Key factors governing alkaline pretreatment of waste activated sludge☆

Xianli Shi,Li Deng,Fangfang Sun,Jieyu Liang,Xu Deng*

College of Life Sciences,Shenzhen Key Laboratory of Marine Bioresources and Ecology,Shenzhen University,Shenzhen 518060,China

Keywords:Waste activated sludge Alkaline pretreatment Waste treatment Optimization Biodegradation Bioprocess design

A B S T R A C T Alkaline pretreatment is an effective technology to disintegrate sew age sludge,w here alkali dosage and sludge concentration are two important factors.pH value or alkali concentration is usually adjusted in order to determine a proper dosage of alkali.Our work has found that this is not a good strategy.A new parameter,the ratio of alkali to sludge(R a/s),is more sensitive in controlling the alkali dosage.The sludge concentration C s and retention time t are two other important factors to consider.The validity of these arguments is confirmed with modeling and experiments.The individual effect of R a/s,C s and t was studied separately.Then the combined effect of these three factors was evaluated.The sludge disintegration degree of 44.7%was achieved with the optimized factors.Furthermore,an alkaline-micro wave combined pretreatment process was carried out under these optimized conditions.A high disintegration degree of 62.3%was achieved while the energy consumption of microwave was much lower than previously reported.

1.Introduction

Waste activated sludge is becoming a great concern worldwide in recent years[1].Due to its high organic content,how ever,after proper treatment waste activated sludge could be a good raw material for production of biog as.It could also be a feedstock in the manufacture of high value products such as polyhydroxylalkanoates[2].In order to warrant bioavail ability of waste activated sludge,suitable pretreatments are needed to disintegrate the hard cell wall of microbes in waste activated sludge to improve substrate solubility and digestibility[3].Among previously pretreatment methods,alkaline pretreatment has been proved to render relatively high disintegration efficiency[4].Furthermore,alkali pretreatment can be combined with other disintegration methods such as thermal,ultrasonic and microwave treatments,and high disintegration degree has been achieved with such synergetic designs[5,6].Almost all the studies on such combined pretreatment reveal that alkaline treatment plays a crucial role in the disintegration process.

In alkaline pretreatment,the dosage of alkali is a very important parameter.In most studies,the dosage is controlled in two ways.One is to add alkali into sewage sludge to adjust the initial pH of the suspension[7,8],but the pH value is independent of the amount of treated sludge.Shao et al.[1]found that it was difficult to correlate the degree of sludge disintegration with the required alkali dosage by solely controlling the pH value.The other method is to adjust the alkaline concentration while the sludge amount is kept constant[9,10].Similar to the first method,a certain amount of sludge was concerned in these studies,so the ‘optimized’alkaline concentration might not be optimal when the amount of treated sludge changed.

The influence of the amount of sludge on pretreatment efficiency could not be ignored.At a certain pH or alkaline concentration,different amounts of treated sludge might result in different disintegration degrees.For example,large amount of sludge may need more alkali,while with small quantity of sludge,alkali may be subject to overdose,making the treatment economically ineffective.Therefore,dosage of alkali should be determined based on the amount of sludge to be pretreated.From this point of view,a new parameter,the ratio of alkali dosage to the dry mass of treated sludge,designated as Ra/s(%),should be a more important parameter than p H value or alkaline concentration.

The sludge concentration,i.e.dry mass of sludge per liter suspension,designated as Cs(g·L?1),is also an important factor in the pretreatment process.Even if Ra/sremains constant,the variance of Csmay affect the efficiency of pretreatment.For example,if Csis too high,the mixing of dense sludge suspension with alkali may become difficult.Then the insufficient mixing would result in inefficient mass transfer and hinder the disintegration.In contrast,diluted alkali/sludge suspension warrants adequate mixing,but the low concentration requires a reactor tank with large volume capacity and will make the process less economical.Hence,a suitable Csshould be determined in the design of sludge pretreatment.Surprisingly few papers concern the effect of sludge concentration,although alkaline pretreatment has been widely studied.In most cases,the study focuses on one certain sludge concentration.

The present study,therefore,aims to improve the efficiency of the alkaline pretreatment process,taking into account the key factors of Ra/s,Csand t.The effects of these factors are evaluated separately,adjusting one of these factors while keeping the other two constant.Then their combined effect is investigated via a matrix of experiments designed with the Box–Behnken Design to find the optimal values of the three factors[11].Finally,an integrated pretreatment process with combined treatments of alkali and microwave is carried out for further comparative evaluation of our result with other similar studies.

2.Materials and Methods

2.1.Sewage sludge

The partly dewatered sample of sewage sludge for this work was taken from a local wastewater treatment plant in Shenzhen,China.The bulk of the sew age sludge was stored in a plastic bucket,covered tightly and kept at 4°C.Some quantitative parameters of the sludge were:pH 7.35,total chemical oxygen demand(TCOD)730 mg·g?1,soluble COD(SCOD)11.5 mg·g?1,and water content 77.5%,and the ratio of volatile solid(VS)content to total solid(TS)content was 0.40.

2.2.Alkaline pretreatment:Individual effects of Ra/s,Cs,and t

To obtain different Ra/swith a constant Cs,a master batch of sludgeat concentration Cswas prepared by diluting with deionized water and homogenized by agitation before being equally split into several parts.Afterwards NaOH was added to each part with different dosages.To obtain different Csat a constant Ra/s,another master batch at a defined Ra/swas prepared.Then it was split into several parts and diluted with water to desired Cs.

All suspensions were cultivated at room temperature,with agitation of 150 r·min?1.To evaluate the individual effects of Ra/sand Cs,the soluble COD was measured after incubating for 24 h.To evaluate the individual effect of retention time t,the soluble COD was detected at different time intervals.The degree of sludge disintegration(DD,%)is calculated as follows[10]:

w here TCOD0is the total COD of untreated sludge sample,SCOD0is the original soluble COD of untreated sludge,and SCODtis the soluble COD of sludge after being treated for retention time t.

2.3.Alkaline pretreatment:Combined effect

The Box–Behnken Design is employed to design the experiment to evaluate the combined effects of the three factors.Further optimization of these factors is based on the individual effects obtained previously[12].A polynomial model of quadratic equation is used to fit the experimental data in order to correlate the response to independent variables Ra/s,Cs,and t:

where Y is the predicted response,Xiis independent variable,and bijis the fitting coefficient.

Design Expert v7.1.2 is used for analysis of variance(ANOVA).The model coefficients are determined by the approach of multiple regression and the optimum operating variables could be obtained from the model equation.

2.4.Alkaline–microwave combined disintegration

The combined disintegration process was carried out under the optimal condition of alkaline pretreatment obtained.After alkali was added,the sludge suspension was irradiated with microwave at 250 W for 3 min.Then the whole pretreatment process was accomplished.

2.5.Analytical methods

TS,VS,pH and water content were measured according to Standard Methods[13].The values of TCOD and SCOD were detected from diluted sludge suspensions as described previously[14]and quantified as milligram of chemical oxygen demand per gram of dry mass pretreated sludge(mg·g?1).The values of Cswere obtained by suspending different grams of dry sludge in 1 L of water.All measurements were repeated at least three times.The average results are reported in this work.

3.Results and Discussion

3.1.Individual effect of influential factors

3.1.1.Ratio of alkali to sludge,Ra/s

With initial Cskept constant,the change of Ra/sis equivalent to changing the dosage of alkali used in the pretreatment process.Fig.1 shows that at initial Csof 45 g·L?1the increase of Ra/sfrom 1%to 16%resulted in a substantial increase of DD from 1.5%to 37%.As Ra/sfurther increased beyond 20%,however,DD was not increased effectively.Higher Ra/s,25%or more,even led to a slight decrease of DD.At such initial Cs,DD maximized to 38.4%with Ra/sbeing 23.3%.The corresponding SCOD was 287.40 mg·g?1,showing a 27-fold increase from that of original sludge.The decrease of DD at high dosage of alkali might be due to the formation of refractory compounds at extreme pH values,which was as high as 12.95 at Ra/sof 31%.This phenomenon was also reported in other studies[1,15].

On the other hand,it should be noted that in Fig.1 DD values nearly kept the same level when Ra/salmost doubled from 16%to 30%.It is demonstrated hereby that there is an optimum Ra/sfor effectiveness of disintegration and cost.Overdose of alkali increases the consumption of raw materials without the return of increased DD.It is noteworthy that this correlation could not be revealed by solely controlling the initial pH values.The corresponding initial pH only changed a bit from 12.5 to 12.9(Fig.1)when Ra/sincreased from 16%to 30%.Therefore,we believe that Ra/sis a far more sensitive parameter than pH,to observe how much alkali is really needed for sludge disintegration.

3.1.2.Sludge concentration,Cs

Fig.1.Effect of R a/s on sludge disintegration and corresponding initial pH.C s=45 g·L?1;t=24 h.

In most previous studies,alkaline pretreatment experiments were carried out at a certain sludge concentration(about 10–30 g·L?1)in an attempt to find the optimum alkali dosage[9,10,16].Our work reveals that Csis another important factor.Even though the dosages of alkali and sludge were kept the same(with Ra/sof 4.5%in Fig.2),Csposed a remarkable impact on sludge disintegration.When Cswas as low as 10 g·L?1,DD was lower than 6%,indicating low efficiency of disintegration in dilute suspension.Nonetheless,when Cswas more than 30 g·L?1,a rapid rise of DD was observed with the increase of Cs.DD reached 30%at Csof 90 g·L?1,almost 4 times higher than that at 10 g·L?1.On the other hand,higher Cs(＞90 g·L?1)would reverse the course,leading to a remarkable decline in DD.This decline might be attributed to hindrance in mass diffusion caused by insufficient dispersion of sludge clusters.Similarly,the initial values of pH could not,again,be correlated to the efficiency of sludge disintegration at different Cs(Fig.2).

Fig.2.Effect of C s on sludge disintegration and corresponding initial pH.R a/s=4.5%;t=24 h.

The effect of Cson the disintegration process was out of the focus of some researchers,though it was mentioned in their work.Toreci et al.[17]evaluated the release of SCOD by microwave irradiation at two Csand found that the effect of Cson the SCOD/TCOD ratio depended on the microwave intensity.The improvement of SCOD/TCOD was greater at low Cs(60 g·L?1)under low microwave intensity,while high Cs(118.5 g·L?1)exhibited greater improvement in enhancing the SCOD/TCOD ratio under high microwave intensity.In the work of Yang et al.[18],Ra/swas kept at 5%.They found that the ratio of SCOD/TCOD was about 12%at 5 g·L?1of Csand only show ed a negligible increase at 10 g·L?1Cs.In comparison,arise of the SCOD/TCOD ratio up to about 30%was observed when Cswas further increased to 30 g·L?1.Xiao and Liu's work[19]also disclosed that sludge concentration was a marked factor affecting the efficiency of alkaline treatment,though the SCOD/TCOD ratio did not change much in the range of Csbeing 10 to 40 g·L?1.In summary,sludge concentration is an important factor affecting the sludge disintegration and should be considered in process design.

3.1.3.Retention time t

The retention time needed for alkaline pretreatment varied in previous studies.Kim et al.[7]reported that 1 h was sufficient to complete the alkaline pretreatment because of the fast reaction rate.The required retention time was independent of pH level.In the work of Li et al.[20],the optimum retention time to achieve the highest efficiency was only 30 min.How ever,Vlyssides and Karlis[21]demonstrated that under intensive hydrolysis conditions,the SCOD of treated sludge was still increasing significantly after 10 h of alkaline pretreatment.

Fig.3.Effect of retention time t on sludge disintegration.C s=45 g·L?1;R a/s=4.5%.

Our experiments(Fig.3)demonstrated that a rapid increase of DD occurred in the first 9 h,and then the disintegration rate slowed down.DD only increased from 30%at 12 h to 32%after 24 h.It suggests that retention time longer than 12 h is not necessary.However,we caution that the required retention time may depend on Ra/sand Cs.A good process should take into account all these factors and achieve a synergetic effect.

3.2.Combined effect of influential factors

Evaluation and optimization on combined effect of various factors have been carried out in sludge disintegration,but the focus has mainly been on the comprehensive effects of alkaline pretreatment combined with other processes such as alkaline–thermal[5],alkaline–ultrasonic[7],and alkaline–microwave[22].No study has been conducted to improve the efficiency of alkaline pretreatment itself by analyzing the combined effect of alkaline-related factors such as Ra/s,Csand t.Therefore,this work evaluates the combined effects of these three factors on the basis of their individual effects found previously.Moreover,optimum operating variables are further determined.

3.2.1.Experimental result of Box–Behnken Design

In the Box–Behnken Design method,the range and levels of the factors are selected based on the results of single factor tests,which are listed in Table 1.The values of the corresponding degree of disintegration(DD)are also included.The highest DD value of 45.14%is obtained with the highest Cs(130 g·L?1)and Ra/s(24%).Those relatively high values of DD are all achieved with relatively high levels of Cs(100 or 130 g·L?1)and Ra/s(16%or 24%)but not necessarily with long retention time t.For example,in the run of No.11 DD came up to 40.32% within6 h.If either Csor Ra/swas kept at low level,no high DD could be obtained.At Csof 70 g·L?1,the highest DD was 37.99%;with Ra/skept at 8%,the highest DD could only reach up to 30.91%.In both cases,extended retention time did not make DD beyond these limits.These limits suggest that Ra/sand Cspose more impact than t on alkaline treatment.

Table 1Box–Behnken Design for alkaline pretreatment of sewage sludge

Furthermore,Table1 shows that high values of DD could be obtained with Cshigher than 90 g·L?1,while in the experiment of individual effects,DD started to decline as Cswas more than 90 g·L?1(Fig.2).This discrepancy implies that the effects of Ra/s,Csand t are not independent of each other.There could be synergetic effects between these factors.

3.2.2.Modeling and analysis of variance

By applying modeling regression with the data in Table 1,the relationship between DD and the influential factors could be established as

where X1,X2and X3are t,Csand Ra/s,respectively,and Y is DD.

The analysis of variance(ANOVA)of the model is carried out to determine the adequacy and significance of each variable.The results are displayed in Table2.The model F-value of 283.32 with a low probability value(P＜0.0001)suggests that the model is highly significant.Since P-value less than 0.05 indicates that model terms are significant,the interactive effects of X1(t)with X2(Cs)and X1(t)with X3(Ra/s)are not significant.Furthermore,the P-value of linear and quadratic terms of X1(t)is 0.0315,which is close to 0.05 and much higher than that of the other two variables.It is believed that the effect of t on sludge disintegration is less significant than that of Ra/sand Cs.This is in agreement with the experimental result.Moreover,the value of ‘lack of fit’is 3.61 with a P-value of 0.2246,indicating that it is not significant.This implies that the model is good for prediction.Finally,the multiple correlation coefficient(R2)is as high as0.9987,suggesting that 99.87%of the variability in the response could be defined by the model.Therefore,the model[Eq.(3)]could be used to predict the alkaline pretreatment process.

Table 2ANOVA analysis for alkaline pretreatment model

By differentiating the regression model of Eq.(3),the optimal values of X1,X2and X3could be obtained to be11 h,22%and 108 g·L?1for t,Ra/sand Cs,respectively.Consequently,the predicted maximum value of DD is 45.4%.

3.2.3.Confirmatory experiment

Three parallel experiments were carried out to confirm the optimum alkaline pretreatment conditions predicted by the model.The average value of DD achieved was 44.7%,close to the predicted 45.4%,indicating that the model is valid.

As most previous studies in terms of sludge pretreatment concerned combined processes of alkaline treatment with other technologies,we review several papers and extract the results of disintegration degree obtained by the alkaline treatment to compare with our result.The values of Ra/sin those studies are calculated based on the dosage of alkali and sludge provided.The comparison is shown in Table 3.A relatively high DD can be obtained in our study when Csand Ra/sare optimized.Generally,an effective alkaline treatment should be achieved with high Cs,low Ra/s,and short t.This will benefit in lowering the cost,as the pretreatment could be accomplished with as mall volume of reactor,low dosage of alkali,and short retention time.From this point of view,although high DD was achieved in the work of Yang et al.[18],excessive dosage of alkali,low concentration of sludge and long retention time made the process economically ineffective.

Table 3Comparison on degree of sludge disintegration by alkaline pretreatment

3.3.Alkaline–microwave combined pretreatment

Now a days alkaline treatment is usually combined with other technologies in order to obtain higher efficiency of sludge disintegration.In a typical combined process,alkaline treatment generally plays a crucial role in the disintegration of sludge[7,25].Significant improvement on the sludge disintegration of the w hole combined pretreatment could be expected by improving alkaline treatment.Subsequently the energy input in other parts of the process such as thermal,mechanical,sonication and microwave treatments could be reduced.In order to prove this hypothesis,a combined pretreatment of alkaline and low intensity microwave has been performed in our work.The result is compared with similar combined processes in previous studies(Table 4).The energy consumption in Table 4 is quantified as kilo Joule per gram of dry mass sludge to be pretreated(kJ·g?1).After optimizing alkaline treatment conditions(Ra/s,Cs,t)we achieved a DD of 62.3%,the highest among the pretreatments compared.Furthermore,it is noteworthy that the high DD is achieved with much lower energy consumption compared to other studies[8,22,25]with similar dosage of alkali.The next highest DD is 50.2%reported in Jang and Ahn's work[26],with less alkali but high energy consumption,more than 80 times of ours.Consequently,their overall process could be much more expensive.

Table 4Comparison on efficiency of alkaline-microwave combined pretreatment

4.Conclusions

The evaluation on individual factors of alkaline pretreatment discloses that the ratio of alkali to sludge(Ra/s)and sludge concentration(Cs)pose remarkable impacts on sludge disintegration.By optimizing the alkaline pretreatment process,the maximum degree of disintegration(DD)is predicted to be 45.4%,which was confirmed experimentally.With optimal Ra/s,Csand retention time t being 22%,108 g·L?1and 11 h,respectively,a DD of 44.7%was obtained,close to the predicted value.In an alkali–microwave combined pretreatment process under the optimized alkaline condition,a DD of 62.3%was achieved.This is the highest level that has been achieved,to the best of our knowledge.Compared with similar pretreatment processes in other studies,our process features the lowest energy input.