259

Vol.48, n. 2 : pp. 259-266, March 2005 ISSN 1516-8913 Printed in Brazil

BRAZILIAN ARCHIVES OF BIOLOGY AND TECHNOLOGY

A N I N T E R N A T I O N A L J O U R N A L

Extraction of Pectin From Apple Pomace

Maria Helene Canteri-Schemin1, Helosa Cristina Ramos Fertonani2, Nina Waszczynskyj3 and Gilvan Wosiacki2*
Centro Federal de Educao Tecnolgica do Paran; Unidade de Ponta Grossa; Coordenao de Alimentos; Ponta Grossa - PR - Brazil; 2 Universidade Estadual de Ponta Grossa; Departamento de Engenharia de Alimentos; Av. Carlos Cavalcanti, 4748; 84030-900; Ponta Grossa - PR - Brazil; 3Universidade Federal do Paran; Departamento de Engenharia Qumica; Ps-Graduao em Tecnologia de Alimentos; Curitiba - PR - Brazil.
1

ABSTRACT
As apple-processing units are now in expansion in Brazil, industrial by-products like pomace play an important role in pectin manufacture. The objective of this article was to determine a pratical follow-up to the extraction of pectin from apple pomace and to characterize it in a laboratory, on a small scale, aiming at establishing the optimum conditions for acid extraction. The highest yields were obtained when [1]apple pomace was dried and ground to obtain an apple flour to be used as raw material, [2] citric or nitric acids were used and [3] when the citric acid concentration was 6.2 g/100 ml and the time of reaction was 153 minutes. The apple variety in itself was not significant in pectin yield. The degree of esterification (DE = 68.84 %) of the product obtained, as well as its physical looks, show the success of pectin extraction. Key words: Apple pomace, pectin, agricultural wastes, yield, extraction, Response Surface Methodology (RSM)

INTRODUCTION
Citrus albedo and apple pomace are rich in pectic substances and important raw materials for pectin production all around the world (Fox, 1984). There is only one factory in Brazil producing citrus pectin, in Limeira, State of So Paulo (Cpkelco, 2002) but none producing apple pectin. The Brazilian apple production, however, quickly spread in the Southern States of the Country in the 70s (Wosiacki, 2001) and in the 2002 / 2003 crop, the apple production was estimated at 980,000 tons, with approximately 180,000 tons for industrial processing. Nowadays, about 50,000 tons of wet pomace are produced in Brazil. A possibility for large enterprises is the drying of this wet raw material aiming at the production of apple pectin (Endress, 2000). Apple pomace is produced
*

as a by-product of the juice factories and currently it is either used for animal feeding or is disposed of as an industrial waste. If apple pomace could be used the production of pectin, around 2000 tons of pectin could be produced each year. Pectin is a family of complex variable polysaccharides extracted from the primary cell wall of higher plants. Chemically, pectin consists of linear polymers of D--(14) anhydrogalacturonic acid. Part of the carboxyl groups of the anhydro-galacturonic acid is esterified with methanol (Wosiacki, 1977). Vauquelin stated its chemical nature in 1790 and Braconnot showed the characteristic of geleification and gave it the name pectin (Berk, 1976). Pectin is widely used as a functional ingredient in the food industry due to its ability to form aqueous gels and has been used in jams and jellies, fruit preparations, fruit drink

Author for correspondence

Brazilian Archives of Biology and Technology

260

Canteri-Schemin, M. H. et al.

concentrates, fruit juice, desserts and fermented dairy products. Commercial pectin is currently classified according to the degree of esterification (DE). There are three classifications of pectin: HM (high ester); LMC (low ester conventional) and LMA (low ester amidated) according to CPKelco (2002). Pectin in apple pomace is mainly present in the form of protopectin, an acid soluble polysaccharide. The condition for the acid extraction of pectin is somewhat obscure in the literature and it is very difficult to visit processing units. There are complicated procedures in its study (Thibault et alli, 1991), although pectin has been under investigation since the start of 20th century (Kertesz, 1951). Some experimental research should be done in order to clarify this important industrial step so as to utilize apple pomace as the main raw material for the extraction of pectin. The objectives of this work were to extract pectin from apple pomace on a laboratory scale and to characterize it in order to observe the influence of some factors on the yield of pectin.

MATERIALS AND METHODS

Materials Samples from nine different apple varieties (Belgolden, Catarina, Eva, Fred Hough, Fuji, Gala, MRC, Rainha and Sansa) were used as raw material in this work. They were processed together in order to obtain an apple pomace pool, according to industrial procedure. In the laboratory, all the fruits were first washed and ground in an electric grater. The crushed pulp was then pressed and the apple pomace dried, initially at room temperature and then at 50C, to a constant weight with air circulation. This dry apple pomace pool was then crushed and mixed and the product was called apple flour and it was used as the raw material for all the assays made concerning pectin extraction and characterization. Three apple flour varieties were selected (Belgolden, Catarina and Eva) in order to establish the influence of apple variety on the yield of pectin extraction and the results were compared with a pomace pool of all the apple varieties. Pectin extraction Five replicates for each assay were made. Pectin was extracted under reflux in a condensation system at 97o C for 30 min (solute/solvent 1:50),

using water acidified with citric acid to pH 2.5, using apple flour (pool) as raw material. Next, the pool was classified with four metallic sieves (1,18 mm, 600 m, 250 m and 106 m) in a shaker device in order to separate them in five portions with different particle sizes. Three batches, as well as crude dried pomace without crushing were used to observe the effect of particle size on pectin yield. The extraction of pectin was performed with apple pomace from selected varieties under equal conditions. Pectin extraction was also performed under reflux at 97oC for 30 min (solute/solvent 1:50), using water acidified with different acids, to investigate the effect of each kind at pH 2.5 and to ratify the use of citric acid. In this assay, the raw material selected was the apple flour pool with particles between 250 and 105 m. The selected acids utilized for extraction were citric, phosphoric, hydrochloric, nitric, malic, tartaric and sulfuric acids. Pectins extracted in each trial were all mixed and the degree of esterification was analyzed (Gee et al., 1959). The analysis of variance (ANOVA) was used to define the optimum conditions concerning the acid used for pectin extraction, the apple variety and the size of particles of the apple pomace. Isolation of pectin Hot acid extract was pressed in a cheese cloth bag and the concentrated juice was cooled to 4o C. The apple pectin was precipitated by alcohol-juice treatment 2:1 (v/v). The mixture of solvent and precipitate was stirred for ten minutes and then left to rest for one hour in order to allow pectin flotation. With this procedure the pectic substances remain at the surface of the alcohol/water mixture and thus it is easier to remove them in a quantitative way. The floating pectin was filtered through cheesecloth, rinsed with 95o GL alcohol and then pressed. The pressed pectin was dried to constant weight at 55 oC, cooled in a dessicator and the yield calculated on a dry weight basis (inicial weight of sample). The hard pectin cake was broken up, ground and sieved in order to obtain powdered pectin. Experimental design Response surface methodology was used to optimize pectin extraction with respect to time and

Brazilian Archives of Biology and Technology

Extraction of Pectin from Apple Pomace

261

acid concentration; the temperature was maintained constant at 97C, the boiling point of water in the laboratory. A five-level two-factor fractional design was adopted. The response surface model equation is (1): Y = bo + b1x1 + b2x2 + b11x1 2 + (1) b22x2 2+b12x1x2 (1)
where Y is the response variable of pectin yield (%); b1 and b2 are regression coefficients for linear effects; b12 is the regression coefficient for interaction effects; b11 and b22 are regression coefficients for quadratic effects, and x1 and x2 are coded experimental levels of the variables. Statistica for Windows was used to match the quadratic equation with the experimental data.

Optimization The apple flour suspensions were heated at 97oC under reflux with a glass condensation device. The factors studied were citric acid concentration (CAC = 0.05-9.95g %) and heating time (HT = 10210 min). Citric acid was always added to the flasks when the system reached the boiling point. The experiment was conducted using a five-level two-factor composite design with three assays at the central point. In order to maximize the pectin extraction, coded and real levels, based on previous experiments, were selected. Both factor variations can be seen in Table 1. The test runs were performed in a random order according to a Statistic for Windows tool.

Table 1 - Coded and real levels of variables Coded level of variables Variable X1 : citric acid, g/100 ml X2 : time, min. 10 39.3 110 180.7 210 0.05 1.5 5 8.5 9.95 -2 -1 0 +1 +2

13 12 11

Yield (%)

10 9 8 7 6 5 Apple flour Pomace Maximum Minimum 75% 25%

Median

Raw

Figure 1 - Effect of particle size on the yield of extraction using apple flour and apple pomace as a pectin source.

Brazilian Archives of Biology and Technology

262

Canteri-Schemin, M. H. et al.

RESULTS AND DISCUSSION

Effect of the size of the particles According to the results (Fig. 1), the pectin yield was significantly higher with the use of flour as raw material (9.73 %); when the extraction was made from the pomace the yield was 6.13%. There are several explanations for this fact, but protopectin is more available in small particles than in large ones. Three apple flour fractions were collected during the study of the particle size and used to confirm these results. The flour fractions with particles larger than 600 m showed lower pectin yield while when smaller particles were used the yield was higher (Fig. 2). Effect of apple variety Although some varieties do have a high concentration of soluble pectin in the apple juice, the pectin extracted from the pomace did not show a pronounced variation, although there was a statistical difference amongst the varieties. The blend of varieties results in a raw material more similar to industrial raw pomace constituted of various apple varieties in indeterminate

proportions and several maturation stages (Fig. 3). Effect of different acids on pectin extraction Several acids can be utilized for the extraction of pectin. According to Kertesz (1951), the acids used for pectin extraction were the tartaric, malic, citric, lactic, acetic and phosphoric acids but there was a tendency to use the cheaper mineral acids, such as sulfuric, hydrochloric and nitric acids. Articles published recently recommend the use of hydrochloric (Kalapathy and Proctor, 2001; Hwang et al., 1998; Dinu, 2001) and nitric acids (Pagn et al., 2001). According to Fig. 4, the lowest yields were obtained when phosphoric and malic acids were used. Although nitric acid showed the highest yield, the variation was very large. Citric acid had the highest average value (13.75%) and it is better than the other acid from an economic as well as from an environmental point of view. The analysis of variance indicated statistical significance in the yield.

Figure 2 - Effect of particle size on the pectin extraction yield.

Brazilian Archives of Biology and Technology

Extraction of Pectin from Apple Pomace

263

16 15 14 13 YIELD (%) 12 11 10 9 8 7 Pool Belgolden Catarina VARIETY

Figure 3 - Effect of apple variety on the pectin extraction yield.

Maximum Minimum 75% 25%

Eva

Median

16 15 14

YIELD (%)

13 12 11 10 Citric Phosphoric Nitric Tartaric Hydrochloric Malic Sulfuric

Maximum Minimum 75% 25% Median

KIND OF ACID
Figure 4 - Effect of the kind of acid on the pectin extraction yield.

Optimization of yield Table 2 shows the experimental design and its results. The average results of the central coordinates were 20 0.86, which represented a low variation coefficient of 4.8%. The regression analysis showed that 93.55 % of the variations was explained by the model. The predictive equation explains a surface with a maximum point with coordinates X1=152.85 minutes and X2= 6.2 g % of citric acid with a

theoretical maximum yield of 17.82 g % of pectin on a dry basis (Equation 2). z=19.9957 + 0.8819x 2.4457x2 +1.3865y (2) 1.5613y2 + 1.4215xy + 0 The results indicated that the yield of pectin was dependent on the linear terms, on the quadratic terms and also on the interactions of both variables. Table 3 illustrates the coefficients of the regression model for the determination of pectin.

Brazilian Archives of Biology and Technology

264

Canteri-Schemin, M. H. et al.

Both linear and quadratic effects showed significance in accordance with the analysis of variance. The experimental design indicated the formation of a surface where the pectin extraction yield was optimum, between 18 and 20%. Fig. 5 illustrates this surface in three dimensions, clearly
Table 2 - Experimental design and pectin extraction yield.

showing a critical point of high yield with an equation statistically significant according to ANOVA, since Fcal (6.005665) >F crt (4.256492). The degree of esterification of the apple pool pectin was 68.84% on average, as could be expected from mature apples.

Level Run 1 2 3 4 5 6 7 8 9 10 11 Coded Real Citric acid (g%) Time (min.) Citric acid (g%) Time (min.) -1 -1 1.5 39.3 -1 1 1.5 180.7 1 -1 8.5 39.3 1 1 8.5 180.7 0 0.05 110 -2 0 9.95 110 2 0 5 10 -2 0 5 210 2 0 0 5 110 0 0 5 110 0 0 5 110 Yield (g%) 15.59 15.88 13.30 19.27 12.98 17.18 15.14 18.56 20.92 19.23 19.83

Table 3 - Analysis of variance of regression model of experimental design.

Factor Mean/ Interaction (1) Citric acid (L) (2) Citric acid (Q) (3) Time (L) (4)Time (Q) 1L * 2L
L= linear effects Q= quadratic effects

Regression Coefficient 19.9956 1.7639 -4.8913 2.7730 -3.1225 2.8430

Standard Error 0.5580 0.6835 0.8136 0.6835 0.8136 0.9665

T(s) 35.8305 2.5805 -6.0114 4.0568 -3.8376 2.9412

P 3.19E-07 0.0493 0.0018 0.0097 0.0121 0.0322

-95% Cnf.Limt 18.5611 0.0068 -6.9829 1.0159 -5.2142 0.3583

+95% Cnf.Limt 21.4302 3.5209 -2.7997 4.5301 -1.0309 5.3277

CONCLUSIONS
When the effects of particle size were studied it was found that the highest average yields of around 14% were obtained when the particle was larger than 106 m and smaller than 250m. Extraction of pectin using pomace as raw material produced a lower pectin yield than when apple flour was used. This result indicated that it was necessary to produce apple flour as an intermediary step in the acid extraction of pectin from pomace. Pectin extraction from different

varieties of apple pomace did not show any significant effect on pectin yield as compared with a pool of apple samples. Citric acid and nitric acid showed the highest yield among the organic and mineral acids tested. The coordinates of the stationary point were found to correspond to 6.2 g/100 ml of citric acid with a reaction time around 150 min. The apple pool pectin showed a degree of esterification of approximately 68.84%.

Brazilian Archives of Biology and Technology

Extraction of Pectin from Apple Pomace

265

YIELD
24 20 16 12 8 4
3.231 4.963 6.695 8.426 10.158 11.890 13.622 15.353 17.085 18.817 above

1.8 1.2 1.8 1.2 0.6 0.6 0.0 0.0 -0.6 -0.6 -1.2 -1.2 -1.8 -1.8

TIME

ACID

Figure 5 - 3-D dimension response surface of experimental design (five-level two-factor composite) on the pectin extraction yield.

RESUMO ACKNOWLEDGEMENTS
No aproveitamento de resduos das indstrias processadoras de ma, em expanso no Brasil, a pectina alimentcia surge como importante subproduto. Os objetivos deste trabalho foram os seguintes: extrao e caracterizao da pectina de bagao de ma em bancada, em que as condies para maior rendimento de obteno em relao ao tipo de cido, ao tamanho das partculas e variedade de ma foram estabelecidas. Objetivouse tambm a otimizao da produo, levando-se em conta o tempo e a concentrao do cido ctrico. Foram considerados mtodos de extrao mais eficientes quanto ao rendimento: [1] acidificao com cido ctrico ou cido ntrico, [2] cominuio do bagao, usando a farinha como matria-prima, com partculas retidas no tamis de 106 m, [3] concentrao de cido de 6,2 g% e tempo de extrao de 153 minutos. A variedade da ma no interferiu significativamente no rendimento. A alta metoxilao (DE= 68,84%) das pectinas produzidas em bancada confirma o grau de maturidade das matrias-primas usadas nos experimentos. The experimental work was done at the Ponta Grossa State University and the authors are deeply grateful to the staff of GTM - Apple Research Team (Departamento de Engenharia de Alimentos) for the technical and the scientific support. The authors are also grateful to CNPq for scholarship and to ABPM, to CEFET/PR-U-PG, to AGRICOLA FRAIBURGO, to CPKelco and to EPAGRI for technical cooperation.

REFERENCES
Berk, Z. (1976), Braverman's introduction to biochemistry of foods. Amsterdam: Elsevier, 375 pp. Chang, K. C.; Dhurandhar, X. and You, M. A. (1994), Cultivar/Location and processing methods affect yield and quality of sunflower pectin. Journal of Food Science, 59, 602-605. CPKelco (2002), Fluxograma de produo de pectina ctrica,Folder, Limeira, So Paulo. Dinu, D. (2001), Extraction and characterization of pectins from wheat bran. Roumanian Biotechnology Letter, 6, 37-43. Endress, H. U. (2000), High quality resulting from product integrated environment protection - PIUS. Fruit Processing, 10, 273-276.

Brazilian Archives of Biology and Technology

266

Canteri-Schemin, M. H. et al.

Fox, G. (1984), Zur Wirtschaftlichkeit der Trocknung von Apfeltrester. Confructa Studien, 28, 174-182. Gee, M.; Reeve, R. M. and Mcready, R. M. (1959), Reaction of hydroxylamine with pectinic acids. Chemical studies and histochemical estimation of the degree of esterification of pectic substances in fruit. Journal of Agricultural and Food Chemistry, 07, 34-38. Hwang, Jae-Kwan; Kim, Chul-Jin and Kim, Chong-Tai (1998), Extrusion of Apple Pomace Facilitates Pectin Extraction. Journal of Food Science, 63, 841-844. Kalapathy, U. and Proctor, A. (2001), Effect of acid extraction and alchool precipitation conditions on the yield and purity of soy hull pectin. Food Chemistry, 72, 393-396. Kertesz, Z. I. (1951), The pectic substances. New York: Interscience. pp. 438. Pagn, J.; Ibarz, A.; Llorca M.; Pagn A. and BarbosaCnovas G. V. (2001), Extraction and characterization of pectin from stored peach pomace. Food Research International, 34, 605-612. Thibault, J. F.; Saulnier, L.; Axelos, M. A. V. and Renard, C. M. G. C. (1991), Difficults exprimentales de l`tude des macromolecules pectiques. Bulletin Societ Botanique Franaise, 138, 319-337. Wosiacki, G. (1977), Enzimas pectinolticas de Fusarium oxysporum Schlecht ex. Fr. Isolado de frutos de caf, Thesis, Universidade Estadual de Campinas, So Paulo, Brasil. Wosiacki, G. and Nogueira, A. (2001), Apple varieties growing in subtropical areas. The situation in Paran State. Fruit Processing, 11, 177-182.

Received: May 14, 2003; Revised: December 08, 2003; Accepted: December 13, 2004.

Brazilian Archives of Biology and Technology