Simplified Batch and Fixed-Bed Design System for Efficient and Sustainable Fluoride Removal from Water Using Slow Pyrolyzed Okra Stem and Black Gram Straw Biochars.

Okra stem biochar (OSBC) and black gram straw biochar (BGSBC) were prepared by slow pyrolysis at 500 and 600 °C, respectively. OSBC and BGSBC were characterized using S BET, Fourier transform infrared, X-ray diffraction, scanning electron microscopy (SEM), transmission electron microscopy, SEM-energy dispersive X-ray, and energy dispersive X-ray fluorescence. High carbon contents (dry basis) of 66.2 and 67.3% were recorded in OSBC and BGSBC, respectively. The OSBC surface area (23.52 m2/g) was higher than BGSBC (9.27 m2/g). The developed biochars successfully remediate fluoride contaminated water. Fluoride sorption experiments were accomplished at 25, 35, and 45 °C. Biochar-fluoride adsorption equilibrium data were fitted to Langmuir, Freundlich, Sips, Temkin, Koble-Corrigan, Radke and Prausnitz, Redlich-Peterson, and Toth isotherm models. The sorption dynamic data was better fitted to the pseudo-second order rate equation versus the pseudo-first order rate equation. The Langmuir sorption capacities of Q OSBC 0 = 20 mg/g and Q BGSBC 0 = 16 mg/g were obtained. Biochar fixed-bed dynamic studies were accomplished to ascertain the design parameters for developing an efficient and sustainable fluoride water treatment system. A column capacity of 6.0 mg/g for OSBC was achieved. OSBC and BGSBC satisfactorily remediated fluoride from contaminated ground water and may be considered as a sustainable solution for drinking water purification.

Introduction

Excessive fluoride in drinking water has been reported in different parts of the world.[1] Bureau of Indian standard (BIS) recommended 1.5 mg/L as the permissible limit for fluoride in drinking water.[2] Fluoride shows health benefits at low concentrations (0.5–1.0 mg/L). But at concentrations >1.5 mg/L, fluoride is toxic and can cause dental and skeletal fluorosis.[3,4] Fluoride-bearing minerals include cryolite, sellaite, fluorspar, fluormica, fluorapatite, topaz, biotite, epidote, and tremolite are the major fluoride contamination sources in groundwater. Groundwater fluoride concentration is controlled by the physical, chemical, and geological properties of the source aquifer including its depth, porosity, acidity, and temperature of the surrounding soil and sediments, action of chemical elements, and weathering intensity.[5,6] Fluoride anthropogenic sources include effluents from metallurgy, fertilizers, semiconductors, and glass manufacturing industries.[7]

Approximately twenty developing and developed nations are endemic to fluorosis.[1] India, China, and African countries of the rift valley region are the most severely affected.[1] In India, fluorosis is prevalent in many states.[1,8−10]

Precipitation,[11] ion-exchange,[12,13] adsorption,[14,15] membrane separation,[16] nanofiltration,[17] reverse osmosis,[18,19] and electrodialysis,[20] were used for water defluoridation. Adsorption is considered to be a sustainable method due to its low operational cost, better adsorption capacity, adsorbent regeneration and reuse, and excellent water quality without any sludge generation.[1]

Several adsorbents including activated and impregnated alumina,[21] activated carbon,[15] bone char,[22,23] zeolite,[24] bauxite,[25] calcite,[26] calcined clay,[27,28] activated fly ash,[29] biosorbent,[30,31] and biochar[32,33] were applied for water defluoridation. Many other suitable adsorbents for the fluoride remediation have been reviewed.[1,34] The biochar use for water decontamination is a relatively new approach.[35] Biochar is characterized by a high surface area which plays an important role in contaminant removal. Micropores in biochar contribute toward most of the biochar’s surface area.[36] Biochar is characterized by plenty of surface functional groups which are also responsible for contaminants removal.[37] In the present investigation, fluoride was successfully eliminated using okra stem biochar (OSBC) and black gram straw biochar (BGSBC) derived from okra stem (OS) and black gram straw (BGSs), respectively. Large availability of these agricultural residues makes them suitable candidates for biochar development and application in fluoride removal. Biochars were also successfully applied in actual groundwater defluoridation. The possible adsorption mechanisms are also elucidated.

Results and Discussion

Characterization

OSBC and BGSBC were obtained via the slow pyrolysis of OSs at 600 °C and BGSs at 500 °C for 30 min (Figure ). These biochars were used for water defluoridation without any further modification.

Figure 1

Experimental set-up for biochar preparation and fluoride sorption.

The physicochemical properties[32,38] of OSBC and BGSBC are summarized in Table . OSBC and BGSBC were characterized by 66.2 and 67.3% carbon, respectively. OSBC is characterized by higher nitrogen (6.55 wt %) than BGSBC (1.71 wt %). Slightly lower oxygen content (calculated by difference) was recorded in OSBC (21.99 wt %) than BGSBC (24.67 wt %). The surface area of OSBC (23.52 m2/g) was higher versus BGSBC (9.27 m2/g). The total pore volume at P/P0 = 0.97 of OSBC (0.0156 cm3/g) was more than BGSBC (0.0083 cm3/g). The Barrett–Joyner–Halenda (BJH) adsorption average pore diameter of OSBC (29.83 nm) was higher versus BGSBC (16.70 nm). The average pore size indicates that both biochars are mesoporous in nature.

Table 1

Properties of OSBC and BGSBC Prepared at 600 and 500 °C, Respectively

properties	OSBC	BGSBC
biochar yield (%)	26.25	26.12
moisture (%)	0.67	0.75
ash (wt %)	3.48	3.79
C (wt %)	66.16	67.33
H (wt %)	1.45	2.49
N (wt %)	6.55	1.71
O (wt %)a	21.99	24.67
SBET (m2/g)	23.52	9.27
VT (cm3/g)	0.0156	0.0083

Oxygen was determined by difference.

The X-ray diffraction (XRD) spectra of BGSBC and OSBC show a sharp peak at 2θ = 29.3° with d value = 3.03 Å indicated the calcite (CaCO3) presence (Figure ).[39,40] The presence of carbonate minerals is possibly due to the reaction of cations (K+, Ca2+, and Na+) in feedstock with carbon during pyrolysis.[41] Quartz (SiO2) presence in BGSBC (Figure ) is shown by a sharp and small peak at 2θ = 26.64° (d = 3.33 Å). The peak at 2θ = 72.61° in OSBC and BGSBC is also assigned to SiO2.[42] The small peak at 2θ = 43.6° corresponds to periclase (MgO) in both the biochars.[43]

Figure 2

XRD spectra of OSBC and BGSBC.

Various peaks were present in the Fourier transform infrared (FTIR) spectra of OSBC and BGSBC. Both OSBC and BGSBC exhibit almost similar FTIR spectra (Figure ). The peaks at 3630 and 3337 cm–1 correspond to the −OH stretching vibration of phenolic and/or alcoholic groups.[44,45] The peak at 1575 cm–1 in both biochars indicates alkene stretching vibrations.[46] However, this peak is slightly diminished in OSBC (Figure ). The band at 1454 cm–1 in both biochars is assigned to methyl C–H bending vibrations.[47] However, this peak shows a little shifting toward left in OSBC. A sharp peak around 875 cm–1 assigned to strong bending of C–H phenyl rings.[48] The peaks at 755 and 619 cm–1 in OSBC correspond to aromatic C–H vibrations.[49−51]

Figure 3

FTIR spectra of OSBC and BGSBC.

Surface morphology plays an important role in solid–liquid interactions.[32] The surface morphology of OSBC and BGSBC was examined using scanning electron microscopy (SEM) imaging at different magnifications (Figures and 5). Both biochars show the presence of highly porous structures. Surface canals are clearly visible in OSBC [Figure b]. Honeycomb-like structures are visible in BGSBC [Figure c]. Partially melted structures also appear [Figure c] which may be due to the exposure to higher temperatures.[32] The energy dispersive X-ray (EDX) analysis provides proportionate elemental composition of biochars [Figure S2a,b]. Both OSBC and BGSBC are characterized by carbon and oxygen as a major element with traces of calcium, magnesium, sodium, and potassium. OSBC and BGSBC show gold peaks at ∼2 and ∼10 keV which are due to gold coating on the biochar’s surface to make it conductive. The distribution of calcium, oxygen, and silica on the surfaces of OSBC and BGSBC is shown in EDX mapping images (Figures and 7).

Figure 4

SEM images of OSBC at (a) 49×, (b) 731×, (c) 6.40k×, and (d) 7.77k× magnifications.

Figure 5

SEM images of BGSBC at (a) 61×, (b) 2.75k×, (c) 5.81k×, and (d) 15.32k× magnifications.

Figure 6

SEM-EDX images of OSBC (a) multi-elements, (b) calcium, (c) oxygen, and (d) silicon distribution at 120× magnification.

Figure 7

SEM-EDX images of BGSBC (a) multi-elements, (b) calcium, (c) oxygen, and (d) silicon distribution at 103× magnification.

Energy dispersive X-ray fluorescence (EDXRF) analysis results obtained for OSBC and BGSBC is compiled in Table S1. Calcium, chlorine, potassium, phosphorus, and sulfur were present in both the biochars. Other elements were found in traces. OSBC has relatively higher concentrations of magnesium, aluminum, calcium, phosphorus, and sulfur versus BGSBC.

Transmission electron microscopy (TEM) images of OSBC and BGSBC at different magnifications are shown in Figure a–d. The primary particles of OSBC are globular in shape having a diameter in the range of 64–228 nm at 15 000× magnification [Figure b]. The shape of primary BGSBC particles is polygonal flakes with a diameter of 78–170 nm at 30 000× magnification [Figure d].

Figure 8

TEM images of OSBC at (a) 10 000× and (b) 15 000×, and BGSBC at (c) 10 000× and (d) 30 000× magnifications.

Equilibrium and Dynamic Studies

Batch equilibrium sorption studies at different pHs, contact time, temperature, and liquid to solid ratios were conducted. Sorption dynamics and equilibrium uptake isotherms are explained in subsequent sections.

Effect of pH

These experiments were conducted at a pH of 2, 4, 6, 8, and 10 (initial fluoride concentration = 10 mg/L and biochar dose = 0.5 g/L) (Figure ). Fluoride sorption decreased with the rise in the solution pH (Figure ). Best fluoride adsorption was recorded at pH 2.0 followed by a sudden decrease at pH > 2.0. Similar results were reported for fluoride adsorption on other biochars.[32,33,52] Biochar exhibits hydrophobic and hydrophilic, acidic and basic properties which are dependent on feedstock types, pyrolysis temperature, and residence time.[32] Hydroxyl, carboxylic acid, anhydride, ketone, ether, quinone, lactone, catechol, pyrone, and hydroxyl ketone were reported on the surface and subsurface of biochar.[53] Oxygenated functional groups in biochar contain lone pair electrons which do not participate in the adsorption of negatively charged fluoride ions due to repulsion. However, at low pH (<2.5), protonation of these basic active sites takes place on the biochar’s surface.[32] This protonation creates a net positively charged surface resulting in the electrostatic attraction of negatively charged fluoride ions.[32] Moreover, protonated groups can hold fluoride by electrostatic attractions as shown below[54]

Figure 9

Effect of initial pH on fluoride uptake on OSBC and BGSBC [initial F– conc = 10 mg/L; biochar concentration = 0.5 g/L; temp = 25 °C].

At lower pH, an electrophilic carbon is formed by the protonation and dehydration of hemicelluloses or cellulose residue if any remains on the surface of biochar (Figure a). The empty p-orbital on carbon overlaps with the filled p-orbital on an adjacent oxygen stabilizing the carbocation. This electron deficient cation reacts with the fluoride ion (Figure a). The ion exchange mechanism can also play a role in fluoride removal. The ion exchange occurs though quaternary ammonium sites (if available) on biochar surfaces (Figure b). The fluoride ion is strongly exothermically solvated by water. Thus, it needs strongly acidic groups to directly protonate. H-bonding by COOH to the solvated anion assists electrostatic attraction of fluoride (Figure c). Figure c shows how carboxyls and phenols might compete with H2O to solvate fluoride (Figure c). As this occurs on surfaces and within the swollen bulk of oxygen-rich pyrolysis biochars. An outside chance also exists for fluoride ion to react with SiO2 present in biochar (Figure d). These possible interactions are shown in Figure a−d.

Figure 10

Possible fluoride sorption mechanisms on OSBC and BGSBC (a) fluoride substitution onto protonated biochar at low pH, (b) ligand exchange mechanism, (c) hydrogen bonding onto biochar, and (d) fluoride ion reaction with SiO2.

At high pH, fluoride uptake decreased which can also be explained due to an increase in the competition between negatively charged hydroxyl and fluoride ions.[55] The equilibrium solution pH shifts toward neutral to alkaline as shown in Figure . Therefore, all the kinetic and equilibrium studies were carried out at an optimum pH of 2.0.

Effect of the Adsorbent Dose

The adsorbent concentration effect was studied at 1.0–2.5 g/L for OSBC and 0.5–2.5 g/L for BGSBC (pH = 2.0; fluoride concentration = 10 mg/L and equilibrium time = 24 h). The OSBC concentration effect on fluoride sorption is shown in Figure S3a. The fluoride uptake increases substantially from 70 to 85% with an increase in OSBC concentration from 1.0 to 2.0 g/L. However, fluoride uptake rises by only 6% on additional increase in the biochar concentration from 2.0 to 2.5 g/L. The time required by an adsorbent to reduce 50% of the initial adsorbate concentration is known as half-life (t50). The t50 of fluoride uptake decreased with the rise in the biochar concentration. The half-life values are provided in Table [t50 = 0.5 h (at 2.5 g/L), t50 = 1 h (at 2 g/L), and t50 = 2 h (at 1 g/L)]. The BGSBC concentration effect on fluoride sorption is shown in Figure S3b. Fluoride sorption increases from 40 to 95% when the BGSBC concentration increases from 0.5 to 2.5 g/L. The half-life (t50) obtained for fluoride uptake on BGSBC at different doses is summarized in Table [t50 = 0.5 h (at 2.5 g/L); t50 = 1 h (at 1.25 g/L); and t50 = 3 h (at 0.5 g/L)]. Thus, t50 values for BGSBC decrease with increasing adsorbent concentration.

Table 2

Rate constants along with Experimental and Calculated qe Values at Different Biochar Concentrations [Initial F– Conc = 10 mg/L; pH = 2.0; Temp = 25 °C]

biochar concentration (g/L)	first-order rate constant k1 (h–1)	R2	second-order rate constant (k2) (g mg–1 h–1)	R2	qe experimental (mg/g)	qe, calculated using first-order kinetic model (mg/g)	qe, calculated using second-order kinetic model (mg/g)	half-life (t50) (h)
OSBC
1	0.099	0.263	0.247	0.981	6.514	2.999	6.993	2.0
2	0.041	0.149	0.210	0.997	4.084	1.130	4.292	1.0
2.5	0.034	0.041	0.062	0.998	3.910	0.596	4.032	0.5
BGSBC
0.5	0.069	0.652	0.053	0.980	8.783	3.656	9.346	3.0
1.25	0.051	0.506	0.091	0.987	5.809	2.163	6.173	1.0
2.5	0.005	0.004	0.468	0.996	4.248	0.313	4.255	0.5

Kinetic Modeling

Kinetics studies along with thermodynamic parameters are important to determine adsorbent performance and sorption mechanism. The kinetic performance of an adsorbent is of great significance in determining pilot scale applications of an adsorbent. With the kinetic data, solute uptake rate can be determined. Thus, adsorption kinetics can be considered as the foundation for fixed-bed studies.[56] Fluoride kinetic data were modeled to the pseudo-first and pseudo-second order rate equations [Figure a,b].[57−59] All pseudo-first and pseudo-second order parameters are compiled in Table . Excellent regression coefficients [0.981, 0.997, and 0.998 for OSBC (adsorbent dose 1.0, 2.0, and 2.5 g/L) and 0.980, 0.987, and 0.996 for BGSBC (adsorbent dose 0.5, 1.25, and 2.5 g/L)] were obtained for the pseudo-second order rate equation. The theoretical qe values calculated by the pseudo-second order equation are also very close to experimental qe values. Thus, fluoride adsorption on biochars is controlled by pseudo-second order rate kinetics. Thus, electrostatic attractions between the fluoride ions and adsorbent can be assumed to be the rate controlling step.[54]

Figure 11

Fluoride sorption kinetic data fitted to the pseudo-second order equation (a) OSBC and (b) BGSBC [initial F– conc = 10 mg/L; pH = 2.0; temp = 25 °C].

Isotherm Modeling

Fluoride adsorption isotherm studies on OSBC and BGSBC were performed at 25, 35, and 45 °C (pH of 2.0). The sorption data were modeled to Langmuir,[60] Freundlich,[61] Koble–Corrigan,[62] Redlich–Peterson,[63] Sips,[64] Radke and Prausnitz,[65] Toth,[66] and Temkin[67] isotherm equations using SigmaPlot 12.0 software. Discussions about isotherms with their equations are provided in the Section S3. All the fluoride adsorption isotherms for OSBC and BGSBC are positive and regular in shape (Figures and S4–S10). Fluoride uptake is independent of temperature. Isotherm parameters along with their regression coefficients (R2) are compiled in Table .

Figure 12

Langmuir sorption isotherms of fluoride removal by (a) OSBC and (b) BGSBC at 25, 35 and 45 °C [pH = 2.0; contact time = 48 h; biochar concentration = 2.5 g/L].

Table 3

Isotherm Parameters Obtained for Fluoride Uptake at 25, 35, and 45 °C [pH = 2.0; Contact Time = 48 h; Biochar Concentration = 2.5 g/L]

	OSBC			BGSBC
isotherm parametersa	25 °C	35 °C	45 °C	25 °C	35 °C	45 °C
Freundlich
KF (mg/g)	5.77	5.50	6.10	4.70	4.09	4.43
1/n	0.31	0.33	0.29	0.30	0.33	0.32
R2	0.70	0.69	0.68	0.83	0.88	0.85
Langmuir
Q0 (mg/g)	19.67	19.95	18.84	15.48	15.45	16.05
B	0.23	0.21	0.27	0.27	0.22	0.23
R2	0.75	0.64	0.69	0.91	0.91	0.92
Redlich–Peterson
KRP (l/g)	3.60	3.74	4.88	3.97	4.64	3.48
aRP (l/mg)	0.11	0.11	0.24	0.23	0.46	0.20
βRP	1.14	1.30	1.02	1.02	0.89	1.02
R2	0.75	0.76	0.69	0.91	0.91	0.92
Sips
KLF (l/g)	0.0086	3.91 × 10–5	6.79	3.82	3.87	3.12
aLF (l/mg)	0.0005	2.37 × 10–6	0.29	0.26	0.23	0.20
nLF	6.6424	14.0	0.62	1.33	0.82	1.77
R2	0.84	0.92	0.70	0.91	0.91	0.93
Koble–Corrigan
a	0.0086	3.78 × 10–5	6.79	3.82	3.87	3.11
b	0.0005	2.28 × 10–6	0.29	0.26	0.23	0.20
β	6.64	14.04	0.62	1.13	0.82	1.17
R2	0.84	0.92	0.70	0.91	0.91	0.93
Radke and Prausnitz
a	4.18 × 106	6.43 × 107	1.49 × 107	1.53 × 108	4.3 × 107	3.45 × 107
r	5.76	5.50	6.10	4.70	4.09	4.43
b	0.03	0.33	0.28	0.30	0.33	0.32
R2	0.70	0.69	0.68	0.83	0.88	0.85
Toth
KT	16.06	16.96	29.42	14.85	17.88	15.20
BT	0.16	0.17	1.23	0.24	0.30	0.19
βT	0.38	0.14	2.60	0.82	1.47	0.78
R2	0.77	0.79	0.69	0.91	0.91	0.92
Temkin
B		5.06	4.63	3.89	3.83	4.00
a		1.00	1.00	1.00	1.00	1.00
R2		0.60	0.17	0.65	0.76	0.75

Terms in this column are defined in Supporting Information.

Koble–Corrigan and Sips models better fitted the isotherm data versus other models for OSBC as shown by the regression coefficient (R2) values (>0.91). In the case of BGSBC, Koble–Corrigan, Langmuir, Redlich Peterson, and Sips isotherms better fitted the sorption data. Furthermore, isotherm models better fitted the fluoride adsorption data obtained for BGSBC versus OSBC. The maximum Langmuir adsorption capacities (Q°) of 19.95 mg/g (at 35 °C) and 16.05 mg/g (45 °C) were obtained for OSBC and BGSBC, respectively. Fluoride is better removed by OSBC than BGSBC. Both OSBC and BGSBC performed better or comparable to reported sorbents applied for aqueous fluoride removal (Table ).

Table 4

Comparative Evaluation of Defluoridation Capacities of Different Adsorbents in Aqueous Solution

adsorbents	pH	Temp. (°C)	surface area (m2/g)	conc range (mg/L)	adsorption capacity (mg/g)	refs
OSBC	2.0	25	23.052	2–100	19.67	this study
		35			19.95
		45			18.84
BGSBC	2.0	25	9.27	2−100	15.48	this study
		35			15.45
		45			16.05
pine wood char	2.0	25	2.73	1–100	7.66	(32)
		35			6.34
		45			4.46
pine bark char	2.0	25	1.88	1–100	9.77	(32)
		35			10.53
		45			8.40
corn stover biochar	2.0	25	ND	1–100	6.42	(33)
		35			5.17
		45			5.00
magnetic corn stover biochar	2.0	25	3.61	1–100	4.11	(33)
		35			3.45
		45			3.41
activated alumina	7.0	RTa	250	2.5–14	2.41	(85)
hydrous manganese oxide-coated alumina	5.2	30	315.5	10–70	7.09	(86)
magnesia-amended activated alumina	7.0	30	193.5	5–150	10.12	(21)
manganese dioxide-coated activated alumina	4.0	RTa	203	1–25	0.17	(87)
bone char	6.0	RTa	99.1	2.5–100	6.1	(88)
aluminum-coated bone char			91.8		11.4
aluminum-impregnated wood char			284		16.8
activated bagasse carbon	6.0	28		2.5–15	1.15	(89)
raw sawdust					1.73
raw wheat straw					1.93
calcium chloride-modified natural zeolite	6.0	25		25–100	1.77	(24)
activated kaolinites	3.0	30	32.43	2–10	0.60	(16)
magnesia-loaded fly ash cenospheres	3.0	45		5–100	6.0	(29)
aluminum hydroxide-coated rice husk ash	5.0	27	50.4	10–60	15.08	(90)
granular red mud	4.7	25	10.2	5–150	0.85	(91)
magnetic douglas fir biochar	3.0	25	663	1–60	7.81	(92)
		35			9.04
		45			7.58

Room temperature.

Defluoridation of Groundwater Samples

Details of the groundwater samples are summarized in Table . The groundwater of the study area is moderately alkaline with a mean pH of 8.23 (range 7.61–8.51). The mean value of total dissolved solids (TDS) and conductivity are 1771 mg/L (range 145–3683 mg/L) and 3614 μs/cm (range 295–7500 μs/cm), respectively. The mean fluoride concentration in the groundwater samples is 2.0 mg/L (range 0.1–5.9 mg/L). About 45% of the total groundwater samples exceed a BIS permissible limit of 1.5 mg/L. Therefore, the groundwater samples are characterized by high fluoride require defluoridation before consumption.

Table 5

Groundwater Quality Parameters of Collected Samples from the Study Area

sample. no.	landmark	pH	Cond. (μs/cm)	TDS (mg/L)	Na+ (mg/L)	K+ (mg/L)	Ca2+ (mg/L)	T. Alk. (mg/L as CaCO3)	HCO3– (mg/L as CaCO3)	SO42– (mg/L)	NO3– (mg/L)	F– (mg/L)	Cl– (mg/L)
1	Haily Mandi	7.82	1550	766	376	4.4	30.6	600	812	57	11.7	1.8	10
2	Haily Mandi	8.22	3850	1885	813	7.4	69.4	380	478	489	6.6	1.1	82
3	Haily Mandi	8.36	3110	1529	670	6.5	58.6	380	517	396	7.3	1.2	67
4	Haily Mandi	8.34	2789	1367	641	6.4	56.9	590	800	379	6.7	1.0	37
5	Haily Mandi	8.05	7500	3681	1622	10.8	113.3	180	219	1179	4.5	1.2	217
6	Patauda	8.06	7500	3683	1662	11	109.9	520	705	1183	11.3	2.9	217
7	Patauda	8.48	3810	1864	912	7.2	52.7	650	844	579	12.6	1.5	57
8	Patauda	8.37	4810	2362	1152	8.1	54.9	430	566	425	14.4	2.0	100
9	Patauda	8.17	3430	1676	804	4.6	42.6	620	807	256	10.2	1.0	70
10	Patauda	8.35	295	145	22	22.5	10.6	180	146	34	0.5	0.4	3
11	Patauda	8.51	1951	956	357	2.8	25.1	400	461	183	4.4	3.7	40
12	Kulana	8.21	727	356	91	81.6	22.5	180	215	35	3.4	0.1	15
13	Kulana	8.23	5840	2862	550	604	64.9	210	226	429	12.9	1.5	182
14	Amadalpur	8.42	1879	920	255	4.2	29.1	220	258	210	2.7	4.2	50
15	Amadalpur	8.50	5530	2702	1256	146	55.8	760	1027	651	11.4	5.9	130
16	Amadalpur	8.35	3570	1752	733	21.7	45.7	340	422	186	12.7	3.5	102
17	Amadalpur	8.08	3160	1547	433	38.1	47.3	430	546	261	10.9	2.1	87
18	Amadalpur	7.61	3750	1839	585	6.6	63.6	440	541	310	13.2	1.0	107
min		7.61	295	145	22	2.8	10.6	180	146	34	0.5	0.1	3
max		8.51	7500	3683	1662	604	113.3	760	1027	1183	14.4	5.9	217
mean		8.23	3614	1771	718	55.2	53.0	417	533	402	8.7	2.0	88

Thus, groundwater defluoridation was carried out using OSBC and BGSBC. The adsorption experiments were performed at 25 °C [dose = 2.5 g/L, pH 2.0 and contact time = 24 h]. Both initial and final fluoride concentrations after adsorption are given in Figure a,b. After adsorption, the fluoride concentration in almost all groundwater samples decreased to <1.5 mg/L of the permissible limit.[2] This clearly demonstrates that developed OSBC and BGSBC have the potential to successfully remediate fluoride from the groundwater.

Figure 13

Groundwater defluoridation using (a) OSBC and (b) BGSBC at pH = 2.0; adsorbent dose = 2.5 g/L and temp = 25 °C [*sample number location is provided in Table ].

Fixed-Bed Studies

OSBC was used in fixed-bed studies to evaluate Vb, V, Cb, C, tb, tf, t, tδ, δ, f, percent saturation, breakthrough capacity, and the capacity at exhaustion point (Table ). The fluoride removal breakthrough curve for OSBC is shown in Figure . The shape of this curve is the “S” type.

Table 6

Fixed-Bed Column Parameters for Fluoride Uptake onto OSBC

parameters	OSBC
Co (mg/mL)	0.0053
Cb (mg/mL)	0.0004
Cx (mg/mL)	0.0038
Vb (mg/cm2)	6.07
Vx (mg/cm2)	15.15
Vx – Vb (mg/cm2)	9.08
D (cm)	8.0
Fm (mg/cm2/min)	0.004
tx (min)	3594
Tf (min)	612
tb (min)	997
Tδ (min)	2155
f	0.54
δ (cm)	5.78
EBCT (min)	6.28
% saturation	66.57
char usage rate (g/L)	1.39

Figure 14

Breakthrough curve at an initial fluoride concentration = 5 mg/L; flow rate = 4 mL/min; pH = 2.0; bed depth = 8 cm and temperature = 25 °C.

The total time (tx) taken by the primary sorption zone for moving downward out of bed is 3594 min. The (V – Vb) value of 9.08 mg/cm2 indicates additional fluoride adsorbed per unit OSBC area. The primary adsorption zone length (δ) is 5.78 cm. The time required for initial formation (tf) of PAZ is 612 min. Breakpoint was achieved after time (Tb) 997 min. The time, tδ, required for the PAZ downward movement in the column after its formation is 2155 min. An empty bed contact time of 6.28 min and char usage rate of 1.39 g/L were obtained. The fractional capacity “f” at the breakpoint to continue in the adsorption zone of the column for fluoride removal from the solution is ∼0.54. The percent saturation at a breakpoint is 66.57. A laterite fixed-bed percent saturation range of 72–92 for fluoride removal was reported earlier.[68]

The breakthrough and column (exhaustion) capacities of 2.6 and 6.0 mg/g were obtained. Lower column capacity (6.0 mg/g) than batch sorption capacity (19.7 mg/g) was observed. This may be due to the short contact time of fluoride with OSBC, which require more time to be absorbed. These column parameters may be useful in designing a large scale fixed-bed reactor for aqueous fluoride removal.

Materials and Methods

All chemicals used in this study were of general or analytical grade. SPADNS, sodium fluoride (NaF) (99%), HCl (35%), NaOH, HNO3 (69%), and zirconium(IV) oxide chloride octahydrate were purchased from Merck India Pvt. Ltd. Fluoride (1000 mg/L) stock solution was made by taking 2.21 g of sodium fluoride and dissolving in double distilled water. All fluoride solutions (2−100 mg/L) were freshly prepared from the stock solution. The pH of working solutions was maintained using 0.1 N NaOH and 0.1 N HNO3.

Adsorbents Preparation

Okra (Abelmoschus esculentus) stems (OSs) and black gram (Vigna mungo) straws (BGSs) were used for biochar preparation. Both feedstocks (OS and BGS) were collected from the agriculture field. These were chopped into small pieces, sun dried, and used for biochar preparation. OS and BGS were slow pyrolyzed for 30 min in the absence of oxygen at 600 and 500 °C, respectively. Pyrolysis temperatures and residence time were decided after performing preliminary fluoride sorption studies on different biochars obtained at (300, 400, 500, 600, and 700 °C). The detailed optimization results are summarized in Section S1. Prepared biochars were grinded, sieved, and different sizes were stored in air-tight containers. Biochars of 30–50 B.S.S. mesh sizes were used for further sorption experiments.

Characterization of Developed Biochars

Proximate and Ultimate Analysis

Carbon, hydrogen, and nitrogen were determined using a CHNS analyzer (model LECO CHNS-932). The ash content, moisture content, and volatile matter were estimated according to the American Society for Testing and Materials method (ASTM D-1762-84). The oxygen content in biochar is evaluated by the difference. The surface area, average pore size, and total pore volume of OSBC and BGSBC were determined using a surface area analyzer (model Micromeritics ASAP 2010). The OSBC and BGSBC samples were degassed for 12 h at 250 °C and <10–3 Torr pressure prior to surface area measurements. The Brunauer–Emmett–Teller (BET) method was used to quantify the specific surface area at a relative pressure P/P0 of 0.2.[69] The pore size distribution was measured using the BJH method.[70] The total micropore volume (W0) of the biochars was measured using the Dubinin–Radushkevich (D–R) equation (eq ).[71]where, W is the micropore volume, W0 is the total micropore volume, P/P0 is the relative pressure, and D is the adsorbent’s characteristic constant.

Powdered X-ray Diffraction

The XRD spectra of OSBC and BGSBC were obtained by a PANalytical X-ray diffractometer (X’Pert PRO) through Cu Kα (λ = 1.54 Å) X-ray irradiation@40 mA and 45 kV. The biochar samples were scanned at a 2θ range of 5.025–89.975° with 2° per min scan speed and 0.33° step size.

FTIR Spectroscopy

The functional group’s presence in OSBC and BGSBC were identified using a FTIR spectrometer (model Spectrum Two PerkinElmer). Biochar was homogeneously mixed using a pestle–mortar with potassium bromide in a 1:20 ratio and its pellets were developed by applying 10 ton pressure through a hydraulic press (model PCI Analytics, CAP 15T). FTIR spectra of OSBC and BGSBC were obtained from 500 to 4000 cm–1 in the transmittance mode.

Scanning Electron Microscopy and Energy Dispersive Analysis

The surface morphologies of OSBC and BGSBC were investigated using SEM (model Zeiss, EVO40) at 20 kV/mA. The powdered biochar sample was mounted on a copper stab with carbon tape and was gold coated to make its surface conducting. The sample surfaces were examined at different magnifications. A Bruker EDX system conjugated with same SEM was used for biochar elemental composition analysis.

Transmission Electron Microscopy

TEM, model JEOL 2100F (200 kV potential) was employed to examine biochar samples. The biochar sample was suspended and ultrasonicated for 10 min in ethanol. A suspension drop was put up on the copper grid and dried. This sample-loaded grid was observed using TEM at various magnifications.

Energy Dispersive X-ray Fluorescence Spectroscopy

A EDXRF spectrometer (model PANalytical Epsilon 5) was used for analyzing biochar’s chemical composition. The powdered biochar sample (∼1.0 g) was uniformly spread and sealed in a plastic cup using transparent mylar foil and analyzed.

Sorption Experiments

Batch sorption experiments were performed to determine kinetic and isotherm parameters. Sorption experiments were conducted at different temperatures, pH values, adsorbent doses, and contact times. Sorption isotherms were in a 2–100 mg/L fluoride concentration range. Fluoride concentrations were determined by the SPADNS method.[72] The fixed biochar dose was mixed in a 50 mL fluoride solution and the suspension was agitated in a water bath shaker (model MSW275, Macro Scientific, India) for a specific time (up to 48 h). The solutions were filtered and analyzed at 570 nm using a UV/vis spectrophotometer model Lamda 35, PerkinElmer. Fluoride uptake on biochar was determined using eq .where, qe (mg/g) is the amount of fluoride adsorbed by per gram biochar, V (L) is the fluoride solution volume, M (g) is the biochar amount, and Co and Ce (mg/L) are the initial and equilibrium fluoride concentrations, respectively.

Solution pH is a crucial factor in fluoride adsorption. About 0.025 g of biochar was added into 50 mL fluoride solution (10 mg/L). Fluoride adsorption was carried out at a pH of 2, 4, 6, 8, and 10. The pH adjustment of fluoride solutions was made using 0.1 N HNO3 and 0.1 N NaOH. The adsorption study was conducted at an equilibrium time of 48 h and at 25 °C under constant stirring in a water bath shaker. The percent fluoride uptake was determined using eq .where, Co and Ce (mg/L) are the initial and equilibrium fluoride concentrations.

Effect of Adsorbent Dose

To optimize the adsorbent concentration, different biochar doses (0.5, 1.25, 2.5 g/L for BGSBC and 1, 2, 2.5 g/L for OSBC) were added to the 50 mL fluoride solution (10 mg/L and pH 2.0) at 25 °C. An optimum biochar dose was determined by calculating the percent fluoride removal at different time intervals.

Sorption Equilibrium Experimentations

Sorption equilibrium experiments are necessary to determine the adsorption capacity required for designing a fluoride removal system. Adsorption isotherm gives the relation between adsorbent capacity and fluoride concentration at a constant pH and temperature.[73] The equilibrium sorption experiments were performed at 25, 35, and 45 °C (initial fluoride conc = 2–100 mg/L; equilibrium time = 48 h).

Groundwater Defluoridation

Many cations and anions are present in actual groundwater.[74−78] These ions may affect the biochar adsorption capacity determined using simulated water in lab conditions. The presence of these ions may decrease, increase, or do not change the adsorption of a desired contaminant.[79] Therefore, interference due to the coexistence of ions on fluoride sorption on developed biochars was investigated. Finally, biochar suitability was investigated by applying it to the groundwater samples collected from fluoride contaminated areas.

Amadalpur village (28° 27′ 12.52″N and 76° 39′ 58.45″E), Haily Mandi town (28° 20′ 50.72″N and 76° 45′ 33.45″E), Kulana village (28° 25′ 31.18″N and 76° 39′ 32.94″E), and Patauda town (28° 24′ 12.34″N and 76° 40′ 38.80″E) in Haryana, India were chosen for the defluoridation study. Eighteen groundwater samples were taken in precleaned and sterilized 2 L capacity polyethylene bottles from hand-pumps or tube wells. Physicochemical parameters of the groundwater samples were determined using the following standard methods.[72] Both OSBC and BGSBC were applied for groundwater defluoridation. Fluoride sorption studies were conducted in a similar fashion as discussed in Section .

Fixed-Bed Studies

A glass column with an internal diameter of 2 and 40 cm height with a feed reservoir of 10 000 mL capacity at the top of the column was used (Figure ). Hot double distilled water was used to make a biochar slurry. The slurry was then poured into the glass wool supported column to prepare a biochar fixed bed. A fluoride solution of 5 mg/L (pH 2.0) was fed into the column at a flow rate of 4 mL/min. The column was operated in a downward flow mode at 25 °C. Eluent was collected at an interval of 30 min. The absorbance was determined at a wavelength of 570 nm using a UV/vis spectrophotometer following the SPADNS method.[72]

Fixed-bed parameters were determined using a mass transfer model developed by Weber and later used previously in many sorption studies.[46,80−82] The column performance can be explained by means of a breakthrough curve.[81−83]Figure S11 shows an ideal breakthrough curve, where C and Ve are the fluoride concentrations in the effluent and volume of fluoride free water passing through 1.0 cm3 of the biochar, respectively. Column breakthrough point was selected at a low effluent concentration, Cb and exhaustion point at C close to Co (influent concentration). The fixed-bed parameters including total time (t), mass flow rate (Fm), time required for PAZ movement (tδ), depth (δ), time needed for initial PAZ formation (tf), amount of fluoride adsorb (Ms), column fractional capacity (f), percent saturation, empty-bed-contact-time (time required for influent to be in contact with biochar bed), and biochar usage rate were calculated using eqs –10.[46,80−84]

Conclusions

OSBC and BGSBC were successfully prepared by the slow pyrolysis of OS and BGS at 600 and 500 °C, respectively. These biochars were characterized and successfully applied for groundwater defluoridation. The surface area of OSBC (23.52 m2/g) was smaller than that of BGSBC (9.27 m2/g). OSBC and BGSBC are characterized by carbon and oxygen as main elements and calcium, magnesium, sodium, and potassium in traces. The oxygen content of OSBC (21.99 wt %) was lower than that of BGSBC (24.67 wt %). For both biochars, maximum fluoride sorption was obtained at a pH of 2.0. The pseudo-second order rate expression better fitted the kinetic data as compared to the pseudo-first order rate expression, suggesting that chemical sorption is the dominating mechanism for fluoride adsorption on biochar. Langmuir adsorption capacities of QOSBC0 = 20 mg/g (pH 2.0; 35 °C; biochar dose 2.5 g/L) and QBGSBC0 = 16 mg/g were obtained (pH 2.0; 45 °C; biochar dose 2.5 g/L). OSBC and BGSBC adsorption capacities are comparable to the adsorption capacities of other adsorbents used for fluoride removal (Table ). Column study was performed with OSBC to determine the necessary parameters for a fixed-bed rector design. A column capacity of 6.0 mg/g for OSBC was achieved. Thus, developed biochars may be considered as effective adsorbents for fluoride removal from water.