**2.4. Operating conditions**

The operational conditions of the combined AH–PTF are shown in Table 1. Both reactors were operated for 330 days, 30–83; 134–212; and 234–324 days at HRTs of, respectively 14.5, 10 and 7.2 h. The first 30 days of operation were considered as a start-up period, while the periods from day 83 to 134 and from 212 to 234 were considered as acclimatization periods to the new HRT.

Polyurethane Trickling Filter in Combination with

Anaerobic Hybrid Reactor for Treatment of Tomato Industry Wastewater 359

fixation, samples were rinsed three times in 0.1 mol/l of phosphate buffer solution (pH 7.3) and dehydrated gradually by successive immersions in ethanol solutions of increasing concentration (30, 50, 70, 80, 90, and 95%). The samples were then washed three times in 100% ethanol. The drying process was then completed by incubating the samples for 2 h at 40 °C. The sponge were then coated with gold powder and attached to the microscope

**3.1. Efficiency of AH reactor as a pretreatment of tomato wastewater industry** 

Figs 2a, b and c show the effect of HRT on the percentage reduction of COD fractions (COD total, COD particulate and COD soluble). By increasing the HRT from 4.3 to 8.6 h, the CODtotal of the effluent significantly reduced from 377±88 to 267±48 mg/l, and the removal efficiency of CODtotal substantially increased from 51±12 to 71±7%. However, the residual values of CODtotal in the treated effluent of the AH reactor remained unaffected by increasing the HRT from 6.0 to 8.6 h. Likely, the results in Fig. 2b show that the effluent quality of CODsoluble and removal efficiency was maintained at the same level of 117 ±11mg/l and 64±9% respectively by decreasing the HRT from 8.6 to 6 h. This indicates that the AH reactor was operated under substrate limiting conditions at an HRT of 8.6 h. Accordingly it is recommended to apply such a system at OLR 3.5 kgCOD/m3. d and HRT not exceeding 6.0 h. An increase in the HRT would result in a decrease in the wastewater linear velocity through the support material, improving the mass transfer from the liquid phase to the biomass and, therefore,

(a)

support with silver glue. SEM photographs were taken at 25 and 20 kV.

favoring the process performance (Elmitwalli et al., 2000).

**3. Results and discussion** 


**Table 1.** Operational conditions of AH reactor in combination with PTF for the treatment of tomato industry wastewater

#### **2.5. Analytical methods**

Composite samples of the influent wastewater and the treated effluents were biweekly analyzed. COD, TSS, volatile suspended solids (VSS), total Kjeldahl nitrogen (TKj-N), ammonia (NH4-N), nitrite (NO2-N), nitrate (NO3-N) and protein were analyzed according to standard methods (APHA, 2005). Raw wastewater samples were used for CODtotal, 0.45 μm membrane-filtered samples for COD soluble. The COD particulate was calculated by the differences between COD total and COD soluble, respectively. Biogas composition was measured using a gas chromatograph fitted with a thermal conductivity detector (TCD) and Poropak Q stainless steel column. The oven, injector, and detector temperatures were set as 40, 60 and 60 °C, respectively and hydrogen was used as the carrier gas. The instrument was calibrated using a mixture of 50% methane and 50% carbon dioxide. Volatile fatty acid (VFA) concentration was measured after centrifuging the samples to remove the suspended solids. A gas-liquid chromatograph equipped with a Flame Ionization Detector (FID) and Chromasorb 101 column was used for the analysis of VFA. The detector, injector and oven temperature were 200, 195 and 180 °C, respectively. The carrier gas used was nitrogen, and a mixture of hydrogen and air was used to sustain the flame in the detector.

#### **2.6. Scanning electron microscope (SEM)**

The surface of sponge carriers and the attached microorganism species in the PTF reactor were analyzed by a JSM-5600 LV scanning electron microscope (JEOL, Japan). A sample of the microorganisms attached to the carriers was withdrawn from the PTF and placed in bottles. After drying for 10 h under vacuum at 40 °C, these samples were fixed in 0.1 mol/l phosphate buffer solution (pH 7.3) containing 2.5% glutaraldehyde for 12 h at 4 °C. After fixation, samples were rinsed three times in 0.1 mol/l of phosphate buffer solution (pH 7.3) and dehydrated gradually by successive immersions in ethanol solutions of increasing concentration (30, 50, 70, 80, 90, and 95%). The samples were then washed three times in 100% ethanol. The drying process was then completed by incubating the samples for 2 h at 40 °C. The sponge were then coated with gold powder and attached to the microscope support with silver glue. SEM photographs were taken at 25 and 20 kV.

#### **3. Results and discussion**

358 Polyurethane

**2.4. Operating conditions** 

HRT (h)

OLR ( kgCOD/m3.d)

to the new HRT.

Operational conditions/ reactors

industry wastewater

**2.5. Analytical methods** 

The operational conditions of the combined AH–PTF are shown in Table 1. Both reactors were operated for 330 days, 30–83; 134–212; and 234–324 days at HRTs of, respectively 14.5, 10 and 7.2 h. The first 30 days of operation were considered as a start-up period, while the periods from day 83 to 134 and from 212 to 234 were considered as acclimatization periods

> HRT (h)

AHreactor 8.6 2.8 6 3.5 4.3 4.5

PTF reactor 5.9 1.0 4 1.43 2.9 3 **Table 1.** Operational conditions of AH reactor in combination with PTF for the treatment of tomato

Composite samples of the influent wastewater and the treated effluents were biweekly analyzed. COD, TSS, volatile suspended solids (VSS), total Kjeldahl nitrogen (TKj-N), ammonia (NH4-N), nitrite (NO2-N), nitrate (NO3-N) and protein were analyzed according to standard methods (APHA, 2005). Raw wastewater samples were used for CODtotal, 0.45 μm membrane-filtered samples for COD soluble. The COD particulate was calculated by the differences between COD total and COD soluble, respectively. Biogas composition was measured using a gas chromatograph fitted with a thermal conductivity detector (TCD) and Poropak Q stainless steel column. The oven, injector, and detector temperatures were set as 40, 60 and 60 °C, respectively and hydrogen was used as the carrier gas. The instrument was calibrated using a mixture of 50% methane and 50% carbon dioxide. Volatile fatty acid (VFA) concentration was measured after centrifuging the samples to remove the suspended solids. A gas-liquid chromatograph equipped with a Flame Ionization Detector (FID) and Chromasorb 101 column was used for the analysis of VFA. The detector, injector and oven temperature were 200, 195 and 180 °C, respectively. The carrier gas used was nitrogen, and a

The surface of sponge carriers and the attached microorganism species in the PTF reactor were analyzed by a JSM-5600 LV scanning electron microscope (JEOL, Japan). A sample of the microorganisms attached to the carriers was withdrawn from the PTF and placed in bottles. After drying for 10 h under vacuum at 40 °C, these samples were fixed in 0.1 mol/l phosphate buffer solution (pH 7.3) containing 2.5% glutaraldehyde for 12 h at 4 °C. After

mixture of hydrogen and air was used to sustain the flame in the detector.

**2.6. Scanning electron microscope (SEM)**

Run 1 Run 2 Run 3

OLR ( kgCOD/m3.d) HRT (h)

OLR ( kgCOD/m3.d)

#### **3.1. Efficiency of AH reactor as a pretreatment of tomato wastewater industry**

Figs 2a, b and c show the effect of HRT on the percentage reduction of COD fractions (COD total, COD particulate and COD soluble). By increasing the HRT from 4.3 to 8.6 h, the CODtotal of the effluent significantly reduced from 377±88 to 267±48 mg/l, and the removal efficiency of CODtotal substantially increased from 51±12 to 71±7%. However, the residual values of CODtotal in the treated effluent of the AH reactor remained unaffected by increasing the HRT from 6.0 to 8.6 h. Likely, the results in Fig. 2b show that the effluent quality of CODsoluble and removal efficiency was maintained at the same level of 117 ±11mg/l and 64±9% respectively by decreasing the HRT from 8.6 to 6 h. This indicates that the AH reactor was operated under substrate limiting conditions at an HRT of 8.6 h. Accordingly it is recommended to apply such a system at OLR 3.5 kgCOD/m3. d and HRT not exceeding 6.0 h. An increase in the HRT would result in a decrease in the wastewater linear velocity through the support material, improving the mass transfer from the liquid phase to the biomass and, therefore, favoring the process performance (Elmitwalli et al., 2000).

Polyurethane Trickling Filter in Combination with

Anaerobic Hybrid Reactor for Treatment of Tomato Industry Wastewater 361

The removal efficiency of CODtotal in an AH reactor at an HRT of 4.3 h was higher than those obtained by Demirer and Chen, (2005) who used AH reactor for treatment of dairy wastewater at longer HRT of 15 days.Also, Gu¨ngo¨r and Demirer, (2004) achieved a lower COD removal efficiency of 37.9–50% in anaerobic batch reactor treating food industry wastewater.. The improved removal efficiency of CODtotal in this study was mainly due to a higher removal efficiency of COD particulate as shown in Fig. 2c. In previous studies on opaque beer wastewater with UASB, 57% CODtotal reduction was achieved at HRT of 24 h (Parawira et al., 2005). Similarly, studies of Cronin and Lo (1998) and Driessen and Vereijken (2003) on UASB with brewery wastewater showed that the CODtotal reduction of 75–80% with the HRT in the range of 12–36 h. In the present study AH reactor could be optimally operated at an OLR of 3.5 kg COD/m3.d and HRT not exceeding 6 h with CODtotal reduction of 71% and methane yield of 0.48 m3 CH4/kg CODtotal reduced. This high efficiency of AH reactor as compared to UASB reactor can be due to the presence of polyurethane carrier material in the sedimentation part which overcome sludge washout and improve the biodegradation process. Moreover,polyurethane carriers provide a much larger surface area for the attachment of biofilm which then leads to an increase of anaerobic biodegradation process. Variations of VFAs in the influent and effluent of AH reactor are shown in Fig. 3a. Although, there was a significant fluctuation in the VFAs of the feed between 198 and 689 mg/l, the AH reactor showed that VFAs in the feed was effectively utilized by methanogenesis bacteria. VFAs in the effluent was quite low (below 121±23 mg/l) at HRTs of 8.3 and 6 h.However, the residual values of VFAs in the treated effluent was increased at decreasing the HRT(4.3 h) as shown in Fig.3a). Apparently, this can be attributed to limited activity of methanogens in the reactor under these operating conditions. Likely, Amit et al., (2007), found that the VFAs concentration increased in the treated effluent of AH reactor

treating industrial cluster wastewater, when the HRT reduced from 12 to 4 h.

(ABR) fed with dairy wastewater at a HRT of 5 d. (Chen and Shyu, 1996).

as shown in Fig.3b.

The variations of biogas production at different HRTs are shown in Fig. 3b. The biogas production was low (2.6 l/d) at an HRT of 4.3 h. HRT was prolonged up to 6 and 8.3 h and the gas production reached as high as 4.0 l/d, equivalent to 0.48 m3/ kg COD removed. d. Similarly, Oscar et al., (2008) found that the value of methane yield in an AH treating food industry wastewater increased from 0.07 to 0.18 l CH4/g COD added when the HRT increased from 1.0 to 5.5 days. The average methane yield in the gas composition was 67%

AH reactor was found to be very effective for removal of TSS and VSS as shown in Figs 4a and b. TSS and VSS removal efficiencies increased from 57 ±10 to 70±8 % and from 70±8 to 78±4 % when the HRT rose from 4.3 to 6 h and from 6 to 8.3 h., respectively. The results obtained demonstrate that clogging of the support polyurethane media in the AH reactor was not evident in spite of the high concentration of TSS contained in the influent (Fig.4a). A previous study (Vartak et al., 1997) reported VSS removal efficiencies of up to 91% in upflow anaerobic attached film reactors with a combination of limestone and polyester as support media treating diluted dairy wastewaters but operating at a longer HRT of 33 d. Lower VSS removal efficiencies (66%) have been achieved in an anaerobic baffled reactor

(b)

**Figure 2.** (a) COD total removal efficiency in an AH reactor treating Tomato industry wastewater at different HRTs ; (b) COD soluble removal efficiency in an AH reactor treating Tomato industry wastewater at different HRTs; (c) COD particulate removal efficiency in an AH reactor treating Tomato industry wastewater at different HRTs

The removal efficiency of CODtotal in an AH reactor at an HRT of 4.3 h was higher than those obtained by Demirer and Chen, (2005) who used AH reactor for treatment of dairy wastewater at longer HRT of 15 days.Also, Gu¨ngo¨r and Demirer, (2004) achieved a lower COD removal efficiency of 37.9–50% in anaerobic batch reactor treating food industry wastewater.. The improved removal efficiency of CODtotal in this study was mainly due to a higher removal efficiency of COD particulate as shown in Fig. 2c. In previous studies on opaque beer wastewater with UASB, 57% CODtotal reduction was achieved at HRT of 24 h (Parawira et al., 2005). Similarly, studies of Cronin and Lo (1998) and Driessen and Vereijken (2003) on UASB with brewery wastewater showed that the CODtotal reduction of 75–80% with the HRT in the range of 12–36 h. In the present study AH reactor could be optimally operated at an OLR of 3.5 kg COD/m3.d and HRT not exceeding 6 h with CODtotal reduction of 71% and methane yield of 0.48 m3 CH4/kg CODtotal reduced. This high efficiency of AH reactor as compared to UASB reactor can be due to the presence of polyurethane carrier material in the sedimentation part which overcome sludge washout and improve the biodegradation process. Moreover,polyurethane carriers provide a much larger surface area for the attachment of biofilm which then leads to an increase of anaerobic biodegradation process.

360 Polyurethane

(b)

(c)

**Figure 2.** (a) COD total removal efficiency in an AH reactor treating Tomato industry wastewater at different HRTs ; (b) COD soluble removal efficiency in an AH reactor treating Tomato industry wastewater at different HRTs; (c) COD particulate removal efficiency in an AH reactor treating Tomato

industry wastewater at different HRTs

Variations of VFAs in the influent and effluent of AH reactor are shown in Fig. 3a. Although, there was a significant fluctuation in the VFAs of the feed between 198 and 689 mg/l, the AH reactor showed that VFAs in the feed was effectively utilized by methanogenesis bacteria. VFAs in the effluent was quite low (below 121±23 mg/l) at HRTs of 8.3 and 6 h.However, the residual values of VFAs in the treated effluent was increased at decreasing the HRT(4.3 h) as shown in Fig.3a). Apparently, this can be attributed to limited activity of methanogens in the reactor under these operating conditions. Likely, Amit et al., (2007), found that the VFAs concentration increased in the treated effluent of AH reactor treating industrial cluster wastewater, when the HRT reduced from 12 to 4 h.

The variations of biogas production at different HRTs are shown in Fig. 3b. The biogas production was low (2.6 l/d) at an HRT of 4.3 h. HRT was prolonged up to 6 and 8.3 h and the gas production reached as high as 4.0 l/d, equivalent to 0.48 m3/ kg COD removed. d. Similarly, Oscar et al., (2008) found that the value of methane yield in an AH treating food industry wastewater increased from 0.07 to 0.18 l CH4/g COD added when the HRT increased from 1.0 to 5.5 days. The average methane yield in the gas composition was 67% as shown in Fig.3b.

AH reactor was found to be very effective for removal of TSS and VSS as shown in Figs 4a and b. TSS and VSS removal efficiencies increased from 57 ±10 to 70±8 % and from 70±8 to 78±4 % when the HRT rose from 4.3 to 6 h and from 6 to 8.3 h., respectively. The results obtained demonstrate that clogging of the support polyurethane media in the AH reactor was not evident in spite of the high concentration of TSS contained in the influent (Fig.4a). A previous study (Vartak et al., 1997) reported VSS removal efficiencies of up to 91% in upflow anaerobic attached film reactors with a combination of limestone and polyester as support media treating diluted dairy wastewaters but operating at a longer HRT of 33 d. Lower VSS removal efficiencies (66%) have been achieved in an anaerobic baffled reactor (ABR) fed with dairy wastewater at a HRT of 5 d. (Chen and Shyu, 1996).

Polyurethane Trickling Filter in Combination with

Anaerobic Hybrid Reactor for Treatment of Tomato Industry Wastewater 363

(a)

(b)

**Figure 4.** (a) TSS removal efficiency in an AH reactor treating Tomato industry wastewater at different HRTs ; (b) VSS removal efficiency in an AHreactor treating Tomato industry wastewater at different

HRTs

(a)

**Figure 3.** (a) VFAs removal efficiency in an AH reactor treating Tomato industry wastewater at different HRTs; (b) Biogas production and gas composition in an AH reactor treating Tomato industry wastewater at different HRTs

(a)

(b)

**Figure 3.** (a) VFAs removal efficiency in an AH reactor treating Tomato industry wastewater at different HRTs; (b) Biogas production and gas composition in an AH reactor treating Tomato industry

wastewater at different HRTs

(a)

**Figure 4.** (a) TSS removal efficiency in an AH reactor treating Tomato industry wastewater at different HRTs ; (b) VSS removal efficiency in an AHreactor treating Tomato industry wastewater at different HRTs

No significant difference was found in the removal of protein in the AH reactor between different HRTs as shown in Fig. 5. The maximum conversion of protein was achieved and accounted for 19.8±8.5% at an HRT of 4.3 h of the protein content. The conversion of protein dropped at an HRT of 8.6 and 6 h (14±5%). The drop in protein hydrolysis might be due to chemical precipitation of NH4-N ( Miron et al., 2000).

Polyurethane Trickling Filter in Combination with

Anaerobic Hybrid Reactor for Treatment of Tomato Industry Wastewater 365

kgCOD/m3. d., and HRT not exceeding 4.0 h. The results obtained in this investigation were higher than those obtained by El-kamah et al., (2010 & 2011) who used down flow hanging sponge (DHS) system for post treatment of anaerobically pretreated onion industry wastewater. The system was operated at an OLR of 5.1 kgCOD/m3.d. and a similar HRT of 4.2 h. The system provided an effluent quality of 80 mg/l for COD and 30 mg/l for TSS.

(a)

(b)

**Figure 5.** Protein removal efficiency in an AH reactor treating Tomato industry wastewater at different HRTs

#### **3.2. Polyurethane trickling filter (PTF) as a post-treatment system**

The results presented in Figs. 6a, b and c show the effect of OLR on the removal efficiency of the different COD fractions (COD total , COD soluble and COD particulate) in the PTF system treating AH reactor effluent. The results reveal a significantly improved COD total removal at decreasing the OLR. The system provided a mean effluent quality of 35±9 mg/l for COD total at an OLR of 1.0 kgCOD/m3.d., which is significantly lower than that at an OLR of 3.0 kgCOD/m3.d (86±16 mg/l). The improved removal efficiency of COD total was mainly due to a higher removal efficiency of COD soluble and COD particulate (Figs. 6b and c). This excellent performance towards the removal of COD soluble and COD particulate matter can be attributed to entrapment or/and adsorption followed by hydrolysis and degradation in the polyurethane packing material. Low removal efficiency of COD total at an OLR of 3 kgCOD/m3.d can be explained by excess biofilm accumulation, filling in pores of the polyurethane packing material and reducing the mass transfer capabilities (Chen et al., 2006; Tawfik & Klapwijk, 2010) and DO concentration dropped from 5.2 to 3.2 mg/l in the PTF as the OLR increased from 1.0 to 3 kgCOD/m3. d. However, the results presented in Fig. 6a show that the residual value of CODtotal in the treated effluent of the PTF system remained unaffected by decreasing the OLR from 1.43 to 1.0 kgCOD/m3.d, as a result of increasing the HRT from 4.0 to 5.9 h. Accordingly it is recommended to apply such a system at loading rate of 1.43 kgCOD/m3. d., and HRT not exceeding 4.0 h. The results obtained in this investigation were higher than those obtained by El-kamah et al., (2010 & 2011) who used down flow hanging sponge (DHS) system for post treatment of anaerobically pretreated onion industry wastewater. The system was operated at an OLR of 5.1 kgCOD/m3.d. and a similar HRT of 4.2 h. The system provided an effluent quality of 80 mg/l for COD and 30 mg/l for TSS.

364 Polyurethane

HRTs

No significant difference was found in the removal of protein in the AH reactor between different HRTs as shown in Fig. 5. The maximum conversion of protein was achieved and accounted for 19.8±8.5% at an HRT of 4.3 h of the protein content. The conversion of protein dropped at an HRT of 8.6 and 6 h (14±5%). The drop in protein hydrolysis might be due to

**Figure 5.** Protein removal efficiency in an AH reactor treating Tomato industry wastewater at different

The results presented in Figs. 6a, b and c show the effect of OLR on the removal efficiency of the different COD fractions (COD total , COD soluble and COD particulate) in the PTF system treating AH reactor effluent. The results reveal a significantly improved COD total removal at decreasing the OLR. The system provided a mean effluent quality of 35±9 mg/l for COD total at an OLR of 1.0 kgCOD/m3.d., which is significantly lower than that at an OLR of 3.0 kgCOD/m3.d (86±16 mg/l). The improved removal efficiency of COD total was mainly due to a higher removal efficiency of COD soluble and COD particulate (Figs. 6b and c). This excellent performance towards the removal of COD soluble and COD particulate matter can be attributed to entrapment or/and adsorption followed by hydrolysis and degradation in the polyurethane packing material. Low removal efficiency of COD total at an OLR of 3 kgCOD/m3.d can be explained by excess biofilm accumulation, filling in pores of the polyurethane packing material and reducing the mass transfer capabilities (Chen et al., 2006; Tawfik & Klapwijk, 2010) and DO concentration dropped from 5.2 to 3.2 mg/l in the PTF as the OLR increased from 1.0 to 3 kgCOD/m3. d. However, the results presented in Fig. 6a show that the residual value of CODtotal in the treated effluent of the PTF system remained unaffected by decreasing the OLR from 1.43 to 1.0 kgCOD/m3.d, as a result of increasing the HRT from 4.0 to 5.9 h. Accordingly it is recommended to apply such a system at loading rate of 1.43

**3.2. Polyurethane trickling filter (PTF) as a post-treatment system** 

chemical precipitation of NH4-N ( Miron et al., 2000).

Polyurethane Trickling Filter in Combination with

Anaerobic Hybrid Reactor for Treatment of Tomato Industry Wastewater 367

(a)

(b)

**Figure 7.** (a) The efficiency of PTF for removal of TSS different OLRs; (b) The efficiency of PTF for

removal of VSS at different OLRs

**Figure 6.** (a) The efficiency of PTF for removal of COD total at differentOLRs; (b) The efficiency of PTF for removal of COD soluble at different OLRs; (c) The efficiency of PTF for removal of COD particulate at different OLRs

The results in Figs. 7a and b revealed that the removal efficiencies of TSS and VSS in the PTF reactor significantly decreased at increasing the OLR from 1.43 to 3.0 kgCOD/m3.d., while decreasing the OLR from 1.43 to 1.0 kgCOD/m3.d did not affect seriously on the removal efficiencies. The reactor achieved removal efficiencies of 87.1; 87 and 78.6% for TSS and 89.3; 88.5 and 79.5 % for VSS at OLRs of 1,1.43 and 3.0 kgCOD/m3.d. respectively. This high removal efficiency for coarse suspended solids in PTF reactor were mainly due to the high entrapment capacity, high specific surface area and porosity of the polyurethane packing material. Tawfik &klapwijk, (2010) found that polyurethane is better than polystyrene packing media for removal of TSS and VSS.

The nitrification efficiency in the PTF treating AH reactor effluent at different OLRs is shown in Fig. 8a. The results show that increasing the OLR from 1.0 to 1.43 and from 1.43 to 3.0 kg COD/m3.d, results in an increase of the ammonia concentration in the final effluent from 2.7±1.3 to 2.8±1.3 mg/l and from 2.8±1.3 to 17.8±3.7 mg/l, respectively. At OLR of 1.1, 1.43, and 3.0 kg COD/ m3.d, ammonia was removed by values of 89.4±5.9%, 89.7±4.6 % and 25 ±10 %, while at the same time 17.7 ±3.5, 17±4.4 and 1.7±0.9 mg/l of nitrate were, respectively produced as shown in Fig. 8a. Based on these results, it can be concluded that the OLR imposed to the PTF reactor should remain below 3 kg COD/m3.d to achieve a high nitrification efficiency as also found by El-kamah et al., (2011) for down flow hanging sponge (DHS) system treating anaerobically pretreated onion industry wastewater.

different OLRs

wastewater.

packing media for removal of TSS and VSS.

(c) **Figure 6.** (a) The efficiency of PTF for removal of COD total at differentOLRs; (b) The efficiency of PTF for removal of COD soluble at different OLRs; (c) The efficiency of PTF for removal of COD particulate at

The results in Figs. 7a and b revealed that the removal efficiencies of TSS and VSS in the PTF reactor significantly decreased at increasing the OLR from 1.43 to 3.0 kgCOD/m3.d., while decreasing the OLR from 1.43 to 1.0 kgCOD/m3.d did not affect seriously on the removal efficiencies. The reactor achieved removal efficiencies of 87.1; 87 and 78.6% for TSS and 89.3; 88.5 and 79.5 % for VSS at OLRs of 1,1.43 and 3.0 kgCOD/m3.d. respectively. This high removal efficiency for coarse suspended solids in PTF reactor were mainly due to the high entrapment capacity, high specific surface area and porosity of the polyurethane packing material. Tawfik &klapwijk, (2010) found that polyurethane is better than polystyrene

The nitrification efficiency in the PTF treating AH reactor effluent at different OLRs is shown in Fig. 8a. The results show that increasing the OLR from 1.0 to 1.43 and from 1.43 to 3.0 kg COD/m3.d, results in an increase of the ammonia concentration in the final effluent from 2.7±1.3 to 2.8±1.3 mg/l and from 2.8±1.3 to 17.8±3.7 mg/l, respectively. At OLR of 1.1, 1.43, and 3.0 kg COD/ m3.d, ammonia was removed by values of 89.4±5.9%, 89.7±4.6 % and 25 ±10 %, while at the same time 17.7 ±3.5, 17±4.4 and 1.7±0.9 mg/l of nitrate were, respectively produced as shown in Fig. 8a. Based on these results, it can be concluded that the OLR imposed to the PTF reactor should remain below 3 kg COD/m3.d to achieve a high nitrification efficiency as also found by El-kamah et al., (2011) for down flow hanging sponge (DHS) system treating anaerobically pretreated onion industry (a)

**Figure 7.** (a) The efficiency of PTF for removal of TSS different OLRs; (b) The efficiency of PTF for removal of VSS at different OLRs

Polyurethane Trickling Filter in Combination with

Anaerobic Hybrid Reactor for Treatment of Tomato Industry Wastewater 369

The results revealed that the nitrification rate in PTF was strongly dependant on VSS/ TN ratio. A low nitrification rate was achieved in the PTF at the high influent VSS/TN ratio of 5±1, the nitrification rate was 0.013 kg NOx-N/m3.d as compared to VSS/N ratio of 2.8, the nitrification rate amounted to 0.1 kgNOx-N/m3.d. This can be attributed to the attachment and degradation of volatile suspended solids on the surface of the nitrifying biofilm where they take away oxygen which otherwise would have been available for nitrifiers (Tawfik et al., 2010). The TKj-N removal in the PTF treating AH reactor effluent was 82.8 ±6.4% at an OLR of 1.0 and 1.43 kg COD/m3.d as compared to 20 ±10% at higher OLR of 3.0 kg COD/m3.d (Fig. 8b). The nitrogen loss amounted to 20% (Fig. 8a) which can be due to (1) assimilation of biomass (2) denitrification occurring in the anoxic zone of the biofilm

Profile of dissolved oxygen (DO) concentration along the height of PTF shows a gradual increase in the concentration of DO as the wastewater flows down. DO in the final effluent was in the range of 4-4.6 mg/l as shown in Fig.9a. The profile results of PTF showed that in the upper part of the PTF system, mainly COD was oxidized while nitrification was taken place in the lower part of the system, where nitrifiers are available. The results in Fig. 9 b,c and d show that most of the COD fractions (COD total, COD soluble and COD particulate) were

The 3rd segment provided a little additional removal of COD fractions as shown in Figs. 9 b, c and d. This can be explained by the fact that the most of the coarse and soluble organic matter were adsorbed and degraded in the segments 1 and 2. Likely TSS and VSS concentrations were gradually decreased from segment 1 to 3 as shown in Figs. 10 a and b. The results in Figs. 11a and b show that nitrification was very limited in the 1st segment of PTF system at OLR of 4.2 kg COD/m3 .d. This was due to the presence of an insufficient ammonia oxidizer population at high loading rate as they cannot compete with heterotrophs for space and oxygen. In the 2nd and 3rd segment of PTF system, a high nitrification rate was

These results demonstrate that at OLR exceeding 4.2 kg COD/m3 d heterotrophic bacteria still prevail in the 1st segment of PTF system, but the nitrifying bacteria promoted in the 2nd , and 3rd segment of PTF system when the OLR drops to 2.1 and 1.4 kg COD/m3.d, respectively. The ammonia oxidation and TKj-N removal (Fig. 11a and c) was virtually approximately complete, only 1.7 mgNH4-N/l and 4.0 mg TKj-N/l provided in the final

**3.4. Efficiency of the combined system (AH+PTF) treating tomato industry** 

The results presented in Table 2 revealed that decreasing the total HRT from 14 to 10 h was not significantly affected on the removal efficiency of COD fractions (COD total, COD soluble

(Holman & Wareham, 2005).

effluent of PTF system.

**3.3. Profile of polyurethane trickling filter (PTF) reactor** 

removed in the 1st and 2nd segment of PTF reactor.

achieved at lower OLRs of 2.1, and 1.4 kg COD/m3 .d.

**wastewater at different OLRs and HRTs** 

(a)

**Figure 8.** (a) Nitrification efficiency and total nitrogen removal in PTF at different OLRs; (b) The efficiency of PTF for removal of TKj-N at different HRTs and OLRs

The results revealed that the nitrification rate in PTF was strongly dependant on VSS/ TN ratio. A low nitrification rate was achieved in the PTF at the high influent VSS/TN ratio of 5±1, the nitrification rate was 0.013 kg NOx-N/m3.d as compared to VSS/N ratio of 2.8, the nitrification rate amounted to 0.1 kgNOx-N/m3.d. This can be attributed to the attachment and degradation of volatile suspended solids on the surface of the nitrifying biofilm where they take away oxygen which otherwise would have been available for nitrifiers (Tawfik et al., 2010). The TKj-N removal in the PTF treating AH reactor effluent was 82.8 ±6.4% at an OLR of 1.0 and 1.43 kg COD/m3.d as compared to 20 ±10% at higher OLR of 3.0 kg COD/m3.d (Fig. 8b). The nitrogen loss amounted to 20% (Fig. 8a) which can be due to (1) assimilation of biomass (2) denitrification occurring in the anoxic zone of the biofilm (Holman & Wareham, 2005).

#### **3.3. Profile of polyurethane trickling filter (PTF) reactor**

368 Polyurethane

(a)

(b)

**Figure 8.** (a) Nitrification efficiency and total nitrogen removal in PTF at different OLRs;

(b) The efficiency of PTF for removal of TKj-N at different HRTs and OLRs

Profile of dissolved oxygen (DO) concentration along the height of PTF shows a gradual increase in the concentration of DO as the wastewater flows down. DO in the final effluent was in the range of 4-4.6 mg/l as shown in Fig.9a. The profile results of PTF showed that in the upper part of the PTF system, mainly COD was oxidized while nitrification was taken place in the lower part of the system, where nitrifiers are available. The results in Fig. 9 b,c and d show that most of the COD fractions (COD total, COD soluble and COD particulate) were removed in the 1st and 2nd segment of PTF reactor.

The 3rd segment provided a little additional removal of COD fractions as shown in Figs. 9 b, c and d. This can be explained by the fact that the most of the coarse and soluble organic matter were adsorbed and degraded in the segments 1 and 2. Likely TSS and VSS concentrations were gradually decreased from segment 1 to 3 as shown in Figs. 10 a and b. The results in Figs. 11a and b show that nitrification was very limited in the 1st segment of PTF system at OLR of 4.2 kg COD/m3 .d. This was due to the presence of an insufficient ammonia oxidizer population at high loading rate as they cannot compete with heterotrophs for space and oxygen. In the 2nd and 3rd segment of PTF system, a high nitrification rate was achieved at lower OLRs of 2.1, and 1.4 kg COD/m3 .d.

These results demonstrate that at OLR exceeding 4.2 kg COD/m3 d heterotrophic bacteria still prevail in the 1st segment of PTF system, but the nitrifying bacteria promoted in the 2nd , and 3rd segment of PTF system when the OLR drops to 2.1 and 1.4 kg COD/m3.d, respectively. The ammonia oxidation and TKj-N removal (Fig. 11a and c) was virtually approximately complete, only 1.7 mgNH4-N/l and 4.0 mg TKj-N/l provided in the final effluent of PTF system.
