v... .

 

silk .0!
:23... .3 .2!-
s :l:.!.o:1lla
x3... 3.. .2 alxx‘i
1113.121."
5.. 2 2!
.23‘...

hrzgmwi ”.22 $§m¥m
3.111“. .11. .«Khfllmb at I: Iii
2.2. :a@ .itiksnyaviLiéixflfiuu .
$32.”? .22... 22.2»
a... $2.25.“... , .
my... . . . mw .
.9...22...m.....ux...2., 29523....

v :3
2. .2 2.!
k .2.. .2... 2.. 23.2.2...

 

c.3-

.l ..

.I .01.. i“
I .
Ill AI.

Chums?”
. . .
a 3.1. .HPAL‘IJ all!“
13.0132... l‘wufi!
tar"... .3233... ‘3!

3.2“}

day

 

. u . 225.. 2. . . . . . . 2 .
.. .2 2.223262 . 2h
. .. .2. .2.. .11.: 2..
.5323... 2...... .1. 1 «$2.32....
0". 338..“- . I {gotta-:1 h 2 17!
323.520.. :3 .. .I|~.!)3 1.1:: {(1%)
.40., 5.7.. an...:.«.uhfiu.n|.: .3: I 5. hail. :LuuH-WII. Jul thuhwtlt Jordan
PK". . _ islua... Ednifii $5.38....
. II IDS-kKI't’. a “rag? uttliilaufll

.
8 l. 32.23:... 5 3:.

:Iu £233.12! I
1412“.-
..

2.. r.222.....21ih .2)...
. a» “an. $2.23.... .

3
.31 .Etxpét z (.12.... . . t.
1 ,4 3 ..u2a..lt!2.2:4tfctl.lt.. .

3.)... 1.2.22 22!...
5.2.2:... (5222......222. . .
2:22.22..2.£.2\..x..!..22 3.22.5.2! .2252):
2.222222225523222fn . .2....th
(21‘:- .t1. ‘1: I .
Stats. in:

i: 11...}.50955012732!‘ 2 L
.iV ).A)I..'2§2...z)Vzr(".V§2|")|lvnl
..... Ki! 2

41‘ . . . . .

mESlS

’L‘O/Z/

This is to certify that the
thesis entitled
THE EFFECT OF YEAST STRAINS USED

FOR FERMENTATION ON THE CONGENER
CONCENTRATION IN DISTILLED FRUIT BRANDIES

presented by

Matthew D. Berg

has been accepted towards fulfillment
of the requirements for

M. S . degree inflheminaLEngineering

2&4:

Major professor

Date 7/2/47 Xxx/*2,

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

LIBRARY
Michigan State
University

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE ' DATE DUE

 

“P3? I0'71 @033

 

ug’crohszoma

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/ClRC/DateDue.p65~p.15

THE EFFECT OF YEAST STRAINS USED
FOR FERMENTATION ON THE CONGENER
CONCENTRATION IN DISTILLED FRUIT BRANDIES
By

Matthew Derek Berg

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE
Department of Chemical Engineering

2002

ABSTRACT

THE EFFECT OF DIFFERENT YEAST STRAINS USED FOR FERMENTATION ON
THE CONGENER CONCENTRATION IN DISTILLED FRUIT BRANDIES

By

Matthew D. Berg

The goal of this research was to demonstrate that different yeast strains produce
varying amounts of flavor components when fermenting and distilling fruit to produce
brandy. Five strains of yeast were used for fermentation of nine different fruits that are
commonly grown in Michigan. The resulting distillates were analyzed using gas
chromatography to identify ethanol, methanol, fusel alcohols, benzaldehyde, acetone, and
acetaldehyde. Based on an objective ranking system, the five yeasts were compared to
one another to determine which yeast was best for fermentation by analyzing the
resulting distillates. These data will benefit distillers by increasing the predictability and

quality of brandy production thus making the process more profitable.

TABLE OF CONTENTS

LIST OF TABLES ................................................................... vi
LIST OF FIGURES .................................................................. vii
INTRODUCTION

1.1 Michigan Fruit Brandy Industry .......................................... 1
1.2 The History of Distillation ................................................. 1
1.3 Different Distillation Approaches ........................................ 2

l .4 Objectives .................................................................... 2

BACKGROUND INFORMATION

2.1 Raw Materials ............................................................... 4

2.2 Fruit Mashing ................................................................ 5

2.3 Ferrnentor ..................................................................... 6

2.4 Yeast .......................................................................... 9

2.5 Yeast Strains ................................................................. 14
2.6 Fermentor Storage .......................................................... 15
2.7 Mash Distillation ............................................................ 15
2.8 Distillate Preparation and Storage ........................................ 16
2.9 Gas Chromatographic Analysis ........................................... 19
2.10 Ethanol ........................................................................ 20
2.11 Methanol ..................................................................... 2O
2. 1 2 Acetaldehyde ................................................................ 2 1
2.13 Ethyl Acetate and Ethyl Formate ......................................... 23
2. l4 2-Furaldehyde ............................................................... 23

iii

2. l 5 Benzaldehyde ................................................................ 24
2. l 6 Fuse] Alcohols ............................................................... 25
MATERIALS AND METHODS

3.1 Preferrnentation Set-Up .................................................... 28
3.2 Selection and Preparation of Fresh Fruit ................................. 29
3.3 Yeast Preparation ............................................................ 31
3 .4 Fermentation .............................................................. g . .. 3 1
3.5 Distillation ................................................................... 32
3.6 Distillate Analysis .......................................................... 32
RESULTS AND DISCUSSION

4.1 Quantitative Identification of the Compounds .......................... 36
4.2 Ethanol Calibration ......................................................... 38
4.3 Ethanol as an Internal Standard ........................................... 40
4.4 Ranking of the Yeasts ...................................................... 40
4.5 Bartlett Pears ................................................................ 42
4.6 Black Plums ................................................................. 44
4.7 Braebum Apples ............................................................ 46
4.8 Gala Apples .................................................................. 48
4.9 Granny Smith Apples ...................................................... 50
4.10 Green Seedless Grapes ..................................................... 52
4.1 1 Montmorency Cherries ..................................................... 54
4.12 New Haven Peaches ........................................................ 56
4.13 Red Seedless Grapes ........................................................ 58
COMPUTER SIMULATION ...................................................... 60

iv

6. SUMMARY AND CONCLUSIONS ............................................. 66

7. FUTURE WORK .................................................................... 68

BIBLIOGRAPHY ................................................................... 69

Table 3.1
Table 3.2

Table 4.1

Table 4.2

Table 4.3

Table 4.4

Table 4.5

Table 4.6

Table 4.7

Table 4.8

Table 4.9

Table 4.10

Table 5.1

Table 6.1

LIST OF TABLES

On/Off Micromega CN7700 Controller Settings ........................ 3O
Shimadzu GC-17A Run Conditions ...................................... 34

Typical Fruit Distillate Components, Boiling Points, and
Retention Times Seen Experimentally Using a Shimadzu Gas

Chromatograph and FID Detector ........................................ 37
Bartlett Pear Distillate Composition and Rank .......................... 42
Black Plum Distillate Composition and Rank ........................... 44
Braeburn Apple Distillate Composition and Rank ..................... 46
Gala Apple Distillate Composition and Rank ........................... 48
Granny Smith Distillate Composition and Rank ........................ 50
Green Seedless Grape Distillate Composition and Rank .............. 52
Montmorency Cherry Distillate Composition and Rank ............... 54
New Haven Peach Distillate Composition and Rank .................. 56
Red Seedless Grape Distillate Composition and Rank ................ 58

The variation in Montmorency cherry distillate congener
concentration estimated by using ChemCAD Batch

simulation software. The Montmorency cherries were

fermented using the CY3079 strain of yeast ............................. 63

The best performing yeast determined by ranking the congener
composition of each fruit distillate ....................................... 67

vi

Figure 2.1

Figure 2.2

Figure 2.3

Figure 2.4
Figure 2.5
Figure 2.6
Figure 2.7
Figure 2.8

Figure 2.9

Figure 2.10

Figure 2.1]
Figure 3.1
Figure 4.1
Figure 4.2
Figure 4.3
Figure 4.4
Figure 4.5
Figure 4.6

Figure 4.7

LIST OF FIGURES

Gay-Lussac Reaction: conversion of glucose to ethanol

And carbon dioxide by Saccharomyces cerevisiae ..................... 7
Diagram of an Airlock ...................................................... 8
The Embden-Meyerhof-Parnas fermentation scheme

showing the intermediates in the production of ethanol

from glucose .................................................................. 11
Active transport of glucose into a yeast cell ............................. l3
Schematic of a Christian Carl Still Set-Up ............................... 18
Pectin and the activity of pectinesterase to produce methanol ........ 22
The production of ethyl acetate ............................................ 23
The production of benzaldehyde by the hydrolysis of amygdalin. . ..24
The production of urethane ................................................ 25
Schematic of pathways for fusel alcohol formation .................... 26
The production of 3-methyl-l-butanol ................................... 27
Simple Distillation Set-Up ................................................. 33
Ethanol Calibration for Gas Chromatography ........................... 39
Bartlett Pear Distillate Ethanol and Methanol Concentrations ........ 43
Black Plum Distillate Ethanol and Methanol Concentrations. . . . . 45
Braeburn Apple Distillate Ethanol and Methanol Concentrations... 47
Gala Apple Distillate Ethanol and Methanol Concentrations ......... 49
Granny Smith Distillate Ethanol and Methanol Concentrations ...... 51
Green Grape Distillate Ethanol and Methanol Concentrations ........ 53

vii

Figure 4.8
Figure 4.9

Figure 4.10

Figure 5.1

Figure 5.2

Montmorency Cherry Distillate Ethanol and Methanol
Concentrations ............................................................... 55

New Haven Peach Distillate Ethanol and Methanol
Concentrations ............................................................... 57

Red Grape Distillate Ethanol and Methanol Concentrations .......... 59

CY3079 Montmorency Cherry Distillate Methanol
Profile Using ChemCAD Batch Simulation ............................. 64

CY3079 Montmorency Cherry Distillate Ethanol
Profile Using ChemCAD Batch Simulation ............................ 65

viii

1. INTRODUCTION

1.1 Michigan Fruit Brandy Industry

Agricultural advances in the United States and particularly in the state of
Michigan have increased the number of apples, cherries, grapes, peaches, pears, and
plums grown each year. Despite this increased growth, the demand or consumption of
these fruit crops has remained relatively stable resulting in large surpluses of fruit crops.
These surpluses combined with fruit determined to be unfit for market due to cosmetic
defects have created a need for the development of alternative uses for excess fruit crops.

A relatively new use currently being introduced into the United States for excess
fruit crops is the production of brandy. This technique has been extensively used in
Europe, but is still in its infancy in the United States. The number of stills in the state of
Michigan used for the distillation of fruit spirits has grown from zero in 1996 to seven in
the year 2002. This growth in the number of distilleries can be accounted for by the need
to develop a market for excess fruit crops combined with changes in the state laws of
Michigan regarding the licensing of fruit distilleries to facilitate the creation of a
distilling industry.
1.2 The History of Distillation

Although the distillation of fruit spirits is a relatively new technology being
developed in the United States, it has been utilized for many centuries. Distillation is a
very old technique which was used by the Chinese 3000 years BC, the East Indians 2500
years BC, the Egyptians 2000 years BC, the Greeks 1000 years BC, and the Romans 200
BC.6 Early stills consisted of a copper boiler heated over an open flame. The alcohol

vapors were condensed and collected using wool fibers. Distillation of this type is called

alambic distillation. The Arabs brought the technique of distillation to Europe around the
sixth century AD. In 1250, Arnaud de Villenueve was the first to distill wines in France;
he called the product that resulted from this process “eau-de-vie” or water of life.6 Over
750 years later Villenueve’s name for distilled wines is still used to identify distilled fruit
spirits. The Dutch, French, Irish, Scottish and others started producing distilled spirits
around the 15‘h and 16h century.6 They created gin (Holland), whiskey (Scotland and
Ireland), arrnanac (France), and cognac (France).
1.3 Different Distillation Approaches

There are two styles of fruit distillation currently in use. The “French” style
involves distillation through a simple pot still, which is called alambic. Multiple
distillations are required in order to obtain high proof spirits. Multiple distillations cause
a loss of ethanol and more importantly a loss of aromatic compounds that lend flavor to
the distillate. Spirits from the “French” style of distillation are termed cognac style
Spirits. The “German” style of distillation involves a single pass through a batch still
with a reflux column to obtain high proof Spirits. These spirits are usually called eau-de-
vie (French) or schnapps (German). They are traditionally stored in glass and served as
water clear brandies.‘8 The batch column still is designed to trap the fruit essence, which
adds to the brandy’s aroma and flavor. The additional aroma and flavor of fruit spirits
distilled using the “German” style of distillation makes it the preferred method for most
fruits except for grapes (cognac and armanac) and apples (calvados).
1.4 Objectives

The objective of this study was to determine the relationship between the strain of

yeast used to ferment various fruits and the resulting composition of the distillate. The

five strains of yeast studied were Lalvin Bourgoblanc CY3079, Lalvin ICV 254D, Lalvin
Prise de Mousse EC-1118, Lalvin Rhone L2056, and Lalvin S6U. The following
chemical compounds in the distillate were analyzed: acetaldehyde, acetone, methanol, 2-
propanol, ethanol, 2-methyl-2-butanol, 2-butanol, l-propanol, l-butanol, 3-methyl-1-
butanol, l-pentanol, l-hexanol, l-heptanol, benzaldehyde, and benzyl alcohol. A

comprehensive guide for yeast strain selection for a given variety of fruit was developed.

2. BACKGROUND INFORMATION

2.1 Raw Material

Generally all sugar containing and ferrnentable materials can be considered as
basic materials for the production of brandies.18 In this study the focus was on the
production of brandy from fresh fruit sources grown in the state of Michigan. The
domestic fruits studied were Bartlett pears, black plums, Braeburn apples, Gala apples,
Granny Smith apples, green grapes (seedless), Montmorency cherries, Newhaven
peaches, and red grapes (seedless).

Fruit for fermentation was selected that was at an advanced stage of ripening yet
visibly free of mold. With advanced ripening the sugar content increases at the cost of
the acid content, a fact usually desired.‘8 Higher sugar content leads to the production of
greater quantities of ethanol; however, fruit with an acid content too low (high sugar/acid
ratio) not only affects the taste in an unsatisfactory way, the mashes obtained from them
are more susceptible to the development of undesired microorganisms.‘8 Acetic acid,
lactic acid, and butanoic acid bacteria are among the most common microorganisms
potentially found in overly ripe or moldy fruit. The microorganisms utilize sugars such
as glucose and fructose to produce unwanted products other than ethanol. 2-Butanol, 1-
propanol, and ethyl acetate are typical compounds formed due to bacterial activity.2

The main constituents of fruit include water, carbohydrates, fruit acids, protein
like substances, phenolic substances, vitamins, aromatic agents, and mineral agents. The
water content of fresh fruit is between 80 and 85%.18 Water decreases the viscosity of a

fruit mash thus assisting fermentation.

The main carbohydrates found in fresh fruit include glucose, fructose, and
sucrose. Glucose and fructose are monosaccharides and thus are directly fermentable by
yeast. Sucrose is a disaccharide and thus is not directly fermentable by yeast. Bakers
yeast typically possesses a potent invertase activity.10 This enzyme splits sucrose into
glucose and fructose.lo Invertase therefore allows the yeast cells to use sucrose in the
production of ethanol.

Citric, malic, and tartaric acid are the main acids found in fresh fruit. These acids
maintain a lower mash pH and thus aid in preventing the growth of unwanted
microorganisms. Protein like substances comprise 1% of the mass of fresh fruit. They
provide amino acids which serve as a breeding ground for the yeast cells and they are
affiliated with enzymes. Phenolic substances give fruits their distinctive colors. These
molecules can condense to larger molecules that can give distillates a harsh and
astringent taste. Vitamin C is present in fruits but does not effect the distillate aroma or
flavor. Aromatic agents such as alcohols, volatile acids, esters, aldehydes, acetals, and
ketones comprise less than 1% of the mass of fresh fruit. These compounds increase with
fruit ripening and give distillates their distinctive aroma and flavor. Fresh fruit contains
mineral agents such as calcium, iron, magnesium, phosphorus, and potassium that assist
the growth of yeast cells.

2.2 Fruit Mashing

The primary purpose of mashing is to increase the surface area of the fruit.

Increased surface area gives yeast cells increased exposure to carbohydrates such as

glucose, fructose, and sucrose. The type of utensil or machinery chosen to mash fresh

fruit is dependent on the volume and/or type of fruit. For small volumes of fresh fruit
mashing can be accomplished through squeezing, grinding, or mixing. A wooden pestle
is often the utensil of choice to aid in the mashing of small volumes of fruit.

Larger volumes of fresh fruit are mashed using a rolling mill or a ratz mill.
Rolling mills are used to mash stone and berry fruit. They consist of parallel rollers that
rotate in opposite directions. The distance between rollers can be adjusted to allow large
stones to pass through the mill. Ratz mills are used to crush seed fruits. The fruit is
pressed against a grinding casing by means of a rotor with several blades. This casing
has milling blades and slits at the bottom that are arranged axially. The crushed fruit is
pressed through the slits to the outside.18
2.3 Fermenter

The vessel for fresh fruit mash storage is called a fermenter and is usually
constructed of glass, stainless steel, or plastic. These materials are non-reactive with the
slightly acidic fruit mash. This non-reactivity prevents the distillates from developing a
metallic or other objectionable taste.

The fermentation of fresh fruit by yeast to produce ethanol is an exothermic
reaction. In large scale ferrnentations, even with some cooling, it is almost impossible to
prevent a significant temperature rise or peaking because of the large amount of heat
evolved during fermentation.5 To counteract the evolved heat, fermenters are equipped
with some type of cooling device. A water jacket or copper piping running through the
mash are the typical methods employed to cool the mash. The optimal temperature range

of a mash is between 15 and 20°C. There are two competing mechanisms as temperature

increases: on the one hand there is a tendency for the rate of reaction to increase; on the
other hand there is a tendency for the rate of inactivation of the yeast to also increase.lo

Temperatures below 15°C slow down the yeast kinetics and subsequently slow
down the production of ethanol. Temperatures above 20°C accelerate the fermentation
process but they also favor the growth of undesired microorganisms. High temperatures
can lead to increased production of acetic acid, lactic acid, and butanoic acid.

Another concern within the fermenter is a temperature variation within the mash
itself. Such inhomogeneity can occur as pockets of mash ferment, with the subsequent
release of heat. The fresh fruit mash is agitated to maintain a relatively consistent
temperature profile. Agitation is typically accomplished using a motor, shaft, and
propeller setup. The final aspect of a quality fermenter involves proper containment of
the mash. The fermenter should allow for the release of carbon dioxide and prevent
oxygen from contacting the mash. The descriptive equation for alcoholic fermentation by

yeast is typically referred to as the Gay-Lussac reaction (Figure 2.1):

"——> 2 HO/\ 2 0=C=O

ethanol carbon dioxide

 

glucose
Figure 2.1 Gay-Lussac Reaction: Conversion of Glucose to Ethanol and Carbon
Dioxide Using Saccharomyces Cerevisiae (Dry Yeast)lo

To allow for the release of carbon dioxide the fermenter was equipped with an air
lock (Figure 2.2) which can be constructed of plastic and is U-shaped. The lower part of
the U-shape is filled with water. The production of carbon dioxide within the sealed

fermenter causes an increase in pressure above atmospheric levels. This pushes carbon

bang—m
332%

235
cow—Sm

sue—LE :a he again N.~ «Ema

 

 

Hi

 

 

 

‘—

 

NOO +l

 

 

 

 

 

 

 

 

 

Iv NOD

 

2%:
Steam

dioxide bubbles through the water in the bottom of the U-shaped air lock. The water in
the bottom of the U-shape prevents the entry of oxygen into the fermenter. Oxygen
contact with the mash promotes the formation of undesired bacteria while the yeast used
for the fermentation of sugar does not require any oxygen.

2.4 Yeast

The similarity between the aroma compounds formed in nitrogen free sugar
fermentation and the aroma fraction of alcoholic beverages indicates that the yeast is
responsible for the bulk of the aroma compounds produced.I6 This statement summarizes
the current study quite well.

The intensive aroma research work carried out during recent years has shown that
the aroma composition of alcoholic beverages consists of several hundred distinct
chemical compounds.'7 Amino acids, pectin, and sugars such as fructose, glucose, and
broken down sucrose provide the substrate for yeast cells to produce the majority of the
chemical compounds that give fruit spirits their distinct aroma and flavor.

Baker’s yeast or Saccharomyces cerevisiae is classified as a fungus. It is capable
of reproducing either sexually or asexually. Under the nutrient conditions prevailing in
fruit mashes the reproduction takes place through budding, i.e. asexually.lo After
complete development, a daughter cell separates from the parent cell. The doubling
period takes between three to six hours depending on the strain of yeast, pH, temperature,
and mineral supply.’8

The fermentation process starts when the amount of living yeast cells has
increased to 100,000 to 1,000,000 per cubic centimeter. To give S. cerevisiae a

competitive advantage over wild yeast and bacteria these cellular levels should be

obtained immediately after mashing of the fruit. This is accomplished by starting with a
large number of cells. Thick mashes or fruit types rich in tanning agents (danger of
fermentation hold up) require an adequately higher dosage, e. g. 40g/hl.lo In industrial
fermentations of complex wortS the rate of alcoholic production is restricted by some
inhibitors resulting from accumulation of organic or inorganic compounds.9 These
compounds include acetic acid, butanoic acid, and lactic acid all formed due to bacterial
activity.

S. cerevisiae is capable of producing ethanol and carbon dioxide from fructose
and glucose. The classic reaction describing this process is called the Gay-Lussac
reaction as previously described. S. cerevisiae is also capable of utilizing sucrose after it
is broken down into fructose and glucose by an invertase enzyme. Yeast invertase is
located at the cell wall, or perhaps just inside the cell at the cell membrane.10 It has been
Shown that very shortly after the yeast comes into contact with sucrose there is no longer
any sucrose. The route by which yeast metabolizes common sugars, such as glucose or
fructose under anaerobic conditions is referred to as the Embden-Meyerhof-Parnas
scheme.'0 This scheme is shown in Figure 2.3. Glucose, fructose, and sucrose are very
quickly fermented.H Glucose, however, is fermented faster than fructose.

The theoretical yield of such a process is 64.5 L of ethanol produced from 100 kg
of sugar.18 This theoretical yield does not hold up due to incomplete fermentation of the
sugars, formation of unwanted side products, and alcohol loss due to distillation. The

actual ethanol yield is approximately 56 L from 100 kg of sugar.'8

10

Figure 2.3 The Embden-Meyerhof-Parnas fermentation scheme

showing the intermediates in the production of ethanol from glucose.10

D-Glucose

l

Glucose-6-phosphate

Fructose-6-phosphate <— Fructose

i

F ructose- l ,6-diphosphate

Di-hydroxyacetone"—> D-Glyceraldehyde-3-phosphate

phosphate
Pi DPNH

1,3-Di-phosphoglyceric acid

3-Phosphoglyceric acid

2-Phosphoglyceric acid

2-Phosphoenol pyruvate

Pyruvate

Lactate j ‘- DPNH

Acetaldehyde + CO2

*— DPNH
Ethanol

ll

Fructose, glucose, and sucrose must enter a yeast cell through active transport
(Figure 2.4). Active transport serves to explain the rapid rate of fermentation of these
sugars in yeast, while at the same time permitting accumulation against a concentration
gradient.‘0 Active transport requires the use of metabolic energy. The binding of sugar to
the carrier is visualized as catalyzed by a perrnease enzyme: a phosphorylation reaction
with polyphosphate as a phosphate donor is involved in this binding.lo

Another aspect of substrate entry into yeast cells involves the penetration of
organic acids. Carboxylase is an enzyme located inside the yeast cell. In order for a keto
acid molecule to be decarboxylated by yeast carboxylase, it must first penetrate the cell
wall and plasma membrane. The plasma membrane tends to be hydrophobic in nature so
keto acids with longer carbon chains enter the yeast cell more rapidly. Once inside the
yeast cell, keto acids are decarboxylated and reduced into fusel alcohols. Fusel alcohols
have a damp cloth odor that has a negative aromatic influence on fruit spirits.

Environmental factors can be extremely influential in the success or lack thereof
of yeast fermentation. Nevertheless, it is good operational precaution to maintain pH in
the range of 4.0 to 6.0 for best yeast activity.'0 This slightly acidic pH range inhibits the
growth of some bacteria. Fruit mashes naturally tend to fall within the suggested pH
range. For fruit mashes that do not, lowering the pH is accomplished using sulfuric acid,
and the pH is increased using calcium carbonate. The ideal temperature range for yeast
fermentations is between 15 and 20°C. This range of temperature balances two
competing mechanisms as the temperature rises; the tendency for the reaction rate to

increase and the tendency of the rate of inactivation of the yeast to also increase.‘0

12

Phosphatidyl glycerol

+ Phosphate

 

V

Phosphatidylglycerolphosphate

+ carrier

Perrnease
+ substrate

 

V

Carrier-Phosphate-Substrate+Phosphotidylglycerol

 

Cell Membrane

 

Cell

Figure 2.4 Active Transport of Glucose into a Yeast Celllo

2.5 Yeast Strains

The Lalvin D254 strain was isolated from Syrah fermentations after screening
3,000 isolates and putting 450 of them through trials for their enological properties.12
The D254 strain was selected for its ability to ferment in low nitrogen musts, while being
a low foamer with an alcohol tolerance up to 16%.12 The D254 strain has a temperature
range of 11-35°C.12 The sensory profile of the D254 strain is described as showing
pronounced butterscotch, creamy, smooth, hazelnut, and almond aromas.l2

The Lalvin CY3079 strain was selected from fermentations in the Burgundy
region, with the objective to isolate a strain that would complement the typical white
Burgundy styles of wine making. A slow, steady fermenter even at cooler temperatures,
this strain demonstrates a good alcohol tolerance and low production of volatile acids and
hydrogen sulfides. It has an alcohol tolerance of 14%.12 The CY3079 strain releases

peptides at the end of fermentations that are believed to enhance many of the aromas such
as fresh butter, honey, white flowers, and pineapple.12 The CY3079 strain has a
temperature range of l3-29°C.12

The Lalvin S6U strain was isolated and studied by Dr. Cioffi of the Institute
Sperimentale per I’Enolagia near Rome. The enological characteristics include the
ability to ferment at low temperatures in musts with low levels of suspended solids. The
temperature range for S6U is 4-32°C and it has an alcohol tolerance of 15%.12 Aroma

profiles of the wines produced by S6U are described as floral and spicy.'2

Lalvin EC1118 Prise de Mousse is the original, steady, low foamer, excellent for

barrel fermentations. It ferrnents well at low temperatures with a range of 4-35°C.l2 It

14

has an alcohol tolerance that is greater than 18%.'2 EC] 118 has excellent organoleptic
properties.

Lalvin L2056 was isolated from the Cotes du Rhone region because of its ability
to maintain a variety of fruit aromas and flavors. The strain demonstrates good alcohol
tolerance to 16% and has a temperature range of 15-28°C.12
2.6 Fermenter Storage

Fresh fruit fermented between 15-20°C requires a fermentation period of 10-20
days.18 For consistency in this study, all fermentations were run for 14 days.
Ferrnentations that do not run for ten days run the risk of not utilizing all of the available
fructose and glucose. This would mean that the resulting fruit distillate would have a
decreased volume of ethanol. Fermented mashes can be stored up to three to four weeks
without disadvantageous changes.18 Beyond four weeks fermentations run the risk of
bacterial infection.

2.7 Mash Distillation

The basic concept of distillation involves separating volatile from non-volatile
compounds using some type of heat source. Volatility is a physical property of a
compound described by its boiling point. Compounds described as highly volatile have
low boiling points and compounds described as non-volatile compounds have high
boiling points. Heat sources include direct fire, steam, and a water/steam combination.
The fermented mash is placed into a still constructed of copper, glass, or high-grade steel.
Several reasons favor the use of copper including its ability to conduct heat effectively, it

Shows optimal resistance to fruit acids, and it effects the quality of the distillate

positively.l8 Copper forms non-volatile products with volatile sulfur compounds, thus
improving the fruit distillate’s aroma and flavor by removal of sulfur compounds.

Once vaporized, the volatile mash components travel into the partial condenser.
The partial condenser is a cylindrical structure with several trays. The use of a water
jacket with cold water flowing through it creates zones of varying temperature. AS the
vapors travel upward within the partial condenser the temperature gradually falls. With
falling temperatures components that are less volatile condense, get trapped on trays, and
fall back into the fermentation mash. This process is commonly known as reflux.'8

Compounds that are more volatile continue to ascend to the top of the condenser
and enter a spirit tube. The Spirit tube is a pipe that connects the partial condenser to the
total condenser. The total condenser also has a water jacket with cold water flowing
through it. This condenses the volatile compounds that are collected from a discharge
point on the total condenser. The volatile compounds collected from the total condenser
comprise the fruit distillate. A schematic of a typical still set-up is shown in Figure 2.5.
2.8 Distillate Preparation and Storage

Distillates collected from the total condenser typically are comprised of 60-80
percent ethanol. Distillates are stored at this high concentration of ethanol for a three to
Sixth month period. Reactions that occur in the aging process include oxidation,
esterification, and acetalization. Storage is typically done in an environment that is free
of light. It is theorized that the production of ethyl carbamate is a process catalyzed by
light. An attempt to minimize the amount of ethyl carbamate found in distillates is made

because it is a known carcinogen. Before consumption distillates are diluted to a

Figure 2.5 Schematic of a Christian Carl Still Set-Up3

l7

“l

Distillation Vapor Pat
Cooling Water Path

 

 

 

1C

atalyt
Converter

}

 

. . ..-
. .
.......

 

  

 

 
 

0
L.
‘3
H
M
$-
01...
a:
8::
r00
E—tn

  

 

 

'5
.1:
C
O
U
>\
a
In
[—

 

./I/

   

i
t

_T'

 

 

 

 

 

 

 

 

 

a, , I -
l
I .
l o
E
3 it; on '5‘;
5 m .E a
3 1: 5 x 2
'23 2 '2 8 .. I
a U o a- 3
H __ U l— E: I E
E .‘2 7-. g. E .3
N
2 :3 s s 2.; a a
m a. [- 0 LL. J

 

 

L evers

 

 

 

 

18

.
a

Cooling Water

Connol

Steam

lnkt

Distillate
Discharge

   
 

 

W ater
Inlet

Bottoms
Rem oval

 

To Drain

   
  

   

concentration of ethanol that falls below governmental regulations (40%). This process
is termed as cutting the distillate and is done using purified water.
2.9 Gas Chromatographic Analysis

Gas chromatography is a technique used to separate and analyze compounds that

are capable of being vaporized in a temperature range of 25 to 250°C. Samples are

injected into a flash vaporizer. The flash vaporizer is at least 50°C above the boiling
point of the least volatile component of the sample.13

A carrier gas such as helium, nitrogen, or hydrogen then carries the sample into
the column. The carrier gas acts as the mobile phase. The stationary phase is usually a
liquid attached to the walls of an open tubular column.l3 Choice of the stationary phase
depends on the polarity of the components that are being separated. Fruit distillates are
comprised mostly of alcohols that are polar in nature. Polyethylene glycol therefore is
the liquid stationary phase of choice.13 To improve separation within the column
temperature programming is often employed. For samples with a broad boiling range, it
is often desirable to employ temperature programming, whereby the column temperature
is increased either continuously or in steps as the separation proceeds.'3 The gradual
increase in temperature is accomplished by housing the column within an oven.

A flame ionization detector is the most widely used and generally applicable
detector for gas chromatography.13 A hydrogen flame pyrolyzes organic compounds to
form ions and electrons that conduct electricity through the flame. A potential of a few
hundred volts is passed through the flame and the resulting current is directed into an

amplifier for measurement.

19

2.10 Ethanol

Ethanol is formed by yeast cells in the breakdown of glucose and fructose to form
two moles of carbon dioxide and two moles of ethanol. The main goal of the distillation
of fruit spirits is to maximize the production of ethanol. Increased ethanol production
means that a greater number of bottles of 40% ethanol can be produced. At the end of the
day this means that net profits will increase. Making the distillation of fruit spirits more
profitable will lead to the continued growth of this industry in the state of Michigan and
throughout the United States.

2.11 Methanol

The production of methanol is not quite as straight forward as the production of
ethanol. Methanol is a positive aroma and flavor component within fruit spirits yet it is
toxic at high levels. The toxicity of methanol has led to its regulation by the United
States Environmental Protection Agency and the Food and Drug Administration. The
United States Environmental Protection Agency mandates a minimum acute toxicity
concentration of methanol in drinking water at 3.9 parts per million.4 The United States
Bureau of Alcohol, Tobacco, and Firearms regulates the amount of methanol to 0.35%
volume/volume for both domestic and imported fruit brandy.

Methanol is formed from the breakdown of pectin by pectinesterase. Pectin is a
polysaccharide that occupies intercellular Spaces in fruit tissue. Pectin contributes to the
adhesion between cells and the overall mechanical strength of the cell wall. Fruit cells
contain approximately 60% water and 40% biopolymers. Pectin makes up 20-3 5% of the

polymers.8

20

Pectinesterase is an enzyme that catalyzes the hydrolysis of the methyl esters of
pectin to generate methanol. Pectinesterase has been found to naturally occur in a wide
variety of plants. The structure of pectin and the activity of pectinesterase is Shown in
Figure 2.6.2
2.12 Acetaldehyde

Acetaldehyde has a distinctive aroma characteristic, which can cause an eau—de-
vie to have a poor aromatic characteristic at high concentrations.14 The production of
acetaldehyde should therefore be minimized if possible. The reaction mechanism that
describes the production of acetaldehyde is shown in Figure 2.3 (The Embden-Meyerhof-
Parnas fermentation scheme). In this scheme, acetaldehyde and carbon dioxide are the
last intermediates used to produce ethanol from glucose. Conversion of the carbon
dioxide and acetaldehyde to ethanol is not 100% efficient. This leaves some
acetaldehyde remaining in the mash. Despite the low boiling point of acetaldehyde it is
still seen in the distillate. This is due to the fact that acetaldehyde is completely soluble

in both water and ethanol, thus making it difficult to separate from the fruit spirits.

21

Figure 2.6 Pectin and the Activity of Pectinesterase to Produce
Methanol

Stucture of Pectin

 

 

OCH3
R
0:0 OH OH 0/
EW0 % o
0 c
C|\OCH3 |\OCH3
l0 Pectin 0
Mode of Action of Pectinesterase
E—N Represents the enzyme
H/O\
H
O 0.
Q I
R
R
CH3OH
0
II o- o
R—C—OH I
4—— E—N—c—OH<— E—N—C
E—N | |
R R

22

2.13 Ethyl Acetate and Ethyl F ormate

Ethyl acetate and ethyl fonnate are both esters. Esters add fruity aromas to the
fruit distillate. If the concentration of the esters is too high, the quality of the fruit spirits
will decrease. Esters are typically formed during the aging process. This study does not
age the fruit distillate so the ester content should be minimal. The reaction mechanism

used in the production of ethyl acetate is shown in Figure 2.7.

acetaldehyde Oxygen :0

acetic acid

 

O
HO/\ + 0 catalyst > O/\ + H/ \H

Fe, Cu

ethanol Oxygen acetaldehyde water
ethanol water
acetic acid ethyl acetate

Figure 2.7 The Production of Ethyl Acetate2
2.14 2-Furaldehyde

Heating the mash too quickly or using a direct flame can increase the production
of 2-furaldehyde. 2-furaldehyde has a burnt-bitter taste. 2-furaldehyde is therefore a
compound that has a negative effect on the flavor and aroma of the fruit spirit.

Distillation of the mash should be done in a methodical fashion to limit the production of

2-furaldehyde.

23

2.15 Benzaldehyde

Benzaldehyde has a bitter almond aroma and flavor that has a positive influence
on the fruit spirit. Benzaldehyde is formed from amygdalin found in the pits of fruit.

Under hot, mineral conditions amygdalin is hydrolyzed as Shown below.2

 

CHZOH
HO o 0 CH2 0 c
O |
CN
HO
OH
HO OH
OH
OH
Amygdalin
O

+ \ + H—CEN

 

2 glucose benzaldehyde hydrogen cyanide
Figure 2.8 The Production of Benzaldehyde by the Hydrolysis of Amygdalin2

The production of hydrogen cyanide presents a potential problem. Hydrogen
cyanide is apparently converted to urea.2 Urea then reacts with ethanol to form urethane
in the reaction mechanism shown in Figure 2.9. Urethane is a carcinogen; therefore,

benzaldehyde production should be carefully controlled.

24

NH2

H2N

o
+ /U\
HO/\ ——>/\o NH2 + NH

urea ethanol urethane ammonia
Figure 2.9 The Production of Urethane2

O 3

2.16 Fusel Alcohols

Fusel alcohols refer to any alcohol that contains more than two carbons. Excess
fusel alcohols can contribute an unpleasant, damp cloth aroma to fruit spirits. This
unpleasant aroma makes the minimization of the production of fusel alcohols a goal of
every fruit Spirit fermentation/distillation. Fusel alcohols analyzed in this study were 2-
butanol, 2-methyl-2-butanol, l-propanol, 2-methyl-2-propanol, l-butanol, 3-methyl-l-
butanol, l-pentanol, l-hexanol, l-heptanol, l-octanol, benzyl alcohol, 2-methyl-l-
butanol, and 2-methyl-l-propranol. The length of this list demonstrates that fusel
alcohols are the largest group of aromatic compounds in alcoholic beverages.15

There are two primary pathways in which fusel alcohols are produced. They are
catabolically produced from amino acids and anabolically produced from glucose (Figure
2.10). At the beginning of the century, Erlich Showed that several higher alcohols can be
formed from amino acids: he demonstrates the formation of 3-methyl-l-butanol from
leucine, 2-methyl-1-butanol from isoleucine, and isobutanol from valine.l Amino acids
are found to occur naturally in domestic fruit and are produced due to the anaerobic
nitrogen metabolism of yeast cells. The reaction method used to produce 3-methyl-1-

butanol is shown in Figure 2.11.

25

 

 

_o=s=m-_-_b_az-m

 

 

:~+ a =e=aEhem 3:32 .92....— ..8 93.55:.— ..e 9.352% SR 8:3...

canon—Sufism—
SO- »

Eon 20538682-le 0:623

8.0.
Eon Poamoofi

 

-.mxoncmu-u->xo€>:-d

 

.8395- 2.38.2-N

 

 

E a 0658.5

 

 

 

 

_o=s=m-_-.§oz-~

 

 

I? a

 

 

 

 

 

 

 

 

-Axon80-a$xo€m:-n Im+
o a o Eb: A E .6
8; \% 03528998. a g 2 A: g .2
._ > <00 _boo< NOD- oEoso—om—A NOU-VH
0 . . Pom .
Eon u_cu_m>0m_o.ov_-5 ore—SELEEéOBVTU
»
mam
Eon n :2? o_._o_m>Om_»xo€b—_Q-n.d
-Eazasxeuéaaa . w .
NOL + Bum
NOD + Eon o_.oa_o~oo<-d BESDAxEnEBoon
D W .
Eu< 22:»; H
5m 0: 5 2o -5
so... swim a. . a E v_
05.8th
330298305 %
3820 F 0538on
3:305 a

 

 

 

2..—33‘

26

 

 

 

 

 

 

-'_‘-‘H2
HO ? OH
NH2 . o o
. transaminase
leucrne glutamic acid
0
O O
ketoglutaric acid ketoisocaproic acid
0
HO
decarboxylase , /
0 -C02 0
ketoisocaproic acid isovaleraldehyde
/ reductase >HOA/k
0 +2H
isovaleraldehyde

3-methyl l-butanol
Figure 2.11 The Production of 3—Methyl-l-Butanol7

S. cerevisiae has the ability to synthesize all amino acids. Sugars such as glucose

and fi'uctose can be used to synthesize leucine.2 Leucine then enters the reaction

mechanism described above to produce 3-methyl-1-butanol. The production of 3-methyl-

l-butanol is anaerobic.

27

3. MATERIALS AND METHODS

3.1 Prefermentation Set-up

A suitable fermenter set-up was designed before fermentation could be initiated.
The major difficulty facing the project involved finding a method that would allow for a
large number of relatively small volume fermentations. To accomplish this thirty, 300-
mL Fleaker Beakers® were purchased. The Fleaker Beakers® came with an air tight
rubber cap. The rubber caps assured that the fermentations would be run in an anaerobic
environment. To facilitate the insertion of an air lock, one hole was drilled into each
rubber beaker cap. Each hole was just small enough to allow for an air lock to pass
through the rubber beaker cap. Each air lock was then partially filled with water. This
set-up allowed for each fermentation to be run in an anaerobic environment with the
release of carbon dioxide.

A chest freezer, thermocouple, and On/Off controller were used for temperature
control. Kenmore manufactured the chest freezer and its model name was Galaxy. The
model number was 253.19501992 and the serial number was WB10805488. The
thermocouple was manufactured by Omega. It was a K-type thermocouple and was
constructed of chromium and aluminum. The On/Off controller was manufactured by
Omega. It was called a Micromega Series CN7700 Controller.

The On/Off controller received power from a wall-mounted outlet. The On/Off
controller was placed outside the chest freezer. The thermocouple was connected to the
On/Off controller and the temperature probe was placed in a water bath inside the chest
freezer. A water bath was used to minimize the temperature fluctuation due to the

opening and closing of the freezer. The chest freezer received power from the On/Off

28

controller. By connecting the chest freezer and the On/Off controller in series to the wall
mounted power supply, the On/Off controller could control the power input to the chest
freezer based on temperature. If the temperature within the chest freezer was below the
set point, the On/Off controller would not allow power to enter the chest freezer.
Without power, the compressor was unable to run and the temperature within the chest
freezer rose. If the temperature within the chest freezer was above the set point, the
On/Off controller would allow power to enter the chest freezer. With power, the
compressor was able to run and the temperature within the chest freezer dropped. This
set up allowed for equilibration of the chest temperature control within a 01°C (Table
3.1).
3.2 Selection and Preparation of Fresh Fruit

Fruit was selected that was of optimal ripeness yet free of any visible blemishes
such as bruising, animal/insect bites, and mold. All stems, leaves, and/or vines were
removed from each piece of fruit. The fruit was then rinsed with water to remove any
residual soil matter that might have been on the surface of the fruit. The fruit was then
dried using toweling and cut into smaller pieces using a pairing knife. The fruit pieces
were placed in a five-gallon bucket and then crushed using a wooden pestle. This created
a mixture of juice and small pieces of fruit.

A 300-mL Fleaker Beaker® was placed on a mass balance and tared. 200 grams

of the previously mashed fruit tissue and juice were placed in a F leaker Beaker® using a
large weigh boat. The weigh boat was used in a Shovel like manner to scoop the juice

and fruit pieces into each Fleaker Beaker®. The mashed fruit tissue and juice in the five-

29

P9P!"

10.

ll.

12.

13.

Table 3.1 On/Off Micromega CN 7700 Controller Settings

Set Point 1=15 degrees

Set Point 2=15 degrees

Input Type= Thermocouple
Reading Configuration

A. Decimal Point=F F F .F

B. Temperature Unit=degrees Celsius
C. Filter Constant=0001
Alarm 1

Enable

Deviation

Unlatched

Normally Open

High-Low

Alarms Enable/Disable at Power On
Alarm 1 Low=.5

. Alarm 2 High=.5

Alarm 2

A. Disable

Loop Break Alarm

A. Disable

Output 1

A. Control Type=On/Off

B. Action Type=Direct Acting
C. Dead Band=000.l

Output 2

A. Control Type=On/Off

B. Action Type=Direct Acting
C. Dead Band=000.l

Ramp and Soak

A. Disable

Angle Out

A. Not Installed
Communication Option

A. Not Installed

Reset Point

A. Not Installed

meawpowe

30

gallon bucket was stirred using the weigh boat after removal of each 200 gram sample to
assure a consistent mash profile. This process was repeated until a total of fifteen F leaker
Beakers® were each filled with 200 grams of mashed fruit. The F leaker
Beakers® were then sealed with the aforementioned rubber cap/air lock set up. The
sealing of the F leaker Beakers® was done to prevent the invasion of external bacteria
while the yeast was prepared.
3.3 Yeast Preparation

Twenty mL of sterile water preheated to 40°C was added to 0.08 grams of yeast.
Each yeast/water mixture was placed in a 40°C water bath for fifteen minutes. While in
the heated water bath, the yeast/water mixture was stirred. After fifteen minutes the
yeast/water mixture was removed from the heated water bath. The yeast/water mixture
was allowed to cool at room temperature for fifieen minutes. Cooling of the yeast/water
mixture was done to prevent temperature shock of the yeast cells when added to the fruit
mash. This process was repeated with each of the four additional strains of yeast. Each
of the five yeast strains was done in triplicate for a total of fifieen fermentations per fruit
type.
3.4 Fermentation

Fermentation was initiated by adding the cooled yeast/water mixture to the fruit
mash contained within each Fleaker Beaker®. Each F leaker Beaker® received a separate
yeast/water sample. The fermentations were done in triplicate to improve the accuracy of
the results. The yeast/water mixture was added to the Fleaker Beaker® and stirred to
assure that the yeast/water mixture penetrated all areas of the fermentation mash. The

rubber cap/air lock was then firmly placed on the Fleaker Beaker® that was then placed

31

into the chest freezer. The chest freezer was held at 15°C using the On/Off controller and
the thermocouple. Each fermentation was allowed to run for a total of 14 days. Each
fermentation was determined to have been run to completion by the lack of production of
carbon dioxide. Carbon dioxide is a by-product in the conversion of glucose to ethanol.
When no carbon dioxide was seen bubbling out of the airlock, the fermentation had
reached completion.
3.5 Distillation

The entire fermented fruit mash was removed from the F leaker Beaker® and
placed into a 1000 mL round bottom flask. The flask was fitted into the Simple
distillation apparatus shown in Figure 3.1. The fermented mash was heated using a hot
plate/stirrer apparatus (Corning Model PC-420) set on a heat setting of four in
combination with a silicon oil bath which typically reached temperatures of around
115°C. This temperature was needed so that the vapor temperature in the simple
distillation column reached 99°C. At 99°C the distillation was stopped and the products
were collected from the SO-mL round bottom flask and placed in a 20-mL storage vile.
3.6 Distillate Analysis

All distillates were analyzed using gas chromatography. The analysis of the
distillates using gas chromatography was done within 48 hours of distillation to prevent
highly volatile compounds from evaporating before analysis. The run conditions for the
gas chromatograph are described in Table 3.2 and were used for all trials throughout the
study.

Standard retention times were generated for the following compounds:

acetaldehyde, acetone, ethyl formate, ethyl acetate, 2-methyl-2-propanol, methanol,

32

Figure 3.1 Simple Distillation Set-Up

 

   
  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Thermometer
g Inlet Adapter
W? Three-way
'" - Condenser
Adapter Vacuum Adapter
1% m ,
Clamp Egg; j\
__,_. 'Water Outlet Water Inlet I
1000 mL Flask

50 mL Flask

Ice Water Bafih

 

I Silicon on Bath 1

 

33

Column: 30 meters, 0.32 pm Stabilwax®
0.5 uL split injection

Oven temperature: 40°C (hold for one minute); to 190° @ 2.5°C/min.
190°C (hold for five minutes)

Injector temperature: 240°C

Detector temperature: 255°C

Carrier gas: Helium @ 30 cm/second

Split ratio: 65:1

Total GC rum time: 65 minutes

Table 3.2 Shimadzu GC-17A Run Conditions

34

ethanol, 2-methyl-2-butanol, 2-butanol, 1-propanol, 2-methyl-l-propanol, l-butanol, 3-
methyl-l-butanol, 2-methyl-1-butanol, l-pentanol, l-hexanol, l-heptanol, furaldehyde,
benzaldehyde, l-octanol, and benzyl alcohol. The retention times in triplicate were then
averaged to increase the accuracy of the analysis. Compounds that have low boiling
points will have relatively short retention times while compounds that have high boiling
points will have relatively long retention times.

The next phase of the experiment involved generating a calibration curve for
ethanol. The following concentrations of ethanol were prepared using 100% ethanol and
HPLC water: 25%, 27%, 30%, 33%, 35%, 38%, 40%, 43%, 45%, 47%, 50%, 53%, 55%,
57%, and 60%. Each of these concentrations was run five times using the
aforementioned gas chromatographic conditions. From this, the peak areas were
integrated and averaged. A calibration curve was generated by graphing the percent
ethanol versus the peak area. This curve allowed for the determination of the
concentration of ethanol and all other compounds within the distillate. Ethanol therefore

served as the internal standard.

35

4. RESULTS AND CONCLUSIONS

4.1 Qualitative Identification of the Compounds within Each Distillate

The first step in determining the influence that a particular yeast strain might have
on the taste of a fruit spirit is to determine which compounds make up the Spirit.
Compounds that have low boiling points have Shorter retention times in gas
chromatography than do compounds that have higher boiling points because compounds
with low boiling points vaporize at lower temperatures. Solubility can also impact the
retention time of a compound. Compounds that are soluble in ethanol will have shorter
retention times because they are vaporized along with ethanol. Ethanol has a relatively
low boiling point (78.3° C) compared to water (100° C). Compounds that are more
soluble in water than in ethanol will remain in a liquid state until the temperature within
the column reaches the boiling point of water. Compounds that are more soluble in water
than in ethanol therefore have increased retention times. Once vaporized, compounds are
swept through the column by the carrier gas to the detector.

Table 4.1 lists the compounds of interest that were identified within each fruit
spirit distillate along with their respective boiling points. This table illustrates that the
boiling point of a compound is not a precise measure of retention order. Compounds
with higher boiling points in some instances do have shorter retention times. This
supports the fact that compounds that are more soluble in ethanol than in water can have

shorter retention times than would be expected based on boiling point alone.

36

Acetone

Formate

Acetate

Methanol

Ethanol

-Butanol

2-Butanol

1

1

1-Butanol

1-Butanol

1-Butanol

1-Pentanoi

1-Hexanol

1

F

1-Octanol

Alcohol

94.0

97.0

107.7

118.0

132.0

128.0

138.0

156.5

176.0

161.7

179.0

194.5

205.0

Retention Time min

1.6

2.0

2.0

2.5

2.7

2.8

3.2

4.5

4.8

5.1

7.4

8.6

11.0

 

41.7

Table 4.1 Typical Fruit Distillate Components, Boiling Points, and Retention Times
Seen Experimentally Using a Shimadzu Gas Chromatograph and FID Detector

37

4.2 Ethanol Calibration

The main goal of every fermentation/distillation is to produce ethanol. Increased
ethanol production while maintaining product quality means an increased yield of fruit
distillate thus leading to increased income. It is therefore necessary to precisely quantify
the amount of ethanol in each distillation. Gas chromatography lends itself well to the
precise quantification of ethanol. The quantification of ethanol is directly related to the
integrated ethanol peak area in a gas chromatogram.

Fruit distillates can have a broad range of ethanol concentrations depending on
such parameters as the sugar content of the fruit, the length of fermentation, the quality of
the yeast, and the end temperature of a distillation. The potential range of ethanol
concentrations of fruit distillates makes it necessary to use a wide range of ethanol
concentrations when producing a calibration curve. Various ethanol concentrations were
made by diluting absolute (100%) ethanol with HPLC water. HPLC water was used as
the agent to dilute absolute ethanol to other concentrations because it has relatively few
impurities. The lack of carbon in HPLC water was important because the flame
ionization detector used in this study responds only to substances that contain a carbon
source. The HPLC water passes through the gas chromatography detector without
registering a peak. The ethanol calibration curve is shown in Figure 4.2. The range of
the ethanol calibration curve is between 25% and 60% ethanol. This range encompassed

all ethanol concentrations seen in the results that follow.

38

Bea—am E881 g:.o>

om

 

 

 

 

signage“. do 3... 825:8 .295 5. 23m

12er need ogudufiozeworqa see

Hit

39

4.3 Ethanol as an Internal Standard

Ethanol was used as the internal standard in the determination of the
concentration of all other flavor components in the fruit spirit distillates. Ethanol was
chosen as the internal standard because it is present in the highest concentration next to
water and all brandies must be normalized to ethanol content.

The integrated area of all peaks within a single chromatogram produced from a
single fruit distillate were integrated and added. Individual peak areas were then divided
by the sum of all peak areas to give an area percent for each peak. The volume percent
composition of ethanol was determined using the ethanol calibration curve shown in
Figure 4.2. By comparing ratios of the unknown peak volume percent composition
divided by the unknown peak area percent to the ethanol volume percent composition
divided by the ethanol peak area percent the unknown volume percent composition was
calculated for each peak. The final calculation involved converting each volume percent
composition to a 40% ethanol solution by multiplying by forty divided by the actual
ethanol volume percent composition found from the calibration curve. The end result is
that each chromatographic peak represents a volume percent composition of a given
flavor component in a 40% ethanol solution. These calculations standardized each
distillate to a 40% ethanol solution and allowed for the comparison of the production of
flavor components between all five varieties of yeast used in this study.

4.4 Ranking of the Yeasts

An objective method of determining which yeast was the best for a given variety

of fruit was developed. Each yeast was compared to all other yeasts and assigned an

alphabetical ranking using the Duncan Range Test Method for each congener category

40

identified using gas chromatography. The letter “a” represented the best possible
performance in a specific congener category. The letter “e” represented the worst
possible performance in a specific congener category. Compounds could have the same
rank if they were determined to be statistically similar by the Duncan Range Test
Method. The alphabetical ranks were then converted to numerical ranks (a=1, ab=l .5,
b=2, bc=2.5, c=3, cd=3.5, d=4, de=4.5, and e=5). The production of ethanol is the
primary goal of a fruit fermentation/distillation. The yeast that produced the most
amount of ethanol was given an “a” ranking. Yeasts that produced less ethanol were
given lower rankings. Methanol is a positive flavor component. The yeast that produced
the most methanol was given an “a” ranking. Yeasts that produced less methanol were
given lower rankings. Benzaldehyde imparts a positive bitter almond aroma and taste
into a fruit distillate. The yeast that produced the most benzaldehyde was given a ranking
of “a” and yeasts that produced less benzaldehyde were given lower rankings. Acetone,
acetaldehyde, and fusel alcohols all impart negative aroma and flavor characteristics into
a fruit distillate. In each of these categories the yeast that produced the smallest amount
of each of these congeners was given a ranking of “a” with lower rankings being assigned
to yeasts that produced more of each of these congeners. After rankings were assigned
for each congener category, an overall average ranking was generated for each yeast.

The yeast with the lowest average ranking was determined to be the yeast that performed

best for a specific variety of fruit.

41

 

4.5

Bartlett Pears

Table 4.2 shows the results of fermenting and distilling bartlett pears. Figure 4.2

shows the production of ethanol and methanol along with the corresponding standard

deviation in the three trials used to generate averages. Ethanol, methanol, acetaldehyde,

acetone and a variety of fusel alcohols were identified in the bartlett pear distillates using

gas chromatography. The fusel alcohols found in bartlett pear distillates included 2-

methyl-Z-propanol, l-propanol, 2-methyl-1-propanol, l-butanol, 3-methyl-l-butanol, l-

hexanol, and l-hetanaol. These fusel alcohols were summed together to generate a single

fusel alcohol category.

Table 4.2 Bartlett Pear Distillate Composition and Rank

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Yeast Strain CY3079 0254 EC1118 L2056 SGU
40% Ethanol mL 19.01.737 18.71.306 18.511.25 17,111.21 18.511.10
Duncan Test a ab ab b ab
Ethanol Rank 1 1.5 1.5 2 1.5
Methanol %VN .0506100261 .01 13100984 .25110324 .225100799 .31510237
Duncan Test c c ab b a
Methanol Rank 3 3 1.5 2 1
Fusel Alcohol %VN .64411 17 .60210127 .49910541 .50710244 .590101 10
Duncan Test 0 b a a b
Fusel Alcohol Rank 3 2 1 1 2
Acetaldehyde %VN 0171100347 .01421000960 .0155100226 0131100171 .0230100340
Duncan Test d b c a e
Acetaldehyde Rank 4 2 3 1 5
Acetone %VN .00008001 .00003001 .001111 .0009801 0005201

000140 0000600 00026 0000500 000310
Duncan Test b a e d c
Acetone Rank 2 1 5 4 3
Avegge Rank 2.60 1.90 2.40 2.00 2.50
Overall Rank 5 1 3 2 4

 

 

 

 

 

 

42

 

uonmos
Iouema %ov a u: Iouetnew °/. ewnIOA

Dow 000m;

525 68>
8 P 8m

END

0300>0

 

oomad
memos ,1

Ema; -. M 1

1

F0-wm.w

5mg -- 4....

 

 

n+1

l

813 -

.1

 

 

_.0-w0.m L

 

 

 

 

T
‘3 ‘ ,.\fi“ 3.”

.1

 

 

 

 

F0-wm.m
_

_
1 .

:8262 n;

6.55.”. a 1

 

 

r "l‘ullL

 

‘r_s.'.¢

 

 

 

 

 

 

 

 

 

 

 

_.0-mo.v

«cor—«.5350 .2552: new Begum 22.300 Eon. Ext—mm «6 2:9".

0N

06

0.0

0.0

- 0.0V

0.N—.

- 0.3.

0.0V

0.0..

. 0.0m

0.NN

Iouetna %017 "ll”

43

 

4.6 Black Plums

Table 4.3 shows the results of fermenting and distilling black plums. Figure 4.3

shows the production of ethanol and methanol along with the corresponding standard

deviation in the three trials used to generate averages. Ethanol, methanol, acetaldehyde,

and a variety of fusel alcohols were identified in the black plum distillates using gas

chromatography. The fusel alcohols found in black plum distillates included 2-methyl-2-

propanol, l-propanol, 2-methyl-l-propanol, 3-methyl-1-butanol, and l-hexanol. These

fusel alcohols were summed together to generate a single fusel alcohol category.

Table 4.3 Black Plum Distillate Composition and Rank

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Yeast Strain CY3079 0254 EC1 1 1 8 L2056 SBU
40% Ethanol mL 19.411.73 19.01916 18.911.34 19.21.0823 17.51.638
Duncan Test a ab ab ab b
Ethanol Rank 1.0 1.5 1.5 1.5 2.0
Methanol %VN 0346100488 0950100635 .25210661 .36110500 .14010143
Duncan Test d c b a c
Methanol Rank 4 3 2 1 3
Fusel Alcohol %VN 64810737 .59510435 .62710562 .5081254 .48210308
Duncan Test e c d b a
Fusel Alcohol Rank 5 3 4 2 1
Acetaldehyde %VN 006991 .005171 006981 .01 131 03361

00201 .00131 000610 .000610 00128
Duncan Test b a b c d
Acetaldehyde Rank 2 1 2 3 4
Average Rank 3.00 2.13 2.38 1.88 2.50
Overall Rank 5 2 3 1 4

 

 

 

 

 

 

44

 

uonmos
Iouema %ov e U! Iouemew % ewnIOA

 

Egm umao>

 

 

 

 

 
 

 

 

 

:8 183 2:8 ES
8.03 1 . on
8.3m -
603 -- o a
503 --
qfi w
amaze. -- w
%
a u
503 : m
qt m
3.QO .
5mg -, - emu
Ewes -, Egan
1. 13E;
5&3. QR

aces—«5:50:00 .0552: new 3.35m. 35.500 E:_n_ zoo—m «.6 0.50.".

4.7

Braeburn Apples

Table 4.4 shows the results of fermenting and distilling braebum apples. Figure

4.4 shows the production of ethanol and methanol along with the corresponding standard

deviation in the three trials used to generate averages. Ethanol, methanol, acetaldehyde,

acetone and a variety of fusel alcohols were identified in the braebum apple distillates

using gas chromatography. The fusel alcohols found in braebum apple distillates

included 2-methyl-2-propanol, l-propanol, l-butanol, 3-methyl-1-butanol, and l-hexanol.

These fusel alcohols were summed together to generate a single fiisel alcohol category.

Table 4.4 Braeburn Apple Distillate Composition and Rank

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Yeast Strain CY3079 0254 EC11 18 L2056 SGU
40% Ethanol mL 32217.67 22017.83 23.91130 22,613.65 27.517.32
Duncan Test a b b b ab
Ethanol Rank 1 2 2 2 1.5
Methanol %VN 064010277 0420100270 0440100322 0480100321 .3681265
Duncan Test b b b b a
Methanol Rank 2 2 2 2 1
Fusel Alcohol %VN .80710687 .63510404 .7441154 .61510234 .8741164
Duncan Test 0 a b a d
Fusel Alcohol Rank 3 1 2 1 4
Acetaldehyde %VN 021010127 00600100278 0130100135 00500100338 0190100554
Duncan Test e b c a d
Acetaldehyde Rank 5 2 3 1 4
Acetone %VN 1.591.470 .5131286 .0001000 .0110100698 .3211556
Duncan Test b ab a a ab
Acetone Rank 2 1.5 1 1 1.5
Average Rank 2.60 1.70 2.00 1.40 2.40
Overall Rank 5 2 3 1 4

 

 

 

 

 

 

46

 

mlos

UOI

£25 «53>

 

 

 

 

 

 

 

 

 

now 183 S F 6m ES @396

8+mod as
A 3.1: . . , 99
w 1
n
w
w Ema.” - qt
o/
w
m.
w 813 . QNN
u
W l
m. 313. QR
.7
o
%
a 519m -, .92
..u..
e
w . - 1 - .
I F0-mo 0 . .mcmfilofifli .. 0 km

- , 685mm
smog owe

mcozmbcoocoo Eco—to: ucm 3.85m 35:35 2.34. 5.535 1v 230E

Ioueuia %O‘P 'l'»u

47

4.8 Gala Apples

Table 4.5 shows the results of fermenting and distilling gala apples. Figure 4.5
shows the production of ethanol and methanol along with the corresponding standard
deviation in the three trials used to generate averages. Ethanol, methanol, acetaldehyde,
acetone and a variety of fusel alcohols were identified in the gala apple distillates using
gas chromatography. The fusel alcohols found in gala apple distillates included 2-
methyl-Z-propanol, l-propanol, 2-methyl-1-propanol, 3-methyl-l-butanol, 2-methyl-l-
butanol, and l-hexanol. These fusel alcohols were summed together to generate a single

fusel alcohol category.

Table 4.5 Gala Apple Distillate Composition and Rank

 

 

 

 

 

 

 

 

Yeast Strain CY3079 0254 EC11 18 L2056 SGU
40% Ethanol mL 20.814.63 22.316.81 22.81634 19.514.15 22.612.39
Duncan Test a a a a a
Ethanol Rank 1 1 1 1 1
Methanol %VN 065110106 05311000830 0406100292 046410106 080610121
Duncan Test b c e d a
Methanol Rank 2 3 5 4 1

 

 

F usel Alcohol %VN .8391. 160 1.061.157 65610644 9791.201 6861.139

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Duncan Test b c a c a
Fusel Alcohol Rank 2 3 1 3 1
Acetaldehyde %VN 008911 008811 007641 01041 007621
000400 00104 .00136 00274 000650
Duncan Test d c b e a
Acetaldehyde Rank 4 3 2 5 1
Acetone %VN .001651 001211 0003701 0005301 0006101
000190 000630 000330 000200 000220
Duncan Test e d a b c
Acetone Rank 5 4 1 2 3
Average Rank 2.80 2.80 2.00 3.00 1.40
Overall Rank 3 3 2 5 1

 

 

 

 

 

 

 

48

uonmos
Ioueuaa %ov a U! Iouetnew % ewnIOA

seam 38>
8m 88.. 8:8 88 2°86

 

Nomad . . . ..

 

._. .fi w. _. .1...” w. 1.... a. .
memos: . _ .., a M U

 

 

T

mom—0.0 .-

 

 

 

memos

 

 

 

 

 

No-03 .- .1 .1. r1

 

 

 

 

 

 

 

 

 

 

 

 

Nomad _ 1.

 

 

N0-m_0.0 .

 

1-655820
_ sesame

l
g
l

 

  

 

F0-mo.w

 

0N

1 0K

- 0.N_.

1 0.x...

-, 0.NN

I 0.5m

 

 

mcozabcoocou .0552: new .225 3.2.sz o_aa< ~30 mi 0.52...

0N0

Iouetna %07 ll“

49

 

4.9

Granny Smith Apples

Table 4.6 shows the results of fermenting and distilling granny smith apples.

Figure 4.6 shows the production of ethanol and methanol along with the corresponding

standard deviation in the three trials used to generate averages. Ethanol, methanol,

acetaldehyde, acetone, benzaldehyde, and a variety of fusel alcohols were identified in

the granny smith apple distillates using gas chromatography. The fusel alcohols found in

granny smith apple distillates included 2-methyl-2-propanol, l-propanol, 3-methyl-l-

butanol, l-hexanol, and 2-butanol. These fusel alcohols were summed together to

generate a single fusel alcohol category.

Table 4.6 Granny Smith Distillate Composition and Rank

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Yeast Strain CY3079 0254 EC1 1 18 L2056 SGU
40% Ethanol mL 16.214.69 23.114.19 23.117.82 20.811.28 19.212.66
Duncan Test a a a a a
Ethanol Rank 1 1 1 1 1
Methanol %VN .137100278 .30610524 .26010513 .211100536 .53010495
Duncan Test d b be c a
Methanol Rank 4 2 2.5 3 1
Fusel Alcohol %VN 0831140 .8571. 168 .8871299 22212.15 .8671. 163
Duncan Test a a a b a
Fusel Alcohol Rank 1 1 1 2 1
Benzaldehyde %VN .00170100248 .0001000 .002041000 .0001000 .0001000
Duncan Test a a a a a
Benzaldehyde Rank 1 1 1 1 1
Acetaldehyde %VN 042010526 0148100556 0187100382 .032910213 0200100420
Duncan Test e a b d c
Acetaldehyde Rank 5 1 2 4 3
Acetone %VN 0005801 0001201 0001801 0002801 0003001

000180 00021 000210 000130 0000500
Duncan Test e a b c d
Acetone Rank 5 1 2 3 4
Avegge Rank 2.83 1.17 1.58 2.33 1.83
Overall Rank 5 1 2 4 3

 

 

 

 

 

 

50

 

uonnlos

! Iouemaw % ewnIOA

Iouema %0? e u

£05m “93>
000mg 0 v F wow

0N00>0

 

_.0-w0._.

_.0-mo.N :

r0-w0.m l

v0-M0.¢ -

F0-M0.m 1

F0-w0.0 -r

F0-MON i.

 

 

 

 

11191.26: 0
1. ficmfimifl

  

    

 

51

IOUBHE %07 "ll"

 
 

. 0.5m

 

 

F0-w0.0

:ozobcoocoo Eco—=22 can .2.—Em 22.000 .EEw >555 04. 230E

 

4.10 Green Seedless Grapes

Table 4.7 shows the results of fermenting and distilling green seedless grapes.

Figure 4.7 shows the production of ethanol and methanol along with the corresponding

standard deviation in the three trials used to generate averages. Ethanol, methanol,

acetaldehyde, acetone and a variety of fusel alcohols were identified in the green seedless

grape distillates using gas chromatography. The fusel alcohols found in green seedless

grape distillates included 2-methyl-2-propanol, l-propanol, 2-methyl-l-propanol, 3-

methyl-l-butanol, 2-methyl-1-butanol, and l-hexanol. These fusel alcohols were

summed together to generate a single fusel alcohol category.

Table 4.7 Green Seedless Grape Distillate Composition and Rank

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Yeast Strain CY3079 0254 EC1118 L2056 $6U
40% Ethanol mL 45.511 .31 43.71.327 47,714.12 45.81262 42012.67
Duncan Test a a a a a
Ethanol Rank 1 1 1 1 1
Methanol %VN 0353100112 03571000860 0465100163 0555100225 077210130
Duncan Test (1 d c b a
Methanol Rank 4 4 3 2 1
Fusel Alcohol %VN .41910145 .41610129 .56210582 .50610245 .42110165
Duncan Test a a c b a
Fusel Alcohol Rank 1 1 3 2 1
Acetaldehyde %VN 0120100123 00755100035 041710250 0280100151 0365100127
Duncan Test b a e c d
Acetaldehyde Rank 2 1 5 3 4
Acetone %VN .0001000 0003201 .0001000 .0001000 .0001000

000320
Duncan Test a b a a a
Acetone Rank 1 2 1 1 1
Average Rank 1.80 1.80 2.60 1.80 1.60
Overall Rank 2 2 5 2 1

 

 

 

 

 

 

52

 

ewnloA

uogmos

loam-113 %07 3 u! loumflallll %

swam 63>

Dow @003 0r _. vow #mmo 0500>0
oo+mo.o . ...J..-w . .... . ,. a . q 1 1- .., 1 . .-M. O.N

 

 

N000 N .- ,1 - 0N?

 

 

 

H

803 .,

 

- 0.NN

 

J;‘,".‘.‘..t.et
.,, .

 

 

n .. .
.1 . .. .
e U 1... .
. .1 . .
.1 H, v. .1 .. . .. . ,. h ,
u . . . .. .. . 1
. .. 1 . _. .. . .. 0 N0
. . . . r. . . .. . .
.. . .

 

 

N0-m0.0 ..

 

 

 

- 0N»q

 

 

 

 

 

 

. .rvp-qv ,-u )1Al'31 .‘. . . ,
l

 
 

 

 

803 r... L.

@135 L
1 .0005wfi1

 

. 0N0

 

 

 

rommé

coach—.8200 .0552: ucm .0525 30.300 2.20 59.0 he 2:9”.

53

musing %07 101

 

4.11

Montmorency Cherries

Table 4.8 shows the results of fermenting and distilling montmorency cherries.

Figure 4.8 shows the production of ethanol and methanol along with the corresponding

standard deviation in the three trials used to generate averages. Ethanol, methanol,

acetaldehyde, acetone, benzaldehyde, and a variety of fusel alcohols were identified in

the montmorency cherry distillates using gas chromatography. The fusel alcohols found

in montmorency cherry distillates included 2-methyl-2-propanol, l-propanol, 3-methyl-l-

butanol, l-hexanol, benzyl alcohol, and 2-butanol. These fusel alcohols were summed

together to generate a single firsel alcohol category.

Table 4.8 Montmorency Cherry Distillate Composition and Rank

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Yeast Strain CY3079 0254 EC1 1 18 L2056 SBU
40% Ethanol mL 17.315.78 21017.54 20.211.56 18912.44 13011.91
Duncan Test a a a a a
Ethanol Rank 1 1 1 1 1
Methanol %VN .33710830 .32910430 .295100319 .30110227 .34610563
Duncan Test b c e d a
Methanol Rank 2 3 5 4 1
Fusel Alcohol %VN .8101157 .65711 12 .6601133 .52910960 .726100319
Duncan Test d b b a c
Fusel Alcohol Rank 4 2 2 1 3
Benzaldehyde %VN 0495100832 0350100431 0389100519 019810177 0548100525
Duncan Test b d c e a
Benzaldehyde Rank 2 4 3 5 1
Acetaldehyde %VN 00706100407 00792100123 00701100128 0198100549 0279100893
Duncan Test a b a c d
Acetaldehyde Rank 1 2 1 3 4
Acetone %VN .4181409 3.821308 0001105 .4111712 652.4401
Duncan Test a a a a a
Acetone Rank 1 1 1 1 1
Average Rank 1.83 2.17 2.17 2.50 1.83
Overall Rank 1 3 3 5 1

 

 

 

54

 

on

Figure 4.8 Montmorency Cherry Distillate Ethanol and Methanol

uonnlos

Iouema %ov a U! Iouetnew % ewnIOA

 

 

 

 

 

555555555
uJUJLuu'JuJuJuJuJuJ
Nqochwtmqoqo.
VV'COMCOOOMNN

-—8

52

0.1%

an

 

 

 

 

 

 

 

 

 

26.0

21.0 ,

o C».
1

Ioueuza %0? 1‘"

55

6.0

L2056

EC1118
Yeast Strain

D254

CY3079

 

 

4.12 New Haven Peaches

Table 4.9 shows the results of fermenting and distilling new haven peaches.

Figure 4.9 shows the production of ethanol and methanol along with the corresponding

standard deviation in the three trials used to generate averages. Ethanol, methanol,

acetaldehyde, acetone and a variety of fusel alcohols were identified in the new haven

peach distillates using gas chromatography. The fusel alcohols found in new haven

peach distillates included 2-methyl-2-propanol, l-propanol, 2-methyl-1-butanol, 3-

methyl-l-butanol, benzyl alcohol, and l-hexanol. These fusel alcohols were summed

together to generate a single fusel alcohol category.

Table 4.9 New Haven Peach Distillate Composition and Rank

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Yeast Strain CY3079 0254 E01118 L2056 SGU
40% Ethanol mL 16312.73 15.711.85 16.213.68 22011.35 21.51304
Duncan Test a a a a a
Ethanol Rank 1 1 1 1 1
Methanol %VN 073710176 .63010591 .4721212 .19710532 .32810673
Duncan Test d a b cd bc
Methanol Rank 4 1 2 3.5 2.5
Fusel Alcohol %VN .5961. 165 .39110600 .6911124 .43710433 .5081192
Duncan Test 0 a d a b
Fusel Alcohol Rank 3 1 4 1 2
Acetaldehyde %VN 0338100513 0369100628 056610304 0463100666 045810149
Duncan Test a b e d c
Acetaldehyde Rank 1 2 5 4 3
Acetone %VN .5901747 .0001000 210013.63 .0001000 08411145
Duncan Test a a a a a
Acetone Rank 1 1 1 1 1
Average Rank 2.00 1.20 2.60 2.10 1.90
Overall Rank 3 1 5 4 2

 

 

 

 

 

 

56

 

uonnlos
louema %07 e u! louemew % eumloA

£95m unu0>

 

 

 

 

 
 

 

 

 

 

88.. 9:8 $8 285
803 . . . L ed
803 --
.L ox
503. :
90
503 -,
503 -. 92
8.03 -, 11:11
1.8212201 - QNN
5......
803

scrub—.0230 .0552: can .2.—Em 32:65 semen. :05: 302 mi 230E

Iouetna %07 1'“

57

 

4.13 Red Seedless Grapes

Table 4.10 shows the results of fermenting and distilling red seedless grapes.
Figure 4.10 shows the production of ethanol and methanol along with the corresponding
standard deviation in the three trials used to generate averages. Ethanol, methanol,
acetaldehyde, acetone and a variety of fusel alcohols were identified in the red seedless
grape distillates using gas chromatography. The fusel alcohols found in red seedless
grape distillates included 2-methyl-2-propanol, l-propanol, 2-methyl-1-propanol, 3-
methyl-l-butanol, and l-heptanol. These fusel alcohols were summed together to

generate a single fusel alcohol category.

Table 4.10 Red Seedless Grape Distillate Composition and Rank

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Yeast Strain CY3079 0254 E01118 L2056 SBU
40% Ethanol mL 49.214.27 47.312.16 50.61432 49.01.707 42.411.92
Duncan Test a a a a a
Ethanol Rank 1 1 1 1 1
Methanol %VN 0352100028 0595100675 0565100394 0697100997 .12610144
Duncan Test (1 c c b a
Methanol Rank 4 3 3 2 1
Fusel Alcohol %VN .45710338 .47110160 .405100776 .55610261 .47410551
Duncan Test b c a d c
Fusel Alcohol Rank 2 3 1 4 3
Acetaldehyde %VN .01391 .009131 .01391 00003001 .08181 00500

.0107 00621 000280 000055

Duncan Test c b c a d
Acetaldehyde Rank 3 2 3 1 4
Acetone %VN 00006001 00007001 00006001 0001501 0001601

000110 .000110 000100 000142 000216
Duncan Test a b a c d
Acetone Rank 1 2 1 3 4
Average Rank 2.20 2.20 1.80 2.20 2.60
Overall Rank 2 2 1 2 5

 

 

 

 

 

 

 

58

uonmos

IOUBIIE %07 3 u

% ewnloA

! Iouetnew

Eabm “93>
Dow 0003 0.. _. row 3ND 0500>o

 

 

803
Nomad -.
803 --
No.03 -.
No-03
303 --

F0-MN.—. %

 

F0-w¢.w 1

  
 

Pom—0.? .,

F000... ., __ocm£o_>_g
,1 35%.01

 

 

 

 

woflod

coast—50:00 .0552: 05.... .0525 32.300 320 com 3.6 230E

0N

0.N_.

0&0

0.N¢

0N0

Iouema %017 110

59

5. COMPUTER SIMULATION

5.1 ChemCAD Batch

ChemCAD Batch® is a type of batch distillation software marketed by
Chemstations. Common applications of this software include modeling existing batch
column equipment, exploring alternative processes for existing products, design of new
equipment, and validation of thermodynamics using batch column equipment. In this
study ChemCAD Batch® was used to model the distillation of a fruit mash using the
experimentally determined congener concentrations. By simulating a fruit mash
distillation using ChemCAD Batch® it was possible to produce the profile of each
congener as the fruit distillate was collected. Congener profiles provide a quantitative
guide to the quality of the fruit spirit being produced. These profiles are useful in
predicting when to make volumetric cuts of a fruit distillation.

The experimentally determined percent volume/volume composition of the
CY3079 Montmorency cherry distillate was as follows: acetaldehyde (.00706), acetone
(.418), 2-methyl-2-propanol (.00773), methanol (.337), ethanol (40.0), l-propanol (.369),
3-methyl-1-butanol (.432), l-hexanol (.000770), and benzaldehyde (.0495). It was then
assumed that a Montmorency cherry mash has an ethanol percent volume/volume
composition of roughly eight percent. Based on this estimate, the other congener percent
volume/volume compositions were calculated for the Montmorency cherry mash. The
estimated congener percent volume/volume compositions in the Montmorency cherry
mash were as follows: acetaldehyde (.00141), acetone (.0837), 2-methyl-2-propanol

(.00155), methanol (.0830), l-propanol (.0738), 3-methyl-l-butanol (.0864), l-hexanol

60

(.000154), and benzaldehyde (.00989). The remainder of the mash composition was
assumed to be water and had a percent volume/volume composition of 91.7.

The estimated mash composition values were used in the ChemCAD Batch®
software as the pot charge. The mash was assumed to be at standard temperature and
pressure (25°C and 1 atm). The column specifications included selecting a four stage
column with a partial condenser followed by a complete condenser at a pressure of one
atmosphere. The operational specifications included selecting a reflux ratio of 1.5, a
distillate rate of .333 L/min, and a stopping criterion when the ethanol liquid volume
fraction reached 0.50. The display specifications included selecting units of liquid
volume fraction versus time in hours.

Table 5.1 outlines the % volume/volume of acetaldehyde, acetone, 2-methyl-2-
propanol, methanol, ethanol, l-propanol, 3-methyl-1-butanol, l-hexanol, and
benzaldehyde in each cut. Acetaldehyde, acetone, and 2-methyl-2-propanol are seen in
the highest percent volume/volume concentrations in the initial distillate or heads cuts
because of their relatively low boiling points. The first three cuts are termed the head
cuts and are not usually used for consumption. Methanol initially has its highest percent
volume/volume concentration and then decreases rapidly. It then increases somewhat
towards the middle or heart distillate cuts and then once again decreases. Cuts number
four through fourteen are designated as hearts cuts and are used for consumption. The
methanol percent volume/volume concentration versus the distillate volume is shown in
Figure 5.1. Methanol’s profile can be attributed to its complete solubility in both ethanol
and water. Ethanol (Figure 5.2), l-propanol, and 3-methyl-1-butanol have a large initial

increase and then gradually decrease in percent volume/volume. This decline can be

61

attributed to the increased concentration of water in the distillate as the temperature
within the partial condenser rises closer to the boiling point of water. Finally, l-hexanol
and benzaldehyde have a gradual increase in percent volume/volume as the temperature

inside the partial condenser rises closer to their respective boiling points.

62

Ana?» a: 5.23 $3.90 2: “En: 685.50.. 9.53 8:..vo zones—5.35 2:. 63.5.«8 acts—=55.
£8.3— G<UE£U win: 3 nous—=38 notch—59:3 Snow—.3 35:25 Page tho-5.8.5.52 E 555...; 2:. fim 935.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

swamd moflowe vo-m_:.v vo-wfin.m mo-moo.© rommmh wommwé SQQNM Tom 93 _. VF
Fo-m_mm.w moflmwd vo-mvm.m vomwmm momvmwn womvvé ro-m_mo.N Fo-mvm.m Adm m. I. _. up
Fomvmé moflmfim vow—05m vomomd mo-moo.m Fommww womNmN remand m.mm m.o—. F Np
romwoh mom—09m vomvvd vommvd woman: FOMMNN aflood am Fan nNm omd _. :.
_.o-w$\® mommNN vo-wom.m vow—mod NOMKNé Sflmmx rOMmNh Fomwmfi wdm omd r cw
rommNN mommfim vomomd vomvfiw Nomnvé vow—MKS wo-m\.©.o Pom—mad w. F0 095 w m
Powwow mom—mme vow KN voflove No-mm®._. Fomwmk eo-mvm.w Fommmd chm om.w _. a
Fomvww No-wmn.r VOMNVN smear No-wvme EMS—mm oo+wwoe Fommmd Two omh w 5
rowQO Nomomd vomwNN momnmd No-mow.N Sflmfw oo+mMN._. Powwow Ewo 09v F u
oo+wmo._. No-m_mm.N vommme mo..wmm.o Nomtud rem—w F .m com—owe rommmd owm omd e m
oo+m_m_..w NO-Mw©.m vomvne moflofw NommKN rem—mod oo+mvw._. woman Nam omN _. v
003”..me No-m SUV vomtbe modmmd Nomomd Fo-m_mm.n oo+wmoN wommwd m.mw owe md n
oo+wmwe Newmfiv vomov... momowd ~9va .m wommwfi oo+MmNN wo-wnw.m ”on 009 md N
remand Novaw momoox mom: .r No-mNN.v ro-wmm.m Fo+mmve romw—mv Qmw oomd m6 r

_o>=o>.x. _o>=o>°\a _o>=o>.x. 32.023 233 235

32.023 2122...” _o>=o>..\o 2932.3 _o:mn2m-u 353-5- F 32.33 _o>=o>.x. _o>=o>.x. 0E:_o> «E:_o> 30:52

3:82.. a8< .888: .28 _E.os-~ 3522-” 2683 .8235. .255 9:53 So So

 

 

 

63

3.

.283. oE=_o> 82:35
2 NF 3 2 m m n o m. w m N F

 

 

mnmd

.- ommd

.- mwmd

. ommd

. mmmd

.- oovd

. movd

I 036

 

 

53:55 seam 98:55
9.3: mack. 35522 32:35 >55 353.55: 2.35 9.... 2:9".

mrvd

|oueqtew wanted ewnloA

64

3.

2

3.8:. 35:5 32.39
NF S 2 m m N o m w m N F

 

 

99V

I 0.0m

7 9mm

.- Odo

.- ohm

- odn

 

 

5:236 seam 9852.0
9.1.5 mack. .225 32:65 2:28 55.2582 285 3 2:9“.

. own

|oueq13 wasted awmoA

65

6. SUMMARY AND CONCLUSIONS

The fermentation and distillation of fruit beverages has been in practice since at
least 3000 BC. Despite this extensive history, the process of fermenting and distilling
and its effect on the resulting product has not been fully understood. It was the goal of
this research to develop some insight into the effects of altering the strain of yeast used to
ferment and distill fresh fruit to produce brandy. By determining the effects of altering
yeast strains, these data could then be used to improve the process and profitability of the
Michigan fruit brandy industry.

This research has proven that yeast strains directly influence the congener
concentration of a fruit brandy. The congeners identified using gas chromatography were
ethanol, methanol, fusel alcohols, benzaldehyde, acetaldehyde, and acetone. When
comparing the performance of different yeast strains for the same variety of fruit, it can
be seen that different congener concentrations were produced. These fluctuations in
congener concentration were used to compare the performance of five different strains of
yeasts for the same fruit. The method of ranking the yeasts was done in an objective
manner. Yeasts were ranked using the Duncan Range Test Method with one being the
best ranking and five being the worst ranking. An overall average was then generated to
compare the performance of different strains of yeast for a single variety of fruit. The
nine varieties of fruit used in this research included Bartlett Pears, Black Plums, Braeburn
Apples, Gala Apples, Granny Smith Apples, Green Seedless Grapes, Montmorency
Cherries, New Haven Peaches, and Red Seedless Grapes.

Table 6.1 shows the yeast that was determined to be the best to ferment and distill

a specific variety of fruit. This figure shows that of the five yeast strains studied all five

66

were determined to be the best performing yeast for at least one variety of fruit. This

indicates the importance of the yeast strain selected to be used to ferment and distill fruit

brandy. The intent of this research was accomplished by providing distillers in the state

of Michigan a guide to yeast selection that will allow them to increase the amount and

quality of fruit brandy that they produce.

 

 

 

 

 

 

 

 

 

 

 

 

Variety of Fruit Best Performing Yeast
Bartlett Pears D254
Black Plums L2056
Braeburn Apples L2056
Gala Apples SSU
Granny Smith Apples 0254
Green Seedless Grapes SGU
Montmorency Cherries CY3079 & SGU
New Haven Peaches 0254
Red Seedless Grapes EC1118

 

Table 6.1 The Best Performing Yeast Determined by Ranking the Congener
Composition of Each Fruit Distillate.

67

 

7. FUTURE WORK

The research in this study focused on the fermentation of nine varieties of fruit
with five different commercial strains of yeast. All fermentations within this study were
done at 15°C. Future work could be done at different temperatures to see if the yeast
performance is altered by temperature changes. Work of this nature would provide an
optimal yeast and temperature combination to ferment and distill particular fruit varieties.
Additional work could also be done on other varieties of fruit and yeast that were not
studied in this research.

Other future work could involve using alternative analytical methods when
determining the production of congener compounds. The research in this study focused
on the major congener categories (ethanol methanol, fusel alcohol, benzaldehyde,
acetone, and acetaldehyde). By using analytical techniques such as gas chromatography-
mass spectroscopy instead of gas chromatography, congeners in smaller concentrations

could be detected and quantified.

68

10.

11.

12.

13.

BIBLIOGRAPHY

. Ayrapaa, T. On the formation of higher alohols by yeasts and its dependence on

nitrogenous nutrients. Soensk Kemish Tidslaift. Vol. 83: pp. 79-90 (1971).

Berglund, K. CEM 924. Fall Semester Class Notes. Department of Chemistry,
Michigan State University, East Lansing, MI (1998).

. Claus, M. An Investigation of Relationship Between Tray Usage of the Still and

Congener Compound Concentration in Distilled Fruit Spirits. Department of
Chemistry, Michigan State University, East Lansing, MI (2001).

Cleland, J .G. and Kingsbury, G.L. Multimedia Environmental Goals for
Environmental Assessment. US. Environmental Protection Agency: EPA-600/7-77-
1366, November (1977).

Guymon, J .F . Higher alcohols in beverage brandy: feasibility of control levels. Wines
and Vines. Vol. 53: pp. 37-40 (1972).

Guymon, J .F. James F. Guymon Lecture: Distillation in Alambic. American Journal
of Enology and Viticulture. Vol. 41: pp. 90-103 (1989).

Guymon, J .F. Mutant strains of Saccharomyces Cerevisiae applied to higher alcohol
formation during fermentation. Developments in Industrial Microbiology. Vol. 7:
pp. 88-96 (1966).

Jarvis, MC. The proportion of calcium bound pectin in plant cell walls. Planta. pp.
344-346 (1982).

Ngang, J .J ., Letoumeau, and Villa P. Alcoholic fermentation of beet molases: effects
of lactic acid on yeast fermentation parameters. Applied Microbiology and
Biotechnology. Vol. 31: pp. 125-128 (1989).

Pomper, S. Biochemistry of yeast fermentation. The Bakers Digest. pp. 32-38
(1969)

Pyke, M. The manufacture of scotch grain whiskey. Journal of Institutional Brewing.
Vol. 71: pp. 209-218 (1965).

Scott Laboratories. Fermentation Han—dbook (1988).

Skoog, D.A. Holler, F .J . and Nieman, T.A. Principles of Instrument An_alvsis 5th
edition. pp. 704-705, Harcourt Brace and Company, Philadelphia (1998).

69

14. Stinson, E.E., Dooley, C.J., F ilipic, V.J., and Hills, C.H. Composition of
montmorency cherry essence. 1. Low boiling components. Journal of Food Science.
Vol. 34: pp. 246-248 (1969).

15. Stinson, E.E., Dooley, C.J., Filipic, V.J., and Hills, OH. Composition of
montmorency cherry essence. 2. High boiling components. Journal of Food Science.
Vol. 34: pp. 554-546 ( 1969).

16. Suomalainen, H. and Nykanen, L. The aroma components produced by yeast in
nitrogen free sugar solutions. Journal of Institutional Brewing. Vol. 72: pp. 469-474
(1966)

17. Suomalainen, H. and Lehtonen, M. The production of aroma compounds by yeast.
Journal of Institutional Brewing. Vol. 85: pp. 149-156 (1978).

18. Tanner, H. and Brunner, H.R. Fruit Distillation Today 3rd edition. Heller Chemical
and Administration Society, Germany (1982).

19. Webb, D. and Ingraham, J .L. Fusel Oil. Advanced Applied Microbiology. Vol. 5:
pp. 317-353 (1963).

20. Wong, D.W.S. Food Enzymes: Structure and Mechanism. Chapman and Hall, New
York (1995).

70