general enquiries on this form should be made...

General enquiries on this form should be made to:Defra, Science Directorate, Management Support and Finance Team,Telephone No. 020 7238 1612E-mail: [email protected]

SID 5

Research Project Final Report

SID 5 (Rev. 3/06) of 28

NoteIn line with the Freedom of Information Act 2000, Defra aims to place the results of its completed research projects in the public domain wherever possible. The SID 5 (Research Project Final Report) is designed to capture the information on the results and outputs of Defra-funded research in a format that is easily publishable through the Defra website. A SID 5 must be completed for all projects.

This form is in Word format and the boxes may be expanded or reduced, as appropriate.

ACCESS TO INFORMATIONThe information collected on this form will be stored electronically and may be sent to any part of Defra, or to individual researchers or organisations outside Defra for the purposes of reviewing the project. Defra may also disclose the information to any outside organisation acting as an agent authorised by Defra to process final research reports on its behalf. Defra intends to publish this form on its website, unless there are strong reasons not to, which fully comply with exemptions under the Environmental Information Regulations or the Freedom of Information Act 2000.Defra may be required to release information, including personal data and commercial information, on request under the Environmental Information Regulations or the Freedom of Information Act 2000. However, Defra will not permit any unwarranted breach of confidentiality or act in contravention of its obligations under the Data Protection Act 1998. Defra or its appointed agents may use the name, address or other details on your form to contact you in connection with occasional customer research aimed at improving the processes through which Defra works with its contractors.

Project identification

1. Defra Project code NF0615

2. Project title

New Polymer Plasticisers & Stabilisers from UK-Grown Crambe

3. Contractororganisation(s)

The BioComposites Centre, BCUniversity of Wales, BangorDeiniol roadLL57 2UW, BangorWales Aston UniversityPolymer Processing and Performance Research UnitSchool of Engineering and Applied ScienceAston TriangleBirmingham B4 7ET

Hydro Polymers LtdSchool Aycliffe LaneCounty DurhamDL5 6EA

Warwick International LtdMostynHolywellCH8 9HE

Springdale Crop Synergies LtdRudstonDriffieldEast YorkshireYO25 4DJ

54. Total Defra project costs £ 199,385(agreed fixed price)

5. Project: start date................ 01 July 2005

end date................. 30 September 2006

SID 5 (Rev. 3/06) of 28

6. It is Defra’s intention to publish this form. Please confirm your agreement to do so...................................................................................YES NO (a) When preparing SID 5s contractors should bear in mind that Defra intends that they be made public. They

should be written in a clear and concise manner and represent a full account of the research project which someone not closely associated with the project can follow.Defra recognises that in a small minority of cases there may be information, such as intellectual property or commercially confidential data, used in or generated by the research project, which should not be disclosed. In these cases, such information should be detailed in a separate annex (not to be published) so that the SID 5 can be placed in the public domain. Where it is impossible to complete the Final Report without including references to any sensitive or confidential data, the information should be included and section (b) completed. NB: only in exceptional circumstances will Defra expect contractors to give a "No" answer.In all cases, reasons for withholding information must be fully in line with exemptions under the Environmental Information Regulations or the Freedom of Information Act 2000.

(b) If you have answered NO, please explain why the Final report should not be released into public domain

SID 5 (Rev. 3/06) of 28

Executive Summary7. The executive summary must not exceed 2 sides in total of A4 and should be understandable to the

intelligent non-scientist. It should cover the main objectives, methods and findings of the research, together with any other significant events and options for new work.In Western Europe approximately 1,300,000 tonnes of plasticisers are used each year, of which 900,000 tonnes are used to plasticise PVC (polyvinyl chloride). The UK market comprises about 200,000 tonnes/year (plasticisers) and 40,000 tonnes/year (stabilisers), worth a total of £850 million/year. Currently, phthalates comprise 93% of all plasticisers used worldwide, with around 800,000 tonnes being consumed annually in Western Europe alone.

The plasticiser industry has become increasingly concerned with the use of phthalates over the recent years amid greater regulatory pressure concerning their toxicity, mainly evidence that phthalates cause endocrine disruption by reducing testosterone production, hence fertility. Furthermore, the REACH (Registration, Evaluation and Authorisation of Chemicals) framework, will replace some of the EU chemicals’ legislation, create a European Chemicals Agency by 2007 and produce a list of banned chemicals in the European space. In that respect, the European Union directive means that where there is an uncertainty concerning toxicity, a product will be banned as a precautionary principle.

Into this context, the discovery and development of new plasticiser candidates ‘free of phthalates’ but with comparable or enhanced properties has grown considerably. Accordingly, some new plasticisers have already been developed worldwide. Trimellitates, citrates, adipates and more recently a new sustainable plasticiser based on castor oil, called Grindsted Soft-n-safe have been proposed as alternative plasticisers. Nonetheless, most of the alternatives don’t attain the high performance that phthalates offer and, when this is the case as it happens with Grindsted Soft-n-safe, it is three times more expensive than current commercial offerings.

Especially attractive would be those candidates derived from renewable sources. In particular, diesters of brassylic acid have been demonstrated to have performance properties comparable to dioctyl phthalate (DOP) in the plastisation of PVC, but without the toxicity and the risks to human reproduction attributed to the phthalates. These molecules have not yet found widespread use largely due to the cost and commercial availability of brassylic acid. Therefore the development of a new cost effective methodology in order to obtain brassylic acid from renewable sources would be crucial. Into this context falls the UK-produced agricultural crop called Crambe abyssinica, which has a high erucic acid content averaging 56-58% of the total oil in the seed. This erucic acid could be derivatised using clean and effective technology in order to obtain brassylic acid. As a result of that, the main objectives of this project were:

Objective 1: To prepare and characterise four new products (two stabilisers and two plasticisers) based on derivatives of crambe oil (by conversion of erucic acid into brassylic acid) using modern “green” chemistry and cost effective commercial methods.

Objective 2: To develop polymer formulations for use with the newly synthesised additives required for polymer processing based on modification of currently used formulations involving commercial products having the same additive function.

Objective 3: To conduct a technological evaluation and appraisal of the new additives in well defined melt processed polymer formulations, in order to generate critical, industry recognised, technical data and information on the new stabilisers and plasticisers.

Objective 4: To transfer the laboratory/pilot scale synthetic procedures used for additive production to the industrial scale.

Objective 5: To transfer the plasticiser and stabiliser additives to industrial end users, initially through industrial partners, but also through other interested groups, including academia and public institutes.

The present interdisciplinary project has been accomplished in collaboration between industry and research centres. In particular, Springdale Crop Synergies as the main producer of crambe oil and supplier of erucic acid, Hydro Polymers as a leading company in the plasticiser industry and an end user of the polymer additives, Warwick International as an additives manufacturer expert in ozonisation and esterification technology, Aston’s Polymer Processing and Performance Laboratory (PPP) research centre as an expert in polymer formulation and technical evaluation and The BioComposites Centre (BC) as an established institute for the development of renewable materials.

The objectives outlined above have been divided into six well defined interlinked work approaches: the synthesis of additives, formulation and processing of fabricated polymer-additive systems, technological

SID 5 (Rev. 3/06) of 28

evaluation and appraisal of performance and efficiency of the additives in polymers, optimisation and scale-up production, dissemination, exploitation and implementation of results and management and co-ordination of project.

The synthesis of the four additives was completed between the BioComposites Centre and Warwick international. The main component of Crambe is erucic acid (55-60%), a monounsaturated omega-9 fatty acid known as cis-13-docosenoic acid, which oil is produced and supplied by Springdale Crop Synergies. Ozonolysis (a clean and environmentally friendly method that only uses oxygen and electricity) of erucic acid, followed by oxidation with molecular oxygen yielded two main products: nonanoic-also known as pelargonic acid-and brassylic acid, which is a C13-α,ω-dicarboxylic acid. Subsequent esterification of brassylic acid with various alcohols (cyclohexanol, brassylic alcohol, piperidinol and irganox alcohol) produced the corresponding four ester additives using an effective and relatively clean methodology, which uses very small amounts of organic solvent. An added value to the process is that the waste generated during the process is kept to a minimum as the by-product of the reaction (pelargonic acid) also finds many different useful applications. It is mainly utilised as a raw material for the production of sodium nonanyl oxybenzenesulphonate (SNOBS), which is employed as a bleach activator in detergents in the USA, and also has a smaller use in synthetic lubricants.

As the synthesis of each one of the four additives was being completed, they were transferred to the other partners (plasticisers to Hydropolymers, stabilisers to Aston) in order to develop formulations required for polymer processing.

The brassylate-based antioxidants and plasticisers were used in different polymer formulations and tested against commercial controls selected as performance benchmarks. Whereas, Aston University developed the formulations of the stabilisers in polypropylene (PP) according to standard industrial procedures; Hydropolymers developed the formulations for the two plasticisers in PVC, namely PLB and PLOB, plus an initial evaluation of their potential as alternatives to DEHP, one of the market’s most effective plasticisers. Initial evaluation of PLOB would suggest that the degree of compatibility with PVC is too low for it to be considered as a possible alternative plasticiser to DEHP for flexible PVC. On the other hand, initial evaluation of PLB shows it to behave in a very similar manner to DEHP as a plasticiser for flexible PVC, therefore making it a very attractive alternative. In its present form as a liquid it would pose some handling difficulties in large scale manufacturing processes, adding an extra cost to the process. As a result of these findings, there are a number of issues that need to be addressed in future projects: complete the weathering evaluation of PLB, further investigate the behaviour of PLB systems at different concentration levels, further investigate the behaviour of PLB during processing and fabrication and evaluate different derivatives which are either completely solid or liquid at ambient temperature. The formulated samples were sent to Aston for further evaluation.

Aston University evaluated the performance of the new stabilisers. A commercial grade polypropylene was used for testing the performance of both the newly synthesised UV and thermal stabilisers and the benchmark commercial stabilisers. The brassylate-based hindered amine light stabiliser HABS and hindered phenol antioxidant HPB as thermal stabiliser, separately and together, were compared against the commercial benchmark stabilisers Tinuvin 770, Irganox 1010 and Irganox 1076. Extensive tests were carried out and the main results have been outlined in the paragraphs below.

Thermogravimetric analysis (TGA) demonstrates that HABS is more stable and less volatile than the control Tin 770 under both inert and oxidising environments.

Under isothermal conditions, at 210ºC, the rate of volatilisation of HPB in neat form is twice that of Irg1010 under inert and oxidising atmospheres. Clearly both stabilisers suffer from mass losses through volatilisation and product transformation even at 150ºC. These losses, nonetheless, should be considered as extreme cases and treated with caution and not in isolation from field results under real test conditions and materials.

Both melt flow index MFI and capillary rheometry give information on the rheological behavior of polymers and changes in their molecular structure and properties due to degradation upon processing. MFI values for multi- and single- pass extrusion of PP formulations reveal that both benchmark and the new brassylate-based stabiliser formulations show similar trends. The extent of discoloration observed with the brassylate-based light stabiliser HABS (formulation PPP3) is comparable to that of the control Tin 770. Moreover, HPB is not an appropriate melt stabiliser at the low concentrations used but it becomes more effective at higher concentrations (0.1% w/w) as can be seen from its shear viscosity-time profile compared to the benchmark formulation at the same concentration. However, it is also clear that HABS does not contribute to polymer stabilisation as Tin770 does at the concentration level used, hence the need to further work to optimising the levels of the brassylate-based stabilisers for cost effective end-use performance.

SID 5 (Rev. 3/06) of 28

Thermal ageing of polymer films and plaques shows that HPB offers greater stability than Irg1010; this advantage diminishes both with higher temperatures and with the number of extrusion passes. Furthermore, it is clear from color measurements that the brassylate-based thermal stabiliser, HPB, offers overall greater color stability to the formulation compared to the benchmark stabiliser Irg1010.Accelerated weathering tests reveal that HABS is far more effective in the stabilisation of PP films, both in the presence and absence of the acid scavenger (calcium stearate); its superiority in the absence of the acid scavenger is very interesting. Additional UV exposure tests reveal that HABS is more effective than Tin770 as a light stabiliser.

Aston University also evaluated the performance of the new plasticisers. Suspension grade polyvinylchloride (sPVC), Norvinyl S7066 (Hydro) was used for testing the performance of the newly synthesised plasticisers against commercial control materials. The commercial plasticiser di-(2-ethylhexyl) phthalate (DEHP) was used as a control for testing the performance of the newly synthesised plasticisers.

The available results suggest that PLB contributes more to thermal stability of PVC than to melt stability. Both DEHP and PLB appear to have similar effects on the melt stability of the dryblend formulations especially at the high plasticiser levels. Whereas, the loss of DEHP and PLB from plasticised PVC formulations (at 32.4% w/w) under isothermal TGA at 150ºC for both is less than one percent per hour at 150ºC under both inert and oxidising environments, it is clear that DEHP is slightly more extractable by both heptane and ethanol compared to PLB; PLOB is less extractable in both solvents.

The effect of plasticisers on modulus and glass transition temperature (Tg) using dynamic mechanical analysis (DMA) tests provides a measure of the performance efficiency and flexibility for low temperature applications. These tests revealed that DEHP is slightly more efficient than PLB and that PLOB is the least efficient. However, higher plasticiser levels lead to lowering of the Tg relative to the unplasticised polymer and reduced modulus: unlike the case for DEHP, where the plot deviates from linearity at higher concentrations, lower levels of pure PLB would be needed to achieve comparable flexibility and low temperature performance so as to make it cost-effective.

Warwick and the BioComposites collaborated to build a scale-up production plant with the purpose to transfer benchscale technology to industrial production. Although this study is based on the ozonolysis of oleic acid, the technology developed can be transferred for the ozonolysis of erucic acid, and virtually to any other unsaturated fatty acid. Whilst the synthesis is easily achieved at laboratory bench scale, manufacturing in commercial quantities was recognised to present some unusual difficulties, not least because an intermediate product in the synthesis is unstable with the potential to decompose rapidly, yielding heat and permanent gases, and therefore posing significant challenges for the design of reaction equipment. These issues, along with a financial study, have been addressed. Based on a manufacturing capacity of 3000 tonnes/year of azelaic acid, the operating costs of the process (£8.998 M/yr) would outweigh the sales revenue (£ 8.400 M/yr) in the first year of full production sales in the case of azelaic. On the other hand, considering that at least one of the brassylate additives goes into the market demand for the production of brassylic would easily outweigh the operating costs, making it a profitable business.

BC took charge of the overall management by means of constant communication with the other partners either by telephone or email and organising periodic meetings every three months in order to evaluate the progress of the project and future actions.

To conclude, it is essential to remark the many benefits and the impact that these results will have, not only for DEFRA, but also for industry and the consumers as a whole. It benefits DEFRA because the development of these additives can increase crambe production as a non-food crop, providing diversification opportunities for farmers and expanding biodiversity in the countryside. It is also highly beneficial for industry due to various reasons. Firstly, it opens new markets for crambe as a renewable source of brassylic acid, which will impact positively on the agricultural sector in terms of new hectarage of industrial crops grown. Secondly, the use of ozonisation technology will allow the competitive production of new vegetable oil derivatives, thus revitalising the UK chemical industry. And thirdly, the availability in the market of substitutes for phthalates such as the ones developed by this project will allow the plastics industry to comply with rigid legislation by presenting bio-derived and phthalate-free additives. Finally, consumers will be the ultimate beneficiary as the market will offer the option to purchase phthalate-free goods based on more environmentally friendly production methods.

Project Report to Defra8. As a guide this report should be no longer than 20 sides of A4. This report is to provide Defra with

details of the outputs of the research project for internal purposes; to meet the terms of the contract; and to allow Defra to publish details of the outputs to meet Environmental Information Regulation or

SID 5 (Rev. 3/06) of 28

Freedom of Information obligations. This short report to Defra does not preclude contractors from also seeking to publish a full, formal scientific report/paper in an appropriate scientific or other journal/publication. Indeed, Defra actively encourages such publications as part of the contract terms. The report to Defra should include: the scientific objectives as set out in the contract; the extent to which the objectives set out in the contract have been met; details of methods used and the results obtained, including statistical analysis (if appropriate); a discussion of the results and their reliability; the main implications of the findings; possible future work; and any action resulting from the research (e.g. IP, Knowledge Transfer).

1. Introduction

In Western Europe approximately 1,300,000 tonnes of plasticisers are used each year, of which 900,000 tonnes are used to plasticise PVC (polyvinyl chloride). The UK market comprises about 200,000 tonnes/year

SID 5 (Rev. 3/06) of 28

(plasticisers) and 40,000 tonnes/year (stabilisers), worth a total of £850 million/year. Currently, phthalates comprise 93% of all plasticisers used worldwide, with around 800,000 tonnes being consumed annually in Western Europe alone.

The plasticiser industry has become increasingly concerned with the use of phthalates over the recent years amid greater regulatory pressure concerning their toxicity, mainly evidence that phthalates cause endocrine disruption by reducing testosterone production, hence fertility. Furthermore, the REACH (Registration, Evaluation and Authorisation of Chemicals) framework, will replace some of the EU chemicals legislation, create a European Chemicals Agency by 2007 and produce a list of banned chemicals in the European space. In that respect, the European Union directive means that where there is an uncertainty concerning toxicity, a product will be banned as a precautionary principle.

Into this context, the discovery and development of new plasticiser candidates free of phthalates but with comparable and/or enhanced properties has grown considerably in the last few years. In this direction, some new plasticisers have been developed worldwide. Trimellitates, citrates, adipates and more recently a new sustainable plasticiser based on castor oil, called Grindsted Soft-n-safe, have been proposed as alternative plasticisers. Nonetheless, most of them don’t attain the high performance that phthalates offer and, when this is the case as with Grindsted Soft-n-safe, it is in the order of three times more expensive than current commercial offerings.

As a general rule, a good plasticiser must be flexible but without resulting in migration due to volatilisation, solution or biodegradation. Especially attractive would be those candidates deriving from renewable sources. In particular, diesters of brassylic acid have been demonstrated to have performance properties comparable to dioctyl phthalate (DOP) in the plasticisation of PVC, but without the toxicity and the risks to human reproduction attributed to the phthalates. These molecules have not yet found widespread use largely due to the cost and commercial availability of brassylic acid. Therefore the development of a new cost effective methodology in order to obtain brassylic acid from renewable sources would be crucial. Into this context falls the UK-produced agricultural crop called Crambe abyssinica, which has a high erucic acid content averaging 56-58% of the total oil in the seed. This erucic acid could be derivatised using clean and effective technology in order to obtain brassylic acid. As a result of that, the main objectives of this project were:

Objective 1: To prepare and characterise new polymer additives (two stabilisers and two plasticisers, Figure 1) based on derivatives of crambe oil using modern “green” chemistry and cost effective commercial methods.

Objective 2: To develop polymer formulations for use with the newly synthesised additives required for polymer processing based on modification of currently used formulations involving commercial products having the same additive function.

Objective 3: To conduct a technological evaluation and appraisal of the new additives in well defined melt processed polymer formulations in order to generate critical, industry recognised, technical data and information on the new stabilisers and plasticisers as to their characteristics and performance in specified polymer formulations.

Objective 4: To transfer the laboratory/pilot scale synthetic procedures used for additive production to the industrial scale.

Objective 5: To transfer the plasticiser and stabiliser additives to industrial end users, initially through industrial partners, but also through other interested groups, including academia and public institutes.

The present interdisciplinary project has been accomplished in collaboration between industry and research centres. In particular, Springdale Crop Synergies as the main producer of crambe oil and supplier of erucic acid, Hydro Polymers as a leading company in the plasticiser industry, Warwick International as an additives manufacturer expert in ozonisation and esterification technology, Aston’s Polymer Processing and Performance Laboratory (PPP) research centre as an expert in polymer formulation and technical evaluation and The BioComposites Centre (BC) as an established centre for the development of renewable materials.

SID 5 (Rev. 3/06) of 28

Figure 1. Stabilisers and plasticisers derived from Crambe oil.

2. Production of erucic acid and its implications

At present, the majority of plant derived erucic acid comes from the high erucic acid rape (HEAR) varieties. However, increased production of HEAR may lead to risk of cross pollination, which could alter the oil profiles, thus be inadequate for their intended applications. Agricultural crops may provide a significant percent of the raw materials for industry and therefore researchers and breeders are continually looking at crops that may be of benefit to industry. One such crop is Crambe abyssinica, which is family cruciferae and therefore does not cross with commercial oil seed rape (OSR) varieties.

Crambe, which is believed to be of Mediterranean origin, was introduced in to the US in the 1940’s and became commercially grown in the 1990’s. Nonetheless, it has been grown in many other places such as tropical and subtropical Africa, the Near East, Central and West Asia, Europe, the United States and South America.

One of the advantages of Crambe is that it has a high erucic acid content averaging 56-58% of the total oil in the seed, approximately 9-10% higher than HEAR.

2.1. Agronomy

In mainland UK the crop is sown in April to May when there is less risk of severe frosts. The crop emerges within five to seven days in warm soil conditions and is harvested approximately 100-120 days from emergence, making it a very fast growing crop. The highest yields of Crambe are on sandy loams and soils with a pH of 6.0, with nitrogen applications of 150 kg per hectare.

The crop can be harvested by direct cutting. Post harvest handling of the seed is straightforward for transportation and drying. The crop is an excellent break crop in the UK agricultural crop rotation and cereals grown after crambe often show yield benefits. Replacing OSR with crambe in the rotation can alleviate weed, pest and disease build up.

SID 5 (Rev. 3/06) of 28

2.2. Processing

The seed is cleaned of impurities (down to a maximum level of 2%) and dried to 9% moisture so that it can be stored without risk of deterioration. After extrusion of the oil, the resulting cake (oil content 10 to 12%) is solvent extracted with hexane to remove more of the oil, leaving a meal with approximate oil content of 1 to 2%.

The crude vegetable oil is then sent for processing to purify erucic acid from the oil. The unrefined vegetable oil goes through a reactor called a splitter. Heat and pressure are used to split the triglycerides in to glycerine and crude fatty acids. The crude fatty acids are then purified via fractional distillation. The fractionation goes through two columns the first column separates the erucic acid and high boiling point impurities out and then the second column separates the erucic acid from the impurities. Overall, approximately 95% of the erucic acid in the original crude oil is recovered.

2.3. Present Production

Crambe is grown primarily for the extraction of erucic acid from the oil. The erucic acid content of the oil is 58%. Erucic acid is predominantly converted into erucamide and used as a non-stick agent in polyolefin films. Table 1 shows some of the main applications of erucic acid derivatives.

Crambe has been commercially cultivated in the US since 1990 and has proven to be a viable crop. In the UK commercial production of crambe started in 2001. At the present time up to 4000 hectares of crambe are grown in the UK, with production also in Australia, South Africa and have trials in Chile, France, Zambia, India, NZ, Italy, Holland, Sweden and Saudi Arabia, in order to establish a year round supply of oil for industry.

Table 1. Major uses of erucic acid and its derivatives.

Derivative Application

Erucamide Polymer additive

Behenyl fumarate vinyl copolymer Oil field chemical

Stearyl erucamide Polymer additive

Behenyl trimethylammonium chloride Personal care product

Brassidolide Perfumery

Glyceryl trierucate Pharmaceutical

Erucyl erucate Cosmetics

Nylon 1313 Apparel

Croda presently take the majority of oil produced in the UK. They have highlighted the benefits of erucic acid extraction from crambe oil compared with HEAR oil. The higher level of erucic acid in the oil tends to give a purer erucamide sample the low level of polyunsaturated fatty acids simplifies downstream processing. For these reasons crambe oil tends to be preferred for erucamide production. The lower levels of production are due to crambe been lower yielding in some cases than HEAR and farmers and advisors being less familiar with the crop. New varieties are being trialled for yield improvements. Also the increased demand for crambe oil for specialist applications will lead to an increase in price to the farmer.

In summary, good agronomic practices have been established for crambe and it has already proven to be a commercially viable crop in the US and UK. Crambe oil has benefits over HEAR oil when further processed and these will be of particular importance when producing stabilisers and plasticisers. Trial work is already in progress to increase the economic viability of this oil and to create a stable, year round supply for industry. The full report on crambe production can be found in Annex A.

SID 5 (Rev. 3/06) of 28

3. Synthesis of brassylic acid from the ozonolysis of erucic acid

The main component of crambe oil is erucic acid (55-60%), a monounsaturated omega-9 fatty acid known as cis-13-docosenoic acid. The synthesis of brassylic acid derived from the ozonolysis of erucic acid is a crucial step because it represents the starting point from which stems the production of all the additives proposed in this project. Although the ozonolysis of erucic acid has been reported previously, this report describes a clean and cost effective methodology, which uses simply oxygen and electricity. The procedure was carried out in one reaction pot, subdivided in three very distinctive steps: ozonisation, decomposition of the ozonides and oxidation (Scheme 1). The full experimental procedures corresponding to the ozonolysis and the synthesis of all additives can be found in Annex B.

Scheme 1. Ozonolysis of erucic acid.

The ozonisation step was developed and scaled-up to 600g of erucic acid of starting material. Erucic acid of 87% purity was ozonised at 40ºC, using nonanoic acid as the co-solvent. In fact, the use of nonanoic acid offers three main advantages in comparison to already existing methods: firstly, it reduces the viscosity of the reaction mixture, thus allowing more effective diffusion of ozone; secondly, it simplifies the work-up of the reaction as nonanoic acid is also formed as a by-product of the ozonolysis/oxidation; and thirdly, nonanoic acid can be isolated from the reaction mixture and recycled or used as a commercial product in its own.

The extent of the reaction was followed by study of both, 1H and 13C NMR spectroscopy, at different reacting times. The 13C NMR spectra clearly show how the peak corresponding to the double bond present in erucic acid at δ130.0 disappears progressively, notably after 4h. Additional follow-up of the reaction using the 1H NMR spectra shows the formation of ozonides, which are identified at δ(6.15-6.39) and δ5.20 respectively. Moreover, the double bond signal of erucic acid, at δ5.35, becomes absent after five hours, indicating full consumption of ozone.

Data from NMR studies were also complemented with the calculation of the rate of ozone being consumed during the reaction. This was accomplished by titrating of the rate of ozone not being consumed in the outlet pipe, using a potassium iodide buffer solution. These results, summarised in Table 2 and Figure 2, corroborate that an optimised time for the ozonisation of 600g of erucic acid should be between five to six hours at 9.7 mmol/min rate of ozone and 40ºC.

Table 2. Rate of the ozonisation of erucic acid.

Time(min)

O3 not consumed (mmol/min)

O3 consumed(mmol/min)

O3 consumption(%)

0 9.70 0.00 0.0

10 0.39 9.31 96.0

60 0.44 9.26 95.5

120 0.78 8.92 92.0

180 1.67 8.03 82.8

240 3.50 6.20 63.9

300 6.89 2.81 29.0

360 9.52 0.18 1.9

SID 5 (Rev. 3/06) of 28

Rate of ozonisation of Erucic acid

0.0%

25.0%

50.0%

75.0%

100.0%

0 50 100 150 200 250 300 350 400

Time (min)

Perc

enta

ge o

f ozo

ne

com

sum

ptio

n

Figure 2. Rate of the ozonisation of erucic acid.

The ozonisation step was followed by decomposition of the intermediate ozonides. Due to the high risk of uncontrolled reaction, decomposition was accomplished by alternate heating/cooling. It was commenced at 75ºC, from where the temperature was raised slowly until 110ºC over a period of an hour, when all ozonides had been destroyed. Subsequent oxidation using molecular oxygen at 110ºC was then carried out in order to convert any aldehyde by-products to their corresponding acid derivatives.

Separation of the main reaction products was achieved by distillation of nonanoic acid under reduced pressure (0.025 mmHg, 82ºC). The resulting residue contained about 5% of nonanoic acid as indicated per NMR analysis. Further crystallisation from hot toluene afforded highly pure brassylic acid in a very good 63.3% yield. Determination of its melting point (110-111)ºC was consistent with literature values. Additional analysis of the 13C NMR spectrum showed the carbonyl corresponding to the acid moiety at 185.8, which was also observed as a broad band at 3020 cm-1 in the IR spectrum.

4. Manufacture of brassylic acid from renewable feed stocks using ozonolysis design study: optimisation and scale-up

A pilot scale-up production plant was designed in order to transfer benchscale technology to industrial production. Although this is based on the ozonolysis of oleic acid, the technology developed can be transferred for the ozonolysis of erucic acid.

Whilst the synthesis is easily achieved at laboratory bench scale, manufacturing in commercial quantities was recognised to present some unusual difficulties, not least because an intermediate product in the synthesis is unstable with the potential to decompose rapidly, yielding heat and permanent gases, and therefore posing significant challenges for the design of reaction equipment. These issues, along with a financial study, are explored in the following summary. For a full report, see Annex C.

4.1. Process route (chemistry, separation and safety analysis)

The overall chemistry can be split into three process steps: ozonolysis, thermal decomposition of polymeric peroxides and oxidation.

Analyses of the general conditions indicate that process temperatures range between ambient and about 250ºC. Whilst the ozonolysis reaction takes place at between 20 and 50ºC, the oxidation reaction takes place at 100 to 120ºC. Both, ozonolysis and oxidation stages take place at pressures above atmospheric, expected to be between 0 and 0.25 bar g.

SID 5 (Rev. 3/06) of 28

One of the main considerations when upscaling the ozonisation process is that the viscosity of the reaction mixture increases as the ozonide is formed. In order to reduce the viscosity of the reaction mixture, a substantial quantity of solvent (nonanoic acid) is required.

At the end of the reaction, brassylic and pelargonic acids can readily be separated by distillation, where pelargonic acid of the required purity can be obtained by distillation alone.

Technical as well as safety issues had also to be considered. Since the reaction between ozone and erucic acid is fast and exothermic, the reaction system was designed with the aim to be able to remove heat at a greater rate than the maximum feasible rate of generation. Besides, the reaction can be stopped immediately once the supply of ozone is removed.

Oxygen will act as an accelerant for an existing fire, and hence bulk oxygen should be stored remotely from plant buildings. Since the reactions are between gas and liquid phases in both the ozonolysis and oxidation steps, there is very limited scope for accumulation of unreacted materials. However, the intermediate ‘ozonide’ (or polymeric peroxide) is known to be unstable, and to decompose exothermically. For this reason the process should be designed such that there is no intermediate storage of ‘ozonide’, and the inventory of ‘ozonide’ is designed to be the minimum feasible.

Information relating to the chemicals being used indicates that most of the liquid and solid materials handled are classified as irritants and some components as harmful. Whilst the majority of the compounds involved in the process are found in nature, and used in foodstuffs, none of the solid or liquid compounds identified are known to be toxic.

4.2. Process development (reactor design and separation processes)

The semi-technical scale reaction unit was designed to provide confirmation that the reaction equipment planned for a full-scale plant would perform as expected. It was intended that the reaction stage in the semitechnical scale unit could be modified easily and that –if necessary– widely different reactor styles could be tested. The semi-technical scale unit was instrumented to a similar extent to the intended full-scale plant, and fitted with PC-based data logging, so that critical parameters of the plant operation could be monitored and compared against the expected 'as-designed' behaviour. Accordingly, a number of potential ozonolysis reactor designs were considered:

Stirred Tank and Continuous Stirred Tank reactors: not appropriate due to high inventory, poor heat transfer.

Bubble column reactors: potentially viable, but may require internal cooling surface and/or CFD, computational fluid dynamics.

Loop reactors: potentially viable and scale-up feasible by replicating hydraulic conditions in the reaction loop.

4.3. Full Scale Production

Market research carried out during the initial stages of the development process indicated that a manufacturing capacity of 3000 tonnes/year of azelaic acid, together with 2500 tonnes/year of pelargonic acid, would be realistically achievable.

Financially, the total capital cost was estimated at 14,465 £k and the total operating cost at 8,998 £k, which would outweigh the sales revenue (£8.400 M/yr) in the first year of full production sales for azelaic acid, based in the production of 3000 tonnes/year of azelaic acid. In the case of brassylic acid, considering just one of the brassylic-derived polymer additives is marketed, it is foreseen that the demand would outweigh any operational costs.

SID 5 (Rev. 3/06) of 28

5. Synthesis of plasticisers

The production of all plasticiser additives was completed following a general standard procedure, which involves condensation of brassylic acid with an alcohol substrate in the presence of an acid catalyst, p-toluene sulfonic acid (PTSA.H2O). The main advantage of this process is that it requires the use of a very small amount of solvent (toluene) and a little amount of catalyst, hence minimising the amount of waste. A small amount of toluene is used in order to drive the equilibrium towards the formation of product by actively removing the water being formed during the course of the reaction by azeotropic distillation using a Dean-Stark apparatus.

The synthesis of dicyclohexyl brassylate was carried out by condensation of the brassylic acid with cyclohexanol using p-toluene sulfonic acid (0.1 eq.) as the catalyst.

Scheme 2. Synthesis of PLB.

Analysis of the final product was confirmed by 1H NMR spectroscopy, which shows the protons of the cyclohexyl group that are next to the ester group at δ4.59, and the protons adjacent to the carbonyl group at δ2.09, as predicted. The 13C NMR spectrum reveals the carbonyl of the ester group at δ173.1 and the carbon next to the carbonyl at δ72.2, as expected. High resolution mass analysis confirmed the product with an m/z 409.3315, as required.

The second plasticiser, oligomeric brassylate named PLOB, was chosen in order to reduce migration of the plasticiser by increasing its molecular weight but, at the same time, keeping the same characteristics as for analogous plasticisers (i.e. similar ratio of polar groups within the molecule).

The brassylic alcohol intermediate needed for condensation was produced in quantitative yields by reduction of brassylic acid with lithium aluminium hydride in THF solution. The use of brassylic alcohol to produce the oligomeric additive adds an advantage to the whole process because it can be derived directly from already synthesised brassylic acid, minimising the number synthetic steps, thus minimising the overall cost.

The synthesis of PLOB was completed in a similar way to dicyclohexyl brassylate (see above, synthesis of PLB). In order to reduce the number of oligomeric products of higher molecular weight, one equivalent of brassylic alcohol was condensed with two equivalents each of brassylic acid and cyclohexanol, hence terminating the condensation chain at an early stage. Once the reaction reached its equilibrium, the work-up afforded a white solid, which was found to be a mixture of three main oligomeric components of different molecular weights: dicyclohexyl brassylate (PLB, monomer), trimer (PLOB) and pentamer, as illustrated in Figure 3.

Figure 3. Mixture of oligomers derived from brassylic acid.

Characterisation of the mixture by MALDI analysis confirmed the presence of the three oligomers at m/z 431.41 (rel. int., 1.0), 855.76 (0.8) and 1280.08 (0.1), respectively.

Successful condensation of brassylic acid with cyclohexanol and brassylic alcohol broadens the prospects for the production of a whole range of new additives using a wider choice of other alcohol precursors. It is envisaged that this could form the basis for the development of other research projects exploring the possibility of improving existing additives’ properties.

SID 5 (Rev. 3/06) of 28

6. Synthesis of stabilisers

The production of the stabiliser additives was completed using similar esterification techniques as for the plasticisers described above.

Tinuvin 770 is a hindered amine light stabiliser (HALS) with a ten carbon backbone derived from sebacic acid and is one of the most commercially successful stabilisers in the polymer industry. A simple modification of its structure, using brassylic instead of sebacic acid, afforded the hindered amine brassylate stabiliser (HABS), which was synthesised from the condensation of brassylic acid and piperidinol in the presence of p-toluene sulfonic acid as the catalyst (Scheme 3). The basic nature of piperidinol, conferred by its amine moiety, means that it can form a complex with PTSA.H2O, therefore sequestering the catalyst and inhibiting the reaction. This impediment was overcome by modifying the synthetic procedure with the use of three equivalents of catalyst.

Scheme 3. Synthesis of HABS.

The second stabiliser additive, the hindered phenol brassylate (HPB), was synthesised from the condensation of brassylic acid with the irganox alcohol (Scheme 6).

Brassylic acid can be obtained from the ozonolysis of erucic acid, as already described previously in this report. The irganox alcohol, however, was not available from chemical catalogues and it had to be derivatised from another commercially available irganox ester precursor. In the first instance, direct reduction of the octadecyl irganox ester precursor with lithium aluminium hydride afforded a mixture of both, irganox alcohol and octadecyl alcohol (Scheme 4).

Scheme 4. Reduction of the octadecyl irganox ester.

However, difficult separation of these two alcohols by either crystallisation (similar melting points) or column chromatography (almost identical retention times), prompted us to use an alternative route into the production of the irganox alcohol.

The new route involved the synthesis in two separate steps. Firstly, saponification of the octadecyl irganox ester precursor gave the irganox acid derivative in nearly quantitative yields. Subsequent reduction of the irganox acid using lithium aluminium hydride, afforded the irganox alcohol with an overall 91.6% yield (Scheme 5).

SID 5 (Rev. 3/06) of 28

Scheme 5. Synthesis of Irganox alcohol.

The irganox alcohol was condensed with brassylic acid in the presence of PTSA.H2O to give the irganox ester stabiliser (Scheme 6).

Scheme 6. Synthesis of HPB.

Purification of the crude product by column chromatography eluting with petrol/diethyl ether 9:1 afforded pure HPB in 88.5% yield. Analysis by IR spectroscopy identified the phenolic group at 3544 cm-1 and the ester moiety at 1734 cm-1, as expected. Analysis by 1H NMR spectroscopy shows the aromatic protons at δ6.99 as a singlet and the phenolic protons at δ5.08, also as a singlet, as predicted. Further analysis by high resolution mass spectrometry confirmed the product formation at m/z 736.5483, as required.

7. Formulation of plasticiser additives and initial evaluation

7.1. Objectives

1. To take the raw materials generated and turn them into working compounds for further evaluation by the project partners.

2. To evaluate the raw materials generated in the project as possible plasticisers for flexible PVC (Poly Vinyl Chloride).

7.2. Sample preparation

The raw materials as received from source were either, liquid, solid or a semi mixture of solids/liquid. A PVC powder blend needed to be produced. This was achieved by mixing the raw materials together in a coffee grinder for 3 minutes (when an homogenous powder was formed).

SID 5 (Rev. 3/06) of 28

7.3. Formulation

The following formulations were prepared: initial sample of dicyclohexyl brassylate (PLB) compared to diethylhexyl phthalate (DEHP) in a simple clear PVC formulation of nominal hardness 75 shore A (Table 3).

Table 3. Formulation for PLB.

PVC Polymer (K Value 70) 100 phr

Plasticiser 50 phr

Ca/Zn stabiliser 1.5 phr

Epoxidised Soya bean oil 3.0 phr

This was subsequently repeated with the addition of a second brassylate derivative (PLOB). A proportion was kept for analysis by Hydro Polymers, whilst the remainder was sent to Aston University for further investigations.

A second series of samples, which were sent to Aston University for further investigation and study, were produced at different concentration levels (Table 4).

Table 4. PVC-plasticiser formulations.

DEHP-PVC samples (controls) A B C D E F

PVC resin 300 100 50 50 50 300

DEHP 0 9.1 9.25 17.4 34.9 0

ESBO 0 3 1.5 1.5 1.5 9

Ca/Zn stabiliser 4.5 1.5 0.75 0.75 0.75 4.5

Amounts shown in phr (wt %) 0.0 8.0 15.0 25.0 40.0 0.0

Brassylate-PVC samples AA BB CC DD EE FF

PVC resin 300 100 50 50 50 300

Brassylate plasticiser 0 9.1 9.25 17.4 34.9 0

ESBO 4.5 1.5 0.75 0.75 0.75 4.5

Ca/Zn stabiliser 0.0 8.0 15.0 25.0 40.0 0.0

Amounts shown in phr (wt %) 0.0 8.0 15.0 25.0 40.0 0.0

7.4. Plasticiser testing

The following tests, in accordance with British Standard test methodology, were conducted on the initial blends of PLB and DEHP (Table 5). Subsequent further testing carried out on blends containing DEHP, PLB, and PLOB (Table 6).

The results summarised in Table 5 and Table 6 represent an initial evaluation of two potential PVC plasticisers as alternatives to DEHP, namely PLB and PLOB. These indicate that PLB is similar in performance to DEHP at this particular concentration level. On the other hand, PLOB proved to be somehow less efficient and would require higher addition levels to obtain the same performance. Besides, PLOB also showed evidence of post fabrication exudation and this would preclude its use as a PVC plasticiser.

SID 5 (Rev. 3/06) of 28

Table 5. Testing carried out on initial blends of PLB and DEHP.

Series 1 DEHP PLB

Specific gravity 1.24 1.23

Tensile strength and elongation 21.66 20.5

Elongation 300 340

Shore A hardness 79 78

Heat stability (200ºC) 15 15

British standard softness (BSS) 38 39

Table 6. Testing carried out on blends containing DEHP, PLB, and PLOB.

Series 2 DEHP PLB PLOB

Specific gravity 1.231 1.231 1.220

BSS (7 days, 30 sec.) 42 38 26

Shore A (7 days, 15 sec.) 76 79 85

Tensile strength (500 mm/min) 18.23 18.74 18.05

Elongation 380 330 335

Static heat stability (Congo Red) 50 40 40

Dynamic heat stability (Haake) 47 12 17

Cold flex (ºC) -23.5 -17.5 -15.5

Volume resistivity (ohm/cm) 12.76 13.10 12.85

Heat loss (%) at 105ºC 0.110 0.169 0.147

Migration (polycarbonate) None None None

Hereaus weathering (4000 h) RW05374 RW05375 RW05376

Additional comments Colorlesstransparent

Very yellowtransparent

Dull yellow/orangeOpacity/exudation

over time

SID 5 (Rev. 3/06) of 28

Concentrating on PLB, then this displayed good colour and clarity in the initial sample but the second sample was more yellow and less clear. For it to be considered as an alternative, the quality of the original sample would need to be achieved. PLB also has a melting point close to ambient temperature. This has implications for material handling. At present the common plasticisers are liquids capable of being pumped around. There are a few solid plasticisers and this in itself does not cause too many problems. With PLB is almost paste like in its consistency and does pose handling issues. One solution would be to store and transport in heated vessels and containers. Again this is not unknown in the PVC additive arena but does add extra cost.

In conclusion, evaluation of PLB shows it to behave in a similar manner to DEHP as a plasticiser for flexible PVC. In its present form it would pose handling difficulties in large scale manufacturing processes. On the other hand, evaluation of PLOB would suggest that the degree of compatibility with PVC is too low for it to be considered as a possible alternative plasticiser to DEHP for flexible PVC.

As a result of all of this initial evaluation, there are a number of issues that need to be developed further:

1. Complete the weathering evaluation of PLB2. Further investigate the behaviour of PLB systems at different concentration levels.3. Further investigate the behaviour of PLB during processing and fabrication.4. Evaluate different derivatives which are either completely solid or liquid at ambient temperature.

8. Formulation, processing and optimisation of the production of fabricated polymer-additive systems

8.1. Polymer formulations and polymer melt processing

Two polymers were selected for this work. For formulations containing the newly synthesised brassylate-based UV and thermal stabilisers and the benchmark commercial stabilisers, an extrusion grade polypropylene, PP (Moplen HF500N, ex. Basell, with a melt flow rate of 12 g/10min and melt volume flow rate of 16 cm3/10min (ISO 1133) was used. For formulations containing the new brassylate-based and commercial plasticisers, a suspension grade polyvinylchloride, sPVC, (Norvinyl S7066, ex. Hydro, having the following characteristics: K-value of 69.5, intrinsic viscosity 121, Mw/Mn 2.768, Mw 106,100, bulk density 0.460 g/cm3, plasticiser absorption capacity 0.360g/g) recommended for films and flexible PVC applications was used. Table 7 shows structures for the newly synthesised additives and the commercial stabilisers and plasticisers used in the project. Table 8 and Table 9 show the different formulations developed for the stabilisation of PP and the plasticisation of PVC and are based on typical industrial formulations.

The melt processing of PP containing stabiliser samples was conducted in a Twin-Screw intermeshing co-rotating extruder, TSE, (24mm diameter (D), L/D = 28/1 and 7 heating zones; PTW24, Thermoprism/Haake) with real-time monitoring of extrusion parameters (melt temperature and torque) using Haake software. The TSE used is a small version of typical industrial extruders used for PP processing. The extrudate was water-cooled and pelletised on-line. Two sets of processing experiments were conducted with PP (single pass extrusion, die temperature 210ºC, screw speed 250 rpm) and multi-pass extrusions (die temperature 260ºC, and screw speed 100 rpm) using pre-blended polymer-additive feed, see Table 8. Formulations of PVC containing plasticiser samples (Table 9) were initially dry blended (this step was done by Hydro) and then melt processed in an internal mixer (Brabender Plasticorder W50) at 170ºC, 60rpm for 3min using 50g charge with on-line monitoring of processing parameters, melt temperature and torque, using Pico software.

SID 5 (Rev. 3/06) of 28

Table 7. Structures of additives (stabilisers and plasticisers) used to develop polymer formulations.

Newly synthesised stabilisers Newly synthesised plasticisers

Name Structure Name Structure

HABS PLB

HPB PLOB Oligomer of PLB

Commercial stabilisers Commercial plasticisers

Name Structure Name Structure

Tinuvin

770DEHP

Irganox

1076

Ca/Zn Stearate

Soap

(in ESBO),

Irganox

1010ESBO

ORH

RH

Irgafos 168 --- ---

Irganox B215 Irganox 1010 + Irganox 168 (1:2) --- ---

Table 8. PP polymer formulations containing newly synthesised and commercial stabilisers.

SampleCode

Sample composition (wt%)/TSEsingle pass/210ºC/250rpm

SampleCode

Sample composition (wt%)/TSEsingle pass/210ºC/100rpm, contain 0.1w% CaSt

CaSt B215 Tin770 HABS Irg1010 Irg1076 Irgf168 Tin770 HPB HABS

PP00† 0.0 0.0 0.0 0.0 PPP1 0.05 0.0 0.10 0.0 0.0 0.0

PP0‡ 0.0 0.0 0.0 0.0 PPP2 0.05 0.0 0.10 0.15 0.0 0.0

PP1 0.1 0.0 0.0 0.0 PPP3 0.05 0.0 0.10 0.0 0.0 0.15

PP2 0.1 0.1 0.0 0.0 PPP4 0.0 0.0 0.10 0.0 0.05 0.15

PP3 0.1 0.1 0.1 0.0 PPP5 0.0 0.005 0.10 0.0 0.0 0.0

PP4 0.1 0.1 0.0 0.1

PP5 0.0 0.1 0.1 0.0SampleCode

Sample composition (wt%)/TSEmulti pass/260ºC/100rpm, contain 0.1 w% CaStPP6 0.0 0.1 0.0 0.1

† Virgin sample.

‡ Processed samples.

Irg1010 Irgf168 HPB

PP3T 0.1 0.2 0.0

PP4T 0.0 0.2 0.1

SID 5 (Rev. 3/06) of 28

Table 9. PVC polymer formulations containing newly synthesised and commercial plasticisers.

DEHP (phthalate) A B C D E F G

PVC resin 300 100 50 50 50 300 100

DEHP or PLB 0 9.1 9.25 17.4 34.9 0 50

ESBO 0 3 1.5 1.5 1.5 9 3

Ca/Zn stabiliser 4.5 1.5 0.75 0.75 0.75 4.5 1.5

DEHP or PLB (% w/w) 0.0 8.0 15.0 25.0 40.0 0.0 32.4

8.2. Characterisation of rheological properties and processing performance of polymer formulations

The rheological properties of PP and PVC that reflect their processing characteristics were measured. For PVC, the torque and temperature profiles monitored during compounding in the Plasticorder were monitored and gelation time for formulations containing different concentrations of synthesised (PLB) and commercial (DEHP) plasticisers (Figure 4). It is clear from this that the processing behaviour of synthesised PLB is very similar to that of the commercial plasticiser analogue. In both cases, increasing plasticiser concentration leads to lowering of torque values and reduced time to fusion of PVC aggregates (time to gelation), hence lower process energy demand for plasticised formulations containing 32.4% plasticiser are used predominantly for evaluation. At this plasticiser level, the evidence suggests that both DEHP and PLB appear to have similar effects on the melt stability of the dry-blend PVC formulations as reflected by similar colours of their processed and compression moulded plaques shown in Figure 5. For PP, degradation test at low shear rate conducted by capillary rheometric measurements of formulations containing the synthesised brassylate-hindered phenol stabiliser HPB and that of the commercial benchmark analogue Irg1010 (compared at the same concentrations) showed that HPB is as effective as the commercial benchmark stabiliser system (Figure 6). Melt rheological stability may also be assessed by following changes in the polymer colour after each pass in a multi-pass extrusion process, and indeed both HPB and Irg1010 formulations gave similar extent of colour upon multi-pass extrusion confirming HPB a good melt and thermal stabiliser.

Figure 4. Torque-time profiles and time to gelation of PVC formulations with different plasticiser concentrations in w%.

Figure 5. Colour plaques of plasticised PVC.

SID 5 (Rev. 3/06) of 28

Figure 6. PP Degradation Test at low.

The processing performance of the PP-stabiliser formulations was further examined by dynamic mechanical analysis, DMA, which determines, amongst other parameters, the polymer glass transition temperature (Tg). Changes in Tg of polymers are used for monitoring differences arising from polymer degradation and processing effects. It is clear from Figure 7, which shows the effect of composition of PP stabiliser formulations on the dynamic mechanical properties of the polymer extrudates determined over temperature range between -50 to and +160ºC, that the Tg, which is also related to the molar mass of the polymer, remains almost unaffected for all the formulations examined when compared to the behaviour of the virgin polymer (see Table 8 for compositions) indicating negligible changes to molecular weights of the polymer during melt extrusion.

Temp °C150.0100.050.00.0-50.0

tanD

0.1400

0.1200

0.1000

0.0800

0.0600

0.0400

0.0200

PP 2

PP 1 PP 3

PP 5

DMA for PP Stabiliser Formulations

PP 6

PP 4 PP 00

PP 0

Tg

Figure 7. Effect of PP formulations on Tg of DMA plaques (bend mode).

8.3. Preparation of different test samples for various performance and characterisation tests

Extruded stabilised PP samples and melt processed plasticised PVC samples obtained from all formulations given in Table 8 and Table 9 were compression moulded into thin films or plaques using a heated hydraulic press (Turton & Bradley) at 190ºC. Thin films samples of about 25μm thickness were prepared for spectral and thermal characterisation, solubility and diffusion studies, weathering and oven ageing experiments. Plaque samples of about 3mm thickness were prepared for colour and thermal ageing, dynamic mechanical tests, diffusion by ATR-FTIR, and for solubility tests by the stack method.

9. Technological evaluation and appraisal of performance and efficiency of the additives in polymers

9.1. Characterisation and appraisal of the properties and performance of the new stabilisers and plasticisers

The newly synthesised brassylate-based thermal and UV-stabilisers and plasticisers were characterised by different thermal-analytical and spectroscopic methods in order to test and assess their properties and performance against the commercial benchmarks. Spectroscopic analysis using FTIR and NMR was used to confirm the identity and purity of the synthesised additives (Annex B). The thermal behaviour of the new additives, polymers and polymer formulations was assessed using three techniques; thermogravimetric analysis (TGA) to follow the effects of temperature on changes in mass, differential scanning calorimetry (DSC) for assessing the effect of temperature on energy, and dynamic mechanical analysis (DMA) to follow changes in

SID 5 (Rev. 3/06) of 28

performance properties under the combined effects of temperature and mechanical stress. These tests give important information on end-product characteristics and short and long term performance of these additives in the polymers.

Figure 8 shows clearly from the dynamic and isothermal TGA that the brassylate UV-stabiliser HABS is both more thermally stable (peak decomposition temperature of 396ºC compared to 312ºC for Tin770) and less volatile (isothermal TGA) than the commercial UV-stabiliser analogue Tin770. Comparison of the thermal behaviour of the brassylate thermal stabiliser HPB with that of the commercial antioxidant Irg1010, reveals that both stabilisers suffer losses through volatilisation and product transformation even at 150ºC (isothermal TGA), but that HPB behaves differently (upon decomposition) to both Irg1076 and Irg1010 in that it shows a three-step decomposition profile (dynamic TGA) which may be attributable to transformation to higher molecular fragments, or it may be that the synthesised HPB is not very pure and contains higher molecular weight components. It is also clear that the rate of volatilisation of HPB is almost twice that of Irg1010 under both inert and oxidising atmospheres (isothermal TGA). The volatilisation behaviour of the synthesised brassylate-based plasticisers is also given in Figure 8 which shows clearly (from isothermal TGA run at 160ºC) that all three neat brassylate plasticisers, PLB and PLOB are much less volatile than the phthalate DEHP which is lost much more rapidly. Moreover, and as would be expected, the % weight loss of the oligomeric plasticiser PLOB is lower than that of its monomeric analogue, PLB (almost half as volatile; from isothermal TGA). However, it is also clear from dynamic TGA of PLOB (Figure 8), a method that can give a clear separation in a sample of different components based on molecular masses, that the sample is a mixture of the monomer and oligomers comprising three oligomeric components of different molecular masses in addition to the monomeric component PLB.

Figure 8. Isothermal and dynamic TGA analysis of different synthesised and commercial stabilisers & plasticisers.

9.2. Evaluation of melt, long term thermal, and UV (weathering) stabilising efficiency of the new stabilisers in PP and appraisal of their performance by comparison with commercial products based on similar antioxidant functions

The melt stabilising efficiency of the synthesised brassylate stabilisers in extruded PP formulations was examined and their performance appraised against commercial stabiliser analogues using polymer formulations subjected to multi-pass extrusions (to exaggerate the effect of processing-an approach typically used by industry when developing new melt stabilisers) and assessed by both melt flow index, MFI (measured according to ASTM standard D 1238-95) and TGA measurements. Figure 9 shows that the brassylate-based thermal (HPB) and UV (HABS) stabiliser formulations, compounded in single and multi-pass extruded PP samples, show high efficiency of melt stabilisation comparable to the commercial controls. This is corroborated by results from TGA for the multi-pass extruded polymers which shows changes in the decomposition temperature (onset temp. from derivative TG curve) to be of similar magnitude in both the brassylate HBP and the commercial Irg1010 stabiliser formulations examined during multi-pass extrusions. A similar trend to the MFI is observed in the melting temperature (measured by DSC) for the brassylate HABS UV-stabiliser to that of the commercial UV-stabiliser Tin770 in that both cause a slight lowering of the polymer melting point (compared to a similar sample but without the light stabiliser). Further evidence of the excellent melt stabilising efficiency of the new brassylate thermal stabilisers HPB was given earlier in Figure 6 above.

SID 5 (Rev. 3/06) of 28

Figure 9. Melt stabilising efficiency of brassylate-based stabilisers in PP compared to commercial benchmarks.

The long-term thermal stability, LTTS, (performance in-service at high temperature) of extruded PP films and plaques containing the brassylate and commercial stabilisers was assessed by both ageing in a multi-cell Wallace oven (standard BS ISO 188) and by measurement of the oxidation induction period, OIT, by DSC (ASTM D3895); a method used generally by industry to quickly assess the LTTS of stabilisation packages. The superior long term thermal stabilisation efficiency offered by HPB compared to the benchmark Irg1010 is clearly illustrated from DSC-OIT measurements particularly at 160ºC; this apparent advantage diminishes both with higher temperatures of OIT testing and with the number of extrusion passes (Figure 10). This excellent LTTS efficiency of the brassylate HPB is further confirmed from colour measurements of the polymer samples following thermal ageing in air oven at 140ºC where the HPB stabiliser offers overall greater colour stability to the polymer formulation compared to the benchmark stabiliser Irg1010, see Figure 10.

Figure 10. Long term thermal stabilising efficiency of the brassylate stabilisers compared to benchmark additives.

The light stabilising efficiency of the new brassylate UV-stabiliser HABS in PP has been assessed by exposure to accelerated weathering conditions in both the industry-standard Xenon arc Weather-Ometer ATLAS device (black panel temperature 60ºC and irradiance measurements of 0.5 W/m2 at 340nm) and in SEPAP 12/24 UV-ageing device (medium pressure quartz mercury lamps with cabinet air temp of 60ºC). Figure 11 shows clearly that the newly synthesised brassylate UV-stabiliser HABS offer very high UV-stabilisation to PP films under accelerated photo-ageing conditions of both the weather-Ometer and SEPAP and is superior to that of the commercial analogue Tin 770.

SID 5 (Rev. 3/06) of 28

Figure 11. Photostabilising efficiency of a brassylate stabiliser system compared to benchmark stabilisers.

9.3. Evaluation of the permanence (solubility and diffusion) of new and benchmark stabilisers in PP

Solubility and diffusion characteristics of stabilisers are important factors in determining permanence and, ultimately, their performance in end-use products. The solubility and diffusion coefficient of the newly synthesised stabilisers were measured and compared to those of the commercial analogues. The solubility was determined at 60ºC using a solubility stack assembly made up of three layers of thin additive-free films (ca. 50μm) sandwiched between polymer plaques saturated with the stabiliser. Diffusion of the stabilisers was determined at 80ºC using FTIR coupled with a horizontal attenuated total reflection (HATR) accessory modified for use with thin films under an inert atmosphere, and operated at elevated temperatures. Figure 12 and Table 10 show the concentration-time profile of additives during monitoring of the diffusion in PP films and solubility data, respectively. It is clear from this that diffusion coefficient and solubility of the brassylate- and commercial stabilisers are very similar.

Figure 12. Diffusion coefficient and their concentration-time profiles derived from time-base FTIR-HATR.

Table 10. Solubility and diffusion characteristics of stabilisers.

StabiliserSolubility at 60ºC

(% w/w)Solubility (% w/w)

Literature1

Diffusion x 1010 at 80ºC, Cm2/s

HPB 1.5 --- 0.29

HABS 2.0 --- 3.20

Tin 770 1.0 1.04 (70ºC) 3.90

Irg 1010 0.5 0.40 (70ºC) 0.49 (Lit 2)

Irg 1076 8.0 7.5 (70ºC) ---1. N C Billingham, in “Plastics Additives Handbook”, H Zweifel, 5th edn, Hanser, Ch. 20 (2001).2. Ferrara, et.al., Polymer Deg & Stab, 73, 411-416 (2001).

9.4. Evaluation of the plasticisation efficiency and permanence of brassylate- plasticisers in PVC and appraisal of their performance by comparison to phthalate-based commercial system

The processing stability of PVC-plasticiser formulations was assessed under objective 2.1 (see also Figure 4 and Figure 5). The effect of the brassylate plasticisers and their performance efficiency was assessed by examining changes in the modulus and glass transition temperature (Tg) of formulated PVC using DMA.

Figure 13 shows the DMA characteristics (Tg, storage, E’, and loss, E” modulii; measured in bend mode, 1Hz, 2ºC/min) for plasticised PVC formulations. It is clear that higher plasticiser levels lead to lowering of the Tg and storage modulus (not shown) relative to the unplasticised polymer. It is also clear from this figure that plasticised (at 32.4% loading) samples show a shift to lower temperatures in their Tg values relative to unplasticised PVC

SID 5 (Rev. 3/06) of 28

(Tg ca. 80ºC); DEHP show the lowest Tg suggesting slightly higher efficiency than PLB and PLOB. However, it is important to point out here that when a high purity sample (white) of PLB was tested, it was found to give a lower Tg than DEHP at 32.4% loading (see green circuled point in Figure 13). The larger the shift (to lower temperatures) in Tg the greater is the efficiency of the plasticiser in terms of its flexibility and performance particularly in low temperature applications. Thus, based on DMA measurements, it is clear that at a typical loading of 32.4%, a well purified brassylate-PLB plasticiser shows higher plasticisation efficiency in PVC than the phthalate DEHP benchmark.

Figure 13. Plasticisation efficiency of plasticisers in PVC determined from DMA measurements.

The extent of loss by volatilisation of plasticisers from PVC was examined by assessing their weight loss from isothermal TGA measured at 150ºC. It is clear from Figure 14 that in the polymer DEHP is slightly less volatile than PLB (the not highly purified sample). This apparently lower extent of volatility of the phthalate compared to the brassylate plasticiser when tested in PVC contrasts their behaviour as neat compounds whereby DEHP was shown to be lost by volatilisation much more rapidly compared to the brassylates (Figure 8). The reason for this is that in the neat form, the brassylates are solids (albeit PLB and PLOB are low melting solids, 15-20ºC) whereas DEHP is a liquid that would have high partial pressure under the conditions of the TGA experiment under nitrogen flow. In polymer formulation at 32.4%, on the other hand, the observed lower volatilisation of DEHP (Figure 14) is probably due to enhanced inter-molecular attraction forces between the phthalate groups and PVC compared to the brassylate groups, albeit the difference in volatilisation from the film samples tested are only small. As for extraction in organic solvents (H: heptane and E: ethanol), the loss of PLB is slightly less than that of DEHP and the oligomer brassylate PLOB is substantially more resistant to extraction. The diffusion coefficient of PLB was found to be very similar to that of DEHP (5.1 X 10-9 cm2/s) and that PLB is somewhat more soluble in PVC (9%w/w for PLB vs. 7%w/w for DEHP at 60ºC).

Figure 14. Volatility, diffusion and extraction in heptane (H) and ethanol (E) of plasticisers.

Overall, therefore, the work carried out by the AU-PPP partner, was successfully completed and shows clearly that the newly synthesised brassylate-based thermal and UV stabilisers, HPB and HABS, are highly efficient in their performance as melt, thermal and UV stabilisers in polypropylene, both in terms of their antioxidant activity and physical permanence, and compare favourably to commercial controls of similar antioxidant function used as benchmarks in this work.

Similarly, the brassylate-based plasticisers, PLB and PLOB, have been shown to possess good processing characteristics in PVC with effective plasticisation efficiency and retention in the polymer comparable to the commercial phthalate control used as a benchmark. It should be emphasised here that further work will be needed to establish performance of products manufactured in the presence of the new stabilisers and plasticisers on a large industrial scale.

SID 5 (Rev. 3/06) of 28

10. Dissemination of results

It is our intention to publish the outcomes of this research in due course. Currently, publication is on hold until a full review of the IP situation is completed. We are currently discussing further funding opportunities with respect to both the plasticiser and stabiliser candidates that have been developed. Upon completion of these discussions, data which can enter the public domain without jeopardising any patent application will be published via the appropriate channels, e.g. J. Polym. Sci.; JACS.

SID 5 (Rev. 3/06) of 28

References to published material9. This section should be used to record links (hypertext links where possible) or references to other

published material generated by, or relating to this project.

SID 5 (Rev. 3/06) of 28

general enquiries on this form should be made...

Documents