Initial Page Layout
The potential use of shear viscosity to monitor polymer conditioning of sewage sludge digestates
I. Oliveira*[1], J. P. Reed**, M. Abu-Orf ***, V. Wilson****, D Jones**** and S. R. Esteves*
* Wales Centre of Excellence for Anaerobic Digestion, **Sustainable Environmental Research Centre, Faculty of Computing, Engineering and Science, University of South Wales, Pontypridd, CF37 1DL, UK. (E-mail: ; ; )
*** Vice President & Residuals Group Practice Leader, One South Broad Street, Suite 1630, Philadelphia, PA 19107, USA (E-mail: )
**** Welsh Water, Nelson, Treharris CF46 6LY, UK (E-mail: ; )
Abstract
The work assessed the use of shear viscosity at 0.1 s-1 (η[0.1s-1]) as a parameter to detect changes in the conditioning and dewatering of digestates. Total and soluble fractions of organic matter of digestate samples before and after storage were also assessed in regards to their conditioning and dewatering performance. Digestate from a conventional mesophilic anaerobic digestion (CMAD) and advanced anaerobic digestion (AAD) plants were used. Linear regression and correlation analysis of 29 different parameters showed that soluble and total fractions of organic matter (Norg, Sc, Sp, Tp, TKN/COD, tCOD and sCOD) during plant operation and storage conditions correlated (r between 0.80 and 0.99) with the variation in polymer dose, floc strength and CST of conditioned digestate samples. The variations occurred within the content of soluble and total fractions of organic matter, and showed to correlate with both conditioning requirements and the variation in η[0.1s-1]. The work concluded that η[0.1s-1] measurements of unconditioned digestate samples have the potential to be used as a parameter to monitor conditioning requirements during digestate storage or during process changes. It was found important to analyse soluble and total fractions of organic matter in order to understand the changes in η[0.1s-1] within specific process conditions.
Keywords
Digestate management
Digested sewage sludge
Storage conditions
Conditioning requirements
Polymer dose monitoring
Rheological characteristics
Abbreviations:
AAD- Advanced Anaerobic Digestion
AC- Ash Content (% DM)
AD- Anaerobic Digestion
CMAD- Conventional Mesophilic Anaerobic Digestion
CSTmin- Minimum capillary suction time
DM- Dry Matter (%)
EPS- Extracellular Polymeric Substances
HT- Holding Tank
Norg- Total organic bound nitrogen
ODM- Organic Dry Matter (% DM)
PDmax- Polymer dose required to achieve the maximum network strength (g/kgTS)
PDmin- Polymer dose required to achieve the minimum capillary suction time (g/kgTS)
Sc- Soluble fraction of total carbohydrates (mg/l)
Sc/Tc- Ratio between soluble and total fraction of carbohydrates (%)
sCOD- Soluble fraction of chemical oxygen demand (mg/l)
sCOD/tCOD- Ration between soluble and total fraction of chemical oxygen demand (%)
SMP- Soluble microbial products
sNorg- Soluble organic bound nitrogen
Sp- Soluble fraction of total protein (mg/l)
Sp/Sc- Ratio between soluble fractions of protein and carbohydrates (%)
Sp/Tp- Ratio between soluble and total fraction of protein (%)
sTKN- Soluble fraction of total kjeldhal nitrogen (mg/l)
sTKN/sCOD- Ratio between soluble fraction of total kjeldhal nitrogen and chemical oxygen demand (%)
sTKN/tTKN- Ratio between soluble and total fraction of total kjeldhal nitrogen (%)
suCOD- Soluble fraction of unidentified chemical oxygen demand (mg/l)
suCOD/tuCOD- Ration between soluble and total fraction of unidentified chemical oxygen demand (%)
Tc- Total fraction of total carbohydrates (mg/l)
tCOD- Total fraction of chemical oxygen demand (mg/l)
TH- Thermal hydrolysis
Tp- Total fraction of total protein (mg/l)
Tp/Tc- Ratio between total fractions of protein and carbohydrates (%)
tTKN- Total fraction of total kjeldhal nitrogen (mg/l)
tTKN/tCOD- Ratio between total fraction of total kjeldhal nitrogen and chemical oxygen demand (%)
tuCOD- Total fraction of unidentified chemical oxygen demand (mg/l)
Wumax- Maximum network strength (MJ/gTS)
η [0.1 s-1] - Limit shear viscosity measured at 0.1 s-1 for 600 s (Pa.s)
1. INTRODUCTION
The management of sewage sludges from wastewater treatment works (WwTW) has been an important topic of research because rapidly increasing population, urbanization and industrialization, extended sewer distribution and new installations of WwTWs have elevated sludge production rates. Sewage sludge generation has been increasing and the requirements for enhanced treatment and resource recovery has become higher on the agenda of the waste water treatment sector since the implementation of the water framework directive (Directive 2000/60/EC) (WssTP, 2015). For many years anaerobic digestion (AD) has been the chosen process for the management and final treatment of biosolids in WwTW, it is estimated that 66% of sewage sludge in the UK is treated via anaerobic digestion (EBA, 2015). With the need to comply with the Nitrates Directive (1991) and local competition for land based markets, the costs of transportation and spreading of digested sludges will likely increase. Digestates will therefore be required to find alternative markets, and as part of this the need to dewater will become even more important. The economic performance of WwTW is critically dependent on the ‘dewatering’ process that is applied to the biosolids before and after AD. It is estimated that biosolids management can contribute to 60% of the annual WwTW operating costs (Bharambe et al. (2014)). Additionally, the expense of conditioning chemicals may constitute a significant portion of the wastewater treatment operating costs, and may, in some instances be as high as 20% of the total treatment plant operating costs (Abu-Orf et al., 2004). Reduction in chemical costs and improvements in dewatering are thus very desirable and increased knowledge on how to accomplish this is required.
Dewatering is a process that usually involves three key stages: coagulation and flocculation, filtration of the conditioned sludge and consolidation phase of the remaining solids cake (Tchobanoglous et al., 2014). Dewatering can be a complex process to optimize since each stage is directly dependent on sludge chemical and physical characteristics (Christensen et al., 2015), polymer type and dosage (Saveyn et al., 2005) (Ayol et al., 2005), mixing energy applied during conditioning (Wang and Dentel, 2010) (Sievers et al., 2008) and pressure applied during filtration and expression phases (Olivier and Vaxelaire, 2005) (Skinner et al., 2015). Typical practices for measuring dewaterability include jar test settling rates, specific resistance to filtration (SRF) test, time to filter (TTF) test and capillary suction time (CST) measurements. However, these measurements are all ‘off-line’ techniques that require manual sampling and thus are poorly suited to the monitoring of full scale plants. In a treatment facility with existing dewatering equipment and with digestate flows and characteristics that could both vary over time, improved conditioner use and control may be an important way to improve and optimize dewatering operation (Örmeci, 2007). Several researchers Yen et al. (2002), Abu-Orf and Örmeci (2004), Ayol et al. (2005), Dentel and Dursun (2006) and Wang and Dentel (2010), have chosen rheology as a way to predict, control, or optimize conditioning and dewatering processes using different rheometric measurements. However, as conducted by Marinetti et al (2010) the correlations between rheological parameters and dewatering properties were not consistent or strong enough to indicate that these rheological tests (dynamic and rotational measurements) could be used to provide useful information regarding full scale dewatering performance. The sludge dewaterability was only related with its rheological properties in an indirect manner, as polymer dose increased network strength increased up to a threshold beyond which further network strength conferred no improvement in dewatering.
It has been reported that there is a positive correlation between soluble EPS (measured as soluble proteins and carbohydrates) with polymer demand (Novak et al. (2003), or increased resistance to filtration for sludges with higher EPS and SMP (Li and Yang, 2007). In addition, in Miller et al. (2008) it is shown that the sludge biofloc structure (size, shape and strength) of unconditioned sludges influences the residual amount of water in the cake rather than the characteristics of the flocs produced after polymer conditioning. Nevertheless, more studies are required that relate these compounds and sludge biofloc characteristics with the conditioning and dewatering performance of full scale processes, which could potentially lead to the development of online tools that could measure digestate changes and adjust polymer dose accordingly. Significant operational performance changes can be realized by implementing process control techniques that use real-time monitoring and ‘on-line’ control of polymer dosing in dewatering processes (Gillette and Joslyn, 2000). The present work, aims to assess the applicability of shear viscosity (η[0.1s-1]) to detect small variations of total and soluble fractions of organic matter and its impact on flocculation and dewatering. Rheological measurements were chosen due to the potential ease of interpretation of the results, relative rapidity and the potential for online implementation. Two established rheometric tests were used to measure conditioned sludge network strength (Wu) and unconditioned sludge biofloc characteristics expressed as η[0.1s-1].
2. MATERIAL AND METHODS
Different types of digestates were used i.e. a conventional mesophilic AD digestate and advanced anaerobic digestion (AAD) digestate samples. Digestate samples were stored at different temperatures (20, 35, 80, 100, 120 and 165°C), or stored for longer periods with and without aeration (from periods of a couple of hours to 1 and 9 days) to drastically change their initial characteristics for comparison purposes. Flocculation was assessed by determining the Wu of the conditioned samples using the method proposed by Abu-Orf and Örmeci (2004) and dewaterability was assessed in terms of filterability using CST measurements, all as a function of polymer dose. In addition, η[0.1s-1]was used to assess unconditioned sludge biofloc characteristics. Linear regression analysis was used to evaluate correlations between total/soluble fractions of organic matter, shear viscosity, polymer dose, network strength and dewaterability.
2.1. Overview of Anaerobic Digestion plant operation
Samples were collected from two AAD and a single CMAD plant. AAD samples were collected from Afan and Cardiff WwTW anaerobic digesters, which operate within a range of mesophilic temperatures of 38 - 42°C with an organic loading rate (OLR) and HRT of approximately 4 kg VS/m3∙d and 15 days, respectively. However, this will vary depending on the quantity of imported sewage sludges at specific times. In Cardiff approximately 90% of the treated sludges are secondary sludges whilst Afan imports a greater proportion of primary thickened sludges. Sludges are pre-treated prior to digestion by thermal hydrolysis (165°C, 6 bar, 30 mins). After digestion, digestate is pumped continuously to an aerated HT before being dewatered. Digestate is kept in the HT for around 5 hours (depending on the digesters OLR and HRT) before being polymer conditioned and dewatered through belt filter presses (BFP).
CMAD samples were collected from Cog Moors WwTW AD plant which at the time of sample collection was operating within a lower range of mesophilic temperatures 26 - 30°C due to a fault in the heating system. The plant typically operated with an OLR and SRT that varied from 1.3 - 2.6 kg VS/m3·d and between 15 and 30 days, respectively. The resulting digestate was then pumped to post-digesters before being dewatered through centrifuges.
2.2. Experimental procedure
Digestate samples of 160 l were collected at different times from the three sites throughout a two month period and different storage conditions were applied (Table 1). Digestate samples were stored at room temperature for a period of time while the different storage conditions and polymer conditioning were being carried out. High temperature storage at varying duration was applied to 10 l of digestate as shown in Table 1. Aeration storage was applied to another 10 l of digestate using a custom built stainless steel continuous stirred tank reactor (CSTR). A stainless steel sparger connected to a stainless steel downpipe was used to provide aeration using compressed air at 1 bar and a flow rate of 10 l/min. Two CSTRs were used, one as a control (without aeration) and the other with aeration as shown in Table 1. Conditioning and dewatering assessment was conducted 24 h after a given period of storage was applied to allow samples to come to the same temperature and remove temperature effects from the results.
2.2.1. Conditioning and dewatering assessment
Conditioning was performed using a range of polymer doses between 4 – 50 g polymer/kg of dry matter (DM), and for each polymer dose, 6 replicates of 200 ml were conditioned using a Jar Tester (Phipps & Bird). Wu and dewatering of conditioned samples produced were assessed according to Abu-Orf and Örmeci (2004) by measuring Wu and CST to assess floc strength and dewaterability of conditioned samples. In each conditioning and dewatering assessment, the results for PDmin (polymer dose required to achieve the minimum CST), CSTmin (minimum capillary suction time), PDmax (polymer dose required to achieve the maximum network strength) and Wumax (maximum network strength) were used as conditioning and dewatering indicators. An example of the output of conditioning and dewatering for two different samples is shown in Figure 1. Where PDmin corresponded to the polymer dose to achieve CSTmin and PDmax corresponded to the polymer dose to achieve Wumax. For each unconditioned digestate sample η[0.1s-1] was measured. For each unconditioned digestate, samples were taken and stored in a freezer (-20°C) until further analysis were conducted for the characterization of total and soluble fractions of organic matter. Conditioning was performed using a cationic polyelectrolyte FO4490 (SNF, UK) of medium molecular weight (8x106 Da) collected as a powder from Cardiff WwTW and for each conditioning and dewatering assessment, 5 l of 0.6% and 1% of DM were prepared using a homogenizer. The prepared polymer solution was also analysed in terms of shear viscosity, conductivity and DM after each batch.
2.2.2. Analytical characterization of samples
Unconditioned samples were analysed in terms of DM, organic dry matter content (ODM), ash content (AC), total (tCOD) and soluble (sCOD), CST, pH and conductivity according to the standard methods for the examination of water and wastewater (Eaton et al., 1995). Soluble cations (Na+, NH4+, K+, Mg2+ and Ca2+) were analysed according to Dionex application note 141 (Sunnyvale, California). Total (tTKN) and soluble (sTKN) were performed according to DIN 38409. Organic bound nitrogen (Norg) was estimated by subtracting NH4+ content from the tTKN content. Total (Tp) and soluble (Sp) proteins were determined according to Hartree (1972) using bovine serum albumin as the standard. Total (Tc) and soluble (Sc) carbohydrates assay was conducted according to Dubois et al (1956) using glucose as the standard. Total unidentified (tuCOD) and soluble unidentified (suCOD) COD was derived from the difference between COD and the equivalent COD from protein and carbohydrate according to Grady et al. (2011). The COD equivalent from VFA in the digested sludges was considered negligible (VFA concentration in digestate was below 200-400 mg/l). All measurements were conducted in triplicate.