Membrane Fouling Mechanisms Using Blocking Models
Membrane fouling mechanism is another important challenge for SEF of protein solution. For the sake of better
reveling fouling mechanism and controlling membrane
fouling, pore-blocking model (Table 5) proposed by Hermia
was utilized to fit the experimental data and calculate the
blocking coefficient (Zhang and Ding 2015). During the SEF
of milk solution, casein micelles (200 nm) and whey
proteins (10 nm) are main foulants producing complete pore
blocking (Luo et al. 2012). Like a homogeneous matrix in
“cherry stones,” the casein micelle is fabricated by caseins
and colloidal calcium phosphate nanoclusters (Kruif et al.
2012). As casein micelles have a much larger size than UF
pore, they could be easily rejected, but for whey proteins,
their size is similar with UF pores, thus easily causing pore
blocking (Zhang and Ding 2015).
Table 1 Experimental and
predicted values of COD and total
protein
Runs
Variable
Effluent quality
Q (L/h)
TMP
(bar)
N (rpm)
COD (mg/L)
Total protein (mg/L)
Experimental
Predicted
Experimental
Predicted
1
60
2
1500
7.7
7.7
2.72
2.69
2
120
2
1500
8.0
8.0
2.95
2.89
3
60
7
1500
7.5
7.5
2.02
2.08
4
120
7
1500
7.7
7.7
2.28
2.31
5
60
4.5
750
8.2
8.2
3.06
2.99
6
120
4.5
750
8.4
8.4
3.31
3.27
7
60
4.5
2250
7.3
7.3
1.29
1.34
8
120
4.5
2250
7.5
7.5
1.41
1.49
9
90
2
750
8.2
8.2
3.14
3.25
10
90
7
750
7.9
7.9
2.94
2.96
11
90
2
2250
7.2
7.2
1.85
1.84
12
90
7
2250
6.9
6.9
1.04
0.94
13
90
4.5
1500
7.6
7.7
2.31
2.33
14
90
4.5
1500
7.7
7.7
2.36
2.33
15
90
4.5
1500
7.8
7.7
2.32
2.33
Table 2 Experimental and
predicted values of average flux,
flux decline, and permeability
index
Runs
Variable
Flux behavior
Q (L/h)
TMP
(bar)
N (rpm)
Average flux (L/m
2 h)
Flux decline
Experimental
Predicted
Experimental
Predicted
1
60
2
1500
41.32
44.72
12.09
12.46
2
120
2
1500
45.81
50.38
15.24
14.88
3
60
7
1500
125.11
120.54
15.26
15.62
4
120
7
1500
136.36
132.96
16.79
16.42
5
60
4.5
750
79.43
78.12
19.86
20.01
6
120
4.5
750
94.60
92.12
20.59
21.48
7
60
4.5
2250
96.70
99.19
6.41
5.53
8
120
4.5
2250
101.95
103.27
7.43
7.28
9
90
2
750
37.04
34.96
19.84
19.32
10
90
7
750
108.34
114.23
21.75
21.23
11
90
2
2250
57.02
51.14
4.02
4.54
12
90
7
2250
128.18
130.27
6.81
7.33
13
90
4.5
1500
89.64
89.99
14.14
14.19
14
90
4.5
1500
90.09
89.99
14.18
14.19
15
90
4.5
1500
90.25
89.99
14.26
14.19
130
W. Zhang et al.
Membrane fouling mechanism is another important challenge for SEF of protein solution. For the sake of better
reveling fouling mechanism and controlling membrane
fouling, pore-blocking model (Table 5) proposed by Hermia
was utilized to fit the experimental data and calculate the
blocking coefficient (Zhang and Ding 2015). During the SEF
of milk solution, casein micelles (200 nm) and whey
proteins (10 nm) are main foulants producing complete pore
blocking (Luo et al. 2012). Like a homogeneous matrix in
“cherry stones,” the casein micelle is fabricated by caseins
and colloidal calcium phosphate nanoclusters (Kruif et al.
2012). As casein micelles have a much larger size than UF
pore, they could be easily rejected, but for whey proteins,
their size is similar with UF pores, thus easily causing pore
blocking (Zhang and Ding 2015).
Table 1 Experimental and
predicted values of COD and total
protein
Runs
Variable
Effluent quality
Q (L/h)
TMP
(bar)
N (rpm)
COD (mg/L)
Total protein (mg/L)
Experimental
Predicted
Experimental
Predicted
1
60
2
1500
7.7
7.7
2.72
2.69
2
120
2
1500
8.0
8.0
2.95
2.89
3
60
7
1500
7.5
7.5
2.02
2.08
4
120
7
1500
7.7
7.7
2.28
2.31
5
60
4.5
750
8.2
8.2
3.06
2.99
6
120
4.5
750
8.4
8.4
3.31
3.27
7
60
4.5
2250
7.3
7.3
1.29
1.34
8
120
4.5
2250
7.5
7.5
1.41
1.49
9
90
2
750
8.2
8.2
3.14
3.25
10
90
7
750
7.9
7.9
2.94
2.96
11
90
2
2250
7.2
7.2
1.85
1.84
12
90
7
2250
6.9
6.9
1.04
0.94
13
90
4.5
1500
7.6
7.7
2.31
2.33
14
90
4.5
1500
7.7
7.7
2.36
2.33
15
90
4.5
1500
7.8
7.7
2.32
2.33
Table 2 Experimental and
predicted values of average flux,
flux decline, and permeability
index
Runs
Variable
Flux behavior
Q (L/h)
TMP
(bar)
N (rpm)
Average flux (L/m
2 h)
Flux decline
Experimental
Predicted
Experimental
Predicted
1
60
2
1500
41.32
44.72
12.09
12.46
2
120
2
1500
45.81
50.38
15.24
14.88
3
60
7
1500
125.11
120.54
15.26
15.62
4
120
7
1500
136.36
132.96
16.79
16.42
5
60
4.5
750
79.43
78.12
19.86
20.01
6
120
4.5
750
94.60
92.12
20.59
21.48
7
60
4.5
2250
96.70
99.19
6.41
5.53
8
120
4.5
2250
101.95
103.27
7.43
7.28
9
90
2
750
37.04
34.96
19.84
19.32
10
90
7
750
108.34
114.23
21.75
21.23
11
90
2
2250
57.02
51.14
4.02
4.54
12
90
7
2250
128.18
130.27
6.81
7.33
13
90
4.5
1500
89.64
89.99
14.14
14.19
14
90
4.5
1500
90.09
89.99
14.18
14.19
15
90
4.5
1500
90.25
89.99
14.26
14.19
130
W. Zhang et al.