Hydrogen-rich gas production from disposable COVID-19 mask by steam gasification.

Globally, the demand for masks has increased due to the COVID-19 pandemic, resulting in 490,201 tons of waste masks disposed of per month. Since masks are used in places with a high risk of virus infection, waste masks retain the risk of virus contamination. In this study, a 1 kg/h lab-scale (diameter: 0.114 m, height: 1 m) bubbling fluidized bed gasifier was used for steam gasification (temperature: 800 °C, steam/carbon (S/C) ratio: 1.5) of waste masks. The use of a downstream reactor with activated carbon (AC) for tar cracking and the enhancement of hydrogen production was examined. Steam gasification with AC produces syngas with H2, CO, CH4, and CO2 content of 38.89, 6.40, 21.69, and 7.34 vol%, respectively. The lower heating value of the product gas was 29.66 MJ/Nm3 and the cold gas efficiency was 74.55 %. This study showed that steam gasification can be used for the utilization of waste masks and the production of hydrogen-rich gas for further applications.

volume at 273.15 [K] and 101,325 Pa [m3]

concentration based on volume m3/m3*100 %

concentration based on weight kg/kg*100 %

volumetric flow rate of the product gas [Nm3/min]

input of the feed material [g/min]

LHV of the product gas [kcal/Nm3]

LHV of the feed material [kcal/g]

hydrogen

carbon mono oxide

carbon dioxide

methane

acetylene

ethylene

ethane

propene

propane

Introduction

Due to the COVID-19 pandemic, personal protective equipment (PPE) is extensively used as one of the main tools to protect against the infection [1], [2], [3], [4]. The demand for PPE has increased rapidly, particularly because of the recommendation to wear a mask in public [5], [6]. For instance, Italy, with a population of 60.4 million, is estimated to utilize 1 billion masks per month, and if other countries consume similarly, 7.8 billion people worldwide would consume 129 billion masks per month [7]. This estimate corresponds to approximately 491,201 tons of disposed masks. In Korea, on average, 130 disposable masks are used per person per year, and the total domestic consumption reached 6.7 billion masks in 2020, which corresponds to about 25,512 tons [8].

Masks are extensively used in medical facilities and the general public. Masks disposed of in medical facilities are difficult to recycle because they are considered biohazardous. Masks are made up of heterogeneous polymers (polypropylene (PP), polyurethane (PU), nylon, etc.) and metal compounds; therefore, it is difficult to build a recycling system for them [9]. Hundreds of millions of disposable masks are currently being disposed of in Korea, with about one-third being landfilled and about two-thirds incinerated [8]. However, in landfills, polymers are expected to take 450 years to decompose, and instead of fully decomposing, they break down into microplastics that contaminate groundwater [10]. Incineration releases toxic gases containing substances such as dioxins and furans into the environment [11]. Therefore, it is necessary to investigate alternative utilization methods for the disposal of waste masks to alleviate the harmful effects on the environment. The gasification process can thermally utilize waste masks in an environmentally friendly way while producing heat and combustible gases containing H2, CO, and CnHm [12]. Currently, transition to a hydrogen economy is considered a promising route for decarbonization, and the demand for hydrogen is increasing[13], [14]. This factor makes steam gasification the most effective thermochemical method for the utilization of waste masks and the production of hydrogen [15].

Generally, air, oxygen, and steam are used as gasifying agents in gasification processes, and steam gasification produces abundant hydrogen gas and increases the calorific value of the gas [16], [17], [18]. Samimi et al. [16] used biomass as a feedstock and compared air, steam, and air/steam gasification at 800 °C. The H2 content reached 23.65 % in air, 59.47 % in steam, and 36 % in air/steam gasification. Thus, it was concluded that the highest hydrogen yield can be obtained with steam gasification. Li et al. [19] studied steam gasification using municipal solid waste (MSW) as a feed and catalyst bed at 800 °C. The steam/MSW (S/M) ratio varied in the range from 0 to 2.66, affecting the H2 composition in the syngas. From the experimental results, it was concluded that the H2 concentration reached a maximum at an S/M ratio of 1.33.

Previously, few studies considered the steam gasification of plastics to investigate the effect of process variables on the process output [20], [21], [22], [23]. Wilk and Hofbauer studied the characteristics of steam gasification of various plastics (PE, PP, PE + PS, PE + PET, PE + PP) in a dual fluidized bed gasification pilot plant [20]. To enable a steam-gasification endothermic process with a higher steam/carbon (S/C) ratio, a dual fluidized bed to supply heat was used [24]. They used virgin PP as the feed to represent the main component of masks and olivine as the bed material, the S/C ratio was set to 2.3, and the temperature was set to 850 °C. The composition of the product gas was 34 % H2, 4 % CO, 8 % CO2, 40 % CH4, and the lower heating value (LHV) of the gas was 27.2 MJ/m3.

Only two previous studies considered the use of AC to remove tar from the gas during steam gasification of plastics [24], [25]. Kim et al. [25] conducted air gasification experiments in a two-stage gasifier using mixed plastic waste as feed. When 640 g of AC was added, the amount of tar production was 2.5 times lower than when AC was not used, and it was determined that the amount of H2 production increased twice. Cho et al. [26] also used a two-stage gasifier to investigate the effect of bed material (dolomite) and the effect of AC on the gasification of mixed plastic waste. Dolomite and AC substantially increased the concentration of H2 in the product gas. Han et al. [27], [28] conducted steam gasification experiments for solid refuse fuel (SRF) in various S/C ratio ranges using the same apparatus as used in this study, and air gasification was studied using various temperature ranges. It was found that the steam/carbon(S/C) ratio is the most effective in terms of hydrogen generation and tar removal at 1.5, and the calorific value of product gas and CCE were the highest at 800 ℃ between 600 and 800 ℃.

Waste mask steam gasification is innovative because it enables the eco-friendly utilization of disposable COVID-19 masks that are otherwise difficult to recycle. Additionally, waste mask steam gasification can potentially produce substantial amounts of hydrogen, thus making waste mask gasification suitable for the hydrogen economy, production of energy, and product gas. In this study, a 1 kg/h lab-scale (diameter: 0.114 m, height: 1 m) externally heated bubbling fluidized bed gasifier was used to investigate the characteristics of waste mask steam gasification. Steam gasification experiments were carried out at a steam/carbon ratio (S/C) of 1.5, temperature of 800 °C, and a feeding rate of 4.09–4.66 g/min based on previous study [27].

Experimental

Feed material

The waste masks used in this study consisted of KF94 masks, which can filter out more than 94 % of particles with an average size of 0.4 μm. Each mask consists of a filter, ear straps, and a nose wire, which are separated from each other [9]. Fig. 1 shows the process of preparing the feed material for the experiments. The injection of raw material of mask filter and ear strap into apparatus causes an unstable feeding since the density of raw material is very low. For smooth feeding into the lab-scale 1 kg/h gasifier, filter parts were shredded to approximately 1 cm × 1 cm in size. Shred masks containing filter and strap parts in a weight ratio of 8:1 were pelletized using an externally heated pelletizer. This ratio corresponds to the weight ratio of the mask filters and their straps. The screw-type pelletizer temperature was set to 250 °C, and the RPM of the screw was set to 15–20. The pellets were then ground and sieved to produce a sample with a particle size of 4 mm.

Fig. 1

Feed material preparation process.

Basic data for sample characterization were obtained using proximate analysis (TGA-701, LECO. Co), ultimate analysis (TruSpec, LECO. Co), sulfur analysis (SC-832DR, LECO. Co), and lower heating value (LHV) analysis (AC600, LECO. Co), and X-ray fluorescence (XRF) analysis (ZSX Primus II, Rigaku). The results of the basic analyses are presented in Table 1 . As shown in Table 1, waste masks mostly comprise volatile matter, accounting for 87.81 %. Fixed carbon is the second-largest component, accounting for 10.83 %, followed by the moisture and ash content of 0.63 % and 0.73 %, respectively. Mask pellet proximate analysis was compared with other mono-component polymers, such as PP, PUSRF, and nylon. The mask pellet ash content and fixed carbon content were higher compared to the given examples. Elemental analysis revealed that C, H, O, and N accounted for 72.02 %, 11.32 %, 16.69 %, and 0.61 %, respectively. The C and H contents are similar to those reported in the literature, and the content of O is higher than that reported in the literature. The LHV of the mask pellets is approximately 35.86 MJ/kg, which is similar to the heating value of PP. XRF analysis of ash revealed a high content of TiO2, reaching 44.34 wt%. The remaining constituents included Fe2O3, CaO, SiO2, Al2O3, P2O5, MgO, K2O, and others.

Table 1

Characteristics Analysis results of Mask pellet.

	Mask pellet	PP[29]	PUSRF[30]	Nylon[31]
Proximate analysis (As received) [wt%]
Volatiles	87.81	93.77	82.91	93.7
Moisture	0.63	0.16	1.86	0.1
Ash	0.73	4.45	5.05	5.5
Fixed Carbon	10.83	1.62	10.18	0.7
Ultimate analysis (Dry basis) [wt%]
C	70.99	83.65	71.29	78.75
H	11.48	14.27	6.47	12.41
O	16.92	0.15	15.52	8.71
N	0.61	0.67	6.72	0.12
S	< 0.01	0.1	ND	0.02
Cl	0	1.16	< 0.01	-
LHV [MJ/kg](As received)	35.86	45.20	24.16-27.99	40.42

Bed material and AC

In this study, silica sand was used as the bed material. The true density (ρs) of the sand was measured to be 2527 kg/m3, the bulk density (ρb) was 1548 kg/m3, and the average particle size was approximately 333 μm.

Granulated AC originating from palm tree residue was supplied by Kaya Carbon (Korea). To obtain basic data on AC, proximate analysis, ultimate analysis, BET, and Iodine number analysis were performed, the corresponding analyses results are shown in Table 2 . The fixed carbon content in AC was 86.25 %, with respective moisture and ash content of 6.08 %, and 2.67 % on an as-received basis. Ultimate analysis revealed that C content accounted for 96.39 %, and the remaining constituents, H, O, N, and S were 0.21 %, 0.56 %, 0.09 %, and 0.08 %, respectively. Brunauer-Emmett-Teller (BET) surface area analysis indicated that the surface area was 506.62 m2/g, which is within the range of the surface area of AC derived from agricultural residues (300–1100 m2/g) [32]. The iodine number is widely used as an indicator of the adsorption capacity of AC, and was calculated from the iodine adsorption capacity test based on the Korean Industrial Standards (KS M11802) [33].

Table 2

Characterization of AC.

Proximate analysis (As received, [wt%])
Moisture	6.08
Volatiles	5.00
Fixed Carbon	86.25
Ash	2.67
Ultimate analysis (Dry basis, [wt%])
C	96.39
H	0.21
O	0.56
N	0.09
S	0.08
BET analysis
Surface area [m2/g]	506.62
Total pore volume [cm3/g]	0.21
Average pore diameter [nm]	1.69
Iodine number [mg/g]	642.96

Experimental procedure

In this study, mask steam gasification was performed in a BFB reactor, as shown in Fig. 2 . The experimental apparatus consisted of an externally heated BFB reactor, preheater, cyclone, AC bed reactor, and condenser. The BFB reactor (SUS 316S stainless steel, diameter: 0.114 m, height: 1 m) was filled with sand forming a 0.2 m high bed. The fuel injection system consists of one silo and two screw feeders, upper and lower, installed consecutively with a window between them to check whether the feed injection was smooth. To prevent the backflow of sand and hot gas during the experiment, and for smooth feed input, 5 L/min of N2 was injected between the silo and screw feeder. During preheating before the experiment, 10 L/min of N2 was injected into the reactor, and when the reactor was heated to the desired temperature, the N2 flow was increased to 15 L/min, which corresponds to the fluidizing number (Ug/Umf) of 2.3. After stabilizing the pressure, fuel feeding was started by setting the rotation of the screw feeders. The feeding rate was adjusted using the upper screw feeder rotation rate, and the lower screw feeder rotation was set to a maximum to ensure uniform and constant feeding. When the NDIR gas analyzer indicated changes in gas composition, steam was injected through the peristaltic pump (77800–60, Masterflex) at a rate of 8.5 L/min, and N2 feed into the pump was adjusted to 6.5 L/min. Steam and N2 were preheated to 300 °C using a preheater, and this temperature was maintained using an external heating band between the preheater and the reactor. A preheated nitrogen-steam mixture was introduced into the reactor through an air distributor at the bottom of the reactor, thus fluidizing the bed and reacting with the fuel. The product gas from the reactor was then passed through the cyclone to remove the char and dust. Then, the gas was passed through the AC reactor to remove the tar. The AC bed was externally heated to 800 °C and filled with 800 g of AC, which corresponds to 70 % of the AC bed capacity. The gas ducts and cyclone temperature were maintained at 300 °C using heating bands to prevent tar from condensing. After the AC bed, gas was passed through the condenser, which removed tar and moisture. The condenser was cooled using an anti-freeze liquid that cooled to −10 °C and recirculated through the chiller. After the condenser, the gas was analyzed using an NDIR (Advanced Optima, Hartmann & Braun Co.) gas analyzer, indicating CO, CO2, CH4, and H2 concentrations in a real-time. Additionally, gas samples were taken three times during each of the experiments using Tedlar bags. The remaining gas was pumped into the vent line. Tar sampling was conducted by passing the gas through impinger bottles filled with isopropyl alcohol (IPA). The sampling points are after the cyclone (SP1), after the AC bed (SP2), and after the condenser (SP3) as shown in Fig. 2. After the experiments, the solution containing tar was collected and filtered using a filter (8 μm) to determine the amount of entrained particle. The filtered solution was evaporated using a rotary evaporator (N-1210B, Eyela) to obtain the tar sample.

Fig. 2

Schematic diagram of the fluidized bed reactor.

The experimental conditions are presented in Table 3 . Case 1 represents an experiment with steam, without fuel feeding, and with AC filled in the AC reactor. Case 2 denotes an experiment with fuel and steam feeding but without AC. Case 3 is an experiment with fuel and steam feeding and AC in the AC bed. The fuel feeding rate was 4.17 g/min, and the steam was injected at rate of 8.5 L/min (=6.83 g/min at room temperature).

Table 3

Experimental condition of mask pellet steam gasification.

	Case 1	Case 2	Case 3
BFB temperature [℃]	800	800	800
AC bed temperature [℃]	800	800	800
Feeding rate [g/min]	–	4.09	4.46
Steam mass flow rate [g/min]	6.89	6.77	6.83
S/C ratio [-]	–	1.58	1.46
AC [g]	800	–	800

Product analysis

Steam gasification products include product gas, char, tar, and condensate liquid. The product gas was analyzed in a real-time using an online NDIR gas analyzer, which measured the CO, CO2, CH4, and H2 concentrations. During the experiment, gas samples were collected three times using Tedlar bags (232 series, SKC Inc.). The concentrations of the product gas constituents in each sample, including CO, CO2, CH4, H2, and C 2-C3 were analyzed using gas chromatography with a thermal conductivity detector (GC-TCD). The average concentrations were calculated based on the GC-TCD analysis results, and the amount of gas was determined by calculating the weight of each component using gas densities. The amount of char was determined by measuring the weight of char inside the BFB reactor, char captured by the cyclone, and adding the particulate matter filtered from the tar sample.

Tars contained in gases from the sampling (SP1, SP2, SP3) are dissolved in the IPA solvent filled in the impingers, and the gas flow was measured using three wet gas meters (W-NK-1A, Shinagawa). The tar and IPA solution samples, obtained from the impinger bottles, were filtered. Then, IPA in the filtered solution was evaporated at 60 °C and pressure below atmospheric pressure, thus separating IPA from tar. After weighing, the relative content of specific components of tar was determined using a gas chromatography-mass spectrometry (GCMS) (QP2010Plus, Shimadzu) analysis instrument. The amount of condensate liquid was measured before and after the experiment with the condenser, and the difference between the two was determined as the weight of the condensate liquid. Condensate liquid is mostly condensed water and contains trace amounts of tar. Other is defined as a value that subtracted from the input to output.

The AC amount was weighed before and after the experiment, and the AC specific surface area, pore area, and pore diameter were determined by BET (ASAP 2020 PLUS, Micromeritics) analysis. The surface characteristics of the AC samples were analyzed using scanning electron microscopy (SEM) (S-4800, HITACHI). During the experiment, the internal temperature, pressure, and flow rate of the experimental apparatus were controlled through a human–machine interface (HMI), measured in a real-time, and stored on a computer. The performance of mask steam gasification was evaluated by calculating the LHV, carbon conversion efficiency (CCE), and cold gas efficiency (CGE) for the product gas. CCE is defined as the ratio of the carbon contained in the product gas to the carbon contained in the gasified fuel, which can be expressed as Eq. (1).

The higher the CCE, the higher is the efficiency of the gasifier, as expressed in Eq. (1). CGE represents the amount of heating value of product gas by gasification with respect to the amount of heating value contained in the fuel as,where is the volumetric flow rate of the product gas (Nm3/min), denotes the LHV of the product gas [kcal/Nm3], is the input of the feed material [g/min], and represents the LHV of the feed material (kcal/g).

Results and discussion

Reaction characteristics

Fig. 3 shows the changes in the CO, CO2, CH4, and H2 contents of the product gas during the experiments which were analyzed using NDIR gas analyzer. Fig. 3(a) shows the reaction conditions with steam injection and AC bed and without fuel feeding (Case 1). Fig. 3(b) shows the experimental conditions with waste mask pellet feeding and steam injection without AC bed (Case 2). Fig. 3(c) shows the experimental condition with mask pellet feeding, steam injection, and AC bed (Case 3). Table 4 lists the reaction equations that can occur during steam gasification [23], [25], [34], [35], [36], [37]. The heterogeneous carbon-steam reaction rates are proportional to the surface area available for the reaction. The homogeneous steam reforming reaction rates are proportional to the reactant concentrations [38]. Gas-solid phase heterogeneous reactions are significantly slower than reactions such as evaporation and devolatilization [39].

Fig. 3

Temporal distributions of volumetric concentrations of CO, CO2, CH4, and H2 in product gas at reaction conditions of (a) Case1, (b) Case 2, and (c) Case 3.

Table 4

Possible reaction during steam gasification.

Number	Reactions	Type
(R1)	C(S) + H2O ↔ CO + H2	Steam gasification reaction (SGR)
(R2)	C(S) + 2H2 ↔ CH4	Methane production reaction
(R3)	C(S) + CO2 ↔ 2CO	Boudouard reaction
(R4)	CO + H2O ↔ CO2 + H2	Water gas-shift reaction (WGSR)
(R5)	CH4 + H2O ↔ CO + 3H2	Steam reforming reaction
(R6)	CH4 + 2H2O ↔ CO2 + 4H2
(R7)	CnHm + nH2O ↔ nCO + (n + m/2)H2
(R8)	CnHm + n/2H2O ↔ n/2CO + (m-n)H2 + n/2CH4
(R9)	CH4 + CO2 ↔ 2CO + 2H2	CO2 reforming reaction
(R10)	CnHm + nCO2 ↔ 2nCO + (m/2)H2
(R11)	CnHm + n/4CO2 ↔ n/2CO + (m-3n/2)H2 + (3n/4)CH4
(R12)	C7H8 + H2 → C6H6 + CH4	H2 reforming reaction
(R13)	C2H6 → C2H4 + H2	Hydrocarbon pyrolysis
(R14)	C2H4 → C2H2 + H2
(R15)	C2H2 → C + CH4
(R16)	C3H8 → C2H4 + CH4

In Case 1, the composition of the gas generated by the reaction of AC and steam is the highest for H2 at about 7.5 vol%, CO and CO2 for about 2 vol%, and CH4 for about 0.1 vol%. First, CO and H2 are produced, which proceeds through R1, a reaction between steam and AC. As CO and H2 atmospheres are formed in the atmosphere in the reactor, the reaction of R4 proceeds to produce CO2. The reaction of R3, which is a reaction between AC and CO2, proceeds with the formation of an atmosphere of CO. At this time, in R3, the forward reaction occurs predominantly at a temperature of more than 700 ℃, and the reaction proceeds in the direction of CO production [40]. According to a study by Chen and Lin [34], trace amount of CH4 is produced by R2. They studied the reaction characteristics of H2O and CO2 in AC and determined that AC acts as a catalyst to induce R2 and R4 reactions. It is expected that the remaining reactions (R5, R6, and R9) proceed continuously according to the generation of CH4.

In Case 2, CH4 and H2 in the product gas were produced at about 6 vol% and 4 vol%, and 0.5 vol% of CO and 0.2 vol% of CO2 were generated, and hydrocarbons that could not be detected by NDIR gas analyzer were generated which will be explained in section 3.3. H2, CH4, and shorter chain hydrocarbon are produced first by thermal cracking through hydrocarbon pyrolysis (R13-R16). After that, the CO atmosphere is formed with R7 and R8 by the reaction of hydrocarbon and steam. After that, as in Case 1, the reaction of R4 generates CO2 and proceeds to the CO2 reforming reaction (R9-11).

In Case 3, the average composition of H2 shows 8.4 vol%, 4.46 vol% of CH4, and CO and CO2 are 1.39 and 1.56 vol%, respectively. In the early stage of Case 3, the thermal decomposition reaction of tar and tar cracking reaction in AC occur actively to generate H2, CH4, and CO. Compared to Case 2 at the beginning stage, the high CO gas comes from the reaction with AC [41]. When the steam is injected into the gasifier, there is a time gap where the actual steam is reacted with fuel since the passage of steam is relatively long. When the steam reacts with fuel, H2 shows the highest value as can be seen in Case 1. Gas composition, excluding methane, has been shown to decrease over time as the specific surface area decreases due to soot/carbon deposition from tar catalyst removal from AC, leading to deactivation [26], [42]. Cho et al. [26] conducted a study on the air gasification of mixed plastic wastes using AC. It was found that the composition of H2, CO, CH4, and CO2 and the composition of tar in the production gas decreased according to the operation time, and the composition of tar increased. It was confirmed that the H2 content decreased while the activity of the tar cracking reaction decreased due to AC deactivation.

Mass balance

Fig. 4, Fig. 5 show the mass balance of steam gasification in the absence and presence of AC, respectively. The actual feeding rate, steam injection rate, and loss of AC rate were determined by measuring the weight of the injected mask pellet and steam before and after the experiment. The gas composition was determined using GC-TCD analysis, and the gas composition was converted to weight using the gas densities at room temperature. The total weight of the gas was determined by summing the yields of each gas component. The amount of char was determined by adding the amount of char particulate matter filtered from the tar samples and char recovered from the BFB reactor and cyclone. The weight of the tar was determined by the amount of tar obtained from the three SPs. The condensate liquid was determined by measuring the weight of the bottom of the condenser before and after the experiment. Others indicates the remaining char, tar, liquid in the connection parts between reactor to AC bed, AC bed to condenser, passing line to vent line, etc.

Fig. 4

Mass balance of steam gasification absence of AC.

Fig. 5

Mass balance of steam gasification presence of AC.

The main output material for steam gasification is the condensate liquid. It comes from unreacted steam, steam generated from reactions, and a marginal amount by tar. Fig. 4, Fig. 5 demonstrate that the output steam accounts for 76–78 % of the injected amount of mask pellet and steam. The others account more than 22–30 % since the remaining char, tar, liquid in the connection parts and vent line is not marginal. Steam gasification produces gas with a high yield of C1-C3 light hydrocarbons, reaching 2.78 g/min, which corresponds to 0.68 g/gfuel. Gasification with downstream AC substantially reduces the C1-C3 hydrocarbons yield to 2.11 g/min, which corresponds to 0.47 g/gfuel. Among C1-C3 hydrocarbons, in both cases, ethylene and propylene had notably high yields. On the other hand, the use of AC increases H2 yield, from 0.04 g/min to 0.16 g/min, which corresponds to an increase from 0.01 g/gfuel to 3.6 g/gfuel. The increase in H2 yield can be attributed to the cleavage of C—H bonds in hydrocarbons, increasing unsaturated hydrocarbon yields, such as acetylene, and a decrease in hydrocarbons with a higher degree of saturation. Cleavage of C—H bonds also suggests that this affects the tar components, increasing the relative percentage of PAHs and naphthalene.

Fig. 6 demonstrates the yield of gas, char, and tar yield ignoring the condensate liquid. The amount of gas is the largest, and the gas generation rate is 87.8 % in Case 2 and 89.7 % in Case 3. Char is 1.2 % in Case 2 and 1.3 % in Case 3, with the lowest amount of char. Tar was generated in 11 % of Case 2 and 9 % of Case 3, and less tar was generated in Case 3 where AC was present. Tar is a major factor affecting the flow of the process, and tar removal is essential. The use of AC can reduce the amount of tar that interferes with whole process.

Fig. 6

Product yield of mask steam gasification.

Product gas analysis

Fig. 7 shows the composition of the product gas, excluding nitrogen. In the absence of AC (Case 2), the composition of product gas was H2 15.90 vol%, CO 3.06 vol%, CH4 30.74 vol%, CO2 0.45 vol%, hydrocarbons (C 2-C3) 49.85 vol%. Whereas product gas composition in Case 3 is 38.89, 6.40, 21.69, 7.34, and 25.69 vol% respectively for H2, CO, CH4, CO2, and hydrocarbons (C 2-C3).

Fig. 7

Product gas composition (a) H2, CO, CO2, CH4, (b) C2H2, C2H4, C2H6, C3H6, C3H8.

In Case 3, the product gas contained 2.4 times higher H2 and 2.1 times higher CO compared to Case 2. In Case 3, AC affected the SGR and tar cracking, substantially increasing the CO and H2 components. The composition of CO2 was approximately 16 times higher in Case 3 than in Case 2, suggesting that an increased concentration of CO increases the reaction rate of both WGSR and Boudouard reactions, thus raising the CO2 content. It can also be seen that in Case 2 the content of C2H4 and C3H6 are 1.4 times and 1.8 times higher in Case 2, respectively, suggesting that AC induces cleavage of C—H and C—C bonds and corresponding thermal cracking reactions. This is confirmed by the 1.8 times higher C2H2 content and slightly lower C2H6 content in Case 3 compared to Case 2. The CH4 content was lower in Case 3, suggesting intensified combined steam and dry reforming of methane [43].

In Case 1, fuel is not fed into the gasifier; thus, gas is generated by the reaction of steam and AC; the mass flow rate of the product gas according to the conditions is shown in Table 5 . The reduction of AC weight in Case 1 and Case 3 were 132.07 g and 79.58 g, respectively. The reduction of AC weight is associated with AC drying and devolatilization and with reaction of AC with steam [44]. Substantially higher loss of AC weight in Case 1 was associated with higher amount of steam available for reaction with AC [26]. In contrary, in Case 3, substantial part of steam participated in reactions with the waste mask decomposition products, thus less unreacted steam is available for reactions with AC. Additionally, soot forming while tar cracks on the surface of AC reduces the available specific surface area, blocks the pores, and reduces available surface functional groups, thus also reducing carbon and steam reaction. The reaction of AC and steam in Case 1 produces 0.12 g/min of H2. When waste masks are supplied in Case, 0.16 g/min of H2 is produced despite the lower AC consumption. It can be assumed that a substantial amount of H2 can be produced under the influence of the AC. CO and CO2 are decreased in Case 3, suggesting that SGR and WGSR rates are reduced. A small amount of CH4 was generated in Case 1 through the methane production reaction. Owing to the thermal decomposition of polyolefins in waste masks, a substantial amount of C1-C3 hydrocarbons was generated in Case 3. The LHV of the product gas in Case 2 was 47.26 MJ/Nm3, and the LHV was 29.66 MJ/Nm3 in Case 3.

Table 5

Mass flow rate of product gas under the influence of AC.

	Case 1	Case 3
AC weight loss [g]	132.07	79.58
Injected rate of steam [g/min]	6.89	6.83
H2 [g/min]	0.12	0.16
CO [g/min]	0.50	0.37
CH4 [g/min]	0.01	0.67
CO2 [g/min]	1.02	0.65
HCs (C2-C3) [g/min]	0.01	1.43

Fig. 8 shows the CCE and CGE of steam gasification without AC (Case 2) and with AC (Case 3) in the AC reactor. In Case 2, the CCE and CGE reached respective values of 82.61 % and 95.61 %. In Case 3, the CCE was 63.06 %, and the CGE was 74.55 %. The reason that LHV, CCE, and CGE are higher in Case 2 than in Case 3 can be attributed to the higher content of hydrocarbons that are not cracked in the absence of AC. Fig. 7 shows a higher content of hydrocarbon gases in Case 2 than in Case 3, affecting the LHV.

Fig. 8

(a)CCE, (b)CGE of mask steam gasification according to the presence or absence of AC.

Tar analysis

Tar is an undesirable by-product of solid fuels – steam gasification. When the temperature of the product gas is lowered after the gasifier, tar is condensed and tar deposits are formed, which causes problems in the gas conditioning and in downstream processes. AC was used to lower the tar concentration in the product gas. Fig. 9 shows the weight of tar per volume of gas collected from the three SPs: after cyclone, after the AC bed, and after the condenser. Both Case 2 and Case 3 had similar tar concentrations obtained after the cyclone. In Case 2, the concentration gradually decreased through the device owing to thermal cracking, and the tar removal rate before and after the empty AC bed vessel was 25 %. In Case 3, when compared with before and after the AC bed, the tar concentration dropped sharply, and the tar removal rate reached 91 % when the tar content was compared before and after the AC. In Case 3, the tar content after the condenser increased when compared with the tar content after the AC bed due to the cooling of gas to −10 °C in the condenser. While high temperatures lead to thermal cracking of tars, low temperatures have previously been observed to cause repolymerization and oxygenation of hydrocarbon compounds, increasing the tar content of the product gas. Similarly, the increase in tar content after the condenser can be attributed to the formation of oxygenated compounds [45]. The following GC–MS analysis was performed to investigate relative percentages of tar components.

Fig. 9

Tar concentration in product gas according to absence (Case 2) and presence (Case 3) of AC and after Cyclone, AC bed and Condenser.

Fig. 10 (a) shows the relative percentage of tar components, and Fig. 10 (b) shows the GCMS analysis results of the study conducted by Wilk and Hofbauer [20], who classified tar components by dividing the tar obtained from PP steam gasification into groups including phenols, furans, aromatics, naphthalene, and PAHs. In this study, aldehydes and ketones, esters, and alcohol, PAHs, and aromatic compounds are added to their tar classification items. More than half of the tar in the mask steam gasification is naphthalene, and the rest are other PAHs and aromatic compounds. The phenols, furans, esters, and naphthalene groups were hardly detected and trace amounts of aldehydes and ketones group and alcohol group were detected. As shown in Fig. 10(b), it can be deduced that furans are not detected because the reactions occur at high temperatures thus promoting their decomposition.

Fig. 10

Result of GCMS analysis (a) tar in Case 3, (b) tar in PP steam gasification [20].

Resultant activated carbon analysis

Table 6 shows the results of the BET analysis of the surface area, pore-volume, and pore size of AC. The raw AC has a surface area of 506.62 m2/g, with a pore volume of 0.21 cm3/g, and a pore size was 1.69 nm. In Case 1 absence of fuel, the surface area of ​​AC is 649.26 m2/g, the pore volume is 0.06 cm3/g, and the pore size was 0.52 nm. Compared with raw AC, the surface area of the AC in Case 1 increased. Since AC is manufactured using steam, the activation site of AC increases in reaction with steam and AC, increasing the surface area [46]. Under the condition of Case 3 when fuel is gasified, the surface area of AC after the experiment is 325.76 m2/g, the pore volume is 0.05 cm3/g, and the pore size is 1.22 nm. The AC surface area of Case 3 reduced to 64.3% of the raw AC surface area. This is because the fuel produces tar through gasification, and the AC adsorbs and decomposes tar. In this process, tar blocks the pores of the AC, and the surface area decreased.

Table 6

BET analysis of AC according to conditions.

	Raw	Case 1	Case 3
Surface Area [m2/g]	506.62	649.26	325.76
Pore Volume [cm3/g]	0.21	0.06	0.05
Pore size [nm]	1.69	0.52	1.22

Fig. 11 shows images of the surface morphology of the AC with SEM. Fig. 11(a) and (b) show the images viewed at 10 k and 20 k magnifications of raw AC, Fig. 11(c,d) shows the images viewed at 10 k and 20 k of AC in Case 1, and Fig. 11(e,f) shows the images of AC of Case 3 as 10 k and 20 k magnification. Raw AC had a smooth surface and some large pores, Case 1 AC had many small pores on the surface, and Case 3 AC had many small particles on the surface. The reason for the presence of numerous pores on the surface of Fig. 11(c,d) is due to increased activated site is from steam, and the small particles on the surface of Fig. 11 (e,f) show that tar and dust are attached to the surface of AC, thus obstructing the pores.

Fig. 11

SEM images of surface morphology of AC. ((a) raw AC (at 10 k × magnification), (b) raw AC (at 20 k × magnification), (c) AC in Case 1 (at 10 k × magnification), (d) AC in Case 1 (at 20 k × magnification), (e) AC in Case 3 (at 10 k × magnification), (f) AC in Case 3 (at 20 k × magnification)).

Conclusions

The steam gasification characteristics of disposable masks used during the COVID-19 pandemic were determined using a fluidized bed gasifier operated at 800 °C with and without AC in the downstream reactor. Characteristics include process mass balance, product gas composition, tar content, tar composition, the effect of AC, and AC characteristics after the experiments.

Waste mask gasification produces gas with substantial LHV, reaching 47.26 MJ/Nm3 when AC is not used, and 29.66 MJ/Nm3 in presence of AC. Waste mask steam gasification produces gas with a high yield of C1-C3 hydrocarbons, reaching 0.68 g/gfuel without AC. When AC is used C1-C3 hydrocarbons yield is reduced to 0.47 g/gfuel due to catalytic decomposition. The use of AC substantially increases the hydrogen yield by 2.4 times, when compared to the process without AC. The process with AC also had better tar cracking performance, demonstrating a 1.8 times higher removal rate than in the absence of AC. When AC was used, the composition of the product gas was as follows: H2 38.89 vol%, CO 6.4 vol%, CH4 21.69 vol%, CO2 7.34 vol%, HCs (C2-C3) 25.69 vol%. The use of AC decreased CCE from 82.61 % to 63.06 %, and CGE decreased from 95.61 % to 74.55 % when compared to the case without AC. The use of AC substantially reduced the tar content, removing 87.1 % of tar compared to non-AC case. The surface area and pore volume of AC decreased as the tar blocks the surface of AC. The regeneration of AC is needed for stable and long-term operation of tar cracking reaction of AC.

CRediT authorship contribution statement

Ji young Nam: Conceptualization, Investigation, Data curation, Writing – original draft. Tae Ryeon Lee: Data curation. Diyar Tokmurzin: Conceptualization, Methodology, Visualization, Data curation, Writing – original draft, Writing – review & editing. Sung Jin Park: Data curation, Visualization. Ho Won Ra: Investigation, Funding acquisition. Sang Jun Yoon: Investigation, Data curation, Visualization. Tae-Young Mun: Investigation, Data curation, Visualization. Sung Min Yoon: Formal analysis, Data curation, Visualization. Ji Hong Moon: Investigation, Data curation. Jae Goo Lee: Formal analysis, Funding acquisition. Dong Hyun Lee: Formal analysis, Funding acquisition, Writing – review & editing. Myung Won Seo: Conceptualization, Validation, Funding acquisition, Writing – review & editing, Supervision.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.