INTERNATIONAL SYMPOSIUM TOWARDS SUSTAINABLE LIVELIHOODS AND ECOSYSTEMS IN MOUNTAINOUS REGIONS 7-9 March 2006, Chiang Mai, Thailand

To whom correspondence should be addressed: neidhasy@uni-hohenheim.de

Integrated mango processing into pulp products rich in provitamin A and value-adding by-products

Sybille Neidhart‡,, Ana Lucía Vásquez-Caicedo‡, Suparat Sirisakulwat‡,+ , Susanne Schilling‡, Pittaya Sruamsiri+ , and Reinhold Carle‡

‡ Institute of Food Technology, Section Plant Foodstuff Technology, University of Hohenheim, 70593 Stuttgart, Germany; and

+ Department of Horticulture, Faculty of Agriculture, Chiang Mai University, Chiang Mai 50200, Thailand

AbstractWith regard to the selection of suitable mango cultivars at appropriate processing ripeness, ac-cumulation of β-carotene during postharvest ripening of nine Thai mango cultivars was assessed. The vitamin A potential was evaluated at different ripening stages unequivocally defined by a ripening index. When subjected to identical postharvest ripening conditions, only those cultivars that developed a bright yellow-orange mesocarp color yielded high vitamin A values of 892-1,573 retinol equivalents (RE)/100 g of mesocarp dry weight at their fully ripe stage. For each cultivar, the postharvest development of mesocarp color and all-trans-β-carotene levels, respec-tively, were described by their exponential dependencies on the ripening index, allowing the se-lection of fruits of high vitamin A value for processing through easily accessible quality parame-ters. Similarly, the influence of postharvest fruit ripeness on extractability, composition and functional properties of pectins from selected cultivars rich in provitamin A was studied to evaluate the suitability of these varieties in terms of integrated mango processing. The presented data proved the manifold potential of selected northern Thai mango cultivars as high-value agri-cultural products. From a technological point of view, further options for product differentiation were opened by high β-carotene stability in the production of fluid mango products at various scales and the recovery of pectin from industrial mango waste. Various possibilities and prereq-uisites for mountainous regions to benefit from this potential were discussed.

Keywords: Mangifera indica L., β-carotene, pectin, postharvest ripening, processing technol-ogy, food quality

1 IntroductionLike vegetables, fruits belong to the high-value agricultural commodities, the production and marketing of which have been an important source of cash income for small-scale farmers in the northern parts of Thailand, since the specific agro-ecological situation renders the production of common staple crops less efficient than in other regions. Cultivation of perennial fruit trees has been assumed to contribute to erosion control to a much larger extent and to consume less water in irrigated agriculture than most annual crops. To increase overall productivity by reduction of postharvest losses resulting from seasonal overproduction due to short harvest periods and insuf-ficient access to markets, technological options are required for diversified fruit utilization at a small- and medium-sized level with respect to regional added value. Diversification aims at the use of various quality types and size classes of raw fruits in addition to fresh fruit marketing. Concomitantly, optimized fruit products for various markets are needed, since local industrial fruit processing has mostly been oriented towards export markets. As recent free trade agree-

2

ments and market liberalization have increased national and international competition, partly leading to significant price decreases, product differentiation promoting niche products is com-ing to the fore. Globalization may induce further changes in agricultural and food markets in developing countries, also due to changing demands in urban centers and supply chains. High-value products and the need for product differentiation may increasingly become important, in particular for the small farm sector. Depending on the scale and type of production systems, sub-tropical mangoes offer interesting options for farmers in the Southeast Asian highlands as a provitamin A source and as cash crops with high export potential both as fresh and processed fruit. The aim of this contribution was to identify the potential of northern Thai mango cultivars for product differentiation, particularly as to health attributes of mango products, contributing to a generally improved provitamin A supply.

The mango (Mangifera indica L.) is naturally adapted to tropical lowlands between 25 °N and 25 °S of the equator and up to elevations of 3,000 ft (915 m) (Morton, 1987), but can be grown up to 1,200 m in tropical latitudes (Bally, 2004). However, most commercial varieties do not produce consistently above 600 m (Bally, 2004), as this fruit grows best in seasonally wet/dry climate zones of the lowland tropics, or frost-free subtropical areas. Mango trees made up re-spectively 42 %, 34 %, and 23 % of the fruit trees on foothills, hills (500-1,800 m), and terraces (400-500 m) in 64 orchards investigated in the northern Thai region of Phrao in 1993 (de Bie, 2004). According to the comparative performance analysis of northern Thai orchards for this region, one of the factors that contributed to increase in yield was that the orchard was situated on a hill or on soils with relatively high pH or poor water-holding capacity, concordant with good drainage required for mango, although application of supplementary irrigation water had to be possible.

Vitamin A deficiency has been known as a public health problem in more than half of all coun-tries, especially in Africa and South-East Asia (WHO, 2006). Children and pregnant women are the major risk groups. The promotion of home gardens with fruits and vegetables rich in provi-tamin A has been one of the suggested long-term strategies. Here, mangoes play a key role among the tropical fruits, as they are generally known to be particularly rich in provitamin A. On the other hand, these fruits are important cash crops with a high export potential. After India (38.6 %) and China (13.1 %), Thailand is the third largest mango producing country with 6.4 % of the world production, closely followed by Mexico (5.4 %). In 2005, world production of mango fruits was 27.97·106 Mt, while world export amounted to 0.92·106 Mt in 2003 (FAOSTAT, 2006). But also their products have economic significance as end products for the consumers or as intermediates in food processing. World production of mango juice and pulp was 153·103 Mt and 931·103 Mt, respectively, in 2005, with corresponding world exports of11.9·103 Mt and 9.5·103 Mt in 2003 (FAOSTAT, 2006). The pulp can be directly consumed as fruit purées or processed into beverages or ingredients for bakery and dairy products. Hence, both the nutritive and the economic benefits of these fruits are interesting for the fruit producers either at small or at large scale.

Modern industrial year-round mango juice production is mostly based on purée intermediates produced during peak harvest seasons. The fruit component in the final nectar usually undergoes several heating treatments in the form of steam peeling, thermal inactivation of endogenous en-zymes prior to enzymatic pulp liquefaction, and pasteurization of purée and nectar, respectively (Dube et al., 2004). However, heat application in continuous industrial processes is restricted to periods below 1 min. In contrast, simple small-scale batch processes at the household level re-quire only the final pasteurization of the filled product, but by heat application for an extended process time. As the contribution of mango fruits and products to provitamin A supply are de-termined by their contents of β-carotene stereoisomers, the effects of thermal treatments and concurrent exposure to light and oxygen on β-carotene degradation and isomerization are of ut-most interest in mango processing (Pott et al., 2003a).

3

Generally, the production of high-quality food equally depends on the selection of suitable raw material and on the appropriate processing technology. In industrialized mango processing, pro-ductivity could be further increased by the transformation of waste into by-products. Compared to fruits of temperate zones, considerably higher ratios of by-products arise from tropical and subtropical fruit processing due to higher amounts of inedible waste material such as peels and seeds, amounting to 35-60% of the total fruit weight for mango with 15-30% of peel and 10-30% of stone (Larrauri et al., 1996; Vásquez-Caicedo et al., 2002). Efficient, inexpensive and envi-ronmentally sound utilization of plant waste material is gaining importance to increase profitabil-ity of fruit processing (Schieber et al., 2001). Mango is considered a potential source of pectin because of high pectin contents (Vásquez-Caicedo et al., 2002) and its overall availability in large quantities (Rehman et al., 2004). Pectins extracted from the peels and residual pulp may bepromising by-products for further use as gelling agents, stabilizers or biological fibers in food and non-food applications. However, composition and functional properties of pectins also de-pend on the processed raw material and the processing technologies used in the production of main and by-product. Hence, the potential of any pectin source strongly depends on the design of processing lines and suitable waste pretreatments prior to pectin extraction, since endogenous or microbial enzymes may degrade the pectin. Functional pectin quality is determined by the mo-lecular structure and heterogeneity of the extractable pectins, especially by the degree of esterifi-cation (DE) of their galacturonan backbones and their molecular weight (Neidhart et al., 2003). Due to their role in fruit softening (Varanyanond et al., 1999), quality and yield of extractable pectins are strongly related to the maturity of the processed raw material.

Concurrent with the development of options for integrated mango processing, the aim of thisongoing study is to evaluate the potential of northern Thai mango cultivars. The research ap-proach comprised studies on the influence of proper raw material selection and thermal process-ing steps involved in small-scale batch and continuous industrial manufacture of fluid mango products on resulting vitamin A values. Presently, focus is on various aspects of pectin recovery from processing waste produced from northern Thai mango cultivars. On the whole, plant physiological and technological prerequisites regarding optimum product qualities in integrated mango processing were evaluated to identify options for further diversification in the production, processing, and utilization of northern Thai mango cultivars.

2 Mango Products of High Vitamin A ValueRegarding the evaluation of the potential of northern Thai mangoes for products naturally rich in provitamin A, this report focused on the relevant aspects of mango processing into fluid prod-ucts. Because of trade specifications, fruit processing into solid products (canning, drying, freez-ing) mostly requires the utilization of particular grades. In contrast, the production and utilization of fruit pulp is far less dependent on the pre-sizing of the raw material and offers further possi-bilities for product diversification. Therefore, it considerably extends the technological options for the reduction of postharvest losses and increase in regional added value.

ß-caroteneß-caroteneRawRaw materialmaterialRawRaw materialmaterial ProcessingProcessing technologytechnologyProcessingProcessing technologytechnology

High High vitaminvitamin A A valuevalue

ß-Carotene biosynthesis Natural equilibrium of

ß-carotene stereoisomers

Vitamin A value of cultivars? Availability and physiologi-

cal consumption ripeness?

Oxidation and isomerization

Bioavailability Exposure to light, oxygen,

and heat? Thermal sensitivity? Protective matrix effects?

all-trans-

13-cis-isomer

9-cis-isomer

all-trans-

13-cis-isomer

9-cis-isomer

Figure 1: Factors influencing the vitamin A value of mango fruits and products therefrom.

4

The vitamin A value of mango products is based on the β-carotene content of the raw material and its retention by adequate processing (Fig. 1). Both pillars are briefly discussed in the subse-quent paragraphs. Since mangoes are generally consumed at very different maturity stages, de-pending on the cultivar, the extent of β-carotene biosynthesis until typical consumption ripeness is crucial. During processing, β-carotene degradation through oxidation and isomerization may reduce the vitamin A value, depending on the exposure to heat, oxygen, and light during various operations. Since for 13-cis- and 9-cis-ß-carotene (Fig. 1), respective efficiency of β-carotene conversion into vitamin A is only 53 % and 38 % of that reported for the all-trans-stereoisomer(Zechmeister, 1962), isomerization of the prevailing all-trans-β-carotene affects the vitamin A value. On the other hand, improved liberation of the β-carotene from homogenized tissue was shown to enhance provitamin A uptake from various processed foods, and hence the bioavail-ability of β-carotene (van het Hof et al., 2000).

2.1 The Impact of Optimal Fruit Quality and Raw Material SelectionAs Thai mango cultivars have been scarcely considered in international literature, Vásquez-Caicedo et al. (2002) comprehensively characterized the quality profiles at full fruit ripeness of nine varieties cultivated in the northern Thai region. Additionally, the postharvest ripening be-havior of these cultivars was studied in detail (Vásquez-Caicedo et al., 2004). Mangoes belong to the climacteric fruits. After harvest at their mature-green stage, they have to undergo postharvest ripening to reach prime quality for eating or processing (Lesham et al., 1986). The climacteric phenomenon is characterized by a rise in respiratory gas exchange. Proper physiological maturity at harvest and control of postharvest ripening are decisive factors in the production of high-quality mangoes at consumption. During this ripening period, the fruit softens (Fig. 2A) and the color of the edible part (mesocarp) changes from green to yellow or even orange-yellow (Fig. 2B), while sugar accumulation and acid degradation lead to an increasing sugar-acid ratio (TSS/TA; Fig. 2). Specific sensory profiles of the cultivars result from the differential postharvest development of their major properties. Cultivars suitable for purée production, such as ‘Kaew’, ‘Nam Dokmai #4’ (Fig. 2) or ‘Maha Chanok’, softened rather fast, already at low sugar and still high acid contents, and developed a bright yellow-orange pigmentation at the same time (Vásquez-Caicedo et al., 2004; Fig. 3B). Based on the quality profiles (Vásquez-Caicedo et al., 2004) and a concomitantly introduced ripening index (Mahayothee, 2005; Vásquez-Caicedo et al., 2005), we specified ripeness stages for various applications of the cultivars, considering re-quired processing properties.

0

5

10

15

20

25

30

35

40

45

0 30 60 90 120TSS / TA [ ]

Kiew Sawoei

Nam Dokmai #4R2 = 0.854

R2 = 0.921

Mesocarp firmness F WB = f (TSS/TA ):power law behaviourF WB = A (TSS/TA )B, B<0

FW

B [N

/cm

2 ]

A.

45

60

75

90

105

120

135

0 30 60 90 120TSS / TA [ ]

Kiew Sawoei

Nam Dokmai #4

Mesocarp hue H° = f (TSS/TA ):power law behaviourH° = A (TSS/TA )B, B<0

R2 = 0.9498

R2 = 0.8647

H°

[°]

B.

Figure 2: Postharvest ripening behavior of selected varieties among nine Thai mango cultivars studied (Vásquez-Caicedo et al., 2004): A. mesocarp firmness as maximum shear load (FWB); B. mesocarp color. Proceeding ripeness is indicated by increasing sugar-acid ra-tios (TSS/TA). Hue angles (H°) decreasing from 135° to 45° imply a color change from greenish-yellow to orange, with a hue angle of 90° indicating pure yellow in the CIELAB color space (cf. Fig. 3C).

5

β-Carotene is among the predominant carotenoids in this fruit (Pott et al., 2003b; Chen et al., 2004). Hence, this pigment is amongst the most important for the development of the natural mesocarp color as the fresh mango fruit ripens (Fig. 3A). The color change of the ripening mango can be described either by the decreasing hue angle (H°, Figs. 2B, 3C) or the increasing green-red color coordinate a* (Fig. 3A,C) that indicates a red color shift. As shown by Fig. 3A, the color of the mango mesocarp is not only interesting for sensory reasons. It is closely related to the β-carotene biosynthesis in the ripening mango fruits. By comparison of the cultivars at the fully ripe stage (Fig. 3B), those rich in β-carotene (MC, Kaew, ND#4) could be clearly distin-guished from others. They always produced an intensely yellow to orange flesh during ripening.

0100020003000400050006000700080009000

-20 -15 -10 -5 0 5 10 15 20Hue coordinate a *

all-t

rans

- -c

arot

ene

[µg/

100

g D

W]

strongpoor

Postharvest -carotene biosynthesis:

Proceeding ripening process

A.

0

20

40

60

80

100

120

140

OK KS

MDG Ra

dO

T

CAND

#4Ka

ew MC

-10

10

30

50

70

90b* a*beta-carotene TSS/TA

-c

arot

ene

[mg/

kg D

w] o

r TS

S/T

A

a*

or b

*

B.

cv. Maha Chanokcv. Kiew Sawoei

a +

a -

b +

b -

H °

C *

Mango mesocarp:red

blue

green

yellow

C.

Figure 3: (A.) Development of mesocarp color (CIELAB hue coordinate a*) by cultivar-specific postharvest β-carotene accumulation (according to Vásquez-Caicedo et al., 2005), exem-plarily shown for cultivars with strong and poor postharvest β-carotene biosynthesis (ND#4 and KS, respectively); (B.) β-carotene content and color of all nine cultivars at the fully ripe stage (according to Vásquez-Caicedo et al., 2004); and (C.) illustration of the Cartesian (a*, b*) and polar (C*, H°) hue coordinates of the CIELAB color space for the color development of ripening mangoes. Cultivars: OK, ‘Okrong Kiew’; KS, ‘Kiew Sa-woei’; MDG, ‘Mon Duen Gao’; Rad, ‘Rad’, OT, ‘Okrong Thong’; CA, ‘Chok Anan’;ND#4, ‘Nam Dokmai #4’; Kaew, ‘Kaew’; and MC, ‘Maha Chanok’.

Table 1: Extent of β-carotene biosynthesis during postharvest ripening, natural distribution of β-carotene stereoisomers, and resulting provitamin A values of the mesocarp in selected Thai mango cultivars (according to Vásquez-Caicedo et al., 2005).

Cultivar TSS/TA a RPIWBb β-carotene Vitamin A c

Total [µg/100g DWd] All-trans [%] 13-cis [%] 9-cis [%] [RE/100g DW]Cultivar examples e with strong postharvest β-carotene biosynthesis:

9.0 5.00 1,190 ± 75 86.7 13.3 n.d. 186 ± 1326.9f 2.62 5,912 ± 231 85.8 8.8 5.4 912 ± 34Kaew50.2f 1.50 8,249 ± 14 82.5 9.6 7.8 1,246 ± 28.9 5.28 1,658 ± 12 73.4 12.2 14.4 236 ± 1

26.6f 3.14 5,513 ± 30 68.8 15.4 15.8 762 ± 2Nam Dok-mai #4

47.6f 1.61 11,249 ± 939 70.5 14.8 14.7 1,573 ± 140Cultivar examples e with poor postharvest ß-carotene biosynthesis:

n.a. n.a. n.a. n.a. n.a. n.a. n.a.24.1f 4.69 673 ± 30 100.0 n.d. n.d. 112 ± 552.8 3.13 1,544 ± 68 74.2 12.7 13.1 221 ± 10

Kiew Sawoei

111.3 0.62 1,831 ± 33 73.5 13.9 12.6 261 ± 514.5 3.49 289 ± 10 100.0 n.d. n.d. 48 ± 223.4 2.88 329 ± 12 100.0 n.d. n.d. 55 ± 265.1f 1.47 1,977 ± 39 66.5 17.4 16.1 270 ± 6

Okrong Thong

114.1 0.63 2,538 ± 62 66.0 18.1 15.8 345 ± 8a TSS/TA, sugar-acid ratio; b RPIWB, ripening index; c vitamin A values in retinol equivalents (RE) calculated ac-cording to Zechmeister (1962); d DW, dry weight; e among 9 Thai cultivars studied; f TSS/TA at typical consump-

6

tion ripeness of this cultivar; n.a., not analyzed; n.d., not detectable. Analytical methods as described by Vásquez-Caicedo et al. (2005). Increasing TSS/TA ratios and decreasing RPIWB indicate proceeding postharvest ripening.

The cultivars revealing strong β-carotene biosynthesis reached vitamin A values around 1,000retinol equivalents (RE)/100 g of mesocarp dry weight at their typical consumption ripeness (Ta-ble 1). The provitamin A levels were thus five times higher than those of the other cultivars. Constant portions of naturally occurring cis-ß-carotene isomers were observed in the range of 14 to 40 %, depending on the cultivar (Vásquez-Caicedo et al., 2005). Hence, high provitamin A levels were the result either of elevated total β-carotene contents (e.g., cv. ‘Nam Dokmai’ in Ta-ble 1) or of low percentages of cis-ß-carotene isomers (e.g., cv. ‘Kaew’ in Table 1).

Regarding product differentiation by the manufacture of mango products rich in provitamin A, the proper selection of the raw fruits to be processed relies on three factors that comprise fruit ripeness, attainable total β-carotene content of a cultivar, and the cultivar-specific ratio of the β-carotene stereoisomers. In practice, mesocarp color may be used as a nutritive indicator due toestablished cultivar-specific relationships between color, all-trans-β-carotene levels, and the rip-ening index (Vásquez-Caicedo et al., 2005). According to our results, suitable mango varieties are found among the cultivars grown in northern Thailand.

2.2 Retention of Provitamin A in Mango ProcessingBecause of short harvesting periods, the vitamin A value of the fresh mango fruit may be con-served by adequate processing. Due to improved bioavailability of β-carotene and further possi-bilities of product diversification, fluid mango products may be of particular interest. However, as modern industrial year-round mango processing into fluid products, such as purées, juices, and nectars, involves a complex process outlined above, there is a widespread notion that this type of processing is knowledge and input intensive and therefore limited to a large scale not relevant for fruit production in the small farm sector. On the other hand, it may be argued that participation of small fruit producers, who may provide fruits of particular processing quality and/or at harvest times extending the use of processing capacities of juice manufacturers, largely depends on the institutional and organizational conditions of processing and supply chains. From the technological point of view, purée and juice production may be also performed by small- and medium-sized private companies or cooperatives operating their facilities at full capacities by the use of either various fruit species differing in harvest periods or raw material of various prove-niences where different climates cause a shift in harvest periods. Particularly the latter may pro-vide comparative advantages for farmers in the highlands as additional suppliers of raw material for processing, since other quality profiles without any limits in fruit size are required for juiceprocessing than in fresh fruit marketing. Further health attributes qualifying the raw fruit for spe-cific applications such as purée production for baby-food rich in natural vitamin A may open the access to special markets. Furthermore, the transport to distant factories may be additionally fa-cilitated by suitable postharvest handling extending shelf life. Finally, production of storable purées may be even performed at the household level in a simplified way using bottle pasteuriza-tion to improve the vitamin A supply of mango homegarden owners beyond the harvest times through the manifold integration of mango purée into the diet. On the whole, many options for the participation of upland farmers in industrial utilization of mangoes through purée production may arise with an increasing importance for product differentiation and novel demands of chang-ing markets.

Because of the manifold technological importance of thermal operations in the production of fluid mango products irrespective of the processing scale, we evaluated the provitamin A reten-tion of mango products under exposure to heat at various levels. On the one hand, the complex production of a typical mango nectar produced at pilot-plant scale was studied by means of step-wise process control to monitor β-carotene degradation and isomerization throughout processing according to an industrial process comprising thermal-mechanical peeling with steaming of the

7

fruits, enzymatic pulp maceration at 30 °C for viscosity standardization, and three continuous heating operations at pasteurization termperatures (93 °C). Continuous heating according to the high-temperature-short-time (HTST) principle is known to generally conserve the nutritive and sensory quality of foods better than the exposure to lower temperatures for longer times needed for the same preservation effect. Therefore, this principle is generally applied in a diversified food industry based on the production of storable intermediate and various end products.

0.7

1

1.3

1.6

1.9

0 4 8 12 16ET [min]

13-cis

9-cis

all-transC

/Co

Figure 4: -Carotene stability in mango purée (cv. ‘Tommy Atkins’) subjected to various pasteurization routines. Effects of rising exposure times (ET) at temperatures of 85 °C () and 93 °C () on the relative contents (C/Co) of β-carotene isomers.C = isomer content of the pasteurized sample [µg/100 g of purée fresh weight (FW)]; Co = isomer content of the unheated control [µg/100 g of purée FW].

Laboratory-scale heating of mango purée at different pasteurization conditions was investigated to mimick a wide range of thermal basic operations from continuous HTST-heating to small-scale bottle pasteurization. Fig. 4 exemplarily shows the effects of a single exposure of mango purée to heat for 1 min up to 16 min on β-carotene stability for the lowest and the highest pas-teurization temperature studied (85 °C and 93 °C, respectively). Poor degradation of the pre-dominant all-trans-β-carotene allowed the retention of total β-carotene at = 93 % in the mango purée, even under the extremest conditions (holding time of 16 min). Concomitant trans-cisisomerization led to the retention of at least 84.6 % of its vitamin A value, depending on the thermal stress applied. For comparison, cumulative retention of total β-carotene and provitamin A of the fruit dry matter were 93 % and 83 %, respectively, in the storable mango nectar pro-duced at pilot-plant scale as mentioned above with three continuous HTST-treatments of the fruit component.

The obtained results documented the high β-carotene stability in the production of fluid mango products at various scales. Mango purées and nectars were shown to be valuable sources of provitamin A. Since fruits must be processed at full ripeness after sufficient softening required for pulping, they are concurrently used after maximum β-carotene biosynthesis. General prereq-uisites for high provitamin A levels of products include the selection of cultivars with strong β-carotene biosynthesis and intensive yellow-orange mesocarp coloration as well as minimization of exposure to light and oxygen throughout processing and storage. The latter was shown by the complex process of mango nectar production studied, where the total thermal stress accounted for approximately one half of the overall all-trans-β-carotene loss.

3 Integrated Mango Processing with Recovery of PectinTo reveal the attainable mango pectin quality, promising Thai mango cultivars rich in both β-carotene and pectin (Vásquez-Caicedo et al., 2002) were screened for the effect of fruit ripeness on pectin yield and quality prior to characterization of fine structure and rheological properties of mango pectins. Finally, the influence of peel and pomace treatments at the local mango process-ing industry has been studied to assess the potential of industrial mango waste for pectin produc-

8

tion. With respect to the aim of this contribution, this report focused on the extractable pectin yields.

3.1 Technological Aspects of Pectin Recovery from MangoesMimicking industrial practice, pectin extraction with hot acid was performed at laboratory scale. Extracted pectinaceous material was precipitated in alcohol (isopropanol), separated from the liquid by pressing, dried at 60°C for 1-2 h, and ground to the desired particle size (< 0.25 mm). Yield of dried pectin was calculated as alcohol-insoluble substance (AIS) per 100 g of dry po-mace used for extraction. Additionally, pectin yield was corrected for the starch content of the AIS after enzymatic determination of starch (AISS-corr).

To standardize pectin extraction from mango, two standard protocols for pectin quantification in apple pomace and citrus peels, respectively, were adopted from an industrial pectin producer. Extraction at two different pH values allowed the comparison of mango pectins with those from apple and citrus. Furthermore, the antagonistic effect of decreasing pH values on yield and func-tional properties of pectins was considered. Increasing acidity facilitated the isolation of the pec-tin molecules from the primary cell walls, but also enhanced de-esterification and molecular cleavage, reducing molecular weight and DE of the pectins and the setting temperature of the gels. When compared to chemical composition and yield, gelation of the mango pectins was much more influenced by the extraction method (Fig. 5). For comparison, unstandardized com-mercial pectins from apple and citrus were selected as references of similar DE levels.

Pectin yield [g/100 g]Pectin samplecode AIS AISS-corr

Peel A – EM C 15.7 ± 1.0 12.9

Peel A – EM A 13.5 ± 0.9 11.4

Peel B – EM C 13.8 ± 1.0 13.1

A. Peel B – EM A 13.1 ± 0.8 12.5

0102030405060708090

100

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7C AUA [g/100g of gel]

T=

45° [

°C]

pericarp A-1.5 pericarp B-1.5pericarp A-2.0 pericarp B-2.0Peel A - EM Peel B - EM Peel A - EM Peel B - EM

B.

Figure 5: Influence of extraction conditions. Yield (A.) and setting properties (B.) of pectins produced from mango peels (cv. ‘Kaew’ at ripeness stages A and B), applying two extraction methods differing in pH values (EM A and EM C). AIS, alcohol-insoluble substance; AISS-corr, net pectin yield as AIS without starch; cAUA, anhy-drogalacturonic acid content (= pectin content) in the gel; Tδ=45�, setting tempera-ture of the gel (gelation temperature).

3.2 Influence of Fruit Ripeness on the Quality of Pectins from Various Mango CultivarsBecause of its industrial use and its particularly high pectin content (Vásquez-Caicedo et al., 2002), cv. ‘Kaew’ was selected for screening of suitable ripeness stages. Cvs. ‘Nam Dokmai #4’ and ‘Kiew Sawoei’ were considered because of their potential availability from off-season fruit production, with the latter cultivar also representing the use green-eaten varieties (Vásquez-Caicedo et al., 2005). Fruits were obtained from the research plot of the Uplands Program and from a local farm in San Sai district, Chiang Mai, Thailand, during main season 2003. Approx. 40-80 kg of each cultivar were harvested at the mature-green ripe stage and ripened for 5 days at 25 ± 2 °C and a relative humidity of 70-80 %, using calcium carbide (CaC2) as ripening accelera-tor. Samples of approx. 2 kg were collected randomly for daily control of postharvest ripening until full ripeness (Fig. 6). Pectin was extracted from peel and mesocarp at different ripeness stages (Fig. 6) that were specified by the ripeness index RPIWB (Vásquez-Caicedo et al., 2005).

9

By estimating necessary ripening times based on the linear development of RPIWB (Fig. 6), ex-periments were repeated in 2004 at a larger scale using 400 kg of mangoes cv. ‘Kaew’ for a more comprehensive characterization of pectins obtained at ripeness stages used in processing (Vásquez-Caicedo et al., 2004).

R 2 = 0.9898R 2 = 0.9589

R 2 = 0.9757

-1 012345678

0 2 4 6 8 10RT [days]

RP

I WB [

]

KS 2003NDM 2003KA 2003KA 2004

Figure 6: Postharvest ripening of mango fruit lots of cvs. ‘Kiew Sawoei’ (KS), ‘Nam Dok-mai’ (NDM), and ‘Kaew’ (KA) at 25-28 °C in years 2003 (with CaC2 as ripening accelerator) and 2004 (without CaC2). Arrows indicate ripeness stages used for pectin extraction. RPIWB, postharvest ripeness index; RT, postharvest ripening time after harvest on day 0; —— linear ripening; ----- initial lag phase.

Percentages of peel, seed, and flesh as well as fruit size of all fruit lots studied were in the range of our previous findings (Vásquez-Caicedo et al., 2002). AIS yields from mango peels at the fully ripe stage reported so far (Sudhakar & Maini, 2000; Rehman et al., 2004) were confirmed. How-ever, as a key result of this screening, AIS markedly decreased during postharvest ripening due to starch degradation, whereas contents of extractable pectins (AISS-corr) hardly declined in the rip-ening peels. Pectin degradation involved in postharvest mesocarp softening resulted in declining pectin yields from mesocarp. At the green-ripe stage (RPIWB 6.43-4.34), AIS from mesocarp and peels still consisted of 66.2-78.0 % and 47.1-52.6 % of starch, respectively, depending on culti-var and extraction method. Even in the AIS of fully ripe mesocarp (RPIWB from 0.64 to -0.23), starch levels were in the range of 1.9 and 15.7 %, with lower contents in the AIS of peels (1.22-5.12 %). Net pectin yields (AISS-corr) of mango peels ranged between 11.4 and 20.9 % depending on cultivar, ripeness, and extraction method, whereas 4.0-13.2 % were found in mango meso-carp. Hence, mangoes are promising pectin sources at various ripeness stages, provided that starch is removed by enzymatic degradation during pectin production. Otherwise, riper fruits would be preferred due to the reduced starch content. These results are of particular importance, when adequate processing ripeness for different applications is considered (Vásquez-Caicedo et al., 2004; Pott et al., 2005).

By analogy with Neidhart et al. (2003), mango and reference pectins were chemically and rheologically characterized. To facilitate transfer of knowledge into practice, analytical and em-pirical methods commonly used by pectin producers were also included. On the whole, the Thai mangoes were shown to be promising sources for high-esterified pectins. Whereas other studies mostly focused on suitable pectin extraction methods for mango, we could identify essential raw material factors influencing pectin quality and characterize their effects on composition and functionality of mango pectins.

3.3 The Potential of Industrial Mango Waste as Source of PectinIn main harvest season 2005, the systematic evaluation of industrial mango processing waste in northern Thailand was started (Table 2). Mango peels were collected in two local canning facto-ries in Chiang Mai and Lampang. For immediate peel drying, a portable laboratory-dryer was temporarily used in the canning factories. The second problem mentioned in Table 2 (B.) re-ferred to the tolerable period between waste production (peeling) and drying of the waste. Dif-

10

ferent operational situations in mango processing were simulated in this experiment regarding exposure of waste to conditions favoring microbial or physiological pectin degradation. Finally, prompt large-scale drying of peels from industrial mango canning was performed in cooperation with a nearby local drying company (C. in Table 2).

Table 2: Evaluation of industrial mango processing waste in northern Thailand

Scientific approach(A.) Variability of mango waste material during the processing campaign and among the factories(B.) Sensitivity of the waste to pectin degradation (optimization of processing logistics)(C.) Suitability of industrially dried mango peels from local industrial canning processes

Table 3 summarizes major results of this study as to the pectin yield. The net pectin yield was in the same range previously found for the peels of fresh fruits (Fig. 5). Moreover, it hardly varied among the mango processing companies and throughout the mango campaign. Chemical and functional analyses of all the pectins obtained according to Table 2 are currently under way.

Table 3: Yields of pectins extracted from industrial mango waste dried at laboratory and at in-dustrial scale

Mango proces-

sor

Ob-served period a

Number of pectin lots studied b

Extrac-tion

method

AIS[g/100 g]c

AISS-corr[g/100 g]c

[days] Min. Median Max. Min. Median Max.Peels dried at laboratory scale (A. in Table 2):

I 32 5 EM C 14.8 16.2 16.5 12.9 13.4 14.8I 32 5 EM A 13.8 15.3 15.8 12.3 12.4 14.5II 23 5 EM C 15.7 16.1 18.5 12.3 13.2 14.0II 23 5 EM A 14.5 15.3 16.9 11.4 11.6 13.6

Peels dried at industrial scale (C. in Table 2):I 23 5 EM C 15.4 16.2 16.7 12.3 13.2 14.2I 23 2 EM A 14.8 --- 15.9 11.2 --- 13.4

a Period studied within the mango processing campaign of the canning company; b Pectin lot: extraction of pectin from one pooled sample of the peels occurring as waste per mango processing day; c total yield (AIS) and net pectin yield (AISS-corr) per 100 g of dry peel

4 Conclusions - The Technological Potential of Northern Thai Mango CultivarsIn conclusion, necessary plant physiological and technological prerequisites were identified. Mango purées and nectars were shown to be valuable sources of provitamin A, because fruits are processed at their full ripeness, the latter as premise ensuring sufficient fruit softening for ac-ceptable yields in pulping. Consequently, fruits are processed after maximum biosynthesis of β-carotene. However, as discussed previously (Vásquez-Caicedo et al., 2005), the selection of cul-tivars with strong β-carotene biosynthesis and an intensive yellow-orange mesocarp color for processing is a prerequisite. Furthermore, thermal stability of β-carotene in typical processes used in the production of purées and nectars of industrial relevance or at the household level was confirmed, while improved liberation of the β-carotene from the ground tissue may enhance bioavailability and provitamin A uptake. As recently shown for drying of mango slices (Pott et al., 2003a), the necessity to reduce the exposure of the material to light and oxygen to a mini-mum throughout processing is a major technological requirement for provitamin A retention.

Although industrial sources for mango pectin are still under evaluation, some essential conclu-sions can already be drawn. Thai mangoes can be considered a promising alternative source of pectin under certain prerequisites. As the peels seem to be notably suitable, canning waste of

11

northern Thai mango processors could also contribute to the supply of pectin manufacturers with dry raw material. Because of limited waste volumes per canning factory, a network of raw mate-rial suppliers with mango processors operating at different scales would be the premise. The im-portance of this study was to identify the potentials and limits for this type of integrated mango processing. Depending on future demands for mango pectin, the value-adding effect might be further increased by the combined recovery of pectins and polyphenols (Berardini et al., 2005).

From the technological point of view, the wide potential of mango cultivars grown in the north-ern Thai uplands for further product differentiation was shown. Mango products rich in provita-min A, combined with pectin recovery from peels, may open novel options. However, further research as to institutional conditions and up-scaling of the presented laboratory findings would be needed prior to realization in food industry.

AcknowledgmentThis research was funded by Deutsche Forschungsgemeinschaft (DFG), Bonn, Germany: Pro-jects SFB 564-E2 and E2.2. It is part of the Special Research Program ‘Research for Sustainable Land Use and Rural Development in Mountainous Regions of Southeast Asia’ (Uplands Pro-gram). The authors thank Herbstreith & Fox KG, Neuenbürg, Germany, providing laboratory facilities for pectin extraction at an extended laboratory scale as well as the Thai companies Northern Food Co., Ltd., Universal Food Public Co., Ltd., and Nithi Foods Co., Ltd. for their assistance in the collection of industrial mango waste.

ReferencesBally, I.S.E., 2004. Mangifera indica (mango). In: Elevitch, C.R. (Ed.), Species profiles for Pa-

cific Island agroforestry. URL: http://www.tradiotionaltree.org. Permanent Agriculture Re-sources (PAR), Holualoa, Hawai’i. [http://www.agroforestry.net/tti/Mangifera(mango).pdf].

Berardini, N., Knödler, M., Schieber, A., Carle, R., 2005. Utilization of mango peels as a source of pectin and polyphenolics. Innovative Food Science & Emerging Technologies 6, 442-452.

de Bie, C.A.J.M., 2004. The yield gap of mango in Phrao, Thailand, as investigated through com-parative performance evaluation. Scientia Horticulturae 102, 37-52.

Chen, J.P., Tai, C.Y., Chen, B.H., 2004. Improved liquid chromatographic method for determi-nation of carotenoids in Taiwanese mango (Mangifera indica L.). Journal of Chromatography A 1054, 261-268.

Dube, M., Zunker, K., Neidhart, S., Carle, R., Steinhart, H., Paschke, A., 2004. Effect of techno-logical processing on the allergenicity of mangoes (Mangifera indica L.). Journal of Agricul-tural and Food Chemistry 52, 3938-3945.

FAOSTAT, 2006. FAOSTAT Agricultural data, last updated December 2005 (Agriculture & Food Trade) and January 2006 (Agricultural Production). In: FAO, Economic and Social De-partment, Statistics Division, Statistical Databases of the Food and Agriculture Organization of the United Nations. Internet edition, Rome, Italy. [http://www.fao.org].

Larrauri, J.A., Ruperez, P., Borroto, B., Saura-Calixto, F., 1996. Mango peels as a new tropical fiber: preparation and characterization. Lebensmittel-Wissenschaft und -Technologie 29, 729-733.

Lesham, Y.Y., Halevy, A.H., Frenkel, C., 1986. Fruit Ripening. In: Lesham, Y.Y., Halevy, A.H., Frenkel, C. (Eds.), Process and control of plant senescence. Elsevier, Amsterdam, The Nether-lands, pp. 162-210.

Mahayothee, B., 2005. The influence of raw material on the quality of dried mango slices (Mangifera indica L.) with special reference to postharvest ripening. Ph.D. thesis, Hohenheim

12

University. Carle, R., (Ed.), Schriftenreihe des Lehrstuhls Lebensmittel pflanzlicher Herkunft, Vol. 2. Shaker Verlag, Aachen, Germany.

Morton, J.F., 1987. Fruits of warm climates. Julia F. Morton, Miami, FL, pp. 221–239. [http://www.hort.purdue.edu/newcrop/morton/mango_ars.html].

Neidhart, S., Hannak, C., Gierschner, K., 2003. Sol-gel transitions of high-esterified pectins and their molecular structure. In: Voragen, F., Schols, H., Visser, R. (Eds.), Advances in pectin and pectinase research. Kluwer Academic Publishers, Dordrecht, The Netherlands, pp. 431-448.

Pott, I., Marx, M., Neidhart, S., Mühlbauer, W., Carle, R., 2003a. Quantitative determination of -carotene stereoisomers in fresh, dried, and solar-dried mangoes (Mangifera indica L.). Journal of Agricultural and Food Chemistry 51(16), 4527-4531.

Pott, I., Breithaupt, D.E., Carle, R., 2003b. Detection of unusual carotenoid esters in fresh mango (Mangifera indica L. cv. ‘Kent’). Phytochemistry 64, 825-829.

Pott, I., Neidhart, S., Mühlbauer, W., Carle, R., 2005. Quality improvement of non-sulphited mango slices by drying at high temperatures. Innovative Food Science & Emerging Tech-nologies 6(4), 412-419.

Rehman, Z.U., Salariya, A.M., Habib F., Shah, W.H., 2004. Utilization of mango peels as a source of pectin. Journal of the Chemical Society of Pakistan 26, 73-76.

Schieber, A., Stintzing, F.C., Carle, R., 2001. By-products of plant food processing as a source of functional compounds - recent developments. Trends in Food Science and Technology 12, 401-413.

Sudhakar, D.V., Maini S.B., 2000. Isolation and characterization of mango peel pectins. Journal of Food Processing and Preservation 24(3), 209-227.

van het Hof, K.H., de Boer, B.C.J., Tijburg, L.B.M., Lucius, B.R.H.M., Zijp, I., West, C.E., Hautvast, J.G.A.J., Weststrate, J.A., 2000. Carotenoid bioavailability in humans from toma-toes processed in different ways determined from the carotenoid response in the triglyceride-rich lipoprotein fraction of plasma after a single consumption and in plasma after four days of consumption. Journal of Nutrition 130(5), 1189-1196.

Varanyanond, W., Naohara, J., Wongkrajang, K., Manabe, M., 1999. Changes in pectin content and average molecular weight of pectin during maturation of mango ‘Kaew’. Food Science and Technology Research 5(4), 362-364.

Vásquez-Caicedo, A.L., Neidhart, S., Pathomrungsiyounggul, P., Wiriyacharee, P., Chattrakul, A., Sruamsiri, P., Manochai, P., Bangerth, F., Carle, R., 2002. Physical, chemical and sensory properties of 9 Thai mango culitvars and evaluation of their technological and nutritional po-tential. International Symposium ‘Sustaining Food Security and Managing Natural Resources in Southeast Asia: Challenges for the 21st Century’, 08.-11.01.2002, Chiang Mai, Thailand. [http://www.uni-hohenheim.de/symposium2002/pa_full/Full-Pap-S3B-3_Vasques_best.pdf].

Vásquez-Caicedo, A.L., Neidhart, S., Carle, R., 2004. Postharvest ripening behavior of nine Thai mango cultivars and their suitability for industrial applications. Acta Horticulturae 645, 617-625.

Vásquez-Caicedo, A.L., Sruamsiri, P., Carle, R., Neidhart, S., 2005. Accumulation of all-trans-β-carotene and its 9-cis and 13-cis stereoisomers during postharvest ripening of nine Thai mango cultivars. Journal of Agricultural and Food Chemistry 53(12), 4827-4835.

WHO, 2006. [http://www.who.int/nutrition/topics/vad/en/index.html].Zechmeister, L., 1962. Cis-trans-isomeric carotenoids, vitamin A and arylpolyenes. Springer

Verlag, Wien, Austria.