High temperature ammonia modification of rice husk char to enhance CO₂ adsorption: influence of pre-deashing

Abstract

To enhance the CO₂ adsorption capacity of rice husk char, pre-deashing of raw materials and high temperature ammonia treatment were combined to prepare the nitrogen-enriched char in this work. Two different pre-deashing methods were compared: HCl pre-deashing and HF–HCl pre-deashing treatment. The physicochemical properties and CO₂ adsorption capacity of the prepared nitrogen-enriched chars were investigated. The experimental results show that the micropore surface area of nitrogen-enriched char derived from HF–HCl deashed rice husk (HF–HCl–N-Char) is 549.61 m² g⁻¹, much greater than the HCl deashed (HCl–N-Char) of 313.72 m² g⁻¹ and the non-deashed (N-Char) of 303.10 m² g⁻¹. The nitrogen content of HF–HCl–N-Char (3.32 wt%) is also the highest among the three types of chars, while that of HCl–N-Char is only 1.01 wt%, even less than the N-Char (1.64 wt%). In addition, at adsorption temperatures of 30 °C and 120 °C, HF–HCl–N-Char demonstrates the largest CO₂ adsorption capacity among the three types of chars, which is 85.90 mg g⁻¹ at 30 °C and 19.27 mg g⁻¹ at 120 °C, respectively. However, compared with N-Char, HCl–N-Char exhibits no obvious increase in CO₂ adsorption capacity. This indicates that HF–HCl pre-deashing improves the CO₂ adsorption capacity of nitrogen-enriched char, but HCl pre-deashing does not.

---

1. Introduction

Rice husk is a by-product of rice-milling. Annually about 40 millions of tons is produced; a major source of agricultural waste in China.¹ It possesses a high content of organic constituents (i.e., cellulose, hemicellulose and lignin), and is a potential source of renewable energy.² Unfortunately, only a small portion of the rice husk is used to prepare poultry feed. Most of it is used as bedding material for animals or in direct combustion.³ Those disposal methods cause resource wastes and environmental problems.

Thus, more eco-friendly and economical uses of rice husk must be identified. Using it as precursor for preparing nitrogen-enriched biochar by high temperature ammonia treatment (hereafter ammonification) is a relatively new developed method for rise husk reuse.^4–9 The derived nitrogen-enriched biochar demonstrates both porous structure and good surface chemical properties, making it an alternative material for CO₂ adsorption. In addition, waste nitrogen-enriched biochar that has adsorbed CO₂ can be used as organic carbon, nitrogen fertilizer, and nutrient for amending soil quality.¹⁰ However, the performance improvement from using ammonification on rice husks is less obvious than that of other precursors (i.e., eucalyptus wood,⁸ almond shells,¹¹ and cotton stalks¹²). The main reason for the difference in ammonification performance is that rice husk has a high ash content,^13–15 whereas eucalyptus wood, almond shells, and cotton stalks have relatively low ash content.

In general, HF leaching is used to remove ash from rice husk, because its main ash component is silica.^16,17 However, HF is effective in eliminating Si, but not Mg, Ca, and Fe, and the deashed rice husk still contains considerable amounts of Mg, Ca, and Fe. The likely mechanism of this phenomenon is that compounds containing Mg, Ca, or Fe are inclined to react with HF to form insoluble or just slightly soluble compounds, such as MgF₂, CaF₂, or FeF₃. These insoluble or slightly soluble compounds remain in the precursor and may block the pore structures, degrading the textual characteristics of the derived biochar. Furthermore, Yin et al.¹⁸ also reported that Ca, Al, and Fe minerals in active carbon are disadvantageous for CO₂ adsorption.

Accordingly, in this work, two different pre-deashing treatments for developing efficient CO₂ adsorbents were compared: HCl pre-deashing and HF–HCl pre-deashing treatment. The main objective of this work is to optimize the modification performance of high temperature ammonia treatment while eliminating the adverse effects of Ca and Fe. In both cases, the final aim is to enhance CO₂ adsorption.

2. Materials and methods

2.1. Materials and treatments

Raw rice husks were ground and sieved, and a particle size between 0.1 and 1 mm was selected for further treatment. The first step in the treatment was acid-washing with HCl or HF–HCl at 60 °C for 24 h (details are shown in Table 1). The deashed rice husk samples were then washed with distilled water until the pH reached 7 and dried at 50 °C for 12 h. Then carbonization and ammonification were carried out at 600 °C in N₂ (400 ml min⁻¹) for 30 min and at 700 °C in a mixture of N₂ and NH₃ (N₂ = 400 ml min⁻¹, NH₃ = 80 ml min⁻¹) for 30 min (referred to as one-step ammonification). Raw rice husk, and rice husk deashed with HCl or HF–HCl were labeled as RH, HCl-RH, and HF–HCl-RH, respectively. The obtained nitrogen-enriched biochars from RH, HCl-RH, and HF–HCl-RH were labeled as N-Char, HCl–N-Char, and HF–HCl–N-Char, respectively. RH was carbonized without ammonification treatment as a control, and the derived char was labeled as R-Char.

Table 1 Pretreatment method

Sample	Treatment	Implementation
HCl-RH	HCl deashing treatment	First, deashing with HCl at a ratio of RH : HCl (36–38%) : deionized water = 25 g : 30 ml : 100 ml. Second, filtrating and washing with distilled water until pH reaches 7
HF–HCl-RH	HF–HCl deashing treatment	First, deashing with HF at a ratio of RH : HF (40%) : deionized water = 25 g : 30 ml : 60 ml. Second, filtrating and washing with distilled water until pH reaches 7. Third, deashing the derived RH with HCl solution (details refer to HCl deashing treatment). Last, filtrating and washing with distilled water until pH reaches 7

---

2.2. Characteristic analysis

Proximate analysis was used to determine the ash content. The composition of the ash was analyzed by X-ray Fluorescence (XRF) instrument (EAGLE III, EDAX Inc., USA). Field emission scanning electron microscopy (FESEM, Sirion 200, FEI, Holland) coupled with an energy-dispersive X-ray (EDX, EDAX Inc., USA) was used to analyze the morphology and elemental composition of the RHs and chars. The Brunauer–Emmett–Teller surface area (S_BET), micropore surface area (S_mic), micropore volume (V_mic) and pore size distribution were determined by automatic adsorption equipment (ASAP2020, Micromeritics, USA). S_mic and V_mic were calculated using the Dubinin–Radushkevich (DR) method. The pore size distribution was analyzed by the density function theory (DFT). The surface functional groups of samples were characterized using Fourier Transform Infrared Spectroscopy (FTIR, VERTEX 70, Bruker, Germany). The ultimate analysis of the RHs and chars was performed with a CHNS elementary analyzer (Vario Micro Cube, Germany).

2.3. CO₂ adsorption

The CO₂ adsorption–desorption circle of biochars was evaluated in a fixed bed reactor with a thermogravimetric analyzer (TGA, shown in Fig. 1). During the process of adsorption, CO₂ was purged (100 ml min⁻¹) through the thermogravimetric system at the targeted adsorption temperature of 30 °C/120 °C for 1 h. The weight increase when exposed to CO₂, was defined as the CO₂ adsorption capacity of the samples. Once the desorption process began, the CO₂ atmosphere was switched to N₂ (100 ml min⁻¹) and the sample was kept at 200 °C for 1 h to achieve regeneration.

Fig. 1 The TGA adsorption–desorption system for CO₂.

3. Results and discussion

3.1. Ash content and composition analysis

The ash content of the deashed and raw rice husk samples was evaluated using proximate analysis. Results are presented in Table 2. After HCl pre-deashing, the ash percentage (A) clearly increases, while the volatile content (V) significantly decreases. Moisture content (M) and fixed carbon (FC) show negligible change. This phenomenon is consistent with previous studies which demonstrate that the removal performance of HCl is better for volatile components than for other components.¹⁹ In contrast to HCl pre-deashing, HF–HCl pre-deashing treatment leads to an apparent reduction in ash content, but a distinct increase in V and FC. This indicates that the deashing effectiveness varies with different treatments.

Table 2 The proximate analysis and ultimate analysis of rice husks

Sample	Proximate analysis (ad, wt%)				A/FC	Ultimate analysis (ad, wt%)
	M	V	A	FC		C	H	N	S	O^a
a Calculated by difference.
RH	5.52	67.49	15.96	11.03	1.45	42.04	5.28	0.38	0.17	30.64
HCl-RH	5.59	63.13	18.77	12.50	1.50	41.52	4.99	0.22	—	28.91
HF–HCl-RH	5.88	79.04	0.77	14.31	0.054	52.58	5.82	0.41	0.21	34.33

---

The chemical composition of the ash was obtained by XRF analysis (shown in Table 3). The HCl deashing treatment removes almost all of the inorganic elements, except Si. By contrast, HF–HCl deashing washes out Si effectively, as well as Mg, Ca, and Fe. The MgF₂, CaF₂, and FeF₃ formed during the HF deashing treatment can be removed by the following HCl deashing treatment. A possible mechanism is illustrated by eqn (1)–(3).^20–22 Moreover, HF–HCl deashing leads to an apparent increase in S, which is ascribed to the insolubility of CaSO₄ in both HF and HCl solutions.

Table 3 XRF results for untreated and treated rice husk ashes

Sample	Composition (wt%)
	MgO	Al2O3	SiO2	P2O5	SO3	Cl	K2O	CaO	MnO	Fe2O3
RH	0.48	1.03	91.25	0.75	1.17	0.61	2.56	1.62	0.19	0.34
HCl-RH	—	—	99.96	—	—	—	—	0.04	—	—
HF–HCl-RH	—	1.69	11.03	3.89	50.88	—	2.00	28.08	—	2.43

---

3.2. Textural characterization

3.2.1 FESEM analysis. FESEM photographs of the deashed RHs, raw RH, HCl–N-Char and HF–HCl–N-Char are shown in Fig. 2. The RH and HCl-RH demonstrate similar outer surface structure, which is smooth and has an organized ripple structure. This indicates that HCl deashing treatment does not destroy this outer epidermal structure, where Si is mainly localized. Compared with RH and HCl-RH, however, the outer surface of the HF–HCl-RH is rough, and the corrugated structure disappears. This is attributed to the removal of the Si-rich protective layer after HF deashing treatment, which exposes the internal xylem tissue. Fig. 2d–f show the surface morphology of the HCl–N-Char and HF–HCl–N-Char. The outer epidermis, internal vascular bundle, and inner epidermis of the samples are respectively marked with P1, P2 and P3 in Fig. 2d and f. The HCl–N-Char and HF–HCl–N-Char are similar in internal vascular bundle and inner epidermis, yet have very different outer epidermis. The HCl–N-Char retains its corrugated Si-rich structure in the one-step ammonification process, and silica particles (P4, shown in Fig. 2e) will form under high temperature treatment.³ However, the outer epidermis of the HF–HCl–N-Char is opened and a richer porosity is created by the one-step ammonification treatment. This indicates that the removal of the Si-rich protective layer is favorable for pore structure formation.

Fig. 2 FESEM micrographs of RH ((a), 200×, 100 μm), HCl-RH ((b), 200×, 100 μm), HF–HCl-RH ((c), 200×, 100 μm), HCl–N-Char ((d), 200×, 100 μm; (e), 20 000×, 1 μm) and HF–HCl–N-Char ((f), 200×, 100 μm).

3.2.2 Pore distribution and surface area. The BET apparent surface area (S_BET), micropore surface area (S_mic), and micropore volume (V_mic) of all samples calculated from the N₂ and CO₂ adsorption isotherms are shown in Table 4. HCl deashing increases the S_BET, S_mic, and V_mic of rice husk, whereas there is no obvious increase in these parameters of the corresponding nitrogen-enriched biochar. The reason is that HCl removes almost all of the inorganic elements (except Si), resulting in the formation of micropores and mesopores which are the dominant contributors to BET surface area, and this can be validated by the micro-, meso-, and macropores distribution, shown in Fig. 3a–c. The S_BET, S_mic, and V_mic of HCl–N-Char are 254.95 m² g⁻¹, 313.72 m² g⁻¹, and 0.126 ml g⁻¹, respectively. Compared with HCl treatment, HF–HCl treatment tends to reduce the S_BET, S_mic, and V_mic of rice husk. The reason is that HF removes the porous Si-rich protective layer on the outer epidermis of rice husk, resulting in the damage of micro-, meso-, and macropores,²³ and counteracts the influence of HCl in developing micro- and mesopores. Furthermore, this can be validated by the micro-, meso-, and macropores distribution (in Fig. 3a and b), where the pore distributions of RH and HF–HCl-RH also show small differences. However, HF–HCl treatment obviously increases these parameters of the corresponding nitrogen-enriched biochar. The S_BET, S_mic, and V_mic of HF–HCl–N-Char are 420.01 m² g⁻¹, 549.61 m² g⁻¹ and 0.220 ml g⁻¹, respectively. The reason for this may be that the Si-rich protective layer on the outer epidermis of rice husk forms amorphous silica and hinders the formation of pore structure during the high temperature process.^24,25 HF–HCl deashing treatment of rice husk disintegrates the Si-rich protective layer and leads to the release of more volatile matter, which benefits the pore structure formation of nitrogen-enriched biochar.

Table 4 Textural properties and CN elementary analysis results

Samples	N2 adsorption	CO2 adsorption		C (wt%)	N (wt%)
	SBET (m^2 g^−1)	Smic (m^2 g^−1)	Vmic (ml g^−1)
RH	0.79	45.93	0.018	42.04	0.38
HCl-RH	7.67	62.61	0.025	41.52	0.22
HF–HCl-RH	0.76	36.85	0.015	52.81	0.41
R-Char	243.06	296.69	0.119	52.62	0.52
N-Char	231.27	303.10	0.121	53.36	1.64
HCl–N-Char	254.95	313.72	0.126	47.74	1.01
HF–HCl–N-Char	420.01	549.61	0.220	91.86	3.32

---

Fig. 3 Micro-, meso-, and macropores distribution of RHs and chars.

Fig. 3a–c present the DFT pore size distributions of RHs. The narrow micropore (<0.7 nm) and mesopore of HCl-RH are significantly better than those of RH, but the macropore structure is poorer. This indicates that the HCl deashing pretreatment improves the narrow micropore and mesopore of the precursor, but damages the macropore. This phenomenon may be attributed to the cleanup of all inorganic minerals except Si by HCl deashing. Unlike HCl-RH, though the HF–HCl deashing treatment removes most of the minerals, including Si, there is a decreasing trend in the pore structure of HF–HCl-RH. It may be attributed to the destruction of the Si-rich protective layer. The pore size distributions of nitrogen-enriched biochars are illustrated in Fig. 3d–f. The HF–HCl–N-Char exhibits the best pore structure among all chars, whereas the pore structure of the HCl–N-Char has not been improved, and even shows a reducing trend in the narrow mesoporous pores (2 to 4 nm). It can be concluded that although HF–HCl deashing treatment deteriorates the pore structure of the precursor, it is beneficial for the pore structure development of its nitrogen-enriched biochar. Conversely, HCl deashing improves the pore structure of the precursor, but does not facilitate the pore structure development of its nitrogen-enriched biochar.

3.3. Chemical characterization

3.3.1 Elemental analysis. The EDX elemental microanalysis and surface atomic ratio of RHs are listed in Tables 5 and 6, respectively. The HF–HCl treatment clearly decreases the surface Si content of the precursor, while HCl deashing only slightly reduces it. In addition, HCl deashing leads to an increase in the O^s and N^s contents of the rice husk, but a small reduction in C^s. These changes are consistent with their surface atomic ratios. By contrast, HF–HCl deashing treatment increases the O^s and C^s content of the precursor, while reducing N^s. Because N, O, and C have different atomic masses, the changes in their surface atomic ratio and weight ratio are not identical. After HF–HCl pre-deashing treatment, the complete removal of Si in the precursor affects the distribution of the surface atom ratio of N, O, and C. There is an increase in C^a, but a decrease in O^a and N^a. This indicates that HF–HCl deashing treatment removes the O and N functional groups when eliminating the Si-rich protective layer.

Table 5 EDX element microanalysis for the sample surface, at 10 μm^a

Sample	Element (wt%)
	C^s	N^s	O^s	Mg^s	Al^s	Si^s	P^s	S^s	Cl^s	K^s	Ca^s	Mn^s
a ^sSurface element.
RH	18.80	2.28	33.86	0.05	—	43.71	1.30	—	—	—	—	—
HCl-RH	18.56	3.11	39.66	0.1	0.24	37.18	0.07	0.19	0.15	0.08	0.18	0.48
HF–HCl-RH	56.86	1.37	40.74	0.1	—	—	0.1	0.15	0.12	0.15	0.12	0.29
HCl–N-Char	35.26	2.04	17.91	—	—	44.57	0.05	0.03	—	—	—	0.12
HF–HCl–N-Char	90.51	5.17	3.81	0.08	0.06	0.07	—	—	—	—	0.09	0.22

---

Table 6 Surface atomic ratio of the sample surface, at 10 μm^a

Sample	C^a (at%)	N^a (at%)	O^a (at%)	Si^a (at%)	N^a/C^a	O^a/C^a	Si^a/C^a
a ^aSurface atomic.
RH	28.75	2.99	38.87	28.58	0.10	1.35	0.99
HCl-RH	27.54	3.95	44.18	23.6	0.14	1.60	0.87
HF–HCl-RH	63.93	1.32	34.38	—	0.02	0.54	—
HCl–N-Char	50.66	2.51	19.3	27.44	0.05	0.38	0.54
HF–HCl–N-Char	92.38	4.52	2.92	0.03	0.05	0.03	0.0003

---

Tables 5 and 6 also summarize the surface elemental analysis and atomic ratio of HCl–N-Char and HF–HCl–N-Char. Although HCl deashing increases the nitrogen and oxygen contents of the precursor, the surface nitrogen weight and atomic ratio of HCl–N-Char were only 2.04 wt% and 2.51 at%, much less than those of HF–HCl–N-Char. Furthermore, its nitrogen content (1.01 wt%) is the lowest among all nitrogen-enriched biochars (Table 4), even less than N-Char (1.64 wt%). Unlike HCl–N-Char, HF–HCl–N-Char shows greater surface nitrogen weight and atomic ratio, 5.17 wt% and 4.52 at% respectively, and its nitrogen content is also the highest, 3.32 wt%. This indicates that HF–HCl pre-deashing treatment increases the nitrogen content of nitrogen-enriched biochar, and promotes the performance of ammonification, even though it reduces the surface nitrogen and oxygen of the precursor.

Shafeeyan et al.²⁶ suggest that surface oxygen functional groups are important active sites for the introduction of nitrogen functional groups during high temperature ammonia treatment. However, in this work, though HCl deashing increases the surface oxygen functional groups of the precursor, it fails to promote the introduction of nitrogen functional groups. By contrast, HF–HCl deashing treatment decreases the surface oxygen functional groups, but enhances the introduction of nitrogen functional groups. This indicates that although surface oxygen functional groups can provide active sites for the introduction of nitrogen functional groups during the ammonification process, not all of them form these active sites. Further, HF–HCl deashing treatment may increase the effective active oxygen functional groups for the introduction of nitrogen functional groups.

3.3.2 Surface functional group analysis. To study the effects of two pre-deashing methods on the surface functional groups of precursors and the derived nitrogen-enriched biochars, FTIR analysis was performed. Fig. 4a presents the FTIR spectrums of the pre-deashed and non-deashed RHs. All RHs show similar peaks at 3428 cm⁻¹, 2921 cm⁻¹, and 2857 cm⁻¹, which may be associated to O–H stretching vibrations and asymmetric and symmetric C–H stretching vibrations in aliphatic CH, CH₂, and CH₃, respectively.^8,27 In addition, a sharp peak at 1639 cm⁻¹ is the evidence of C=O stretching in ketones and carboxylic acids.⁵ At the same time, the spectrums of all RHs exhibit differences. For instance, peaks at 1089 cm⁻¹ and 802 cm⁻¹, which can be ascribed to Si–O and Si–H stretching vibrations,²⁸ appear in RH and HCl-RH, but not in HF–HCl-RH. This is because HF deashing can break Si–O and Si–H bonds, but HCl deashing cannot.

Fig. 4 FTIR spectra of RHs (a) and chars (b).

Fig. 4b depicted the FTIR spectrums of R-Char and nitrogen-enriched biochars. The spectrums of R-Char, N-Char, and HCl–N-Char are almost similar. Peaks at 1100 cm⁻¹ and 802 cm⁻¹ which may be attributed to Si–O and Si–H observed in the precursors may still be observed.²⁹ This indicates that the high temperature ammonia treatment cannot decompose Si–O and Si–H bonds. However, the silicon functional groups hinder the introduction of nitrogen functional groups in ammonification. This can be verified by the fact that the HF–HCl–N-Char contained more nitrogen than the N-Char (Table 4) because of its high efficiency in removing Si, while HCl deashing failed to enhance the introduction of nitrogen functional groups because it failed to remove Si. The spectrum of the HF–HCl–N-Char also demonstrates that Si functional groups obstruct the introduction of nitrogen functional groups. When HF–HCl pre-deashing removes the Si functional groups in the rice husk, O–H (at 3430 cm⁻¹) and aliphatic C–H (at 2920 and 2857 cm⁻¹) of HF–HCl–N-Char break down during ammonification, while the peak at 1126 cm⁻¹, which can be assigned to C–N vibration, becomes stronger.³⁰ In addition, the C=O and C=N groups (at 1580 cm⁻¹) of HF–HCl–N-Char are slightly reduced compared with the R-Char.³¹ This is probably because the reactions of ammonia with hydroxyl and carboxylic acids lead to pyrroles and pyridines.³² The reaction mechanism is shown in Fig. 5.

Fig. 5 Mechanism for ammonification reaction resulting in pyrroles and pyridines.

3.4. CO₂ adsorption

3.4.1 Adsorption–desorption cycle of sorbent. Fig. 6 shows the CO₂ adsorption profiles of the rice husk chars during three adsorption–desorption cycles at different adsorption temperatures. The adsorption process was performed at 30 °C and 120 °C in a CO₂ atmosphere, while desorption was performed at 200 °C in a N₂ atmosphere. At adsorption temperatures of 30 °C and 120 °C, the CO₂ adsorption capacity of HF–HCl–N-Char is the largest among all rice husk chars, and shows good regeneration ability. However, compared with N-Char, HCl–N-Char displays no obvious increase in adsorption capacity, and decreases at 120 °C. This indicates that the HF–HCl pre-deashing can improve the CO₂ adsorption capacity of nitrogen-enriched char, but HCl pre-deashing cannot.

Fig. 6 CO₂ adsorption and desorption profile for three cycles.

3.4.2 Adsorption kinetic study. To analyze the CO₂ adsorption kinetics of rice husk char, two kinetic models, the intra-particle diffusion model and the Bangham model, were used. The intra-particle diffusion model³³ is given as eqn (9):

q_t = k₀t^½ + C (9)

where k₀, t and C are the intra-particle diffusion rate constant, adsorption time and a constant, respectively, and q_t is the amount of adsorbate adsorbed at time t.

The Bangham model³⁴ is given as eqn (10):

where q_t is the amount of adsorbate adsorbed at time t, q_e is the adsorption capacity at equilibrium, n and k are constants.

The coefficients of the intra-particle diffusion model and Bangham model for the CO₂ adsorption are shown in Table 7. At 30 °C and 120 °C, k₀ and q_e of N-Char are greater than those of R-Char. This indicates that the ammonification treatment increases the CO₂ adsorption rate and capacity of rice husk char. k₀ of HCl–N-Char is smaller than that of N-Char, but there is a slight increase in q_e at 30 °C. This may be because the diffusion of CO₂ into the pores of the rice husk char is not the only rate-controlling step, film and pore diffusion both play important roles in CO₂ adsorption.³⁵ HF–HCl–N-Char obtains the maximum value of k₀ and q_e among all rice husk chars at the adsorption temperatures of 30 °C and 120 °C. When the adsorption temperature is 30 °C, the adsorption process is mainly controlled by physical adsorption and the optimal porous structure of HF–HCl–N-Char greatly improves the speed of pore diffusion.³⁶ When the adsorption temperature increases to 120 °C, the adsorption process is dominated by chemical adsorption, and the improved nitrogen content and surface activity of HF–HCl–N-Char benefit chemical adsorption of CO₂.²⁶

Table 7 Coefficients of intra-particle diffusion model and Bangham model for the CO₂ adsorption

Sample	T (°C)	Intra-particle diffusion model			Bangham model
		k0 (mg (g min^1/2)^−1)	C (mg g^−1)	R^2	qe (mg g^−1)	K (min^−n)	n	R^2
R-Char	30 °C	16.26	−59.08	0.91	52.27	3.26 × 10^−5	3.01	0.97
N-Char		20.86	−82.25	0.92	57.35	2.28 × 10^−6	3.81	0.96
HCl–N-Char		19.55	−68.95	0.87	61.95	8.07 × 10^−6	3.47	0.96
HF–HCl–N-Char		30.63	−120.56	0.93	85.90	3.53 × 10^−6	3.67	0.97
R-Char	120 °C	2.60	−8.82	0.96	9.54	2.03 × 10^−4	2.46	0.98
N-Char		3.94	−12.56	0.93	14.81	2.50 × 10^−4	2.44	0.98
HCl–N-Char		3.45	−10.87	0.94	13.23	3.11 × 10^−4	2.37	0.98
HF–HCl–N-Char		6.15	−22.75	0.96	19.27	3.24 × 10^−5	3.02	0.97

---

Table 8 presented a comparison of the CO₂ adsorption capacity of the biochars developed in this work with activated carbons and biochars derived from other precursors in the literature. It can be seen that compared to these carbonaceous adsorbents derived from other precursors, for example, almond shells, coffee grounds and rice straw, the biochar derived from raw rice husk with ammonia modification (N-Char) shows poorer CO₂ adsorption capacity. However, the biochar derived from HF–HCl deashed rice husk (HF–HCl–N-Char) demonstrates significant improvement in CO₂ adsorption capacity (from 57 mg g⁻¹ to 86 mg g⁻¹), making it more competitive to other biochars. It indicates that pre-deashing treatment of rice husk may be helpful to promote the utilization of rice husk in producing carbonaceous adsorbents.

Table 8 A comparison on the CO₂ adsorption capacities of activated carbons and biochars derived from other precursors

Adsorbent	Precursor	SBET (m^2 g^−1)	Temperature (°C)	CO2 adsorption capacity (mg g^−1)	Reference
Activated carbon	—	1323	36	76	5
Activated carbon	Palm shells	889	30	74	26
Activated carbon	Peat	942	30	97	37
Biochar	Almond shells	822	25	97	38
Biochar	Coffee grounds	84	25	62	39
Biochar	Rice straw	122	20	77	40
N-Char	Rice husk	231	30	57	This work
HCl–N-Char	Rice husk	255	30	62	This work
HF–HCl–N-Char	Rice husk	420	30	86	This work

---

4. Conclusions

In this work, HCl pre-deashing treatment and HF–HCl pre-deashing treatment for developing efficient CO₂ adsorbents were compared. The corresponding nitrogen-enriched chars demonstrate significant differences in physicochemical properties, which affect the CO₂ adsorption performance greatly. The most notable conclusions are summarized as follows:

(1) HCl pre-deashing treatment successfully removes Mg, Ca, and Fe from the rice husk, but fails to remove Si, which forms a Si-rich protective layer hindering the pore structure development during one-step ammonification. HF–HCl pre-deashing efficiently removes Mg, Ca, Fe, and Si, which leads to the collapse of the Si-rich layer and greatly improves the pore structure of the derived nitrogen-enriched char. The micropore surface area of HF–HCl–N-Char was 549.61 m² g⁻¹, much greater than those of N-Char (303.10 m² g⁻¹) and HCl–N-Char (313.72 m² g⁻¹).

(2) Though HCl pre-deashing increases the surface oxygen functional groups of the precursor, it fails to enhance the introduction of nitrogen functional groups. By contrast, HF–HCl pre-deashing reduces the surface oxygen functional groups of the precursor, but increases the effective active oxygen functional groups, which introduce nitrogen functional groups. Hence, it promotes the introduction of nitrogen functional groups.

(3) HCl pre-deashing fails to improve the adsorption capacity and adsorption rate of rice husk char, while HF–HCl pre-deashing successfully increases the adsorption capacity and rate at the adsorption temperatures of 30 °C and 120 °C.

Acknowledgements

The authors wish to express their sincere thanks to the financial support from the National Basic Research Program of China (2013CB228102), the National Natural Science Foundation of China (51276075 and 51476067), the Special Fund for Agro-scientific Research in the Public Interest (201303095) and the innovation foundation of Huazhong University of Science and Technology (2014TS118 and CX14-035). The study also benefits from the technical support from Analytical and Testing Center in Huazhong University of Science & Technology (http://atc.hust.edu.cn).

References

1. S. Hu, J. Xiang, L. Sun, M. Xu, J. Qiu and P. Fu, Fuel Process. Technol., 2008, 89, 1096–1105.
2. W. T. Tsai, M. K. Lee and Y. M. Chang, Bioresour. Technol., 2007, 98, 22–28.
3. B. D. Park, S. G. Wi, K. H. Lee, A. P. Singh, T. H. Yoon and Y. S. Kim, Biomass Bioenergy, 2003, 25, 319–327.
4. D. Long, M. Wang, J. Wang, W. M. Qiao and L. Ling, RSC Adv., 2014, 4, 61456–61464.
5. J. Przepiórski, M. Skrodzewicz and A. W. Morawski, Appl. Surf. Sci., 2004, 225, 235–242.
6. N. Sun, C. Sun, J. Liu, H. Liu, C. E. Snape, K. Li, W. Wei and Y. Sun, RSC Adv., 2015, 5, 33681–33690.
7. C. Chen, J. Kim and W. S. Ahn, Fuel, 2012, 95, 360–364.
8. A. Heidari, H. Younesi, A. Rashidi and A. A. Ghoreyshi, Chem. Eng. J., 2014, 254, 503–513.
9. J. Yu, Y. Le and B. Cheng, RSC Adv., 2012, 2, 6784.
10. R. S. Jassal, M. S. Johnson, M. Molodovskaya, T. A. Black, A. Jollymore and K. Sveinson, J. Environ. Manage., 2015, 152, 140–144.
11. M. G. Plaza, S. García, F. Rubiera, J. J. Pis and C. Pevida, Sep. Purif. Technol., 2011, 80, 96–104.
12. X. Zhang, S. Zhang, H. Yang, Y. Feng, Y. Chen, X. Wang and H. Chen, Chem. Eng. J., 2014, 257, 20–27.
13. T. H. Liou and S. J. Wu, J. Hazard. Mater., 2009, 171, 693–703.
14. D. Vamvuka, S. Troulinos and E. Kastanaki, Fuel, 2006, 85, 1763–1771.
15. Y. Shen, C. Areeprasert, B. Prabowo, F. Takahashi and K. Yoshikawa, RSC Adv., 2014, 4, 40651–40664.
16. W. M. A. W. Daud and A. H. Houshamnd, J. Nat. Gas Chem., 2010, 19, 267–279.
17. L. Wang, J. Xue, B. Gao, P. Gao, C. Mou and J. Li, RSC Adv., 2014, 4, 64744–64746.
18. G. Yin, Z. Liu, Q. Liu and W. Wu, Chem. Eng. J., 2013, 230, 133–140.
19. P. Das, A. Ganesh and P. Wangikar, Biomass Bioenergy, 2004, 27, 445–457.
20. H. K. Fouad and J. Radioanal, Nucl. Chem., 2010, 285, 193–197.
21. D. Ma, Z. Cao, H. Wang, X. Huang, L. Wang and X. Zhang, Energy Environ. Sci., 2012, 5, 8538–8542.
22. N. Louvain, A. Fakhry, P. Bonnet, M. El Ghozzi, K. Guérin, M. T. Sougrati, J. C. Jumas and P. Willmann, CrystEngComm, 2013, 15, 3664–3671.
23. X. Zhang, S. Zhang, H. Yang, J. Shao, Y. Chen, Y. Feng, X. Wang and H. Chen, Energy, 2015, 91, 903–910.
24. C. L. Mangun, K. R. Benak, J. Economy and K. L. Foster, Carbon, 2001, 39, 1809–1820.
25. O. Ioannidou and A. Zabaniotou, Renewable Sustainable Energy Rev., 2007, 11, 1966–2005.
26. M. S. Shafeeyan, W. M. A. W. Daud, A. Houshmand and A. Arami Niya, Appl. Surf. Sci., 2011, 257, 3936–3942.
27. H. Yang, Y. Yuan and S. C. E. Tsang, Chem. Eng. J., 2012, 185–186, 374–379.
28. F. Ahmad, W. M. A. W. Daud, M. A. Ahmad and R. Radzi, Chem. Eng. Res. Des., 2013, 91, 1028–1038.
29. T. H. Liou, Mater. Sci. Eng., A, 2004, 364, 313–323.
30. M. S. Shafeeyan, W. M. A. W. Daud, A. Houshmand and A. Shamiri, J. Anal. Appl. Pyrolysis, 2010, 89, 143–151.
31. A. Heidari, H. Younesi, A. Rashidi and A. Ghoreyshi, J. Taiwan Inst. Chem. Eng., 2014, 45, 579–588.
32. R. J. J. Jansen and H. van Bekkum, Carbon, 1994, 32, 1507–1516.
33. V. S. Mane, I. D. Mall and V. C. Srivastava, Dyes Pigm., 2007, 73, 269–278.
34. M. S. Bilgili, J. Hazard. Mater., 2006, 137, 157–164.
35. E. Tütem, R. Apak and Ç. F. Ünal, Water Res., 1998, 32, 2315–2324.
36. M. G. Plaza, C. Pevida, A. Arenillas, F. Rubiera and J. J. Pis, Fuel, 2007, 86, 2204–2212.
37. M. G. Plaza, S. García, F. Rubiera, J. J. Pis and C. Pevida, Chem. Eng. J., 2010, 163, 41–47.
38. M. G. Plaza, C. Pevida, B. Arias, J. Fermoso, F. Rubiera and J. J. Pis, Energ. Procedia, 2009, 1, 1107–1113.
39. M. Plaza, A. González, C. Pevida, J. Pis and F. Rubiera, Appl. Energy, 2012, 99, 272–279.
40. Y. Huang, P. Chiueh, C. Shih, S. Lo, L. Sun, Y. Zhong and C. Qiu, Energy, 2015, 84, 75–82.

---