A ceramic membrane bio-reactor (CMBR) process was first used to treat wastewater collected from a campus of Guilin University of Technology (GUT). A CMBR with a submerged flat-sheet ceramic membrane module was designed for a Qmax of 300 m3/d. With a stable flux at about 33 L/m2 h, the transmembrane pressure was maintained at −10 kPa until the end of the experiment. The results showed that the CMBR process is a robust system capable of producing high-quality service water from campus wastewater. By using the CMBR process, the treated water met the Chinese national standards for landscaping irrigation.

In China, the number of universities increased to 2, 762 by 2011, almost 2.7 times the number in 1998, to accommodate the increasing numbers of students (China Education Annuals 2011). For example, the number of campuses for Guilin University of Technology (GUT) went from two to four, increasing the campus surface area from 5.3 × 105 m2 to 2.7 × 107 m2. Green areas occupy over 50% of the surface area. The annual irrigation water needed for the green areas is estimated to be 7.3 × 104 m3. Tap water is often used for this irrigation. However, as there is a lack of potable water in most Chinese cities, this use is wasteful. Every day, 8 × 106 m3/d of wastewater is generated from the university campuses. Therefore, it is becoming increasingly important to recycle this wastewater.

Recently, attention has been increasing towards the use of membrane bioreactor (MBR) technology for wastewater recycling due to its higher efficiency in pollutant removal, compact size, and lower energy consumption (Melin et al. 2006; Chon et al. 2015; Judd 2015; Khalid et al. 2015; Kim et al. 2015). The material of the ultra-filtrating membrane, the most important part of the MBR technology, has evolved from polypropylene to polyethylene (PE) and, finally, to polyvinylidene fluoride (PVDF) (Tatsuki & Kenji 1999; Jeffrey et al. 2000). Currently, PVDF membranes are widely used because of economic considerations. A PVDF membrane generally has a hollow fibrous or flat-sheet structure. A hollow fibrous membrane can easily break up during operation, with the fracture not easily found, causing a deterioration in water quality. A ceramic membrane is a type of solid membrane. MBR technology based on a ceramic membrane has a very low running cost. The membrane can be recycled, which corresponds with the national industrial policy of promoting a circular economy featuring energy conservation and emission reduction. The ceramic membrane can be used as the recycling membrane module in the treatment of industrial and municipal wastewater. Compared to conventional organic membranes, the advantages of a flat ceramic membrane are: (1) High permeability performance: the water permeability can be increased by backwashing, (2) Energy-saving and cost reducing: the aeration consumption due to membrane cleaning can be reduced by 50%, (3) Durability: the ceramic membrane has a long operating life and potentially can withstand a pH range of 1–14, (4) Easy maintenance and management: the membrane could be backwashed automatically, and (5) No pollution: the ceramic membrane is made of an environmentally-friendly material that will not create secondary pollution.

This paper describes how a flat-sheet ceramic membrane, which can be produced at a low cost, was used as a membrane module to get a higher permeate flux. The newly-developed ceramic membrane bioreactor (CMBR) process, which consisted of a 1 mm screen, raw water tank, and submerged flat-sheet ceramic membrane module, was evaluated for the treatment of campus wastewater.

CMBR

The flat-sheet ceramic membrane has a normal pore size of 0.1 μm, boundary dimensions of W280 × H1046 × T12 mm, effective area of 0.5 m2, and dry weight of 2.2 kg (Figure 1). The capture rate of 0.1 μm particles can reach over 95%. The pH application range is 1–14, and there is an upper temperature limit of 80 °C. The filter type is out-to-in type, and one membrane module contains 100 membrane flats. The setup and technical information are shown in Figure 1.
Figure 1

Membrane modules and flat-sheet membrane.

Figure 1

Membrane modules and flat-sheet membrane.

Close modal

The campus wastewater used in the system was from the primary sedimentation tank of the wastewater treatment plant. First, the wastewater was pumped into the clean screening equipment by a sinking pump, and then it went through the 1 mm screen into the raw water tank. The screen removed the large particles and solid impurities before the wastewater entered the raw water tank. The raw water tank served as a storage pool for the wastewater that was used in the system.

The raw water pump was connected to the raw water tank and the anoxic tank. With the help of the pumping action of the suction/backwashing pump, the wastewater went through the flat-sheet ceramic membrane for drug washing. An aeration blower was used for continuous aeration of the aerobic tank. The aeration provided oxygen for the growth of the microorganisms, and helped clean the flat-sheet ceramic membrane surface through airflow oscillation to prevent membrane clogging (Figure 2).
Figure 2

Diagram of CMBR process.

Figure 2

Diagram of CMBR process.

Close modal

Wastewater composition

The GUT campus wastewater included domestic water from the student dormitories, and water from public use. The water from the student dormitories included drinking water and undrinkable water for washing. The undrinkable water included water used for irrigation, firefighting, and cleaning. Different departments and buildings have different water utilization structures, and the amount of water used for the same purposes is also different. The amounts of water used and discharged by different parts of the GUT Yanshan campus are listed in Table 1.

Table 1

The amount of water used and discharged by different parts of Yanshan campus

PartsIndexUnitThe amount of water usedMaximum amount
Used (m3)Discharged (m3)
Students' dormitory 200 L/(person·d) person 30,000 6,000 4,800 
Dining Hall 20 L/(person·d) person 31,875 637.5 510 
Teaching and Office Buildings 50 L/(person·d) person 31,875 1,593.6 1,275 
Laboratories 170 L/(room·d) room 18 3.1 2.5 
Infirmary 
 Medical stuff 50 L/(person·d) person 12 0.6 4.8 
 Patients 20 L/(person·d) person /d 270 5.4 
Irrigation 3 L/(m2·d) m2 540,270 1,620.8 evaporation 
Playground 3 L/(person·d) person 30,000 90 72 
Subtotal    9,951  
Unexpected amount (10%)    995.1  
Total amount    10,946.1  
PartsIndexUnitThe amount of water usedMaximum amount
Used (m3)Discharged (m3)
Students' dormitory 200 L/(person·d) person 30,000 6,000 4,800 
Dining Hall 20 L/(person·d) person 31,875 637.5 510 
Teaching and Office Buildings 50 L/(person·d) person 31,875 1,593.6 1,275 
Laboratories 170 L/(room·d) room 18 3.1 2.5 
Infirmary 
 Medical stuff 50 L/(person·d) person 12 0.6 4.8 
 Patients 20 L/(person·d) person /d 270 5.4 
Irrigation 3 L/(m2·d) m2 540,270 1,620.8 evaporation 
Playground 3 L/(person·d) person 30,000 90 72 
Subtotal    9,951  
Unexpected amount (10%)    995.1  
Total amount    10,946.1  

In most Chinese universities, cooking is forbidden in student dormitories because of the risk of natural gas leakage. Therefore, only personal hygiene and laundry washing generate wastewater from the dormitories. Three canteens serve all the university students at GUT Yanshan campus. Leftover food was collected in the trash. Tableware were washed by the canteen workers. Oil separation units were used to separate the oil from the wastewater according to Chinese law. The average concentration of oil in the inlet of the aeration chamber was 12.53 mg/L during dry weather and 7.67 mg/L during storm water conditions. Unlike municipal wastewater (Jeffrey et al. 2000), nitrogen concentrations were relatively high throughout the test period (average: TN = 76 mg/L). Table 2 shows the concentrations of relevant parameters measured at the inlet of to the CMBR throughout the test period.

Table 2

Composition of wastewater in winter and summer time at GUT

ParameterUnitWinter timeSummer time
COD mg/L 167.5–401.1 101.2–521.0 
BOD5 mg/L 113.6–252.1 36.1–241.1 
SS mg/L 90–120 37.7–120 
TN mg/L 48.2–80.8 47.4–122.7 
TP mg/L 3.6–8.0 5.5–11.3 
pH  6.34–7.44 6.60–7.56 
Temperature °C 11–16 20–28 
Oil mg/L 3.39–20.46 1.2–21.18 
ParameterUnitWinter timeSummer time
COD mg/L 167.5–401.1 101.2–521.0 
BOD5 mg/L 113.6–252.1 36.1–241.1 
SS mg/L 90–120 37.7–120 
TN mg/L 48.2–80.8 47.4–122.7 
TP mg/L 3.6–8.0 5.5–11.3 
pH  6.34–7.44 6.60–7.56 
Temperature °C 11–16 20–28 
Oil mg/L 3.39–20.46 1.2–21.18 

Analytical method

CODCr, SV30, SS, BOD5, oil: Filtered CODCr (1 μm) was measured by the closed reflux colorimetric method (APHA 1995). SS and volatile suspended solids in effluent and sludge samples were measured in accordance with Standard Methods (Wenjie et al. 2009, 2011). Oil was measured by infrared spectrophotometry (ET1200, EURO-TECH). Total nitrogen (TN) was determined by the persulfate method (Jin et al. 2015; Wenjie et al. 2014, 2015) using the UV spectrophotometric screening method (Yue & Wenjie 2016) for quantification of TN as NO3-N (the oxidization product of the persulfate digestion). The pH was measured by using a pH meter (9010, Jenco, USA), and dissolved oxygen (DO) was measured by using a DO meter (6010, Jenco, USA).

The total sludge content was estimated as mixed-liquor suspended solids (MLSS). For determination of MLSS, a sludge sample of known volume was washed twice by centrifugation (1,000× g for 15 min), decanted and re-suspended in deionized water, dried to a constant weight at 105 °C, and then cooled to room temperature using desiccation prior to weighing.

Reactor performance

As shown in Figure 3(a), the influent CODCr values were in the range of 167.5–379.9 mg/L. During the operation period, the CODCr removal rate reached 82.4%, and the mean value of the CODCr in the output water was 44.8 mg/L. Some of the output water had a higher CODCr, partly due to the high concentration of the dissolvable organic compounds in the input water. These organic compounds could not be degraded fully before they passed through the membrane and were released.
Figure 3

The performance of CMBR process (a) COD variety; (b) BOD5 variety.

Figure 3

The performance of CMBR process (a) COD variety; (b) BOD5 variety.

Close modal

As shown in the Figure 3(b), the effluent BOD5 remained almost stable within an acceptable inflow BOD5 of 113.57–252.1 mg/L. The mean percentage of the BOD5 removal rate was 94.5%, and the mean value of the BOD5 concentration in the effluent was closer to 0 mg/L.

Transmembrane pressure and flux

Transmembrane pressure (TMP) is an important indicator of how much a membrane is clogged by particles. In this study, backwashing was adopted to prevent the clogging of the membrane. During the operation period, the membrane was backwashed for 1 h each week to maintain its functionality.

Aeration is an important parameter in MBR operating conditions, because it not only provides necessary oxygen for the microorganisms, but also increases the turbulent flow in a reactor, causing a waterpower shearing force. Scouring the membrane by aeration relieved the sedimentation of the sludge and reduced the clogging of the membrane. During the hour of backwashing once a week, 0.5 min of backwashing was performed for every 9.5 min of filtration. Figure 4 shows the change in the TMP during the study. During the initial operation, the TMP changed greatly. This may have been caused by tiny floc sludge, most of which was inorganic matter that could adhere easily to the membrane surface. After six days, the TMP had built up to −32.85 kPa. The system was stopped to clean the membrane by using 1% NaClO for 1 h to reduce the TMP. Because of the increase in the sludge concentration, the amount of the input water increased from the 17th day, as did the filtration and backwashing flux. On day 27, the TMP approached the maximum value again. The value returned to normal after the system was cleaned by using 1% NaClO for 1 h. After enhancing the blowing rate on the membrane and the amount of aeration, no obvious increase in the TMP was observed during subsequent operation.
Figure 4

TMP change during the experiment period.

Figure 4

TMP change during the experiment period.

Close modal

SV30, MLSS and viscosity

Figures 57 characterize the changes in sludge concentration. In the initial stage of operation, the concentration of MLSS was about 4,000 mg/L in the aerobic tank. During operation, the microorganisms reproduced and grew quickly, causing a rapid increase in biomass. During the test, the MLSS exceeded 10,000 mg/L. After 64 days, 0.9 m3 of sludge was discharged from the aerobic tanks. In the following days, the MLSS declined and then gradually increased again.
Figure 5

The change of the sludge concentration in anaerobic tank.

Figure 5

The change of the sludge concentration in anaerobic tank.

Close modal
Figure 6

The change of the sludge concentration in aerobic tank.

Figure 6

The change of the sludge concentration in aerobic tank.

Close modal
Figure 7

The change of viscosity in anaerobic tank and aerobic tank.

Figure 7

The change of viscosity in anaerobic tank and aerobic tank.

Close modal

SV30 increased quickly at the initial stage of operation, and the value remained above 95% for most of the test. There was essentially no change in the SV30 value after the sludge was discharged from the system.

Along with the increases in the sludge concentration, the volume of the sludge floc also increased, as did the viscosity (Figure 7). On day 90, DO was increased to inhibit the growth of filamentous bacteria. This caused the viscosity to decrease and mostly level off at a relatively low value.

The results show that the CMBR had stable removal efficiencies for organic compounds. Affected by the temperature and concentrations of the sludge, the CODCr of at the outlet in winter was higher than that in summer, with mean values of 44.8 mg/L and 34.3 mg/L, respectively. The outlet was always free of solids. Thus, except for the coliform group, all effluent values met the standards specified in Table 3 for the use of treated water for irrigation. The filtrate tank is a possible the pollutant source. Therefore, disinfection should be carried out before used.

Table 3

Effluent properties of the CMBR process and requirements on effluent quality for the reuse as irrigation purpose

ItemFiltration waterReclaimed water regulationa
Turbidity (NTU) <0.5 Less than 10 
Coliform Group (CFU) <60 Less than 3 (/1 L) 
BOD5 (mg/L) <0.5 Less than 20 
CODCr (mg/L) <40 – 
NH4-N (mg/L) <20 Less than 20 
TN (mg/L) <15 – 
TP (mg/L) <3 – 
ItemFiltration waterReclaimed water regulationa
Turbidity (NTU) <0.5 Less than 10 
Coliform Group (CFU) <60 Less than 3 (/1 L) 
BOD5 (mg/L) <0.5 Less than 20 
CODCr (mg/L) <40 – 
NH4-N (mg/L) <20 Less than 20 
TN (mg/L) <15 – 
TP (mg/L) <3 – 

aRefer to GB/T18920—2002, reuse of recycling from urban/water quality standard for urban miscellaneous water consumption.

Oil was found to be harmful to the organic membrane. In this study, oil concentrations of up to 20 mg/L had no adverse effect on the TMP and permeate flux (Figure 8).
Figure 8

Oil change following the process.

Figure 8

Oil change following the process.

Close modal

In addition, no oil accumulation was detected in the aeration tank (Figure 8). It can be deduced that oil-degrading microorganisms were successfully cultivated in the CMBR process. The CMBR process ran with a stable flux of about 33 L/m2 h, with the TMP maintained at −10 kPa until the end of the experiment period. To our knowledge, this is the highest flux observed in wastewater treatment to date.

The CMBR plant construction total investment composed of infrastructure cost, design cost, and equipment expenses, was US$85,937. The operation costs, including electricity, chemical costs, and labor costs, are estimated to be about US$0.056/m3. Once the treated water is substituted appropriately for potable water, investment recovery takes only 3 years.

As shown in Table 4, the flat-sheet type of ceramic membrane used in this study had more advantages that that of the other type of organic membranes (PVDF and PE) (Meng et al. 2017). For domestic wastewater treatment, high fluxes of up to 42 L/m2 h can be achieved with this type of membrane, whereas only 25 L/m2 h and 17 L/m2 h fluxes are possible for flat-sheet type organic membranes and hollow-fiber type membranes, respectively. Long life, about two times longer than organic membranes, is another merit for flat-sheet ceramic membranes. The operating cost and maintenance are also superior. The footprint of a flat-sheet ceramic membrane is higher than that of a flat-sheet organic membranes but lower than that of a hollow-fiber type organic membrane. The flat-sheet ceramic membrane can be recycled after used. Therefore, flat-sheet ceramic membranes, as demonstrated in this study, are a good prospect for use in the MBR process.

Table 4

Comparison of membranes

Membrane TypeFlat-sheet Type
Hollow-fiber Type
MaterialCeramicPolymer (PVDF, PE)Polymer (PVDF)
Flux (Domestic wastewater) ++ +
42 L/m2
++
25 L/m2
+
17 L/m2
Stable ++
Permeability was recovered after inline cleaning 
++ ++ 
Life-time 15 years 8 years 7 years 
Operating cost ++ +
50% reduction of energy consumption 
++ ++ 
Maintenance ++ +
Automatic backwash and inline cleaning
Easy to wash the surface of membrane 
++
Backwash is impossible 
+
Washing is not effective due to the hollow-fiber structure 
Footprint ++
Higher flux than Flat-sheet polymer 
++ + 
Recycle ++ +
Recyclable 

Incineration disposal 

Incineration disposal 
Membrane TypeFlat-sheet Type
Hollow-fiber Type
MaterialCeramicPolymer (PVDF, PE)Polymer (PVDF)
Flux (Domestic wastewater) ++ +
42 L/m2
++
25 L/m2
+
17 L/m2
Stable ++
Permeability was recovered after inline cleaning 
++ ++ 
Life-time 15 years 8 years 7 years 
Operating cost ++ +
50% reduction of energy consumption 
++ ++ 
Maintenance ++ +
Automatic backwash and inline cleaning
Easy to wash the surface of membrane 
++
Backwash is impossible 
+
Washing is not effective due to the hollow-fiber structure 
Footprint ++
Higher flux than Flat-sheet polymer 
++ + 
Recycle ++ +
Recyclable 

Incineration disposal 

Incineration disposal 

The CMBR process has proven to be a robust system, capable of producing high quality service water from campus wastewater. Even under difficult local conditions, such as varying inlet concentrations, the quality requirements for the use of treated wastewater were met throughout the whole test period after disinfection. Fresh water demand can be reduced significantly if this service water is used for irrigation. Fluxes above 33 L/m2 h, with MLSS at about 12,000 mg/L, were maintained. By using the CMBR process, treated water meets the Chinese national standards for landscaping irrigation.

This research was supported by Guilin Scientific Research and Technology Development Program (No. 2016012303), Guangxi Scientific Experiment Center of Mining, Metallurgy and Environment (KH2012ZD004), the project of high level innovation team and outstanding scholar in Guangxi colleges and universities (002401013001).

APHA
1995
Standard Method for the Examination of Water and Wastewater
, 19th edn.
American Public Health Association
,
Washington, DC
.
Jin
Y.
Wang
D.
Zhang
W.
2015
Use of bamboo charcoal reduced the cultivated anammox seed sludge dosage during the start-up period
.
Desalin. Water Treat.
doi:10.1080/19443994.2015.1107510
.
Khalid
B. M.
Zakaria
A. Q.
Mohammad
A. S.
Ahmad
Q.
Mohammed
R. Q.
Malek
A.
2015
On the performance of real grey water treatment using a submerged membrane bioreactor system
.
J. Membrane Sci.
476
,
40
49
.
Kim
I.
Choi
D. C.
Lee
J.
Chae
H. R.
Jang
J. H.
Lee
C. H.
Park
P. K.
Won
Y. J.
2015
Preparation and application of patterned hollow-fiber membranes to membrane bioreactor for wastewater treatment
.
J. Membrane Sci.
490
,
190
196
.
Melin
T.
Jefferson
B.
Bixio
D.
Thoeye
C.
De Wilde
W.
De Koning
J.
van der Graaf
J.
Wintgens
T.
2006
Membrane bioreactor technology for wastewater treatment and reuse
.
Desalination
187
,
271
282
.
Meng
F.
Zhang
S.
Oh
Y.
Zhou
Z.
Shin
H.
Chae
S.
2017
Fouling in membrane bioreactors: an updated review
.
Water Res.
114
,
151
180
.
Wenjie
Z.
Xuehong
Z.
Dunqiu
W.
Yasunori
K.
Rouse
J. D.
Furukawa
K.
2011
Trace elements enhance biofilm formation in UASB reactor for solo simple molecule wastewater treatment
.
Biores. Technol.
102
,
9296
9299
.
Wenjie
Z.
Yuanyuan
Z.
Liang
L.
Xuehong
Z.
Yue
J.
2014
Fast start-up of EGSB reactor using stored Anammox sludge
.
Water Sci. Technol.
69
,
1469
1474
.
Wenjie
Z.
Huaqin
W.
Joseph
D. R.
Yue
J.
2015
Granular activated carbon as nucleus for formation of Anammox granules in an expanded granular-sludge-bed reactor
.
Global NEST J.
17
,
508
514
.