Appl Microbiol Biotechnol (2015) 99:2829–2838 DOI 10.1007/s00253-014-6149-x
ENVIRONMENTAL BIOTECHNOLOGY
Nickel biosorption by discharged biomass from wastewater treatment bioreactor: size plays a key role Dandan Zhou & Yang Yang & Yunbao Li & Zhengxue Xu & Shanshan Dong
Received: 4 September 2014 / Revised: 25 September 2014 / Accepted: 9 October 2014 / Published online: 26 October 2014 # Springer-Verlag Berlin Heidelberg 2014
Abstract Biomass size significantly affects the characteristics of the extracellular polymeric substances, which in turn influences the biosorption mechanisms. In this work, nickel biosorption mechanisms and the capacity of excess aerobic granules (AGs) of >850 μm, 500–850 μm, 212–500 μm, and bioflocs 850 μm due to their different biosorption mechanisms. Chemical binding between nickel and protein was dominant for biomass less than 850 μm, whereas nickel adsorption was mainly physical when biogranule size was greater than 850 μm, because the nickel had no opportunity to interact with protein located in the core of the biogranule. Only low levels of ion exchange (850 μm, 500–850 μm, 212–500 μm, and bioflocs 850 μm, 500–850 μm, 212– 500 μm, and bioflocs, respectively (the initial Ni2+ concentration was 100 mg/L and the initial AG concentration was 1.0 g/ L in dry weight). This suggested that the initial formed AG of 212–500 μm had a higher biosorption capacity than that of bioflocs, whose diameter was usually less than 212 μm. It is generally accepted that microorganisms secrete more EPS during aerobic granulation (Zhou et al. 2013b; Miksch and Konczak 2012); thus, the functional groups in the EPS of granules enhanced their biosorption capacity. The biosorption capacity for Ni2+ decreased from 1.28 to 0.93 mEq/g with the increment of AG size from 500–850 to >850 μm, which may be a result of a decrease in the specific surface area (Wu et al. 2010). To summarize, AGs have better settleability and stability compared with that of sludge flocs; however, the biosorption capacity was significantly influenced by their size, particularly granules larger than 850 μm. The following work therefore investigated how the size influenced the biosorption.
X-ray photoelectron spectroscopy (XPS) The XPS measurement was performed in the ultrahigh vacuum chamber of a Thermo XPS system (ESCALAB 250, Thermo, USA). The XPS spectra were recorded using the energy source of monochromatic Al K radiation (1486.71 eV) operated at 15 kV and 15 mA. All binding energies were referenced to the neutral C1s peak at 284.6 eV to compensate for surface charging effects. The XPS peak (version 4.1) was used to fit the XPS spectra peaks.
X-ray diffraction (XRD) XRD analyses were performed with a diffractometer (D8 Advance, Bruker, Germany), with a cobalt tube scattering from 20 to 70° in 2θ. The pretreatment method of the granule sample was conducted according to Angela et al.(2011). The granule samples analyzed by XRD had been previously dried and calcined in an oven at 550 °C for 2 h, in order to remove the organic fraction.
Energy dispersive X-ray (EDX) The interior nickel distribution of the biomass was observed by cryosectioning coupled with EDX. The AGs were fixed by 2.5 % glutaraldehyde in paraformaldehyde for 3 h, reducing the damage of sample morphology by cryosectioning. Then, granules were embedded in Tissue-Tek O.C.T. (Sakura Finetek, CA), before which they were frozen at −20 °C for cryosectioning using a cryomicrotome (Leica, CM1900, German). The pretreatment process of bioflocs was similar with that of AGs, except cryosectioning was not included. The relatively well preserved of at least ten samples were chosen for SEM operation with a scanning electron microscope (Akashi-SX-40, USA), and then the average value of nickel weight ratio was calculated according to the SEM-EDX results.
Binding function in Ni2+ biosorption: protein dominant FT-IR and XPS were integrated to elucidate the interaction mechanism between nickel ions and the biosorbents. FT-IR spectra The FT-IR spectra of the aerobic granules and bioflocs, before and after Ni2+ biosorption, were performed to identify the functional groups involved in the biosorption (Fig. 1). The FT-IR spectra of initial granules displayed adsorption peaks in the range of 500–4000 cm−1. The strong bands in the region of 3100– 3500 cm−1 reflected N–H and O–H stretching vibrations, showing the presence of hydroxyl and amine groups (Yao et al. 2009). The bands at 2962 and 2964 cm−1 are because of an asymmetric vibration of CH2, while the bands at 2852 cm−1 resulted from a symmetric vibration of CH2. The bands at 1738/1732 and 1238/1234 cm−1 could be attributed to the stretching vibration and deformation vibration of C=O of carboxylic acids, respectively. The bands at 1651 cm−1 were the result of the stretching vibration of C=O and C–N (amide I) peptidic bond of protein, while 1539 cm−1 bands show stretching vibration of C–N and deformation vibration of N–H (amide II) peptidic bond of protein (Xu and Liu 2008). The peaks at 1038 cm−1 could be attributed to stretching vibration of OH of polysaccharides. The FT-IR spectra of the initial granules indicated that the functional groups in the biomass of different sizes were the same and generally originate from the EPS of the biomass (see Fig. 1, left panel).
2832
Appl Microbiol Biotechnol (2015) 99:2829–2838
before biosorption
after biosorption
a
a 2852 2962 3429
1234 1732 1539 1038 1651 1398
3431
3176
1238
1732
2852
1539
2964 3196
1653
Transmittance %
b
1385 1038
1232
1738 1539
2852
1734 1541
2852
1038
1651 1402
2964 3431
b
1238
1655
2964
c
3176
3433
1038 1387
3138
c 2852
1738
1238 1539 1038 1651 1402
2964 3429
3429
3176
1234
1738
2852 2962 3176
1038 1539 1387 1651
d
d 2852 3431 3500
1738
2962 3176 3000
1234
1539 1651 1402 1038 2500
2000
1236
1500
1000
3446 500
1734
1038 1541 1387 1653
2852 2962 3151
3500
3000
Wavenumber cm-1
2500
2000
1500
1000
500
Wavenumber cm-1
Fig. 1 FT-IR spectrum of the biomass before (left panel) and after (right panel) Ni2+ biosorption, aerobic granules with sizes of a >850 μm, b 500– 850 μm, c 212–500 μm, and d bioflocs 850 μm, 500–850 μm, 212–500 μm, and bioflocs, respectively. These results showed the participation of C–(O, N) in the biosorption reaction. Correspondingly, the quantity of other functional group, i.e., C–(C, H) and O–C–O+C=O, increased after the biosorption. These findings further revealed the significant role of protein in nickel binding as the group with a C single bonded to N, which attributed to amine or amide (around 285.7 eV), was the main group that responded to nickel binding. As indicated above, the results of FT-IR spectrum and XPS spectrum of N1s and C1s all confirmed protein interaction with nickel ions and that protein had a more significant role in the biosorption process than other EPS components.
and other elements, which mainly contained O, S, and P, were detected. Very small amounts of Ni at the level of 0.05– 0.08 mg/g were found in fresh AG/sludge probably due to the fact that the municipal wastewater used contained trace Ni elements. The Ca content in the fresh AG/sludge was around 10 mg/g, much lower than that reported by Xu and Liu (2008). The possible reason was that a lack of control of the selective pressure leads to microbial cell-to-cell granulation, self-immobilization, and Ca2+ playing a minor role in granulation in the continuous-flow bioreactor (Zhou et al. 2014). Table 2 presents the metal mass balance in the Ni-biosorbed biomass deduced from Table 1. The total detected Ni2+ biosorption concentration was obtained by the difference between fresh and Ni-biosorbed biomass. The predicted Ni concentration by the ion exchange between Ni and Ca, Mg, as well as K was calculated according to the concentration variation of Ca, >Mg, and K ions. The results indicated that ion exchange would be one of the mechanisms involved in Ni biosorption by granules/sludge. However, the quantity of Ni biosorption by ion exchange only accounted for a maximum of 15.22 % of total detected Ni concentration before and after Ni biosorption. Therefore, ion exchange only had a minor role in biosorption of biomass of all sizes investigated.
Ion exchange and precipitation function: minor
XRD spectrum
Elemental compositions analysis
XRD analysis was carried out to distinguish the crystalline minerals in the granules. The XRD peaks in the spectra of all the sizes biomass showed that CaCO3 was the main crystalline mineral (Fig. 4). CaCO3 is the chemical form of calcium found
Table 1 shows the elemental compositions of fresh and Nibiosorbed biomass of different sizes. C, H, N, Mg, K, Ca, Ni,
2834
Appl Microbiol Biotechnol (2015) 99:2829–2838 before biosorption C-(C,H), 50.82%
Relative intensity
C-(C,H), 52.71% C-(O,N), 29.01%
C-(O,N), 33.65%
O-C-O, 18.27% C=O
O-C-O, 15.53% C=O
a
C-(C,H), 57.52%
C-(C,H), 58.63%
C-(O,N), 29.76%
C-(O,N), 28.11%
b
O-C-O, 13.27% C=O
c
O-C-O, 12.72% C=O
d
after biosorption C-(C,H), 61.08%
C-(C,H), 55.39%
C-(C,H), 64.68%
C-(C,H), 56.99%
C-(O,N), 11.71%
C-(O,N), 20.31% C-(O,N), 21.52% O-C-O, 21.49% C=O
292
290
288
a
286
284
282
280
290
288
C-(O,N), 16.32%
O-C-O, 27.21% C=O
O-C-O, 24.30% C=O
b
286
284
282
c
280
290
288
286
284
282
O-C-O, 18.99% C=O
280
290
288
d
286
284
282
280
Binding energy/ eV Fig. 3 XPS spectrum of C1s of the biomass before and after Ni2+ biosorption, aerobic granules with sizes of a >850 μm, b 500–850 μm, c 212–500 μm, and d bioflocs 850 μm, indicating that the larger biomass had the poorer biosorption capacity. Protein binding was the dominant function of nickel biosorption for biomass no greater than 850 μm, and ion exchange and precipitation function played a much minor role for all biomass. Thus, nickel adsorption mechanism for granules greater than 850 μm was still called into question. To obtain further insight into the effect of granule size on the mechanism of nickel biosorption, a narrow XPS scan
Table 1 Elemental compositions of biomass with different sizes (mg/g dry weight of granules) Element mg·g−1
C H N Mg K Ca Ni Others
>850 μm
500–800 μm
200–500 μm
850
Fresh Ni polluted Difference Fresh Ni polluted Difference Fresh Ni polluted Difference Fresh Ni polluted Difference
500–850
212–500
850 μm before and after biosorption (a and b), 500–850 μm before and after biosorption (c and d), 212–500 μm before and after biosorption (e and f), and bioflocs before and after biosorption (g and h)
biosorption in Ni 2p binding energy region were quite similar, which confirmed there was almost no chemical binding/ interaction related to nickel accumulation in bio-granules greater than 850 μm. As ion exchange and precipitation were also negligible for nickel biosorption, the accumulation of nickel in the larger granules could be realized by simple physical adsorption. Thus, the size of biomass influenced nickel adsorption mechanism significantly, i.e., protein binding dominated for biomass no greater than 850 μm, while simple physical adsorption dominated for granules greater than 850 μm. However, biosorption based on simple physical adsorption possessed less biosorption capacity as compared with that of binding function. EPS were found ideally to serve as a natural ligand source and to provide binding sites for other charged particles/ molecules including metals (Bhaskar and Bhosle 2006; Chen et al. 2007). EPS in the aerobic granules were confirmed to be a mixture of macromolecular polyelectrolytes, in which protein and polysaccharides were the main components. EPS could present different molecular mass and distribution in different sizes of granules. McSwain et al. (2005) and Chen et al. (2007) found protein mainly accumulated in the core of the granules greater than 800 μm, and the granule formation is dependent on this core. But, in our previous work (Zhou et al. 2013a), protein distributed throughout the interior of the granule uniformly was no more than 800 μm. As both of XPS and FT-IR analysis revealed that protein dominated the chemical binding/interaction of nickel biosorption, the different adsorption mechanisms of biomass with various sizes were considered to be a result of the different protein distributions within the biomass. Thus, we concluded that nickel had no opportunity to interact with the protein at the core of granules with a
b
Ni 2p1/2
after biosorption 855.3
872.1
860.2
a
868.0
855.1
861.3
868.5
872.3
a
before biosorption
879.6
Appl Microbiol Biotechnol (2015) 99:2829–2838 878.0
2836
Ni 2p3/2
Relative intensity
b
c
Ni 2p1/2
Ni 2p3/2
c
d
d
880
875
870
865
860
855
850
Binding energy/ eV
880
Ni 2p1/2 Ni 2p3/2
875
870
865
860
855
850
Binding energy/ eV
Fig. 5 XPS spectrum of Ni 2p of the biomass before and after Ni2+ biosorption, aerobic granules with sizes of a >850 μm, b 500–850 μm, c 212– 500 μm, and d bioflocs 850
500-850 212-500 Biomass size /μm
ENVIRONMENTAL BIOTECHNOLOGY
Nickel biosorption by discharged biomass from wastewater treatment bioreactor: size plays a key role Dandan Zhou & Yang Yang & Yunbao Li & Zhengxue Xu & Shanshan Dong
Received: 4 September 2014 / Revised: 25 September 2014 / Accepted: 9 October 2014 / Published online: 26 October 2014 # Springer-Verlag Berlin Heidelberg 2014
Abstract Biomass size significantly affects the characteristics of the extracellular polymeric substances, which in turn influences the biosorption mechanisms. In this work, nickel biosorption mechanisms and the capacity of excess aerobic granules (AGs) of >850 μm, 500–850 μm, 212–500 μm, and bioflocs 850 μm due to their different biosorption mechanisms. Chemical binding between nickel and protein was dominant for biomass less than 850 μm, whereas nickel adsorption was mainly physical when biogranule size was greater than 850 μm, because the nickel had no opportunity to interact with protein located in the core of the biogranule. Only low levels of ion exchange (850 μm, 500–850 μm, 212–500 μm, and bioflocs 850 μm, 500–850 μm, 212– 500 μm, and bioflocs, respectively (the initial Ni2+ concentration was 100 mg/L and the initial AG concentration was 1.0 g/ L in dry weight). This suggested that the initial formed AG of 212–500 μm had a higher biosorption capacity than that of bioflocs, whose diameter was usually less than 212 μm. It is generally accepted that microorganisms secrete more EPS during aerobic granulation (Zhou et al. 2013b; Miksch and Konczak 2012); thus, the functional groups in the EPS of granules enhanced their biosorption capacity. The biosorption capacity for Ni2+ decreased from 1.28 to 0.93 mEq/g with the increment of AG size from 500–850 to >850 μm, which may be a result of a decrease in the specific surface area (Wu et al. 2010). To summarize, AGs have better settleability and stability compared with that of sludge flocs; however, the biosorption capacity was significantly influenced by their size, particularly granules larger than 850 μm. The following work therefore investigated how the size influenced the biosorption.
X-ray photoelectron spectroscopy (XPS) The XPS measurement was performed in the ultrahigh vacuum chamber of a Thermo XPS system (ESCALAB 250, Thermo, USA). The XPS spectra were recorded using the energy source of monochromatic Al K radiation (1486.71 eV) operated at 15 kV and 15 mA. All binding energies were referenced to the neutral C1s peak at 284.6 eV to compensate for surface charging effects. The XPS peak (version 4.1) was used to fit the XPS spectra peaks.
X-ray diffraction (XRD) XRD analyses were performed with a diffractometer (D8 Advance, Bruker, Germany), with a cobalt tube scattering from 20 to 70° in 2θ. The pretreatment method of the granule sample was conducted according to Angela et al.(2011). The granule samples analyzed by XRD had been previously dried and calcined in an oven at 550 °C for 2 h, in order to remove the organic fraction.
Energy dispersive X-ray (EDX) The interior nickel distribution of the biomass was observed by cryosectioning coupled with EDX. The AGs were fixed by 2.5 % glutaraldehyde in paraformaldehyde for 3 h, reducing the damage of sample morphology by cryosectioning. Then, granules were embedded in Tissue-Tek O.C.T. (Sakura Finetek, CA), before which they were frozen at −20 °C for cryosectioning using a cryomicrotome (Leica, CM1900, German). The pretreatment process of bioflocs was similar with that of AGs, except cryosectioning was not included. The relatively well preserved of at least ten samples were chosen for SEM operation with a scanning electron microscope (Akashi-SX-40, USA), and then the average value of nickel weight ratio was calculated according to the SEM-EDX results.
Binding function in Ni2+ biosorption: protein dominant FT-IR and XPS were integrated to elucidate the interaction mechanism between nickel ions and the biosorbents. FT-IR spectra The FT-IR spectra of the aerobic granules and bioflocs, before and after Ni2+ biosorption, were performed to identify the functional groups involved in the biosorption (Fig. 1). The FT-IR spectra of initial granules displayed adsorption peaks in the range of 500–4000 cm−1. The strong bands in the region of 3100– 3500 cm−1 reflected N–H and O–H stretching vibrations, showing the presence of hydroxyl and amine groups (Yao et al. 2009). The bands at 2962 and 2964 cm−1 are because of an asymmetric vibration of CH2, while the bands at 2852 cm−1 resulted from a symmetric vibration of CH2. The bands at 1738/1732 and 1238/1234 cm−1 could be attributed to the stretching vibration and deformation vibration of C=O of carboxylic acids, respectively. The bands at 1651 cm−1 were the result of the stretching vibration of C=O and C–N (amide I) peptidic bond of protein, while 1539 cm−1 bands show stretching vibration of C–N and deformation vibration of N–H (amide II) peptidic bond of protein (Xu and Liu 2008). The peaks at 1038 cm−1 could be attributed to stretching vibration of OH of polysaccharides. The FT-IR spectra of the initial granules indicated that the functional groups in the biomass of different sizes were the same and generally originate from the EPS of the biomass (see Fig. 1, left panel).
2832
Appl Microbiol Biotechnol (2015) 99:2829–2838
before biosorption
after biosorption
a
a 2852 2962 3429
1234 1732 1539 1038 1651 1398
3431
3176
1238
1732
2852
1539
2964 3196
1653
Transmittance %
b
1385 1038
1232
1738 1539
2852
1734 1541
2852
1038
1651 1402
2964 3431
b
1238
1655
2964
c
3176
3433
1038 1387
3138
c 2852
1738
1238 1539 1038 1651 1402
2964 3429
3429
3176
1234
1738
2852 2962 3176
1038 1539 1387 1651
d
d 2852 3431 3500
1738
2962 3176 3000
1234
1539 1651 1402 1038 2500
2000
1236
1500
1000
3446 500
1734
1038 1541 1387 1653
2852 2962 3151
3500
3000
Wavenumber cm-1
2500
2000
1500
1000
500
Wavenumber cm-1
Fig. 1 FT-IR spectrum of the biomass before (left panel) and after (right panel) Ni2+ biosorption, aerobic granules with sizes of a >850 μm, b 500– 850 μm, c 212–500 μm, and d bioflocs 850 μm, 500–850 μm, 212–500 μm, and bioflocs, respectively. These results showed the participation of C–(O, N) in the biosorption reaction. Correspondingly, the quantity of other functional group, i.e., C–(C, H) and O–C–O+C=O, increased after the biosorption. These findings further revealed the significant role of protein in nickel binding as the group with a C single bonded to N, which attributed to amine or amide (around 285.7 eV), was the main group that responded to nickel binding. As indicated above, the results of FT-IR spectrum and XPS spectrum of N1s and C1s all confirmed protein interaction with nickel ions and that protein had a more significant role in the biosorption process than other EPS components.
and other elements, which mainly contained O, S, and P, were detected. Very small amounts of Ni at the level of 0.05– 0.08 mg/g were found in fresh AG/sludge probably due to the fact that the municipal wastewater used contained trace Ni elements. The Ca content in the fresh AG/sludge was around 10 mg/g, much lower than that reported by Xu and Liu (2008). The possible reason was that a lack of control of the selective pressure leads to microbial cell-to-cell granulation, self-immobilization, and Ca2+ playing a minor role in granulation in the continuous-flow bioreactor (Zhou et al. 2014). Table 2 presents the metal mass balance in the Ni-biosorbed biomass deduced from Table 1. The total detected Ni2+ biosorption concentration was obtained by the difference between fresh and Ni-biosorbed biomass. The predicted Ni concentration by the ion exchange between Ni and Ca, Mg, as well as K was calculated according to the concentration variation of Ca, >Mg, and K ions. The results indicated that ion exchange would be one of the mechanisms involved in Ni biosorption by granules/sludge. However, the quantity of Ni biosorption by ion exchange only accounted for a maximum of 15.22 % of total detected Ni concentration before and after Ni biosorption. Therefore, ion exchange only had a minor role in biosorption of biomass of all sizes investigated.
Ion exchange and precipitation function: minor
XRD spectrum
Elemental compositions analysis
XRD analysis was carried out to distinguish the crystalline minerals in the granules. The XRD peaks in the spectra of all the sizes biomass showed that CaCO3 was the main crystalline mineral (Fig. 4). CaCO3 is the chemical form of calcium found
Table 1 shows the elemental compositions of fresh and Nibiosorbed biomass of different sizes. C, H, N, Mg, K, Ca, Ni,
2834
Appl Microbiol Biotechnol (2015) 99:2829–2838 before biosorption C-(C,H), 50.82%
Relative intensity
C-(C,H), 52.71% C-(O,N), 29.01%
C-(O,N), 33.65%
O-C-O, 18.27% C=O
O-C-O, 15.53% C=O
a
C-(C,H), 57.52%
C-(C,H), 58.63%
C-(O,N), 29.76%
C-(O,N), 28.11%
b
O-C-O, 13.27% C=O
c
O-C-O, 12.72% C=O
d
after biosorption C-(C,H), 61.08%
C-(C,H), 55.39%
C-(C,H), 64.68%
C-(C,H), 56.99%
C-(O,N), 11.71%
C-(O,N), 20.31% C-(O,N), 21.52% O-C-O, 21.49% C=O
292
290
288
a
286
284
282
280
290
288
C-(O,N), 16.32%
O-C-O, 27.21% C=O
O-C-O, 24.30% C=O
b
286
284
282
c
280
290
288
286
284
282
O-C-O, 18.99% C=O
280
290
288
d
286
284
282
280
Binding energy/ eV Fig. 3 XPS spectrum of C1s of the biomass before and after Ni2+ biosorption, aerobic granules with sizes of a >850 μm, b 500–850 μm, c 212–500 μm, and d bioflocs 850 μm, indicating that the larger biomass had the poorer biosorption capacity. Protein binding was the dominant function of nickel biosorption for biomass no greater than 850 μm, and ion exchange and precipitation function played a much minor role for all biomass. Thus, nickel adsorption mechanism for granules greater than 850 μm was still called into question. To obtain further insight into the effect of granule size on the mechanism of nickel biosorption, a narrow XPS scan
Table 1 Elemental compositions of biomass with different sizes (mg/g dry weight of granules) Element mg·g−1
C H N Mg K Ca Ni Others
>850 μm
500–800 μm
200–500 μm
850
Fresh Ni polluted Difference Fresh Ni polluted Difference Fresh Ni polluted Difference Fresh Ni polluted Difference
500–850
212–500
850 μm before and after biosorption (a and b), 500–850 μm before and after biosorption (c and d), 212–500 μm before and after biosorption (e and f), and bioflocs before and after biosorption (g and h)
biosorption in Ni 2p binding energy region were quite similar, which confirmed there was almost no chemical binding/ interaction related to nickel accumulation in bio-granules greater than 850 μm. As ion exchange and precipitation were also negligible for nickel biosorption, the accumulation of nickel in the larger granules could be realized by simple physical adsorption. Thus, the size of biomass influenced nickel adsorption mechanism significantly, i.e., protein binding dominated for biomass no greater than 850 μm, while simple physical adsorption dominated for granules greater than 850 μm. However, biosorption based on simple physical adsorption possessed less biosorption capacity as compared with that of binding function. EPS were found ideally to serve as a natural ligand source and to provide binding sites for other charged particles/ molecules including metals (Bhaskar and Bhosle 2006; Chen et al. 2007). EPS in the aerobic granules were confirmed to be a mixture of macromolecular polyelectrolytes, in which protein and polysaccharides were the main components. EPS could present different molecular mass and distribution in different sizes of granules. McSwain et al. (2005) and Chen et al. (2007) found protein mainly accumulated in the core of the granules greater than 800 μm, and the granule formation is dependent on this core. But, in our previous work (Zhou et al. 2013a), protein distributed throughout the interior of the granule uniformly was no more than 800 μm. As both of XPS and FT-IR analysis revealed that protein dominated the chemical binding/interaction of nickel biosorption, the different adsorption mechanisms of biomass with various sizes were considered to be a result of the different protein distributions within the biomass. Thus, we concluded that nickel had no opportunity to interact with the protein at the core of granules with a
b
Ni 2p1/2
after biosorption 855.3
872.1
860.2
a
868.0
855.1
861.3
868.5
872.3
a
before biosorption
879.6
Appl Microbiol Biotechnol (2015) 99:2829–2838 878.0
2836
Ni 2p3/2
Relative intensity
b
c
Ni 2p1/2
Ni 2p3/2
c
d
d
880
875
870
865
860
855
850
Binding energy/ eV
880
Ni 2p1/2 Ni 2p3/2
875
870
865
860
855
850
Binding energy/ eV
Fig. 5 XPS spectrum of Ni 2p of the biomass before and after Ni2+ biosorption, aerobic granules with sizes of a >850 μm, b 500–850 μm, c 212– 500 μm, and d bioflocs 850
500-850 212-500 Biomass size /μm
