Valorization of Dacryodes edulis Fruit (Safou) through Improved Drying Procedures, Packaging and Processing

The drying of safou is practiced in countries where the crop is grown but dried products are of low quality due to a high rate of rancidity. The drying procedure and the material used for packaging dried safou may be responsible for the high rate of rancid safou. The use of safou powder as an ingredient in the making of biscuits will open up new markets for dried safou especially for sour clones and this may reduce the quantity of rancid dried safou in the market. In order to investigate this, safou fruits were d ried using two different procedures; “rapid” and “slow” procedures and then packaged in polyethylene (PE), polypropylene (PP), vacuum (GR) bags and bottles. Sensory evaluation was performed at 1, 15, 30, 60, 90 days after storage. Safou powder from sour and non-sour clones was used as a partial and complete substitute of butter in making biscuits and the acceptability of these biscuits tested in comparison with regular b iscuits. The data showed that 100 % of the test panelists accepted the form of the slow-dried safou compared to just 30 % for the rapid-dried (Χ = 25, df =1, N=36). After 90 days of storage,safou dried using the rapid procedure and packaged in PE and PP were completely rancid while safou dried using the slow drying procedure and packaged in PE and PP were rated as slightly rancid (Χ = 19, df =1, N=36). The panellists rated dried safou packaged in GR and bottle and stored for 90 days as not rancid irrespective of the drying procedure used. Biscuits produced were highly acceptable and no difference was recorded in the acceptability of part ial safou biscuits made from sour or non-sour safou powder (Χ = 0.09, df =1, N=36). The proportion of consumers who liked the colour, taste and texture of all biscuits made using non-sour safou powder was higher compared to those who disliked o r had no opinion about them (Χ = 24, d f =1, N=36). Smaller (40 mm )safou slices dried for 4 h got rancid faster than those larger (1500 mm )safou slices dried for 16 h. Vacuum bags and airt ight bottles stored dried safou better than PE and PP. Safou powder from sour and non-sour clone can be used directly as a partial substitute of animal fat (butter) in the making of b iscuits without going through the costly and laborious oil extraction process.


Introduction
In Cameroon like other less-developed countries, post harvest loss is very h igh due to inadequate post harvest p ractices an d p rocessing . Th is is particu larly true fo r Dacryodes edulis fruit (safou) that is highly perishable in nature last ing 2-3 days at amb ient cond it ions [1]. Poor harvesting, handling and storage of fresh fruits is co mp licat ed since value-chain actors have limited knowledge and resources to ensure product quality and increase shelf life. Methods to increase shelf life and maintain quality of fru its during andafter harvest have been proposed [2]. Irrespective of these achievements, storage of fresh fruits could only last for two weeks at ambient conditions.The question one will normally ask is; what beco mes of fresh safou after t wo weeks have elapsed? For safou to be properly valorised and marketed both locally and abroad, new methods to increase shelf life and add value need to be considered. Production of dried safou will be an attractive alternative. Dried safou can be eaten as it is or be a potential raw material for safou oil production or safou powder for the making of biscuits. However, the dried safou recorded in some drying units are of low quality. This is main ly due to non-uniformity of safou clones used, poor handling and packaging of products. Most dried safoufruits are packaged in lo w-grade packages like polyethylene that lead to a high level of rancidity (development of off-flavours). In addit ion, some drying procedures are believed to accelerate the process of product rancidity. It will be important to determine which drying procedure produces products that are acceptable by the population and reduces the rate of off-flavour development during storage. Although dried safou is promising, much still needs to be done especially on how to add-value to these dried products via processing into oil or biscuits.
The existence of an interspecific tree-to-tree variation in safoutraits [3] has greatly influenced the market ing of safou since most farmers do not sort their fruits at harvest. It is not uncommon to find a sachet of dried safou with different clones (sour and not sour). Consumers of safou (boiled, roasted and dried) have expressed the desire to eat fruits coming fro m the same clone and being non-sour with a n ice aroma [4]. A majority of safou trees growing in Cameroon are not selected cultivars and it is common to find a high quantity of sour cultivars amongst these trees [3]. That's why the International Center in Agroforestry Research (ICRAF) and the Institute of Agricultural Research for Develop ment (IRAD) have started the selection and multip licat ion of cultivars with desirable traits for distribution to farmers [5]; [6]). The question one will ask is; what becomes of this sour safou that is not acceptable by the population? In addition, if most drying units were to adopt proper drying and packaging of safou for long-term storage, what will they do with this product given that the consumption of dried safou at moment is not commonplace in the diet of Ca me roonians? The fresh mesocarp of safou contains about 33-65 % o il [7]; about 18-27 % o f this oil is linoleic acid [8] and a good amount of minerals like calciu m, potassium, phosphorus and magnesium [9]. Safou o il has been used to make biscuits [10] but the o il extraction p rocess is known to be costly and laborious. In addition, there is a huge loss of other nutrients present in the fruit pu lp. It will be interesting to evaluate the potential of safou powder fro m sour and non-sour clones as a substitute of animal fat in the making of biscuits. Furthermo re, test the acceptability of these biscuits in co mparison with regular b iscuits.

Study Sites
Fruit preparation, dry ing, packaging and milling were done at TalessDryfoods and Equip ment, Nkolfoulou, Yaoundé. The baking process, sensory evaluations and consumer preference were done at IRAD, Nkolbisson-Yaoundé and during the International Trade Exhib ition (PROM OTE 2011) in Yaoundé.

Plant Materials
Sour and non-sour safou clones were harvested from IRA D experimental farm. Two flour types were used; wheat flour bought from the Mokolo market and rice flour produced by milling rice using a co rn mill fo llowed by sieving.

Preparation of Dried Safou
The safou fruits were washed with 0.1 % bleach by immersing fru its in solution for 20 minutes after which the fruits were rinsed with distilled water. The fruits were then cut into two halves using a sharp knife. The seeds were removed and the pericarp scraped out.

First Procedure "Slow Dry ing Procedure"
For slow dry ing, the pulp was sliced longitudinal to an area of about 1500 mm 2 (Fig 1a). The sliced safou was then spread on a drying board and dried in a Heraeus® D-6450 Hanau drying oven at 60°C fo r 16 hour with convectional air circulat ion.

Second Procedure"Rapid Dry ing Procedure"
For rapid d rying, fru it pulp was sliced lat itudinal to an area of 40 mm 2 using a fruit-slicing mach ine (Fig 1b). The sliced safou was then spread on a drying board and dried in a Heraeus® D-6450 Hanau drying oven at 60°C fo r 5 hours with convectional air circu lation. The dried were then removed, cooled to room temperature and packaged.

Packag ing of Safou
The dried safou was then removed, cooled to room temperature and before packaging.Ten grams (10 g) of safou dried using the two dry ing procedures were packaged immed iately in PE, PP, GR, and bottles. PE and PP were sealed using a sealing machine, GR was sealed using a Vacuum sealer wh ile bottles were corked and sealed using parafilm®. Samp les were evaluated after 1, 15, 30, 60 and 90 days post packaging and storage in a cold dry p lace at room conditions. All experimentations had 3 replications.

Preparation of Biscuits
Three types of biscuits were produced using two kind of safou powder (sour and non sour dried safou milled); one where safou powder was the only lipid source (0% butter: 100% safou), the other where butter was partially substituted Improved Drying Procedures, Packaging and Processing with safou powder (46% butter: 54 % safou) and that where only butter was the source of lip id (100% butter: 0% safou). The biscuits with 0% butter: 100% safou was named complete safou biscuits; that with 46% butter: 54 % safou was named partial safou biscuit and that with 100% butter: 0% safou was named regular biscuit. To produce the different types of biscuits, 100 g of sugar, varying proportions of butter: safou (0:130 g, 60:70 g, 130:0 g ), 2 eggs and 1g of salt were mixed. Two hundred (200) g of wheat flour, 10 g of powdered milk, 5 g baking powder and flavouring were then added and mixed well to obtain the homogenous dough. The dough was then wrapped and refrigerated for an hour. The dough was rolled out to ¼ inches thick on a well-floured surface. The ro lled dough was then cut into the same shape and place on a greased sheet and bake at 190 o C for 10 minutes. Partial safou and regular biscuits were also produced using rice flour produced locally. Biscuits were allo wed to cool on a tray and vacuum packaged and evaluated the next day for sensory evaluation. All experiments had three replications.

Sensory Evaluation
With the aid of a questionnaire, a trained sensory evaluation panel consisting of 12 persons was asked to score the appearance (form and co lour) of dried safou obtained using the rapid and slow drying procedures on a likeness hedonic scale (like, d islike and not sure). The same panel was asked to rank the taste of dried safou packaged in different containers on a rancidity hedonic scale (not rancid, slightly rancid and rancid). These tests were done 1, 15, 30, 60 and 90 days post packaging and storage. Another trained test panel consisting of 12 persons was also asked to score the colour, taste and texture of the biscuits produced on a likeness hedonic scale (like, d islike and not sure). In addition they were to say if the texture was granulated, smooth, crispy or hard. To avoid the influence of packaging material on the choice of the panelists, products were taken out of the packaging containers and placed in dishes before being presented to the test panel. The panelists were not informed about the constituent of the product being tested and all replicated samp les were tested. During POM OTE 2011 in Yaoundé, a stand was put in place to sell dried safou produced from the slow and rapid drying procedures. In addition, regular and safou biscuits were also sold. Consumers were presented with dried safou and biscuits to taste and their preference were recorded. The quantity of each type of dried safou sold during the entire period of the PROMOTE 2011 was also recorded.

Statistical Analysis
The data was entered into the Statistical Package for Social Sciences (SPSS) version 10. The frequencies of the responses were generated and the percentages computed. The Chi Square (Χ 2 ) test was used to compare two independent preferences expressed as proportions of respondents who bought a particular product, liked or disliked a part icular product or scored it as rancid, slightly rancid or not rancid. The Chi Square (Χ 2 ) test was also used to compare the proportion of respondents who scored the 4 different types of biscuits based on texture and the acceptability of those biscuits. All statistical tests were done at the 5 % significance level.

Safou Drying Procedures and Consumer Preference
of Drie d Safou Safou dried using the slow dry ing procedures was preferred by a trained test panel and customers visiting IRA D/Taless Dry Food Partnership stand during the Yaoundé International Trade Exhib ition (PROMOTE 2011) compared to that produced using the rapid drying procedure (Fig 1). The main reasons put forward for this preference was the size of the slices and the fact that they could recognize that the product was safou. The slow drying procedure produced longitudinal slices of 1500 mm 2 and consumers can easily recognize it as safou while the rap id drying procedure produces slices of 40 mm 2 , wh ich wasnot easily recognized as safou. The data showed that 100 % of the panel accepted the form of the slow dried safou compared to just 30 % for the rapid (Χ 2 = 25, df =1, N=36). The co lour of the product was not an important factor influencing consumer preference of dried safou produced using both procedures. One hundred (100 %) of d ried safou packs produced using the slow drying procedure were sold at the PROMOTE 2011 stand compared to 40 % of that produced using the rapid drying procedure (Fig 2) (Χ 2 = 25, df =1, N=100). The results obtained at PROMOTE 2011 were similar to observations made at the level of the test panel at IRAD Yaoundé-Cameroon.

Safou Drying Procedures, Packaging and the Devel opment of Off-Fl avours
The taste of safou dried using both procedures were considered not rancid (no off-flavour) by the test panel a day after drying and 15 days after drying and packaging in PE, PP, GR and bottles (Fig 3a). Thirty (30) days after dry ing, safou dried using the rapid procedure and packaged in PE were slightly rancid and the rest were not rancid. Sixty (60) days after drying, most of the safou dried using the rapid procedure and packaged in PE were rancid while those in PP were slightly rancid. All safou dried using the slow procedure was not rancid irrespective of the packaging material used. In addition, all dried safou packaged in GR and bottles were not rancid irrespective of the drying procedure used (data not shown). After 90 days of storage, the safou dried using the rapid procedure and packaged in PE and PP were co mpletely rancid while safou dried using the slow drying p rocedure and packaged in PE and PP were rated as slightly rancid by more than 80 % of the trained panel (Χ 2 = 19, df =1, N=36). The panel rated dried safou packaged in GR and bottle as not rancid irrespective of the drying procedure used (Fig 3b) Fro m the above results, show that safou produced using the rapid procedure get rancid faster than that produced using the slow dry ing procedure. Safou produced using the slow drying procedure is sliced into thin slices and thus the total surface area of the tissue exposed to oxygen is higher than for safou dried using the slow drying procedure. Since safou pulp contains about 33-65 % lip id [7] mostly polyunsaturated fatty acids, increase exposure of the dried pulp to atmospheric o xygen and light may result in the o xidation of these lipids to produce volatile aldehydes, ketones and alcohols that are responsible for the rancid flavour observed [11]. For the food industry, the detection of oxidative off-flavors by taste or smell is the main procedure of deciding when a lip id-containing food is no longer fit for consumption [12]. In this study, we observe that packaging and storage of dried safou in air-deprived or air-tight containers like vacuum bags and bottle respectively minimize the rate of off-flavour development in safou produced using both rapid and slow drying procedures during storage. The greatest inhibitory effect of vacuum and bottle packaging compared to polyethylene (PE) and polypropylene (PP) was probably because they are impermeab le to gases, mo isture and aroma. PP is moderate permeab le to mo isture, gases and aroma. Although PE is a good moisture barrier, it has relat ively high permeability to gases and aroma.

Safou Powder as a Parti al Substitute to Ani mal Fat in The Making of Biscuits
Powdered dried safouwas effectively used as a partial substitute of animal fat in the making of biscuits (Fig. 4). Consumers' preference for partial safou biscuits made using sour safou was same as that made using non-sour safou clones (data not shown). Ho wever, there was a d ifference in response for co mplete safou biscuits especially at the level of the taste. The proportion of consumers who liked the colour, taste and texture of all biscuits made using non sour safou powder was h igher co mpared to those who disliked or had no opinion about them (Χ 2 = 24, df = 1, N = 36). Un like for non sour safou powder, the proportion of consumers who liked and disliked the taste of complete biscuits made using sour sofou powder was not different (Fig.5) (Χ 2 = 0.11, df = 1, N=36).
A B C Figure 4. Biscuits produced using different proportions of butter and safou powder as lipid source. (A) Regular biscuit (100% butter: 0% safou), (B) Partial safou biscuit (46% butter: 54 % safou), (C) Complete safoubiscuit (0% butter: 100% safou) Improved Drying Procedures, Packaging and Processing Figure 5. Preference of biscuits produced using butter (regular, 100 % butter: 0 % safou), partial sour safou powder (46% butter: 54 % safou) and complete sour safou powder (100% butter: 0% safou). * Indicates the difference is significant at the 0.05 level in that category Figure 6. T exture and acceptability of the texture of biscuits determine by a sensory evaluation panel at IRAD Yaoundé-Cameroon for partial safou biscuit (46% butter: 54 % safou), and Regular biscuit (100% butter: 0% safou) produced using rice or wheat flour. *Indicates the difference is significant at the 0.05 level in that category The main reason advanced by consumers was an after-taste after eating the b iscuits. A slightly high but insignificant proportion of respondent disliked the texture of complete safou biscuits irrespective of the type of safou powder used. The main reasons advanced was that these biscuits were not crispy (Χ 2 = 16, d f =2, N=36)and hard (Χ 2 = 22, df =2, N=36). Biscuits produced using safou flour were rated as smooth while those produced using rice flour were rated as granulated (Fig.6). Respondent generally accepted biscuits produced using both wheat and rice biscuits with or without safou powder but the proportion of those who liked regular rice biscuits was lower (Fig. 6) (Χ 2 = 10, df =3, N=36). Safou powder improved the acceptability of rice biscuits. The reason for this improvement may be because safou powder decreased the granulated nature of the rice biscuits produced.

Val orisationof Sour Safouas a Substitute of Animal Fat In Bakery
Consumers of safou (boiled, roasted and dried) have expressed the desire to eat fruits coming fro m the same clone and being non-sour with a nice aro ma [4]. Most D. edulis trees in Cameroon are gro wing in the wild (not traceable) and it is common to find a good quantity of acidic safou among the fruits fro m these trees [3]. Since most of this sour fruit is not acceptable, they are usually left to get rotten, thus aggravating post harvest loss in these species. The result obtained in this study showed for the first time that value could be added to sour safou by using its powder in making safou biscuits. The biscuits were highly acceptable and no difference was recorded in the acceptability of part ial safoubiscuits made either withsour or non-soursafou powder (Χ 2 = 0.09, df =1, N=36).
Although the results in this study are similar to those obtain other work [13]; [10], in which safou oil was used as an ingredient for making cookies, the present study demonstrated that the laborious oil extraction process can be avoided and the entire dry powdered fruit pulp used as an ingredient in bakery. It has been shown that the fresh mesocarp of safou contains about 33-65 % oil [7]. Thus incorporation of sour safou powder for the making of biscuits will increase the use of safou and spur production. This will undoubtedly increase the inco me of rural communit ies and especially wo men who are the main value-chain actors involved in safou trade.
In addition, the nutritional benefits the population will get fro m consuming this product are expected to be high as demonstrated in several studies. The lipid content of safou pulp is between 33-65% [7]; [14]. This value is considerably higher co mpared to other fru its such as apple (0.4%), guava (0.4%), banana (0.39%) and pawpaw with traces of oil.
The high lip id content in the fruit pulp is of significant value. Apart fro m the fact it contains linoleic acid, an important polyunsaturated fatty acid in human food, which can prevent cardiovascular disorder, the oil is also rich in oleic acid that has oxidative stability importance [15]. The protein content of the fruit pulp of safou has been shown to be low (3.24%), but will make significant contribution in ameliorating protein malnutrition especially in child ren. Carbohydrate composition (11.38%) is moderate but can supply some calories. The fru it pulp is a rich source of mineral elements such as sodium potassium, phosphorus, magnesiu m, iron, copper and zinc [9]. The pulp of safoucontains linoleic and arachidonic acids, which are the essential fatty acids with the percentage of the linoleic acid, found in the pulp to be 23.17%. Dietary fat rich in linoleic acid prevents disorders such as coronary heart diseases [16].
This exp lains why it is advantageous to use safou powder for the making of biscuits rather than just the oil. However, it will be important to evaluate the effect of the baking process on key nutrients in safou pulp and also study the effect of safou and safou biscuit consumption on the level of serum cholesterol.

Conclusions
This study has demonstrated for the first time the drying procedure, packaging materials and duration of storage of dried safou that best reduces rancidity of the dried product. Safou sliced into s maller and thin slices got rancid faster than those sliced into larger and thick slices. Vacuum bags and airtight bottles stored dried safou better than PE and PP.
This work also demonstrated the use of safou powder fro m sour and non-sour clone as a partial substitute of animal fat (butter) in the making of biscuits. Partial safou biscuits made fro m sour and non-sour safou powder were comparable to regular biscuits with both being highly acceptable. However, complete safou biscuits made using sour safou powder were only moderately acceptable because consumer said eating this biscuit left an after-taste. In addition, the biscuits were rated as hard and not crispy. Value addition to sour safou by using its powder in making safou biscuits is an important finding, as this will likely reduce post harvest loss of these fruits that are not acceptable as boiled, roasted or dried [4]. This action will likely open up new markets for dried safou, reduce the duration of dried safou on market shelves thus reduce the quantity of rancid p roducts in the market. In addition, this action will likely protect biodiversity, as farmers will no longer cut down safou trees that produce sour fruits. The biscuit obtained may be very nutritious and may play a vital role in human health especially in reducing cardio-vascular diseases.