Activated carbon derived from hydrothermal treatment of sucrose and its air filtration application

Abstract

As air pollution problems become increasingly serious and the environmental awareness continues to rise, a source of cheap and simple air filtration active material is also required. Here, we demonstrated the use of hydrothermal carbonization-derived activated carbon for such an application. By doing so, we hope to develop a biodegradable filtering material which can combine filtering capability with biodegradable functionalities. To this end, an air filtration paper using activated carbon and pulp as materials was synthesized. Activated carbon was prepared by hydrothermal carbonization of sucrose and further activation. Hydrochar synthesis parameters including reaction time, reaction temperature, sucrose concentration, pH effect of hydrothermal solution, and nitrogen doping were investigated. The air filtration application was demonstrated using activated carbon obtained from a dramatic chemical reaction with KOH at 800 °C as the biodegradable active filler material. The morphology of hydrochar and activated carbon was characterized by scanning electron microscopy (SEM). Gas chromatography-mass spectrometry (GC-MS) was used to investigate the mechanism of hydrochar formation with different pH of hydrothermal solution. Fourier-transform infrared spectrometer (FTIR) was used to confirm activation. The results of elemental analysis showed evidence of nitrogen doping. Surface area and pore size analysis results allow for comparison of the degree of activation. The highest surface area of activated sample obtained in this study is up to 3026 m² g⁻¹ and the filtering abilities of our samples were measured by TSI model 8130 testing method with results of filtering efficiency achieving a quarter of that of the commercial air filter product.

---

1. Introduction

The development of carbon materials has always been an attractive issue due to their excellent advantages such as less damage to the environment, low cost, and relative ease of commercialization. Hydrothermal carbonization (HTC) is a synthesis method that transfers carbon precursors into useful carbon materials under mild synthesis conditions. The basic process involves pre-configured solutions that are placed into sealed Teflon-lined stainless steel autoclaves and heated to a specified temperature for the duration of several hours.

The resulting hydrochar can be obtained from the bottom of the autoclave that display controllable pore size, morphology, and specific chemical functionalities. The features of hydrochar are usually controlled with temperature, concentration of carbon precursor, reaction time and acidic or basic catalysis.¹ Carbohydrates, such as sucrose,^2,3 glucose,⁴ cellulose,^5–7 and natural biomass are the usual carbon precursors used in such a processes. On the other hand, various biomass materials have been used in HTC synthesis, including bagasse,⁸ coconut shell,⁹ banana,¹⁰ corn cob,¹¹ coffee grounds, peanut hull, hemp,¹² pine needles and oak.¹³ Applications for HTC-synthesized hydrochar include fuel, electric capacitor, heavy metal adsorption, and activated carbon synthesis.¹⁴

Activated carbon can be obtained through activation of hydrochar precursor. A highly porous material with large specific surface area will be formed and the resulting performance as adsorbents will be dramatically enhanced by this process. The porosity and pore size of activated carbon depends on activation temperature, time, inert gas flow rate during activation, and species of activated agent. Functions of such highly porous carbon material include adsorbents,^15,16 capacitor electrodes,^12,17,18 electrocatalyst,¹⁹ hydrogen storage,²⁰ and carbon dioxide capture.²¹ In addition, modifying activated carbon by nitrogen-doping has been demonstrated to alter adsorption performance of the material.²² Although there are many applications of activated carbon, there have been very few studies that emphasize its biodegradable functions in air filtration, with the possible exception of commercial surgery masks that are popular in Taiwan.

Air pollution particles form naturally from volcanic eruptions, dust storms, forest fires and from emissions of industrial, traffic, agricultural, and residential origins. To purify the air, there are already many products on the markets nowadays, such as electrostatic filters, HEPA filter, activated carbon filters. However, the abovementioned products are not all manufactured in an eco-friendly fashion, leading to further costs to the environment and resources. In order to improve this problem, we designed an air filtration paper, which not only purifies air pollution particles but is also an environmentally friendly green product. The air filtration paper is made of all biodegradable components, including pulp and activated carbon.

Studies were first performed on the underlying active material, leading to a comprehensive investigation of the effects of pH and nitrogen doping on the active material. pH-Modified and nitrogen source added in the hydrothermal solution make hydrochars with distinct pore size, morphology, and surface chemical functionalities, which were compared to hydrochar obtained from aqueous conditions. Activated carbon from the pH-modified and nitrogen-doped hydrochar were further produced by an activation process, and different pore characteristics were determined.

2. Experimental

2.1 Chemical compounds

The materials we used in this research include sucrose (CAS number: 57-50-1, Showa, 99.5%), sulfuric acid (CAS number: 7664-93-9, Showa, 97%), sodium hydroxide (CAS number: 1310-73-2, Showa, 97%), ammonium sulfate (CAS number: 7783-20-2, Showa, 99.5%) and potassium hydroxide semi pellets (CAS number: 1310-58-3, Aencore, 85%).

2.2 Experimental procedure

In this work, experiments are divided into three parts: (i) the effect of time, concentration and temperature on hydrochar and activated carbon synthesis, (ii) the effect of pH and nitrogen doping on hydrochar and activated carbon synthesis, and (iii) the application of air filtration with resulting activated carbon.

2.2.1 Effect of time, concentration and temperature on hydrochar and activated carbon synthesis.
Hydrothermal treatment. 20 ml of sucrose at varying concentrations in aqueous solution were placed in a 45 ml Teflon-lined stainless steel autoclave and heated to chosen synthesis temperatures for several hours in a conventional oven. The autoclave was naturally cooled to room temperature, and the resulting solid product was washed with deionized water and dried in an oven at 70 °C overnight.
Activation process. Hydrochar with fused (C2) and separated (C6) structures were selected for further activation because they are the most representative samples of eight hydrochars based on their morphology (see Section 3.1.1 for details). The hydrochar was mixed with KOH in a mass ratio of 1 : 4 and 1 : 2 in a ceramic mortar and the mixture was heated to 800 °C for 1 hour in a tubular furnace under nitrogen flow of 200 cm³ min⁻¹. After cool down, still under N₂, the activated carbon was then washed with distilled water until neutral pH was reached and dried in an oven overnight.

2.2.2 Effect of pH and nitrogen doping on hydrochar and activated carbon synthesis.
Hydrothermal treatment. 20 ml of 0.5 M sucrose aqueous solution with addition of sulfuric acid for acidic conditions, sodium hydroxide for basic conditions, or ammonium sulfate for nitrogen doping was placed in a 45 ml Teflon-lined stainless steel autoclave and heated to 200 °C for 4 hours in an oven. The autoclave was naturally cooled to room temperature, and the resulting solid product was washed with deionized water until neutral pH and dried in an oven overnight.
Activation process. Sucrose-derived hydrochar was mixed with KOH in a mass ratio of 1 : 4 in a ceramic mortar and the mixture was heated to 800 °C for 1 hour in a tubular furnace under nitrogen flow of 200 cm³ min⁻¹. After cool down while still under N₂, the activated carbon was then washed with deionized water until neutral pH was reached and dried in an oven overnight.

2.2.3 Application of air filtration with resulting activated carbon. 1 g of activated carbon and 4 g of pulp were mixed with 1.5 liters of water by a physical method, further assisted by rigorous stirring with a homogenizer. Then the mixture was distributed uniformly in a container containing water. By using a traditional papermaking process, we could obtain a wet preliminary paper. The preliminary product was then heated to 70 °C for 4 hours to remove water content. Finally, dry pieces of paper with activated carbon uniformly distributed in it was obtained, which we called the air filtration paper.

3. Results and discussion

3.1 Effect of time, concentration and temperature on hydrochar and activated carbon synthesis

3.1.1 Morphology and yield. Scanning electron microscopy was performed at the National Central University Precision Instruments Center. Fig. 1 shows SEM images of our sucrose-derived hydrochar with different reaction parameters. It was observed that carbon spheres (CSs) with different morphologies result due to distinct synthesis parameters. However, we can roughly categorize them into two types: fused (Fig. 1a–e and g) and separated (Fig. 1f and h). Notation for samples created under varying conditions are listed in Table 1. We observed that reaction time affected particle size, yield and degree of fusing. Particle size of CSs was measured by MB ruler program.²³ From various SEM images of the same sample, we selected a representative area, then measured and averaged the diameter of at least thirty carbon spheres. Yield for each synthesis was calculated from the following equation:

Fig. 1 SEM images of sucrose-derived hydrochar obtained by hydrothermal treatment.

Table 1 Notation, synthesis parameters, yield of hydrochar and average particle size of carbon spheres

Hydrochar	Parameter	Yield (%)	Average particle size (μm)
C1	200 °C, 1.5 M, 12 h	39.2	10.33 ± 2.07
C2	200 °C, 1.5 M, 5 h	34.6	9.27 ± 2.79
C3	200 °C, 1 M, 12 h	36.1	5.18 ± 0.29
C4	200 °C, 1 M, 5 h	31.0	10.86 ± 3.72
C5	190 °C, 1.5 M, 12 h	36.6	6.69 ± 0.93
C6	190 °C, 1.5 M, 5 h	24.0	11.41 ± 3.46
C7	190 °C, 1 M, 12 h	32.2	4.48 ± 0.91
C8	190 °C, 1 M, 5 h	20.1	11.74 ± 2.22

---

When synthesis time was increased to 12 h, CSs were joined together, forming a fused type structure. Concentration of precursor influenced only yield rather than size of CSs and degree of fusing, with C6 sample giving a yield of 24.0% and C8 sample giving a yield of 20.1%. We found that higher temperatures resulted in fused CSs and higher yield, even with a relatively small 10 degree difference between 190 °C and 200 °C, while particle size remained the same. Based on the above discussion, we discuss how these two morphologies further differentiate upon activation. C2 and C6 were investigated further, which represent fused structure and separated structures, respectively.

Table 2 shows the notation of activated carbon. Yield of activated carbon was calculated from

Table 2 Sample names, synthesis parameters and yield of activated carbon

Activated carbon	Parameter	Yield
C2K2	C2 carbon precursor/KOH = 1 : 2, 800 °C, 1 h	32.2%
C2K4	C2 carbon precursor/KOH = 1 : 4, 800 °C, 1 h	20.1%
C6K2	C6 carbon precursor/KOH = 1 : 2, 800 °C, 1 h	31.0%
C6K4	C6 carbon precursor/KOH = 1 : 4, 800 °C, 1 h	18.7%

---

It can be observed that yield only depends on KOH amount and is unaffected by hydrochar morphology. Higher KOH concentration leads to lower yield of products. The effect of KOH amount concentration can be seen in Fig. 2. With lower concentrations (Fig. 2a and c), spheres are still present in parts of the structure (marked with solid line). However, with a higher concentration of KOH (Fig. 2b and d), we could hardly find spheres in the structure because the morphology is dominated by porous fragments that have formed instead of spheres. It could be deduced that at a 1 : 2 ratio of carbon precursor to KOH, the amount of activation agent was not enough to react with all spheres, resulting in a coexistent structure of CSs and porous fragments. The effect of different morphologies of the starting hydrochar precursor was not evident, as there was not much difference between C2K4 and C6K4. Due to the high reaction temperature and the addition of enough KOH, the two different types of morphology of carbon precursors were eliminated.

Fig. 2 SEM images of activated carbon obtained through activation of hydrochar samples. (a) C2K2, (b) C2K4, (c) C6K2, (d) C6K4.

3.1.2 Chemical properties of hydrochar and activated carbon. Functional groups on the surface of various hydrochar samples have been investigated in detail.¹⁶ FTIR spectra of our sucrose-derived hydrochar samples are shown in Fig. 3. Broad bands in the 3000–3700 cm⁻¹ range correspond to O–H stretching in water, and the bands at around 2900 cm⁻¹ were assigned to stretching vibrations of aliphatic C–H. The bands at 1620 cm⁻¹ together with the band at 1510 cm⁻¹ were attributed to C=C vibrations, and the band at 1710 cm⁻¹ was assigned to C=O vibration. Another broad band at the 1000–1450 cm⁻¹ region correspond to C–O stretching. Two bands at 800 and 750 cm⁻¹ were attributed to aromatic C–H out-of-plane bending vibrations. No matter the change of synthesis parameters, the FTIR results for each sample were nearly the same. The results suggest that even with different reaction conditions, hydrochar with similar chemical characteristic can be obtained.

Fig. 3 FTIR spectra of hydrochar samples. Characteristic peaks can be identified in the spectra of samples under various synthesis parameters.

For the activated samples, the number of characteristic peaks decreased due to the dramatic chemical reaction. Only O–H stretching band, C=C vibration band, and C=O vibration band were left, and a peak at around 2350 cm⁻¹ formed which is identified with atmospheric carbon dioxide²⁴ found in FTIR results (absent or barely present in C6K4). Different KOH concentration and different morphologies of carbon precursor seemed to have no significant impact on resulting samples. FTIR results (Fig. 4) show that hydrochar of all preparations were successfully transformed into activated carbon.

Fig. 4 FTIR spectra of activated carbon derived from hydrochar samples.

3.1.3 Nitrogen adsorption isotherms and related surface area results. Nitrogen adsorption isotherms of hydrochar and activated carbon, performed at the National Central University Precision Instruments Center, are shown in Fig. 5 and the related textural properties are summarized in Table 3. Pressure ranges used for BET specific surface area determination were chosen according to selection rules suggested by Rouquerol et al.²⁵ Hydrochar samples displayed very low adsorption quantities no matter the fused (C2) or separated (C6) structure of CSs, as seen from the nearly flat isotherm seen near the baseline in Fig. 5. The four activated samples exhibit a type I isotherm, which implies the microporous nature of our samples as adsorption amount quickly increased at low relative pressure in nitrogen adsorption isotherms. It could be seen that higher KOH concentrations resulted in samples with higher adsorption quantities because of a more complete reaction, which correlates with our observations using the SEM. Of note is the appearance of a hysteresis loop for samples obtained from a hydrochar to KOH mass ratio of 1 : 4, nearly in the same position for both C2K4 and C6K4 samples. The formation of the hysteresis loop might be due to the existence of mesopore structures of activated carbon.^26–28 Specific surface area followed the decreasing order of C2K4 > C6K4 > C6K2 for BET determination and the BET result of C2K4 was up to 3036 m² g⁻¹.

Fig. 5 Nitrogen adsorption and desorption isotherms of fused (C2) and separated (C6) hydrochar, and activated carbon derived from C2 and C6 with different KOH amount.

Table 3 BET surface areas of carbon precursor and activated carbons

Notation	ABET (m^2 g^−1)
C2	3.06
C2K4	3036
C6	3.10
C6K2	1635
C6K4	2837

---

3.2 The effect of pH and nitrogen doping on hydrochar and activated carbon synthesis

3.2.1 Yield and supernatant liquid. All information related to hydrothermal treatment of sucrose are listed in Table 4, including hydrothermal condition, yield of hydrochar, pH value of solution and average particle size of carbon spheres. Particle size of CSs was measured in the same way by using the MB ruler program as mentioned in Section 3.1.1. Each set of data was obtained by averaging three separate experimental results. In this table, initial pH and final pH represent the values for the solution before and after hydrothermal treatment, respectively. It is evident that yield, pH value and particle size depend on pH value of hydrothermal solution and nitrogen doping. For non-nitrogenated samples, yields were nearly the same, but the yield of hydrochar was observed to follow the order of water > acidic > basic conditions, a trend that is also found in nitrogenated samples. With the addition of ammonium sulfate, the yield increased expect for synthesis under basic conditions. Results suggest the addition of ammonium alters the nucleation mechanism by offering an easier path for hydrochar to nucleate and grow, leading to an increased yield and lager particle size (discuss in next section).

Table 4 Samples names, yield of hydrochar, pH value of solution and average particle size of carbon spheres

Hydrochar	Initial pH	Final pH	Yield	Average particle size (μm)
HW	6.50	2.06	27.3%	3.26 ± 0.69
HA	1.54	1.42	26.3%	6.66 ± 2.27
HB	11.53	3.20	25.6%	2.89 ± 0.96
HWN	5.41	1.56	33.9%	6.44 ± 1.11
HAN	1.62	1.41	30.1%	7.99 ± 1.76
HBN	10.33	2.31	24.2%	8.01 ± 1.86

---

Initial pH values of HW, HA and HB solutions confirm neutral, acidic and basic conditions, respectively. When ammonium sulfate was added to the reaction, starting pH values changed slightly to lower acidity and basicity of HA and HB solutions, respectively. All final pH data are lower than initial pH values, as expected from the formation of organic acids such as acetic, lactic, propenoic, levulinic and formic acids or other soluble compounds such as 5-hydroxymethylfurfural, furfural and 5-methylfurfural during hydrothermal treatment of sucrose.² Previous research has demonstrated the importance of such organic acids and soluble compounds to hydrochar yield.^2,29,30 Therefore, the supernatant liquid obtained from hydrothermal treatment were analyzed with gas chromatography-mass spectrometry (GC-MS, National Chiao Tung University Center for Advanced Instrumentation). Two target compounds, 5-hydroxymethylfurfural (HMF) and levulinic acid, were chosen for the reason that these two by-products are competitive compounds within hydrothermal treatment of sucrose.³⁰

Much lower levels of HMF were found in nitrogenated samples compared to non-nitrogenated ones, as summarized in Table 5, and the result corresponds to higher yields achieved with the addition of ammonium sulfate, except for a slight discrepancy in HBN results. The clear decrease in HMF levels upon addition of ammonium was measured in the supernatant liquid and a corresponding elevation in the hydrochar yield.

Table 5 GC-MS results with quantitative analysis for supernatant liquid of hydrothermal treatment

Hydrochar	HMF (mM)	Levulinic acid (mM)
HW	6.3	18.4
HA	0.8	47.4
HB	18.2	8.6
HWN	0.16	28.7
HAN	0.03	35.8
HBN	11.97	1.69

---

Additionally, we found no regular relationship between levulinic acid and yield. Levulinic acid results from GC-MS did not reveal the trend described in the reference paper mentioned above. Many organic acids can be produced during the hydrothermal treatment of sucrose,² and we deduce that levulinic acid may not have be the main acid product from our experiments.

3.2.2 Structural characteristics of hydrochar and activated carbon. Morphology of hydrothermally treated 0.5 M sucrose under 200 °C with different chemical environments can be seen in the SEM images of Fig. 6. It is observed that a typical sample consisted of spherical particles with a diameter ranging from 2–8 μm. The mean diameter and standard deviations of hydrochar for distinct synthesis parameters are listed in Table 4. The particle size of HW and HB are smaller than that of HA, and morphology of HB is different from HW and HA in that the carbon spheres are joined together, forming a fused type structure promoted by the basic conditions during hydrothermal treatment. When ammonium sulfate was added for nitrogen doping, results suggest that CSs develop into larger microspheres. Particle sizes of nitrogenated samples are larger than non-nitrogenated samples in the same solution condition. From these results, it is deduced that the addition of ammonium sulfate results in the promotion of nucleation and growth and makes the reaction of hydrochar more complete as we could observe the apparent difference of particle sizes between HB and HBN. Elemental analysis results prove that nitrogen exist in the structure, detailed in Section 3.2.4.

Fig. 6 SEM images of hydrochar obtained by hydrothermal treatment in six synthesized conditions.

The above discussions conclude that: (i) basic solution for hydrothermal treatment might limit the growth of CSs, leading to smaller and more irregular shapes than HW and HA; (ii) with the addition of ammonium, hydrochar develop to CSs of larger sizes no matter in water, acidic or basic conditions because of more complete reaction occurring when ammonium is present.

After the activation process, particles of entirely different morphology were obtained. Fig. 8 shows SEM images of activated carbon samples obtained from distinctly different initial hydrochar precursors, but by undergoing the same activation process, they all display porous fragments. However, despite similarities in the morphology of these six samples, there are still some differences. Activated carbon fragments can be categorized with the aid of BET results into three types: spherical structure (marked with solid line), sponge-like fragments (marked with dashed line) and fine pore fragments (marked with dotted line) in Fig. 7. Spherical structures retained the original morphology of carbon spheres, large holes existing in some fragments result in sponge-like fragments, while fine pore fragments were defined as structure where the pores cannot be observed at the magnification in Fig. 7. These three structure types provide different specific surface areas because they contribute varying degrees of adsorption capacities.

Fig. 7 Activated carbon was composed of spherical structure (marked with solid line), sponge-like fragments (marked with dashed line) and fine pore fragments (marked with dotted line).

Fig. 8 SEM images of activated carbon obtained through activation of hydrochar samples.

3.2.3 Chemical properties of hydrochar and activated carbon. A reduced peak in FTIR spectra of nitrogenated samples at 1705 cm⁻¹ has been previously described in literature, suggesting that nitrogenated samples contains reduced formation of C=O groups.³⁰ Another band at 621 cm⁻¹ related to SO₄ stretching was observed by the authors in each nitrogenated samples because of the addition of ammonium sulfate. Other evidence that proves nitrogen doping was successful include N–H stretching bands at 3000–3500 cm⁻¹, but this was indistinguishable due to the overlapped region of O–H stretching bands.

Hydrochar FTIR results here show that the C=O reduction band at 1705 cm⁻¹ and SO₄ stretching band at 621 cm⁻¹ were not observed (Fig. 9). The C=O stretching bands of nitrogenated and non-nitrogenated seem to have same intensity. While the absence of the 621 cm⁻¹ peak indicates that SO₄ is not found in the structure for nitrogenated samples of this work, which should be the case since SO₄²⁻ in its ion state would have been removed by deionized water described in the experimental procedure. FTIR spectra cannot prove nitrogen doping of samples because all samples showed the same results. Additionally, there are two different results compared to reference paper. Therefore, elemental analysis (National Chung Hsing University Instrument Center) was employed to provide definitive answer as to whether nitrogen doping was successful in the samples.

Fig. 9 FTIR spectra of hydrochar samples. Two characteristic peaks at 621 cm⁻¹ and at 1705 cm⁻¹ are marked in spectra.

There was no obvious difference between the six activated samples in Fig. 10. Peaks can be identified for O–H stretching band in the range of 3000–3700 cm⁻¹, C=C vibration band in 1630 cm⁻¹, and C=O vibration band ranging from 1450 to 1000 cm⁻¹. The results demonstrated that functional groups on activated carbon were the same after activation process no matter distinct synthesis conditions of hydrochar, whether formed in basic, acidic, or neutral conditions, with our without nitrogen doping. Similar results happened in Section 3.1.2, which illustrate same FTIR results of activated carbon even though different hydrochar precursor. It can be concluded that KOH activation of hydrochar samples can result in the same functional groups on the surface of activated carbon samples.

Fig. 10 FTIR spectra of activated carbon.

3.2.4 Elemental analysis of hydrochar and activated carbon. Tables 6 and 7 report elemental analysis results of hydrochar and activated carbon, respectively. Elemental analysis of C, H, N, S, O, residue and empirical formula are listed in detail. Residue was calculated by the difference of initial sample weight and analyzed elements, and empirical formula was calculated from elemental analysis results. Strong evidence of nitrogen doping of hydrochar can be observed in Table 6. Nitrogenated hydrochar displayed higher nitrogen content than non-nitrogenated ones. Obvious elevation of nitrogen content was also found in activated carbon in Table 7. When comparing activated carbon with hydrochar, the promotion of carbon content and reduction of hydrogen and oxygen can be noted, which is due to the activation process of the hydrochar. Additionally, a relatively higher content of sulfur can be observed in HW-AC, HA-AC and HB-AC, and the sulfur content in nitrogenated activated carbon cannot be explained because there were nearly no sulfur existing in the hydrochar precursors used. Trace amounts can be found in the impurities of sucrose with 0.003 wt% of SO₄ and KOH with 0.001 wt% of SO₄. High residue amounts can be seen in HAN-AC and HBN-AC, indicating unidentified impurities in the structure.

Table 6 Elemental analysis of hydrochars

Notation	C (wt%)	H (wt%)	N (wt%)	S (wt%)	O (wt%)	Residue	Empirical formula
HW	67.30	4.425	0.085	0.64	27.50	0.05	CH0.789O0.306
HA	65.94	4.28	0.13	0.03	28.37	1.25	CH0.779O0.323
HB	66.035	4.77	0.095	0.115	28.65	0.335	CH0.867O0.325
HWN	66.50	4.195	1.59	0.535	27.04	0.14	CH0.757N0.0205O0.305
HAN	83.095	4.18	0.87	0.62	11.12	0.115	CH0.604N0.00897O0.100
HBN	74.045	3.775	1.055	0.345	19.96	0.82	CH0.612N0.0122O0.202

---

Table 7 Elemental analysis of activated carbon

Notation	C (wt%)	H (wt%)	N (wt%)	S (wt%)	O (wt%)	Residue	Empirical formula
HW-AC	87.16	0.855	0.48	1.94	8.635	0.93	CH0.118O0.0743
HA-AC	93.07	0.84	0.48	1.94	3.575	0.095	CH0.108O0.0288
HB-AC	87.935	0.87	0.54	3.075	7.415	0.165	CH0.119O0.0632
HWN-AC	88.035	2.165	0.94	0.95	7.69	0.22	CH0.295N0.00915O0.0662
HAN-AC	88.015	1.945	0.905	0.95	2.195	5.99	CH0.265N0.00881O0.0187
HBN-AC	89.215	2.945	0.78	0.775	3.78	2.505	CH0.396N0.00749O0.0318

---

3.2.5 Nitrogen adsorption isotherms and related surface area results. Nitrogen adsorption isotherms for samples synthesized in water, acidic, and basic conditions are reported in Fig. 11. Hydrochars resulting from the three conditions exhibited a non-porous structure, as evident from the flat isotherms found near the baseline. Low surface area results of hydrochar (Table 8) suggest that hydrochar porosity and resulting specific surface areas were not further optimized upon nitrogen doping, as there was hardly any promotion of surface area results. On the other hand, adsorption amounts for activated carbon samples obtained from both nitrogenated and non-nitrogenated hydrochar showed drastic improvements, as seen from their isotherms shown in Fig. 11.

Fig. 11 Nitrogen adsorption isotherms of hydrochar and activated carbon in water, acidic and basic conditions.

Table 8 BET surface areas of hydrochars in the conditions of water, acidic, basic or nitrogen doping

Notation	ABET (m^2 g^−1)
HW	4.75
HA	1.44
HB	2.60
HWN	1.31
HAN	2.34
HBN	3.54

---

The adsorption isotherms indicated that adsorption capacities of HW-AC > HWN-AC, HAN-AC > HA-AC and HBN-AC > HB-AC. With nitrogenated hydrochar precursors from acidic and basic conditions, higher surface area of activated carbon could be obtained (Table 9). However, an opposite effect was found for neutral conditions. Surface area results followed the order of HAN-AC > HBN-AC > HW-AC > HA-AC > HWN-AC > HB-AC. The same functional groups were found in all non-nitrogenated activated carbon samples, and likewise for all nitrogenated samples as proven by using FTIR. Hence, we could eliminate the effect of different functional groups on the surface of the samples. Pore size distribution for activated carbon samples show mainly micropores with significant portion of pores with diameters down to 5 Å for non-nitrogenated samples. With nitrogen doping, however, only samples processed in acidic and basic environments showed a significant increase in distribution of larger diameter pores (Fig. S1†).

Table 9 BET surface areas of activated carbons in the conditions of water, acidic, basic or nitrogen doping

Notation	ABET (m^2 g^−1)
HW-AC	2430
HA-AC	2392
HB-AC	2239
HWN-AC	2317
HAN-AC	2693
HBN-AC	2681

---

Furthermore, material morphology has a great impact on adsorption capacities. From SEM images, three morphology types, described before as the spherical structure, sponge-like fragments, and fine pore fragments, can thus be identified. Spherical structure contributed the lowest adsorption as hydrochar because of the limited surface area and pore structure, while fine pore fragments were considered to provide the most nitrogen adsorption sites because of the micropores present in the broken structure. Sponge-like fragments with large holes supplied a lower degree of adsorption because such large cavities can be thought of as flat surfaces which are not able to adsorb much nitrogen internally.

It was observed that large areas of fine pore fragments existed in HAN-AC and HBN-AC, leading to a relatively increased surface area. Sponge-like fragments and spherical structure occurred in HW-AC, HA-AC and HWN-AC, resulting in moderate surface areas between the six activated samples. The lowest surface area was found in the HB-AC sample, which was composed of small pieces of porous fragments and spherical structures. We deduce that hydrochar precursor particle size affects the final form of the active material. The nitrogenated hydrochar with larger particle sizes seemed to obtain higher surface area after activation. Such large-sized hydrochar tend to form larger porous fragments of activated carbon via activation process.

The conclusion of nitrogen adsorption measurement was that surface area of activated carbon synthesized from these six conditions is significantly influenced by the three types of morphologies, which are in turn influenced by hydrochar particle size.

3.3 The application of air filtration with resulting activated carbon

In this section, air filtration papers were tested by TSI model 8130 testing method, which is a commercial respirator mask standard, at the Taiwan Textile Research Institute. In this method, carefully regulated test parameters include aerosol concentration, aerosol diameter, aerosol flow rate, aerosol temperature and relative humidity of the environment. NaCl with mass median diameter (MMD) of 0.26 μm was used as solid aerosol that was actively filtered by test samples. The testing model was chosen because of the limitation of our sample size. The results of TSI model 8130 testing method include filtration efficiency and pressure loss which together can determine the filtering capability. The comparison of these two results between air filtration papers and commercial air filter product are discussed in this section.

3.3.1 Filtration efficiency and pressure loss. Higher filtration efficiency indicates higher capability of intercepting dust particles. Low pressure loss represents that air can flow through the filter with less resistance. Filters with great functionality have results of both high filtration efficiency and low pressure loss.

The filtration efficiency of air filtration paper created in this work is about 14–15% regardless of different activated carbon added (Table 10, additional information in ESI†), and the value is around a quarter of commercially available air filter product. Pressure loss of air filtration papers is higher than commercial air filter product. Although lower in performance, these samples serve to demonstrate several key characteristics. These results suggested that four different activated carbon has no significant difference for air filter application. Hence, we can speculate that the efficiency of filtering mechanism for activated carbon is mainly a contribution from surface interception which can capture particles within a gas flow when particles touch the surface³¹ instead of material porosity because of the close value of filtration efficiency for our samples.

Table 10 Filtration efficiency, pressure loss and thickness of air filtration paper, paper and commercial air filter product

Filtration paper	Solid aerosol filtration efficiency (%) (0.26 μm, NaCl, MMD)	Pressure loss (mm H2O)	Thickness (μm)
C2K2	14.03	1.5	99 ± 2.9
C2K4	14.79	2.1	146 ± 18.1
C6K2	14.79	1.7	88 ± 10.4
C6K4	15.36	1.8	112 ± 15.6
Paper	5.00	1.6	Not measured
Commercial	57.99	0.6	150 ± 11.4

---

With addition of activated carbon, there is approximately ten percent improvement on filtration efficiency while keeping the same pressure loss. Based on the above discussion, we can see further potential in optimizing filtering mechanisms in activated carbon for use in air filtration application.

4. Conclusions

In this work, the effect of reaction time, reaction temperature, sucrose concentration, pH effect of hydrothermal solution and nitrogen doping that affected sucrose-derived hydrochar and activated carbon were investigated. All reaction conditions that affected yield, particle size of CSs, SEM morphology, functional groups and BET surface area were discussed here. These different reaction parameters make hydrochar a controllable material. Through KOH activation process, highly porous material was synthesized and one of the highest specific surface area was up to 3026 m² g⁻¹. Moreover, it was shown that the amount of activation agent affected the pore characteristics of activated carbon greatly. pH effect and nitrogen doping of hydrochar precursor also had great influence on surface area results of activated carbon. In our results, nitrogenated activated carbon had larger surface area than non-nitrogen activated carbon, but hydrochar displayed similar surface area results.

The air filtration application of activated carbon was performed by air filtration paper which used activated carbon as adsorbent and pulp as substrate making it a biodegradable material. The filtering ability of air filtration paper achieved one quarter of commercial air filter product but different synthesized activated carbon demonstrated similar filtration efficiency. By determining material porosity is not a significant factor in efficiency while surface trapping mechanism is important, future optimization of active filler materials can greatly increase the use of carbon materials for filtration.

Acknowledgements

The authors are grateful to the National Central University and the Bureau of Energy, Ministry of Economic Affairs, Taiwan, R.O.C. for funding (project title: Biodegradable Air Filter Development). SWH thanks JM Huang and JD Lin at NCU Precision Instruments Center for assistance in BET and SEM. The authors also thank the National Chung Hsing University and the National Chiao Tung University for elemental analysis and GC-MS measurements, respectively. BKC thanks Prof AST Chiang for providing equipment and Prof Tu Lee for FTIR used in this research.

References

1. F. Jin and H. Enomoto, Energy Environ. Sci., 2011, 4, 382–397.
2. M. Sevilla and A. B. Fuertes, Chem.–Eur. J., 2009, 15, 4195–4203.
3. Y. Qi, M. Zhang, L. Qi and Y. Qi, RSC Adv., 2016, 6, 20814–20823.
4. M.-M. Titirici, M. Antonietti and N. Baccile, Green Chem., 2008, 10, 1204.
5. X. Lu, P. J. Pellechia, J. R. Flora and N. D. Berge, Bioresour. Technol., 2013, 138, 180–190.
6. X. Cao, K. S. Ro, J. A. Libra, C. I. Kammann, I. Lima, N. Berge, L. Li, Y. Li, N. Chen, J. Yang, B. Deng and J. Mao, J. Agric. Food Chem., 2013, 61, 9401–9411.
7. S. Kang, X. Li, J. Fan and J. Chang, Ind. Eng. Chem. Res., 2012, 51, 9023–9031.
8. W. H. Chen, S. C. Ye and H. K. Sheen, Bioresour. Technol., 2012, 118, 195–203.
9. Z. Liu, A. Quek, S. Kent Hoekman and R. Balasubramanian, Fuel, 2013, 103, 943–949.
10. H. Yuan, L. Deng, Y. Qi, N. Kobayashi and J. Tang, Sci. World J., 2014, 2014, 832850.
11. X. Chen, G. Chen, L. Chen, Y. Chen, J. Lehmann, M. B. McBride and A. G. Hay, Bioresour. Technol., 2011, 102, 8877–8884.
12. H. Wang, Z. Xu, A. Kohandehghan, Z. Li, K. Cui, X. Tan, T. J. Stephenson, C. K. King'ondu, C. M. B. Holt, B. C. Olsen, J. K. Tak, D. Harfield, A. O. Anyia and D. Mitlin, ACS Nano, 2013, 7, 5131–5141.
13. M. M. Titirici, A. Thomas, S.-H. Yu, J.-O. Müller and M. Antonietti, Chem. Mater., 2007, 19, 4205–4212.
14. A. J. Romero-Anaya, M. Ouzzine, M. A. Lillo-Ródenas and A. Linares-Solano, Carbon, 2014, 68, 296–307.
15. A. S. Mestre, E. Tyszko, M. A. Andrade, M. Galhetas, C. Freire and A. P. Carvalho, RSC Adv., 2015, 5, 19696–19707.
16. Y. Shi, X. Zhang and G. Liu, ACS Sustainable Chem. Eng., 2015, 3, 2237–2246.
17. L. Mao, Y. Zhang, Y. Hu, K. H. Ho, Q. Ke, H. Liu, Z. Hu, D. Zhao and J. Wang, RSC Adv., 2015, 5, 9307–9313.
18. N. Subramanian and B. Viswanathan, RSC Adv., 2015, 5, 63000–63011.
19. M. Nunes, I. M. Rocha, D. M. Fernandes, A. S. Mestre, C. N. Moura, A. P. Carvalho, M. F. R. Pereira and C. Freire, RSC Adv., 2015, 5, 102919–102931.
20. M. Sevilla, A. B. Fuertes and R. Mokaya, Energy Environ. Sci., 2011, 4, 1400–1410.
21. M. Sevilla and A. B. Fuertes, Energy Environ. Sci., 2011, 4, 1765–1771.
22. J. Bimer, P. D. Sałbut, S. Berłożecki, J.-P. Boudou, E. Broniek and T. Siemieniewska, Fuel, 1998, 77, 519–525.
23. M. Bader, MB-Ruler (5.3), 2014.
24. G. Socrates, Infrared and Raman Characteristic Group Frequencies: Tables and Charts, John Wiley & Sons Ltd., Chichester, 2001.
25. J. Rouquerol, P. Llewellyn and F. Rouquerol, Stud. Surf. Sci. Catal., 2007, 160, 49–56.
26. C. Solar, F. Sardella, C. Deiana, R. M. Lago, A. Vallone and K. Sapag, Mater. Res., 2008, 11, 409–414.
27. J.-M. Beralus, R. Ruiz-Rosas, D. Cazorla-Amorós and E. Morallón, Frontiers in Materials, 2014, 1, 28.
28. Y.-D. Chen, W.-Q. Chen, B. Huang and M.-J. Huang, Chem. Eng. Res. Des., 2013, 91, 1783–1789.
29. M. Sevilla and A. B. Fuertes, Carbon, 2009, 47, 2281–2289.
30. K. G. Latham, G. Jambu, S. D. Joseph and S. W. Donne, ACS Sustainable Chem. Eng., 2014, 2, 755–764.
31. Application Note ITI-041, T. Incorporated, 2012.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra23958g

---