Operation of the A-stage with a mesh sieve coupled to the MF unit

In order to prevent the passage of large size suspended solids to the membrane MF units, a first barrier with a mesh sieve was installed at the A-stage outlet.

The outlet coming from the mesh sieve was pumped to three MF alternative combinations, namely, the thin mesh sieve, the support layer from the MF membrane coupled with thin mesh sieve, and the MF support layer only. However, when using the first two combinations an overpressure built up occurred which the system could not hold. Thus, the following experiment was carried out with the MF unit fitted with the membrane support layer only.

In this experiment, the permeate flow rate value was controlled at around the desired value of 8 L/d (corresponding to a flux of 14.1 L/m².h) before the sludge bleed was started, and 7.75 L/d (corresponding to a flux of 13.7 L/m².h) after starting the sludge bleed (Figure 30). To do this, the flow rate was measured on an hourly basis and the permeate pump velocity was changed when necessary. It could be observed that before the washing of the membranes (twice a week), the flux value tended to decrease but was restored after washing (Figure 30).

0 2 4 6 8 10 12

0 20 40 60 80

Flux (L/m2.h)

Conductivity (mS/cm)

2,14 M 0,75 M

50

Figure 30 – Permeate flux values in the MF system coupled to the A-stage fitted with mesh sieve, along the experimental time. Setpoint values for the permeate flow rate without (red) and with (green) sludge bleed are also given.

During the first days of this experiment, WW from the Ossemeersen WWTP was used as feed. On day 16 the feed was changed to WW from the Maria Middelares Hospital WWTP. It was observed that the latter WW has higher soluble and total COD values but lower TAN and Kj-N concentrations (see Figure 32 to Figure 35).

The dissolved oxygen concentration was monitored in the A-stage along the experimental time (Figure 31). Values were high in the first days but later decreased and leveled out with the majority of readings between 0.1 and 1 mg/L O2.

0 2 4 6 8 10 12 14 16 18

0 200 400 600 800 1000

Flux (L/m2.h)

Time (h)

Desired flux for 7,750 L/d

Desired flux for 8L/d

51

Figure 31 – Dissolved oxygen concentration in the A-stage coupled to a mesh sieve and MF unit, along the experimental time.

From the COD values given in Figure 32, it is visible that the sCOD and tCOD removal efficiencies were higher with the first than with the second WW feed. When using Maria Middelares WW, the results from samples taken from the daily permeate mix at its receiving vessel (Figure 32), give tCOD removal values between 39 and 83%. For sCOD, there is no significant difference between the two kinds of WW used, removal rate values being between 5 and 42 % for permeate samples taken at given experimental times (Figure 33) and between 2 and 40 % for the permeate daily mix samples (Figure 32). The latter samples were generally more reliable in providing an overall picture of the system’s efficiency.

0 1 2 3 4 5 6 7 8 9 10

0 200 400 600 800 1000

Oxygen concentration (mg/L)

Time (h)

52

Figure 32 – Soluble COD and total COD values in the permeate mix and fed WW from the MF system coupled to the A-stage fitted with a mesh sieve, along the experimental time.

Figure 33 - Soluble COD and total COD values in permeate time samples and in the fed WW from the MF system coupled to the A-stage fitted with a mesh sieve, along the experimental time.

When analyzing the TAN and Kj-N results, it can be concluded that the WW from the Ossemeersen WWTP had its nitrogen content essentially in the form of ammonia, since the TAN and Kj-N concentrations were very similar. However, the WW from Hospital Maria Middelares WWTP presents a significant difference between these two determinations (Figure 34 and Figure 35). In the first days, there is a slight removal of ammonia in the system, which is not apparent in the Kj-N values. From the

0 100 200 300 400 500 600 700

0 10 20 30 40

COD (mg O2/L)

Time (days)

sCOD fed WW tCOD fed WW sCOD Permeate mix tCOD Permeate mix

0 100 200 300 400 500 600 700 800 900

0 10 20 30 40

COD mg O2/L

Time (days)

sCOD fed WW tCOD fed WW sCOD time samples tCOD time samples

53 16^th day on, there is no significant ammonia and Kj-N removal in the system (Figure 34 and Figure 35). It should be noted that between days 18 and 22 low concentrations of ammonia were measured in the system.

Figure 34 - TAN values in permeate time samples and in the fed WW from the MF system coupled to the A-stage fitted with a mesh sieve, along the experimental time.

Figure 35 - Kj-N values in permeate time samples and in the fed WW from the MF system coupled to the A-stage fitted with a mesh sieve, along the experimental time.

0 20 40 60 80 100 120

0 10 20 30 40

TAN (mg/L)

Time (days)

TAN fed WW TAN time samples

0 20 40 60 80 100 120 140 160

0 10 20 30 40

Kj-N (mg/L)

Time (days)

Kj-N fed WW Kj-N time samples

54 The WW from the Ossemeersen WWTP exhibited less than half the chloride concentrations measured in the WW from the Maria Middelares WWTP. The system did not remove any of the chloride present in either WW feeds (Figure 36).

Figure 36 - Chloride concentration values in the permeate mix and fed WW from the MF system coupled to the A-stage fitted with a mesh sieve, along the experimental time.

The nitrite concentrations in both of the fed WW were always below the detection limit. There was some production of nitrite in the system during the essay (Figure 37), essentially in the early stages when the dissolved oxygen concentration was high (Figure 31).

Figure 37 - Nitrite concentration values in the permeate mix and fed WW from the MF system coupled to the A-stage fitted with a mesh sieve, along the experimental time.

0 5 10 15 20 25 30 35

0 10 20 30 40

mg Cl- /L

Time (days)

Fed WW Permeate mix

0 1 2 3 4 5 6

0 10 20 30 40

mg NO2- /L

Time (days)

Fed WW Permeate mix

55 Also, the WW added did not generally contain detectable nitrate levels and there was some production of nitrate in the system essentially in the first days (Figure 38), again coinciding with higher dissolved oxygen concentrations (Figure 31).

Figure 38 - Nitrate concentration values in the permeate mix and fed WW from the MF system coupled to the A-stage fitted with a mesh sieve, along the experimental time.

The phosphate concentration in the fed WW remained stable at around 2 mg/L. The system generally did not remove phosphate, with the exception of days 18 and 22 when all the phosphate was apparently removed (Figure 39).

0 0.05 0.1 0.15 0.2 0.25 0.3

0 10 20 30 40

mg NO3- /L

Time (days)

Fed WW Permeate mix

56

Figure 39 - Phosphate concentration values in the permeate mix and fed WW from the MF system coupled to the A-stage fitted with a mesh sieve, along the experimental time.

The sulphate concentration profile shows that there was some sulphate in the fed WW, frequently between 5 and 10 mg/L, but the system generally did not remove it, with the exception of days 18 and 22 when almost all the sulphate was apparently removed (Figure 40).

Figure 40 - Sulphate concentration values in the permeate mix and fed WW from the MF system coupled to the A-stage fitted with a mesh sieve, along the experimental time.

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5

0 10 20 30 40

mg PO43- /L

Time (days)

Fed WW Permeate mix

0 5 10 15 20 25 30 35

0 10 20 30 40

mg SO42- /L

Time (days)

Fed WW Permeate mix

57 When analyzing the TS and VS levels in the A-stage samples taken during the test, it can be concluded that the system presented higher VS/TS ratio values in the first days. With the exception of the samples around day 18, showing abnormally high TS and VS values, the system showed a steady increase in the solids contents in the A-stage, after an initial drop. After 36 days and considering the VS and TS content from the initial sludge in the experiment, the increase was of 45% in TS and 40% in VS.

Figure 41 – VS and TS values in the A-stage coupled to a mesh sieve and MF unit, along the experimental time.