Maximization of volatile fatty acids production from alginate in acidogenesis

  • Published on

  • View

  • Download

Embed Size (px)


  • uc


    feedstock was approached for the rst time.and initial pH was determined using RSM.tion was evaluated.of the a

    , while alcohols (i.e., ethanol, butanol, and propanol) were not detected.

    biofuels and biomaterials (Chang et al., 2010). For instance, bio-alcohol (Borines et al., 2013) and bio-hydrogen (Shi and Yu,2006) can be produced from these algae by anaerobic fermentationand bio-oil (Ross et al., 2009) can be produced by pyrolysis. Among

    t in brown algaeraget et al.omass prer liquid b

    bial substrate, compared to terrestrial biomass components (Pawarand Edgar, 2012).

    In this study, we investigated the experimental conditions thatare necessary to maximize TVFA and alcohol production by the fer-mentation of alginate. The objective of this study was to identifythe optimal conditions (alginate concentration and initial pH) re-quired to maximize the efciency of the bioconversion of alginateinto TVFAs and alcohols by anaerobic fermentation.

    Corresponding authors. Tel.: +82 51 629 7646; fax: +82 51 629 6429 (M. Song).Tel.: +82 51 629 6436; fax: +82 51 629 6429 (H.-C. Woo).

    Bioresource Technology 148 (2013) 601604

    Contents lists availab

    Bioresource T

    elsE-mail addresses: (M. Song), (H.-C. Woo).Recently, marine macro-algae, classied as green-, red-, andbrown-algae, have attracted attention as biomass resources for

    the use of alginate in fermentation processes is still a challengedue to its low solubility in water and its limited usage as a micro-energy sources. Among these, marine biomass from sources suchas microalgae and macro-algae, represent one of the most promis-ing resources due to its cleanliness and sustainability (Kim et al.,2013).

    which accounts for up to 40% of the dry weighand is a principal component of the cell wall (DJung et al., 2013). Although this new marine bihigh potential as a polysaccharide feedstock fo0960-8524/$ - see front matter 2013 Elsevier Ltd. All rights reserved., 2005;sents aiofuels,Anaerobic fermentationMarine brown algaeResponse surface methodologyVolatile fatty acids

    2013 Elsevier Ltd. All rights reserved.

    1. Introduction

    Fundamental issues in the current energy system such as dra-matic increase in fossil fuel prices, sharp depletion of petrol, andclimate change, have led scientists to investigate new alternative

    the macro-algae, massive brown algae are primarily composed ofpolysaccharides such as alginate, laminaran, fucoidan, mannitol,and cellulose (Chang et al., 2010) and the feasibility of the fermen-tative conversion of these polysaccharides into liquid biofuels hasdemonstrated (Horn et al., 2000). Alginate is a polysaccharide,Keywords:Alginate

    mixtures (i.e., 71.995.5%) The VFA production from alginate as a Optimization of alginate concentration The effect of variables on VFAs produc Acetic acid was the major component

    a r t i c l e i n f o

    Article history:Received 22 June 2013Received in revised form 20 August 2013Accepted 21 August 2013Available online 30 August 2013lginate fermentation.

    a b s t r a c t

    In this study, the response surface methodology (RSM) was applied to determine the optimum fermen-tative condition of alginate with the respect to the simultaneous effects of alginate concentration and ini-tial pH to maximize the production of total volatile fatty acids (TVFAs) and alcohols. The results showedthat the alginate fermentation was signicantly affected by initial pH than by alginate concentration andthere was no interaction between the two variables. The optimum condition was 6.2 g alginate/L and ini-tial pH 7.6 with a maximum TVFAs yield of 37.1%. Acetic acids were the main constituents of the TVFAsh i g h l i g h t sShort Communication

    Maximization of volatile fatty acids prodacidogenesis

    Hong Duc Phama, Jiyun Seon b, Seong Chan Lee a, MaDepartment of Chemical Engineering, Pukyong National University, 365 Sinseon-ro, Nab The Institute of Cleaner Production, Pukyong National University, 365 Sinseon-ro, Nam

    journal homepage: www.tion from alginate in

    yung Song b,, Hee-Chul Woo a,u, Busan 608-739, Republic of KoreaBusan 608-739, Republic of Korea

    le at ScienceDirect


    evier .com/locate /bior tech

  • 2. Methods

    2.1. Feedstock

    Sodium alginate (80120 mPas, Wako Pure Chemical IndustriesLtd., Japan) was dissolved in distilled water and autoclaved (121 Cfor 15 min) then used as a microbial growth substrate. Alginatewas the sole carbon source in the medium, which also containedNH4HCO3, 2.0 g/L; KH2PO4, 1.0 g/L; MgSO4.7H2O, 0.01 g/L; NaCl,0.001 g/L; Na2MoO4.2H2O, 0.001 g/L; CaCl2.2H2O, 0.001 g/L;MnSO4.7H2O, 0.0015 g/L; and FeCl2.4H2O, 0.00388 g/L as nutrientadditives. The initial pH was adjusted as required using 5 N NaOHor 5 N HCl.

    Liquid and gas samples were taken daily for analysis. The liquid

    602 H.D. Pham et al. / Bioresource Tech2.2. Inoculum preparation and fermentation

    Anaerobically digested sludge was obtained from a municipalwastewater treatment plant in Busan, Korea. In order to enhancethe activity of VFA-producing bacteria, an acid pre-treatment(2 N HCl) was applied at 35 C for 24 h (Lee et al., 2009). A contin-uous anaerobic fermentation was operated in a 3 L bioreactor witha working volume of 2 L at 35 C and pH was maintained at 5.5with 5 N NaOH or 5 N HCl. Part of the fermentation broth was re-moved daily and replaced (the retention time was 1 day) with afresh feed (Lee et al., 2009). The concentrations of TVFAs and alco-hols in the system were maintained at 1214 g/L. The efuent fromthe inoculum system was used as seed culture (equivalent to 10%of working volume) for a series of 500 mL amber reactors with aworking volume of 400 mL.

    The alginate fermentation was operated at 35 C and 120 rpm.Chloroform (CHCl3; 100 lM) was used as a methanogen inhibitorfrom both H2/CO2 and acetate. It also inhibited acetate consump-tion by sulfate reducers (Hu and Chen, 2007).

    2.3. Experimental design and selection of variables

    Response surface methodology (RSM) was used to determinethe effects of alginate concentration and initial pH. It was appliedto evaluate the relative signicance of the experimental variablesand to nd the optimum conditions within the design boundaryof the independent variables, under which TVFA and alcohol yieldswere maximum. The experiment was based on the central compos-ite in cube design (Montgomery et al., 2011) and consisted of a2 2 orthogonal design (alginate concentration and initial pH) inorder to minimize the number of trials needed to obtain statisti-cally valid results (Table 1). The ranges of independent variableswere set 4.09.0 g of alginate/L and pH 6.010.0 based on prelimin-ary results (data not shown). Each trial with a center point (i.e.,

    Table 1Experimental design and observed total volatile fatty acids (TVFAs) production in theanaerobic alginate fermentation.

    Trials Independent variables TVFAsyield(%)




    Linear design 1 4 6 30.92 9 6 24.03 4 10 15.14 9 10 20.75a 6.5 8 34.7 0.2


    6 2.9 8 23.97 10.0 8 21.68 6.5 5.2 13.69 6.5 10.8 8.0Validation 10 6.2 7.6 37.0 0.1

    a Center point was repeated by three times.samples were centrifuged for the detection of VFAs (C2C6) andalcohols (ethanol, butanol, and propanol) at 3000 rpm for 10 min.The VFAs prole was detected by UV/VIS detector at 210 nm, andalcohols were determined by Refractive Index detector using HPLC(Ultimate 3000, Dionex, USA) with column Aminex HPX-87H.Every analysis was performed at 65 C under isocratic conditionwith 2.5 mM H2SO4 as mobile phase. Total organic carbon (TOC)was analyzed by a TOC analyzer (TOCVCPH, Shimadzu, Japan).The volatile solids (VS) concentration was determined accordingto the procedures in Standard Methods (APHA-AWWA-WEF.,1998). The carbohydrate concentration was determined using thephenolsulfuric acid method (Dubois et al., 1956) and pH wasmonitored by a pH meter (Istek, model 720P, Korea).

    Gas samples for hydrogen were analyzed by GC-HP5890 with apacked column Hayesep Q (SS, 1.8 m 1/800, and 80/60 mesh) anda thermal conductivity detector (TCD) of 90 C, 35 C, and 120 C.And methane and carbon dioxide were measured by GC-HP5890with a ame ionization detector of 180 C, 35 C, 280 C, and350 C using Ni catalyst and a packed column Porapak Q (SS, 2 m,1/800, 80/100 mesh).

    2.5. Calculation

    The yield of TVFAs (g carbon in TVFAs/g carbon in substrate) wascalculated as the amount of carbon in the TVFAs produced dividedby the amount of soluble carbon in the substrate feed.

    TVFAs yield % nTVFAsnalginate


    where: nTVFAs = the carbon amount of TVFAs produced as observed(mole carbon).nalginate = the carbon amount of alginate feeding as ob-tained (mole carbon).

    3. Results and discussion

    In this study, acetic-, butyric-, propionic-, lactic-, and valericacids were the bioconversion metabolites, whereas alcohols (i.e.,ethanol, butanol, and propanol) were not detected. In individualVFA, acetic acid was the main constituent with 71.995.5% of theTVFAs. Ethanol production from lignocellulosic biomass whichhas similar chemical structure with alginate has been reported(Nakashima et al., 2011). In contrast, in our study, the utilizationof alginate as acidogenic feedstock did not lead to the productionof alcohols.

    A total of 11 trials, including a center point, were run to approx-imate the response surface for TVFAs production. To nd the max-imum bioconversion efciency, increasingly complex equations6.5 g of alginate/L and initial pH 8.0) was replicated 3 times aspreviously described. This type of design was used to minimizethe number of trials needed to obtain statistically valid results(Song et al., 2007).

    A sequential procedure of collecting data, estimating polynomi-als, and checking the adequacy of the model was applied. Themethod of least squares was used to estimate the parameters inthe approximating polynomials. For the statistical analysis,Minitab software (version, Minitab Inc., State College,Pennsylvania, USA) was applied to establish the experimentaldesign and to test complex polynomials to model the data.

    2.4. Analytical methods

    nology 148 (2013) 601604from linear to quadratic were sequentially tested to model the dataobtained from the trials in Table 1. When the data were analyzedusing the various models, the P-value of regression was signicant

  • pH).Thus, this equation was used to determine the condition that

    would maximize the bioconversion efciency of TVFAs by setting

    of 7.6 indicates that slight alkaline environment was suitableranges for enhancing alginate solubility (Borchard et al., 2005;Haug et al., 1963). The solubility of alginate depends strongly on

    H.D. Pham et al. / Bioresource Techat the 1% a-level, whereas the lack of t was not signicant at the5% a-level only for the quadratic model in Eq. (2).

    gTVFAs 35:241 0:592 x1 3:397 x2 4:769 x21 10:745 x22 3:130 x1x2 2

    Fig. 1. Two- and three-dimensional contour plots of the quadratic model for theTVFAs yields (%) with the respect to alginate concentration and initial pH within thedesign boundary.

    Fig. 2. Residual plots of the quadratic model for the TVFAs yields (%). Each residualwas calculated using Eq. (2).the dissociation of the backbone carboxylic groups by pH.Alginate is a water soluble polysaccharide containing a linear

    unbranched chain of b (1? 4)-linked-D-mannuronic acid (M)-and a (1? 4)-linked-L-guluronic acid (G)- residues. It consists ofa widely varying composition of M- and G- residues with no regu-lar repeating unit (Kim et al., 2010). The low soluble fractions arecomposed of molecules that are either predominantly M rich orG rich (i.e., MM and GG), whereas the hydrolysable fractions aremade up of a high proportion of alternating MG residues (Pawarand Edgar, 2012). In this study, the maximum yield of TVFAs (i.e.,37.1%) was lower than that of lignocellulosic biomass as a polysac-charide feedstock i.e., 63.0%; (Hu et al., 2004). This suggests thatthe irregular sequences of alginate presented a challenge duringthe anaerobic fermentation.

    In order to verify the accuracy of the model predictions, anadditional validation trial was run under the optimal conditionspredicted by the model (6.2 g alginate/L and pH 7.6) (Table 1).The residual plots, another indication of the adequacy of the tof the model, were randomly distributed without any patternsand trends (Fig. 2). Therefore, it was concluded that the modelwas able to accurately predict the maximum bioconversion condi-tions for TVFA production using alginate as a biofuels feedstock inanaerobic fermentation.

    4. Conclusions

    The RSM was successfully applied to determine the optimalconditions with respect to alginate concentration and initial pHfor TVFAs production. In the quadratic model, the initial pH signif-icantly affected the yield of TVFAs and no signicant interaction be-tween the independent variables was observed. The maximumTVFA yield of 37.1% was determined at 6.2 g/L of alginate and aninitial pH of 7.6. Acetic acid was the predominant component ofTVFAs mixture (i.e., 95.5%). This novel approach suggests that algi-nate is a potential biomass resource for biofuel production inanaerobic fermentation.

    Acknowledgementthe quadratic derivatives of the equation to zero with respect tothe independent variables. The RSM model estimated a maximalTVFAs yield of 37.1% at 6.2 g of alginate/L and initial pH 7.6. Two-dimensional response surfaces of the quadratic model for TVFAsyield with the corresponding estimated optimums are illustratedin Fig. 1. The response surface of TVFAs yield indicated a clear peak,which showed the optimum condition was inside the designboundary.

    Response surface analysis showed that the initial pH was signif-icant for TVFAs yield at the 1% a-level, whereas the alginate con-centration was not. And none of the interaction between thesetwo variables was signicant at the 1% a-level. The optimum pHwhere g = experimental value of the TVFAs yield (%), and xi = inde-pendent variables i (i = 1 for alginate concentration and 2 for initial

    nology 148 (2013) 601604 603This work was nancially supported by the Korea Fisheries Re-source Agency of Ministry of Oceans and Fisheries (20120631251-00).

  • References

    APHA-AWWA-WEF, 1998. Standard Methods for the Examination of Water andWastewater, 20th ed. American Public Health Association, Washington, DC.

    Borchard, W., Kenning, A., Kapp, A., Mayer, C., 2005. Phase diagram of the systemsodium alginate/water: a model for biolms. Int. J. Biol. Macromol. 35 (5), 247256.

    Borines, M.G., de Leon, R.L., Cuello, J.L., 2013. Bioethanol production from themacroalgae Sargassum spp. Bioresour. Technol. 138, 2229.

    Chang, H.N., Kim, N.-J., Kang, J., Jeong, C.M., 2010. Biomass-derived volatile fattyacid platform for fuels and chemicals. Biotechnol. Bioprocess Eng. 15 (1), 110.

    Draget, K.I., Smidsrod, O., Skjak-Braek, G., 2005. Alginates from algae, . rst ed.. In:Steinbuchel, A., Rhee, S.K. (Eds.), Polysaccharides and Polyamides in the FoodIndustry: Properties, Producti...


View more >