Morphological Assessment of Annatto (Bixa orellana L.) Fruit and Seed for the Development of Mechanical Seed Separator

Annatto or lipstick tree belongs to the family Bixaceae (Kumaran et al., 2014) It is a perennial and tropical tree, that originated in Central and South America (Pandey et al., 2019). The trees are extensively distributed in parts of Africa and Asia. In India, it is widely cultivated in the forest areas of Madhya Pradesh, Orissa Andhra Pradesh, Kerala, Tamil Nadu, West Bengal, Gujarat, Maharashtra and Chhattisgarh (Math et al., 2016; Bindyalaxmi et al., 2022). The trees can reach heights up to 10 m. The flowers are pinkish-white in colour. The fruits are of different shapes such as ovoid and flattened, ellipsoid, or conical. The fruits are capsules that may contain seeds from 20 to 60 on average. (Akshatha et al., 2011; Vilar et al., 2014).  The trees are cultivated for their seeds.

In general, fruiting starts after 3 to 4 years of planting. A 3-year-old tree may yield 0.5-1.0 kg of seeds annually. An average of 300-600 kg of seed may be expected to yield per hectare (Math et al., 2016). A gradual decline in the yield may be found after 10 years of planting. The shapes of the seed differ from pyramidal to nearly conical (Vilar et al., 2014). Once, the fruit gets matured and dried, the capsules split open, showing the seed out.

The seed produces yellow-red-orange pigments. The main pigments are bixin and norbixin. Bixin is a di-apocarotenoid, an oil-soluble compound, extracted from the pericarp of the seed. While norbixin is a water-soluble pigment. Both bixin and norbixin are safe, eco-friendly and approved food colorants for their antioxidant and antibacterial properties (Balakrishnan et al., 2021). It is widely used in the food industries, cosmetics, pharmaceutical and textile sectors (Vilar et al., 2014).

The bixin content of the seed may vary depending on the cultivar and the climatic conditions (Vilar et al., 2014). Heavy rain and dehiscence of pods are the major issues that may cause injury to the fruit and leads to the loss of seed. Due to the lack of a single cultivar in abundance, in most countries, different cultivars are planted in a single area. In India, Southern states like Andhra Pradesh, Tamil Nadu, Karnataka and Kerala are the major producers of annatto seeds. Different cultivars of annatto are produced through advanced biotechnological or plant breeding technology to meet the demand for quality bixin (Aksatha et al., 2011).

Post-harvest operations such as drying, separation of seed, cleaning, packaging and storage plays a vital role to retain the quality of the seeds. In general, the seeds are separated manually by beating the fruits. The pressure applied to the seed may cause pigment loss (Kumaran, 2014; Math et al., 2016). The use of power operated seed separator or decorticator may reduce the loss of pigment and minimize the time of processing. Insight from the above-mentioned content, the study aims to analyze the engineering properties of two different variety annatto (Bixa orellana L.) fruit. This aid to understand their morphological features that are useful to design a mechanical decorticator for annatto seed separation.

Annatto fruits were collected from Forest College and Research Institute, TNAU, Mettupalayam, Tamil Nadu, India from November to February 2021-2022. The engineering properties such as physical, gravimetric and frictional properties were analyzed for two varieties (A₁ and A₂) of dried annatto fruit (100 nos.) and their seeds. The moisture content of the fruit and seeds was determined as per the standard method for the analysis of oil seed, IS:3579; using a hot air oven at a temperature of 105±1°C for 24 hours (Math et al., 2016). The basic dimensions like length (l), breadth (b) and thickness (t) of fruits and seeds were measured using a digital vernier caliper having 0.01 mm accuracy. Using the basic dimensions of the fruits and seeds, the arithmetic mean diameter (Da), geometric mean diameter (Dg) and equivalent mean diameter (De) were calculated by the standard formulae stated by Mohsenin (1986). The projected area of the fruits and seeds such as long (length), intermediate (breadth) and short (thickness) dimensional axes were calculated as per the standard formulae given by Mohsenin (1986). From the values of the projected area of length, breadth and thickness, the criteria projected area was calculated. To determine the shape of the fruits and seeds, parameters like sphericity, aspect ratio and flakiness ratio were calculated using standard formulae Mohsenin (1986). The surface area and volume (oblate, spheroid and ellipsoid) were also calculated by definite formulae. The pod and seed ratio was determined by measuring the mass of the seeds and pods of the fruit. The gravimetric properties such as true density, bulk density and porosity were determined using the water displacement method, mass-by-volume ratio method and the relationship between true and bulk densities of the samples, respectively Mohsenin (1986). In frictional properties, the angle of repose was determined by pouring the materials from a set height on a circular base to form a heap to find the angle formed. The static coefficient of friction was determined for four different base materials like stainless steel, aluminum, mild steel and rubber (Kaliniewicz et al., 2015). The colour of the seed was determined using a handheld tintometer which gives L*(Lightness-darkness), a*(red to green), b* (yellow-blue) and h° (0-45°, redness) values (Pathak et al., 2020). Bixin content was analyzed by the spectrophotometric method given by FAO (2006). The significant difference between the two varieties of fruits and seeds was statistically analyzed by Z- the test and t-test (p<0.05) based on the sample size.

Physical properties
 
The mean values of the physical properties of annatto fruits and seeds (A₁ and A₂) (Fig 1) are presented in Table 1 and Table 2. The mean moisture content of fruits and seeds was 9.27±1.2% (w.b.) and 6.6±1.50% (w.b.), respectively, an optimum and safer moisture level for storage and easy removal of seed from the pod (Math et al., 2016). The mean mass of A1 and A2 fruit was 2.39±0.48 g and1.35 ±0.38 g, respectively. The mass of the A₁ fruit was 77.03% higher than the A₂ fruit. The pod ratio of A₁ and A₂ fruit was 57.09% and 43.14%, respectively. The number of seeds observed in the A₁ and A₂ fruit was around 15-20 and 20-40, respectively. This observation was on par with Vilar et al., (2014), Math et al., (2016) and Umadevi et al., (2020).







The mean length (38.45±4.65 mm), breadth (27.64±3.10 mm) and thickness (24.34±2.26 mm) of the A₁ fruits were found to be larger than the A₂ fruits but it was contrary in the case of the A₁ variety seeds. There was no significant difference in breadth and thickness between A₁ and A₂ seeds. The geometric mean diameter (Dg), equivalent mean diameter (De) and arithmetic mean diameter (Da) of A₁ and A₂ fruit were 29.43±2.71 mm, 29.49±2.73 mm and 30.15±2.86 mm; 23.00±2.39 mm, 23.24±2.37 mm and 23.00±2.39 mm, respectively. The equivalent and arithmetic mean diameter of the fruit and seed exhibited very close resemblance in values but showed significant (p<0.05) difference between the variety (Pathak et al., 2020), this helps to design the hopper, selection of screen size and shape, provision of concave clearances between the screen and the beaters for the separation of seed from the pod based on the shape of the fruit and seed.

The sphericity of A₁ and A₂ fruits and their seeds were 76.89%, 70.7%; 80.78% and 76.50%, respectively. This shows that fruit and seed were merely spherical or oblong (Vilar et al., 2014; Pandey et al., 2019).  The sphericity and aspect ratio defines the good flowability of the fruit and seed. The flakiness ratio is relative to the sphericity and aspect ratio; the obtained value shows that the fruits and seeds of both varieties were not flaked. These properties indicate the rolling or sliding nature of the fruit and seed. There were no significant differences in sphericity, aspect ratio and flakiness ratio between the seeds (p>0.05).

There was a significant difference in the projected area perpendicular to the length, breadth and thickness between the fruits. Since the A₁ variety fruits are large, the criteria projected area value was higher. In seed, the criteria projected area showed a significant difference based on the projected area perpendicular to the length of the seeds. These parameters help to determine the water loss during drying and forecasting of harvest time (Pathak et al., 2020). The surface area and volume of the A1 fruit were 38.78% and 52% higher than the A₂ fruit. Whereas the A₂ fruit seeds showed 28.68% and 39.28% higher surface area and volume. This help to calculate the capacity of a machine, storage structures, etc. A₁ fruits were large and required more space than A₂ (Pathak et al., 2020). ­
 
Gravimetric properties
 
The gravimetric properties like bulk density, true density and porosity were determined to find the storage volume, resistance to the airflow, etc (Table 3 and Table 4). The bulk and true density of A1 and A2 fruit were 38.26±0.28 kg m-3, 44.98±0.14 kg m-3; 1033±10.73 kg m-3 and 515.00±13.2 kg m^-3, respectively. The porosity value of A₁ and A₂ fruit was 96.26% and 91.26%, respectively. The higher porosity was due to the expansion and splitting of dried pods (Kumaran, 2014; Pandey et al., 2019). The true density of the seed was higher than the pod because of the individual mass of the seeds. The mean porosity of the A1 and A2 seeds was 61.16±0.006% and 69.08±0.016%, respectively. All the gravimetric properties between the variety of fruits and seeds showed significant differences with p<0.05.




 
Frictional properties
 
The mean angle of repose of A₁ and A₂ fruit was 40.44±0.74° and 43.53±0.62°, respectively. The higher repose angle represents the higher friction between the fruits, the presence of bristles on the outer surface resulted in higher internal friction. A similar angle of repose value was observed in both varieties of seeds (37.2±0.4°). Internal friction between the seeds was minimum than in fruits. The lowest static friction (0.039, 0.121) was observed for stainless steel (Murakonda et al., 2022) followed by other materials like aluminium, rubber and mild steel for fruits. Stainless steel can be suggested for conveying systems for fruit with minimum friction that reduces energy losses. There was a significant difference in frictional properties between annatto fruit due to their difference in mass, shape and texture. In terms of seeds, mild steel and stainless steel showed the lowest static friction. Higher static friction was observed in materials like rubber and aluminium. Higher frictional value was observed in seeds because of their resinous pericarp (Bindyalaxmi et al., 2022). There was no significant difference (p>0.05) in the angle of repose between the seeds.
 
Colour and bixin content
 
The annatto seed colour was represented by L*, a*, b* and h° values; the mean colour values were 21.70, 18.81, 6.16 and 40.60°, respectively. The positive a* and b* value signifies the reddish and yellowish nature. A huge angle of less than 45° confirms that annatto falls under the red-orange colour group (Grillitsh, 2019). The primary pigment in annatto seeds are bixin and norbixin (carotenoids) a reddish-yellow colour (Cevallos et al., 2021). A₁ and A₂ seeds had a bixin concentration of 1.5±0.4% and 1.84±0.57%, respectively. Location, soil, environment and weather all have an impact on bixin content (Math et al., 2016; Umadevi et al., 2020).
 

Assessment of the engineering properties of two different annatto fruit and seed varieties showed a significant difference in morphological characteristics. The A₁ fruit was long and wide, with thick pods. The A₁ variety fruit was large while the A₂ fruit was small. A₂ variety fruit possessed a large number of seeds than A₁. The fruit and the seed were spherical or oblong and merely heart-shaped or flattened. There was a significant difference in the morphological characters between the variety. The variety showed a difference in bixin content. These parameters will help to design a mechanical decorticator to separate the seeds from the pods without much loss in the pigments.

Sincere thanks to the Department of Science and Technology, Science and Engineering Research Board, Core Research Grant (DST-SERB-CRG), for the financial support and Forest College and Research Institute, Mettupalayam for providing the raw materials.

None