- Email: [email protected]
The chemical and mechanical differences between alginate-like exopolysaccharides isolated from aerobic flocculent sludge and aerobic granular sludge
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 5 7 e6 5
Available online at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/watres
The chemical and mechanical differences between alginate-like exopolysaccharides isolated from aerobic flocculent sludge and aerobic granular sludge Y.M. Lin a,*, P.K. Sharma b, M.C.M. van Loosdrecht a a
Department of Biotechnology, Delft University of Technology, Julianalaan 67, 2628BC Delft, The Netherlands Department of Biomedical Engineering, University Medical Center Groningen, University of Groningen, P.O. Box 196, 9700 AD Groningen, The Netherlands b
article info
abstract
Article history:
This study aimed to investigate differences in the gel matrix of aerobic granular sludge and
Received 7 June 2012
normal aerobic flocculent sludge. From both types of sludge that fed with the same
Received in revised form
municipal sewage, the functional gel-forming exopolysaccharides, alginate-like exopoly-
21 August 2012
saccharides, were isolated. These two exopolysaccharides were chemically fractionated,
Accepted 6 September 2012
and investigated by FT-IR spectroscopy. The isolated polymers were made into a gel by
Available online 5 October 2012
calcium addition and the mechanical properties of these reconstituted gels were measured by a low load compression tester. The viscoelastic behavior of the gels was described by
Keywords:
a generalized Maxwell model. The alginate-like exopolysaccharides derived from aerobic
Exopolysaccharides
granules had significantly higher amount of poly(guluronic acid) blocks but lower amount
Gel-forming property
of poly(guluronic acid-manuronic acid) blocks in the chemical structure, while the
Aerobic granular sludge
alginate-like exopolysaccharides derived from aerobic flocculent sludge had equal amount
Activated sludge
of poly(guluronic acid) blocks and poly(guluronic acid-manuronic acid) blocks. These differences result in a perfect gel-forming capability of alginate-like exopolysaccharides derived from aerobic granules and bestowed this exopolysaccharides gel a stronger mechanical property as compared to alginate-like exopolysaccharides derived from aerobic flocculent sludge. The different chemical and mechanical properties of these two exopolysaccharides contributed to the distinguished characteristics between aerobic granular sludge and aerobic flocculent sludge. ª 2012 Elsevier Ltd. All rights reserved.
1.
Introduction
Exopolysaccharides are recognized as key elements that shape and provide structural support for biofilms (Sutherland, 2001). Likewise, they are highly involved in aerobic flocculent
sludge and aerobic granular sludge matrix structure (Seviour et al., 2009a, 2009b). Some efforts have been made to understand how and why these two microbial aggregates differed, by comparing the quantity of exopolysaccharides in the isolated extracellular polymers from these two microbial
Abbreviations: FT-IR, Fourier transform infrared; G, Guluronic acid; GG, Poly(guluronic acid); M, Mannuronic acid; MM, Poly(mannuronic acid); MG, Alternating mannuronic and guluronic acid; ALE, Alginate-like exopolysaccharides; ALEflocs, Alginate-like exopolysaccharides derived from aerobic flocculent sludge; ALEgranules, Alginate-like exopolysaccharides derived from aerobic granular sludge. * Corresponding author. E-mail addresses: [email protected] (Y.M. Lin), [email protected] (P.K. Sharma), [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.2012.09.017
58
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 5 7 e6 5
aggregates respectively (Adav et al., 2008; McSwain et al., 2005). However, research focusing on isolating and characterizing the functional exopolysaccharides by comparing their physical and chemical properties, and investigating the contribution of these properties on the difference between aerobic flocculent sludge and aerobic granules is just started very recently. Based on identifying the crossover between storage modulus relative to the loss modulus by rheological measurement, a study displayed that the exopolysaccharides isolated from aerobic granules were able to form gels at pH 6e8.5, while the exopolysaccharides isolated from flocs couldn’t. Thereafter, difference in gel-forming capability of exopolysaccharides at different pH was considered as a distinctive property between granules and flocs (Seviour et al., 2009b). However, explanation for the different gelforming capability based on exopolysaccharides structure was not provided. In our previous research, one of the major exopolysaccharides extracted from aerobic granules resembled commercial alginate but differed in the behavior in a few chemical reactions (Lin et al., 2010). These polymers were denoted as alginate-like exopolysaccharides (EPS). On the other hand, evidence also existed that an alginate-like crosslink of exopolymers was important for the aerobic flocculent sludge structure (Bruus et al., 1992). Therefore, alginate-like EPS could be involved in the basic structures of both granular and flocculent sludge. It is worth pointing out that alginate is not a single polymer, but a family of polymers with different chemical structures. The gel-forming capability of alginate was decided by their chemical structures (Draget et al., 2000). Therefore, aiming at differentiating aerobic granular sludge from aerobic flocculent sludge, a hypothesis is put forward in the present research: the alginate-like exopolysaccharides in aerobic granular sludge and aerobic flocculent sludge are dissimilar in their chemical structures, and this fundamental difference decides the gel-forming capability of exopolysaccharides and mechanical property of the gel. To test the hypothesis, alginate-like exopolysaccharides were isolated both from aerobic granular sludge from a pilot plant and aerobic flocculent sludge from a wastewater treatment plant fed with the same municipal sewage. These two exopolysaccharides were chemically fractionated, investigated by FT-IR spectroscopy, and compared. The mechanical property of the gel formed after alginate-like exopolysaccharides crosslink with Ca2þ was measured by a low load compression tester. The viscoelastic behavior of the gels was described by a generalized Maxwell model.
2.
Average parameters of the influent were: CODtotal 585 mg/L, suspended solids 195 mg/L, NH4eN 55 mg/L and PO4eP 6.3 mg/L. The reactor was operated in Sequencing Batch (SBR) mode for biological phosphate and nitrogen removal. Operational details were described in Lin et al. (2010). After start-up, biomass concentration in the reactor was maintained around 8e10 g TSS/L. Oxygen in the reactor was controlled between 2 and 3 mg/ L during aeration. Temperature and pH were not controlled in this system and depended on the incoming sewage. Aerobic granular sludge was collected during steady operation of the reactor. Aerobic flocculent sludge of the same wastewater treatment plant was collected from the aeration basin.
2.2.
Alginate-like exopolysaccharides (ALE) extraction
Alginate-like exopolysaccharides (ALE) were extracted from both aerobic granular sludge and activated sludge according to Lin et al. (2010). Dried biomass (0.5 g) was extracted in 0.2 M Na2CO3 at 80 C. After centrifugation at 15,000 rpm for 20 min, the pellet was discarded. The supernatant pH was adjusted to 2 by adding 0.1 M HCl. The precipitate was collected by centrifugation (15,000 rpm, 30 min), washed by di-deionized water until effluent pH reached 7, and dissolved in 0.1 M NaOH. The ALE in the supernatant was precipitated by the addition of cold absolute ethanol to a final concentration of 80% (vol/vol). The precipitate was collected by centrifugation (15,000 rpm, 30 min), washed three times in absolute ethanol and lyophilized. Ash content of the extracted exopolysaccharides was measured according to the standard method (APHA, 1998).
2.3.
Gel formation property in CaCl2
To test ALE’s gel formation property in CaCl2, ALE-miliQ water solution (10 ml, 2% (w/v)) was added drop wise into CaCl2 solution (150 ml, 2% (w/v)) using a peristaltic pump through a syringe needle (0.9 40 mm) (Fig. 1). The ALEs were crosslinked in the CaCl2 solution at room temperature for 16e18 h.
Materials and methods
2.1. Activated sludge and aerobic granular sludge for investigation Granules were sampled from the Nereda pilot plant, operated by DHV at the wastewater treatment plant Epe, The Netherlands. The reactor was fed with municipal sewage. The influent consisted of approximately 25% of slaughterhouse wastewater, which was discharged in the sewage system.
Fig. 1 e Making alginate-like exopolysaccharides gel by dropping 2% alginate-like exopolysaccharides sodium solution into 2% CaCl2 solution.
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 5 7 e6 5
The morphology of Ca-ALEs was photographed by using an Olympus SZH10 stereomicroscope. The water content of both Ca-ALEs was measured by drying the samples at 105 C and recording the loss of the weight.
2.4. Fractionation of the isolated alginate-like exopolysaccharides by partial acid hydrolysis Alginates are characterized as a family of copolymers comprised of mannuronic acid (M) and guluronic acid (G) units arranged in an irregular block pattern of varying proportions of GG, MG and MM blocks. The amount of these three blocks in alginate molecules is considered as the fine chemical structure of alginate polymer. This parameter determines the alginate properties, especially the gel-forming behavior (Hampson et al., 2005). To study the fine chemical structures of ALE derived from granules (ALEgranules) and from flocs (ALEflocs), both the isolated ALEs were fractionated according to Leal et al. (2008). This partial hydrolysis method is able to separate fractions enriched in different kind of blocks (Leal et al., 2008). ALE (0.250 g) were dissolved into 9 ml di-deionized water. After the addition of 1 ml 3.0 M HCl, it was heated at 100 C for 0.5 h. After cooling, the mixture was centrifuged (15,000 rpm) and the supernatant solution was neutralized with 1.0 M NaOH and poured over 80 ml of ethanol, yielding a precipitate that was dissolved in di-deionized water and freeze-dried (Fraction 1). The insoluble material was dissolved in 1.0 M NaOH; the pH was decreased to 2.85 by the addition of 1.0 M HCl. The soluble fraction was neutralized with 1.0 M NaOH and precipitated by the addition of ethanol to a final concentration of 80% (vol/vol) and freeze-dried (Fraction 2). The insoluble fraction was dissolved by neutralization with 1.0 M NaOH, precipitated by ethanol and freeze-dried (Fraction 3). Fraction 1 was mainly composed of MG blocks; fraction 2 which was soluble at pH 2.85 was enriched in MM blocks; and fraction 3 which was insoluble at pH 2.85 was enriched in GG blocks (Leal et al., 2008).
The LLCT consists of a linear positioning stage (Intellistage M-511.5IM, physic Instrument, Karlsruhe, Germany) connected to a cylindrical, moving upper plate (10 mm diameter), a bottom stationary plate is fixed to an automated force compensating balance. Load cell and linear positioning stage are interfaced to a PC for data acquisition and control using LabVIEW 7.1 (Sharma et al., 2011). Movement of top plate (position, h) and force registered (Force, F ) by the load cell were computer-stored for further analysis (Fig. 2). A microscope glass slide was placed on the bottom plate of the LLCT stage and the top plate was moved downwards until it touched the glass surface. This position was recorded as the bottom of the sample (position ¼ 0). Subsequently, a sample was placed on top of the microscope slide and the top plate was moved downwards at a speed of 10 mm/s until it experienced a counter force of 104 N (t ¼ 0 in Fig. 2). This position was recorded too and the sample size was calculated as the height of this position (position ¼ h0 in Fig. 2). The resolution in position, load and time determination for LLCT was 0.1 mm, 2 mg and 25 ms respectively and the velocity of motion was controlled in feedback mode. Before each measurement, the free water from the sample was removed by filter paper. Next, the sample was compressed to 90% of its original size (strain, ε ¼ 0.1) within 0.5 s and the loading force was monitored for 100 s. Position of the top plate and the loading force were continuously recorded during the compression process (Ft and ht in Fig. 2). The data was analyzed in the following manner. The strain was calculated as straint ¼
FT-IR spectroscopic analysis
ðh0 ht Þ h0
(1)
While h0 is the size of the granule, and ht is the position of the top plate at any time t. Next, a plot of force versus strain was created. The stiffness of the polymer was defined as the slope of linear region in the force versus strain curve (Sharma et al., 2011).
2.6.2. 2.5.
59
Maxwell analysis
Viscoelastic response is often used as a probe in polymer science, since it is sensitive to the materials chemistry and
The FT-IR spectra of the isolated ALEs in KBr pellets (98 mg KBr þ 2 mg sample) were recorded in the 4000e400 cm1 region using a Nicolet 670 instrument (Thermo, USA).
2.6.
Viscoelasticity
2.6.1.
Viscoelasticity measurement
Biopolymers exhibit both viscous and elastic characteristics when undergoing deformation. This property is called viscoelasticity. Parameters derived from viscoelasticity measurement were used to describe the physical characteristics (e.g. mechanical strength) of the polymer (Wineman and Rajagopal, 2000). Alginate is a family of polysaccharides that forms gel. The mechanical strength is one important parameter of polysaccharides gel. To make comparison of ALEgranules and ALEflocs, the viscoelasticity of the gel formed by ALEs in CaCl2 was measured by the low load compression tester (LLCT) at room temperature (22 C) according to Sharma et al. (2011).
Fig. 2 e Viscoelasticity measurement by the low load compression tester. Both the position h and the force F are computer monitored, ht and Ft are the position and force at time t.
60
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 5 7 e6 5
microstructure (Tschoegl, 1989). To further understand how the ALE structure affects gel strength, the Maxwell analysis is conducted. In Maxwell model, each element is described as a Hook spring and a Newtonian dashpot connected in series; their viscoelastic response in time can be described by differential equations in time. The spring is visualized as representing the elastic or energetic component of the response, while the dashpot represents the conformational or entropic component (Tschoegl, 1989). Data treatment was as follows: The compressive force F(t) was first converted into a relaxation modulus G(t) by dividing with the applied strain (ε ¼ 0.1). GðtÞ ¼ FðtÞ=0:1
(2)
Subsequently, a generalized Maxwell model was fitted to the measured values of G(t) using the Solver tool in Microsoft Excel 2003. t= s1
GðtÞ ¼ G1 e
t= s2
þ G2 e
t= s3
þ G3 e
t= s4 :::::
þ G4 e
(3)
Where si ¼ hi/Gi, Gi is the spring constant and hi is the viscosity of the ith element and i has the value 1,2,3. The generalized model is composed of various Maxwell elements connected in parallel, each composed of a spring representing the elastic response to the loading force and a dashpot accounting for the viscous response. Each element is characterized by a relaxation time constant (s), which is the ratio of the dashpot viscosity and spring constant for that particular element, and an elastic spring constant (G). The optimum number of Maxwell elements to be used was determined using the Chi-square function expressed by (Eq. (4)). x2 ¼
2 n X Gi Gðti ; s1 ::::::sm Þ i¼1
si
(4)
Where Gi is the measured and G(ti;s1......sm) is the calculated relaxation spring constant at time point ti. si is the standard error of the device which was measure to be 2.76 105 N. Chisquare value was accepted if it was of the same order as the degrees of freedom, i.e. total number of data points minus the number of variables. The relative importance of the ith Maxwell element from the total of n elements was determined using Eq. (4)
Relative importance ðEt Þ ¼ 100
3.
G1 G1 þ G2 þ ::::::Gn
(5)
Results
3.1. The extractable alginate-like exopolysaccharides in aerobic granular sludge and aerobic flocculent sludge Alginates are exopolysaccharides which can be produced by several bacteria genera among which Azotobacter and Pseudomonas (Sutherland, 1990). Alginate-like EPS were one of the major exopolysaccharides in aerobic granules (Lin et al., 2010). To investigate their presence in flocculent sludge, alginatelike EPS were isolated from aerobic flocculent sludge as well. The extractable alginate-like EPS from aerobic flocculent sludge was 72 6 mg/g (VSS ratio), and from aerobic granular sludge was 160 4 mg/g (VSS ratio). Apparently, ALE existed in both aerobic granular sludge and flocculent sludge. The amount of ALE in granular sludge is roughly twice the amount in activated sludge.
3.2.
Gel-formation property of alginate-like EPS in CaCl2
The uniqueness of alginates is that they are able to form gels with multivalent ions (e.g. Ca2þ, Fe3þ), in a broad range of temperature and pH. To check the gel-forming property of alginate-like EPS derived both from flocculent sludge and granular sludge, the same concentration of ALEflocs and ALEgranules were dropped into CaCl2 solution. Interestingly, crosslink of these two kinds of alginate-like EPS with Ca2þ reflected the morphology of granules and flocs, respectively: condensed beads formed immediately once the drops of ALEgranules in contact with CaCl2 solution; while fluffy flocs generated in the case of the ALEflocs (Fig. 3). The water content of these two kinds of Ca-ALE was around 93%, displaying that they were hydrogels. The significant difference in their morphology indicates that these two ALEs have different characteristics.
3.3.
Alginate-like exopolysaccharides fractionation
Alginate molecules consist of two monomers (mannuronic acid, M and guluronic acid, G) (Simdsrød and Draget, 1996).
Fig. 3 e Morphology of alginate-like exopolysaccharides after crosslink with Ca2D under the stereomicroscope. a: alginatelike exopolysaccharides derived from aerobic flocculent sludge in CaCl2 solution; b: alginate-like exopolysaccharides derived from aerobic granular sludge in CaCl2 solution.
61
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 5 7 e6 5
These two monomers form three different kinds of blocks (GG, MG and MM blocks). The percentage of each kind of block is one parameter describing the fine chemical structure of alginate. The recovery yields after fractionation and percentages of these blocks are listed in Table 1. The two ALEs had clearly different fine chemical structures: ALEgranules had significantly higher percentage of GG blocks, but a lower percentage of MG blocks. The monomer ratios of G:M clearly differ from roughly 8:1 for granules and 2:1 for flocs. In addition, despite that both the two isolated polymers had quite low amount of MM blocks, the percentage of MM blocks in ALEflocs was still more than two times higher than in ALEgranules.
3.4.
FT-IR spectroscopy analysis
To identify and compare the block fractions of ALEgranules and ALEflocs, their FT-IR spectra were also recorded.
3.4.1. Similarities in spectra between the two ALEs and their three fractions Not only the spectra of the two ALEs, but also the spectra of their three fractions manifested typical bands of polysaccharides (Fig. 4): a broad rounded absorption band above wavenumber 3000 cm1 assigned to OeH stretching vibrations, CeH stretching peaks at 2800e2974 cm1 and a strong and broad stretch of CeOeC, CeO at 1000e1200 cm1 (Bramhachari et al., 2007). In addition, the bands at 1660e1684 cm1 and 1400e1414 cm1 in those spectra correspond to the asymmetric and symmetric stretching of eCOO vibration in uronic acid residues respectively (Silverstein et al., 1991), confirming that the isolated polymers and their enriched block fractions were salts of uronic acid polysaccharides. Furthermore, despite of their origin (either derived from aerobic granular sludge or from activated sludge), the same fraction (i.e. fraction enriched in GG blocks, MG blocks and MM blocks, respectively) displayed almost identical spectrum. This proves that, both ALEs are indeed composed of three similar building blocks and it is mainly the ratio of these blocks which make the distinction between the gel forming properties of aerobic flocculent and granular sludge.
3.4.2. Differences in the spectra between two ALEs and their fractions Spectrum of ALEgranules almost completely resembled the spectrum of their GG fraction, verifying that GG blocks were the dominant component in this ALE. In contrast, the spectrum of ALEflocs was a combination of the spectra of the three
blocks more than resembling a single fraction, which is in agreement with the result of chemical fractionation indicating no single fraction significantly dominant over the other fractions.
3.5. gel
Viscoelasticity of alginate-like exopolysaccharides
Normally, Ca2þ is one of the common divalent cations in wastewater; in addition, it easily binds with polysaccharides (e.g. alginates, polygalacturonic acid) as a counter ion mediating gelation (De Kerchove and Elimelech, 2007). It was found in this research that Ca-ALEs reflect the main morphology of aerobic granules and activated sludge flocs, respectively. This indicates that the gels formed between ALE and Ca2þ could indeed be the main form of ALE in both kinds of sludge. Alginate gels are hydrogels. They are neither pure solids (which display only elasticity), nor pure liquids (which display only viscosity). They simultaneously show viscous and elastic behaviors, thus they are viscoelastic in nature (Donati and Paoletti, 2009). The mechanical strength of a viscoelastic material (e.g. alginateecalcium gel) can be quantitatively determined by the stress relaxation tests, in which, a sample is compressed at a controlled speed to a desired strain, this strain is maintained constant while the stress required to maintain this constant strain is measured as a function of time (Fig. 5a) (Del Nobile et al., 2007). In order to determine the difference in the mechanical strength of Ca-ALEgranules and Ca-ALEflocs, stress relaxation tests were conducted using a low load compression tester. The force-time and strainetime curves are shown in Fig. 5a. Within the first 0.5 s, the Ca-ALEs was compressed to 90% of their original height (strain, ε ¼ 0.1), and then the strain was held constant at 0.1 for 100 s to allow force relaxation. The forceestrain curve within the first 0.5 s was plotted in Fig. 5b. The force required at lower strains (up to 0.02) was extremely low and dominated by the adhesive interactions at the interface between Ca-ALEs and the top plate (Johnson et al., 1971). Between the strain 0.02 and 0.1, a linear increase in loading force was visible as a function of strain for both Ca-ALEgranules and Ca-ALEflocs, the slope of this forceestrain curve was defined as the stiffness of the material, i.e. resistance of the material to deformation. Ca-ALEgranules was found to be 6 times stiffer than Ca-ALEflocs, judging from the slope of 31.5 8.5 mN as compared to the slope of 5.3 1.3 mN of Ca-ALEflocs. This means that Ca-ALEgranules is more difficult to deform. When the strain was kept constant at 0.1, the force required to maintain this constant strain decreased with time, faster in the beginning and then gradually to a non-zero
Table 1 e Fraction of different building blocks obtained by partial hydrolysis of alginate-like exopolysaccharides isolated from both aerobic granular sludge and conventional aerobic flocculent sludge. Recovery is given as percentage recovered from original ALE amount. Sludge source Granular sludge Flocculent sludge
Fractionation recovery (%)
Poly(guluronic acid) blocks (GG) %
Poly(mannuronic acid) blocks (MM) %
Heteropolymeric blocks (MG) %
85.4 7.2 73.1 8.0
69.1 8.9 35.3 8.4
2.1 1.4 5.7 2.4
14.6 2.3 32.4 1.6
62
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 5 7 e6 5
Fig. 4 e FT-IR spectra of alginate-like exopolysaccharides isolated from aerobic granular sludge and flocculent sludge.
residual value for Ca-ALEgranules. This is a typical behavior of a viscoelastic solid: a certain amount of internal force is required to maintain a constant strain (Wineman and Rajagopal, 2000). This behavior was in accordance with findings by Mancini et al. (1999), the well cross-linked alginate gels
demonstrate an asymptotically decreasing trend in stress relaxation. Ca-ALEflocs showed different trend where, the force decreased continuously and attained a value of zero in about 19.4 s. This is a typical behavior of a viscoelastic fluid (Wineman and Rajagopal, 2000).
Fig. 5 e Data acquisition and analysis from the low load compression tester. (a) Raw output of force and strain as a function of time (b) Derivation of stiffness from the linear region where deformation is imposed on the Ca-ALEs. Derivation of the viscoelastic parameters from the relaxation part of Ca-ALEflocs (c) and Ca-ALEgranules (d) force-time curve by fitting a generalized Maxwell model.
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 5 7 e6 5
To further understand how the ALE micro-structure affect gel strength, the force-time data at relaxation region was fitted with a generalized Maxwell model (Fig. 5c and d). The optimum number of Maxwell elements chosen to fit the experimental data is obtained by minimizing the Chi square and the number of the considered fitting parameters. The model fitting parameters are reported in Table 2. It is evident that, two Maxwell elements are required to satisfactorily fit the experimental data of Ca-ALEflocs. The first element relaxes extremely fast, with the relaxation time s 1 s; the second element relaxes slowly, with the relaxation time 1 < s 10 s. In contrast, three Maxwell elements are required to fit the experimental data of Ca-ALEgranules. Besides the two elements that fell in the same ranges of relaxation time of Ca-ALEflocs (s 1 s and 1 s < s 10 s), a third element, which has the highest importance (E ¼ 59.5%) and the longest relaxation time (s > 10 s) has to be presented as well. The decrease of stress at constant strain is called stress relaxation. As ideal elastic materials (ideal solids) do not relax (s / N) whereas ideal viscous materials (ideal liquids) instantaneously relax (s / 0), the viscoelastic materials display relaxation in between (0 < s < N) (Steffe, 1992). It is demonstrated from the relaxation time derived by Maxwell model that, both Ca-ALEgranules and Ca-ALEflocs have viscoelastic property. However, Ca-ALEgranules is much more elastic than Ca-ALEflocs due to the presence of the third element, which is the most important element with the highest spring constant (representing elasticity) and longest relaxation time, implying that Ca-ALEgranules is well cross-linked gel (Riande et al., 2000). Therefore, in comparison with Ca-ALEflocs, Ca-ALEgranules has not only higher stiffness but is also well cross-linked.
4.
63
4.1. Influence of the fine chemical structure on gelforming capacity of ALE Alginate is composed of three different blocks, GG, MG and MM blocks. These three kinds of blocks are distributed randomly in the molecular chain of alginate. As each kind of block has its specific property, alginate with different fine chemical structure manifests a different behavior. From this point of view, alginate is indeed a family of polymers consists of same monomers, but with different chemical structures (Davis et al., 2003). Alginate is able to form a gel matrix. The three different kinds of blocks contribute unequally to the gel-forming property of alginate: GG blocks have higher affinity towards divalent cations than the other two blocks (Braccini et al., 1999); the order of gel-forming capability of the three blocks is: MM blocks MG blocks << GG blocks (Smidrød, 1974). Therefore the content of GG blocks is the most important factor for gel formation. In the chemical structure of ALEgranules, about 69% was GG blocks. If the amount of the ALE in granules is also considered (16% w/w), the GG rich ALEgranules benefited the granules with the formation of a compact gel structure (Fig. 3b). In contrast, there are less GG blocks in ALEflocs, which results to a weaker affinity to divalent ions and limits the polymer’s gel-forming capability. Furthermore, the percentage of MG blocks is almost equal to GG blocks in ALEflocs. As flexibility of the three kinds of blocks increase in the order: GG blocks < MM blocks < MG blocks (Melvik and Dornish, 2004), larger amount of MG blocks increase flexibility of alginate molecules. Consequently, the ALEflocs with comparatively less GG blocks and more MG blocks endowes flocculent sludge with a flexible and loose structure (Fig. 3a).
4.2. Influence of the fine chemical structure on the mechanical property of ALE gel
Discussion
The alginate-like exopolysaccharides were isolated both from activated sludge and aerobic granular sludge. The quantity of ALE in aerobic granular sludge is two times higher than in the flocculent activated sludge. These two ALEs manifested different gel forming property, chemical composition and mechanical stiffness. The distinguished chemical composition is the fundamental cause for the differences in gel formation and mechanical properties.
The stiffness of alginate gel is affected by the amount of crosslink points, the concentration and composition of alginate (Donati and Paoletti, 2009). The three kinds of blocks in
Table 2 e Generalized Maxwell parameters of CaALEgranules and Ca-ALEflocs. Spring constants (G), relaxation time constants (s), and relative importance (E ) of Maxwell elements. Ca-ALEgranules
Ca-ALEflocs G (mN) E (%) s (s)
G1 2.8 0.1 51.7 2.0
G2 2.6 0.3 48.3 3.0
G1 9.4 0.5 28.1 0.8
G2 10.3 0.7 30.7 1.6
G3 13.9 2.4 59.5 2.8
s1 0.4 0.03
s2 6.1 1.3
s1 1.0 0.8
s2 9.6 2.9
s3 77.2 8.0
Fig. 6 e Crosslink between alginate-like exopolysaccharides and Ca2D. a: strong gel structure formed between alginate-like exopolysaccharides derived from aerobic granules and Ca2D; b: weak structure formed between alginate-like exopolysaccharides derived from aerobic flocculent sludge and Ca2D.
64
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 5 7 e6 5
alginate molecules have substantially different geometries. GG blocks are buckled. If two GG blocks are aligned side by side, diamond shaped holes result. Their dimensions are ideal for the crosslink with divalent cations (especially Ca2þ). This is the so-called “egg-box” model (Christensen et al., 1990) (Fig. 6). Thus, in alginate gel, the “egg-boxes” formed by GG blocks are the crosslink points, they decide the stiffness of the gel. The large numbers of GG blocks in Ca-ALEgranules contributes to its higher resistance to deformation. Alginate gels consist of more than 90% water in a supporting matrix (or network) of crosslinked polymer chains (Nussinovitch, 1997). In function, the polymer network resembles a sponge with water holding in the pores of the matrix. When a rapid external compression force is applied to Ca-ALEs at the start of relaxation tests, it causes deformation of the sponge structure by reducing the pore size and reorientation the molecular linkage from the original configuration (Wineman and Rajagopal, 2000). Decrease of the pore size results in a build-up of hydraulic pressure. The difference of hydraulic pressures in different pores might force water to migrate, therefore, water migration could be the fastest (s 1 s) element in Maxwell model in both Ca-ALEs. Ca-MG and Ca-MM are comparatively weak and flexible (Fig. 6). Their structures can easily reoriented, thus the second element with 1 s< s 10 s could be attributed to the weak structure of (Ca-MG þ Ca-MM) in both Ca-ALEs. The third element of Ca-ALEgranules with s > 10 s could be regarded as Ca-GG, because of the well crosslinked structure (Fig. 6), the reorientation is limited. A force is always required to maintain it in a distorted state. Thanks to the presence of this element, the stress-relaxation of Ca-ALEgranules displays as viscoelastic solid within the 100 s. In the case of Ca-ALEflocs, due to its lower GG content, a well crosslinked structure could not be formed; a reorientation of GG-Ca could therefore be not necessary with the applied deformation. The stress-relaxation curve of Ca-ALEflocs displays a viscoelastic liquid behavior due to a lack of the GG-Ca element. Therefore, when a same external force is applied to both Ca-ALEflocs and Ca-ALEgranules, water is going to migrate and the flexible part quickly deforms. As a continuous network structure is present in Ca-ALEgranules by crosslink of GG blocks with Ca2þ, deformation of the flexible part hardly causes the whole gel structure to break. However, Ca-ALEflocs is less crosslinked and lacks the third element, deformation of the flexible part will induce a continuous movement of ALE molecules, eventually resulting in a damage of the whole CaALEflocs structure. If Ca-ALE is the main form of ALE in activated sludge flocs and aerobic granules, evidently, the existence of Ca-ALEgranules endows aerobic granules a stronger mechanical property.
5.
Conclusions
The fine chemical structure of alginate-like exopolysaccharides isolated from aerobic flocculent sludge and aerobic granular sludge are different. Alginate-like exopolysaccharides derived from aerobic granules had significantly higher amount of poly(gluronic acid) blocks but lower amount of poly(guluronic acidmanuronic acid) blocks, while alginate-like exopolysaccharides
derived from activated sludge has an equal amount of poly(gluronic acid) blocks and poly(guluronic acid-manuronic acid) blocks. These differences result in a perfect gel-forming capability of alginate-like exopolysaccharides derived from aerobic granules and bestowed this exopolysaccharides gel a stronger mechanical property as compared to alginate-like exopolysaccharides derived from flocculent sludge, eventually contribute to the distinguished characteristics between aerobic granular sludge and flocculent sludge.
Acknowledgments This research was supported by a Marie Curie Intra European Fellowship (PIEF-GA-2009-253096) within the 7th European Community Framework Programme. We thank DHV, The Netherlands for providing aerobic granular sludge and flocculent activated sludge.
references
Adav, S.S., Lee, D.J., Show, K.Y., Tay, J.H., 2008. Aerobic granular sludge: recent advances. Biotechnology Advances 25, 411e423. APHA, 1998. Standard Methods for the Examination of Water and Wastewater, twentieth ed. American Public Health Association, Washington, DC. Bramhachari, P.V., Kavikishor, P.B., Ramadevi, R., Kumar, R., Rao, B.R., Dubey, S.K., 2007. Isolation and characterization of mucous exopolysaccharide (EPS) produced by Vibrio furnissii strain VB0S3. Journal of Microbiology and Biotechnology 17, 14e51. Braccini, I., Grasso, R.P., Perez, S., 1999. Conformational and configurational features of acidic polysaccharides and their interactions with calcium ions: a molecular modeling investigation. Carbohydrate Research 317, 119e130. Bruus, J.H., Nielsen, P.H., Keiding, K., 1992. On the stability of activated sludge flocs with implications to dewatering. Water Research 26, 1597e1604. Christensen, B.E., Indergaard, M., Smidsrød, O., 1990. Polysaccharide research in Trondheim. Carbohydrate Polymers 13, 239e255. Davis, T.A., Volesky, B., Mucci, A., 2003. A review of the biochemistry of heavy metal biosorption by brown algae. Water Research 37, 4311e4317. De Kerchove, A.J., Elimelech, M., 2007. Impact of alginate conditioning film on deposition kinetics of motile and nonmotile Pseudomonas aeruginosa strains. Applied and Environmental Microbiology 73, 5227e5234. Del Nobile, M.A., Chillo, S., Mentana, A., Baiano, A., 2007. Use of the generalized Maxwell model for describing the stress relaxation behavior of solid-kike foods. Journal of Food Engineering 78, 978e983. Donati, I., Paoletti, S., 2009. Material properties of alginates. Biology and Applications 13, 1e53. Draget, K.I., Strand, B., Hartmann, M., Valla, S., Smidsrød, O., Skja˚k-Bræk, G., 2000. Ionic and acid gel formation of epimerized alginates; the effect of AlgE4. International Journal of Biological Macromolecules 27, 117e122. Hampson, F.C., Farndale, A., Strugala, V., Sykes, J., Jolliffe, I.G., Dettmar, P.W., 2005. Alginates rafts and their characterisation. International Journal of Pharmaceutics 294 (1e2), 137e147.
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 5 7 e6 5
Johnson, K.L., Kendall, K., Roberts, A.D., 1971. Surface energy and the contact of elastic solids. Proceedings of the Royal Society A324, 301e320. Leal, D., Matsuhiro, B., Rossi, M., Caruso, F., 2008. FT-IR spectra of alginic acid block fractions in three species of brown seaweeds. Carbohydrate Research 343, 308e316. Lin, Y.M., de Kreuk, M.K., van Loosdrecht, M.C.M., Adin, A., 2010. Characterization of alginate-like exopolysaccharides isolated from aerobic granular sludge in pilot plant. Water Research 44, 3355e3364. McSwain, B.S., Irvine, R.L., Hausner, M., Wilderer, P.A., 2005. Composition and distribution of extracellular polymeric substances in aerobic flocs and granular sludge. Applied and Environmental Microbiology 71 (2), 1051e1057. Mancini, F., Moresi, M., Rancini, R., 1999. Mechanical properties of alginate gels: empirical characterisation. Journal of Food Engineering 39, 369e371. Melvik, J.E., Dornish, M., 2004. Alginate as a carrier for cell immobilization. In: Nedovic, V., Willaert, R. (Eds.), Fundamentals of Cell Immobilization Biotechnology. Kluwer Academic Publishers. Nussinovitch, A., 1997. Hydrocolloid Applications Gum Technology in the Food and Other Industries. Blackie Academic & Professional, London, pp. 292e311. Riande, E., Diaz-Calleja, R., Prolongo, M.G., Masegosa, R.M., Salom, C., 2000. Polymer Viscoelasticity: Stress and Strain in Practice. Marcel Decker, New York, p. 879. Seviour, T., Pijuan, M., Nicholson, T., Keller, J., Yuan, Z., 2009a. Understanding the properties of aerobic sludge granules as
65
hydrogels. Biotechnology and Bioengineering 102 (5), 1483e1493. Seviour, T., Pijuan, M., Nicholson, T., Keller, J., Yuan, Z., 2009b. Gel-forming exopolysaccharides explain basic differences between structures of aerobic sludge granules and floccular sludges. Water Research 43, 4469e4478. Sharma, P.K., busscher, H.J., Terwee, T., Koopmans, S.A., van Kooten, T.G., 2011. A comparative study on the viscoelastic properties. Experimental Eye Research 93 (5), 681e688. Silverstein, R.M., Clayton Bassier, G., Morrill, T.C., 1991. Spectrometric Identification of Organic Compounds. Willey, New York. Smidrød, O., 1974. Molecular basis for some physical properties of alginates in the gel state. Faraday Discussions of the Chemical Society 57, 263e274. Simdsrød, O., Draget, K.I., 1996. Chemistry and physical properties of alginates. Carbohydrate in Europe 14, 6e13. Steffe, J.F., 1992. Rheological Methods in Food Process Engineering. Freeman Press, East Lansing, Mich. USA. Sutherland, I.W., 1990. Biotechnology of Microbial Exopolysaccharides. Cambridge University Press, Cambridge. Sutherland, I.W., 2001. Biofilm exopolysaccharides: a strong and sticky framework. Microbiology 147, 3e9. Tschoegl, N.W., 1989. The Phenomenological Theory of Linear Viscoelastic Behavior. Springer-Verlag, Heidelberg. Wineman, A.S., Rajagopal, K.R., 2000. Mechanical Response of Polymers: An Introduction. Cambridge University Press, UK.
Available online at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/watres
The chemical and mechanical differences between alginate-like exopolysaccharides isolated from aerobic flocculent sludge and aerobic granular sludge Y.M. Lin a,*, P.K. Sharma b, M.C.M. van Loosdrecht a a
Department of Biotechnology, Delft University of Technology, Julianalaan 67, 2628BC Delft, The Netherlands Department of Biomedical Engineering, University Medical Center Groningen, University of Groningen, P.O. Box 196, 9700 AD Groningen, The Netherlands b
article info
abstract
Article history:
This study aimed to investigate differences in the gel matrix of aerobic granular sludge and
Received 7 June 2012
normal aerobic flocculent sludge. From both types of sludge that fed with the same
Received in revised form
municipal sewage, the functional gel-forming exopolysaccharides, alginate-like exopoly-
21 August 2012
saccharides, were isolated. These two exopolysaccharides were chemically fractionated,
Accepted 6 September 2012
and investigated by FT-IR spectroscopy. The isolated polymers were made into a gel by
Available online 5 October 2012
calcium addition and the mechanical properties of these reconstituted gels were measured by a low load compression tester. The viscoelastic behavior of the gels was described by
Keywords:
a generalized Maxwell model. The alginate-like exopolysaccharides derived from aerobic
Exopolysaccharides
granules had significantly higher amount of poly(guluronic acid) blocks but lower amount
Gel-forming property
of poly(guluronic acid-manuronic acid) blocks in the chemical structure, while the
Aerobic granular sludge
alginate-like exopolysaccharides derived from aerobic flocculent sludge had equal amount
Activated sludge
of poly(guluronic acid) blocks and poly(guluronic acid-manuronic acid) blocks. These differences result in a perfect gel-forming capability of alginate-like exopolysaccharides derived from aerobic granules and bestowed this exopolysaccharides gel a stronger mechanical property as compared to alginate-like exopolysaccharides derived from aerobic flocculent sludge. The different chemical and mechanical properties of these two exopolysaccharides contributed to the distinguished characteristics between aerobic granular sludge and aerobic flocculent sludge. ª 2012 Elsevier Ltd. All rights reserved.
1.
Introduction
Exopolysaccharides are recognized as key elements that shape and provide structural support for biofilms (Sutherland, 2001). Likewise, they are highly involved in aerobic flocculent
sludge and aerobic granular sludge matrix structure (Seviour et al., 2009a, 2009b). Some efforts have been made to understand how and why these two microbial aggregates differed, by comparing the quantity of exopolysaccharides in the isolated extracellular polymers from these two microbial
Abbreviations: FT-IR, Fourier transform infrared; G, Guluronic acid; GG, Poly(guluronic acid); M, Mannuronic acid; MM, Poly(mannuronic acid); MG, Alternating mannuronic and guluronic acid; ALE, Alginate-like exopolysaccharides; ALEflocs, Alginate-like exopolysaccharides derived from aerobic flocculent sludge; ALEgranules, Alginate-like exopolysaccharides derived from aerobic granular sludge. * Corresponding author. E-mail addresses: [email protected] (Y.M. Lin), [email protected] (P.K. Sharma), [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.2012.09.017
58
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 5 7 e6 5
aggregates respectively (Adav et al., 2008; McSwain et al., 2005). However, research focusing on isolating and characterizing the functional exopolysaccharides by comparing their physical and chemical properties, and investigating the contribution of these properties on the difference between aerobic flocculent sludge and aerobic granules is just started very recently. Based on identifying the crossover between storage modulus relative to the loss modulus by rheological measurement, a study displayed that the exopolysaccharides isolated from aerobic granules were able to form gels at pH 6e8.5, while the exopolysaccharides isolated from flocs couldn’t. Thereafter, difference in gel-forming capability of exopolysaccharides at different pH was considered as a distinctive property between granules and flocs (Seviour et al., 2009b). However, explanation for the different gelforming capability based on exopolysaccharides structure was not provided. In our previous research, one of the major exopolysaccharides extracted from aerobic granules resembled commercial alginate but differed in the behavior in a few chemical reactions (Lin et al., 2010). These polymers were denoted as alginate-like exopolysaccharides (EPS). On the other hand, evidence also existed that an alginate-like crosslink of exopolymers was important for the aerobic flocculent sludge structure (Bruus et al., 1992). Therefore, alginate-like EPS could be involved in the basic structures of both granular and flocculent sludge. It is worth pointing out that alginate is not a single polymer, but a family of polymers with different chemical structures. The gel-forming capability of alginate was decided by their chemical structures (Draget et al., 2000). Therefore, aiming at differentiating aerobic granular sludge from aerobic flocculent sludge, a hypothesis is put forward in the present research: the alginate-like exopolysaccharides in aerobic granular sludge and aerobic flocculent sludge are dissimilar in their chemical structures, and this fundamental difference decides the gel-forming capability of exopolysaccharides and mechanical property of the gel. To test the hypothesis, alginate-like exopolysaccharides were isolated both from aerobic granular sludge from a pilot plant and aerobic flocculent sludge from a wastewater treatment plant fed with the same municipal sewage. These two exopolysaccharides were chemically fractionated, investigated by FT-IR spectroscopy, and compared. The mechanical property of the gel formed after alginate-like exopolysaccharides crosslink with Ca2þ was measured by a low load compression tester. The viscoelastic behavior of the gels was described by a generalized Maxwell model.
2.
Average parameters of the influent were: CODtotal 585 mg/L, suspended solids 195 mg/L, NH4eN 55 mg/L and PO4eP 6.3 mg/L. The reactor was operated in Sequencing Batch (SBR) mode for biological phosphate and nitrogen removal. Operational details were described in Lin et al. (2010). After start-up, biomass concentration in the reactor was maintained around 8e10 g TSS/L. Oxygen in the reactor was controlled between 2 and 3 mg/ L during aeration. Temperature and pH were not controlled in this system and depended on the incoming sewage. Aerobic granular sludge was collected during steady operation of the reactor. Aerobic flocculent sludge of the same wastewater treatment plant was collected from the aeration basin.
2.2.
Alginate-like exopolysaccharides (ALE) extraction
Alginate-like exopolysaccharides (ALE) were extracted from both aerobic granular sludge and activated sludge according to Lin et al. (2010). Dried biomass (0.5 g) was extracted in 0.2 M Na2CO3 at 80 C. After centrifugation at 15,000 rpm for 20 min, the pellet was discarded. The supernatant pH was adjusted to 2 by adding 0.1 M HCl. The precipitate was collected by centrifugation (15,000 rpm, 30 min), washed by di-deionized water until effluent pH reached 7, and dissolved in 0.1 M NaOH. The ALE in the supernatant was precipitated by the addition of cold absolute ethanol to a final concentration of 80% (vol/vol). The precipitate was collected by centrifugation (15,000 rpm, 30 min), washed three times in absolute ethanol and lyophilized. Ash content of the extracted exopolysaccharides was measured according to the standard method (APHA, 1998).
2.3.
Gel formation property in CaCl2
To test ALE’s gel formation property in CaCl2, ALE-miliQ water solution (10 ml, 2% (w/v)) was added drop wise into CaCl2 solution (150 ml, 2% (w/v)) using a peristaltic pump through a syringe needle (0.9 40 mm) (Fig. 1). The ALEs were crosslinked in the CaCl2 solution at room temperature for 16e18 h.
Materials and methods
2.1. Activated sludge and aerobic granular sludge for investigation Granules were sampled from the Nereda pilot plant, operated by DHV at the wastewater treatment plant Epe, The Netherlands. The reactor was fed with municipal sewage. The influent consisted of approximately 25% of slaughterhouse wastewater, which was discharged in the sewage system.
Fig. 1 e Making alginate-like exopolysaccharides gel by dropping 2% alginate-like exopolysaccharides sodium solution into 2% CaCl2 solution.
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 5 7 e6 5
The morphology of Ca-ALEs was photographed by using an Olympus SZH10 stereomicroscope. The water content of both Ca-ALEs was measured by drying the samples at 105 C and recording the loss of the weight.
2.4. Fractionation of the isolated alginate-like exopolysaccharides by partial acid hydrolysis Alginates are characterized as a family of copolymers comprised of mannuronic acid (M) and guluronic acid (G) units arranged in an irregular block pattern of varying proportions of GG, MG and MM blocks. The amount of these three blocks in alginate molecules is considered as the fine chemical structure of alginate polymer. This parameter determines the alginate properties, especially the gel-forming behavior (Hampson et al., 2005). To study the fine chemical structures of ALE derived from granules (ALEgranules) and from flocs (ALEflocs), both the isolated ALEs were fractionated according to Leal et al. (2008). This partial hydrolysis method is able to separate fractions enriched in different kind of blocks (Leal et al., 2008). ALE (0.250 g) were dissolved into 9 ml di-deionized water. After the addition of 1 ml 3.0 M HCl, it was heated at 100 C for 0.5 h. After cooling, the mixture was centrifuged (15,000 rpm) and the supernatant solution was neutralized with 1.0 M NaOH and poured over 80 ml of ethanol, yielding a precipitate that was dissolved in di-deionized water and freeze-dried (Fraction 1). The insoluble material was dissolved in 1.0 M NaOH; the pH was decreased to 2.85 by the addition of 1.0 M HCl. The soluble fraction was neutralized with 1.0 M NaOH and precipitated by the addition of ethanol to a final concentration of 80% (vol/vol) and freeze-dried (Fraction 2). The insoluble fraction was dissolved by neutralization with 1.0 M NaOH, precipitated by ethanol and freeze-dried (Fraction 3). Fraction 1 was mainly composed of MG blocks; fraction 2 which was soluble at pH 2.85 was enriched in MM blocks; and fraction 3 which was insoluble at pH 2.85 was enriched in GG blocks (Leal et al., 2008).
The LLCT consists of a linear positioning stage (Intellistage M-511.5IM, physic Instrument, Karlsruhe, Germany) connected to a cylindrical, moving upper plate (10 mm diameter), a bottom stationary plate is fixed to an automated force compensating balance. Load cell and linear positioning stage are interfaced to a PC for data acquisition and control using LabVIEW 7.1 (Sharma et al., 2011). Movement of top plate (position, h) and force registered (Force, F ) by the load cell were computer-stored for further analysis (Fig. 2). A microscope glass slide was placed on the bottom plate of the LLCT stage and the top plate was moved downwards until it touched the glass surface. This position was recorded as the bottom of the sample (position ¼ 0). Subsequently, a sample was placed on top of the microscope slide and the top plate was moved downwards at a speed of 10 mm/s until it experienced a counter force of 104 N (t ¼ 0 in Fig. 2). This position was recorded too and the sample size was calculated as the height of this position (position ¼ h0 in Fig. 2). The resolution in position, load and time determination for LLCT was 0.1 mm, 2 mg and 25 ms respectively and the velocity of motion was controlled in feedback mode. Before each measurement, the free water from the sample was removed by filter paper. Next, the sample was compressed to 90% of its original size (strain, ε ¼ 0.1) within 0.5 s and the loading force was monitored for 100 s. Position of the top plate and the loading force were continuously recorded during the compression process (Ft and ht in Fig. 2). The data was analyzed in the following manner. The strain was calculated as straint ¼
FT-IR spectroscopic analysis
ðh0 ht Þ h0
(1)
While h0 is the size of the granule, and ht is the position of the top plate at any time t. Next, a plot of force versus strain was created. The stiffness of the polymer was defined as the slope of linear region in the force versus strain curve (Sharma et al., 2011).
2.6.2. 2.5.
59
Maxwell analysis
Viscoelastic response is often used as a probe in polymer science, since it is sensitive to the materials chemistry and
The FT-IR spectra of the isolated ALEs in KBr pellets (98 mg KBr þ 2 mg sample) were recorded in the 4000e400 cm1 region using a Nicolet 670 instrument (Thermo, USA).
2.6.
Viscoelasticity
2.6.1.
Viscoelasticity measurement
Biopolymers exhibit both viscous and elastic characteristics when undergoing deformation. This property is called viscoelasticity. Parameters derived from viscoelasticity measurement were used to describe the physical characteristics (e.g. mechanical strength) of the polymer (Wineman and Rajagopal, 2000). Alginate is a family of polysaccharides that forms gel. The mechanical strength is one important parameter of polysaccharides gel. To make comparison of ALEgranules and ALEflocs, the viscoelasticity of the gel formed by ALEs in CaCl2 was measured by the low load compression tester (LLCT) at room temperature (22 C) according to Sharma et al. (2011).
Fig. 2 e Viscoelasticity measurement by the low load compression tester. Both the position h and the force F are computer monitored, ht and Ft are the position and force at time t.
60
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 5 7 e6 5
microstructure (Tschoegl, 1989). To further understand how the ALE structure affects gel strength, the Maxwell analysis is conducted. In Maxwell model, each element is described as a Hook spring and a Newtonian dashpot connected in series; their viscoelastic response in time can be described by differential equations in time. The spring is visualized as representing the elastic or energetic component of the response, while the dashpot represents the conformational or entropic component (Tschoegl, 1989). Data treatment was as follows: The compressive force F(t) was first converted into a relaxation modulus G(t) by dividing with the applied strain (ε ¼ 0.1). GðtÞ ¼ FðtÞ=0:1
(2)
Subsequently, a generalized Maxwell model was fitted to the measured values of G(t) using the Solver tool in Microsoft Excel 2003. t= s1
GðtÞ ¼ G1 e
t= s2
þ G2 e
t= s3
þ G3 e
t= s4 :::::
þ G4 e
(3)
Where si ¼ hi/Gi, Gi is the spring constant and hi is the viscosity of the ith element and i has the value 1,2,3. The generalized model is composed of various Maxwell elements connected in parallel, each composed of a spring representing the elastic response to the loading force and a dashpot accounting for the viscous response. Each element is characterized by a relaxation time constant (s), which is the ratio of the dashpot viscosity and spring constant for that particular element, and an elastic spring constant (G). The optimum number of Maxwell elements to be used was determined using the Chi-square function expressed by (Eq. (4)). x2 ¼
2 n X Gi Gðti ; s1 ::::::sm Þ i¼1
si
(4)
Where Gi is the measured and G(ti;s1......sm) is the calculated relaxation spring constant at time point ti. si is the standard error of the device which was measure to be 2.76 105 N. Chisquare value was accepted if it was of the same order as the degrees of freedom, i.e. total number of data points minus the number of variables. The relative importance of the ith Maxwell element from the total of n elements was determined using Eq. (4)
Relative importance ðEt Þ ¼ 100
3.
G1 G1 þ G2 þ ::::::Gn
(5)
Results
3.1. The extractable alginate-like exopolysaccharides in aerobic granular sludge and aerobic flocculent sludge Alginates are exopolysaccharides which can be produced by several bacteria genera among which Azotobacter and Pseudomonas (Sutherland, 1990). Alginate-like EPS were one of the major exopolysaccharides in aerobic granules (Lin et al., 2010). To investigate their presence in flocculent sludge, alginatelike EPS were isolated from aerobic flocculent sludge as well. The extractable alginate-like EPS from aerobic flocculent sludge was 72 6 mg/g (VSS ratio), and from aerobic granular sludge was 160 4 mg/g (VSS ratio). Apparently, ALE existed in both aerobic granular sludge and flocculent sludge. The amount of ALE in granular sludge is roughly twice the amount in activated sludge.
3.2.
Gel-formation property of alginate-like EPS in CaCl2
The uniqueness of alginates is that they are able to form gels with multivalent ions (e.g. Ca2þ, Fe3þ), in a broad range of temperature and pH. To check the gel-forming property of alginate-like EPS derived both from flocculent sludge and granular sludge, the same concentration of ALEflocs and ALEgranules were dropped into CaCl2 solution. Interestingly, crosslink of these two kinds of alginate-like EPS with Ca2þ reflected the morphology of granules and flocs, respectively: condensed beads formed immediately once the drops of ALEgranules in contact with CaCl2 solution; while fluffy flocs generated in the case of the ALEflocs (Fig. 3). The water content of these two kinds of Ca-ALE was around 93%, displaying that they were hydrogels. The significant difference in their morphology indicates that these two ALEs have different characteristics.
3.3.
Alginate-like exopolysaccharides fractionation
Alginate molecules consist of two monomers (mannuronic acid, M and guluronic acid, G) (Simdsrød and Draget, 1996).
Fig. 3 e Morphology of alginate-like exopolysaccharides after crosslink with Ca2D under the stereomicroscope. a: alginatelike exopolysaccharides derived from aerobic flocculent sludge in CaCl2 solution; b: alginate-like exopolysaccharides derived from aerobic granular sludge in CaCl2 solution.
61
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 5 7 e6 5
These two monomers form three different kinds of blocks (GG, MG and MM blocks). The percentage of each kind of block is one parameter describing the fine chemical structure of alginate. The recovery yields after fractionation and percentages of these blocks are listed in Table 1. The two ALEs had clearly different fine chemical structures: ALEgranules had significantly higher percentage of GG blocks, but a lower percentage of MG blocks. The monomer ratios of G:M clearly differ from roughly 8:1 for granules and 2:1 for flocs. In addition, despite that both the two isolated polymers had quite low amount of MM blocks, the percentage of MM blocks in ALEflocs was still more than two times higher than in ALEgranules.
3.4.
FT-IR spectroscopy analysis
To identify and compare the block fractions of ALEgranules and ALEflocs, their FT-IR spectra were also recorded.
3.4.1. Similarities in spectra between the two ALEs and their three fractions Not only the spectra of the two ALEs, but also the spectra of their three fractions manifested typical bands of polysaccharides (Fig. 4): a broad rounded absorption band above wavenumber 3000 cm1 assigned to OeH stretching vibrations, CeH stretching peaks at 2800e2974 cm1 and a strong and broad stretch of CeOeC, CeO at 1000e1200 cm1 (Bramhachari et al., 2007). In addition, the bands at 1660e1684 cm1 and 1400e1414 cm1 in those spectra correspond to the asymmetric and symmetric stretching of eCOO vibration in uronic acid residues respectively (Silverstein et al., 1991), confirming that the isolated polymers and their enriched block fractions were salts of uronic acid polysaccharides. Furthermore, despite of their origin (either derived from aerobic granular sludge or from activated sludge), the same fraction (i.e. fraction enriched in GG blocks, MG blocks and MM blocks, respectively) displayed almost identical spectrum. This proves that, both ALEs are indeed composed of three similar building blocks and it is mainly the ratio of these blocks which make the distinction between the gel forming properties of aerobic flocculent and granular sludge.
3.4.2. Differences in the spectra between two ALEs and their fractions Spectrum of ALEgranules almost completely resembled the spectrum of their GG fraction, verifying that GG blocks were the dominant component in this ALE. In contrast, the spectrum of ALEflocs was a combination of the spectra of the three
blocks more than resembling a single fraction, which is in agreement with the result of chemical fractionation indicating no single fraction significantly dominant over the other fractions.
3.5. gel
Viscoelasticity of alginate-like exopolysaccharides
Normally, Ca2þ is one of the common divalent cations in wastewater; in addition, it easily binds with polysaccharides (e.g. alginates, polygalacturonic acid) as a counter ion mediating gelation (De Kerchove and Elimelech, 2007). It was found in this research that Ca-ALEs reflect the main morphology of aerobic granules and activated sludge flocs, respectively. This indicates that the gels formed between ALE and Ca2þ could indeed be the main form of ALE in both kinds of sludge. Alginate gels are hydrogels. They are neither pure solids (which display only elasticity), nor pure liquids (which display only viscosity). They simultaneously show viscous and elastic behaviors, thus they are viscoelastic in nature (Donati and Paoletti, 2009). The mechanical strength of a viscoelastic material (e.g. alginateecalcium gel) can be quantitatively determined by the stress relaxation tests, in which, a sample is compressed at a controlled speed to a desired strain, this strain is maintained constant while the stress required to maintain this constant strain is measured as a function of time (Fig. 5a) (Del Nobile et al., 2007). In order to determine the difference in the mechanical strength of Ca-ALEgranules and Ca-ALEflocs, stress relaxation tests were conducted using a low load compression tester. The force-time and strainetime curves are shown in Fig. 5a. Within the first 0.5 s, the Ca-ALEs was compressed to 90% of their original height (strain, ε ¼ 0.1), and then the strain was held constant at 0.1 for 100 s to allow force relaxation. The forceestrain curve within the first 0.5 s was plotted in Fig. 5b. The force required at lower strains (up to 0.02) was extremely low and dominated by the adhesive interactions at the interface between Ca-ALEs and the top plate (Johnson et al., 1971). Between the strain 0.02 and 0.1, a linear increase in loading force was visible as a function of strain for both Ca-ALEgranules and Ca-ALEflocs, the slope of this forceestrain curve was defined as the stiffness of the material, i.e. resistance of the material to deformation. Ca-ALEgranules was found to be 6 times stiffer than Ca-ALEflocs, judging from the slope of 31.5 8.5 mN as compared to the slope of 5.3 1.3 mN of Ca-ALEflocs. This means that Ca-ALEgranules is more difficult to deform. When the strain was kept constant at 0.1, the force required to maintain this constant strain decreased with time, faster in the beginning and then gradually to a non-zero
Table 1 e Fraction of different building blocks obtained by partial hydrolysis of alginate-like exopolysaccharides isolated from both aerobic granular sludge and conventional aerobic flocculent sludge. Recovery is given as percentage recovered from original ALE amount. Sludge source Granular sludge Flocculent sludge
Fractionation recovery (%)
Poly(guluronic acid) blocks (GG) %
Poly(mannuronic acid) blocks (MM) %
Heteropolymeric blocks (MG) %
85.4 7.2 73.1 8.0
69.1 8.9 35.3 8.4
2.1 1.4 5.7 2.4
14.6 2.3 32.4 1.6
62
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 5 7 e6 5
Fig. 4 e FT-IR spectra of alginate-like exopolysaccharides isolated from aerobic granular sludge and flocculent sludge.
residual value for Ca-ALEgranules. This is a typical behavior of a viscoelastic solid: a certain amount of internal force is required to maintain a constant strain (Wineman and Rajagopal, 2000). This behavior was in accordance with findings by Mancini et al. (1999), the well cross-linked alginate gels
demonstrate an asymptotically decreasing trend in stress relaxation. Ca-ALEflocs showed different trend where, the force decreased continuously and attained a value of zero in about 19.4 s. This is a typical behavior of a viscoelastic fluid (Wineman and Rajagopal, 2000).
Fig. 5 e Data acquisition and analysis from the low load compression tester. (a) Raw output of force and strain as a function of time (b) Derivation of stiffness from the linear region where deformation is imposed on the Ca-ALEs. Derivation of the viscoelastic parameters from the relaxation part of Ca-ALEflocs (c) and Ca-ALEgranules (d) force-time curve by fitting a generalized Maxwell model.
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 5 7 e6 5
To further understand how the ALE micro-structure affect gel strength, the force-time data at relaxation region was fitted with a generalized Maxwell model (Fig. 5c and d). The optimum number of Maxwell elements chosen to fit the experimental data is obtained by minimizing the Chi square and the number of the considered fitting parameters. The model fitting parameters are reported in Table 2. It is evident that, two Maxwell elements are required to satisfactorily fit the experimental data of Ca-ALEflocs. The first element relaxes extremely fast, with the relaxation time s 1 s; the second element relaxes slowly, with the relaxation time 1 < s 10 s. In contrast, three Maxwell elements are required to fit the experimental data of Ca-ALEgranules. Besides the two elements that fell in the same ranges of relaxation time of Ca-ALEflocs (s 1 s and 1 s < s 10 s), a third element, which has the highest importance (E ¼ 59.5%) and the longest relaxation time (s > 10 s) has to be presented as well. The decrease of stress at constant strain is called stress relaxation. As ideal elastic materials (ideal solids) do not relax (s / N) whereas ideal viscous materials (ideal liquids) instantaneously relax (s / 0), the viscoelastic materials display relaxation in between (0 < s < N) (Steffe, 1992). It is demonstrated from the relaxation time derived by Maxwell model that, both Ca-ALEgranules and Ca-ALEflocs have viscoelastic property. However, Ca-ALEgranules is much more elastic than Ca-ALEflocs due to the presence of the third element, which is the most important element with the highest spring constant (representing elasticity) and longest relaxation time, implying that Ca-ALEgranules is well cross-linked gel (Riande et al., 2000). Therefore, in comparison with Ca-ALEflocs, Ca-ALEgranules has not only higher stiffness but is also well cross-linked.
4.
63
4.1. Influence of the fine chemical structure on gelforming capacity of ALE Alginate is composed of three different blocks, GG, MG and MM blocks. These three kinds of blocks are distributed randomly in the molecular chain of alginate. As each kind of block has its specific property, alginate with different fine chemical structure manifests a different behavior. From this point of view, alginate is indeed a family of polymers consists of same monomers, but with different chemical structures (Davis et al., 2003). Alginate is able to form a gel matrix. The three different kinds of blocks contribute unequally to the gel-forming property of alginate: GG blocks have higher affinity towards divalent cations than the other two blocks (Braccini et al., 1999); the order of gel-forming capability of the three blocks is: MM blocks MG blocks << GG blocks (Smidrød, 1974). Therefore the content of GG blocks is the most important factor for gel formation. In the chemical structure of ALEgranules, about 69% was GG blocks. If the amount of the ALE in granules is also considered (16% w/w), the GG rich ALEgranules benefited the granules with the formation of a compact gel structure (Fig. 3b). In contrast, there are less GG blocks in ALEflocs, which results to a weaker affinity to divalent ions and limits the polymer’s gel-forming capability. Furthermore, the percentage of MG blocks is almost equal to GG blocks in ALEflocs. As flexibility of the three kinds of blocks increase in the order: GG blocks < MM blocks < MG blocks (Melvik and Dornish, 2004), larger amount of MG blocks increase flexibility of alginate molecules. Consequently, the ALEflocs with comparatively less GG blocks and more MG blocks endowes flocculent sludge with a flexible and loose structure (Fig. 3a).
4.2. Influence of the fine chemical structure on the mechanical property of ALE gel
Discussion
The alginate-like exopolysaccharides were isolated both from activated sludge and aerobic granular sludge. The quantity of ALE in aerobic granular sludge is two times higher than in the flocculent activated sludge. These two ALEs manifested different gel forming property, chemical composition and mechanical stiffness. The distinguished chemical composition is the fundamental cause for the differences in gel formation and mechanical properties.
The stiffness of alginate gel is affected by the amount of crosslink points, the concentration and composition of alginate (Donati and Paoletti, 2009). The three kinds of blocks in
Table 2 e Generalized Maxwell parameters of CaALEgranules and Ca-ALEflocs. Spring constants (G), relaxation time constants (s), and relative importance (E ) of Maxwell elements. Ca-ALEgranules
Ca-ALEflocs G (mN) E (%) s (s)
G1 2.8 0.1 51.7 2.0
G2 2.6 0.3 48.3 3.0
G1 9.4 0.5 28.1 0.8
G2 10.3 0.7 30.7 1.6
G3 13.9 2.4 59.5 2.8
s1 0.4 0.03
s2 6.1 1.3
s1 1.0 0.8
s2 9.6 2.9
s3 77.2 8.0
Fig. 6 e Crosslink between alginate-like exopolysaccharides and Ca2D. a: strong gel structure formed between alginate-like exopolysaccharides derived from aerobic granules and Ca2D; b: weak structure formed between alginate-like exopolysaccharides derived from aerobic flocculent sludge and Ca2D.
64
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 5 7 e6 5
alginate molecules have substantially different geometries. GG blocks are buckled. If two GG blocks are aligned side by side, diamond shaped holes result. Their dimensions are ideal for the crosslink with divalent cations (especially Ca2þ). This is the so-called “egg-box” model (Christensen et al., 1990) (Fig. 6). Thus, in alginate gel, the “egg-boxes” formed by GG blocks are the crosslink points, they decide the stiffness of the gel. The large numbers of GG blocks in Ca-ALEgranules contributes to its higher resistance to deformation. Alginate gels consist of more than 90% water in a supporting matrix (or network) of crosslinked polymer chains (Nussinovitch, 1997). In function, the polymer network resembles a sponge with water holding in the pores of the matrix. When a rapid external compression force is applied to Ca-ALEs at the start of relaxation tests, it causes deformation of the sponge structure by reducing the pore size and reorientation the molecular linkage from the original configuration (Wineman and Rajagopal, 2000). Decrease of the pore size results in a build-up of hydraulic pressure. The difference of hydraulic pressures in different pores might force water to migrate, therefore, water migration could be the fastest (s 1 s) element in Maxwell model in both Ca-ALEs. Ca-MG and Ca-MM are comparatively weak and flexible (Fig. 6). Their structures can easily reoriented, thus the second element with 1 s< s 10 s could be attributed to the weak structure of (Ca-MG þ Ca-MM) in both Ca-ALEs. The third element of Ca-ALEgranules with s > 10 s could be regarded as Ca-GG, because of the well crosslinked structure (Fig. 6), the reorientation is limited. A force is always required to maintain it in a distorted state. Thanks to the presence of this element, the stress-relaxation of Ca-ALEgranules displays as viscoelastic solid within the 100 s. In the case of Ca-ALEflocs, due to its lower GG content, a well crosslinked structure could not be formed; a reorientation of GG-Ca could therefore be not necessary with the applied deformation. The stress-relaxation curve of Ca-ALEflocs displays a viscoelastic liquid behavior due to a lack of the GG-Ca element. Therefore, when a same external force is applied to both Ca-ALEflocs and Ca-ALEgranules, water is going to migrate and the flexible part quickly deforms. As a continuous network structure is present in Ca-ALEgranules by crosslink of GG blocks with Ca2þ, deformation of the flexible part hardly causes the whole gel structure to break. However, Ca-ALEflocs is less crosslinked and lacks the third element, deformation of the flexible part will induce a continuous movement of ALE molecules, eventually resulting in a damage of the whole CaALEflocs structure. If Ca-ALE is the main form of ALE in activated sludge flocs and aerobic granules, evidently, the existence of Ca-ALEgranules endows aerobic granules a stronger mechanical property.
5.
Conclusions
The fine chemical structure of alginate-like exopolysaccharides isolated from aerobic flocculent sludge and aerobic granular sludge are different. Alginate-like exopolysaccharides derived from aerobic granules had significantly higher amount of poly(gluronic acid) blocks but lower amount of poly(guluronic acidmanuronic acid) blocks, while alginate-like exopolysaccharides
derived from activated sludge has an equal amount of poly(gluronic acid) blocks and poly(guluronic acid-manuronic acid) blocks. These differences result in a perfect gel-forming capability of alginate-like exopolysaccharides derived from aerobic granules and bestowed this exopolysaccharides gel a stronger mechanical property as compared to alginate-like exopolysaccharides derived from flocculent sludge, eventually contribute to the distinguished characteristics between aerobic granular sludge and flocculent sludge.
Acknowledgments This research was supported by a Marie Curie Intra European Fellowship (PIEF-GA-2009-253096) within the 7th European Community Framework Programme. We thank DHV, The Netherlands for providing aerobic granular sludge and flocculent activated sludge.
references
Adav, S.S., Lee, D.J., Show, K.Y., Tay, J.H., 2008. Aerobic granular sludge: recent advances. Biotechnology Advances 25, 411e423. APHA, 1998. Standard Methods for the Examination of Water and Wastewater, twentieth ed. American Public Health Association, Washington, DC. Bramhachari, P.V., Kavikishor, P.B., Ramadevi, R., Kumar, R., Rao, B.R., Dubey, S.K., 2007. Isolation and characterization of mucous exopolysaccharide (EPS) produced by Vibrio furnissii strain VB0S3. Journal of Microbiology and Biotechnology 17, 14e51. Braccini, I., Grasso, R.P., Perez, S., 1999. Conformational and configurational features of acidic polysaccharides and their interactions with calcium ions: a molecular modeling investigation. Carbohydrate Research 317, 119e130. Bruus, J.H., Nielsen, P.H., Keiding, K., 1992. On the stability of activated sludge flocs with implications to dewatering. Water Research 26, 1597e1604. Christensen, B.E., Indergaard, M., Smidsrød, O., 1990. Polysaccharide research in Trondheim. Carbohydrate Polymers 13, 239e255. Davis, T.A., Volesky, B., Mucci, A., 2003. A review of the biochemistry of heavy metal biosorption by brown algae. Water Research 37, 4311e4317. De Kerchove, A.J., Elimelech, M., 2007. Impact of alginate conditioning film on deposition kinetics of motile and nonmotile Pseudomonas aeruginosa strains. Applied and Environmental Microbiology 73, 5227e5234. Del Nobile, M.A., Chillo, S., Mentana, A., Baiano, A., 2007. Use of the generalized Maxwell model for describing the stress relaxation behavior of solid-kike foods. Journal of Food Engineering 78, 978e983. Donati, I., Paoletti, S., 2009. Material properties of alginates. Biology and Applications 13, 1e53. Draget, K.I., Strand, B., Hartmann, M., Valla, S., Smidsrød, O., Skja˚k-Bræk, G., 2000. Ionic and acid gel formation of epimerized alginates; the effect of AlgE4. International Journal of Biological Macromolecules 27, 117e122. Hampson, F.C., Farndale, A., Strugala, V., Sykes, J., Jolliffe, I.G., Dettmar, P.W., 2005. Alginates rafts and their characterisation. International Journal of Pharmaceutics 294 (1e2), 137e147.
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 5 7 e6 5
Johnson, K.L., Kendall, K., Roberts, A.D., 1971. Surface energy and the contact of elastic solids. Proceedings of the Royal Society A324, 301e320. Leal, D., Matsuhiro, B., Rossi, M., Caruso, F., 2008. FT-IR spectra of alginic acid block fractions in three species of brown seaweeds. Carbohydrate Research 343, 308e316. Lin, Y.M., de Kreuk, M.K., van Loosdrecht, M.C.M., Adin, A., 2010. Characterization of alginate-like exopolysaccharides isolated from aerobic granular sludge in pilot plant. Water Research 44, 3355e3364. McSwain, B.S., Irvine, R.L., Hausner, M., Wilderer, P.A., 2005. Composition and distribution of extracellular polymeric substances in aerobic flocs and granular sludge. Applied and Environmental Microbiology 71 (2), 1051e1057. Mancini, F., Moresi, M., Rancini, R., 1999. Mechanical properties of alginate gels: empirical characterisation. Journal of Food Engineering 39, 369e371. Melvik, J.E., Dornish, M., 2004. Alginate as a carrier for cell immobilization. In: Nedovic, V., Willaert, R. (Eds.), Fundamentals of Cell Immobilization Biotechnology. Kluwer Academic Publishers. Nussinovitch, A., 1997. Hydrocolloid Applications Gum Technology in the Food and Other Industries. Blackie Academic & Professional, London, pp. 292e311. Riande, E., Diaz-Calleja, R., Prolongo, M.G., Masegosa, R.M., Salom, C., 2000. Polymer Viscoelasticity: Stress and Strain in Practice. Marcel Decker, New York, p. 879. Seviour, T., Pijuan, M., Nicholson, T., Keller, J., Yuan, Z., 2009a. Understanding the properties of aerobic sludge granules as
65
hydrogels. Biotechnology and Bioengineering 102 (5), 1483e1493. Seviour, T., Pijuan, M., Nicholson, T., Keller, J., Yuan, Z., 2009b. Gel-forming exopolysaccharides explain basic differences between structures of aerobic sludge granules and floccular sludges. Water Research 43, 4469e4478. Sharma, P.K., busscher, H.J., Terwee, T., Koopmans, S.A., van Kooten, T.G., 2011. A comparative study on the viscoelastic properties. Experimental Eye Research 93 (5), 681e688. Silverstein, R.M., Clayton Bassier, G., Morrill, T.C., 1991. Spectrometric Identification of Organic Compounds. Willey, New York. Smidrød, O., 1974. Molecular basis for some physical properties of alginates in the gel state. Faraday Discussions of the Chemical Society 57, 263e274. Simdsrød, O., Draget, K.I., 1996. Chemistry and physical properties of alginates. Carbohydrate in Europe 14, 6e13. Steffe, J.F., 1992. Rheological Methods in Food Process Engineering. Freeman Press, East Lansing, Mich. USA. Sutherland, I.W., 1990. Biotechnology of Microbial Exopolysaccharides. Cambridge University Press, Cambridge. Sutherland, I.W., 2001. Biofilm exopolysaccharides: a strong and sticky framework. Microbiology 147, 3e9. Tschoegl, N.W., 1989. The Phenomenological Theory of Linear Viscoelastic Behavior. Springer-Verlag, Heidelberg. Wineman, A.S., Rajagopal, K.R., 2000. Mechanical Response of Polymers: An Introduction. Cambridge University Press, UK.