Ena Sensory PDF
Ena Sensory PDF
Ena Sensory PDF
ALCOHOLIC BEVERAGES
by
DON MCGREW
A REPORT
MASTER OF SCIENCE
Food Science
2011
Approved by:
Major Professor
Dr. Delores Chambers
Abstract
Tools are available, through various reference books, to develop a purposeful sensory
quality program. Some companies already have a strong sensory program in place; others may
require a cultural change to facilitate the implementation. This paper indicates some of the
challenges to be overcome, covers some current quality control (QC) sensory practices and
addresses advantages and disadvantages of expert tasters Some specific issues regarding sensory
evaluations of alcohol beverages are discussed and critical factors in production are reviewed
with discussion on the potential for off taint development.
Key words: alcohol beverages, sensory quality, expert tasters, off odors, quality control
Table of Contents
iii
List of Figures
iv
List of Tables
v
Acknowledgements
"The final version of this report, with additional sections, is included as a book chapter
titled "Quality Control and Assurance through Sensory Evaluation in the book, "Alcoholic
Beverages: Sensory Evaluation and Consumer Research (Editor, J. Piggott, Woodhead
Publishing) .
vi
Chapter 1 - Introduction
There are numerous reference materials on sensory quality and its place within the
Quality Control (QC) / Quality Assurance (QA) function. These reference materials give explicit
detail on how to plan, set up, and implement a sensory program for QC. Most current sensory
analysis books devote at least a chapter on the subject. The American Society for Testing and
Materials (ASTM), for example, published a 52 page manual detailing the intricacies of
implementing a sensory program in a manufacturing plant s QC environment. [T]o bring
sensory testing for quality control out of the touchy/feely and value judgment realm into the
world of quality control data reporting (Yantis, 1992, p 49).
Despite availability of such reference materials, the integration of sensory evaluation into
the areas of QC/QA has achieved only moderate success throughout the food and beverage
industries. As Stone and Sidel (1993, p 295) mention, most QC sensory programs have not
been as successful as anticipated. Even after nearly two decades that statement holds true; and
the success of these programs in the alcohol beverage industry is no exception.
Stone and Sidel (1993) speculate that the lack of success in the implementation of
sensory programs is likely a result of the organizational environment. This is reasonable because
many manufacturing plants operate under the old axiom, if it s not broke, don t fix it . The
challenge is for QC sensory professionals that understand the benefits of a well run sensory
program to continue to enlighten industry management by demonstrating the efficiency and
practicality of a sensory quality model. The difficulty of this challenge is in describing the
benefits of a sensory program and how it can save time and money, while concurrently
increasing value for consumers. Phrased in this manner the business case in support of a QC
sensory program is more meaningful to management than scientific jargon about statistical
power, repeatability or reproducibility.
1
Chapter 2 - Sensory quality concerns and issues in the alcohol
beverage industry
Most alcohol beverage manufacturers perform some form of flavor evaluation on their
products. Some companies, typically the larger ones, have an established sensory
group/department responsible for the overall sensory programs used in the production facilities
QC departments. In these instances, the sensory group will communicate, as needed, standard
operating procedures, conduct and monitor panel training and interpret results for corrective
actions. Other companies perform sensory testing on an ad hoc basis where the evaluations are
conducted to meet the needs of a specific situation. On this point, it should be understood that
many recognizable brands in the alcohol beverage industry have been producing products much
longer than current organized sensory practices have been in use. The implication is not that
sensory quality is an unimportant aspect in these products production. In fact; the United States
Bottled-in-Bond Act of 1897 was enacted to protect whiskey consumers from illicit and often
lethal productions of blended whiskey. The Bottled-in-Bond Act is considered a precursor for the
U.S. Pure Food and Drug Act of 1906. Despite providing no direct guidance for sensory quality,
whiskey labeled as Bottled-in-Bond came to known as the good stuff because the Act did
provide assurance to the consumer (Veach, nd).
Meiselman and Schutz (2003) note that the old Seagram s company was one the first
companies, in any industry, to embrace sensory techniques established in the 1940 s by the U.S.
Army s Quartermaster Corps. Throughout the 1900 s, Seagram s was one of, if not, the largest
distilled spirits companies in the world. The Seagram s brand still exists, but is currently
owned by Diageo . Meiselman and Schutz (2003, p200) describe the hiring of David Peryam
and Norman Giradot; two of Seagram s quality control scientists, by the US Army s
laboratory in 1949 as a critical step that led to the development of many sensory analysis
techniques still in use today. It is apparent that the alcohol beverage industry has made an
important impact in the development of sensory evaluation methods and in general the field of
sensory analysis as a whole. Despite this link in the development of sensory evaluation methods
to the alcohol beverage industry, many companies still lack proper support for and use of sensory
quality practices. Munoz (2002, p327) stated, Current QC/sensory programs in industry cover
2
the whole gamut of quality , from expert measures similar to those used a century ago to
state-of-the-art programs.
At some point in the history of QC evaluations of alcohol beverages the phrase
organoleptic testing crept into industry s vocabulary. Though organoleptic testing does refer to
the use of the senses for product evaluations that is where the comparison to sensory evaluations
stops. Sensory evaluation involves greater attention to details such as methodological assessor
training, sample preparation, and recording and reporting steps, that are useful to the scientific
evaluation of products. Therefore, organoleptic tests are a more qualitative, subjective method,
whereas sensory evaluations typically have a more scientific basis. In this context it is not
surprising that most sensory texts avoid discussing the use of organoleptic testing. Since at least
the early 1980 s, sensory professionals have been trying to eliminate the use of the word
organoleptic from the practice of product evaluations (Jellinek, 1985; McDaniel, 1985). The
problem is that many people continue to use the words, organoleptic and sensory, in the same
context. The misuse of these words is especially prevalent in the alcohol beverage industry. It is
important for the laboratory analyst, sensory scientist, or whoever is charged with the function of
product quality assessment, to fully understand these differences, and the deficiencies related to
organoleptic testing. It is also important to understand that the terms; organoleptic testing and
sensory evaluation, are not synonymous and should never be used interchangeably.
A similar but perhaps more damaging effect to the integrity of sensory practices, has been
its misuse. It is not at all uncommon to find QC staffs combining difference tests, e.g., triangle
tests or paired comparisons with a preference test. Many researchers have commented on the
misapplication or combining of test methods and the resulting problems from this practice (Stone
and Sidel, 1993; O Mahony, 1986). Stone and Sidel explained the inherent problems (1993, pp
196) as follows:
Odd sample bias Panelists tend to prefer samples that are not perceived as
different (or odd in the case of a triangle test) from a reference. Since the initial
test is to determine sample difference, panelists are now predisposed to a
preference bias.
Skewed results Panelists for a discrimination test are recruited based on their
acuity and usually specific experience looking for differences in a product
3
Small sample sizes Discrimination tests typically use 20 30 panelists while
preference/acceptance tests require much larger panel sizes.
Transition of tasks It is more difficult than it might appear to mentally flip the
switch from trained discriminator to consumer, which potentially leads to odd
sample bias.
Making sense of the results Do you only include those preferences from
subjects that were correct on the difference test? Do you include the preference
from all subjects?
Additionally, the use of preference tests as part of plant sensory quality does not make
practical sense, because a primary role of QC is to make a product consistently, irrespective of
personal likes or dislikes. Other common misuses of sensory practices include the use of
discrimination test methods (i.e. triangle or duo-trio tests) with too few respondents for statistical
relevance, repeated test panels to achieve a desired result and having panelists evaluate more
than one set of samples on the same test to increase the number of responses.
4
of testing requires that the panelists have some training and background in typical or
representative product and then use good judgment in determining whether finding small,
sometimes inherent, differences result in the product being comparable.
When product is not found to be in compliance with standards, the corrective action may
involve blending off the unapproved product in small amounts with approved batches and
retesting to ensure that the new blend is not compromised in quality. Other times the decision is
to re-work the product by going through additional processing, i.e., additional filtration.
Occasionally, when a panel cannot reach a consensus to accept or reject, the questionable
product will go to a larger difference panel, i.e., triangle, duo-trio or paired comparison. This
panel will determine whether there is a significant difference between two samples the test
sample and a control batch selected as typical or representative. It is critical in these cases to
have large enough panels that decisions of no-difference can correctly be interpreted.
Also it is common to find multiple sample comparison tests in use by QC areas. These
types of tests are useful when multiple production samples of the same product are available;
usually 3 to 5 products compared to a reference. This test type can be conducted as either a
scaled test whose results are analyzed using ANOVA, or a ranking of samples for difference by
Friedman analysis (Meilgaard et al., 2007). These types of tests often require 8 12 panelists, so
their use may be restricted based on the number of panelist available.
Descriptive analysis methods are less common throughout the alcohol beverage industry.
Some producers, most notably the larger beer and wine companies, use descriptive analysis
methods to define product flavor profiles. The information gathered from descriptive testing on a
product s flavor profile can be used to train panelists in the identification and intensity of critical
product characteristics. Such information also may help the sensory staff understand sensitivities
that particular individuals on QA/QC panels have to specific flavor compounds that may
influence his/her decisions. Research also suggests that trained sensory panel results may be
useful as predictive indicators of consumer acceptance (Rousseau, 2010; Lattey et al, 2010;
Drake and Civille, 2002) although such predictions would need to be determined on a product by
product basis. Again, it is important to emphasize that a QC/QA test rarely is designed simply to
test overall acceptability or liking , but rather to ensure that products meet agreed upon
standards of typicality and representativeness.
5
Advantages and disadvantages of expert tasters
Master brewer, cellar master, enologist, winemaker, master distiller and master blender,
these are some of the titles used in alcohol beverage industry that indicates experts in their
field. The expertise of these individuals is not only about their acuity as part of a difference panel
or their descriptive abilities, but also refers their knowledge of the production process and the
products (Hughson and Boakes, 2002; Lawless, 1985).
Use of the term expert can be misleading, in regards to who is responsible for the
sensory quality of a product. Is it the individual or the panel that has the responsibility for
sensory quality? It is true, at least in the not too distant past, that experts would be the final
decision makers regarding a product s quality. This way of working has changed in recent years
in many companies. Although there are individuals who have in-depth knowledge of and
experience with specific products, these individuals rarely make sensory quality decisions
exclusive of other s input. In many cases where experts make a final decision, they also
participate on panels and use the panel s results to help form their decisions of a particular
product s quality. These panels usually evaluate products on a go/no go basis. Training of new
panelists in this setting often takes place in what can be described as an apprenticeship with the
expert and the panel. There obviously are problems with this training method i.e., lack of
standardized training protocols, understanding individual s sensitivities and anosmias, and the
effect of panel influences among others.
Training and maintenance protocols for panelists are what set sophisticated sensory
programs apart from other programs. All companies want to produce the most consistent quality
products possible. However, companies that have sensory procedures in place without a sensory
program strategy and strong management support typically have fewer resources (i.e., time,
money and people), made available to them than their counterparts in other companies. Granitto
et al., (2008, p590) stated, the most difficult and time consuming stage of sensory profiling is
probably the selection and training of a panel, in the attempt to reach agreement in the use of
scales and in the meaning of each attribute.
Thorough training and maintenance protocols for the development and assessment of
panelists have been implemented by some companies, but there are many who still rely on
expert tasters to evaluate their products. There have been many published reports on improved
reliability of trained sensory panels over the expert tasters. (Guinard et a.l, 1998; Latreille et
6
al., 2006; and Gawel and Godden, 2008). There is no argument that training of panelists
improves reliability of results. The research by Guinard et al., (1998, p59), for example, suggests
that expert tasters judgments differ significantly from person to person when asked to
evaluate the quality of a variety of beers.
7
Most wines and beers are of low enough alcohol content that alcohol reduction is not
required before testing. However, for those distilled spirit products with greater than 35%
alcohol content, it is common practice to dilute the alcohol content using de-ionized or distilled
water. Depending on the producer of the product, the range for evaluation can be from 15% -
25% alcohol content. It is important that a single diluted alcohol content is adhered to for quality
assessments of a product. Varying the alcohol content from one panel to the next can cause
confusion and erroneous results. There is a simple formula to use when diluting alcohol content
of a product;
The remaining desired volume would be filled with de-ionized water to bring the product
to the desired alcohol content.
8
Other special concerns related to the sensory evaluations of alcoholic beverages
Along with the previously discussed issue regarding alcohol content, Piggott and
Macleod (2010) also provide insight to other areas of special concern when evaluating beverage
alcohol. These areas of concern include: 1) time from sample preparation to serving and
evaluation; 2) sample serving temperature; and 3) the topic of nosing as compared to tasting. The
first of these, time from preparation to serving, suggests serving as close to a fresh preparation as
possible. Nonetheless those authors recognize that this is not always a possibility and conclude
that most sample preparations (e.g., that are not temperature sensitive or time sensitive in the
case of carbonated beverages) may be pre-poured and held for evaluation up to eight hours
without noticeable differences in results. Next is sample serving temperature with the basic
concern being that although products generally may be served at a wide range of temperatures;
product comparisons of a given product should always be served at the same temperature. The
last topic, nosing, is a common practice given the number of samples that a QC lab may evaluate
daily. In many labs the appearance and smell of the product are determined during nosing and
used to determine whether the product will pass QC sensory tests. In those cases, the product
may not actually be tasted. This presents a conundrum that must be understood no matter what
testing decision finally is made. It may be important to control the amount of beverage that is
tasted, because the alcohol content can result both in adaptation and desensitization and can
produce physiological effects if over-consumed,. Even if assessors do not swallow the samples
being evaluated, the alcohol can still be absorbed through the mouth. This information suggests
that nosing, without actual tasting is preferred. However, tasting the sample can release flavors
or heighten certain flavors that would not be detected in the same way if the sample is only
smelled. Again, companies need to make decisions based on informed discussion before
deciding on a specific set of protocols. Regardless, if the samples are being tasted, expectorating
almost always is encouraged when numerous samples (more than three or four depending on
alcohol content) are to be evaluated in a single session.
Two other areas of concern for quality evaluations to be considered include sample order
and the sample number and serving size. It is not uncommon for the quality assessor to be
involved in multiple evaluations of a variety of products. If this is the case, samples should be
organized in an order of least flavorful, i.e., most neutral, such as vodka, to more heavily
flavored products such as gin or a full bodied red wine. Flavor impact is not the only concern
9
because mouth-feel and texture also come into play when considering sample order. More
viscous types of products may leave a mouth coating that can influence subsequent evaluations.
The number of samples and serving size is critical to sensory quality assessment and even
more so if consumers are involved in the QA process as they might be when standards are first
established or are checked to determine if modifications are necessary. The Distilled Spirits
Council (DISCUS) of the U.S. lists one of its main goals as the promotion of responsible
drinking. They have published guidelines comparing serving sizes of distilled spirits, wine and
beer. In a press release on 22 April, 2010 DISCUS proposed, [s]erving sizes are 1.5 fluid
ounces for 80-proof distilled spirits, 12 fluid ounces for regular beer, 5 fluid ounces of wine.
These guidelines are important for all researchers and participants involved with alcohol
beverage evaluations to understand. Likewise, when conducting consumer research it is the
responsibility of the manufacturer to understand the laws regarding total allowable serving
volume, which affects the number of samples that may be presented as well as the serving size
that each respondent can receive. Tables 1 and 2, below, show equivalent number of servings
based on serving size and alcohol content. It should be noted these are not recommended serving
sizes for individual samples for testing purposes, but rather are included only to help make
comparisons for total alcohol consumption during testing based on total amount consumed
during a test.
10
Table 2-1 Equivalent serving volumes (fluid ounces)
11
Table 2-2 Equivalent serving volumes (milliliters)
The most basic similarity of beer, wine and distilled spirit products is quite obvious, they
all contain alcohol in the form of ethanol. The alcohol content of these products arguably affects
their sensory quality more than any other single factor. For example, the fermentation process
that produces ethanol also contributes to the formation of several volatile aromatic compounds
12
including acetaldehyde, diacetyl, acetic acid and numerous other compounds. The equilibrium of
ethanol and other compounds produce further aromatic compounds; an example would be the
formation of ethyl acetate through the esterification of acetic acid in the presence of ethanol.
Additionally, alcohol influences the sensory quality of these products by the fact that it is an
active solvent. This ability to serve as a solvent for other flavor fractions has an impact on the
sensory testing of the products. As products become more complex and more flavorful, QC
testing become more difficult both from the standpoint of the typicality of the product and the
timing and serving protocols that must be considered.
The alcohol content of a product also impacts the maturation process of products that are
aged in wood, the order of addition and solubility of ingredients in compounded products (e.g.
cordials, gins, blended whiskies to name a few) and the shelf stability of products as well.
The differences that affect the sensory quality for beer, wine and distilled spirits are
numerous. The first difference is the fermentable substrates used in the production of each type
of product. Next, the production processes for wine, beer and distilled spirits are quite different.
Some products are carbonated, some are aged in wood and others require filtration. The
operational controls that must be included for sensory testing of the various products are product
specific and must be considered based on the total product and its intended use.
Raw materials
Water is a critical component in the production of all alcohol beverages. The quality of
water used in production often goes unnoticed when there are no problems, but when quality
issues arise in production, water is one of the primary suspects. The use of water is ubiquitous
throughout production. Whether water is limited in use to rinsing and cleaning production
vessels or it is added as an ingredient in the production process (e.g., beer and distilled spirit
productions), the quality of water is essential to the quality of the finished product. Routine
evaluations to monitor water quality should be included as part of a sensory quality program,
because conventional water treatment procedures are not always adequate to remove off-odors
(Srinivasan R and Sorial G, 2011). Geosmin and 2-methyl isoborneol (2-MIB), which have been
13
described as having earthy and musty notes respectively, are two of the most problematic taints
found in water.
All alcohol starts from some sort of fermentable substrate (e.g., grains, fruits or tubers to
name a few). Any material that contains simple sugars, or complex carbohydrates that can be
converted into simple sugars, has the potential to produce alcohol through fermentation.
Establishing sensory procedures to evaluate these materials as they come into the production
facility adds assurance that new shipments meet predetermined quality criteria for flavor and
consistency. A common method used for grain evaluations involves making a tea where hot
water is poured into a glass containing the grain. This sample is compared side-by-side with a
reference sample of the specific grain, prepared in the same way. Often it is the grain
operators/receivers who are charged with this responsibility, which works only if they are given
procedural support and training to conduct these evaluations. Common off-notes that can be
detected by this method include geosmin, 2-MIB, and diesel fuel (used commonly in machinery
to dry grains).
The yeast variety used in fermentation also impacts the flavor of the finished product
(Lilly et al, 2000; Lee et al, 1995). Many companies choose to use dry yeast, because it allows
for more consistent yeast application and is less costly to maintain. Though use of dry yeast does
not prevent cross-contamination from wild strains, there is less opportunity for this interaction.
Companies that choose to propagate their own proprietary strain of yeast must do so with care.
Wild yeasts are everywhere and can mutate a particular strain of yeast if proper control
conditions are not in place. Proper sensory testing to ensure that no extraneous flavors are
formed by unexpected yeast contamination is essential.
Operational controls
Time, temperature and pH are critical operational control points in the production of
alcohol beverage. Beer and many spirits products begin with a mashing process, where grains are
cooked with water (i.e., the mash) to gelatinize the starch content of the grains. Obviously to
cook the mash, the temperature is elevated. However, prior to the addition of enzymes, or malt in
the case of straight whiskies, that is used to convert the starches to fermentable sugar, the
temperature must be reduced so as not to inhibit this enzymatic action. As the mixture is
transferred into a fermenter it must continue to cool before the yeast is added. In some products
14
an additional cooling step takes place prior to the enzyme addition. For bourbon production, for
example, milled rye is added to mash. If the rye is added at too high a temperature it will form
rye balls (i.e., the rye meal will clump together). The interior portion of the rye ball does not
get cooked. This uncooked rye meal can act as a medium for bacteria growth, which can
contaminate the fermenter. Increased levels of phenolic compounds (e.g., ethyl phenol and ethyl
guaiacol) are associated with contaminated fermentations due to the presence of rye balls . In
addition to proper temperature control, the yeast activity in the fermenter is influenced by the pH
of the mash. Each yeast strain operates at its own optimal pH range. Control of pH in the
fermenter will impact the yield and quality of the alcohol produced (Bra nyik et al, 2004; Davis
et al, 1985). Sensory testing during the production process, often by nosing intermediate
products of the production, can help to determine if problems are occurring before additional
processing steps compound the problems.
There are other operational controls to minimize the risk of product contamination.
Proper sanitation and cleaning of all equipment is critical. Cleaning minimizes bacteria and mold
growth potential as well as reducing the chance of cross-contamination from shared equipment.
In plants with stills, keeping copper portions of the still clean reduces the occurrence of sulfur
compounds. Clean, active copper sites will bind sulfurous compounds to minimize their
migration through the still. Many companies have adopted closed-topped stainless steel
fermenting tanks. In the past, fermenters often were open-topped wooden containers. Stainless
steel is easier to clean and sanitize than wood. The closed system also reduces contamination
from air-born wild yeast and bacteria (Bra nyik et al, 2005).
Maturation/Storage
Some wines and spirits are aged, often in oak barrels. This process is known as
maturation. The sensory impact from maturation can be huge depending on the length of time the
product ages, the type and size of the barrels, alcohol content in the barrel, and whether the barrel
is a new charred or reused barrel. It is possible to underage and overage products. Not enough
time and the product may appear immature. Too much time the product may be considered over-
aged. Product in smaller barrels will mature (i.e., gain more wood character) more quickly than
product aged in larger barrels, because the ratio of surface area to volume is greater in the small
barrels. French oak provides a different flavor (drier more tannic flavor) profile than American
15
white oak (more vanilla notes). Product with higher alcohol content will extract more tannins
than a lower alcohol content product. More delicate distilled spirits (e.g., rum, Canadian whisky
and malted whiskies) are aged in reused barrels. Straight whiskies (e.g., bourbon and rye
whiskies) are aged in new charred oak barrels, as prescribed by U.S. law. Research also suggests
that barrel aging can contribute increased levels of phenolic compounds in wine (Rayne, 2007).
The difficulty in sensory testing of these products is that the product changes considerably during
the aging process and the initial product going into the barrel is quite different from the ending
product. However, it is clear that poor quality, tainted, product entering the aging process will
not improve with age.
Packaging materials
Finished product containers (e.g., cans, glass and PET) and closures also need to be
evaluated for product compatibility. The difference here is that these materials will be in contact
for a much long time period. Packaging materials can initially be reviewed using the 72 test as
described above, but for approval these materials must be placed into long-term storage testing
containing product similar to what it is intended to hold. Test increments for can vary; but
increments of 1, 3, 6, 9 24 months often are recommended.
If closures are being tested it is suggested that the container be laid on its side to
maximize product contact. Historically, natural cork has been used as a closure with many
16
alcohol beverage products. Over the past 20 years, many companies have replaced natural corks
with synthetic cork. Natural corks have the disadvantage of occasionally imparting an off odor,
2,4,6-trichloroanisole (TCA), with odor characteristics similar to geosmin or 2-MIB.
Chapter 6 - Conclusions
Great strides have been made in the alcohol beverage industry to improve sensory
quality programs, but room for further improvement exists. Continued training is the key to the
successful implementation of sensory programs as well as a means of retraining employees to an
enhanced understanding of quality. Training should include discussions with management to
establish quality goals, but should integrate other functional groups within the company,
especially those groups involved in activities that can influence the development of off-odors and
taints.
17
References
Bra´nyik T, Vicente AA, Cruz JM and Teixeria JA (2004), Continuous primary fermentation of
beer with yeast immobilized as spent grain: the effect of operational conditions , Journal of the
American Society of Brewing Chemist, 62(1), 29-34.
Bra´nyik T, Vicente AA, Dosta´lek P, Teixeira JA (2005), Continuous beer fermentation using
immobilized yeast cell bioreactor systems ,Biotechnology Progress, 21, 653 663.
Davis CR, Wibowo D, Eschenbruch R, Lee TH and Fleet GH (1985), Practical implications of
malolactic fermentation: A review ,American Journal of Enology and Viticulture, 36(4), 290-
301.
DISCUS (2010), Distilled Spirits Council urges government support for standard drink
information on alcohol labels, Washington, D.C., Distilled Spirits Industry Council. Available
from: http://www.discus.org/media/press/article.asp?NEWS_ID=589 [accessed 19 November
2010].
Drake MA and Civille GV (2003), Flavor Lexicons , Comprehensive Reviews in Food Science
and Food Safety, 2(1), 33-40.
Gawel R and Godden PW (2008), Evaluation of the consistency of wine quality assessments
from expert wine tasters , Australian Journal of Grape and Wine Research, 14, 1-8.
Granitto PM, Biasioli F, Endrizzi I and Gasperi F (2008), Discriminant models based on sensory
evaluations: Single assessors versus panel average , Food Quality and Preference, 19, 589-595.
Guinard J, Yip D, Cubero E and Mazzucchelli R (1999), Quality ratings by experts, and relation
with descriptive analysis ratings: A case study with beer , Food Quality and Preference, 10, 59-
67.
18
Hughson A and Boakes R (2002), The knowing nose: the role of knowledge in winr expertise
,Food Quality and Preference, 13, 463-472.
Jellinek G (1985), Sensory evaluation of food: theory and practice, Chichester, UK, Ellis
Horwood Ltd.
Lattey KA, Bramley BR and Francis IL (2010), Consumer acceptability, sensory properties and
expert quality judgments of Australian Cabernet Sauvignon and Shiraz wines , Australian
Journal of Grape and Wine Research, 16, 189-202.
Lee S, Villa K and Patino H (1995). Yeast strain development for enhanced production of
desirable alcohols/esters in beer , Journal of the American Society of Brewing Chemist, 53, 153-
156.
McDaniel MR (1985), Sensory evaluation of food flavors , in Bills D et al, Characterization and
management of flavor compounds, Washington, D.C., American Chemical Society, 1-9.
Meilgaard M, Civille G and Carr TB (2007), Sensory evaluation techniques, Boca Raton, Fl,
CRC Press.
19
Meiselman H and Schutz H (2003), History of food acceptance research in the US Army ,
Appetite, 40, 199-216.
Munoz A (2002), Advances in sensory evaluation for quality control , Food Quality and
Preference, 13, 327-328.
Nakamura S, Nakamura N and Ito S (2001), Determination of 2-methylisoborneol and geosmin
in water by gas chromatography-mass spectrometry using stir bar sorptive extraction , Journal of
Separation Science, 24(8), 674 677.
O Mahony M (1986), Sensory evaluation of food, New York, NY, Marcel Dekker.
Piggott JR and Macleod S (2010), Sensory quality control of distilled beverages , in Kilcast D,
Sensory analysis for food and beverage quality control: A practical guide, Cambridge,
Woodhead Publishing, 262-275.
Rayne S and Eggers NJ (2007), 4-ethylphenol and 4-ethylguaiacol in wines: Estimating non-
microbial sourced contributions and toxicological considerations ,Journal of Environmental
Science and Health Part B Pesticides Food Contaminants and Agricultural Waste , 42(8), 887-
897.
Schrader K and Blevins W (2001), Effects of carbon source, phosphorous concentration, and
several micronutrients on biomass and geosmin production by Streptomyces halstedii , Journal
of Industrial Microbiology & Biotechnology, 26, 241-247.
20
Srinivasan R and Sorial G, (2011). Treatment of taste and odor causing compounds 2-methyl
isoborneol and geosmin in drinking water: A critical review , Journal of Environmental Sciences,
23, DOI
Stone H and Sidel JL (1993), Sensory evaluation practices, 2nd edn, San Diego, CA, Academic
Press.
Veach M (nd), The Filson Historical Society: The Filson Newsmagazine, 7:4. Available from:
http://www.filsonhistorical.org/archive/news_v7n4_distilling.html [accessed 19 November
2010].
Yantis J, ed (1992), The role of sensory analysis in quality control, ASTM Manual Series: MNL
14, Baltimore, MD, ASTM International.
21