Lipid composition of natural rubber sheet and relationship with its structure and properties

(1)

WORLD ELASTOMER MARKET

1. Present situation of natural rubber

In year 2007, total production of natural rubber over the world stood at about 9.73 million tons all over the world (IRSG, 2008). Thanks to certain

unequalled properties, natural rubber

(NR) keeps its share in world elastomer market with around 40% for more than 15 years.

Presently, around 93% of the world production comes from Asia as shown in the figure, the major producers being Thailand (32.5%), Indonesia (27.3%), Malaysia (13.3%) and India (8.8%).

For Thailand, the exported NR products are block rubber (37% of exported products and mainly STR20), rubber sheets (33%, mainly RSS3) and concentrated latex (30%, mainly HA and LATZ) (Thai Ministry of Commerce, 2008).

Lipid Composition of Natural Rubber Sheet and Relationship with its Structure and Properties

Centre de coopération Internationale en Recherche Agronomique pour le

Développement Kasetsart University

Siriluck Liengprayoon

1, 2

_{, Frédéric Bonfils}

3

_{, Jérôme Sainte-Beuve}

3

_{, Klanarong Sriroth}

2

_{, Eric Dubreucq}

1

_{, Laurent Vaysse}

4 1-Montpellier SupAgro, UMR IATE, Montpellier, France, 2 -Kasetsart University, Faculty of Agro-Industry, Bangkok, Thailand, 3 -CIRAD, UMR IATE, Montpellier, France, 4 -CIRAD, UMR IATE, Bangkok, Thailand

2. natural rubber structure and properties

Variability of natural rubber properties such as initial plasticity (P0) or

plasticity retention index (PRI) are indeed not appreciated by second transformation industry which use more and more automated processes.

The samplings were carried out in the plantation of Visahakit Thai Rubber Co.,LTD in Chantaburi province, Thailand. Four popular rubber clones were chosen: RRIM600, GT1, BPM24 and PB235. Unsmoked sheet (USS) which was proved to be similar as ribbed smoked sheet (RSS) in terms of properties (Rodphakdeekul et. al, 2008) was used as dry rubber model. Lipids from fresh latex and dry rubber were extracted after methodological optimization (Liengprayoon et. al, 2008). The analyses performed for each sample are presented in the scheme below.

5. Lipid composition of natural rubber

is more complicated than the one of its synthetic counterpart (Bonfils and Vaysse, 2003). Apart from its 100% cis-configuration polyisoprene, natural rubber contains 3-5% of non-isoprene component, a part that could not be mimicked by chemical synthesis. This non-rubber fraction has been suspected to be responsible for both superior

properties and inconsistent quality of natural rubber.

Lipids have been described to play roles on some rubber properties in both raw and vulcanized states. Unfortunately, obtained results are hardly comparable due to the difference of rubber sources, method of extraction and analytical techniques. Only one comprehensive work dealing with the eventual relationships between lipid composition and the lack of consistency of natural rubber properties has been published (Hasma, 1984).

Our research question was :

“IS LIPID COMPOSITION RELATED TO THE LACK OF CONSISTENCY OF NATURAL RUBBER PROPERTIES?”

3. Research question

4. Methodology

From various chromatographic analyses, precise structural information could be obtained. Furan fatty acid (FFA) clearly distinguished PB235 from the other clones as this atypical fatty acid represented more than 70% of its total fatty acid content, while linoleic acid (C18:2) was the main fatty acid in the other clones. Among the detected compounds in

unsaponifiable fraction, β-sitosterol was the major sterol in every clone. Moreover, the presence of Δ5-avenasterol which was

previously claimed to be fucosterol was proved in this study. The structural analysis of fresh latex polar lipids using

HPLC/ESI-MS showed that both FFA and Δ5-avenasterol were more present in neutral lipids and glycolipids than in

phospholipids for every clone even in PB235.

In parallel with lipid analysis, USS samples were characterized for their structures and properties in order to identify possible correlations.

 Poly 1,4-cis isoprene Non – isoprene components

•

• 6. Relationship between lipid composition,

6. Relationship between lipid composition,

structures and properties of NR

The variability of the collected data from 15 samplings (April 2004 – December 2006) i.e. lipid composition, natural

rubber structure (molar mass distribution and gel content) and properties (P0,PRI, breakdown and vulcanization behavior)

was represented using principle component analysis (PCA) as shown in figure below.

From further pair correlation determined by analysis of variance with an observed significance probability less than

0.05 (P≤ 0.05), lipids was found to correlate with various natural rubber structures and properties. Some examples are

illustrated below.

Monoclonal fresh latex

Monoclonal sheet rubber Specification (ISO 2004) DRC, TSC Specification (ISO 2004) P0, PRI, ML, N Mesostructure Gel, MMD, Mn, Mw Mastication BI Vulcanization ts2, tc(90), ML, MH Extraction Lipid extract

Thin Layer Chromatography (TLC) Global qualitative composition Colorimetric spectrophotometry (R6G)

Free fatty acids (% w/w) Saponification

Gas chromatography-mass spectrometry (GC-MS) Total fatty acid composition (% w/w)

Liquid Column Chromatography (LC) Neutral lipids (% w/w)

Glycolipids (% w/w) Phospholipids (% w/w) High performance liquid chromatography-MS (HPLC-MS)

Glycolipid and phospholipid structure Neutral and unsaponifiable composition (% w/w)

General analysis scheme of structure, properties and lipid composition of natural rubber

Negative correlation between P0and total

lipids, neutral lipids and fatty acids (a) especially esterified form of furan fatty acid which indicated a plasticizing effect of lipids. This effect resulted in the drop of consumed energy during mastication assessed by Breakdown Index method (b).

Negative correlation between PRI and fatty acids especially unsaturated fatty acids indicated the

activity of such fatty acids as prooxidant (c).

α-tocotrienol and stigmasterol were found to act as antioxidant in every studied clone (d).

FROM FRESH LATEX TO DRY RUBBER

: Lipids z : Proteins z : Carbohydrates z : Others z z z z z z z z z z

hydrophilic component loss ex. proteins, carbohydrates : Cleaning Rubber Processing Rubber Processing Dry rubber Dry rubber z z z zzzzz z z z z z z Fresh Latex Fresh Latex Dry Rubber Dry Rubber z z zz z z z z z z zz z z z z zz z z z z z z Major non-isoprene component is lipids

Lipids from fresh latex from four H. brasiliensis clones were characterized and found to be clonal dependent (2.5 – 3.7% w/w dry rubber). PB235 clone was the highest lipid containing clone followed by BPM24, RRIM600 and GT1. The amount of lipids in USS was found to be lower (2.0 – 3.3%) than in latex but the ranking remained the same. Comparison of lipid classes between fresh latex and USS revealed that the quantity of polar lipids in USS decreased by around 50% for glycolipids and 80% for phospholipids from their initial quantity in latex. This change could be due to enzyme-catalyzed hydrolysis as lipase activity was observed in fresh latex. However, the clonal specificity of lipid profile observed in fresh latex remained unchanged in the corresponding sheet rubber.

Natural products

Petroleum based products Synthetic Rubber 13,583,000 mt 58.3% Natural Rubber 9,725,000 mt 41.7%

WORLD NATURAL RUBBER PRODUCTION SHARES

Thailand 30.3% Indonesia 27.8% Malaysia 11.9% India 8.0% Viet nam 6.0% China 6.0% Ivor y Coast 1.9% Ot her s 8.2%

NATURAL RUBBER STRUCTURE

Variables -10 -5 0 5 10 15 -15 -10 -5 0 5 10 I PB235 RRIM600 (young) O K RRIM600 (old) BPM24 J GT1 J Axis 1 (26.4%) Axis 2 (13.7%) • Mesostructure (7) • Macrostructure (6) • Breakdown behavior (12) • Vulcanization characteristics (20) • Clones (4) • Season (3) • Ages (2) • Lipid composition (98)

Principal component analysis showed clearly that data from PB235 clone play an important role in the global variance of obtained data. Indeed, in most cases, PB235 clone displayed distinguishable characteristics. To get rid of unknown (or not lipid-dependent) PB235 clonal effect, correlations were therefore observed by putting PB235 apart from the others namely RRIM600

60 70 80 90 100 110 M ean( PR I_R R IT ) 0 .1 .2 .3 .4 .5 .6 .7 .8 .9 Mean(% unsat vs NR) Me a n ( P R I) (a)

Mean (%unsaturated fatty acids w/w dry rubber)

P la s tic it y r e te nt ion in de x ( P R I)

% unsaturated fatty acids w/w dry rubber 80 85 90 95 100 105 110 115 Mean( PR I_R R IT ) .00 .01 .02 .03 .04 .05 .06 .07 .08 .09 .10 .11 .12 .13 Mean(%alpha toco vs NR) Me an (P RI )

Mean (%α-tocotrienol w/w dry rubber) (a)

% α-tocotrienol w/w dry rubber

Pl a s ti c ity r e te n ti o n i n d e x (PR I) (c) (d)

7. Conclusions and perspectives

This study permitted a characterization of lipids composition, structures and properties of fully identified natural rubber samples from various Hevea clones, collected in a database. This allowed through statistical analyses, to provide an overview of the relationships between lipid composition and natural rubber properties. It can be seen that lipids display both advantage and disadvantage effects on natural rubber properties. Some involvements with structure of natural rubber have been also mentioned. Moreover, USS rubber displayed a high but narrow range. With the knowledge from this work, further studies are in progress using a similar approach with various rubber types whose properties cover a wider range of values.

Acknowledgement:Author wish to thank French-Thai committee project, and Thailand Research Fund (TRF) for their financial support and also Visahakit Thai Rubber for samples providing and the hosting of field experiment .

This poster is the part of the Ph.D. thesis entitled “ Characterization of lipid composition of sheet rubber from Hevea brasiliensis and relations with its structure and properties”, Liengprayoon S., 2008 IRRDB Natural Rubber Conference, 13-15 October 2008, Selangor, Malaysia During raw natural rubber

processing, some of non-isoprene components are lost with water used during cleaning process while others, especially lipids are retained in dry rubber due to their hydrophobic properties.

References

4.Hasma, H. 1984. Ph.D thesis, Faculteit Landbouwwetenschappen, Gent, Belgium.

5.Liengprayoon, S. et al., 2008. Eur. J. Lipid Sci. Technol. vol 110. 563-569. 6.Rodphakdeekul, S. et al., 2008. Kasetsart J. (Nat. Sci.) 42 : 306 - 314 1.IRSG report, vol 7., No. 11-12 May-June 2008

2.Thai ministry of commerce report In The Rubber International Magazine, vol 10, No.9, 2008

3.Bonfils, F. and L., Vaysse. IRRDB Annual Symposium, Thailand. 2003 25 30 35 40 45 50 55 M ean(P o_RRI T ) 2.0 2.5 3.0 3.5

Mean(%Lipid extract vs rubber)

M

ean (Po)

Mean (% lipids w/w dry rubber) (a)

% lipids w/w dry rubber

In it ia l W a lla c e P la s tic it y (P 0 ) 28000 30000 32000 34000 36000 38000 M ean( Cum ulat ed ener gy ( J )) 0 .1 .2 .3 .4 .5 .6 .7 .8 Mean(%furan FA vs NR) % furan fatty acid w/w dry rubber

C ons um e d e ne rgy d ur in g M a s tic a tion ( J ) (b) (old and young), GT1 and BPM24.

Lipids : 2.0 – 3.3% (w/w dry rubber)

Neutral lipids: 82 - 86% (w/w total lipids) Glycolipids: 10 - 13% (w/w total lipids) Phospholipids: 4 - 5% (w/w total lipids) Free fatty acids :

0.4 – 0.6% (w/w dry rubber)

Total fatty acids :

0.9 – 1.9% (w/w dry rubber)

C18:2: 33 - 44% (RRIM600, GT1 and BPM24) Furan fatty acid: 81% (PB235) Unsaponifiable :

0.9 – 1.1% (w/w dry rubber)

β-sitosterol: 40 - 59% (w/w total unsaponifiable) 2. Unsmoked sheet (USS)

Δ5_{-avenasterol was identified} C18:2: 37 - 53% (RRIM600, GT1 and BPM24)

Furan fatty acid: 74% (PB235) Neutral lipids: 39 - 61% (w/w total lipids) Glycolipids: 21 - 36% (w/w total lipids) Phospholipids: 18 - 26% (w/w total lipids)

Lipids : 2.5 – 3.7% (w/w dry rubber)

Free fatty acids :

0.04 – 0.06% (w/w dry rubber)

Total fatty acids :

0.8 – 2.0% (w/w dry rubber)

β-sitosterol: 34 - 46% (w/w total unsaponifiable)

1. Fresh latex

Unsaponifiable :

0.9 – 1.6% (w/w dry rubber)

Δ5-avenasterol was identified