Digital image analysis to estimate the inﬂuence of sodium dodecyl sulphate on activated sludge ﬂocs

(1)

Digital image analysis to estimate the influence of sodium dodecyl sulphate on activated sludge flocs

Ewa Liwarska-Bizukojc

^a,

* , Marcin Bizukojc

^b

aDepartment of Environmental Engineering, Technical University of Lodz, Al. Politechniki 6, 90-924 Lodz, Poland

bDepartment of Bioprocess Engineering, Technical University of Lodz, Al. Politechniki 6, ul. Wolczanska 213/215 90-924 Lodz, Poland

Received 30 January 2004; accepted 4 July 2004

Abstract

The results of a quantitative description of activated sludge flocs exposed to anionic surfactant sodium dodecyl sulphate (SDS) are presented. Image analysis procedures are applied to measure the morphological parameters of activated sludge flocs. Surfactant influenced sludge flocs properties even at low concentration from 2.5 to 25 mg L ¹. Mean projected area of flocs decreased by about 30% at SDS concentrations from 2.5 to 25 mg L ¹and by about 60% at concentrations from 250 to 2500 mg L ¹. The decrease of sludge flocs dimensions deteriorates their settleability. Additionally, a linear correlation between total suspended solids (TSS) and mean projected area was found within the duration of the biodegradation process.

Keywords: Activated sludge; Biodegradation; Image analysis; Linear alkyl sulphate; Sodium dodecyl sulphate; Surfactant

1. Introduction

Surfactants, due to their favourable physicochemical properties are extensively used in many fields of technology and research, i.e. in pharmacy, in cosmetics, textile industry, agriculture, biotechnology[1,2]. After use large quantities of surfactants and their derivatives are released to aquatic and/or terrestrial environment. This substantial stream reaches wastewater treatment plants, where sewage and surface water are gathered and processed[1,3].

The average surfactant concentration in raw domestic wastewater is from 10 to 20 mg L ¹, whereas in some industrial wastewater it may achieve 300 mg L ¹ [4,5].

Surfactants can act on biological wastewater treatment processes. They induce foaming in aerated bioreactors and reduce the settling ability of sludge [6]. What is more important, surfactants are toxic to microorganisms.

Proksova´ et al. reported that dialkyl sulphosuccinate anionic surfactants influence the respiration rate as well

as the activity of enzymes and the growth of degrading bacteria[7].

Apart from biochemical activity of activated sludge microorganisms, the physical properties of sludge flocs play also an important role in the run of wastewater treatment processes. The size and shape of sludge flocs are correlated with settleability of sludge, which influences the efficiency of separation processes in the final clarifier. Digital image analysis has been used to examine the morphological parameters of activated sludge flocs [8,9]. Grijspeerdt and Verstraete have quantified the size and shape of sludge flocs using image analysis [9]. The morphology of flocs was expressed by two parameters: the mean equivalent circle diameter and the mean form factor. It was found that there is a possibility to correlate the morphology of the sludge flocs to the traditional settling index. Furthermore, the correlation between biomass concentration and the projected floc area, in a certain concentration range, was established. Motta et al.

developed an automated image analysis procedure for the simultaneous characterisation of flocs (projected diameter and fractal dimension) and filamentous bacteria (filamentous length) of fresh activated sludge observed by optical

www.elsevier.com/locate/procbio

* Corresponding author. Tel.: +48 42 631 35 98; fax: +48 42 631 35 23.

E-mail address: ewaliwar@p.lodz.pl (E. Liwarska-Bizukojc).

doi:10.1016/j.procbio.2004.07.020

(2)

Sodium dodecyl sulphate C₁₂H₂₅OSO₃Na purchased from POCh (Poland) was used. SDS is the most typical alkyl sulphate (AS) compound and is used extensively in scientific research[11]. SDS, in particular, is an essential component of shampoos and foaming agent for toothpaste. Its molecular weight is 288 g mol ¹. Critical micellar concentration (cmc) is equal to 2310 mg L ¹.

The experimental procedure consisted of two parts. In first stage of the experiments, a so called ‘‘screening test’’, were carried out for different concentrations of SDS.

Experiments were conducted at the following concentrations of SDS below 2.5, 25, 250 and above cmc 2500 mg L ¹. The second step comprised of the kinetic investigations of SDS biodegradation, which were conducted at one selected SDS concentration. This was made on the basis of the first stage results and the SDS concentration was 250 mg L ¹.

All experiments were conducted in shake flasks under aerobic conditions. The culture medium was synthetic wastewater prepared according to Polish Norm PN-72/C- 04550 [12]. Activated sludge was taken from activated sludge chamber at the wastewater treatment plant (WWTP) in Zgierz. Fifteen millilitre samples of activated sludge (inoculum) were transferred to 300 ml Erlenmayer flasks, containing 135 ml of fresh synthetic wastewater with or without addition of SDS. The flasks were incubated at 221 8C in a thermostated bath shaker for 48 h. All experiments were conducted according to this procedure.

2.2. Analytical methods

2.2.1. Physicochemical analyses

SDS concentration was determined according to a methylene blue method [13]. SDS forms ion pairs with methylene blue that are extracted by chloroform and determined spectrophotometrically at 652 nm.

Chemical oxygen demand (COD) was measured by a standard dichromatic method [13]. Total suspended solids (TSS) were measured gravimetrically. Turbidity was measured using a HACH spectrophotometer DR/2000 at 450 nm according to method no. 750[14].

pixel count and its multiplication by scaling factor. The other parameters are the derivatives of mean projected area. The perimeter is simply the length of the boundary of the object, while the convex perimeter is the length of the boundary obtained after filling of all concavities of the object. Both perimeters give the insight into the irregularity of the object.

Perimeter is always higher than convex perimeter but they are equal to each other in the boundary case for the ideally smooth objects. Mean diameter was measured as the lengths of lines between two points on the boundary of the object going through its centroid, while the mean feret diameter is equivalent to the diameters measured using a pair of callipers.

They are also the measure of the irregularity of the object.

Finally, the circularity index is the shape factor, that indicate to what extent the measured object is similar to the true circle. If it is equal to one, the object is the true circle. The higher it is, the less circular is the object[15,16].

3. Results and discussion

The results of image and physicochemical analyses of screening tests are shown inFigs. 1–4. The significant effect of SDS on morphology of flocs was observed. In the control run the increase of mean projected area, diameter, perimeter, convex perimeter and feret diameter was found, whereas in sample containing SDS all these parameters decreased significantly after 24 h. The changes in morphology of flocs are connected with four different phenomena (Fig. 1):

growth (1); agglomeration and deagglomeration of flocs (2);

starvation (3); and saponification (4).

In the control run first three phenomena occurred, while the saponification played an important role in the sample with detergent added (Fig. 2). Therefore, the relative degree of flocs saponification was introduced and these values are presented in Table 1. The saponification resulted in a fast decrease of mean projected area, diameters and perimeters and changed the shape of flocs (circularity index). The influence of saponification was dependent on SDS concentration. Taking into account the effect of SDS on flocs morphology two ranges of SDS concentrations can be discriminated: 2.5–25 mg L ¹and 250–2500 mg L ¹.

(3)

Mean projected area of flocs increased by about 8% after 24 h without addition of SDS. In the run containing SDS this parameter decreased by about 30% for concentrations 2.5–

25 mg L ¹ and about 60% for concentrations 250–

2500 mg L ¹ (Table 1). Other morphological parameters diameter and feret diameter, perimeter and convex perimeter also increased by 8–9% after 24 h in the control run. At the same time the decrease of these parameters in the run with SDS addition was lower in comparison to mean projected area. It was between 10 and 15% for concentrations 2.5 and 25 mg L ¹and from 30 to 40% for 250 and 2500 mg L ¹. Analysing data inTable 1andFigs. 1 and 2with regard to the biological properties of the sludge flocs in conjunction with their morphology, the following remarks emerge.

Firstly, the growth of flocs is observed in the control run within the first 24 h and it is associated with the presence of easily available, biodegradable substrate. It was expressed here as the increase of morphological parameters values. In contrast, the decrease of flocs dimensions that appeared even at lower range of SDS concentrations is typical for high loaded systems or when the toxic substances are present in wastewater[17]. Secondly, a relationship between perimeter and convex perimeter was found. At higher range of SDS concentrations the relative degree of saponification for perimeter is higher than for convex perimeter. Taking the definitions of both perimeters into account, it can be stated

Fig. 1. Schematic representation of phenomena observed in the investigated biodegradation processes (masks of flocs are derived from original images snapped within kinetic experiments).

Fig. 2. Comparison of mean projected area values within the screening test. Fig. 3. Comparison of circularity index values within the screening test.

(4)

the level of 2.7 in the control run (Fig. 3). Its changes in SDS runs were different dependently on surfactant concentration.

In the lower range of SDS concentration the increase of circularity index was observed whereas in the higher range circularity index decreased. Saponification was probably responsible for the decrease of circularity index. The irre- gularities were smoothed and the flocs became more circular.

In the lower range of SDS concentration the circularity index of flocs changed similarly to the control sample.

In spite of the unfavourable impact of SDS on flocs morphology, the surfactant was almost completely biodegraded at lower concentrations within 24 h (Fig. 4). At a concentration of 250 mg L ¹ the degree of SDS removal was also high and equal to 91%. However, SDS remained practically unchanged, when its initial concentration was above the critical micellar concentration. The surfactant content as well as COD level in the runs, in which the initial SDS concentration was 2500 mg L ¹, was almost constant despite 48 h duration of the experiments. The results obtained indicate that the negative influence of surfactant on microbiological decomposition of organic matter increases with the initial surfactant concentration. The biodegradation

In the second stage of the experiments the kinetics of SDS biodegradation were examined. The concentration of 250 mg L ¹ was selected for the kinetic investigations because of two major reasons. First of all, the impact of SDS on flocs morphology seemed to be strongest at this concentration. Secondly, such concentrations of detergents are detected in industrial wastewater [4]. Simultaneously, the experiment without surfactant i.e. the control run was conducted. The changes of mean projected area, diameters and perimeters for SDS and control run differed significantly. As an example changes of mean projected area, which is one of the key morphological parameters, are shown in Fig. 5.

In the control run, after a short lag-phase the increase of these parameters was observed within the first 24 h of experiment. After 24 h the decrease of mean projected area and other parameters was observed as a result of starvation.

COD at 30 h of experiment was close to zero (Fig. 6).

Therefore, the processes of endogenous respiration had to dominate. In the SDS run the mean projected area, diameters and perimeters decrease gradually during the whole experiment.

During both runs the removal of organic substances load, expressed as COD, was observed. In the SDS runs changes of the surfactant concentration were well correlated with

Fig. 5. Mean projected area time courses in the kinetic experiments.

Fig. 4. Removal of sodium dodecyl sulphate (SDS) content within the screening test.

(5)

COD course. The degree of SDS removal was equal to about 90% after 48 h and was higher than the degree of COD removal (80%). This difference indicates that some inter- mediates were not biodegraded and remained in wastewater.

It is probably connected with the unfavourable morphological changes of activated sludge flocs. To compare with, the decrease of mean projected area and other morphological parameters in the control run appeared since the level of COD also decreased significantly and was at the level of several milligram O2L ¹. In both cases COD changes were well correlated with biomass concentration expressed by its indicators of changes: TSS and turbidity (Fig. 6).

Moreover, a difference between the biodegradation rates was also found (Fig. 7). The biodegradation rate was the highest between 7 and 15 h in the SDS run, whereas in the control run it was practically constant from the beginn- ing until total substrate utilisation. The initially lower

biodegradation rate in the SDS run proves that an adaptation phase is likely to occur. Further, the biodegradation rate increases significantly exceeding the rate of organic removal in the control run.

Additionally, the linear correlation between TSS and mean projected area was found within the duration of the biodegradation process (Fig. 8). This confirms a previously reposted relationship[9]. Taking this into account, the mean projected area of the flocs could also be used as a biomass indicator for practical use.

4. Conclusions

The obtained results shown that surfactants significantly influenced the morphology of activated sludge flocs even at low concentrations. The mean projected area of flocs decreased by about 30% at SDS concentrations from 2.5 to 25 mg L ¹and by about 60% at concentrations from 250 to 2500 mg L ¹. It did not seriously interfere with biodegradation processes at low anionic surfactant concentrations that are typical for domestic wastewater. The surfactant had a negative effect on biological treatment of industrial wastewater, in which the concentration of surfactants is usually higher. Nevertheless, the surfactant in the whole investigated range of concentrations caused the decrease of sludge flocs dimensions that affected the settling properties of sludge and deteriorated the separation at the final clarifier.

A significant difference in biodegradation run with and without SDS addition was observed. In the control run after a short lag-phase an increase of mean projected area, perimeter, diameter was observed, while in the SDS run all measured morphological parameters decreased gradually. Moreover, the results of kinetics experiments indicate that the removal of organic load for the SDS run was initially lower in comparison with the control run. Additionally, the linear correlation between total suspended solids and mean projected area was found. This means that the mean

Fig. 7. Comparison of biodegradation rates for control and SDS runs in the kinetic experiments; solid line-SDS run; dotted line-control run.

Fig. 8. Linear relationship between total suspended solids and mean projected area.

Fig. 6. Chemical oxygen demand (COD) and turbidity time courses in the kinetic experiments.

(6)

sludge. Water Res 2003;37:1288–95.

[4] Shcherbakova VA, Laurinavichius KS, Akimenko VK. Toxic effect of surfactants and probable products of their biodegradation on metha- nogenesis in an anaerobic microbial community. Chemosphere 1999;29:1861–90.

[5] Scott MJ, Jones MN. The biodegradation of surfactants in the environment. Biochim Biophys Acta 2000;1508:235–51.

[6] Zhang C, Valsaraj KT, Constant WD, Roy D. Aerobic biodegradation kinetics of four anionic and nonionic surfactants at sub- and supra- critical micelle concentrations (CMCs). Water Res 1999;33:115–24.

1991/96

[15] Russ JC. Computer Assisted Microscopy: The Measurement and Analysis of Images. New York: Plenum Press; 1990.

[16] Paul GC, Thomas CR. Characterisation of mycelial morphology using image analysis. Adv Biochem Eng Biotechnol 1998;60:2–59.

[17] Eikelboom DH, van Buijsen HJJ. Handbuch fu¨r die mikrobiologische Schlammuntersuchung. Munich: F Hirthammer Verlag GmbH; 1992 [18] Kayser R. Activated sludge process. In: Rehm H-J, Reed G, editors.

Biotechnology. Environmental processes. vol. 11a. 2nd ed. Weinheim:

VCH Verlagsgesellschaft; 1999