The present application relates to an encapsulated pesticide and a method for producing the same. More particularly in relates to a pesticide which is a herbicide.

Herbicides are commonly used in agriculture to improve the productivity and quality of the grown product. The global market for herbicides is estimated to reach $31.5 billion by 2020. Asia-Pacific is the dominant market accounting for two fifth of herbicide use, whereas North America region equates to one-third of global revenue generated by the herbicide market.

At present, the most widely used herbicide in the world is glyphosate, which has global sales amounting to more than $10 billion per year due to low cost production and low

environmental impact. As a result of the prolonged use of herbicides, weeds are becoming resistant and this is increasing exponentially with over 217 weed species presently resistant around the world. In most cases, it is estimated that weed resistance will occur within three years on a piece of land which has been treated with a given herbicide. Current solutions are either to develop novel herbicides or switch from glyphosate-ready to glufosinate-ready crops. Both alternatives are expensive and there are no guarantees that the weed will not become resistant after 3-5 years. In case of the herbicide glufosinate, weed- resistance has already been reported. The first cases of weed-resistance were reported in 2009 and increasing use of the herbicide is likely to result in widespread resistance.

Other alternatives are to use stronger dosages of glyphosate, which could be detrimental to the environment or in extreme circumstances hand-weeding the weeds, which is labour intensive. Both alternatives lead to higher costs or danger of ruining the fertile land. EP 1 499 183 B2 (Rothamsted Research Institute Limited et al.) discloses a method for preventing or reducing resistance to a pesticide of a substrate pest, which method comprises administering to the substrate or the pest a metabolic enzyme inhibitor (such as piperonyl butoxide - PBO) and (substantially simultaneously) a pesticide (such as a pyrethroid insecticide) encapsulated in a degradable capsule. The capsule prevents an effective dose of the pesticide from coming into contact with the substrate or the pest until the inhibitor has had time to begin its inhibiting effect on the substrate. The formulation used in this case is Karate Zeon® which is a PVA-encapsulated insecticide (lambda-cyhalothrin) produced by Syngenta.

EP 0427991 Al (Sumitomo Chemical Co.) discloses an insecticidal and/or acaricidal and/or nematicidal composition which is a mixture of an encapsulated part which is formed of water-insoluble microcapsules and a flowable part which is emulsified or suspended in water.

Other prior art includes:

Roy et al. (J. Macromol. Sci., Part A: Pure and App. Chem., 46, 847 (2009))

Roy et al. (Carbo. Polymers., 76, 222 (2009))

Davarci et al. (Food Hydrocolloids, 62, 119 (2017)

Wu et al. (Carbo. Polymers, 110, 259 (2014))

Singh et al. (J. Envir. Sci. and Health Part B, 44, 113 (2009))

US 6 248 321 B l (Her Majesty the Queen in right of Canada)

WO 93/01713 (The United States of America)

CN 104 904 710 A (Huazhong Agricultural University)

US 5 599 767 (Micro Flo Company)

EP 0 051 161 Al (Agro-Cap International Inc)

WO 02/080881 (Universite Laval)

WO 2009/062254 (The University of Queensland)

CN 101 889 587 A (Nanjing Forestry University)

CN 103 548 995 A (Meizhou Wuzhishi Technology Co. Ltd. and University of Jiaying)

WO 2016/176764 (Green Advantage Technology Inc.)

CN 105 727 889 A (Huaiyin Institute of Technology)

CN 104 094 930 A (Shandong Agricultural University)

US 5 160 530 (Griffin Corporation)

CN 105 230 612 A (Zhongkai University of Agriculture and Engineering) US 5 629 187 (LVMH Recherche)

CN 105 145 622 A (Shandong Weifang Rainbow Chemical Co. Ltd.)

CN 101 100 646 A (Institute of Microbiology Chinese Academy of Sciences)

CN 106 614 556 A (Institute of Hydrobiology Chinese Academy of Sciences)

KR 100 864 399 B (Industry Academic Cooperation Foundation Gyeongsang National University)

KR 2007 002 4792 A (Hwang Gyeong Suk)

WO 89/07447 (President and fellows of Harvard College)

CN 177 5039 A (South China Agricultural University)

WO 97/20462 (The United States of America)

US 4 908 233 (Lion Corporation)

WO 2014/164418 (North Carolina State University)

WO 89/10117 (Southwest Research Institute) The present application seeks to provide an improved encapsulated pesticide.

In accordance with a first aspect of the present invention, there is provided a method for producing an encapsulated pesticide, including the steps of

(a) providing a mixture of a pesticide and at least one polysaccharide in either (i) an organic solvent or (ii) in aqueous solution,

(b) combining the mixture of step (a) with water in the case of (i) or an oil in the case of (ii) and stirring to create an emulsion of said mixture,

(c) adding to said emulsion a salt in powder form, which salt is formed of a cation which reacts with said polysaccharide to produce a water-insoluble reactant, and

(d) stirring the product of step (c) in order to produce particles of pesticide encapsulated in said water-insoluble reactant.

In a preferred embodiment, the pesticide is water soluble. It has been discovered that a capsule coating prepared in this manner has at least some of the following advantages: The formulation process is simple and inexpensive

The coating material is readily available, cheap and environmental friendly i.e. nontoxic and biodegradable.

Microcapsules of less than- 100 μιη in diameter can be produced which can be sprayed using existing equipment on farms without clogging the spray nozzle

The concentration of glyphosate in the encapsulated solution can be similar to existing commercial glyphosate solutions (range 330-470 g Γ ¹ )

The encapsulation efficiency of glyphosate can be as high as possible (more than

70%)

Delayed release of up to 2 hours and optimally 4 hours can be obtained

The formulation process is relatively unhazardous and avoids use of highly flammable organic solvents

Without wishing to be constrained by theory, it is thought that the glyphosate diffuses out of the microcapsules when deployed, the microcapsules staying intact. This is in contrast to prior art formulations (for example in which glyphosate is encapsulated in PVA capsules) in which the capsules need to rupture in order to deliver the glyphosate to the locus of the substrate. This requires an external force or a chemical degradation of the capsule wall, which is difficult to control and predict.

By contrast, diffusion is a predictable process, the rate of which is proportional to

temperature and has a predictable dependence on pH conditions. Thus, a farmer can be given a chart which gives clear instructions as to the dependence between environmental conditions and glyphosate delivery rates.

Preferably, the at least one polysaccharide includes (or consists of or consists essentially of) alginate, chitosan or any combination thereof, and more preferably additionally includes pectin. Most preferably, the polysaccharide component is a combination of alginate and pectin, with preferred ratios being 25% alginate, 75% pectin; 50% alginate, 50% pectin; and most preferably 75% alginate, 25% pectin. Again, without wishing to be constrained by theory, it is thought that the inclusion of pectin makes the walls of the capsules more compact and affects the diffusion rate of pesticide therethrough.

The cation which combines with the polysaccharide to produce the capsule walls is preferably calcium or barium, most preferably calcium. This is preferably provided in the form of the chloride salt. It has been discovered that the provision of the salt in the form of a powder is necessary in order to create the microcapsules, otherwise a gel is formed.

In a preferred embodiment, the average diameter of the particles is less than 100

micrometres. This is principally so that the formulation can be used with existing farm machinery and does not get clogged in the spray nozzle. More preferably the particle size is from 2-100 micrometres as it has been discovered that the smaller the particle size, the faster the pesticide release and that size range gives the best range of release times.

In order to produce a formulation having the preferred size distribution, it is preferred that the rate of stirring of step (d) is greater than 800rpm, more preferably greater than or equal to lOOOrpm.

The method may include the additional step of including in the water or oil of step (b) a substance which is an inhibitor of a factor causing or contributing to the resistance of the pest to the pesticide (such as piperonyl butoxide). In an alternative aspect of the invention, there may be provided a formulation of an encapsulated pesticide as defined above and said inhibitor.

In a further aspect of the invention, there is provided the use of a method or formulation as defined above in the treatment or prevention of damage to a substrate by a pest.

There are significant advantages to using a formulation in accordance with the invention in the field, including:

1. The product significantly improves the speed of weed kill, compared to standard offering. This feature will provide significant value to growers where there is a tight window to kill weeds between crops. This product will provide growers with more flexibility in their growing rotation.

2. The combined mix of enhanced PBO formulation and encapsulated glyphosate has proven to break down glyphosate resistance in some target weeds. Glyphosate resistance is recognised as one of the largest issues facing global agriculture. This encapsulated glyphosate product has enormous potential in the USA, China, India, Australia, Africa and South America where glyphosate resistant weeds have become an enormous issue as a result of the high use of glyphosate in their cropping systems.

3. The encapsulation of the glyphosate significantly reduces Operator exposure', in effect making glyphosate 'safer' compared to the standard offering. This feature of the product will be of great value to the operator and will create interest in both the amenity and 'home and garden' sectors.

4. From initial trials, it also seems possible to reduce the amount of active ingredient required to achieve the same efficacy as the label rate on present glyphosate herbicide formulations. This will be further tested during this growing season, but if this is the case the upside from an environmental perspective will be enormous. The fact that most countries are trying to reduce the amount of glyphosate being applied as it is being found in water, milk, beer and foodstuffs - means that such a rate reduction will be welcomed by most countries. This should allow a premium to be placed on the CS formulation (over the present standard price of glyphosate active ingredient, in formulations) and result in a price equivalent to a minimum of the full rate, providing an opportunity for good returns and at the same time a reduction of active ingredient in the environment.

A number of preferred embodiments of the present invention will now be disclosed by reference to the accompanying drawings, in which: Figure 1 is a graph showing the effect of agitation speed on the microcapsule size;

Figure 2 shows micrographs of glyphosate encapsulated with various different coating compounds; Figure 3 is a graph showing the cumulative glyphosate release profiles for different coated capsules in 100 % humidity at 25 °C;

Figure 4 shows a series of cumulative glyphosate release profiles for the different coated capsules in 100 % humidity in different pHs and at 25 °C;

Figure 5 is a graph showing the effect of humidity on the release of glyphosate from capsules coated with alginate: pectin (75:25);

Figure 6 is a graph showing the effect of temperature on the release of glyphosate from capsules coated with alginate: pectin (75:25);

Figure 7 is a graph showing the effect of UV/visible light exposure on the release of glyphosate from capsules coated with alginate: pectin (75:25);

Figure 8 is a graph showing the glyphosate release profiles for capsules formed of different ratios of active and coating material;

Figure 9 is a graph showing the glyphosate release profiles for capsules formed in different continuous phases; and

Figure 10 is a graph showing the glyphosate release profiles for capsules formed in

PBO as continuous phase and release done in water and 2 % (w/w) ammonium sulfate salt.

Experimental

Materials

All the chemicals were purchased from either Aldrich or Fisher Scientific, whereas the solvents were purchased from Fisher Scientific.

• Alginic acid sodium salt from brown algae (Aldrich; product number A0682).

• Pectin from citrius peel with galactaronic acid > 74 % (Aldrich; product number P9135).

• Chitosan from crab shell, with at least 85% deacylated (Aldrich; product number 48165).

• Gelatine from bovine skin with around 75 bloom (Aldrich).

• Calcium chloride anhydrous powder (Fisher Scientific; product number

1.02378.2500).

• Glacial acetic acid (Fisher Scientific; A/0360/PB17). • Phosphoric acid (Aldrich; product number 79617).

• Glyphosate (99 %, Aldrich: product number 455251).

• The encapsulated 67 % glyphosate isopropylammonium solution was supplied by Pangea Chemicals.

• Isopropylamine (Aldrich: product number 471291).

• Sunflower oil (supermarket brand; ASDA).

Emulsion-gelation method

• Aqueous solutions of 4 % (w/w) of coating materials were prepared in distilled water.

The only exception being chitosan where 1 % solution was prepared in 1 % (v/v) acetic acid, which was also prepared in distilled water.

• For the formulations containing the mixed coatings; the mixed coating solutions were prepared by stirring the required w/w ratios of the individual coating solutions for 5 min prior to the addition of glyphosate salt to ensure homogeneous mixture.

• An aqueous solution of coating material (25 g) was mixed with aqueous solution of 67 % (w/w) glyphosate isopropylammonium solution (5 g) using a magnetic stirrer at 200 rpm for 2 min.

• The resultant solution was added dropwise to a stirred 250 ml breaker containing sunflower oil (100 ml) to form an emulsion using a homogeniser (IKA) at 1000 rpm with blade diameter 27 mm.

• After 30 min, calcium chloride(l g) powder was added slowly over 10 min. Small portions of powder (0.1 g) was sprinkled using a stainless steel spatula to ensure the powder was evenly distributed over the reaction vessel. The calcium chloride reacts instantaneously when it comes in contact with the coating material.

• The resultant reaction mixture was further stirred for 30 min at 1000 rpm using a homogenizer. Observe formation of white capsules settling to the bottom of the breaker.

• The resultant reaction mixture was separated in two centrifuge tubes (50 ml) and centrifuged for 10 min at 3300 rpm.

• The supernatant was discarded. • Hexane (50 ml) was added to each centrifuge tube (50 ml volume) and the suspension centrifuged for 10 min at 3300 rpm.

• The supernatant was decanted and the process was repeated once again.

• The microcapsules were placed under high vacuum overnight to remove remaining solvent residue.

glyphosate

Scheme 1. Preparation of microcapsules via the Emulsion-gelation method.

Capsule Characterisation

Encapsulation Efficiency (ee)

· The ee for the capsules were determined via thoroughly grinding the capsules (100 mg) periodically for 1 min after intervals of 15 min over 1 h using a mortar and pestle in an aqueous solution of potassium dihydrogen phosphite buffer (5ml).

• After 1 h, the grinded mixtures were filtered (0.45 μπι) and samples were run by HPLC to determine the glyphosate concentration. Each experiment was repeated three times for each capsule.

Determination of Capsule size.

• The microcapsules were imaged and analysed using software Leica Qwin (Opem 32). Glyphosate Release Studies

Glyphosate release of a function of temperature.

• The microcapsules (1.0 g) were placed in a dialysis tube and H ₂ 0 (50 ml) preheated to required temperature and immediately placed in a preheated chamber (25 °C and 40 °C).

· The submerged samples were continuously gently agitated at 130 rpm at the required temperature. • Samples were taken periodically at 0.25, 0.5, 1, 2, 3, 4, 7 and 24 h.

• Sample were analysed by High Performance Liquid Chromatography (HPLC). For each microcapsule the experiment was repeated three times using fresh made microcapsules each time.

Glyphosate release as a function of pH.

Microcapsule release behaviour were studied over three pHs; 5, 7 and 8 at 25 °C under 100 % humidity.

• The pH of the water reservoir was adjusted to the required pH via the addition of aliquots of 1M HCl _(aq ) and 1M NaOH _(aq) .

• The microcapsules (1.0 g) were placed in a dialysis tube and in the required pH

aqueous solution.

• Placed in an oven preheated to 25 °C.

• The submerged samples were continuously gently agitated at 130 rpm at the required temperature.

• Samples were collected after duration of 0.25, 0.5, 1, 2, 3, 4, 6, 7, 8 and 24 h.

• The samples were run by HPLC to determine the glyphosate concentration. Each experiment was repeated three times for each capsule.

Capsule stability to Ultraviolent (UV) light exposure.

• Microcapsules (1.0 g) were placed in a dialysis tube and preheated H ₂ 0 (50 ml) at 25 °C was added.

• The submerged microcapsules were immediately placed in a preheated oven (25 °C) exposed to UV light.

• Samples were collected from the water reservoir after time duration of 0.25, 0.5, 1, 2, 3, 4, 6, 7, 8 and 24 h.

• The samples were run by HPLC to determine the glyphosate concentration. Each experiment was repeated three times for each capsule.

Glyphosate release under 35 % humidity • Droplets (1 μΐ) of 20 mg/ml aqueous suspension of capsules were distributed on a peltier dish whereby the total volume distributed amounted to 1 ml and the dish placed in a humidity chamber preset at 35 % humidity and 25°C (Figure 11). The surface was hydrophobic producing contact angles of 94 ± 4° between the three phases.

• After the required exposure time (0.25, 0.5, 1, 2, 3, 4, 6, 7, 8 and 24 h) the dish was washed with distilled water (1 ml) rapidly followed by filtration through a filter (0.45 μπι) to minimise further glyphosate release.

HPLC Analysis

• The concentrations of glyphosate salt present in the samples obtained for ee and release profiles were determined via running the samples with 20 μΐ injections through a Shimadzu HPLC system using a Phenosphere 5μ SAX 80 A New column 250 x 4.6 mm at 35 °C.

• All the samples were run with aqueous potassium dihydrogen phosphite buffer (pH 1.9) as the mobile phase.

Preparation of Potassium dihydrogen phosphite Buffer

• Potassium dihydrogen phosphite (0.74 g) was dissolved in degassed HPLC water (50 ml) in a 1000 ml volumetric flask.

• A degassed HPLC MeOH (160 ml) was added.

• The resultant solution was further diluted with HPLC to the 1000 ml mark on the volumetric flask.

• The pH of the solution was adjusted to 1.9 with addition of 85 % phosphoric acid and the buffer was filtered before use and degassed for 20 min.

Preparation of calibration plot

• Known concentrations of solutions ranging from 0.01 to 10 mg/ml were prepared from glyphosate (99 %) in distilled water with addition of isopropylamine (1 eq) and each sample was run three times by HPLC as explained above. The coating formulations investigated for the emulsion-gelation method are listed below. First and foremost, single material shells were investigated. To improve the shell strength, permeability, and encapsulation efficiency of process, mixed coatings formed from chemically compactable materials were also studied. The weight by weight ratios used in the formulations are shown in the brackets. A total of 25 g of aqueous solution of coating material was used in each formulation unless otherwise stated. Point to note, low mass polysaccharides were used to ensure that highly concentrated solutions of coating material could be prepared to minimise dilution of the encapsulated glyphosate solution.

• Alginate

• Chitosan

• Pectin

• Gelatine

• Shellac

• Alginate + Chitosan (75:25, 50:50, 25:75)

• Alginate + Pectin (75:25, 50:50, 25:75)

• Alginate + Gelatine (75:25, 50:50, 25:75)

The effect of agitation on capsule size

Prior to fabricating the different coated capsules a study was performed to obtain an understanding on how the agitation speed affects the size of the microcapsules formed (figure 1). The study was performed using alginate and pectin (50:50) coated capsules. From figure 1, it can be observed that the average microcapsule diameter is unaffected by the agitation speed when taking into consideration the standard deviation of the measurements. However, a narrower size distribution is observed when the agitation speed is >1000 rpm as shown by the max and min microcapsule size found in the microcapsule formed. At greater agitation speeds than 1000 rpm there is no major change in the microcapsule size range, hence there is no advantage is using agitation speeds >1000 rpm. Thus, in the experiments hereafter agitation speed of 1000 rpm was employed. Microcapsule Characterisation

Not all of the materials investigated for their viability as coating materials were successful in encapsulating the glyphosate salt. Table 1 shows the coating materials investigated, whether the capsules were formed, the average capsule size with coefficient of valence (CV), encapsulation efficiency (ee) and active load for each type of capsule formed.

Figure 2 shows micrographs of glyphosate encapsulated with (a) alginate; (b) alginate: pectin (75:25); (c) 75 alginate xhitosan (75 :25) and (d) alginate xhitosan (50:50).

• The first point to note, only pure alginate and chistosan encapsulated glyphosate

successfully, the other pure coatings; shellac, pectin and gelatine were unsuccessful. Although quasi-spherical sub- 100 μπι microcapsules (figure 2) which were within requirements, however the alginate coated capsules had low active load of 7 ± 2 whereas chitosan coated capsules had low ee of 36 ± 2.

• To improve the ee and active load alginate and chitosan were mixed with a second

chemically compactible material such as pectin and gelatine. Different compositions between the two materials in the mixed formulations were investigated as shown in table 1.

• With the exception of alginate: gelatine and 25 % (w/w) alginate composition coating formulations, the other coating formulations encapsulated glyphosate.

• All of the mixed coatings which encapsulated the glyphosate salt form quasi-spherical sub 100 μπι microcapsules (figure 2) with high EE > 75 % and improved active load in comparison to pure alginate coated capsules.

Table 1. Successful encapsulation and microcapsule characterisation.

Coating Coating Microcapsule Diameter Encapsulate Active

Material Material (^ ^m ) n Efficiency Load (%)

Ratio (% (%) w/w)

Range Average cv Average Average

Alginate 100 2.41-53.60 10.95 1.12 71 ± 13 7 ± 2

Chitosan 100 2.32-57.54 7.75 1.32 36 ± 2 47 ± 4

Shellac 100 - - - - -

Pectin 100 - - - - -

Gelation 100 - - - - -

Alginate + 75:25 2.31-92.12 12.20 1.47 82 ± 10 11 ± 1

Chitosan 50:50 2.73-51.46 11.23 1.24 96 ± 3 14 ± 1

25:75 - - - - -

Alginate + 75:25 2.86-48.82 15.30 1.10 76 ± 5 12 ± 1

Pectin 50:50 2.1-39.9 6.34 0.98 98 ± 2 22 ± 1

25:75 - - - - -

Alginate + 75:25 - - - - -

Gelatin 50:50 - - - - -

25:75

Note: Active load is relative to the encapsulated glyphosate.

Glyphosate Release Studies from capsules The glyphosate release profile studies for the microcapsules are shown in figure 3 which were obtained at 25 °C and 100 % humidity. The release trigger for glyphosate from the capsules is via contact with water which initiates diffusion of encapsulated glyphosate through the capsule wall to the outer water reservoir. All the plots show a similar release profile whereby there is an initial release burst followed by a slower release which plateaus out which is characteristic for release by diffusion. Other interesting points noted from the release profile are as follows:

• First and foremost, the monitoring of the amount of glyphosate released in the first 2 h is crucial because this is the cumulative time duration required for the preparation, spraying of the formulation onto the crop fields and for the time required for the synergist to inhibit the weed's immune system, hence any glyphosate released before 2 h will most likely be resisted by the weed and hence will be lost. All the polysaccharide coated capsules show a release between 30-36 % the only exception being chitosan which releases 6 ± 1 % (table 2).

The release rate of glyphosate is the most rapid from the alginate: pectin (75:25) coated microcapsules and a release of 91 ± 6 % of the encapsulated glyphosate is observed after 24 h and hence is the most efficient in releasing glyphosate among the capsules formed (figure 3).

Table 2. Cumulative glyphosate release after 2 and 24 h.

Coating Coating Cumulative Glyphosate

Material Material Release (%)

Ratio

(w/w)

2h 24 h

Alginate 100 36 ± 5 64 ± 4

Chitosan 100 3 ± 1 24 ± 1

Alginate 50:50 32 ± 8 50 ± 2

+ 75:25 30 ± 2 81 ± 1

Chitosan

Alginate 50:50 33 ± 3 43 ± 3

+ Pectin 75:25 35 ± 4 91 ± 6

Glyphosate release as a function of temperature and pH

The release behaviour was investigated over pH range 5 -8 at 25 °C as well as pH 7 at 40 °C in 100% humidity. These conditions were specifically chosen as these are conditions that farmers are happy to work under.

The capsule walls are formed primarily through complexation between calcium ions and functional groups such as carboxylic acid. This intermolecular interaction can be influenced by change of pH. Hence, it was purposed that changing the pH would influence the rate of glyphosate release which may lead to more efficient glyphosate release. The data in Figure 4 shows the average release observed for each time interval. There are number of points to note: · The release is more rapid at 40 °C in comparison to 25 °C due to rate of diffusion being directly proportional to temperature.

• For the capsules walls containing alginate the rate of release is slower at pH 5 and increases with increase in pH.

• Whereas, chitosan coated capsules the rate of release is slower at pH 8 compared to 5 and 7 were no significant change is observed.

From the data obtained on ee, active load and glyphosate release profiles for the capsules formed, it was concluded that the alginate: pectin (75:25) coated capsules was the most promising among the capsules formed. Although, the ee, active loads and microcapsule sizes are similar for all the capsules formed with the mixed coatings the alginate: pectin (75:25) showed to be the most promising due to the capsules being the most efficient in releasing glyphosate (55 % released between 2 and 24 h). Thus, from this point onwards, it was decided upon further capsule characterisation and method optimisation would be performed on these capsules.

The effect of humidity on glyphosate release

The release behaviour of alginate: pectin (75:25) coated capsules under 35 % humidity was compared to 100 % humidity at 25 °C because 35 % humidity is lowest level of humidity that the capsules are likely to be exposed in the field.

To investigate the release behaviour of the capsules under 35 % humidity, a micropipette (1 μΐ) was used to replicate the droplets which are going to be formed on the surface of the crops after spraying the formulated herbicide mixture into the fields. The series of droplets on a petri dish were placed in a humidity chamber, where the temperature was maintained at 25 °C.

Figure 5 shows the glyphosate release under humidity of 35 and 100 % at 25 °C. There is no significant change in the glyphosate release behaviour under humidity of 35 % and 100 % for the first 8 h. However, after 8 h there is no further release under 35 % humidity in

comparison to 100 % humidity were release continues over 24 h with cumulative release amounting to 78 ± 7 % and 90 ± 6 %, respectively, over 24 h. The discrepancy in glyphosate release after 8 h could be explained as a result of the water evaporating off after this duration and as water is a stimulus for the release of glyphosate, there is no release observed when the stimuli is no longer present.

The effect of temperature on the glyphosate release.

A more comprehensive study on the effect of temperature on glyphosate release behaviour was performed in the temperature range 8-35 °C. This temperature range was specifically chosen because these temperatures are the most likely to be experienced in the countries that the product is going to be used in. There are several points to note from this experiment:

• Slower release rate at lower temperatures (figure 6). At the temperature of 8 °C

glyphosate release is shown to be much slower than at 35 °C. After 24 h, 70 % of encapsulated glyphosate is released whereas at 35 °C 94 % is released.

• Furthermore, at the higher temperatures due to quicker release more of the glyphosate is released in the first 2 h. At 8 °C, 32 % of the encapsulated glyphosate is released whereas, at 35 °C, 45 % is released. Hence, in the warmer climates more of the glyphosate will be released prior to weeds immune system being inhibited by piperonyl butoxide.

• This was expected as diffusion rate is directly proportional to temperature.

• Interestingly, glyphosate release profile between temperatures 35 and 40 °C is similar suggesting optimum release rate obtained at 35 °C.

UV/vis stability of the emulsion-gelation capsules Figure 7 shows the cumulative glyphosate release over 24 h at 25 °C in 100 % humidity. The plot shows no significant change between UV/vis exposed and unexposed microcapsules.

Therefore, this suggests that the capsules are stable under UV/vis exposure. Hence, the release behaviour will not be affected by the amount of light that the capsules are exposed to and there is no need for special requirements for storing the capsules so to prevent exposure to light. It is also expected that the other capsules will behave similarly due to the chemical makeup of the capsule walls being similar in all the capsules.

Optimization of the formulation for the encapsulation of glyphosate

To increase the performance of the encapsulated glyphosate capsules and reduce the cost of production a number of different formulations were investigated to elaborate whether the active load could be increased by reducing the amount of coating material used in the formulation without affecting the capsule properties. Table 3 shows the formulations and the ee observed for each formulation. Formulation 1 contains the quantities used in the original formulation.

Table 3. Different formulations investigated and the encapsulation efficiency from the formulation. In all formulations 100 ml of sunflower oil was used.

Formulat Glyphosat Coating Calcium Capsule CV ee (%) Active ion e solution material Chloride s size (%) load (%)

(g) (g) (g) (μ ^η ι)

1 5 25 1 15.3 1.10 76 ± 5 12 ± 2

2 10 25 1 19.6 1.25 39 ± 5 14 ± 2

3 5 12.5 0.5 17.8 1.17 60 ± 5 13 ± 2

4 5 6.25 0.25 - - - -

5 5 6.25 0.5 21.5 1.09 48 ± 5 11 ± 1

6 5 6.25 1 21.0 1.03 52 ± 5 13 ± 1

7 5 3.13 0.13 - - - -

8 5 3.13 0.25 - - - -

Note: CV abbreviation for Coefficient of Variance. From this investigation, it was discovered that the minimum quantity of coating material required to obtain sufficient amount of encapsulation was 6.25 g with 0.5 g of calcium chloride, however complete encapsulation was not observed. A layer of aqueous water was seen after the encapsulation process. This is reflected in the ee which is much smaller compared to formulation 1. Increasing the amount of CaCl ₂ from 0.5 g to 1.0 g had no effect on the ee. The capsule sizes (taking the error into consideration) are similar to the capsules formed from formulation 1. Interestingly, the active loads are very similar for the all the capsules, hence reducing the quantity of coating material has not improve active load but has had negative effect on ee.

The glyphosate release behaviour for capsules formed from formulation 2 and 3 in 100 % humidity is shown in figure 8. It can be observed that the glyphosate release from capsules formed from formulation 2 and 3 are faster in comparison to the capsules formed from formulation 1. The release after 2 h is observed as 47 ± 8 % and 58 ± 4 %, respectively, which is more than the 32 ± 4 % observed for the capsules formed from formulation 1. It is postulated that the faster glyphosate release could be as a result of a thinner capsule wall being formed from formulations 2 and 3 in comparison to formulation 1 due to smaller quantities of coating material being used. The effect on glyphosate release in the presence of piperonyl butoxide (PBO)

To determine:

(i) The stability of the glyphosate in the presence of PBO.

(ii) Whether PBO could be used as the continuous phase in place of sunflower oil, therefore removing the need for use of sunflower oil. This would reduce material and wakeup costs. To investigate the points above a number of experiments were performed. To evaluate the first point, a glyphosate release study was performed in a similar manner as before however this time 2 ml of PBO containing surfactant supplied by Pangaea was mixed with the capsules prior to exposure to water. The release profile obtained from this experiment is labelled Sunflower oil + PBO-S, sunflower referring to the continuous phase used in the preparation of the capsules and PBO-S referring to the addition of the PBO with surfactant which was mixed with the capsules prior to performing the release experiment. To evaluate point 2, encapsulation of glyphosate was carried out in PBO without surfactant (formulated PBO) and PBO with surfactant (formulated PBO-S) as the continuous phase in place of sunflower oil. This would lead to a readymade formulation which a farmer can buy straight off a shop shelf, then dilute with water and spray on to their crops. This will save the farmer time on preparative work and labour costs. Table 4 consists of ee, active load and size data of the capsules formed in PBO and PBO-S. There are several points to note:

• Capsules formed in PBO and PBO-S are of similar size to the capsules formed in sunflower oil when taking into consideration the CV.

• The ee is much smaller in PBO and PBO-S compared in sunflower oil over 20 % difference.

• The active loads observed in all continuous phases are similar.

• The glyphosate release rate is unaffected from the capsules formed in sunflower oil in the presence of PBO-S (figure 9).

• The glyphosate release is slower from the capsules prepared in PBO and PBO-S compared to capsules formed in sunflower oil as can be observed in Figure 9.

• However, glyphosate release profiles from capsules formed in PBO and PBO-S are similar and observe release of around 69 % after 24 h, hence the presence of surfactant does not affect the release.

• Only ~20 % of the encapsulated glyphosate is released in the first 2 h compared to 35 % which is observed from the capsules formed in sunflower oil.

From these findings, it can be concluded that the capsules are stable in the presence of PBO as the release profile are unaffected by the presence of PBO-S. Secondly, encapsulation can be performed in PBO and PBO-S, hence removing the need of sunflower oil and providing a product which is ready for the shop shelf, thus saving on labour and manufacturing costs. However, the cost of replacing sunflower oil with PBO and the poorer ee as well as slower release of glyphosate, need to be considered prior to deciding whether this option is economically more attractive. Table 4. Characterisation of microcapsules prepared with alginate: pectin (75:25) as coating material in different continuous phases.

Continuous Microcapsule Diameter Encapsulatio Active

Phase (μηι) n Efficiency Load (%)

(%)

Range Average CV Average Average

Sunflower oil 2.86-48.82 15.30 1.10 76 ± 5 12 ± 1

PBO 2.45-41.67 12.34 0.95 53 ± 3 12 ± 1

PBO-S 2.67-51.76 9.23 0.59 52 ± 2 13 ± 1

To improve the glyphosate release from the capsules formed in PBO, a release study was performed in 2 % w/w ammonium sulfate salt solution in place of water. Ammonium sulfate is a known adjuvant agent for glyphosate as it binds with calcium and iron ions present in water and plant cells. In this case ammonium sulfate would be employed as a scavenger to remove calcium ions present in the capsule wall, which would weaken the wall, leading to a quicker release. Figure 10 shows the presence of ammmonium sulfate increases the rate of glyphosate release and the release profile is similar to what was observed from the capsules prepared in sunflower oil. Hence, improving the viability of preparing the capsules in PBO.

Gelation with dications other than calcium ions.

Furthermore, the possibility of substituting the dication ion used for the gelation process from calcium to other dications (which have previously been used for such purposes) was explored. The ions which have been investigated were barium, copper and nickel ions. From table 5 it can be observed that at present capsules have only been successfully isolated using barium ions in place of calcium ions. Currently, for the other dications no encapsulation has been achieved. The encapsulation efficiency observed for barium ions gelated capsules was 56 ± 2 % which is lower than what was obtained for calcium ions gelated capsules (76 ± 5 %). The active load is very similar to the capsules which have been previously isolated. Table 5. The effect of substituting Ca ions with other dications for initiating the gelation process. In all formulations 4% aqueous solution of coating material, 9.01 mmol of the respective dication and 80 ml of sunflower oil was used.

Formulatio 67 % Coating Dication Capsule CV EE Activ n Glyphosate material ion s size (%) (%) e load salt solution (g) (g) (μηι)

10 5 25 Ba ⁺ 20.6 1.06 56 ± 2 12 ± 2

11 5 25 Cu ²⁺ - - - -

12 5 25 Ni ²⁺ - - - -

Note: CV abbreviation for Coefficient of Variance.

Summary

There are a number of experimental findings from the project, these are listed below:

• By taking into consideration the specifications of the project that were given at the beginning of the project, the emulsion gelation method was seen as an attractive process to fabricate the capsules due to the procedure being simple, easy, quick and most importantly cheap to set up.

There were a number of coatings which could have been used. However, due to the project requirements such as the coating material should be cheap to purchase, readily available, environmental friendly (biodegradable), compactable with glyphosate, easy to handle (non-hazardous) and preferably the material should already be widely used as a coating material so that future licensing is easily obtained for the product. So we decided to focus on polysaccharides as these cover all the points discussed.

From the polysaccharides that were studied it was discovered that a combination of two materials to develop the capsule wall is better than a wall consisting of a single polymer because these capsules had high ee and good active loads.

• All the capsules formed were within the size limit (sub- 100 μπι). Alginate: pectin (75:25) coated capsules showed to be the most promising due the capsules being most efficient in releasing glyphosate. Although, alginate hitosan (75:25) coated capsules were also promising, however, chitosan is more expensive to purchase that pectin therefore increasing cost of production and hence, is less commercially attractive compared to the alginate :pectin combination.

The alginate: pectin (75:25) coated capsules are stable to UV/vis and PBO-S exposure.

Humidity does not affect the release rate form the capsules.

As expected the glyphosate release is temperature sensitive. In warmer climate the release is more rapid, however release remains similar when temperature is increased from 35 to 40 °C.

Glyphosate release is also pH sensitive. For the capsules which contain alginate in its wall, the release rate was shown to be slower at pH 5 which increased with increasing pH.

Encapsulation is also achievable using less amount of coating material than 25 g, however reducing the amount of coating material led to increase rate of glyphosate release.

Encapsulation of glyphosate also achievable in PBO and PBO-S in place of sunflower oil. Similar capsule sizes were formed, however poorer ee were observed. Also the glyphosate release is slower from the capsules formed from PBO and PBO-S compared to the capsules formed in sunflower oil. The release can be enhanced by adding ammonium sulfate in the formulation.

Encapsulation using dications other than calcium ions gave mixed results.

Encapsulation was achieved when barium ions was used as gelling agent, however encapsulation was observed when copper and nickel ions were used. However, water soluble barium compounds are known to be toxic and therefore using barium ions is less environmentally attractive than calcium ions.

All optional and preferred features and modifications of the described embodiments and dependent claims are usable in all aspects of the invention taught herein. Furthermore, the individual features of the dependent claims, as well as all optional and preferred features and modifications of the described embodiments are combinable and interchangeable with one another.

The disclosures in UK patent application number 1710736.8, from which this application claims priority, and in the abstract accompanying this application are incorporated herein by reference.