African Journal of Biotechnology Vol. 11(30), pp.

7693-7700, 12 April, 2012

Available online at http://www.academicjournals.org/AJB
DOI: 10.5897/AJB12.3740
ISSN 1684–5315 © 2012 Academic Journals

Full Length Research Paper

Lactic acid fermentation from refectory waste: Factorial

design analysis
Didem OMAY1* and Yuksel GUVENILIR2
1
Yalova University, Faculty of Engineering, Chemical and Process Engineering, Yalova, Turkey
2
Istanbul Technical University, Faculty of Chemical and Metallurgical Engineering, Department of Chemical Engineering,
Istanbul, Turkey
Accepted 30 March, 2012

A factorial experimental design method was used to optimize the lactic acid production using
Lactobacillus bulgaricus from refectory waste obtained from Istanbul Technical University mess hall,
Turkey. Fermentation experiments were carried out in a batch type reactor system which contains
refectory waste with Lactobacillus bulgaricus bacteria during an exposition time of 30 h. Factorial
design of experiments was employed to study the effect of three factors namely temperature (30 and
45°C), substrate concentration (10 and 60 g/L) and pH (4.0 and 6.0) at two markedly different levels. The
main effects and interaction effects of the three factors were analysed using statistical techniques. A
regression model was recommended and it was found to fit the experimental data very well. The results
were analysed statistically using Student’s t-test, and analysis of variance was used to define the most
important process variables affecting the production of lactic acid by fermentation. In the present
study, the most significant factor affecting lactic acid fermentation was found to be the initial substrate
concentration.

Key words: Refectory waste, lactic acid, factorial design, Lactobacillus bulgaricus.

INTRODUCTION

Lactic acid is an organic acid (α-hydroxy-propionic acid) body and this property has resulted in the application of
used for a wide variety of industrial applications. In food lactic acid in biomaterial and biomedical field (Hunger,
industry, it is used as an acidulant, a preservative and an 1984).
antimicrobial agent. For pharmaceutical applications, Lactic acid is produced by chemical synthesis and by
lactic acid can be used as electrolytes and mineral microbial fermentation. By chemical synthesis method,
sources. For technical applications lactic acid can be racemic mixture of lactic acid is produced, while by
used as neutralizers, solvents, cleaning agents, slow acid microbial fermentation method L(+) and D(-) lactic acids
release agents and metal complexing agents. It has also can be produced according to the type of microorganism
been used in cosmetic industry as pH buffer, which may be homofermentative or heterofermentative.
antimicrobial, skin rejuvenating and skin lightening. A This is an important advantage of the microbial fermen-
large number of carbohydrate materials have been used, tation method compared to the chemical synthesis
tested or proposed for the manufacture of lactic acid by method. At the end of the fermentation process, lactic
fermentation (Vick Roy, 1985). There are two isomers of acid exists in the complex medium of fermentation broth
lactic acid, these are D(-) and L(+) forms, which differ that contains whey proteins, biomass, salts and other
only in their optical properties, but are identical in their impurities. Lactic acid is then recovered from this
physical and chemical characteristics. L(+)-Lactic acid is complex medium. Since the high cost of lactic acid
biodegradable and can be metabolized by the human purification process limits the utilization of this chemical,
in a large scale application, a system with less raw
material and fewer unit operations are needed
(Narayanan et al., 2004).
*Corresponding author. E-mail: [email protected]. Sucrose from cane and beet sugar, whey containing
7694 Afr. J. Biotechnol.

lactose, and maltose and dextrose from hydrolysed fermentations.

starch are presently used commercially for lactic acid
production. Over the years, authors have studied a large Composition of refectory waste
number of carbohydrates and nitrogenous materials for
production of lactic acid. These have been investigated Reducing sugar and total carbohydrates of refectory waste was
on the basis of high lactic acid yields, optimum biomass estimated by dinitrosalicylic acid and phenol sulphuric acid
production, negligible by-product formation, fast methods, respectively. Physiochemical properties of refectory waste
fermentation rate, less pre-treatment, easy downstream were determined using standard protocols. The pH of the sample
was determined using a digital pH meter.
processing, low cost, ease of availability etc. The choice
of the raw material to be used depends on the
microorganisms studied and also on the product desired. Lactic acid fermentation
Sucrose, lactose, maltose, glucose, mannitol etc. have
been commercially used (Narayanan et al., 2004). Batch experiments were performed in a temperature-controlled
incubator shaker operated at 160 rpm, at 37°C. The shake flasks
Batch fermentations are widely used method for the
were 250 ml Erlenmeyer flasks containing 100 ml of refectory waste
production of lactic acid. Fermentation conditions are as fermentation medium (initial pH 6). Unless otherwise indicated,
different for each industrial producer but are typically in refectory waste was dissolved to attain 58 g/L of initial sucrose
the range of 45 to 60°C with a pH of 5.0 to 6.5 for concentration and supplemented with (g/L) yeast extract (10),
Lactobacillus delbrueckii; 43°C and a pH of 6.0 to 7.0 for K2HPO4 (0.5), KH2PO4 (0.5), MgSO4 (0.2) and MnSO4.H2 O (0.05).
Lactobacillus bulgaricus. The acid formed is neutralized Refectory waste medium (pH 6.0) and all salt solutions were
sterilized separately at 121°C for 15 min. Sterile CaCO3 (10% (w/v)
by calcium hydroxide or calcium carbonate. The of the initial sucrose concentration) was added to the medium to
fermentation time is 1 to 2 days for 5% sugar sucrose neutralize the acid. The shake flasks were inoculated aseptically
such as whey and 2 to 6 days for a 15% sugar source with 30-h-old fresh culture propagated in medium at 30 and 45°C
such as glucose or sucrose. Under optimal laboratory (Mel et al., 2008).
conditions the fermentation takes 1 to 2 days. The yield
of lactic acid after the fermentation stage is 90 to 95 wt%
Lactic acid determination
based on the initial sugar or starch concentration (John et
al., 2006). The fermentation rate depends primarily on the Lactic acid concentrations were analysed by high performance
temperature, pH, concentration of nitrogenous nutrients, liquid chromatography (HPLC). The HPLC system was composed
and initial substrate concentration. The undissociated, of Agilent 1100 Series. The mobile phase was 5 mM H2 SO4 for
electroneutral form of lactic acid rather than lactate Aminex HPX-87H column. Mobile phase was filtered through 45 µM
cellulose acetate filter papers after solution preparation (Okano et
appears to be the components which inhibits the
al., 2009).
fermentation (Vick Roy, 1985).
In the present study, lactic acid was produced from re-
fectory waste, which was collected from Istanbul Experimental design and statistical analysis
Technical University mess hall, through fermentation
process and convenient conditions were optimized for the In the optimization studies using factorial design analysis, initial
substrate concentration, pH and temperature were varied as
production of the highest lactic acid yield via experimental parameters, while the levels of other medium components were
design analysis. The most important factors for the kept constant. The statistical analysis of the data was performed
production of lactic acid using fermentation process are using Minitab Statistical Software (Release 14). In this design, there
the substrate concentration, pH and temperature. were two experimental levels (-1, +1) where –1 and +1
Factorial design of experiments was employed to study corresponded to low level and high level of each variable,
the effect of these three factors on the lactic acid respectively (Kotzamanidis et al., 2002).
production. Also, this work represents the first in the
literature to follow the factorial experimental design of RESULTS
lactic acid production using Lactobacillus bulgaricus
from refectory waste. Composition of refectory waste

MATERIALS AND METHODS

Samples of waste collected from Istanbul Technical
University mess hall were analysed for moisture, ash,
Microorganism and culture conditions protein, total reducing sugar and total carbohydrates,
nitrogen and protein contents (Table 1). The refectory
L. bulgaricus (DSMZ 20081) was used through the study. The waste contained mainly carbohydrate components. Total
microorganism was maintained on De Man, Rogosa and Sharpe carbohydrate (total reducing sugar + sucrose) and total
(MRS) agar plates at 4°C and sub-cultured every 15 days. Cells for
inoculation of the production medium at a level of 10% (v/v) were
reducing sugar content of the refectory waste were
obtained from cultures grown on MRS broth (pH 7.0) at 37°C for 24 confirmed as 61.38 and 0.47% respectively. During the
h in the incubator and kept at 4°C in the refrigerator. Twenty-four inversion of 1 mole of sucrose, it reacts with 1 mole of
hour (24 h) old fresh cultures were used as the inoculum for the water and 95 g of sucrose to produce 100 g of reducing
 Omay and Guvenilir 7695

Table 1. Composition of refectory waste.

Composition of medium Percentage (%)

Moisture content 7.50
Total Ash 5.68
Total reducing sugar 0.47
Total carbohydrate 61.38
Nitrogen 2.85
Protein 10.41

Table 2. Levels of factors.

Factor Low (-1) High (+1)

pH 4.0 6.0
Temperature, T (°C) 30 45
Initial substrate concentration, S0 (g/L) 10 60

Table 3. Experimental factorial design results for lactic acid.

Factor Produced lactic acid (g/L)

T S0 pH SF1 SF2 SF3
-1 -1 -1 8 6 6
+1 -1 -1 13 10 17
-1 +1 -1 20 18 22
+1 +1 -1 29 32 30
-1 -1 +1 15 12 13
+1 -1 +1 24 18 23
-1 +1 +1 27 25 29
+1 +1 +1 43 47 40
T, Temperature; S0, ınitial substrate concentration.

sugar. From this principle, sucrose content of the determined. The effect of a factor was in the change in
refectory waste was calculated as difference between response and production of lactic acid by a change in the
total carbohydrates and reducing sugar multiplied by 0.95 level of a factor, pH, temperature and initial substrate
and sucrose content and was estimated at 58 g/L. concentration from lower to higher level (Table 2). Eight
The moisture content of refectory waste was found in experiments were carried out and each of them was
the range of 5 to 10%, the nitrogen content in waste was replicated three times. All possible combinations of
2.85% and ash content at range 5 to 6%. The lactic acid factors were used and a matrix was established
bacteria require substrates with high nitrogen content according to the high and low levels represented by +1
during fermentation. The nutrients were added in the form and -1 respectively (Table 3).
of malt sprout, corn steep liquor, and yeast extract. Lactic The main effects represented deviations of the average
acid production increased with the concentration of the between high and low levels for each one of them. When
supplement especially yeast extract. The highest the effect of a factor was positive, production of lactic
production rate was found with addition of 5 to 15 g/L acid increased as the factor was changed from low to
yeast extract (Lund et al., 1992). Although the refectory high levels. The results were analysed using the software
waste contained very little nitrogen, this concentration for a 95% confidence level (α = 0.05) and main effects
was adequate for lactic acid bacteria growth in the and interactions between factors were examined. The
present study. effects, regression coefficients, standard errors and p
were shown in Table 4. The mathematical model
3
Experimental analysis employed for the 2 factorial design was:

The results were analysed using Minitab 14 for windows. SF = A0 + A1*T + A2*S0 + A3*pH + A4*T*S0 + A5*T*pH +
The main effects and interaction between factors were A6*S0*pH + A7*T*S0*pH........................ (1)
7696 Afr. J. Biotechnol.

Table 4. Statistical parameters for 23 design.

Term Effect Coefficient Standart error t-statistic p

Constant 21.9583 21.9583 0.5052 43.47 0.000
T 10.4167 5.2083 0.5052 10.31 0.000
S0 16.4167 8.2083 0.5052 16.25 0.000
pH 8.7500 4.3750 0.5052 8.66 0.000
T*S0 2.9167 1.4583 0.5052 2.89 0.011
T*pH 1.9167 0.9583 0.5052 1.90 0.076
S0*pH 1.2500 0.6250 0.5052 1.24 0.234
T*S0*pH 1.0833 0.5417 0.5052 1.07 0.300

Figure 1. Main effects plot for produced lactic acid. T, Temperature; S0, ınitial substrate concentration.

Where A0 represents the global mean and Ai represents (parallel to the x-axis) indicated that no main effect was
the other regression coefficients. Substituting the present (the control factor does not influence the
coefficient Ai in Equation (1) by their values from Table 4 objective function). If the line was not horizontal, there
we got model equation: could be a main effect present and in this case the
control factors influence the objective function. The
SF = 21.9583 + 5.2083T + 8.2083S0 + 4.3750pH + greater the slope of the line, the stronger the effect
1.4583T*S0 + 0.9583T*pH + 0.6250S0*pH + produced.
0.5417T*S0*pH…………………………………………… (2)

Equation (2) presented that effects of all factors were Student’s t-test
positive and results in an increase in the value of the
produced lactic acid. The main effects of the control The Pareto chart (Figure 2) gave the relative importance
factors were presented in Figure 1. A horizontal line of the individual and interactions effect. Student’s t-test
 Omay and Guvenilir 7697

2.12

Figure 2. Pareto chart of standardized effects on the produced lactic acid concentration. T, Temperature;
S0, ınitial substrate concentration.

Table 5. Analysis of variance.

Source Degrees of freedom Sum of square Mean square F p

Main effects 3 2727.46 909.153 148.43 0.000
2-way interactions 3 82.46 27.486 4.49 0.018
3-way interactions 1 7.04 7.042 1.15 0.300
Residual error 16 98.00 6.125
Pure error 16 98.00 6.125
Total 23 2914.96
2 2 2 2
R = SSMODEL/SSTOTAL; R adj = 1 – [(SSERROR/DFERROR)/(SSTOTAL/DFTOTAL)]; R = % 96.64 R (adj) =% 95.17.

was performed in order to determine whether the that R2 and R2(adj) values which were important due to test
calculated effects were significantly different from zero obtained mathematical model were close to each other
2 2
and these values for each effect were shown in Pareto and approximately 1.0 (R = 96.64%, R (adj) =95.17%) that
chart by horizontal columns. In our case, all control of expected result statistically. Another important aspect
factors with a significant influence were located over the was the interaction among the control factors. The
line marked at 2.12 (p = 0.05). The vertical line in the interaction between the control factors can be estimated
chart indicated the minimum statistically significant effect from experimental design and the results were presented
magnitude for 95% confidence level. in Figure 3. If the lines were parallel to each other, there
was no interaction present. With the increase of the
deviation degree of line from being parallel, the inter-
Analysis of variance (ANOVA) actions among control factors increase. The residues
were also examined for normal distribution. Figure 4
In Table 5, the sum of squares used to estimate the shows the normal probability plot of residual values. It
factors’ effects and F ratios are shown. It can be said could be seen that the experimental points were
7698 Afr. J. Biotechnol.

Figure 3. Interaction effects for produced lactic acid. T, Temperature; S0, ınitial substrate concentration.

reasonably aligned, thus suggesting normal distribution. growth increased, however, until up to a certain limit.
Beyond initial pH 6.5, its growth rate decreased again.
Therefore, the optimal initial pH for the refectory waste
DISCUSSION fermentation of L. bulgaricus was 6.0, which is similar to
those reported by Goksungur and Guvenc (1997) by
Lactic acid starters are currently produced using pH using beet molasses as a substrate. Various researchers
controlled pure cultures (Beal et al., 1989), during which have studied the effect of pH on lactic acid production
pH is generally regulated at an optimal value by conti- and found that the optimum pH for lactic acid production
nuously adding sodium hydroxide or ammonia in the is in the range of 5.0 to 7.0 (Hofvendahl and Hagerdal,
bioreactor (Savoie et al., 2007). Various growth charac- 2000; Goksungur and Guvenc, 1997). Goksungur and
teristics such as maximal biomass concentration, specific Guvenc (1997) showed that the effect of pH on lactic acid
growth rate, fermentation time, substrate consumption production was important and the optimal pH was 6.0
and product yields are influenced by the pH value with the yield value 79%. That is why when the statistical
(Adamberg et al., 2003). Optimal pH ranges were analysis of the data was performed using Minitab, the
therefore determined for several lactic acid bacteria, such levels of pH factors used in the experimental design were
as Streptococcus thermophilus (pH 6.5), Lactococcus chosen as pH 4.0 and 6.0 (Table 2). Also, temperature is
lactis subsp. cremoris (pH 6.3 to 6.9) and L. bulgaricus one of the most important environment factors that effect
(pH 5.8 to 6) (Beal et al., 1989). The effect of initial pH on the lactic acid production. Various studies on the effect of
the cell growth of L. bulgaricus during the fermentation of temperatureon the lactic acid production have reported
refectory waste was investigated and optimized in the an optimal temperature between 40 to 45°C (Hofvendahl
present study. and Hagerdal, 2000). Goksungur and Guvenc (1997)
According to the experimental results, at the initial pH reported that the optimal temperature was 45°C and this
of 4.0, the bacteria exhibited a prolonged lag phase and might be due to the different substrates used in the lactic
bacteria did not grow as well as at higher initial pH value. acid fermentation. Maximum yield obtained with 53.61 g/L
Moreover, as the initial pH increased above 4.0, the cell of lactic acid when the temperature was 45°C, and the
 Omay and Guvenilir 7699

Figure 4. Normal probability plot of residual values for concentration of lactic acid vs. their expected
values when the distribution is normal.

lactic acid production decreased rapidly to 25.14 g/L whose value was 16.4167, was the most significant effect
Lactobacillus helveticus used in a temperature range of on the production of lactic acid. After that respectively T
35°C. For similar reason the causes of choosing the pH and pH main effects and T*S0 binary interaction
levels, temperature levels were chosen as 30 and 45°C presented the statistical significance. Other binary and
(Rakın et al., 2004). trio interactions were not statistically significant: T*pH,
In the present study, the Pareto chart showed that S0*pH and T*S0*pH. Similar results were obtained when p
substrate concentration has a highly significant effect on values were evaluated. From the p value which was
the produced lactic acid concentration (Figure 2). defined as the smallest level of significance leading to
Additionally, for initial substrate concentration at 60 g/L, rejection of the null hypothesis, it appears that the main
concentration of lactic acid was estimated as 20 g/L. This effect of each factor and the interaction effects were
value was considerably higher than the obtained value statistically significant when p<0.05. So, it can be said
using initial substrate concentration as 10 g/L (Table 3); that the main effects of T, S0, pH, and interaction effect
this situation was considered ideal in batch type T*S0 were statistically significant. On the other hand, p
fermentations. It was observed that as the initial substrate values of T*pH, S0*pH and T*S0*pH interaction effects
concentration increased to a very high point, production were higher than 0.05, so these interaction effects were
of lactic acid decreased clearly (results were not given), a not statistically significant (Table 4).
phenomenon that can occur by substrate inhibition, The effects of the all variables (pH, temperature and
product inhibition or exhaustion of one restricting nutrient initial substrate concentration) and their interactions on
or their combined effect. It was reported that in batch the formation of the lactic acid production were illustrated
type, lactic fermentations have varying substrate with analysis of variance (Table 5). The goodness of fit of
concentrations from 20 to 100 g/L; the results for lactic the analysis of variance model was checked by the
2
acid concentrations and sucrose conversion obtained determination coefficient (R ). In this case, the value of
2
were similar to those reported in this study. It was also the R (0.9664) for ANOVA indicates that the sample
reported inhibition by substrate in fermentations occur variation of nearly 97% for lactic acid was attributed to
using Lactobacillus casei NRRL B-441, and varying the the independent variables and only 3% of the total
glucose concentration between 80 and 160 g/L (Hujanen variation could not be explained by the model. The value
et al., 2001). According to these results, the effects of the of the adjusted determination co efficient (R2adj = 0.9517)
high and low levels for the initial substrate concentration was also high, which stressed the significance of the
were chosen as 10 and 60 g/L in paralleling to the other model. The high value of R (0.9664) demonstrated a high
variables. degree of agreement between the experimental obser-
It was observed that initial substrate concentration (S0) vations and predicted values.
7700 Afr. J. Biotechnol.

Conclusion Lund B, Nordahl B, Ahring B (1992). Production of lactic acid from whey
using hydrolysed whey protein as nitrogen source. Biotechnol. Lett.
14: 851-856.
In this study, important process variable factors which Mel M, Karim MIA, Salleh MRM, Amin NAM (2008). Optimizing media of
affect fermentative lactic acid production were deter- Lactobacillus rhamnosus for lactic acid fermentation. J. Appl. Sci.
mined using a factorial experimental design technique. 8(17): 3055-3059.
The results of statistical study clearly showed that initial Narayanan N, Roychoudhury PK, Srivastava A (2004). L(+) lactic acid
fermentation and its product polymerization. Electron. J. Biotechnol.
substrate concentration was the most important 7(2): 167-179.
parameter. Moreover the main effect of temperature, pH Okano K, Zhang Q, Shinkawa S, Yoshida S, Tanaka T, Fukuda H,
and interaction between temperature-initial substrate Kondo A (2009). Efficient production of optically pure D-Lactic acid
from raw corn starch by using a genetically modified L-Lactate
concentrations had a considerable effect on the amount
dehydrogenase gene deficient and alpha amylase secreting
of lactic acid produced. Other two-way and three-way Lactobacillus plantarum strain. Appl. Environ. Microbiol. 75(2): 462-
interactions did not exhibit any statistical significance. 467.
Rakın M, Baras J, Vukasinovic M, Maksimovic M (2004). The
examination of parameters for lactic acid fermentation and nutritive
value of fermented juice of beetroot, carrot and brewer’s yeast
REFERENCES
autolysate. J. Serb. Chem. Soc. 69(8-9): 625-634.
Savoie S, Champagne CP, Chiasson S, Audet P (2007). Media and
Adamberg K, Kask S, Laht TM, Paalme T (2003). The effect of
process parameters affecting the growth, strain ratios and specific
temperature and pH on the growth of lactic acid bacteria: a pH-
acidifying activities of a mixed lactic starter containing aroma-
auxostat study. Int. J. Food Microbiol. 85: 171-183.
producing and probiotic strains. J. Appl. Microbiol. 103: 163-174.
Béal C, Louvet P, Corrieu G (1989). Influence of controlled pH and
Vick Roy TB (1985). Lactic acid. in Comprehensive biotechnology: the
temperature on the growth and acidification of pure cultures of
principles, applications, and regulations of biotechnology in industry,
Streptococcus thermophilus 404 and Lactobacillus bulgaricus 398.
agriculture and medicine, ed Moo-Young M. (Pergamon Press, New
Appl. Microbiol. Biotechnol. 32: 148-154.
York), 2: 761-776.
Goksungur Y, Guvenc U (1997). Continuous production of lactic acid
from beet molasses by L. Delbrueckii IFO 3202. J. Chem Eng.
Biotechnol. 69: 399-404.
Hofvendahl K, Hagerdal BH (2000). Factors affecting the fermantative
lactic acid production from renewable resources. J. Enzymes,
Microbial Technol. 26: 87-107.
Hujanen M, Linko S, Linko Y, Leisola M (2001). Optimization of media
and cultivation conditions for L(+) (S)-lactic acid production by
Lactobacillus casei NRRL B- 441. Appl. Microbiol. Biotechnol. 56(1-
2): 126-130.
Hunger W (1984). Dextro-rotatory and levo-rotatory lactic acid: Their
significance and occurrence in sour milk products. Danish Dairy Food
Industry Worldwide, 4: 39-42.
John RP, Nampoothiri KM, Pandey A (2006). Simultaneous
saccharification and L(+) lactic acid fermentation of protease-treated
wheat bran using mixed culture of lactobacilli, Biotechnol. Lett. 28(2):
1823-1826.
Kotzamanidis Ch, Roukas T, Skaracis G (2002). Optimization of lactic
acid production from beet molasses by Lactobacillus delbrueckii
NCIMB 8130. World J. Microb. Biot. 18(5): 441-448.