford, U.K., 1993.
42. J. Wang, S. C. Kim and D. Y. H. Pui, J. Aerosol Sci., 2008, 39, 323–334.
43. J. Wang, S. C. Kim and D. Y. H. Pui, Aerosol Sci. Technol., 2008, 42, 722–728.
44. S. H. Huang, C. W. Chen, Y. M. Kuo, C. Y. Lai, R. McKay and C. C. Chen, Aerosol Air Qual. Res., 2013, 13, 162–171.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra24512a

---