THE CLASSIFICATION OF NATURAL GUMS .3. ACACIA-SENEGAL AND RELATED SPECIES (GUM-ARABIC)

Citation
P. Jurasek et al., THE CLASSIFICATION OF NATURAL GUMS .3. ACACIA-SENEGAL AND RELATED SPECIES (GUM-ARABIC), Food hydrocolloids, 7(3), 1993, pp. 255-280
Citations number
17
Categorie Soggetti
Food Science & Tenology","Chemistry Physical","Chemistry Applied
Journal title
ISSN journal
0268005X
Volume
7
Issue
3
Year of publication
1993
Pages
255 - 280
Database
ISI
SICI code
0268-005X(1993)7:3<255:TCONG.>2.0.ZU;2-N
Abstract
Chemometric methods have been used to characterise and evaluate commer cial gum arabic in relation to authenticated Acacia senegal specimens and gums from the Combretum series. Principal component (PCA) and disc riminant component analysis (DCA) were undertaken using firstly parame ters mainly associated with the carbohydrate moiety: specific optical rotation, viscosity, %N, and sugar compositions, galactose, arabinose, rhamnose, equivalent weight, glucuronic acid and 4-methylglucuronic a cid, Secondly, the compositions of the 18 amino acids were utilised, a nd subsequently the 27 features were combined. Of the 54 samples inves tigated, it is evident that the commercial gum arabic and authenticate d Acacia senegal mainly fall into a discrete cluster. The outliers are few and can be explained. The Acacia gums can readily be distinguishe d from the Combretum gums. The samples are drawn from various countrie s: Sudan, Niger, Nigeria, Uganda, Oman, Mauritania, Mali, Senegal, Eth iopia, Kenya and Uganda. No significant difference is found in the gum from the main producing countries. Local variants from Oman and Kenya do not conform to the specification. Using PCA with amino acid compos itions, the family of gums from Uganda can be distinguished. The gums are drawn from periods ranging from 1903 to the present, and no signif icant change has taken place in the gum over this period. Neither clim ate nor nature of the soil introduces any particular difference. When all 27 parameters are used, it can be demonstrated that the amino acid s are dominant in establishing the distinctive character of Acacia sen egal. Using loading-loading plots, the major distinctive features can be selected. As a result a characterisation is now possible using only four features: rotation, viscosity, lysine and hydroxy-proline compos itions. Based only on the two features (specific optical rotation and %N) proposed by the Joint Expert Committee for Food Additives (JECFA), no correlation is possible. It is proposed that such a chemometric cl assification could be used to set the specification for commercial gum arabic and provide the basis for a practical quality control system f or producers and users.