Characterization of raw LLP. Ultimate and proximate analysis. Table 1 shows the results for ultimate and proximate analysis of raw LLP and its comparison with other studies. Elemental analysis results show that raw LLP is mainly composed of carbon (47.6%) and oxygen (38.61%). The hydrogen and nitrogen contents were 6.097% and 2.586% respectively, with no sulfur. The result is similar to those reported by Šoštarić et al. 28, Nuithitikul et al. 32, and Martín-Lara et al.34, who investigated the element composition of raw apricot shells, cashew nut shells, and raw olive stone respectively, which showed high oxygen content. Raw LLP also showed higher ash content as it contains more lignin content. This was confirmed by Sonia and Priya Dasan 35, who found similar data for this finding. This is probably due to sample variety and different biomass sources, reflecting the difference in soil contamination. The proximate analysis in this study in agreement with Šoštarić et al. 28 which found the bulk density of raw LLP is 0.33 g/ml.
Table 1
The ultimate and proximate analysis of raw LLP and its comparison with other studies
Raw material
|
Ultimate, %
|
Proximate, %
|
References
|
C
|
H
|
N
|
O
|
S
|
Ash
|
Bulk density, g/ml
|
moisture
|
|
Raw LLP
|
47.208
|
6.097
|
2.586
|
38.61
|
ND
|
5.5
|
0.33
|
2.3
|
This study
|
Raw olive stone
|
52.34
|
7.11
|
0.03
|
40.47
|
< 0.01
|
0.37
|
-
|
5.43
|
Martín-Lara et al. (2013)
|
Cashew nut shells
|
45.93
|
5.75
|
0.62
|
47.57
|
ND
|
-
|
-
|
-
|
(Nuithitikul et al., 2020)
|
Apricot shells
|
47.6
|
6.24
|
0.12
|
46.03
|
ND
|
1.02
|
1.15
|
7.30
|
(Šoštarić et al., 2018)
|
Persimmon leaf
|
57.8
|
5.4
|
0.83
|
35.94
|
0.03
|
3.23
|
-
|
1.4
|
(Lee and Choi, 2018)
|
Biochemical Analysis. The percentages of hemicellulose, lignin, and cellulose present in the raw LLP were 17.677%, 31.2%, and 18.1545%, respectively as shown in Table 2. These results show that the lignocellulosic component of raw LLP mainly consists of lignin, followed by cellulose and hemicellulose. These results agree well with existing studies 36. Arshad and Imran 37 reported that biosorbent consists of lignin, cellulose, and hemicellulose, which make them effective removal metal cation. However, the percentage of each component varies depending on the material source of the biosorbent, plant genetics, the environment where the plant was grown, and processing conditions of the plant 38.
Table 2
Percentual values of the biomass content in the raw LLP and comparison with other studies
Adsorbent
|
Property
|
Reference
|
Holocellulose,%
|
Cellulose,%
|
Hemicellulose,%
|
Lignin,%
|
extractive,%
|
Raw LLP
|
35.8313
|
18.1545
|
17.6768
|
31.2
|
2.5
|
This study
|
Flamboyant pods
|
-
|
14.7
|
32.9
|
48.5
|
0.4
|
(di Bitonto et al., 2021)
|
Flamboyant pods
|
-
|
34
|
35
|
16.5
|
11.75
|
(Oluwasina et al., 2020)
|
Bauhinia pods
|
-
|
37.5
|
33.5
|
17.5
|
10.3
|
(Oluwasina et al., 2020)
|
Native apricot shells
|
74.46
|
55.23
|
19.23
|
22.72
|
2.56
|
(Šoštarić et al., 2018)
|
Fourier transform infrared spectroscopy (FTIR). Figure 2 shows FTIR analysis of (a) raw LLP and (b) Pb loaded with LLP. Table 3 shows peak positions in the FTIR spectrums of raw LLP and Pb loaded with raw LLP, which follows the range given by Pavia et al. 39. Figure 2a shows the spectrum of raw LLP. The broad and intense peak around 3200–3500 cm− 1 was assigned to the stretching of O–H group, carboxylic acid, and amino groups, in which overlapping bands of O–H and N–H stretching vibrations were observed. Vibrational bands of –OH groups in hydrogen bonded and chemisorbed water, as well as inter and intramolecular hydrogen bonding. This O–H bonding contains polymeric compounds (macromolecular associations), such as alcohols, phenols, and carboxylic acids, which exist in cellulose and lignin. Cellulose connects with hemicellulose or lignin molecules mainly through hydrogen bonds, while the connections between hemicellulose and lignin include both hydrogen and covalent bonds40. The absorption band at 2913.95 cm− 1 could be assigned to asymmetric vibration of C–H stretching due to methyl and methylene groups in the polymers in the biosorbent such as lignin, cellulose, and hemicellulose 41. The stretching vibration band at 2850 cm− 1 is due to C–H stretching of aldehyde hydrogen (C–H–O). The stretching vibration band at 1718 cm− 1 is due to carbonyl group (C = O) stretching from carboxylic acid functional group 42. The small peak at 1718 cm− 1 appears in the spectrum of raw LLP. This peak is due to the stretching vibration of C = O bond (carbonyl group) due to non-ionic carboxyl groups (–COOH, –COOCH3), and may be assigned to carboxylic acids and ketones. The band at 1607.65 cm− 1 is attributed to carbonyl groups of ketone which is describe C = O stretching. The peaks at 1315.45 cm− 1 are due to C–O stretching of carboxylic group. Besides that, the band at 1515.13 cm− 1 corresponds to aromatic ring for the symmetrical C = C vibration in the lignin component. The observations are in line with the results reported by Md Salim et al. 43 and Ibrahim et al. 44. The peaks from 1000 to 1300 cm− 1 indicate C–O stretching of these functional groups with single oxygen bonds of carboxylic acids, ketone, and alcohols. The sharp peak appearing at 1033.46 cm− 1 is due to C–OH stretching in cellulose, hemicellulose, and lignin 45. The deformation vibrations of the C–H bond in the benzene rings give absorption bands in the 840–730 cm− 1 range.
Table 3
Peak position in the FTIR spectrums of LLP and Pb loaded with raw LLP
Transmittance peak wave number indicate by the range of samples of wavenumber (cm-1)
|
Raw LLP
|
Pb loaded with raw LLP
|
3400 − 3200
|
Broad band indicates the carboxylic acid
|
3310.5
|
3295
|
3400 − 3300
|
Broad band OH groups of alcohol
|
3310.5
|
3295
|
3500 − 3300
|
N-H stretching for primary amines
|
3310.5
|
3295
|
3000 − 2840
|
C-H stretching
|
2913.93
|
2918.03
|
2860 − 2700
|
medium bands of C-H stretching in aldehyde
|
2852.45
|
-
|
1730 − 1700
|
carbonyl group (C = O) stretch from carboxylic acid
|
1718
|
-
|
1720 − 1708
|
Aliphatic ketone
|
1718
|
-
|
1670 − 1600
|
Conjugation of two aromatics rings for C = O.
|
1607.65
|
1603.57
|
1650 − 1580
|
N-H bending for primary and secondary amines
|
1607.65
|
1603.57
|
1500 − 1400
|
Deformation C-C bond
|
1438.05
|
-
|
1515
|
Aromatic ring for the symmetrical C = C vibration
|
1513.13
|
-
|
1400 − 1350
|
Aldehyde C-H bending vibration
|
1374.71
|
1374.71
|
1364.49
|
1362.45
|
1300 − 1000
|
C = O stretching vibration of ether groups (R-O-R)
|
1096.8
|
1096.8
|
|
Bending vibration of C-CO-C group of ketone
|
|
|
1320 − 1210
|
C-O stretching of carboxylic group
|
1315.45
|
-
|
1200 − 1100
|
Aliphatic ketone
|
|
|
1050 − 1017
|
Primary alcohol
|
1033.46
|
1027.71
|
840 − 730
|
The deformation vibrations of the C–H bond
|
827.07
|
-
|
730 − 550
|
C-Cl bending vibrations
|
665.64 590.03 510.34
|
-
|
Figure 2b for Pb(II)-adsorbed LLP shows that some peaks decreased. The shifts may be attributed to the changes due to bonding of functional group to Pb(II). The broad peaks detected at 1570–1700 cm− 1 are weaker because of the reduction of the carboxylic acid, ketone, and C–C bond. FTIR spectra of metal Pb(II)-adsorbed LLP show that the peaks expected at 3310.5, 2913.93, 1607.65, 1374.71, 1364.49, and 1033.46 cm− 1 shifted, respectively, to 3295.16, 2918.03, 1603.65, 1374.71, 1362.45 and 1091. Peaks 2852.45, 1718, 1515.70, 1438.05, 1315.45, 827, and 1163 cm− 1 disappeared after adsorption. A decrease in transmittance, small deflection, and disappearance of band frequency indicate that there may be interactions due to the exchange of hydrogen ions with Pb(II) and functional groups 46,47. It is well indicated from the FTIR spectrum of raw LLP that carboxyl and hydroxyl groups were present in abundance. In biopolymers, these groups may function as proton donors; hence, deprotonated hydroxyl and carboxyl groups may be involved in adsorption of Pb(II).
Surface morphology.The SEM micrographs (1500× and 500× magnification) before Pb(II) adsorption are shown in Fig. 3a and b and the micrographs (5000× and 2000× magnification) after Pb(II) adsorption are shown in Fig. 3c and d. Morphological examination by SEM analysis for raw LLP revealed that it is soft-flat shaped with rough pits (Fig. 3a). The cross-section of raw LLP has some distorted pores and the surface of raw LLP has an irregular surface and a fixed shape and size. It also does not have any defined holes (only a few pores on the surface), and it has a small surface area. This is because wax and lignin surround the raw LLP surface as a protective layer 48. No significant fine pore development on the surface of the samples was observed Fig. 3b shows the self-assembled structure of cellulose fibers, which is due to the strong interfibrillar attraction between the surface hydroxyl groups. Cellulose also connects with hemicellulose or lignin molecules in raw LLP. Figure 3c and d clearly show that the Pb(II) adsorbed to the raw LLP surface but not deep within the pores as expected. The rough and uneven surface of raw LLP facilitated the adsorption of Pb(II). This result supports the assumption that the pore structure is not the only determining factor influencing the adsorption capacity of heavy metals.
Energy dispersive X-ray analysis (EDAX). Fig. 4 shows the EDAX pattern for raw LLP before and after Pb(II) adsorption. The EDX pattern in Fig. 4a for the raw LLP did not show the characteristic signal of Pb(II), whereas after adsorption, as shown in Fig. 4b, a clear signal of the presence of Pb(II) was observed. Furthermore, the presence of Ca2+ and K+, which have been shown to be involved in the ion exchange with Pb(II), was indicated in the spectra of unloaded raw LLP. After the adsorption of Pb(II) onto raw LLP, these peaks cations were decreased in the EDX spectrum of Pb(II)-loaded raw LLP. Iqbal et al. 49 reported the use of EDAX analysis of grapefruit peel to confirm the mechanism of ion exchange for the removal of Zn(II) from aqueous solutions. The report also noted that the calcium and potassium ions identified in the EDAX spectrum of the fresh grapefruit peel were absent in the peel used for Zn(II) ion adsorption, thus suggesting that these ions may be involved in the ion exchange with the Zn(II) ions. Hence, it could be assumed that Pb(II) was bound to the raw LLP over ion exchange mechanism between Ca2+ and K+ and Pb(II).
XRF analysis. Table 4 displays the results of the XRF analysis, as well as information on the inorganic chemical composition of raw LLP and Pb(II)-adsorbed raw LLP. Potassium (K) has a significant percentage (14.955%) in raw LLP compared to other inorganic constituents. After the adsorption process, there was a marked reduction of K content in the raw LLP. The elemental composition of the raw LLP is similar to what has been reported by Elhafez et al. 50 for rice husk in adsorption of Cu(II) ion. This analysis reveals that K contributes significantly to the adsorption due to the formation of Pb on the raw LLP surface. The XRF analysis show that K percentage was reduced from 14.955–3.781% after adsorption. These results are in line with a recent study 51, showing that ion exchange mechanism between K+ and Pb2+ is involved in the adsorption process.
Table 4
XRF of raw LLP and Pb adsorbed with LLP
Adsorbent
|
Elements
|
Al
|
Si
|
P
|
S
|
Cl
|
K
|
Ca
|
Mn
|
Fe
|
Ni
|
Cu
|
Cd
|
Pb
|
Raw LLP, wt%
|
0.148
|
0.229
|
0.155
|
0.54
|
0.884
|
14.955
|
4.349
|
0.696
|
69.375
|
6.67
|
0.264
|
0.074
|
1.655
|
Adsorbed LLP with Pb, wt%
|
0.128
|
0.191
|
0.076
|
0.366
|
0.109
|
3.781
|
5.331
|
0.671
|
64.844
|
5.37
|
0.215
|
0.136
|
18.782
|
X-ray diffraction. The XRD patterns of raw LLP before and after adsorption with Pb(II) are shown in Fig. 5. Figure 5 (a) shows a wide broad strong and sharp diffraction peak in the 2θ range of 22.321°–24.403° and a weak peak at 2θ = 14°–16° which indicates crystalline matter. However, the intensity of raw LLP adsorbed with Pb(II) decreased, as shown in Fig. 5 (b). This observation agrees with Pereira et al. 53 who found that biosorbent from carnauba straw and cashew leaf showed an intense peak at about 21.32°, which was attributed to the crystalline structure of cellulose. Gassan and Bledzki54 found that chain scissions in cellulose will increase the crystallinity of the cellulose. These findings demonstrate that the cellulose structure is responsible for the adsorption. Salamun et al. (2015) reported that the cellulose molecules are arranged in ordered lattice and most of the OH groups are bonded by hydrogen bond (C–H-O), which can attract Pb(II) ions.
Thermogravimetric analysis (TGA) .Thermal degradation process of raw LLP can be divided into three regions as shown in Fig. 6a. The degradation temperature was obtained from the TG and DTG curves as shown in Table 5. At region I (< 200°C), raw LLP did not undergo significant thermal degradation because of the hydrophilic character of the lignocellulose materials occurred between 25 and 100°C. The weight loss in the first stage was 5.47% (Table 5), which started around 33.91 to 121.69°C to eliminate all moisture. Meanwhile, for region II (200–400°C), which is showed the maxima in the DTG plot (Fig. 6b ) and main thermal degradation stage of raw LLP, are also described in Table 5. Region II also describes the evolution of volatile component. The second region of DTG as shown in Fig. 6b started at 272.91°C, which then reached the highest peak at 332.54°C with weight loss of 47.92%. Region II has overlapping decomposition temperature of hemicellulose, cellulose, and lignin. This is due to hemicellulose degradation (220–315°C), lignin and cellulose decomposition (315–400°C), and lignin degradation (> 450°C) 56,57,58,59,60,61. Comparing the results from previous studies to this study, most of the weight lost in region II occurred at the temperature between 250 and 450°C. The percentage of volatile matter component of raw LLP (47.92%) was lower than those of corn straw (75.02%), corn cob (80.72%), cocoa pod (68.47%), Jatropha cakes (72.53%), moringa cakes (75.08%), Parinari fruit shell (78.17%), and sugar cane bagasse (79.45%) 62.
Table 5
Onset temperature (Ton), degradation temperature on maximum weight-loss rate (TMax), weight loss (WL) of raw LLP obtained from the TG and DTG cur
Adsorbents
|
Region I
|
Region II
|
Region III
|
Reference
|
Ton oC
|
T max oC
|
Moisture loss(%)
|
Ton oC
|
Tend, oC
|
Weight loss (WL)(%)
|
Tmax oC
|
Weight residue(%)
|
Raw LLP
|
33.91
|
12.6
|
5.47
|
272.91
|
372.95
|
47.92
|
332.54
|
48.17
|
This study
|
Sugarcane bagasse
|
-
|
-
|
-
|
310
|
-
|
58.18
|
370
|
26.8
|
(Sukyai et al., 2018)
|
Phaseolus vulgaris L. pod
|
-
|
-
|
-
|
280
|
-
|
67.34
|
380
|
-
|
(Raulino et al., 2018)
|
After 400°C, the decomposition maintained a slow profile where almost all cellulose had been degraded, and the profile can be attributed to the degradation of the remaining lignin. At temperature higher than 380°C, degradation of the remaining lignin occurs and leads to the formation of residue 63,64. Moreover, residual mass of raw LLP after 400°C was 48.166%. Consistent with the findings of others Dominic et al. 65, we found high % residual mass of raw LLP, which corresponds to the protective waxes and layers of lignin in raw LLP. A similar observation was reported by Sukyai et al. 56 as shown in Table 5, where unmodified bagasse pulp showed high % of residual mass of 26.8%.
Point of zero charge (pH pzc ) and Boehm titration.The point of zero charge is defined as the pH at which the biosorbent surface is completely neutral, which is a useful measurement in identifying the best operating conditions (pH levels)66. The value of pHpzc at 4.88 (Table 6) showed predominance of acidic groups on raw LLP surface. This is because raw LLP had acidic surfaces with many oxygenated functional groups, carboxylic, lactonic, and phenolic groups. Based on the biosorbent’s pHpzc value, it was identified that the biosorbent was primarily acidic in nature; thus, a pH greater than the pHpzc must be chosen to facilitate the adsorption of the positive Pb(II) by the raw LLP. These results are in good agreement when compared to other studies 28,67, where the pHpzc values for Colocasia esculenta (L.) and apricot shells were 3.42 and 4.9, respectively.
Table 6
Boehm titration of the raw LLP and it comparison with other studies
Adsorbent
|
Surface functional group, mmol/g
|
|
pHPZC
|
Reference
|
phenolic
|
lactonic
|
carboxylic
|
Total acid
|
LLP
|
0.045875
|
0.04675
|
0.14375
|
0.160614
|
4.88
|
This study
|
Apricot shells
|
1.14
|
0.46
|
0.02
|
1.2439
|
4.9
|
(Šoštarić et al., 2018)
|
The results for the Boehm titration are given in Table 6. Surface active site analysis revealed that raw LLP contained 0.045875 mmolg− 1 of phenolic group, 0.04675 mmolg− 1 of lactone group, 0.14375 mmolg− 1 of carboxylic group, and a total acidity and 0.236375 mmolg− 1. As shown in Table 6, the surface of raw LLP have acidic (mostly carboxylic) groups dominating with higher mmolg− 1. This is in good agreement with the point of zero charge value (4.88) of the biosorbent.
Zeta potential. The zeta potential values for raw LLP are presented in Fig. 7. Raw LLP shows negative zeta potential values within the pH range from 2.0 to 6. The zeta potential for raw LLP decreased from − 1.2 up to − 22.7 mV. As a result, negative charges on the surface of raw LLP could be a good explanation for cationic heavy metal adsorption to them. The carboxylic group on the raw LLP surface is responsible for this significant amount of negative surface charge that dissociates into H+(aq.) and COO− (aq.) in the solution. Inglesby et al. 68 stated that the increase of magnitude of the negative in the acidic pH is due to the dissociation of functional groups. So, the zeta values of raw LLP remained negative over the entire pH range examined. As the zeta potential decreased, the cell surface becomes more negatively charged, which is favorable for adsorption. This is in agreement with Basu et al. 64, which found that zeta potential of cucumber peel decreased up to − 21.0 mV as pH of the suspension increased from pH 2 to 6.
BET analysis and pore size. The nitrogen adsorption–desorption isotherms of raw LLP are shown in Fig. 8. Type III isotherm (IUPAC classification) was observed for raw LLP, indicating that the raw LLP was macroporous.
Table 7 shows BET, total pore volume, external surface area, and average pore width of raw LLP were determined to be 0.6692 m²/g, 0.00269 cm3/g, 1.0063 m²/g, and 175.346 Å, respectively. This indicates that the raw LLP has a low surface area and this is similar to what has been reported in previous studies for some agricultural residues. Pereira et al., 53 and Demirbas69 stated that plant-based biosorbents do not have a high surface area. The small pore volume of raw LLP may be due to the structure of cellulose, hemicellulose, and lignin in raw LLP, which may result in fewer pores, which correlates with the results of Bota et al. 70 and Asuquo and Martin 41. This results in raw LLP having wide average pore width due to the macropores, making them suitable for liquid-phase adsorption because it facilitates adsorbate diffusion into the biosorbent structure 41,71,72. However, it should be noted that the adsorption of heavy metals to raw LLP is not solely dependent on the presence of large surface areas, as porosity is not the only criterion required for good biosorbents. Functional groups such as phenolic, carboxylic, hydroxyl, amines, and ether groups on raw LLP that serve as active sites for adsorption also contribute significantly to their adsorption potential for target heavy metals. These functional groups act as sources of physical and chemical interactions that facilitate adsorption via a number of mechanisms that include hydrogen bonding, electrostatic interactions, and surface complexation 41.
Table 7
Pore structure of raw LLP
Property
|
BET Surface Area (m²/g)
|
Total pore volume cm³/g
|
External Surface area (m²/g)
|
Average pore width (BJH), Å
|
LLP
|
0.6692
|
0.00269
|
1.0063
|
175.346
|
Pb(II) adsorption analysis by raw LLP.
Effect of pH on Pb(II) adsorption to raw LLP. Figure 9 shows the adsorption capacity of Pb(II) in synthetic solution at different pH with 50 mg/L of initial Pb(II) concentration and 40g/L raw LLP dosage. Adsorption of Pb(II) on raw LLP increased at pH 3.0 and 4.0; it remained constant up to pH 5 but decreased at pH 6.0 (Fig. 9). The highest adsorption capacity of Pb(II) was 1.1 mg/g obtained at pH 5 of the synthetic solution. This is because the hydrogen ion (H+) concentration influences the removal efficiency of metal ions in the aqueous solution and affects the solubility of Pb(II) in the solution. The protonation of the raw LLP surface leads to the competition between protons and metal ions for the binding sites. The plots in Fig. 9 confirm that adsorption of Pb(II) is strongly influenced by pH, which is explained based on the point of zero charge (pHpzc). Free Pb(II) ions were the predominant species when the solution pH was below 6.0 73. The removal percentage increased as the pH of the solution approached 5. In the case of Pb(II) adsorption, most of the active sites such as carboxylic acids were occupied by protons on the raw LLP surface at lower pH. According to Roberts and Caserio74, dissociation of acidic functional groups such as carboxylic acids (pKa values of carboxylic acids range from 3.8 to 5.0). When the pH of the biosorbent (raw LLP) rises (pH > pKa), the surface acidic functional groups deprotonate and the negative charge on the biosorbent rises.
Effect of raw LLP dosage on Pb(II) removal. The effect of raw LLP dosage on the removal of Pb(II) is illustrated in Fig. 10 where the dosage of LLP was varied from 8 to 40 g/L at an initial synthetic Pb(II) solution concentration of 50 mg/L at pH 5 and retention time of 30 min. The percentage of Pb(II) removal increased from 85–93.4% with increasing dosage from 8 to 50 g/L. This result could be explained by the fact that for optimum biosorption, extra sites must be available for biosorption reaction, which was achieved by increasing the dosage75. The maximum biosorption percentage reached 93.4% at 50 g/L. However, further increase in biosorbent dosage beyond 40 g/L did not result in sufficient improvement in the percentage removal of Pb(II) ions by raw LLP as reported by 76. Therefore, the optimum dosage for raw LLP was taken as 40 g/L for subsequent batch experiments.
Comparison study. The performance comparison between LLP and other biosorbents is shown in Table 8. It is clearly demonstrated that the performance of raw LLP on the Pb(II) removal achieved 93% higher than Sophora japonica pods 31, Adsononsia digitata 77 and peanut shells 27 with 60%, 68% and 65%, respectively for plant based biosorbent. Raw LLP performed better than Adsononsia digitata and peanut shells in terms of Pb(II) removal time, which was 30 min at 93% removal. Aside from that, both experiments on pH effect revealed that LLP is effective for Pb(II) removal at high concentration of 50 mg/L.
Table 8
Comparison of raw LLP with other plant-based biosorbent
Adsorbent
|
Type of solution
|
Dosage, g/L
|
Time, min
|
Optimum pH
|
Initial concentration, mg/L
|
Pb percent removal, %
|
Refence
|
Raw LLP
|
Synthetic
|
40
|
30
|
5
|
50
|
93
|
This study
|
Sophora japonica pods
|
Synthetic
|
5
|
120
|
6.0–7.0
|
5-100
|
60
|
(Amer et al., 2015)
|
Adsononsia digitata
|
Synthetic
|
7
|
120
|
5.5
|
25
|
68
|
(Chigondo, 2013)
|
Peanut shells
|
Synthetic
|
2
|
180
|
5.5
|
50
|
65
|
(Taşar et al., 2014)
|