Factors influencing the style of brandy

Bougas, Nina Valleska (2014-04)

Thesis (PhD(Agric))-- Stellenbosch University, 2014.

Thesis

ENGLISH ABSTRACT: Brandy producers in South Africa can produce three categories of brandy, and within those categories, varies styles in order to cater for diverse consumer needs. Thus they have divergent opinions as to what comprises a good base wine and distillate for their individual end-product style in mind. Due to the fact that brandy is the largest locally produced spirit in South Africa, it is crucial to investigate the factors that influence the production of brandy as better understanding and control of these processes leads to the production of a brandy that is more consistent in quality and in style. The factors known to influence the quality of brandy are the grape cultivar, vinification techniques, distillation and the oak maturation process; however limited research has been done on the factors contributing to the style of brandies. Understanding and identifying the factors that contribute to the style of brandy, will allow for better control and manipulation of the process to ultimately produce a spirit product with a desired style classification. In order to do so, one must distinguish as to what differentiates (chemical and sensory analyses) the two most divergent styles of brandy (in this study referred to as class one and class five three-year barrel matured brandy). Over 104 three-year old barrel matured brandy samples were collected and analysed using GC-FID and HPLC analysis over a three year period. Of these samples, only 7 were classified as a class one three-year barrel matured brandy and 23 were found to be classified as class five three-year barrel matured brandy. The results obtained showed that the class one samples were found to contain a higher concentration of total higher alcohols, higher ratio of higher alcohols vs. esters, higher alcohols vs. acids and aldehydes vs. esters, but a lower level of total esters and acids. The class five samples contained a higher concentration of total esters, furans, lactones and acids, but a lower level of total higher alcohols and ratio of higher alcohols vs. esters. The factors investigated that possibly influence the style of brandy were: the base wine prior to distillation, yeast strain, fermentation temperature, condenser water temperature and barrel age. Results of the base wine analysed prior to distillation show that those samples that scored a sensory score of one could produce a class one three-year barrel matured brandy, and base wines that scored a sensory score of five would yield a class five three-year barrel matured brandy. Alchemy 1 yeast was found to produce elevated levels of total esters and could by default result in a class five three-year barrel matured brandy. Similarly if the wine was fermented at 24⁰C it would result in the increased production of higher alcohols and total acids and in-turn could yield a class one three-year barrel matured brandy. The results obtained show that the different condenser water temperatures do not necessarily influence the concentration of esters and higher alcohols. However, if a lower condenser water temperature (8⁰C) was employed, it could yield a class one barrel matured brandy sample. These distillates are associated with a higher concentration of total higher alcohols and even though they may be positively associated with higher levels of esters, the esters could be removed during the head fraction in the distillation process. The results also show that during the maturation process that the use of 18 year old barrels for the maturation of three-year barrel matured brandy samples could possibly produce a class one three-year barrel matured brandy and the use of new barrels could yield a class five three-year barrel matured brandy sample. It is recommended that analysing the base wine prior to distillation using GC-FID to identify the chemical composition is the most important step in determining the outcome of the style of brandy.

AFRIKAANSE OPSOMMING: Brandewyn produsente in Suid-Afrika kan drie kategorieë brandewyn produseer, en binne hierdie kategorieë sekere style om voorsiening te maak vir die diverse behoeftes van die mark, en dus het hulle uiteenlopende menings oor wat 'n goeie basis wyn en distillaat behels vir hul individuele eindproduk se spesifieke styl. As gevolg van die feit dat brandewyn die belangrikste spiritus produk is wat deur die Suid-Afrikaanse bevolking verbruik word, is dit noodsaaklik dat dié faktore wat die produksie van brandewyn beïenvloed ondersoek word om 'n beter begrip van, en beheer oor hierdie prosesse verkry kan. Dit sal lei tot die produksie van 'n brandewyn van konstante in qualitiet en in style. Die faktore wat bekend is om die gehalte van brandewyn te beïnvloed, is die druifkultivar, wynbereidingstegnieke, distillasie en die hout verouderings proses, maar daar is beperkte navorsing oor die faktore wat bydra tot die styl van n brandewyn. Om die faktore wat bydra tot die spesifieke styl van n brandewyn te identifiseer en te verstaan, kan dit lei tot 'n beter beheer en manipulasie van die prosess om uiteindelik n brandewyn met 'n gewenste styl te produseer. Ten einde dit te kan doen, moet 'n mens kan bepaal wat die twee uiterste style van brandewyn (wat in die studie sal bekend wees as klas een en klas vyf drie-jaar vat veroude brandewyn) van mekaar onderskei (chemiese en sensoriese analise). Meer as 104 drie-jaar vat verouderde brandewyn monsters is ingesamel en ontleed met behulp van GC-FID en HPLC analyliese oor 'n tydperk van drie jaar. Van die 104 monsters, is slegs 7 as 'n klas een drie-jaar vat verourde brandewyn geklassifiseer en 23 as klas vyf drie-jaar vat verourde brandewyn geklassifiseer. Die resultate het getoon dat die klas een monsters 'n hoër konsentrasie van die totale hoër alkohole, 'n hoër verhouding van hoër alkohole teenoor esters het, hoër alkohole teenoor sure en aldehiede teenoor esters, maar 'n laer vlak van totale esters en sure. Die klas vyf monsters het 'n hoër konsentrasie van die totale esters, furanen, laktone en sure, maar 'n laer vlak van totale hoer alkohol en verhouding van hoër alkohole teenoor esters. Die faktore wat ondersoek was wat moontlike ‘n invloed op die styl van brandewyn kon hê, was die basis wyn voor distillasie, die gisras, die fermentasie temperatuur, die kondensor water temperatuur en die vat ouderdom. Die ontledings van die basis wyn voor distillasie het gewys dat die monsters wat 'n sensorise telling van een behaal het, moontlik kan lei tot 'n klas een drie-jaar vat verouderde brandewyn, en basis wyne wat 'n sensorise telling van vyf gekry het weer ‘n klas vyf drie-jaar vat verouderde brandewyn sou oplewer. Alchemie 1 gis ras het ‘n verhoogde vlak van totale esters geproduseer wat tot 'n klas vyf drie-jaar vat verouder brandewyn gelei het As die basis wyn gefermenteer is teen 24⁰C sal dit lei tot die verhoogde produksie van hoër alkohole en totale sure wat 'n klas een drie-jaar vat verouderde brandewyn kan oplewer. Die resultate verkry, toon dat die verskillende kondensor water temperature nie noodwendig die konsentrasie van esters en hoër alkohole beïnvloed het nie, maar dit wil voorkom dat indien 'n laer kondensor water temperatuur (8⁰C) gebruik is, kan dit 'n klas een vat verouderde brandewyn oplewer. Dit hou verband met 'n hoër konsentrasie van die totale hoër alkohole en selfs al word die distillate positief geassosieer met hoër vlakke van esters, kan die esters verwyder word tydens die verwydering van die voorloop in die distillasie proses. Die resultate toon ook dat die gebruik van 18 jaar oue vate vir die veroudering van brandewyn moontlik 'n klas een drie-jaar vat verouderde brandewyn kan oplewer en die gebruik van nuwe vate kan 'n klas vyf drie-jaar vat verouderde brandewyn lewer. Dit word egter aanbeveel dat die basis wyn voor distillasie met GC-FID chemies ontleed word om die samestelling van die wyn te identifiseer Die resultate bly die beste aanduiding van die styl van die brandewyn.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/86331
This item appears in the following collections: