4.1.3 The beginning of small-scale commercial neem production

In 1994, the GTZ-IPM Horticulture Project (GTZ-IPMH) in Kenya directed its attention towards neem as an alternative to synthetic pesticides, in the context of integrated pest management programmes (IPM), to increase production of horticultural crops for export and local consumption in an environmentally sound and economically viable manner (Loehr et al 1997).

Based on experience from other countries, GTZ-IPMH focused its efforts on establishing a small-scale industry for producing simple, ready-to-use neem-based pesticides.

As a first step the GTZ IPMH project funded two feasibility studies, which were conducted in the coastal area in Kenya during 1994 -1995. The aim of these studies was to determine the economical and technical feasibility of producing neem-based pesticides. In January 1996, a local agrochemical company called Saroc Ltd (today: Saroneem Biopesticides Ltd) was contracted to organise the first commercial seed collection and to establish a pilot processing unit for extraction of oil from neem seeds. When these studies showed that Kenya had good potential for growing neem and first experiences were gained with processing and seed collection, measures were taken by IPMH and the Pesticide Service Project to initiate a separate project. This was started in mid-1996, as a two-year project called "Small scale industrialisation of neem-based insecticides in Kenya", funded by the GTZ and set up at ICIPE. The aim of this project was to produce simple, standardised, neem-based pesticides which could be purchased on the market at competitive prices.

ICIPE contracted Saroc Ltd to organise collection of seeds, to develop neem-based pesticides and to coordinate marketing of the products. A part-time scientist was contracted by ICIPE to coordinate research on the efficacy of the developed products and to generate the necessary information for registration of the developed products.

The main activities and business of Saroc are confined to the formulation of imported pesticides, mainly fungicides based on copper, and importing fertiliser for the east African market. At the end of 1999 the conventional agricultural supply wing was separated off and sold, while the neem-processing wing continued to operate under the name of Saroneem Biopesticides Ltd (for more data see: "description of the plant", below).

Development of neem products

· Seed collection:

The peak fruiting season in coastal Kenya and Tanzania is from March to May, which coincides with the rainy season. It was found feasible to organise collection of seeds through existing organised groups, especially women's groups, which expressed interest in a collection scheme. However, the feasibility studies concluded that the success of seed collection depended on several factors, such as:

· educating and familiarising local residents with neem and its use as an insecticide well in advance of the fruiting season;

· training on collection, cleaning and drying of neem seeds;

· setting the prices to take into account factors such as seed availability, wage levels for casual workers, the amount of time required, and opportunities of alternative sources of income. This should involve discussion with other organisations collecting seeds (DM 1994, 1995).

Collection started in 1996, when Saroneem Biopesticides Ltd staff met with representatives of the local authorities and rural farmers' organisations, women's groups and pupils from primary and secondary schools. The aim was to identify, motivate and train potential collectors. It turned out that schoolchildren made excellent multipliers, since they pass on their knowledge and information to their families.

During the first collection it was found that a further factor was crucial to seed collection, namely creation of confidence between the collecting communities and the purchaser. Due to bad experiences in the past, villagers were extremely cautious and were not prepared to work if they were not fully assured of payment. Nevertheless, almost 7000 kg of seeds were collected by women's groups, schools, and local farmers. The seeds were then transported to Nairobi for processing (Rocco 1996).

Saroneem Biopesticides Ltd organised meetings in the coastal area and in the north-eastern region to inform the local population of the possibilities of earning money through seed collection, and to train them in harvesting, cleaning and drying the neem seeds. However, no collections were made in the north-eastern region due to the high costs of transporting the seeds to Nairobi. Thus, seed collection was concentrated along the coastal region in Kenya (Kilifi and Malindi Districts) and Tanzania. The average distance to the neem processing plant is about 730 km.

A series of local organisations, women's groups, farmers and their families as well as inmates of prisons were involved in seed collection. Since the harvest period coincides with the period of heaviest rains along the coast, special attention had to be given to the proper storage and drying of seeds to avoid fungal contamination and degradation of the seeds due to the high humidity.

Luckily, at the coast the harvest of seeds is facilitated by large colonies of fruit bats and birds, which pluck the ripe fruit off the tree, feed on the sweet pulp and then spit out the seed, which can be found lying under the tree. This saves depulping, which is a labour-intensive step in processing.

The following table (9) indicates the amount of neem seeds purchased in the last four years (since the beginning of the neem-processing activities).

Table 9: Neem kernels purchased by Saroneem Biopesticides Ltd

Year	Neem seeds purchased (kg)
1996	6,700
1997	16,755
1998	13,474
1999	31,148
Total	68,077

In each season Saroneem Biopesticides Ltd purchased the total amount of collected and usable seeds to create confidence in its collection system and motivate the farmers to collect seeds in the forthcoming seasons.

About 17 tonnes of neem seeds were collected and delivered to Nairobi in 1997. The collection in 1998 was hampered by the onset of unusually heavy and long rains at the end of 1997. Many of the flowers aborted, limiting the amount of fruit produced. Moreover, the seed quality was affected by the prolonged moisture conditions. About 14 tonnes were collected in 1998. In 1999 in total 2000 people were involved in collecting the 31 t of neem seeds. These are individuals and groups, mainly women and children, who see the collection of neem seeds as an additional source of income.

Seeds were delivered to collection points and transported to Nairobi. In 1999 the collections were organised by four agents situated in the coastal region, which purchased the collected neem seeds on behalf of Saroneem Biopesticides Ltd.

During the harvesting season the collected neem seeds are checked at the collection point, and the deliverers paid promptly. There the seeds are packed, and they are collected and transported to Nairobi twice a week.

The collectors receive 20 KSH (70 KSH = US$ 1) for 1 kg of usable neem seeds from the agents, or 5 KSH for lower quality seeds. The agents in their turn get 25 KSH for 1 kg of good quality seeds and 5 KSH as lump sum for packing and transport. The azadirachtin content was about 0.3 or 0.4% in 1999. The seeds are graded (and paid for) according to following quality criteria: contamination by fungi, cleanliness and moisture content. Oil and azadirachtin content are not considered because they cannot be checked on the spot.

It is difficult to obtain reliable figures on the yield per tree due to the heterogeneous growing conditions. According to rough estimates, about 30 kg of neem seeds are produced by a 12-15 year old tree, and about 12 kg per tree are gathered by the collectors. The yields, however, vary from year to year according to climatic conditions. Considering younger neem trees as resource trees, about 3500 to 4500 neem trees are required to obtain 31 t of seeds.

Seeds are transported from the collection area to the district capital by private pick-ups, and from there to Nairobi by frequent buses. Transport costs for the neem manufacturing plant are 30 000 KSH/t.

For the year 2000 Saroneem Biopesticides Ltd intends to purchase 50 t neem seeds, provided sales are according to the expectations. It is considered that it is feasible to extend and intensify the collection of seeds if more seeds are required.

· Processing

In order to overcome the difficulties of storing seeds under the climatic coastal conditions it was decided to undertake post-harvest development in Nairobi. For this purpose a pilot unit was established.

Description of the pilot plant

The machinery used was developed and manufactured by testing and adapting local machinery for groundnut processing, with the exception of a "Comet" oil expeller manufactured in Germany. The machinery included:

· a groundnut sheller, modified to cope with the much smaller neem seeds, for breaking the shell to liberate the seed kernel,

· a winnower to separate the kernels from the shell,

· an oil expeller to extract the oil from the kernels (later a double spindle oil expeller with a higher processing capacity was added),

· a hammer mill to crush the neem cake,

· an automatic packing machine.

The main problems faced during processing were:

1. Drying of seeds was difficult under the prevailing weather conditions and in the limited space available at Saroc Ltd in Nairobi. It was found that oil extraction was very limited unless the kernels were properly dried.

2. Crushing neem kernel cake alone was not easy as it jammed the hammer mill. This problem was solved by mixing the neem kernel cake with neem shells.

3. Contamination of neem products was possible since the company was producing agro-chemicals under the same roof.

4. The lack of equipment for analysis of azadirachtin hampered the formulation of standard products.

After these initial experiences the pilot unit developed by Saroc was moved to ICIPE headquarters in Nairobi. A section of the former livestock centre was made available for the processing unit. It consisted of a large covered shed of about 600 sq m, a large enclosed room, an office and a laboratory block. The covered shed was suitable for drying and shelling the seeds, whilst processing, formulation and packing could be restricted to the enclosed area, thus minimising the risk of contamination.

The processing capacity was later increased (see above).

Formulation of neem products

Pesticides for seed treatment, soil treatment and foliar spray were developed based on neem kernel cake powder and neem oil. Processing of neem seeds into pesticides includes the following steps:

1. Disinfecting and drying the seeds: the seeds are treated with a solution of calcium hypochlorite to prevent fungal growth and subsequent aflatoxin contamination.

2. Drying the seeds: the seeds are spread in thin layers and exposed to the sunshine. During the evening the seeds are covered with polyethylene sheets.

3. Threshing the seeds: 50% of the seeds are passed through a modified groundnut sheller to liberate the kernel from the seed shell.

4. Winnowing: a fan blowing through a constricted space is used to separate shells from kernels.

5. Extraction of oil: a mixture of kernels and seeds is passed through the mill. Trials on the most effective ways of extracting oil from the seeds showed that a mixture of 2/3 kernels to 1/3 shells maximised the extraction of oil.

6. Filtering the oil produces the pharmaceutical by-product "Neemsar "O" "The neem cake can be either used to produce neem powder or for alcoholic extraction. The quantity depends on the demands for the specific product lines.

7. The azadirachtin content in neem cake is quantified using high performance liquid chromatography (HPLC) at the chemistry department of ICIPE. Standardisation of azadirachtin A content in neem cake powder is achieved by reconstituting the neem cake with neem seed shell in appropriate proportions. A hammer mill and mixer are used to grind and mix neem kernel cake and shells. The final product is a neem powder of 0.5% azadirachtin, called "Neemros 0.5% Powder".

8. For alcoholic extraction the neem cake stays in an alcoholic extractor for 8 h. 100 l of alcohol is required for 130 kg cake. The alcohol percolates continuously through the cake, extracting part of the remaining neem oil and other active ingredients. The plant in Kenya has been using isopropanol (isopropyl alcohol), due to its easy availability and cheap price (but it is now using ethanol).

9. The neem oil (32%) is combined with alcohol (63%) and an emulsifier (5%). This marketable product, an emulsifiable concentrate, is sold under the trade name "Neemroc EC 0.03% W/W".

10. The extracted cake mixed with neem shells in a ratio of 1:1 is milled with a hammer mill and can be sold as organic fertiliser. To date Saroneem Biopesticides has not traded this product.

11. Formulation of neem products: pesticides for seed treatment, soil treatment and foliar sprays were developed based on neem kernel cake powder and neem oil. Two formulations were developed:

A) Neemros: Neem cake powder (NCP) standardised at 0.5% azadirachtin and

B) Neemroc EC oil: a water-miscible oil with 0.03% azadirachtin. For clarification the processing of neem cake is explained below:

The shells are removed from 1 t of neem seeds. The shells make up about 25% of the seeds' weight, coming to 250 kg. From the remaining 750 kg kernels 120 kg cold neem oil is pressed (16%). This leaves 630 kg neem cake which is mixed with 100 kg neem shells for technical reasons. Together the neem cake and seed shells amount to 730 kg of neem powder.

Figure 3: Processing of neem seeds into pesticides includes the following steps:

*Officially Neemroc has a minimum azadirachtin content of 0.03%, although it is usually higher than this.

In 1999 Saroneem Biopesticides Ltd purchased 31 t of neem seeds and from this manufactured the following quantities of marketable products:

Table 10: Expected output of marketable neem products

Raw material: 31 t neem seeds
Product	Output
Neemroc	10,000 l
Neemros	10,000 kg
Neemsar "O"	60 l
Organic fertiliser	10,000 kg

The above data concerning marketable products could vary according to the demand. The figures are estimates based on the experience gained in previous years.

Quality Control

Quality control takes place with regard to the following components: aflatoxin (not desired) and azadirachtin content (desired, should be as high as possible).

The seeds are checked for contamination with the storage fungus Aspergillus spp. The absence of aflatoxin in the seeds and in the finished products (neem oil, neem cake and neem powder) is regularly monitored (twice a month) at the chemistry department at ICIPE. The technician in charge of the analysis at ICIPE took part in a training course in extraction and bioassays of biologically active plant products, offered by the GTZ Pesticide Service Project in 1995. During 1996-1997 duplicates of some samples were sent for analysis to Trifolio-M GmbH in Germany for the sake of comparison.

The shelf-life of the products was determined by monthly HPLC azadirachtin A analysis of stored products and seeds. The changes in content of azadirachtin A in neem cake powder, stored at room temperature, during a ten-month period ranged from 0 to 30% reduction. The analyses showed that azadirachtin A was better conserved when the seeds were stored and then processed as needed.

The following table (11) provides data on azadirachtin and aflatoxin content according to the quality control laboratory at ICIPE in the latest analysis of products from Saroneem Biopesticides Ltd:

Table 11: Results of the quality analysis of Saroneem Biopesticides Ltd products in October 1999:

Product/material	Azadirachtin	Aflatoxin (mg per g)
Neem cake	5	below the limit of determination
Neem powder	3.9	below the limit of determination
Neem oil	0.2	not detectable
Filtered neem oil	0.2	not detectable
Neem seeds	3.2	below the limit of determination

The quality criteria for marketable neem-based pesticides are defined as follows:

· constant azadirachtin content
· constant consistency
· reliable efficacy against pests
· aflatoxin content below defined limit

· Registration of neem-based pesticides in Kenya

All pesticides to be used in the Kenya have to be approved by the Pest Control Products Board (PCPB), the authority established by the government of Kenya to take care of pesticide regulation. The requirements for registration of plant derivatives are the same as for synthetic chemicals. Information on chemical composition, toxicity and efficacy is required. Efficacy trials can only be conducted by institutions recognised by the PCPB. The costs of efficacy trials conducted by governmental institutions such as the Kenya Agricultural Research Institute (KARI) have to be covered by the company making the application. Other institutions charge a fixed fee for conducting efficacy trials. Once all requirements are met, a fee of KSh 10,000 (about US$ 140) per product has to be paid for provisional registration and 30000 KSh (ca. US$ 400) in order to apply for permanent registration.

Applications for the registration of the two products, Neemros® and Neemroc® were presented to the PCPB in 1996. Information on the chemical composition and toxicology of neem and neem products was gathered from the literature and from information provided by the GTZ Pesticide Service Project.

Field and laboratory trials of the efficacy of the two products for the control of key pests of several crops were conducted. These trials were also intended to determine the frequency of application and effective rates of application that are economically competitive with other available pesticides. The trials on horticultural crops were carried out in collaboration with the GTZ-IPMH Project, the ICIPE-USAID Export Vegetable IPM Project, and the Kenya Agricultural Research Institute (KARI), as well as fruit, vegetable and cut-flower growers. Trials on maize and banana were conducted by the ICIPE Neem Awareness Project. Some of these studies were done as MSc theses by students from local and German universities under the supervision of scientists from ICIPE and the respective university.

A temporary Certificate of Registration for Neemroc® and Neemros® for use on horticultural crops was issued by the PCPB in March 1998. Further information on efficacy of the products in the field and effects on non-target organisms might be required in order to get a permanent certificate of registration.

At the end of 1999 Neemroc and Neemros were provisionally registered for horticultural crops.

These products are being tested, but are not yet registered in Kenya for potatoes/tobacco.

· Vegetables and fruits
· Flowers and ornamentals
· Potatoes
· Tobacco

· Results of the efficacy trials

Pests of brassica crops

Foliar applications with Neemros® water extracts at concentrations of 25 and 50 g/l, and Neemroc® at 1 - 3% have compared favourably with the Bacillus thuringiensis products Florbac and Dipel, for control of the diamond-back moth, and with Karate for control of aphids on kale and cabbage in greenhouse and field experiments (K 1998, Okoth 1998).

Pests of French beans

Foliar applications with Neemroc® at concentrations of 1 - 3% provided very good control of A. fabae on French beans in greenhouse trials. This aphid was also satisfactorily controlled by foliar applications of Neemros® water extracts at a concentration of 50g/l (Maundu 1997).

Pests and diseases of tomato

Foliar sprays of Neemros® of 10 - 50 g per litre of water controlled dipteran leaf miners Lyriomyza spp. on tomato in field experiments (Pacho, in preparation).

Aqueous Neemros® extracts at concentrations of 5 - 50 g/l inhibited mycelial growth, sporulation and germination of Fusarium oxysporum fsc lycopersici. These applications reduced the pathogenicity of the fungus (Stanley 1998).

Dipping bare-root tomato seedlings into aqueous Neemros® extracts (50g/l) protected the seedlings when planted in soil inoculated with mycelium of F. oxysporum fsc lycopersici compared with plants dipped in water. The reduction in the severity of the disease increased as the duration of dipping increased (Stanley 1998).

Soil additives with Neemros® have proven to be promising for controlling root-knot nematodes on tomatoes. Unfortunately, conclusive data could not be obtained since the field experiments were hampered by diseases which wiped out the experimental plants.

Pests of cut flowers

In a commercial-scale trial, foliar sprays containing Neemros® at 10 - 25 g/l of water applied in combination with Trigard (whose Al is cyromazine) controlled leaf miners on Carthamus in a flower farm in Naivasha. A significant increase in levels of the leaf miner parasitoid Diglyphus isaea Walker was observed after synthetic pesticides were replaced by Neemros® in the spraying programme. Application of chemicals was reduced by 80%. Rejection of flowers due to leaf miner damage was reduced from 70% to 6%.

Management of stemborers

Products based on neem powder have proved to effectively control stemborers on maize and sorghum. This has been demonstrated in trials with neem seed powder conducted by the Neem Awareness Project. The trials were conducted at Mbita Point and at the ICIPE field station at the coast. These results were reconfirmed in trials carried out with Neemros®. Results of trials with neem cake (NC) showed that damage and stemborer infestations on plants treated with NC were similar to those on plants treated with Dipterex and much lower than on untreated plants. Grain yields of NC-treated plants were also comparable to yields from plants treated with Dipeterx and much higher than in control plants (Anon. 1996/97).

Other potential uses of neem-based pesticides

Neem products, particularly neem-seed kernel extract and neem oil, have proved to effectively control nematodes and the banana weevil, serious pests of banana in Kenya (Musabyimana & Saxena 1999).

Trials on tobacco are being conducted by the British American Tobacco Company Group (BAT), Kenya Ltd. Trials on cotton are being conducted by an organic cotton project in Lamu District, Kenya and by the Namulonge Research Institute in Uganda.

In addition, neem-based products have shown potential for the management of ticks and tick-borne diseases on livestock in Kenya. Neem kernel extracts and neem oil are being further tested (Kaaya 1997).

Further opportunities are discussed in section IV 1.5.4 below.