The shock resistance characteristic (SRC) of an anaerobic bioreactor characterizes the ability of the anaerobic community in the reactor to withstand violent change in the living environment. In comparison with an upflow anaerobic sludge blanket reactor (UASBR), the SRC of a spiral symmetry stream anaerobic bio-reactor (SSSAB) was systematically investigated in terms of removal efficiency, adsorption property, settling ability, flocculability and fluctuations in these parameters. A quantitative assessment method for SRC was also developed. The results indicated that the SSSAB showed better SRC than the UASBR. The average value (m value) of chemical oxygen demand removal rates of the SSSAB was 86.0%. The contact angle of granules in the SSSAB present gradient distribution, that is the m value of contact angle increasing from bottom (84.5°) to top (93.9°). The m value of the density at the upper and lower sections of the SSSAB were 1.0611 g·cm−3 and 1.0423 g·cm−3, respectively. The surface mean diameter of granules in the SSSAB increased from 1.164 to 1.292 mm during operation. The absolute m value of zeta potential of granular sludge at the upper and lower sections of the SSSAB were 40.4 mV and 44.9 mV, respectively. The weighted mean coefficient variance () value indicated SSSAB was more stable than the UASBR.

INTRODUCTION

Since the upflow anaerobic sludge blanket reactor (UASBR) was invented by Lettinga et al. (Lettinga et al. 1980), high-rate anaerobic technology has undergone a rapid development period. In order to overcome the defects of both easy acidification and granular sludge ecological instability in traditional anaerobic reactors (Lettinga et al. 1997), the authors invented a spiral symmetry stream anaerobic bio-reactor (SSSAB) (Chen et al. 2015). The reactor has three elliptical plates spirally and symmetrically set in the bed, aiming to form the spiral flow to enhance mass transfer between phases on the premise of maintaining a stable population of functional bacteria.

A novel and effective anaerobic reactor needs not only efficient volumetric loading but also long-term stable operation performance, depending on the reactor shock resistance characteristic (SRC), which characterizes the ability of the anaerobic community in the reactor to withstand the violent change in the living environment, such as fluctuations of influent quantity and quality and temperature variation due to alterations in the production process or seasonal changes. These influent index changes are the main cause for sludge washout, acidification, disintegration and ecological unbalance, as well as the root cause of impossibly sustainable and efficient operation. Therefore, it is of practical significance to research the SRC of anaerobic reactors in chronically unstable environmental conditions.

Past studies of SRC in anaerobic reactors often concentrated on a few indices such as the bacteria, chemical oxygen demand (COD) of influent and effluent, and biogas production (Grobicki & Stuckey 1991; Alves et al. 2000; Gao et al. 2011; Zhao et al. 2013), and the present assessment of SRC was confined to only brief descriptions. In our work, in order to study the SRC of anaerobic reactors systematically, we tested not only the influent and effluent COD, but also contact angle, density, particle size and zeta potential of anaerobic granules in an SSSAB (due to their convenient measurement and typical characterization). So SRC was investigated in terms of removal efficiency, adsorption property, settling ability, flocculability and the fluctuations in these parameters. Meanwhile, a UASBR was adopted as the comparison and statistics were employed for developing a quantitative assessment method for SRC.

MATERIALS AND METHODS

Experimental set-up and operation

A schematic diagram of the SSSAB and UASBR is shown in Figure 1. The effective volume of each of the two reactors was 4 L. The other structure and parameters of SSSAB are given in the literature (Chen et al. 2015). The structure parameters of the UASBR were almost the same as for the SSSAB, other than without elliptic plates and gas collection pipes under the plates.
Figure 1

Schematic diagram of experimental set-up: I – SSSAB; II – UASBR; 1 – influent tank; 2 – water pump; 3 – anaerobic granular sludge; 4 – sampling tube; 5 – three phase separator; 6 – water seal; 7 – wet gas-flow meter; 8 – elliptic plates; 9 – biogas collection tube.

Figure 1

Schematic diagram of experimental set-up: I – SSSAB; II – UASBR; 1 – influent tank; 2 – water pump; 3 – anaerobic granular sludge; 4 – sampling tube; 5 – three phase separator; 6 – water seal; 7 – wet gas-flow meter; 8 – elliptic plates; 9 – biogas collection tube.

The SSSAB and UASBR were run without any heat requirement. The SRC experiment took 144 days in a continuous mode lasting from winter to spring. The ambient temperature ranged from 10 to 25 °C, which was lower than the optimum temperature of 35 °C for mesophilic anaerobic fermentation. During the whole experiment, the influent quantity and quality (COD, pH, etc.) and outside temperature were the same for both reactors. The hydraulic retention time (HRT) was set to 24 h, and the HRT was gradually shortened to 6 h in the whole experiments. The COD ranged from 1,500 to 4,500 mg·L−1. The same COD in the running reactor would not last for more than 3 days, changing by over 300 mg·L−1 at every COD altering time. Details about shock factors applied are given in the Supplementary Material (available in the online version of this paper).

Seed sludge and synthetic wastewater

The seed sludge was obtained from a laboratory-scale spiral anaerobic bioreactor (Chen et al. 2012). The surface mean diameter (SMD, the diameter of a sphere that has the same volume/surface area ratio as a particle of interest) of the seed sludge was 1.136 mm, and its density was 1.049 g·cm−3. The zeta potential of the seed sludge was −48.4 mV and contact angle was 88.7°. Moreover, the inoculation amount of the seed sludge was about 2.5 L. The volatile suspended solids concentration of the inoculum was 40.4 g/L, and the volatile suspended solids to suspended solids ratio was 0.72.

The SSSAB was fed with synthetic wastewater containing sucrose (6,000 mg·L−1), NH4Cl (330.0 mg·L−1), CaCl2·2H2O (100.0 mg·L−1), trace element solution according to Tang et al. (2011), and nutrition solution (32 mL·L−1) containing beef extract 0.6 g·L−1, yeast extract, tryptone 1.8 g·L−1, KH2PO4 7.54 g·L−1, MgSO4 0.22 g·L−1. The NaHCO3 was added to adjust pH value (7.0–8.0) and to satisfy alkalinity requirements.

SRC assessment method

The SRC of anaerobic reactors was investigated by its removal efficiency, adsorption property, settling ability, flocculability and fluctuation. Characterization parameters relevant were the removal rate of the reactor, the contact angle, density, zeta potential and particle size distribution of granular sludge (the justification for parameter selection is given in the Supplementary Material, available online). Among these, the first four parameters needed further analyses by statistical methods and the last one could be compared directly. For granular sludge having a series of characteristic Z from Z1 to Zn, the average m and variance s were the significant eigenvalue. But considering that different reactors and spatial locations might contribute to different Z and m, dispersion degree could not be compared by s directly. Thus coefficient of variation CV was introduced for comparing the discrepancy of dispersion degree between different reactors and spatial locations. For samples tested n times for a specific parameter, the value m, s and CV are as follows: 
formula
1
 
formula
2
 
formula
3
where Z was a certain characteristic parameter of anaerobic granular sludge; n and j were capacity and number of different time samples, respectively. Therefore, m was the mean level of a certain characteristic parameter for multiple sludge samples during operation; CV was the fluctuation of such characteristic parameters. And a certain characteristic would be more stable (its fluctuation was smaller) if its CV value was smaller. Notably, only when the experimental error range is small, can the m, s and CV be calculated by the parameter mean value in each measurement time.
Also, weighted mean was further computed to evaluate the fluctuation of reactors comprehensively: 
formula
4
where W was the weight of a certain characteristic parameter of anaerobic granular sludge; k was the number of parameters tested; and i was the number of a specific characteristic parameter. If CV value of one characteristic parameter differed in different spatial locations, the average was calculated instead.
In our SRC experiment, k was determined as 4, so CVi=1–4 represents removal efficiency, contact angle, density and zeta potential, respectively. And the weight value of CVi=1–4 was set to 0.6, 0.1, 0.2 and 0.1, respectively (the justification for parameter weight selection is given in the Supplementary Material, available online). Then Equation (4) can be transformed into Equation (5) as follows: 
formula
5

Analyses

The measurement of COD was performed following Standard Methods (APHA et al. 2007).

The hydrophobicity of sludge was tested by water contact angle measurements on a prepared granular sludge cake. The test was carried out using a modified axisymmetric drop shape analysis–contact diameter (ADSA-CD) technique (Liao et al. 2001; Dong et al. 2004). About 5.0 g of granular sludge samples from the reactors were first washed with distilled water. Then the samples were baked in an oven at temperature of 105 °C for 4 hours. After being cooled down to room temperature in a glass desiccator, the sample was ground by a vibromill (ZHM-1A, China) for 5 min. Then the sample was compressed to a sludge cake packaged by boric acid under pressure of 25 MPa through a tablet machine (ZHY-401, China). A drop of double distilled deionized water was placed on the sludge cake and the contact angle was measured by a contact angle meter (SL200C, USA). The contact angle of every sample was measured three times, and the mean value of them was the final result.

The zeta potential of sludge was measured by a micro-electrophoresis apparatus (Zeta-Meter 3.0+ , The Netherlands). The density of granular sludge was determined according to the literature (Xiaoguang et al. 2013). The morphology of the granular sludge was taken by a camera (Canon IXUS 115 HS). The granular sludge size and size distribution were analyzed by a dynamic image analyzer for granule size measuring range between 0.002 and 0.5 cm (QICPIC, Germany).

RESULTS AND DISCUSSION

Removal efficiency

The COD removal rates and their fluctuations, as well as the applied organic load, after 144 d operation of the SSSAB and UASBR are shown in Figure 2. With the same shock loads, the COD removal rates of the two reactors fluctuated between about 50 and 97% while the applied organic load increased from 2.08 to 11.3 kgCOD·m−3·d−1. The m value of COD removal rates of the SSSAB was 86.0%, which was 1.2 times as high as that of the UASBR (73.2%). And the highest removal rates of the SSSAB could reach 96.5%, higher than the 90.6% of the UASBR. Further, the CV value of removal rates of the SSSAB was 11.2%, which was 0.6 times as low as that of the UASBR (18.4%). Therefore, both removal efficiency and its fluctuation in the SSSAB are better than those in the UASBR.
Figure 2

COD removal efficiency of SSSAB and UASBR.

Figure 2

COD removal efficiency of SSSAB and UASBR.

Adsorption property

The contact angle of the granular sludge was defined by the mechanical equilibrium of the drop under the action of three interfacial tensions (solid–liquid, solid–vapor and liquid–vapor) when a liquid drops on the surface of sludge (Liao et al. 2001). The contact angle could characterize the hydrophobicity of the granular sludge. As the contact angle of the granular sludge gets larger, the hydrophobicity of granular sludge becomes stronger (Sheng et al. 2010). And the strong hydrophobicity of the granular sludge is beneficial for its adsorption to organic contaminant.

As shown in Figure 3(a) and (b), for the SSSAB, its m value of contact angle at the upper section (located at the height of 600 mm, Sample 2 in Figure 1) was 93.9°, and at the lower section (located at the height of 300 mm, Sample 1 in Figure 1) it was 84.5 °. This suggested the granular sludge at the upper section had stronger hydrophobicity than that at the lower section, indicating that the adsorption organic contaminant property of the granular sludge in the SSSAB was gradually increased from bottom to top. Such distribution provided an environment for the complete degradation of organics, as well as improving effluent quality.
Figure 3

Several physical characteristics of anaerobic granular sludge: (a), (b) contact angle; (c), (d) wet density; (e), (f) zeta potential.

Figure 3

Several physical characteristics of anaerobic granular sludge: (a), (b) contact angle; (c), (d) wet density; (e), (f) zeta potential.

For the UASBR, its m value of contact angle at the upper section was 83.6°, which was much lower than that of the SSSAB at the same location. And m value of contact angle at the lower section of the UASBR was 86.2°, close to the value at the upper section. This suggested that there were no significant differences in the adsorption properties of the granular sludge in the UASBR from bottom to top.

Therefore, the granular sludge in the SSSAB has better adsorption property for organic contaminant, compared with that in the UASBR.

Furthermore, the CV value of contact angle at the upper section of the SSSAB was 6.6%, which was 0.46 times that at the lower section (12.3%). This demonstrated that the ability to resist the influent concentration changes for the adsorption property at the upper section of SSSAB was better than that at the lower section, which creates favorable conditions for bacterial distribution in the SSSAB. This may be due to the flow pattern of plug-flow of high-rate anaerobic reactors (Chen et al. 2010). By comparison, the CV value at the upper and lower sections of the UASBR were 8.2% and 8.1%, respectively. This showed that the ability to resist the influent concentration changes for the adsorption property at the upper section of the UASBR was nearly the same as that at the lower section. This might be due to completely mixed flow of the UASBR (Peña et al. 2006).

Settling ability

Effectively maintaining the high concentration of sludge is necessary for an anaerobic reactor to operate efficiently. And one of the key factors to keep the sludge from washing out is whether the anaerobic granules possess ideal settling abilities. In order to quantify the settling ability, the Stokes formula is given as follows: 
formula
6
where ut is the settling velocity; dp is the particle size; ρp is the particle density; ρ is the solution density; g is the gravitational acceleration; μ is the solution viscosity. According to Equation (6), density and particle size are two important parameters for settling velocity of the granular sludge, which is directly proportional to its density and inversely proportional to its particle size.

Density

As illustrated in Figure 3(c) and (d), the m value of the density at the upper and lower section of the SSSAB (1.0611 g·cm−3 and 1.0423 g·cm−3, respectively) were both greater than those of the UASBR (1.0290 g·cm−3 and 1.0228 g·cm−3, respectively). This was significant for keeping the biomass in the SSSAB. Additionally, the m value of the density at the upper section of the SSSAB and UASBR were both greater than that at the lower section, which was due to the hydraulic screening. This phenomenon maintains the granules with larger density at the upper section, reducing the risk of sludge wash-out. Further, the CV values of the density at the upper and lower sections of the SSSAB (7.7% and 9.3%, respectively) were both less than those of the UASBR (18.3% and 11.0%, respectively). This suggested the fluctuation of the density of granules in the SSSAB was more favorable than that in the UASBR, which was beneficial to improve the settling ability of the granules in the SSSAB.

Particle size

The parameter of SMD was adopted to characterize the mean particle size of the anaerobic granular sludge. The initial SMD of seed sludge was 1.136 mm, and the granular size of 1.076 to 2.341 mm dominated the total biomass of the seed sludge analyzed by QICPIC. After 34 days operation, as shown in Figure 4(a) and (d), the SMD of granules in the two reactors both increased to about 1.2 mm, and the granular size of 1.307 to 2.844 mm dominated the total biomass in both reactors. It suggested that some small particles were washed out, attributable to the hydraulic screening; hence the granules with large diameter accounts for the higher ratio. And due to the same hydraulic loading of the two reactors in initial operation stage, the hydraulic screening effect in the two reactors is nearly the same as well. Thus, the particle size distributions of the SSSAB and UASBR were similar after 34 days operation. Along with the further synchronous change of volumetric loading of the two reactors, to 94 day, the SMD of granules in the SSSAB and the UASBR decreased from 1.272 and 1.228 mm to 1.164 mm and 1.111 mm, respectively. As illustrated in curve q3 in Figure 4(a), (b), (d) and (e), the particle size distribution in the SSSAB changed apparently. For the SSSAB the granular size of 1.307 to 3.454 mm dominated the total biomass. And for the UASBR the granular size of 1.307 to 2.341 mm dominated the total biomass. That was probably due to the higher COD mean removal rate and consequently more gas production for the SSSAB than for the UASBR. Therefore the pneumatic screening effect was enhanced. Then the ‘regroup’ of the particle size distribution in the SSSAB occurred under the influence of both hydraulic and pneumatic screening.
Figure 4

Size distribution of the anaerobic granular sludge: (a)–(c) SSSAB; (d)–(f) UASBR.

Figure 4

Size distribution of the anaerobic granular sludge: (a)–(c) SSSAB; (d)–(f) UASBR.

On day 144 (see Figure 4(c) and (f)), the SMD of granules in the SSSAB increased to 1.292 mm rather than decreased. In contrast, the SMD of granules in the UASBR continued to decrease to 0.941 mm. The higher diameter of granules in the SSSAB leads to larger settling velocity of granules (according to Equation (6)), which may be beneficial for maintaining the sludge. And the color and compactness of granules in the SSSAB did not change judging by appearance. The granules in the UASBR became yellow, the floc sludge emerged obviously, and some granules started to disintegrate. This may be connected with the flocculability of sludge.

Hence, the granular sludge in the SSSAB has better settling ability, compared with that in the UASBR.

Flocculability

The flocculability of the anaerobic granular sludge is an important guarantee of preventing granules from disintegrating. The zeta potential, or the electric potential in the interfacial double layer at the slipping plane in the bulk fluid away from the interface (Coday et al. 2015), could be utilized to evaluate the flocculability of the granular sludge (Yu et al. 2009). The granular sludge with low zeta potentials tends to coagulate or flocculate. So the larger the absolute value of zeta potential, the worse the flocculability of the granular sludge. As given in Figure 3(e) and (f), the absolute m value of zeta potential of granular sludge at the upper section of the SSSAB was 40.4 mV, lower than that at the lower section of 44.9 mV. This indicated that for the SSSAB the granular sludge at the upper section has stronger flocculability than that at the lower section. Figure 3(e) and (f) also show that the absolute m value of zeta potential of granular sludge at the upper section of the UASBR (44.5 mV) was higher than that at the lower section (41.8 mV). This suggested that for the UASBR the granular sludge at the upper section has weaker flocculability than that at the lower section, leading to disintegration happening at the upper section in the UASBR. And this resulted in the decrease of particle size in the UASBR, as well as the emergence of sludge flocs (Figure 4(f)).

In addition, the CV value of the zeta potential at the upper and lower sections in the SSSAB (28.5% and 31.2%, respectively) was less than those in the UASBR (33.4% and 40.0%, respectively). This illustrated that the fluctuation of flocculability of granules in the SSSAB was smaller, contributing to prevent a part of the sludge from disintegrating and then washing out due to its poor flocculability.

Therefore, both granule flocculability and its fluctuation in the SSSAB are better than those in the UASBR.

Comprehensive assessment of fluctuation

According to Equation (5), the value of the SSSAB (12.4%) was calculated. Such value was 0.68 times that of the UASBR (18.5%), indicating the fluctuation of the SSSAB was smaller than that of the UASBR. Therefore, the SSSAB was more stable than the UASBR.

CONCLUSIONS

The SSSAB showed better SRC than the UASBR. The removal efficiency of the SSSAB and adsorption property, settling ability, floculability and fluctuation of granular sludge in the SSSAB were all more favorable than that in the UASBR. The contact angle, density and zeta potential of granules in the SSSAB present gradient distribution, while those in the UASBR present unsatisfactory distribution. During operation, the SMD of granules in the SSSAB increased from 1.164 to 1.292 mm, both values larger than those in the UASBR. The value of the SSSAB was 12.4%, which was 0.67 times that of the UASBR.

ACKNOWLEDGEMENTS

The authors would like to thank the National Natural Science Foundation of China (Grant no. 51208087), Doctoral Fund of Ministry of Education of China (New Teachers) (Project no. 20120075120001), and the Fundamental Research Funds for the Central Universities.

REFERENCES

REFERENCES
Alves
M.
Cavaleiro
A.
Ferreira
E.
Amaral
A.
Mota
M.
Motta
M.
Vivier
H.
Pons
M.
2000
Characterisation by image analysis of anaerobic sludge under shock conditions
.
Water Science & Technology
41
(
12
),
207
214
.
APHA, AWWA & WEF
2005
Standard Methods for the Examination of Water and Wastewater
.
American Public Health Association, American Water Works Association, and Water Environment Federation
,
Washington, DC
.
Chen
X.
Zheng
P.
Guo
Y.
Qaisar
M.
Tang
C.
Ding
S.
2010
Flow patterns of super-high-rate anaerobic bioreactor
.
Bioresource Technology
101
(
20
),
7731
7735
.
Chen
X.
Dai
R.
Xiang
X.
Ma
C.
Li
G.
Hu
T.
Xu
Z.
2015
Rheological behaviors of anaerobic granular sludge in a spiral symmetry stream anaerobic bioreactor
.
Water Science & Technology
72
(
4
),
658
664
.
Coday
B. D.
Luxbacher
T.
Childress
A. E.
Almaraz
N.
Xu
P.
Cath
T. Y.
2015
Indirect determination of zeta potential at high ionic strength: specific application to semipermeable polymeric membranes
.
Journal of Membrane Science
478
,
58
64
.
Dong
P.
Shan
Z.
Li
Z.
2004
Study on the surface wet characteristics of ultrafine coal powder
.
Journal of China Coal Society
29
(
3
),
346
349
(in Chinese with English abstract)
.
Grobicki
A.
Stuckey
D. C.
1991
Performance of the anaerobic baffled reactor under steady-state and shock loading conditions
.
Biotechnology and Bioengineering
37
(
4
),
344
355
.
Lettinga
G.
Van Velsen
A. F. M.
Hobma
S. W.
De Zeeuw
W.
Klapwijk
A.
1980
Use of the upflow sludge blanket (USB) reactor concept for biological wastewater treatment, especially for anaerobic treatment
.
Biotechnology and Bioengineering
22
(
1980
),
699
734
.
Lettinga
G.
Field
J.
Van Lier
J.
Zeeman
G.
Huishoff Pol
L. W.
1997
Advanced anaerobic wastewater treatment in the near future
.
Water Science and Technology
35
(
10
),
5
12
.
Liao
B. Q.
Allen
D. G.
Droppo
I. G.
Leppard
G. G.
Liss
S. N.
2001
Surface properties of sludge and their role in bioflocculation and settleability
.
Water Research
35
(
2
),
339
350
.
Tang
C.-J.
Zheng
P.
Wang
C. H.
Mahmood
Q.
Zhang
J. Q.
Chen
X. G.
Zhang
L.
Chen
J. W.
2011
Performance of high-loaded ANAMMOX UASB reactors containing granular sludge
.
Water Research
45
(
2011
),
135
144
.
Xiaoguang
C.
Ruobin
D.
Xinyi
X.
Mengyan
T.
2013
Optimum fluidization velocity of granular sludge bed for anaerobic fluidized-bed bioreactors
.
Journal of Chemical Technology and Biotechnology
88
(
12
),
2272
2278
.
Zhao
H.
Li
J.
Li
J.
Yuan
X.
Piao
R.
Zhu
W.
Li
H.
Wang
X.
, Cui, Z.
2013
Organic loading rate shock impact on operation and microbial communities in different anaerobic fixed-bed reactors
.
Bioresource Technology
140
,
211
219
.

Supplementary data