Review of Recent Trends in Capillary Suction Time (CST

Sep 30, 2005 - The feasibility of employment of the CST test for characterizing the dewaterability of excess activated sludges was examined. The CST w...
2 downloads 10 Views 189KB Size
Ind. Eng. Chem. Res. 2005, 44, 8157-8163

8157

REVIEWS Review of Recent Trends in Capillary Suction Time (CST) Dewaterability Testing Research Miklas Scholz* Institute for Infrastructure and Environment, School of Engineering and Electronics, The University of Edinburgh, Faraday Building, The King’s Buildings, Edinburgh EH9 3JL, Scotland, United Kingdom

The aim of this review paper is to critically assess recent trends in capillary suction time (CST) dewaterability testing research. The importance of floc structure in explaining practical dewaterability problems such as filter paper clogging is assessed. Current literature on CST for sludge dewaterability is reviewed. The empirical CST test is well-established as the leading method for the determination of sludge dewaterability, despite its shortcomings, such as restricted possibilities to modelsand, therefore, predictsphysical processes, such as the amount of water bound by the filter paper. Nevertheless, the CST apparatus is portable, and the method is easy to conduct, quick, cost-effective, and accurate, if the product of solid concentration and average specific resistance to filtration is of interest. The use of modified CST tests with a rectangular instead of a circular sludge storage and distribution facility, and a filter paper with smaller pore diameter, were most promising. Introduction Aim and Objectives. The aim of this work is to review dewaterability tests, with particular attention to the application of the capillary suction time (CST) apparatus. The objectives are as follows: (1) to review current sludge conditioning aids and methods assessed with the CST device; (2) to assess current knowledge on the divalent cation bridging theory (DCBT), floc structure, and extracellular polymeric substances (EPS); (3) to critically review current literature on CST and other tests for sludge dewaterability; (4) to assess modified CST tests and the potential for modeling processes within CST apparatuses; and (5) to propose laboratory tests and further research work to address open research questions. Dewaterability Tests. Wherever sludges have to be disposed of, be it in restricted land areas or transported over long distances for ultimate disposal, some form of volume reduction is usually necessary. Sludge dewatering is an effective method of volume reduction in such cases. It is also an essential pretreatment where incineration is required. Dewatering processes in common use, such as pressure filters, vacuum filters, and centrifuges, require, for their design, some measure of the sludge dewatering characteristics. Most dewaterability tests are simple but empirical and shed little light on dewaterability mechanisms that could be described mathematically. In comparison, viscosity (resistance to flow) and streaming characterize the effectiveness of sludge conditioning. Both can be * To whom correspondence should be addressed. Tel.: +44 131 6 506780. Fax: +44 131 6 506554. E-mail: [email protected].

used as measurements that characterize the filtrate from the dewatering process. Various measurements of dewaterability, such as the CST, the conditioned filtrate (liquid stream) streaming current, and filtrate total solids were examined previously.1 A significant correlation between the filtrate viscosity and the CST was established, with both being related to dewaterability. There was also a clear relationship between the CST, the filtrate streaming current, and solids recovery. A minimum CST, filtrate streaming current values close to zero for both the filtrate and the conditioning sludge, and the minimum filtrate viscosity correlated well with maximum solids recovery.1 Nevertheless, the two most common methods used to measure the ease of dewatering are specific resistance to filtration (SRF) and CST. A detailed discussion of different dewaterability tests is beyond the scope of this paper. However, detailed discussions can be found elsewhere.1-5 Capillary Suction Time. The CST equipment has long been established as a practical but empirical method for the determination of sludge dewaterability. The main use for the CST is to determine filterability after the addition of coagulant aids. The feasibility of employment of the CST test for characterizing the dewaterability of excess activated sludges was examined. The CST was shown as a good index for sludge filterability, if only the product of solid concentration and average SRF is of interest. On the other hand, the bound-water content cannot be directly evaluated from the CST data.2 The CST apparatus provides a simple, rapid, and inexpensive method for comparing the effects of different agents and dosages. Slurry is poured into a small open tube resting on a piece of filter paper. The filtrate is extracted by capillary suction, and a cake is formed at the bottom of the tube. The rate at which filtrate is

10.1021/ie058011u CCC: $30.25 © 2005 American Chemical Society Published on Web 09/30/2005

8158

Ind. Eng. Chem. Res., Vol. 44, No. 22, 2005

extracted is dependent on the resistance of the cake. By measuring the distance the filtrate is required to travel along the paper as a function of time, the cake resistance can be determined. This method provides a simple technique for estimating dewaterability and, on a comparatively basis, can be useful in practice. The method relies on the varying pressure applied by the movement of water through the filter paper, so that a theoretically complete mathematical treatment of the CST is virtually impossible (see below). Moreover, sludge filterability predominantly governs the output of almost all the various types of dewatering equipment, including drying beds, belt and filter presses, vacuum filters, and centrifuges. The capillary suction pressure generated by the standard filter paper is used to extract water from the sludge. The rate at which water permeates through the filter paper varies, depending on the condition of the sludge and the filterability of the cake formed on the filter paper. The CST is obtained from two electrodes placed at a standard interval from the funnel. The time taken for the water front to pass between these two electrodes constitutes the CST. The force generated by capillary suction is much greater than the hydrostatic head within the funnel, so the test is relatively independent of the amount of sludge, as long as there is a sufficient amount to generate a CST measurement. Brief History of the CST Sludge Filterability Measurement Apparatus. The CST apparatus developed by Gale and Baskerville at Stevenage (U.K.) in 1967 provides a simple, rapid, and inexpensive method for comparing the effects of different agents and dosages.3 The CST simplistically is the time to collect a unit volume of filtrate of a sludge undergoing filtration in a standard-sized CST funnel, when placed upon a standard grade of chromatography paper (Whatman No. 17). Whatman No. 17 is a smooth cellulose filter paper of grade 17 chr with a very high flow rate (190 mm per 30 min). This paper is suitable for the heaviest loadings and is ideal for preparative paper chromatography and electrophoresis. However, the wet front moves faster along the grains than across them. Typical data according to the manufacturer are as follows: basis weight, 413 g/m2; specific typical thickness, 920 µm @ 0.54 kg/ cm2; gurley air porosity (0.14 kg cylinder), 4.13 s/100 mL/cm2; dry burst, 3.36 kg/cm2; wet burst, 0.13 kg/cm2; 98% liquid retention efficiency, 8 µm; and ash content, 0.025%. (Data obtained from a personal communication with G. Barton of Whatman, December 23, 2004, and from the Whatman Product Guide 2004/05, which is a company publication.) Because the CST is an automatic measurement that requires little expertise or training to enable precise measurement of the filterability to be recorded, it is suited for use wherever sludge dewatering (e.g., filter press, vacuum filter, and centrifuge) is undertaken. Sludge almost invariably requires the addition of a chemical conditioner to improve its dewatering characteristics; therefore, the use of the CST apparatus enables an optimum dose of conditioning chemical to be utilized. Moreover, CST measurements may permit the identification of the cause of inadequate performance, such as the excessive shearing forces that occur within worn pumps. The CST values usually correlate well with SRF determinations. However, such a correlation should be

Figure 1. Standard apparatus (model 304B CST) to measure the capillary suction time (CST).

Figure 2. Revised apparatus (model 319 Multi-purpose CST) to measure multiple capillary suction times (CSTs) or measurements at five fixed positions.

confirmed experimentally. The CST values do not allow for the solids content; therefore, it is necessary to specify the solids content of the sludge. It is possible for sludge with a low CST value and high solids content to be easily dewatered. The CST is measured with the standard CST apparatus: model 304B CST (Figure 1). The new CST apparatus, model 319 Multi-purpose CST (Figure 2), can be used with either five single-radius test heads or with one multi-radii test head for the rapid assessment of SRF. The SRF can be calculated from the slope of the filtration curve. The latter application option allows for better mathematical treatment of the corresponding CST values (see below and the Web address www.tritonel.com).

Ind. Eng. Chem. Res., Vol. 44, No. 22, 2005 8159

Using single-radius test heads as the mode of operation, it is possible to test several dosing quantities at the same time, thereby reducing the overall test time or allowing for more discrete tests to be performed. In each case, the model 319 Multi-purpose CST operates in a manner very similar to that of model 304B CST. The multi-radii test head has five probes, set at intervals to give the same volume between each probe. For a wide range of different types of sludge, there is an approximately linear relationship between the SRF, using the standard Bu¨chner funnel apparatus and the multi-radii test head. When using the multi-radii test head, the model 319 Multi-purpose CST automatically divides the time by the probe number. Since filtration fulfils a parabolic relationship between time and the volume of filtrate, then the arithmetic difference in the measurement is constant and directly proportional to the SRF (according to a personal communication with J. Tapp of Triton Electronics, December 23, 2004). Recent Advances in Understanding Capillary Suction Time (CST) Dewaterability Test Variables. Capillary suction time (CST), filtrate streaming current, viscosity, and solids measurements are conditioning indicators. Dewatering indicators include the CST, the filtrate streaming current, and the percentage solids recovery. It follows that CST, for example, can be used as both a conditioning indicator and a dewatering indicator. Moreover, previous research indicated the lack of correlation between certain conditioned sludge control parameters and some conditioning indices over a wide range of mixing conditions.4 Strong correlations between the dispersed (spatially scattered) mass concentration, CST, and rheological parameters for activated sludge suspensions of varying solids content were observed.5 The restriction of flow caused by the development of interparticle networks led to an increased surface erosion of primary particles. The increased dispersed mass concentration, in turn, caused a deterioration of filterability.5 Sludge Conditioning Aids. The addition of alum (up to 4% dosage) rapidly decreased the SRF and CST of oily sludges. Alum dosages beyond 4% only increased the solids content in the sludge cake and increased the sludge volume to be handled. The total suspended solids of the filtrate decreased with alum dosage. The correlations between dewatering characteristics were studied. A correlation between the CST and SRF was established. Using measured CST data and the corresponding correlation equation, a quick prediction on dewaterability can be established.6,7 Fenton’s reagent (Fe2+ and hydrogen peroxide (H2O2)) can also be used as a chemical conditioner, as demonstrated recently.8 The effects of various Fe2+ and H2O2 concentrations on the SRF and CST of biological sludges were investigated. Experimental results indicated that high concentrations provided high dewatering efficiencies. The feasibility of using ferric chloride (FeCl3), in conjunction with polymers, for conditioning wastewater sludges was examined.9 Specific resistance to filtration, CST, and time to filter were evaluated. Optimum dosing resulted in complete charge neutralization. Dual-chemical conditioning showed that a combination of 50% of optimum dosage of FeCl3 and 50% of an optimum dosage of a polymer (Percol 757) had the

lowest SRF, CST, and time-to-filter values. A percentage of the optimum dosage of polymer could be replaced by a similar percentage of the optimum dosage of FeCl3 without adversely affecting the dewatering efficiency. A good correlation was observed between the SRF and CST data.9 Furthermore, the CST and time to filter correlate well. Either dewaterability variable can be used in the laboratory to test the drainage characteristics of the residuals.10 The increase of SRF and CST with increasing oil content in sludge was observed previously.6,7,11 It has been found that addition of municipal solid wastes incinerator fly ash decreased the SRF and CST of oily sludges rapidly, up to 3% dosage. Beyond 3% fly ash, the decrease was less significant and the solids content in the sludge cake increased. This optimum dosage remained the same for sludges with varying oil contents. Correlations of CST with the dewatering characteristics such as SRF, filter yield, and corrected filter yield were established. These correlations were used to predict dewaterability.11 Good correlations between the CST and SRF were established for sludges from water treatment plants, but not for biological sludges containing organic matter, such as flocs. Cationic polymers (conditioning aids) exhibit the best enhancement on moisture removal for these sludges. The dewatering potential of sludge was greatly enhanced by thermal treatment, in conjunction with the use of polymers.12 The optimum dosage of a nonionic polymer during the conditioning of an alum sludge arising from the treatment of colored upland water was studied previously.13 Dewaterability was investigated through CST and SRF tests, together with viscosity measurements on the liquid phase of sludge. The optimum dose associated with the CST and SRF was relatively low. The CST and SRF optima were determined to correspond to a state of ∼75% of the saturation coverage. Divalent Cation Bridging Theory (DCBT) and Cation Balance. The application of the divalent cation bridging theory (DCBT) to improve settling, dewatering, and effluent quality in pilot-scale reactors and full-scale systems treating industrial wastewater was examined.14 This was accomplished by reducing the monovalent-todivalent cation ratio by direct divalent cation addition. Research showed that the monovalent-to-divalent ratio is a potential indicator for settling and dewatering problems at wastewater treatment plants. Monovalentto-divalent ratios of >2 were associated with slow settling, reduced dewaterability, and insufficient effluent quality. Results showed that the addition of calcium improved floc properties (see below). According to the DCBT, the goal of chemical selection should be to reduce the ratio of monovalent cations to divalent cations to achieve improvements. A study was conducted to determine the effect of using magnesium hydroxide as an alternative to sodium hydroxide at a full-scale industrial wastewater treatment plant. The dewatering properties of the activated sludge improved dewatering, as measured by decreased CST and SRF values, along with an increase in cake solids from the SRF test. The DCBT was a useful theoretical tool for assessing chemical-addition strategies and their potential effect on activated sludge performance.15 Liquid hog manure was previously used during a laboratory-based experiment to investigate the effects of Ca2+ and Mg2+ ions and nitrification inhibition on

8160

Ind. Eng. Chem. Res., Vol. 44, No. 22, 2005

the overall settling and dewatering characteristics. The results indicated that settling and dewatering characteristics improved during the course of the experiment. However, the improvement in dewaterability, as quantified by the CST, was not consistent. The greatest filterability in the supernatant was observed for a CST of 40 s and for a monovalent-to-divalent cation ratio of 2:1.16 Sludge Floc Structure. Most sludges are made of flocs produced during the coagulation stage. Their dewatering characteristics are dependent not only on the equipment that is used, but also on the chemical, structural, and physical characteristics of the sludge flocs. Therefore, the relationship between floc structure and sludge dewaterability was investigated.17 Ferric flocs, for example, contained relatively low amounts of bound water but exhibited relatively high CST values, and, therefore, high SRF to water removal, in comparison to alum flocs. The floc structure and sludge dewaterability were determined to be linked for various sludge types. The bound-water content in the sludge was determined to decrease with the floc fractal dimension. Less-compact flocs contained more water but less bound water. The CST was lower for small sludge flocs, in comparison to that for large sludge flocs.17 It follows that CST is a useful variable to indirectly characterize fractal dimensions. Furthermore, both the size and structure of wastewater solids (biosolids) have an impact on the dewaterability. Studies in which the size and structure of activated sludge flocs have been altered in a controlled manner were undertaken.18 The dewaterability of the resultant flocs were examined using the CST test. A small-angle light scattering method was applied for the determination of activated sludge floc size and structure. Sludge floc structure was determined to be a major determinant of CST. In some instances, flocs had almost double the median size but possessed similar fractal dimensions and similar CST values.18 Influence of Extracellular Polymer Substances (EPS) on Dewaterability. The floc physical parameters are important variables, which affect the water binding ability of the sludge flocs. The morphological characteristics and the presence of humic substances are likely to have relatively weak impact on the dewaterability. In contrast, the polymeric components protein and carbohydrate contribute to the enhancement of the water binding ability of the sludge flocs. As shown experimentally, the CST had good statistical correlations with polymers measured in both sludge and the extracted EPS.18 Furthermore, the bound water only correlated well with the individual polymers measured in the sludge. High concentrations of Ca2+, Mg2+, Fe3+, and Al3+ were linked to a significant improvement of sludge dewaterability.18 Anaerobic digestion was shown to alter the quantity of EPS present, with digested sludge containing less EPS than raw sludge. The level of EPS that may give maximum sludge dewaterability was calculated to be ∼20 and ∼10 mg EPS/g suspended solids (SS) for raw and digested sludge, respectively. The organic composition of the EPS extracted from digested sludge contained more protein, with respect to carbohydrates, than that extracted from pre-digested sludge. These changes in the sludge EPS were shown to correlate well with the changes in sludge dewaterability.19

Furthermore, the end point extracted EPS yield was related to the sludge dewaterability, as measured using the CST test.20 There was a very strong, positive correlation between the EPS yield and sludge dewaterability, but no relationship was observed between EPS yield and digester retention time. Further analysis of the extracted polymer organic and inorganic content demonstrated that the digester retention time had no effect on polymer composition, which seemed to have an impact on the sludge dewaterability.20 Moreover, the presence of microbial EPS is important for sludge flocculation, but EPS has also been shown to have a detrimental effect on the dewaterability of certain sludge types.21 Previous results indicated that, for each type of sludge examined, there seems to be an optimum level of EPS at which the sludge should exhibit maximum dewaterability. The optimum levels are ∼20, 35, and 10 mg EPS/g SS for raw, activated, and digested sludge, respectively. The establishment of a trend between sludge dewaterability and the quantity of EPS present opened up the possibility of manipulating the level of microbial polymer present to aid sludge dewatering and, hence, reduce plant-operating costs.21 In laboratory tests, an increase of the potassium concentration beyond requirement as a nutrient for biological growth impeded the activated sludge dewatering properties, as measured by CST and SRF, and associated with an increase in soluble protein. An increase in effluent total organic carbon and effluent turbidity was observed at higher concentrations of potassium. Conversely, an increase in concentration of potassium improved the settling properties of sludge with a low equivalent monovalent-to-divalent cation ratio.22 Modifications of the CST Methodology. Although the standard CST method works well for the majority of sludges, it is often not a useful test for measuring the filterability of highly flocculated sludges or sludges conditioned by other methods such as freeze-thaw. The escape of water from the sludge in such cases is too fast and the rate-limiting step becomes the movement of water through the filter paper instead of the release of water from the sludge. There is a need to improve the CST test so that it can be used to measure the degree of filterability for super-flocculated sludges as well. Therefore, a new CST measurement methodology, which consists of first draining the free water from the well-conditioned (super-flocculated) sludge and only then allowing the flocculated sludge to be exposed to the filter paper, was developed. The filterability of super-flocculated sludges can be measured with acceptable precision using this modified CST test procedure.23 CST Filter Paper Selection. Since the invention of the CST test, the same filter paper has been used to allow for an easy comparison of test results worldwide. There are economic concerns that the standard Whatman filter paper is too expensive as a consumable item. Moreover, the technical concern that depth filtration of colloids might be likely due to the relatively large pore size (8 µm) have not been addressed. The flow of water in this paper (Whatman No. 17 chromatography filter paper; see previous discussion) and the CST method for determining filterability of suspensions have been examined previously.2,24 Filter paper water flow was described by the theory of water flow in porous media. This theory is based on the material balance for the water and Darcy’s law. The researchers assumed that

Ind. Eng. Chem. Res., Vol. 44, No. 22, 2005 8161

the matric potential or suction (arising from the interaction of water with the solid surfaces and their geometry) and the hydraulic conductivity of both the filter paper and the suspension are functions of the water content. The one-dimensional CST was analyzed from considerations of the material balance and continuity of the water potential on the filter paper and suspension interface. Experiments agreed with the theory and illustrated that flow in the filter paper is unsaturated. The advance of the wetting front in the filter paper was assumed to be a function of the interaction of the sorptivity and matric potential functions of both the paper and the suspension.24 Extension of this approach to radial flow in the filter paper was too complicated to permit simple interpretation of the wetting front advance. Therefore, the CST method might remain a useful empirical tool for practicing engineers. Its weaknesses should be recognized, however, and constant pressure filtration tests, for example, which provide both sorptivity pressure and an equilibrium water content pressure relationship simply and accurately, should be preferable in most research applications.24 Furthermore, the water movement in filter paper to measure the CST has also been studied elsewhere.25 It has been emphasized that a rectangular CST apparatus is able to overcome the problem of anisotropic filter paper by making use of only one direction of the filter paper. Both diffusion-like and pistonlike (constant capillary suction pressure) approaches were described. It must be noted that estimations based on the pistonlike approach obviously lead to an overestimation of the liquid invasion volume. This will lead to errors in estimating the specific resistance of the cake (see below). The liquid saturation profiles in the filter paper for pure water and some slurries were experimentally constructed. Results indicated that the unsaturated flow in filter paper was unexceptionable in all tests. However, the CST correlations derived from either pistonlike or diffusion-like approaches resembled each other in form. The shape was subsequently attributed to the S-shaped profile near the wet front, resembling a powerlaw-type behavior.25 CST Test Affected by Sedimentation. A problem with the conventional CST test is the effect of sedimentation, which cannot be neglected for a majority of flocculated slurries and sludges. Sediments such as suspended solids and heavy flocs accumulate on top of the filter paper during the CST test. The CST theory does not take the effect of sedimentation into account. This may lead subsequently to an overestimation of the cake resistance. However, the application of a rectangular chamber should solve this problem. A modified CST theory for rectangular cells (outside dimensions of 40 mm × 40 mm) including sedimentation (see above) was developed recently. Calculated values of resistance, based on the new model, indicated the existence of large errors (e400%) in former CST theories and models that omitted sedimentation.26 The effects of particle sedimentation on the wet front dynamic behavior of the CST test apparatus were investigated experimentally and theoretically.27 A dynamic model incorporating particle sedimentation effects was solved numerically with zone settling velocity and average SRF as model input data. In contrast to filter papers of small to medium sizes, particle sedi-

mentation was the most influential factor in determining the CST test dynamics for relatively large filter paper sizes (regardless of the brand). Furthermore, the idea of stirring the slurry to prevent sedimentation was introduced for the first time.27 CST Theories, Modifications, and New Methods. The CST apparatus has been studied both theoretically and experimentally.2,28-30 The radial CST apparatus, and the filter paper used by it, were investigated in various ways to obtain the capillary suction pressure, the water porosity of the filter paper, and the water saturation in the filter paper, each as a function of the radial distance from the suspension to be filtered.31 Air was shown to coexist with water in the wetted Whatman No. 17 filter paper (see previous discussion), but the capillary suction pressure and the effective water porosity were shown to be practically constant in the filter paper, as used in the CST method. Measurements were made of the desorptivity of a bentonite suspension, using a conventional constant-pressure filter press, and compared with the desorptivity obtained from CST measurements.31 A method for estimating the averaged specific resistance of the filtration cake, which avoids the necessity of measuring the liquid invasion volume in the CST tests, was proposed. When insufficient slurry was used, the slurry was exhausted during an experiment, and the curve of wet front radius versus time showed a transition.32 The liquid saturation under the inner cell was estimated based on the wet front distance and CST data of the transition point. The transition point shifted according to the slurry concentration, cake specific resistance, and the CST parameters. The average specific resistance of cake was calculated without the liquid invasion volume data. Based on the wet front radius and the CST of the transition point, the liquid saturation under the inner cylinder was estimated and the averaged specific resistance of cake calculated without the liquid invasion volume data. The method agreed well with vacuum filtration data.33,34 A theoretical model was presented for the description of the liquid flow in the conventional CST apparatus (model 304B CST, depicted in Figure 1). The filter paper in this CST apparatus was replaced by ceramics in a new CST apparatus. The position of the liquid front was determined continuously by measuring the electrical resistance of the wetted ceramics as a function of time. In contrast, the conventional CST apparatus determined the position of the liquid front at only two fixed positions.30 However, the modified model 319 Multipurpose CST (see Figure 2) is capable of measuring five fixed liquid-front positions (see previous discussion). Fluid flow and cake formation in a novel rectangular CST apparatus were investigated experimentally and theoretically.33,35 The wet front shape is taken as an ellipse in a cylindrical CST apparatus test, and as a straight line in a rectangular capillary suction apparatus test.35 Therefore, the fluid flow field in a rectangular CST apparatus test was shown to be onedimensional (straight line). The particle sedimentation affects the CST in a cylindrical capillary suction apparatus (see previous discussion). In contrast, sedimentation effects were insignificant in rectangular capillary suction apparatus tests. The CST was also little affected by the filter paper shape, the top and the bottom plates, or when the inner cylinder loading was