- Email: [email protected]
Sieving wastewater – Cellulose recovery, economic and energy evaluation
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 4 3 e4 8
Available online at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/watres
Sieving wastewater e Cellulose recovery, economic and energy evaluation C.J. Ruiken a,*, G. Breuer b, E. Klaversma a, T. Santiago a, M.C.M. van Loosdrecht b a b
Waternet, Korte Ouderkerkerdijk 7, 1096 AC Amsterdam, The Netherlands Department of Biotechnology, Delft University of Technology, Julianalaan 67, Delft, The Netherlands
article info
abstract
Article history:
Application of fine-mesh sieves (<0.35 mm) as pretreatment for municipal biological
Received 17 November 2011
wastewater treatment gives an opportunity to recover resources and increase sustain-
Received in revised form
ability of wastewater treatment processes. Sieves are traditionally used for single stage
16 August 2012
mechanical treatment (typical mesh of 0.35 mm) or in combination with an MBR (typical
Accepted 17 August 2012
mesh >0.7 mm). When sieves with a mesh of 0.35 mm are used on raw sewage we observed
Available online 13 September 2012
that cellulose fibres mainly originating from toilet paper are removed efficiently from the influent with a high recovery and purity. The application of sieves as pretreatment for
Keywords:
conventional activated sludge processes has been evaluated based on pilot plant research
Cellulose
at three WWTPs in the Netherlands. With sieving applied to the dry weather flow only the
Influent
overall energy usage of the WWTP including sludge treatment can be decreased by at least
Sieves
40% with a payback time of 7 years. ª 2012 Elsevier Ltd. All rights reserved.
Fibres Sludge production Toilet paper
1.
Introduction
Wastewater treatment by the activated sludge process has been the common technology in the past hundred years for an effective control of environmental and sanitary impact of wastewater discharge in the environment. The activated sludge process has been well studied and state-of-the-art dynamic models and wastewater characterisation methods are widely used (Henze et al., 2000) Currently there is a large focus on decreasing costs and increasing sustainability by minimising net energy usage and recovery of materials (Sutton et al., 2011; Hofman et al., 2011; Verstraete et al., 2009). It is in this light remarkable that behaviour of for instance cellulose fibres in the activated sludge process has been given hardly attention. Only a few studies related to cellulose conversion in wastewater are available (Verachtert
et al., 1982; Edberg and Hofsten, 1975). Toilet paper can be expected to be a major constituent of the wastewater; sales of toilet paper in Netherlands amount around 1 kg per person per month (http://www.worldwatch.org/node/5142). A second reference indicates similar amounts toilet paper sales (http://www.europeantissue.com/). Toilet paper essentially consists of cellulose. Cellulose needs first to be hydrolysed before it can be metabolised. Biological hydrolysis of cellulose strongly depends on temperature and sludge retention time. Unfortunately conditions and results in these studies are not easy to compare. The effects of removal or biodegradation of cellulose on the oxygen demand, sludge production, nutrient removal and dewaterability are clear knowledge gaps. At the same time cellulose fibres could be a potential resource which could be recovered easily from wastewater by for instance sieving.
* Corresponding author. Tel.: þ31 206082833. E-mail addresses: [email protected] (C.J. Ruiken), [email protected] (M.C.M. van Loosdrecht). 0043-1354/$ e see front matter ª 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.watres.2012.08.023
44
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 4 3 e4 8
Sieves with a fine mesh <0.35 mm are widely used in Norway as mechanical treatment, without further biological treatment (Rusten, 2006; Odegaard, 1998). In these cases the system is optimised for maximal removal of all suspended solids. Comparing the removal capacity of sieves and primary clarifiers indicates that a sieve can be a realistic alternative for primary treatment of wastewater. The performance with a 50e80% removal of suspended solids may be even better than a primary clarifier (Rusten, 2006). With low hydraulic loading, the sieve can operated with some cake filtration, in this cake particles smaller as the sieve mesh can be entrapped (Rusten, 2006). Also in combination with membrane bioreactors (MBR) sieves are used to optimise operation of the membrane performance. In this case a larger mesh of 0.8e2 mm, is applied (Frechen and Schier, 2008; Schier and Frechen, 2008). Interestingly we observed in initial tests that cellulose fibres form the major fraction of removed COD. The sieve material has a kind of paper mache consistency and seemed to represent the toilet paper fraction in the wastewater. Application of sieves with a mesh <0.5 mm for influent in combination with biological treatment with activated sludge is not reported in literature. Remarkably also the biodegradation of cellulose fibres in activated sludge is hardly investigated, despite being a significant fraction of the COD in wastewater. The objective of this paper was to discuss that, by removing suspended solids from the influent with fine-mesh sieves, the costs for treatment of wastewater can be reduced and at the same time the energy consumption for the treatment process (including sludge treatment and incineration) can be reduced. During 2008/2009 a pilot study for sieving of influent was executed at the WWTP Blaricum. Based on the results a case study for application of sieves at this WWTP was performed.
2.
Materials and methods
WWTP Blaricum (30,000 pe, 1600 m3/h; coordinates google maps: 52.264474, 5.250595) consist of a course bar screen 6 mm, non-aerated sand trap, selector, oxidation ditch, secondary clarifiers and sludge thickening by gravity. Phosphate is removed by simultaneous dosing of FeClSO4 in the aeration tank. The sludge treatment, digestion and dewatering is centralised at the WWTP Amsterdam West. During the test period the average influent composition of WWTP Blaricum was 441 mg COD/l, 177 mg BOD/l, 43 mg N-kj/l, and 242 mg SS/l. For the evaluation we used results of pilot tests from September 2008eJanuary 2009, with a sieve from Salsnes Filter (Salsnes Filter, PO BOX 279, N-7801 Namsos, Norway) type SF200, with a flow capacity of 35 l/s and 0.35 mesh and 0.5 m2 effective belt surface. For the cost calculation and energy balance calculation, it was assumed that mechanical dewatering of the sieving material gives a 50% solids content, which is confirmed by experiences in Norway (WWTP Tromso, oral communication). The sieving material is incinerated in a biomass energy power station for electricity production. It was further assumed that the specific biological excess sludge production in case of a sieve as primary treatment is the same as for a system with
primary clarification. This is likely an overestimation of the sludge production with sieving since a large fraction of the sievings is expected to be non-biodegradable in the activated sludge tanks. Debris (twisted hair and fibres) is no longer formed in the plant; the operational advantages (cost savings) in pumps and mechanical equipment have not been taken into account. The cellulose fibres in the sievings and sludge can be observed microscopically using polarized light. The fraction of cellulose in sievings and primary sludge was analysed using a Leco TGA-601 analyser. Sludge was dried in an oven at 105 C and grinded, 0.3e1 g of sample was transferred into the TGA analyser. The sample was first heated to 105 C and maintained at this temperature until the weight was stable in order to evaporate remaining water. Hereafter the temperature was increased with 1 C per minute until the temperature was 550 C under an air atmosphere. The mass of each sample was continuously measured. Using these results the mass decrease as a function of temperature was obtained. A graph can be constructed and the surface below the peak corresponds to the mass decrease. For degradation tests, anaerobic batch experiments were done in which cellulose fibres formed the only carbon source. Reactors with a working volume of 4 l were filled with 0.5 l activated sludge from WWTP Blaricum, dry solid concentration 4 g/l, 10 g/l toilet paper or 10 g/l sievings (obtained from the fine sieve pilot at WWTP Blaricum); 0.5 g/l NH4Cl; 12.5 ml/l buffer solution (BOOM bv; pH 7.0; 3.54 g/l KH2PO4; 14.7 g/l Na2HPO4); 0.5 ml/l nutrient solution and filled to 4 l with tap water. The reactors were continuously stirred. At the start of the experiment the pH was set to pH 7.0 by the addition of 1.4 M NaOH or 1 M HCl solution and continuous measuring the pH. Every 48e72 h the pH was measured and adjusted to pH 6.8e7.0 by manual titration of a NaOH solution. The drop in pH in this period was usually 1 pH unit. The dry solid concentration from samples was determined as function of time. At the start of the experiment and after each sampling the reactor was flushed with N2 gas for several minutes in order to maintain anaerobic conditions. The fermentations were stopped when the dry solid concentration was constant and when the pH value didn’t drop significantly anymore.
3.
Results and discussion
3.1.
Toilet paper (cellulose) in wastewater
A continuously operated pilot plant sieve was operated during 5 months in 2008/2009 on the raw influent of Blaricum WWTP after the coarse bar screen of 6 mm and before the sand trap. The constant flow rate of 200 m3/m2 h was at the higher end of the operational window of the sieve. The average influentbased removal efficiencies over the sieve were approx. 50% for suspended solids (range 10e75%), 35% for COD (range 10e60%) and approx. 1% for nitrogen and <1% for phosphorus. Based on the pilot research it was found that a sieve <0.35 mm removes most of the cellulose fibres. Fig. 1a shows a microscopic picture of the sediment from the raw influent and in Fig. 1b from the sediment of sieved influent. Most of the fibres are removed in sieved influent, this becomes clear with visual
45
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 4 3 e4 8
Fig. 1 e Microscopic observation of fibres in settled material from raw influent (a) and fibres of settled material from sieved influent (b) sieve mesh 0.35 mm; sieve rate 30 m3/(m2 h).
microscopic observation (Fig. 1). However no measurements have been done since a good measurement for cellulose in wastewater or activated sludge is still missing. Therefore, the removal efficiency for cellulose is not known yet. However, the removal efficiency will be considerable. The reason for a high fibre recovery on the sieve is that the length of cellulose fibres in toilet paper is in the range of 1e1.2 mm on average. Microscopic observation showed this length is not changed during transport in the sewer network. The fibre length of cellulose fibres in sludge was similar as the fibre length in different types of toilet paper obtained directly from a shop. With a mesh size <0.35 mm most fibres will therefore be retained under low hydraulic load conditions, whereas most of the remaining particulate fraction in the influent passes the sieve. To estimate the fraction of cellulose in sievings and primary sludge thermographic analysis was used. This analysis gives the weight loss of a sample upon heating. Cellulose is oxidized in a small temperature (Fig. 2a) range around 300 C and can thereby be distinguished from other organic components. The same peak in the thermographic curve is obtained for microcrystalline cellulose, filtrate paper or toilet paper. For mixtures of organic components, a broad spectrum is obtained (Fig. 2b), with two peaks of which one represents a cellulose peak. In raw influent it appeared not possible to measure cellulose content this way because no sharp peak is
found most likely because too many other organic components are present. To evaluate the effect of the presence of different solids we observed the effect of addition of butter to resemble fat to toilet paper. The peak of toilet paper occurs at the same temperature as for cellulose. Addition of butter moves the peak to the left just like can be seen in Fig. 2a with the sieving material. So most likely the thermal oxidation of cellulose is occurring at lower temperatures due to the presence of other organic components. This explains the difference between pure cellulose and cellulose in influent, but makes the method not well suited to quantify the actual cellulose content in a sludge sample. Based on thermographic measurements the fraction cellulose found in sievings is 79% of the total mass and 84% of the organic mass. The fraction inorganic matter was 6%. The fine sieve had a removal efficiency for suspended solids of around 40% during the period in which the sievings were obtained. This means that about 35% of the suspended solids in the influent originated from toilet paper. For WWTP Blaricum this would be a consumption of about 10 kg toilet paper per person a year. The cellulose nature (and absence of proteins) is also indicated by the very low removal of kjeldahl nitrogen and phosphate by the sieving. The primary sludge represents 32% of the organic mass in the influent of WWTP ‘Horstermeer’. Similar results were obtained with primary sludge from the WWTP ‘Ronde Venen’.
7
7
a
microcrystalline cellulose sievings pilot Blaricum
5
4
3
2
1
primary sludge Horstermeer primary sludge Horstermeer sievings added
6
d(relative decrease)/dt(%/min)
d(relative decrease)/dt(%/min)
6
toilet paper
5
4
3
2
1
0 150
b
filter paper
0
200
250
300
350
Temperature (oC)
400
450
500
150
200
250
300
350
400
450
500
Temperature (oC)
Fig. 2 e Thermographic spectra of microcrystalline cellulose, cellulose-based filter paper and sievings (a) and primary sludge from WWTP ‘Horstermeer’, primary sludge from WWTP ‘Horstermeer’ with sieving material added and toilet paper (b).
46
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 4 3 e4 8
With primary sludge a large peak in the thermograph at 225 C was found (Fig. 2b), the origin of this peak is unknown. This peak is absent in the sievings. For primary sludge from ‘Ronde Venen’ and ‘Horstermeer’ a peak related to cellulose is found between 250 and 280 C. This results in a fraction of cellulose of 25% and 32% of the total mass and a fraction of cellulose of 32% and 38% of the organic mass for ‘Ronde Venen’ and ‘Horstermeer’. Primary sludge clearly has a significant fraction of cellulose, but this fraction is much lower then for sievings. The removal efficiency of suspended solids for fine sieves and pre settlement tank (a removal efficiency of around 50% SS) is comparable. Therefore, the effluent from a pre-settlement tank will contain much more cellulose than the effluent from a fine sieve. It was assumed that all mass decrease around the thermographic peak for cellulose was caused by oxidation of cellulose. Other organic components could potentially also be oxidized in this range. In that case the assumed fraction of cellulose would be too high. This effect is expected to be larger in samples with a low fraction of cellulose (e.g. primary sludge compared with sievings). The observed fraction of cellulose (especially in case of primary sludge) can therefore best be seen as an upper limit of the fraction of cellulose instead of an absolute concentration. Knowing that the removal efficiency of the pre-settlement tank is 50% for suspended solids and <30% of the primary sludge mass is cellulose then <50% of the cellulose is removed with a pre-settlement tank. Clear is that the composition of the wastewater after a sieve or primary settler is significant different from the effluent of primary clarification. To compare the energy balance and sludge production after a sieve with a primary settlement tank, this apparent difference is ignored for this study. Assumed in the calculation is that effects of the remaining COD on the biology between sieves and primary settlers will be the same. However most likely this is not correct, cellulose fibres are present in large amounts in activated sludge (by microscopic observation) indicating their low biodegradability. Unfortunately the thermographic analysis is not feasible on activated sludge. Likely, cellulose fibres make up a significant part of the inert solids fraction in the wastewater. In aerobic conditions only 60% of cellulose is degraded in activated sludge in 4e5 weeks, whereas from the cellulose in excess sludge only 50% is converted in anaerobic digesters (Verachtert et al., 1982). There is a fraction of very slowly biodegradable COD in the influent (Nowak et al., 1999). The percentage degradation depends strongly on the solids retention time. We suggest this very slowly biodegradable fraction is related to the cellulose compounds, although more in-depth investigations are needed. The better removal of cellulose by sieving might therefore result in a lower specific sludge production on the remaining COD as compared to primary clarification.
3.2.
Conversion of toilet paper in activated sludge
The behaviour of toilet paper present in wastewater has not been a popular topic, and it is neglected in wastewater characterisation studies (Roeleveld and Van Loosdrecht, 2002). This is quite strange, because toilet paper is a considerable fraction of the COD in influent.
The amount of toilet paper sales per person (Table 1) varies from country to country (Matters of Scale e Into the Toilet (http://www.worldwatch.org/node/5142)). A second reference indicates similar amounts of toilet paper sales (http://www. europeantissue.com/). A paper producing company (SCA) gives a consumption in Germany of about 11 kg toilet paper per person, which is believed to be about the same in the Netherlands. The data forming the different sources show some variability (10e20%), but all indicate a significant amount of toilet paper sales. The majority of these sales can be assumed to end up in the municipal wastewater. For interpretation of consequences of application of sieves and the role of cellulose in activated sludge processes significant differences between countries have to be taken into account. Based on the thermographic measurement of material from the pilot sieves at the WWTP Blaricum about 10 kg toilet paper per person a year was found. This is less then the presumed sales of toilet paper in the Netherlands. However, the sieve recovery is not 100% and likely a small fraction of toilet paper might be ending up in the solid waste of a household. Overall it seems that the available information is consistent and gives a good approximation of the cellulose load to a wastewater treatment plant. An amount of 10e13 kg per year of cellulose would represent a considerable fraction of the suspended solids in the wastewater. As an example the situation at Waternet in Amsterdam: Approximately. 1,200,000 persons are connected to the WWTPs of Waternet. They discharge 12,000e15,000 ton toilet paper per year. The total suspended solids mass measured in influent is 32,000 ton per year. This indicates that roughly 40% of the influent suspended solids could be toilet paper derived cellulose. Expressed in COD the amount of toilet paper represents 17,000e21,000 ton COD, this would represent 25e30% of the total COD influent load. Clearly a separate evaluation of the cellulose in wastewater would be needed in order to improve our understanding of the processes occurring in activated sludge processes. To obtain a preliminary indication of biodegradability of cellulose fibres in activated sludge systems batch tests with anaerobic degradation of cellulose have been executed. In Table 2 the main results are reported. Cellulose degradation is observed to be indeed a slow process confirming the observations (Verachtert et al., 1982). From these initial tests it is clear that temperature has a very large impact on cellulose biodegradation. For example, at 9 C only 10% is degraded anaerobic after 20 days whereas at 24 C complete removal is within 12 days anaerobic. This means that under winter conditions there will be less degradation of cellulose and more sludge production (Verachtert et al., 1982). There appears to be a difference in degradability
Table 1 e Per capita annual consumption of toilet tissue (http://www.worldwatch.org/node/5142). North America Western Europe Latin America Asia Africa
23.0 kg 13.8 kg 4.2 kg 1.8 kg 0.4 kg
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 4 3 e4 8
Table 2 e Biodegradation of toilet paper in batch biodegradation tests inoculated with sludge from WWTP Blaricum. Condition pH C Anaerobic Anaerobic Anaerobic Anaerobic Anaerobic Anaerobic Anaerobic Anaerobic
Type fibre
7.0 30 Toilet paper from recycled paper 7.0 24 Toilet paper from recycled paper 7.0 9 Toilet paper from recycled paper 7.0 24 4 layer type 1 toilet paper 7.0 24 4 layer type 2 toilet paper 7.0 24 Sieving material 50% SS removal 7.0 17 Sieving material 50% SS removal 5.0 24 Toilet paper from recycled paper
Removal% Days
separated materials to produce safely toilet paper, a real cradleto-cradle application, but difficulties may be encountered in relation to social acceptance. For the wastewater treatment at Waternet a total maximum yearly production of 13,500 ton dry solids for the whole region is estimated. Recovery of the cellulose fibres can therefore contribute in recovery and reuse of materials from the wastewater plant.
100
8
100
12
10
20
3.4.
100 100 100
15 20 20
50
20
40
20
Based on our operational experience of a pilot plant sieve on the Blaricum wastewater a cost evaluation has been made to evaluate the economic payback time for sieving of wastewater. Important to realise is, that the local situation for Waternet is described. At Waternet sludge from 10 WWPTs after thickening to 2.5e5% dry solids, is transported to a centralised digestion and dewatering at the WWTP Amsterdam. Local dewatering at each WWTP is not economic feasible at the situation of Waternet. For the cost calculations we used the observed average removal of 50% suspended solids and 35% of total COD. It could well be that with a more extensive study the removal could be further optimised. A clear description of the degradation of cellulose in activated sludge is not available yet. Sludge production on sieved wastewater is therefore assumed the same as for a clarifier primary effluent. Nutrient removal is currently assumed not to be negatively influenced. The BOD/N ratio of sieved effluent is still good (4 g BOD/g N), while cellulose is likely for a large fraction normally incorporated in the inert suspended COD fraction of wastewater. Seen to the abundant presence of cellulose fibres even after sludge digestion, surplus sludge production on sieved wastewater is most likely overestimated in this evaluation. The sieving material can with mechanical pressure be concentrated to 40e50% dry solids, compared to 2.8% for surplus sludge after thickening. This means that transportation costs are greatly reduced compared to waste sludge transport. The actual processing costs for sieved material is maximal 35 V per ton. In this case the sieved material is used for compost production. For excess sludge the costs are currently 450 V per ton dry solids, including dewatering and incineration. For the local situation of WWTP Blaricum with one sieve for 400e500 m3/h (dry weather flow) the total investment is V 800,000 including taxes (real costs because installation was taken into operation Januari 2011). Sieving gives a lower load (compared to the current situation without any pretreament) and therefore a lower aeration
for different types of toilet paper. Clearly more effort is needed to understand the conversion processes of cellulose, the kinetics and influences such as temperature, oxygen, nitrate and pH. Mechanism for degradation may be complex because in essence it is a biofilm process with a lag-time for microbial colonisation of the fibre. The lag-time was 1e2 days in the experiments in Table 2. In Fig. 3a the biofilm surrounding the cellulose fibre in activated sludge can be seen. Interesting aspect is also that in mesophilic digesters with 20 days retention time, only partly degradation (a good method to measure cellulose concentration in digested sludge is not available yet) of cellulose fibres is observed (Fig. 3b). The significant presence of cellulose in activated sludge and digested sludge calls for more in-depth studies on the conversion processes of these fibres.
3.3.
47
Processing the sieving material
Various options are available for the treatment and reuse of the sieving material: The heavy metal content of separated materials is low and complies with the decree for application in agriculture as soil conditioner. The separated material has already a high solids content and dries easily, it could therefore be easily used as fuel in a biomass based power plant. The cellulose could be used as feed stock in the fermentation industry for production of for instance biofuels like ethanol. Technically it is possible to use the
Economic evaluation
Fig. 3 e Microscopic pictures of cellulose fibre from in-situ biodegradation experiments with a clear biofilm around the fibre (a), cellulose fibres (and also struvite crystals) abundantly present in digested sludge from WWTP Amsterdam West (b).
48
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 4 3 e4 8
demand. For electricity costs V 0.12/kWh was used, a detailed overview of the cost calculation can be found in the technical Stowa report (Stowa 2010-19). When sieves are applied only for the dry weather flow at the existing low loaded (0.12 g COD/g MLSS day) WWTP Blaricum (which has no primary clarifier), including all investments (sieves, pipes, automation etc.) and operational costs, a payback time of 7 years was estimated. The life time expectancy of the sieve is 15 years. When the entire rain weather flow (RWF) is treated, the break-even point was more than 15 years. Yearly savings were estimated to be V 125,000. Savings related to sludge treatment are most important, since sieving leads to a lower load to the treatment plant. In addition, the use of a fine-mesh sieve may also yield operational benefits for example in the form of reduced debris formation (the intermeshing of hair and fibres). This was not taken into account in the cost comparison. The used sieve did not face any problems with fouling of the belt. However removal of fat and operational attention every one/two weeks takes about 30 min. Clearly an evaluation has to be made in each individual case but it seems that sieving is an economic viable option to improve wastewater treatment plants. Certainly when expansion of wastewater treatment plants are considered the benefits might be larger then here reported. Cellulose forms a significant part of the sludge and removal of this inert material increases the specific activity of activated sludge and thereby the treatment capacity.
3.5.
Energy evaluation
The effects on the energy balance (consumptioneproduction) for the activated sludge process combined with digestion, biogas production, transport of sludge, dewatering, and incineration, were calculated and compared to the reference situation without sieves. The underlying assumption is that the sieve product will be incinerated in a biomass plant. Moreover, separated materials can be dewatered more effectively than sludge, with the result that less transport is required and the caloric value is higher. The energy balance showed clear positive results in the case of the WWTP Blaricum. It appears that fine-mesh sieves represent an alternative to settlement tanks for energy-related considerations. Net energy need (wastewater treatment, sludge treatment and incineration), would amount to at least 40% less compared with the reference without fine-mesh sieving (Stowa 2010-19).
4.
Conclusion
It is remarkably that until now a major fraction of wastewater (cellulose fibres from toilet paper) have hardly been studied.
Preliminary evaluations indicate they might form a major part of the so-called inert COD in the wastewater. Sieves can be used to remove this fraction. The resulting filtered material (approximately. 30% of the COD for Dutch wastewater) can efficiently be used in a biomass power plant for energy generation whereas the resulting sieved wastewater gives advantages in optimizing the wastewater treatment process. This leads to an overall energy optimization and more cost efficient wastewater system. Future evaluation of use of sieves at a full scale application will further refine the evaluation of the potential of cellulose recovery by sieving.
references
Edberg, N., Hofsten, B., 1975. Cellulose degradation in wastewater treatment. Journal of the Water Pollution Control Federation 47 (5), 1011e1020. Frechen, F.B., Schier, W., 2008. Mechanische Abwasser Vorbehandlung auf Kommunalen Membranbelebungsanlagen. Korrespondenz Abwasser 55 (1), 39e44. Henze, M., Gujer, W., Mino, T., van Loosdrecht, M.C.M., 2000. Activated sludge models ASM1, ASM2, ASM2d, and ASM3. IWA scientific and technical report no. 9. IWA Publishing, London, UK. Hofman, J., Hofman-Caris, R., Nederlof, M., Frijns, J., Van Loosdrecht, M., 2011. Water and energy as inseparable twins for sustainable solutions. Water Science and Technology 63 (1), 88e92. Nowak, O., Svardal, K., Franz, A., Kuhn, V., 1999. Degradation of particulate organic matter e a comparison of different model concepts. Water Science and Technology 39 (1), 119e127. Odegaard, H., 1998. Optimised particle separation in the primary step of wastewater treatment. Water Science and Technology 37 (10), 43e53. Roeleveld, P.J., Van Loosdrecht, M.C.M., 2002. Experience with guidelines for wastewater characterisation in The Netherlands. Water Science and Technology 45 (6), 77e87. Rusten, B., 2006. Evaluation and testing of fine mesh sieve technologies for primary treatment of municipal wastewater. Water Science and Technology 54 (10), 31e38. Schier, W., Frechen, F.B., 2008. Efficiency of mechanical pretreatment on European MBR plants, In: International Conference Amsterdam Rai Netherlands, 1st and 2nd October. Sutton, P.M., Melcer, H., Schraa, O.J., Togna, A.P., 2011. Treating municipal wastewater with the goal of resource recovery. Water Science and Technology 63 (1), 25e31. STOWA 2010-19. Influent fijnzeven in rwzi’s, www.stowa.nl. Verachtert, H., Ramasamy, K., Meyers, M., Bever, J., 1982. Investigations on cellulose degradation in activated sludge plants. Journal of Applied Bacteriology 52, 185e190. Verstraete, W., Van de Caveye, P., Diamantis, V., 2009. Maximum use of resources present in domestic “used water”. Bioresource Technology 100 (23), 5537e5545.
Available online at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/watres
Sieving wastewater e Cellulose recovery, economic and energy evaluation C.J. Ruiken a,*, G. Breuer b, E. Klaversma a, T. Santiago a, M.C.M. van Loosdrecht b a b
Waternet, Korte Ouderkerkerdijk 7, 1096 AC Amsterdam, The Netherlands Department of Biotechnology, Delft University of Technology, Julianalaan 67, Delft, The Netherlands
article info
abstract
Article history:
Application of fine-mesh sieves (<0.35 mm) as pretreatment for municipal biological
Received 17 November 2011
wastewater treatment gives an opportunity to recover resources and increase sustain-
Received in revised form
ability of wastewater treatment processes. Sieves are traditionally used for single stage
16 August 2012
mechanical treatment (typical mesh of 0.35 mm) or in combination with an MBR (typical
Accepted 17 August 2012
mesh >0.7 mm). When sieves with a mesh of 0.35 mm are used on raw sewage we observed
Available online 13 September 2012
that cellulose fibres mainly originating from toilet paper are removed efficiently from the influent with a high recovery and purity. The application of sieves as pretreatment for
Keywords:
conventional activated sludge processes has been evaluated based on pilot plant research
Cellulose
at three WWTPs in the Netherlands. With sieving applied to the dry weather flow only the
Influent
overall energy usage of the WWTP including sludge treatment can be decreased by at least
Sieves
40% with a payback time of 7 years. ª 2012 Elsevier Ltd. All rights reserved.
Fibres Sludge production Toilet paper
1.
Introduction
Wastewater treatment by the activated sludge process has been the common technology in the past hundred years for an effective control of environmental and sanitary impact of wastewater discharge in the environment. The activated sludge process has been well studied and state-of-the-art dynamic models and wastewater characterisation methods are widely used (Henze et al., 2000) Currently there is a large focus on decreasing costs and increasing sustainability by minimising net energy usage and recovery of materials (Sutton et al., 2011; Hofman et al., 2011; Verstraete et al., 2009). It is in this light remarkable that behaviour of for instance cellulose fibres in the activated sludge process has been given hardly attention. Only a few studies related to cellulose conversion in wastewater are available (Verachtert
et al., 1982; Edberg and Hofsten, 1975). Toilet paper can be expected to be a major constituent of the wastewater; sales of toilet paper in Netherlands amount around 1 kg per person per month (http://www.worldwatch.org/node/5142). A second reference indicates similar amounts toilet paper sales (http://www.europeantissue.com/). Toilet paper essentially consists of cellulose. Cellulose needs first to be hydrolysed before it can be metabolised. Biological hydrolysis of cellulose strongly depends on temperature and sludge retention time. Unfortunately conditions and results in these studies are not easy to compare. The effects of removal or biodegradation of cellulose on the oxygen demand, sludge production, nutrient removal and dewaterability are clear knowledge gaps. At the same time cellulose fibres could be a potential resource which could be recovered easily from wastewater by for instance sieving.
* Corresponding author. Tel.: þ31 206082833. E-mail addresses: [email protected] (C.J. Ruiken), [email protected] (M.C.M. van Loosdrecht). 0043-1354/$ e see front matter ª 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.watres.2012.08.023
44
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 4 3 e4 8
Sieves with a fine mesh <0.35 mm are widely used in Norway as mechanical treatment, without further biological treatment (Rusten, 2006; Odegaard, 1998). In these cases the system is optimised for maximal removal of all suspended solids. Comparing the removal capacity of sieves and primary clarifiers indicates that a sieve can be a realistic alternative for primary treatment of wastewater. The performance with a 50e80% removal of suspended solids may be even better than a primary clarifier (Rusten, 2006). With low hydraulic loading, the sieve can operated with some cake filtration, in this cake particles smaller as the sieve mesh can be entrapped (Rusten, 2006). Also in combination with membrane bioreactors (MBR) sieves are used to optimise operation of the membrane performance. In this case a larger mesh of 0.8e2 mm, is applied (Frechen and Schier, 2008; Schier and Frechen, 2008). Interestingly we observed in initial tests that cellulose fibres form the major fraction of removed COD. The sieve material has a kind of paper mache consistency and seemed to represent the toilet paper fraction in the wastewater. Application of sieves with a mesh <0.5 mm for influent in combination with biological treatment with activated sludge is not reported in literature. Remarkably also the biodegradation of cellulose fibres in activated sludge is hardly investigated, despite being a significant fraction of the COD in wastewater. The objective of this paper was to discuss that, by removing suspended solids from the influent with fine-mesh sieves, the costs for treatment of wastewater can be reduced and at the same time the energy consumption for the treatment process (including sludge treatment and incineration) can be reduced. During 2008/2009 a pilot study for sieving of influent was executed at the WWTP Blaricum. Based on the results a case study for application of sieves at this WWTP was performed.
2.
Materials and methods
WWTP Blaricum (30,000 pe, 1600 m3/h; coordinates google maps: 52.264474, 5.250595) consist of a course bar screen 6 mm, non-aerated sand trap, selector, oxidation ditch, secondary clarifiers and sludge thickening by gravity. Phosphate is removed by simultaneous dosing of FeClSO4 in the aeration tank. The sludge treatment, digestion and dewatering is centralised at the WWTP Amsterdam West. During the test period the average influent composition of WWTP Blaricum was 441 mg COD/l, 177 mg BOD/l, 43 mg N-kj/l, and 242 mg SS/l. For the evaluation we used results of pilot tests from September 2008eJanuary 2009, with a sieve from Salsnes Filter (Salsnes Filter, PO BOX 279, N-7801 Namsos, Norway) type SF200, with a flow capacity of 35 l/s and 0.35 mesh and 0.5 m2 effective belt surface. For the cost calculation and energy balance calculation, it was assumed that mechanical dewatering of the sieving material gives a 50% solids content, which is confirmed by experiences in Norway (WWTP Tromso, oral communication). The sieving material is incinerated in a biomass energy power station for electricity production. It was further assumed that the specific biological excess sludge production in case of a sieve as primary treatment is the same as for a system with
primary clarification. This is likely an overestimation of the sludge production with sieving since a large fraction of the sievings is expected to be non-biodegradable in the activated sludge tanks. Debris (twisted hair and fibres) is no longer formed in the plant; the operational advantages (cost savings) in pumps and mechanical equipment have not been taken into account. The cellulose fibres in the sievings and sludge can be observed microscopically using polarized light. The fraction of cellulose in sievings and primary sludge was analysed using a Leco TGA-601 analyser. Sludge was dried in an oven at 105 C and grinded, 0.3e1 g of sample was transferred into the TGA analyser. The sample was first heated to 105 C and maintained at this temperature until the weight was stable in order to evaporate remaining water. Hereafter the temperature was increased with 1 C per minute until the temperature was 550 C under an air atmosphere. The mass of each sample was continuously measured. Using these results the mass decrease as a function of temperature was obtained. A graph can be constructed and the surface below the peak corresponds to the mass decrease. For degradation tests, anaerobic batch experiments were done in which cellulose fibres formed the only carbon source. Reactors with a working volume of 4 l were filled with 0.5 l activated sludge from WWTP Blaricum, dry solid concentration 4 g/l, 10 g/l toilet paper or 10 g/l sievings (obtained from the fine sieve pilot at WWTP Blaricum); 0.5 g/l NH4Cl; 12.5 ml/l buffer solution (BOOM bv; pH 7.0; 3.54 g/l KH2PO4; 14.7 g/l Na2HPO4); 0.5 ml/l nutrient solution and filled to 4 l with tap water. The reactors were continuously stirred. At the start of the experiment the pH was set to pH 7.0 by the addition of 1.4 M NaOH or 1 M HCl solution and continuous measuring the pH. Every 48e72 h the pH was measured and adjusted to pH 6.8e7.0 by manual titration of a NaOH solution. The drop in pH in this period was usually 1 pH unit. The dry solid concentration from samples was determined as function of time. At the start of the experiment and after each sampling the reactor was flushed with N2 gas for several minutes in order to maintain anaerobic conditions. The fermentations were stopped when the dry solid concentration was constant and when the pH value didn’t drop significantly anymore.
3.
Results and discussion
3.1.
Toilet paper (cellulose) in wastewater
A continuously operated pilot plant sieve was operated during 5 months in 2008/2009 on the raw influent of Blaricum WWTP after the coarse bar screen of 6 mm and before the sand trap. The constant flow rate of 200 m3/m2 h was at the higher end of the operational window of the sieve. The average influentbased removal efficiencies over the sieve were approx. 50% for suspended solids (range 10e75%), 35% for COD (range 10e60%) and approx. 1% for nitrogen and <1% for phosphorus. Based on the pilot research it was found that a sieve <0.35 mm removes most of the cellulose fibres. Fig. 1a shows a microscopic picture of the sediment from the raw influent and in Fig. 1b from the sediment of sieved influent. Most of the fibres are removed in sieved influent, this becomes clear with visual
45
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 4 3 e4 8
Fig. 1 e Microscopic observation of fibres in settled material from raw influent (a) and fibres of settled material from sieved influent (b) sieve mesh 0.35 mm; sieve rate 30 m3/(m2 h).
microscopic observation (Fig. 1). However no measurements have been done since a good measurement for cellulose in wastewater or activated sludge is still missing. Therefore, the removal efficiency for cellulose is not known yet. However, the removal efficiency will be considerable. The reason for a high fibre recovery on the sieve is that the length of cellulose fibres in toilet paper is in the range of 1e1.2 mm on average. Microscopic observation showed this length is not changed during transport in the sewer network. The fibre length of cellulose fibres in sludge was similar as the fibre length in different types of toilet paper obtained directly from a shop. With a mesh size <0.35 mm most fibres will therefore be retained under low hydraulic load conditions, whereas most of the remaining particulate fraction in the influent passes the sieve. To estimate the fraction of cellulose in sievings and primary sludge thermographic analysis was used. This analysis gives the weight loss of a sample upon heating. Cellulose is oxidized in a small temperature (Fig. 2a) range around 300 C and can thereby be distinguished from other organic components. The same peak in the thermographic curve is obtained for microcrystalline cellulose, filtrate paper or toilet paper. For mixtures of organic components, a broad spectrum is obtained (Fig. 2b), with two peaks of which one represents a cellulose peak. In raw influent it appeared not possible to measure cellulose content this way because no sharp peak is
found most likely because too many other organic components are present. To evaluate the effect of the presence of different solids we observed the effect of addition of butter to resemble fat to toilet paper. The peak of toilet paper occurs at the same temperature as for cellulose. Addition of butter moves the peak to the left just like can be seen in Fig. 2a with the sieving material. So most likely the thermal oxidation of cellulose is occurring at lower temperatures due to the presence of other organic components. This explains the difference between pure cellulose and cellulose in influent, but makes the method not well suited to quantify the actual cellulose content in a sludge sample. Based on thermographic measurements the fraction cellulose found in sievings is 79% of the total mass and 84% of the organic mass. The fraction inorganic matter was 6%. The fine sieve had a removal efficiency for suspended solids of around 40% during the period in which the sievings were obtained. This means that about 35% of the suspended solids in the influent originated from toilet paper. For WWTP Blaricum this would be a consumption of about 10 kg toilet paper per person a year. The cellulose nature (and absence of proteins) is also indicated by the very low removal of kjeldahl nitrogen and phosphate by the sieving. The primary sludge represents 32% of the organic mass in the influent of WWTP ‘Horstermeer’. Similar results were obtained with primary sludge from the WWTP ‘Ronde Venen’.
7
7
a
microcrystalline cellulose sievings pilot Blaricum
5
4
3
2
1
primary sludge Horstermeer primary sludge Horstermeer sievings added
6
d(relative decrease)/dt(%/min)
d(relative decrease)/dt(%/min)
6
toilet paper
5
4
3
2
1
0 150
b
filter paper
0
200
250
300
350
Temperature (oC)
400
450
500
150
200
250
300
350
400
450
500
Temperature (oC)
Fig. 2 e Thermographic spectra of microcrystalline cellulose, cellulose-based filter paper and sievings (a) and primary sludge from WWTP ‘Horstermeer’, primary sludge from WWTP ‘Horstermeer’ with sieving material added and toilet paper (b).
46
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 4 3 e4 8
With primary sludge a large peak in the thermograph at 225 C was found (Fig. 2b), the origin of this peak is unknown. This peak is absent in the sievings. For primary sludge from ‘Ronde Venen’ and ‘Horstermeer’ a peak related to cellulose is found between 250 and 280 C. This results in a fraction of cellulose of 25% and 32% of the total mass and a fraction of cellulose of 32% and 38% of the organic mass for ‘Ronde Venen’ and ‘Horstermeer’. Primary sludge clearly has a significant fraction of cellulose, but this fraction is much lower then for sievings. The removal efficiency of suspended solids for fine sieves and pre settlement tank (a removal efficiency of around 50% SS) is comparable. Therefore, the effluent from a pre-settlement tank will contain much more cellulose than the effluent from a fine sieve. It was assumed that all mass decrease around the thermographic peak for cellulose was caused by oxidation of cellulose. Other organic components could potentially also be oxidized in this range. In that case the assumed fraction of cellulose would be too high. This effect is expected to be larger in samples with a low fraction of cellulose (e.g. primary sludge compared with sievings). The observed fraction of cellulose (especially in case of primary sludge) can therefore best be seen as an upper limit of the fraction of cellulose instead of an absolute concentration. Knowing that the removal efficiency of the pre-settlement tank is 50% for suspended solids and <30% of the primary sludge mass is cellulose then <50% of the cellulose is removed with a pre-settlement tank. Clear is that the composition of the wastewater after a sieve or primary settler is significant different from the effluent of primary clarification. To compare the energy balance and sludge production after a sieve with a primary settlement tank, this apparent difference is ignored for this study. Assumed in the calculation is that effects of the remaining COD on the biology between sieves and primary settlers will be the same. However most likely this is not correct, cellulose fibres are present in large amounts in activated sludge (by microscopic observation) indicating their low biodegradability. Unfortunately the thermographic analysis is not feasible on activated sludge. Likely, cellulose fibres make up a significant part of the inert solids fraction in the wastewater. In aerobic conditions only 60% of cellulose is degraded in activated sludge in 4e5 weeks, whereas from the cellulose in excess sludge only 50% is converted in anaerobic digesters (Verachtert et al., 1982). There is a fraction of very slowly biodegradable COD in the influent (Nowak et al., 1999). The percentage degradation depends strongly on the solids retention time. We suggest this very slowly biodegradable fraction is related to the cellulose compounds, although more in-depth investigations are needed. The better removal of cellulose by sieving might therefore result in a lower specific sludge production on the remaining COD as compared to primary clarification.
3.2.
Conversion of toilet paper in activated sludge
The behaviour of toilet paper present in wastewater has not been a popular topic, and it is neglected in wastewater characterisation studies (Roeleveld and Van Loosdrecht, 2002). This is quite strange, because toilet paper is a considerable fraction of the COD in influent.
The amount of toilet paper sales per person (Table 1) varies from country to country (Matters of Scale e Into the Toilet (http://www.worldwatch.org/node/5142)). A second reference indicates similar amounts of toilet paper sales (http://www. europeantissue.com/). A paper producing company (SCA) gives a consumption in Germany of about 11 kg toilet paper per person, which is believed to be about the same in the Netherlands. The data forming the different sources show some variability (10e20%), but all indicate a significant amount of toilet paper sales. The majority of these sales can be assumed to end up in the municipal wastewater. For interpretation of consequences of application of sieves and the role of cellulose in activated sludge processes significant differences between countries have to be taken into account. Based on the thermographic measurement of material from the pilot sieves at the WWTP Blaricum about 10 kg toilet paper per person a year was found. This is less then the presumed sales of toilet paper in the Netherlands. However, the sieve recovery is not 100% and likely a small fraction of toilet paper might be ending up in the solid waste of a household. Overall it seems that the available information is consistent and gives a good approximation of the cellulose load to a wastewater treatment plant. An amount of 10e13 kg per year of cellulose would represent a considerable fraction of the suspended solids in the wastewater. As an example the situation at Waternet in Amsterdam: Approximately. 1,200,000 persons are connected to the WWTPs of Waternet. They discharge 12,000e15,000 ton toilet paper per year. The total suspended solids mass measured in influent is 32,000 ton per year. This indicates that roughly 40% of the influent suspended solids could be toilet paper derived cellulose. Expressed in COD the amount of toilet paper represents 17,000e21,000 ton COD, this would represent 25e30% of the total COD influent load. Clearly a separate evaluation of the cellulose in wastewater would be needed in order to improve our understanding of the processes occurring in activated sludge processes. To obtain a preliminary indication of biodegradability of cellulose fibres in activated sludge systems batch tests with anaerobic degradation of cellulose have been executed. In Table 2 the main results are reported. Cellulose degradation is observed to be indeed a slow process confirming the observations (Verachtert et al., 1982). From these initial tests it is clear that temperature has a very large impact on cellulose biodegradation. For example, at 9 C only 10% is degraded anaerobic after 20 days whereas at 24 C complete removal is within 12 days anaerobic. This means that under winter conditions there will be less degradation of cellulose and more sludge production (Verachtert et al., 1982). There appears to be a difference in degradability
Table 1 e Per capita annual consumption of toilet tissue (http://www.worldwatch.org/node/5142). North America Western Europe Latin America Asia Africa
23.0 kg 13.8 kg 4.2 kg 1.8 kg 0.4 kg
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 4 3 e4 8
Table 2 e Biodegradation of toilet paper in batch biodegradation tests inoculated with sludge from WWTP Blaricum. Condition pH C Anaerobic Anaerobic Anaerobic Anaerobic Anaerobic Anaerobic Anaerobic Anaerobic
Type fibre
7.0 30 Toilet paper from recycled paper 7.0 24 Toilet paper from recycled paper 7.0 9 Toilet paper from recycled paper 7.0 24 4 layer type 1 toilet paper 7.0 24 4 layer type 2 toilet paper 7.0 24 Sieving material 50% SS removal 7.0 17 Sieving material 50% SS removal 5.0 24 Toilet paper from recycled paper
Removal% Days
separated materials to produce safely toilet paper, a real cradleto-cradle application, but difficulties may be encountered in relation to social acceptance. For the wastewater treatment at Waternet a total maximum yearly production of 13,500 ton dry solids for the whole region is estimated. Recovery of the cellulose fibres can therefore contribute in recovery and reuse of materials from the wastewater plant.
100
8
100
12
10
20
3.4.
100 100 100
15 20 20
50
20
40
20
Based on our operational experience of a pilot plant sieve on the Blaricum wastewater a cost evaluation has been made to evaluate the economic payback time for sieving of wastewater. Important to realise is, that the local situation for Waternet is described. At Waternet sludge from 10 WWPTs after thickening to 2.5e5% dry solids, is transported to a centralised digestion and dewatering at the WWTP Amsterdam. Local dewatering at each WWTP is not economic feasible at the situation of Waternet. For the cost calculations we used the observed average removal of 50% suspended solids and 35% of total COD. It could well be that with a more extensive study the removal could be further optimised. A clear description of the degradation of cellulose in activated sludge is not available yet. Sludge production on sieved wastewater is therefore assumed the same as for a clarifier primary effluent. Nutrient removal is currently assumed not to be negatively influenced. The BOD/N ratio of sieved effluent is still good (4 g BOD/g N), while cellulose is likely for a large fraction normally incorporated in the inert suspended COD fraction of wastewater. Seen to the abundant presence of cellulose fibres even after sludge digestion, surplus sludge production on sieved wastewater is most likely overestimated in this evaluation. The sieving material can with mechanical pressure be concentrated to 40e50% dry solids, compared to 2.8% for surplus sludge after thickening. This means that transportation costs are greatly reduced compared to waste sludge transport. The actual processing costs for sieved material is maximal 35 V per ton. In this case the sieved material is used for compost production. For excess sludge the costs are currently 450 V per ton dry solids, including dewatering and incineration. For the local situation of WWTP Blaricum with one sieve for 400e500 m3/h (dry weather flow) the total investment is V 800,000 including taxes (real costs because installation was taken into operation Januari 2011). Sieving gives a lower load (compared to the current situation without any pretreament) and therefore a lower aeration
for different types of toilet paper. Clearly more effort is needed to understand the conversion processes of cellulose, the kinetics and influences such as temperature, oxygen, nitrate and pH. Mechanism for degradation may be complex because in essence it is a biofilm process with a lag-time for microbial colonisation of the fibre. The lag-time was 1e2 days in the experiments in Table 2. In Fig. 3a the biofilm surrounding the cellulose fibre in activated sludge can be seen. Interesting aspect is also that in mesophilic digesters with 20 days retention time, only partly degradation (a good method to measure cellulose concentration in digested sludge is not available yet) of cellulose fibres is observed (Fig. 3b). The significant presence of cellulose in activated sludge and digested sludge calls for more in-depth studies on the conversion processes of these fibres.
3.3.
47
Processing the sieving material
Various options are available for the treatment and reuse of the sieving material: The heavy metal content of separated materials is low and complies with the decree for application in agriculture as soil conditioner. The separated material has already a high solids content and dries easily, it could therefore be easily used as fuel in a biomass based power plant. The cellulose could be used as feed stock in the fermentation industry for production of for instance biofuels like ethanol. Technically it is possible to use the
Economic evaluation
Fig. 3 e Microscopic pictures of cellulose fibre from in-situ biodegradation experiments with a clear biofilm around the fibre (a), cellulose fibres (and also struvite crystals) abundantly present in digested sludge from WWTP Amsterdam West (b).
48
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 4 3 e4 8
demand. For electricity costs V 0.12/kWh was used, a detailed overview of the cost calculation can be found in the technical Stowa report (Stowa 2010-19). When sieves are applied only for the dry weather flow at the existing low loaded (0.12 g COD/g MLSS day) WWTP Blaricum (which has no primary clarifier), including all investments (sieves, pipes, automation etc.) and operational costs, a payback time of 7 years was estimated. The life time expectancy of the sieve is 15 years. When the entire rain weather flow (RWF) is treated, the break-even point was more than 15 years. Yearly savings were estimated to be V 125,000. Savings related to sludge treatment are most important, since sieving leads to a lower load to the treatment plant. In addition, the use of a fine-mesh sieve may also yield operational benefits for example in the form of reduced debris formation (the intermeshing of hair and fibres). This was not taken into account in the cost comparison. The used sieve did not face any problems with fouling of the belt. However removal of fat and operational attention every one/two weeks takes about 30 min. Clearly an evaluation has to be made in each individual case but it seems that sieving is an economic viable option to improve wastewater treatment plants. Certainly when expansion of wastewater treatment plants are considered the benefits might be larger then here reported. Cellulose forms a significant part of the sludge and removal of this inert material increases the specific activity of activated sludge and thereby the treatment capacity.
3.5.
Energy evaluation
The effects on the energy balance (consumptioneproduction) for the activated sludge process combined with digestion, biogas production, transport of sludge, dewatering, and incineration, were calculated and compared to the reference situation without sieves. The underlying assumption is that the sieve product will be incinerated in a biomass plant. Moreover, separated materials can be dewatered more effectively than sludge, with the result that less transport is required and the caloric value is higher. The energy balance showed clear positive results in the case of the WWTP Blaricum. It appears that fine-mesh sieves represent an alternative to settlement tanks for energy-related considerations. Net energy need (wastewater treatment, sludge treatment and incineration), would amount to at least 40% less compared with the reference without fine-mesh sieving (Stowa 2010-19).
4.
Conclusion
It is remarkably that until now a major fraction of wastewater (cellulose fibres from toilet paper) have hardly been studied.
Preliminary evaluations indicate they might form a major part of the so-called inert COD in the wastewater. Sieves can be used to remove this fraction. The resulting filtered material (approximately. 30% of the COD for Dutch wastewater) can efficiently be used in a biomass power plant for energy generation whereas the resulting sieved wastewater gives advantages in optimizing the wastewater treatment process. This leads to an overall energy optimization and more cost efficient wastewater system. Future evaluation of use of sieves at a full scale application will further refine the evaluation of the potential of cellulose recovery by sieving.
references
Edberg, N., Hofsten, B., 1975. Cellulose degradation in wastewater treatment. Journal of the Water Pollution Control Federation 47 (5), 1011e1020. Frechen, F.B., Schier, W., 2008. Mechanische Abwasser Vorbehandlung auf Kommunalen Membranbelebungsanlagen. Korrespondenz Abwasser 55 (1), 39e44. Henze, M., Gujer, W., Mino, T., van Loosdrecht, M.C.M., 2000. Activated sludge models ASM1, ASM2, ASM2d, and ASM3. IWA scientific and technical report no. 9. IWA Publishing, London, UK. Hofman, J., Hofman-Caris, R., Nederlof, M., Frijns, J., Van Loosdrecht, M., 2011. Water and energy as inseparable twins for sustainable solutions. Water Science and Technology 63 (1), 88e92. Nowak, O., Svardal, K., Franz, A., Kuhn, V., 1999. Degradation of particulate organic matter e a comparison of different model concepts. Water Science and Technology 39 (1), 119e127. Odegaard, H., 1998. Optimised particle separation in the primary step of wastewater treatment. Water Science and Technology 37 (10), 43e53. Roeleveld, P.J., Van Loosdrecht, M.C.M., 2002. Experience with guidelines for wastewater characterisation in The Netherlands. Water Science and Technology 45 (6), 77e87. Rusten, B., 2006. Evaluation and testing of fine mesh sieve technologies for primary treatment of municipal wastewater. Water Science and Technology 54 (10), 31e38. Schier, W., Frechen, F.B., 2008. Efficiency of mechanical pretreatment on European MBR plants, In: International Conference Amsterdam Rai Netherlands, 1st and 2nd October. Sutton, P.M., Melcer, H., Schraa, O.J., Togna, A.P., 2011. Treating municipal wastewater with the goal of resource recovery. Water Science and Technology 63 (1), 25e31. STOWA 2010-19. Influent fijnzeven in rwzi’s, www.stowa.nl. Verachtert, H., Ramasamy, K., Meyers, M., Bever, J., 1982. Investigations on cellulose degradation in activated sludge plants. Journal of Applied Bacteriology 52, 185e190. Verstraete, W., Van de Caveye, P., Diamantis, V., 2009. Maximum use of resources present in domestic “used water”. Bioresource Technology 100 (23), 5537e5545.