Академический Документы
Профессиональный Документы
Культура Документы
Odors in
the Food
Industry
Odors in the Food Industry
ISEKI-FOOD SERIES
Series Editor: Kristberg Kristbergsson, University of Iceland
Reykjavík, Iceland
Edited by
Xavier Nicolay
Institut Meurice
Brussels, Belgium
Xavier Nicolay
Ingénieur
Service de Génie Chimique & Biochimique
Instiut Meurice – HELdB
1, avenue Emile Gryzon, bât. 2
1070 Bruxelles
Belgium
xnicolay@ceria.be
Series Editor
Kristberg Kristbergsson
Professor of Food Science
Dept. Food Science and Human Nutrition
University of Iceland
Hjardarhaga 2-6
107, Reykjavik
Iceland
ISBN-10: 0-387-33510-2
ISBN-13: 978-0387-33510-0
e-ISBN-10: 0-387-34124-2
e-ISBN-13: 978-0387-34124-8
9 8 7 6 5 4 3 2 1
springer.com
SERIES ACKNOWLEDGMENTS
ISEKI Food is a thematic network on food studies, funded by the European Union as
project N° 55792-CP-3-00-1-FR-ERASMUS-ETN. It is a part of the EU program in
the field of higher education called ERASMUS, which is the higher education action
of SOCRATES II program of the EU.
v
SERIES PREFACE
The single most important task of food scientists and the food industry as a whole
is to ensure the safety of foods supplied to consumers. Recent trends in global food
production, distribution, and preparation call for increased emphasis on hygienic
practices at all levels and for increased research in food safety in order to ensure a
safer global food supply. The ISEKI-Food Series is a collection of books where vari-
ous aspects of food safety and environmental issues are introduced and reviewed by
scientists specializing in the field. In all of the books a special emphasis was placed
on including case studies applicable to each specific topic. The books are intended
for graduate students and senior level undergraduate students as well as profession-
als and researchers interested in food safety and environmental issues applicable to
food safety.
The idea and planning of the books originates from two working groups in the
European thematic network “ISEKI-Food,” an acronym for “Integrating Safety and
Environmental Knowledge In Food Studies.” Participants in the ISEKI-Food net-
work come from 29 countries in Europe and most of the institutes and universities
involved with food science education at the university level are represented. Some
international companies and nonteaching institutions have also participated in the
program. The ISEKI-Food network is coordinated by Professor Cristina Silva at
The Catholic University of Portugal, College of Biotechnology (Escola) in Porto.
The program has a website at: http://www. esb.ucp.pt/iseki/. The main objectives of
ISEKI-Food have been to improve the harmonization of studies in food science and
engineering in Europe and to develop and adapt food science curricula, emphasizing
the inclusion of safety and environmental topics. The ISEKI-Food network started
on October 1st in 2002, and recently has been approved for funding by the European
Union for renewal as ISEKI-Food 2 for another three years. ISEKI has its roots in
an EU-funded network formed in 1998 called Food Net where the emphasis was on
casting a light on the different food science programs available at the various univer-
sities and technical institutions throughout Europe. The work of the ISEKI-Food
network was organized into five different working groups with specific task, all aiming
to fulfill the main objectives of the network.
The first four volumes in the ISEKI-Food book series come from WG2 coordi-
nated by Gerhard Schleining at Boku University in Austria and the undersigned.
The main task of the WG2 was to develop and collect materials and methods for
teaching safety and environmental topics in the food science and engineering cur-
ricula. The first volume is devoted to food safety in general with a practical and
case study approach. The book is composed of 14 chapters that were organized
into three sections on preservation and protection, benefits and risk of microorgan-
isms, and process safety. All these issues have received high public interest in recent
years and will continue to be in the focus of consumers and regulatory personnel
for years to come. The second volume in the series is devoted to the control of air
vii
viii Series Preface
pollution and treatment of odors in the food industry. The book is divided into eight
chapters devoted to defining the problem, recent advances in analysis, and methods
for prevention and treatment of odors. The topic should be of special interest to
industry personnel and researchers due to recent and upcoming regulations by the
EU on air pollution from food processes. Other countries will likely follow suit with
stricter regulations on the level of odors permitted to enter the environment from
food-processing operations. The third volume in the series is devoted to utilization
and treatment of waste in the food industry. Emphasis is placed on sustainability of
food sources and how waste can be turned into by-products rather than pollution
or landfills. The book is composed of 15 chapters starting off with an introduction
of problems related to the treatment of waste and an introduction to the ISO 14001
standard used for improving and maintaining environmental management systems.
The book then continues to describe the treatment and utilization of both liquid
and solid waste with case studies from many different food processes. The last book
from WG2 is on predictive modeling and risk assessment in food products and pro-
cesses. Mathematical modeling of heat and mass transfer as well as reaction kinetics
is introduced. This is followed by a discussion of the stoichiometry of migration in
food packaging, as well as the fate of antibiotics and environmental pollutants in the
food chain using mathematical modeling and case study samples for clarification.
Volumes five and six come from work in WG5 coordinated by Margarida Vieira at
the University of Algarve in Portugal and Roland Verhé at Gent University in Bel-
gium. The main objective of the group was to collect and develop materials for teach-
ing food safety-related topics at the laboratory and pilot plant level using practical
experimentation. Volume five is a practical guide to experiments in unit operations
and processing of foods. It is composed of 20 concise chapters each describing differ-
ent food-processing experiments outlining theory, equipment, procedures, applicable
calculations, and questions for the students or trainees followed by references. The
book is intended to be a practical guide for the teaching of food-processing and engi-
neering principles. The final volume in the ISEKI-Food book series is a collection of
case studies in food safety and environmental health. It is intended to be a reference
for introducing case studies into traditional lecture-based safety courses as well as
being a basis for problem-based learning. The book consists of 13 chapters contain-
ing case studies that may be used, individually or in a series, to discuss a range of
food safety issues. For convenience the book was divided into three main sections
on microbial food safety, chemical residues and contaminants, and a final section on
risk assessment and food legislation.
The ISEKI-Food book series draws on expertise from close to a hundred universi-
ties and research institutions all over Europe. It is the hope of the authors, editors,
coordinators, and participants in the ISEKI network that the books will be useful to
students and colleagues to further their understanding of food safety and environ-
mental issues.
ix
CONTENTS
CONTRIBUTORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii
2. ODOR MEASUREMENT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Elefteria Psillakis
INDEX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
xi
CONTRIBUTORS
Simona Benedetti
Department of Food Science, Technology and Microbiology, University of
Milan, Via Celoria 2, 20133 Milan, Italy
Susanna Buratti
Department of Food Science, Technology and Microbiology, University of
Milan, Via Celoria 2, 20133 Milan, Italy
Kristof Demeestere
EnVOC Research Group, Department of Organic Chemistry, Faculty of
Bioscience Engineering, Ghent University, 9000 Gent, Belgium
Vassilis Gekas
Department of Environmental Engineering, Technical University of Crete,
Polytechnioupolis, GR-73100 Chania-Crete, Greece
Kristberg Kristbergsson
Department of Food Science and Human Nutrition, University of Iceland,
Hjardarhaga 2-6, Reykjavík, Iceland
Saverio Mannino
Department of Food Science Technology and Microbiology, University of
Milan, Via Celoria 2, 20133 Milan, Italy
Regina Nabais
C.E.R.N.A.S. -Centro de Recursos Naturais, Ambiente e Sociedade, Escola
Superior Agrária de Coimbra, Instituto Politécnico de Coimbra. Bencanta,
3040-316 Coimbra, Portugal
xiii
xiv Contributors
Gudrun Olafsdottir
Icelandic Fisheries Laboratories, Skulagata 4, 101 Reykjavík, Iceland
João Peixoto
Department of Biological Engineering, CEB, University of Minho, 4710-057
Braga, Portugal
Elefteria Psillakis
Department of Environmental Engineering, Technical University of Crete,
Polytechnioupolis, GR-73100 Chania-Crete, Greece
Bram Sercu
EnVOC Research Group, Department of Organic Chemistry, Faculty of
Bioscience Engineering, Ghent University, 9000 Gent, Belgium
1
Odor Problems in the Food
Industry
Elefteria Psillakis and Vassilis Gekas
1. INTRODUCTION
Odor is complex both because of the large number of compounds that contrib-
ute to it and because it involves a subjective human response. In general, different
people find different odors offensive at different concentrations and this is fre-
quently related to the way different people perceive odors. According to a simple
1
2 E. Psillakis and V. Gekas
Other parameters such as age and gender also may contribute to the ability
to perceive an odor (Bliss et al., 1996) and to a lesser extent health (e.g., cold,
nasal allergy), personality, educational background, training, or hereditary defi-
ciencies in odor sensitivity.
The regulations and guidelines vary from country to country. It is neither pos-
sible nor practical to provide all guidance or regulations of different countries.
As such, the aim of this section is to provide some useful information relevant
to the idea behind odor nuisance. It also should be mentioned here that the term
4 E. Psillakis and V. Gekas
“guidelines” in the context of this chapter implies not only numerical values but
also any kind of guidance given.
4.1. Germany
After World War II, the rapid economic development and the densely populated
areas were responsible for the early regulatory legislation regarding air pollution
in Germany. The legal basis for any requirement with respect to air quality is the
German Federal Protection Act for Ambient Air (1974/1990) and the Technical
Instruction on Air Quality Control (1986). According to the Federal Protection
Act for Ambient Air, odors caused by plants and are treated as an annoyance,
although the problem is to find out whether an annoyance has to be considered
as significant (relevance of the annoyance).
The state of Northrhine-Westfalia is undoubtedly the most densely popu-
lated and industrialized area in Germany and for decades now the authorities of
this state have been developing and testing a new regulation/directive, according
to which a complete system has been designed, beginning with measurement
methods of the existing odor load and calculation of the expected odor load,
and concluding with ambient air quality requirements expressed as limit values
in terms of maximum permitted odor frequency in ambient air in certain areas
(Frechen, 2000; Both, 2001).
These limit values were developed on the basis of investigations in which the
existing odor load measured as odor frequency (Guideline VDI 3940, 1997) and
the degree of odor annoyance of residents assessed by questionnaires according
to guideline VDI 3883 Part 1 (1997) were correlated. The result was to set a limit
impact concentration at 1 odor unit /m3 (see definition in Chapter 2, paragraph
3.2) and then to limit the time percentage during which a higher impact con-
centration is tolerable (Both, 2001). These time percentages (odor frequencies)
were between 10% for residential areas and 15% for industrial areas. In order to
assess more correctly the extent of odorous emission to people it is stipulated
that an hour may be recognized as exceeding the limit value if the limit value is
exceeded during 10% of 1 hour, thus during 6 minutes. In practice, it is assumed
to be sufficient to multiply the hourly mean by a factor of 10 (the so-called factor
10 model). Thus an impact concentration of 0.1 odor unit /m3 (hourly average)
equals to 1 odor unit /m3 in the sense of the directive.
The general policy in Netherlands is to keep the population as free as possible from
annoyance. The term “annoyance” is translated as the percentage of population
perceiving sometimes or even often an annoying odor. In this context in the year
2000, 12% of the population was annoyed by industrial odors (Frechen, 2001).
The pig production sector in the Netherlands is considerable in size, relative
to both the size of the population and the surface area of the country. Annual
production is approximately 30 million pigs, which amounts to 2 pigs per head
Odor Problems in the Food Industry 5
of the population. It is therefore not surprising that odor impact of pig produc-
tion is a major environmental issue, given the high density and proximity of both
residents and pigs.
The first guideline on how to take account of environmental odor aspects
for licensing as a result of application of the existing the Nuisance Law was first
issued in 1971 and revised several times in later years.
The Environmental Protection Act of 1990 provides the legal framework for
avoiding and controlling odor nuisance in the United Kingdom. A comprehensive
overview over “The legal context of odor annoyance” in the United Kingdom can
be found elsewhere (Salter, 2000). The regulations, however, do not set any general
valid emission standards concerning odor and retracts to more general statements
concerning the odor nuisance. This is expected to change in the near future after
publication of Technical Guidance Note H4 (2002), Integrated Pollution Preven-
tion and Control (IPPC), Horizontal Guidance for Odor by the Environment
Agency.
The regulations for odor in the United States of America vary from state to
state. Depending on the state regulation may concern hydrogen sulfide limits,
detection to threshold limits, or use as the main legal basis the nuisance law. An
overview on the regulation of odors in the United States is given by Thomas
Mahin (2001) and is summarized in Table 1.
5. ODORS CHEMISTRY
A number of factors affect the emission of compounds from food industry opera-
tions. Most of the substances emitted are the products of microbial processes and
in most cases it is the microbial environment that will determine which substances
are generated and at what rate. This section intends to describe the chemical and
biological mechanisms that affect the formation and release of emissions. Most of
the information used in this section was taken by the draft report of the US Envi-
ronmental Protection Agency (2001).
5.1. Ammonia
should be mentioned here that during the storage of fresh manure, amino acids
are most likely to undergo decarboxylation producing putrescine, cadaverine,
and ammonia (Zhu, 2000).
Hydrogen sulfide and other reduced sulfur compounds are produced during
anaerobic decomposition and hydrogen sulfide is the predominant reduced sul-
fur compound emitted from animal feeding operations. In the case of animal
manures, there are two primary sources of sulfur (US Environmental Protection
Agency, 2001): (1) sulfur amino acids contained in the feed and (2) inorganic
sulfur compounds, such as copper sulfate and zinc sulfate, serving as growth
stimulants and used as feed additives to supply trace minerals.
In general, the magnitude of hydrogen sulfide emissions depends on the
liquid phase concentration, temperature, and pH. It is well known that the solu-
bility of hydrogen sulfide in water increases at pH values above 7. Therefore, as
pH shifts from alkaline to acidic, the potential for hydrogen sulfide emissions
increases (Stumn and Morgan, 1996).
As already stated, under anaerobic conditions, any excreted sulfur will be
microbially reduced to hydrogen sulfide. Thus, all manures managed as liquids
or slurries are potential sources of hydrogen sulfide emissions; as such emissions
from confinement facilities with dry manure handling systems and dry manure
stockpiles should be negligible if there is adequate exposure to atmospheric oxy-
gen to maintain aerobic conditions.
Other sulfur compounds emitted from animal feeding operations include
methyl mercaptan, dimethyl sulfide, dimethyl disulfide, and carbonyl sulfide. In
general, the very offensive smelling compound methyl mercaptan is a product
of sulfur-containing amino acid decomposition, and it can be oxidized to form
the unpleasant-smelling compounds dimethyl disulfide or dimethyl sulfide (Zhu,
2000).
in emission plumes from swine production facilities. Together with CO2, H2, as
well as ammonia, they can be produced from the deamination of amino acids
that are produced during the process of protein degradation and breakdown of
carbohydrates. Typical acids in this group consist of acetic, propionic, butyric,
iso-butyric, valeric, iso-valeric, caproic, and capric acids.
Phenolic compounds such as phenols and p-cresols are produced from the micro-
bial degradation of tyrosine and phenylalanine in the intestinal tract of animals
(Ishaque et al., 1985). Indole, skatole, cresol, and 4-ethylphenol appear to be
the major components included in this group of compounds. In addition, the
metabolism of tryptophan can result in the production of indoleacetate, which
is subsequently converted into skatole (3-methylindole) and indole by a different
group of bacteria (Zhu, 2000).
The major components in the food industry are depicted on Figure 1. As can
been seen edible products go through a number of stages before reaching the
consumer. It is generally accepted that odor problems in the food industry are
more likely to occur during the first two stages. Food service and food retail
operations are themselves generally not significant sources of air or odor pollu-
tion. No data are available on the portion of total emissions of specific pollut-
ants that can be ascribed directly to the food service and food retail industries.
The one exception is emissions from vent hood systems of food service equip-
ment, where in some cases users of certain food service cooking equipment are
requested to install pollution control measures.
A selection of the food industry activities of the first two stages are given in
Table 2. In the food industry large livestock operations, poultry farms, slaugh-
terhouses, food and meat processing industries, and bone mills are among major
contributors to odor pollution.
Figure 1. Major components in the food industry. Adapted from Davies and Konisky (2000).
Odor Problems in the Food Industry 9
from buildings and storages are relatively constant and vary with seasonal tem-
peratures. In contrast, land application of manure has the potential to emit
large amounts of gases periodically throughout the year. Gas emissions vary
significantly with management practices and manure system design.
The main odorous compound groups associated with animal feeding
operation and wastes are fatty acids, amines, ammonia, aromatics, as well as
inorganic and organic sulfur (US Environmental Protection Agency, 2001).
The most frequently detected odorous compound from livestock operations
is undoubtedly hydrogen sulfide. Another group of malodorous compounds
released during animal feeding operations are the volatile organic compounds.
Next to the odor problem created by these compounds, the majority of them
have an environmental impact, given that some of them participate in atmo-
spheric photochemical reactions, while others play an important role as heat-
trapping gases.
Meat smokehouses are used to add flavor, color, and aroma to various meats,
including pork, beef, poultry, and fish. The operations typically involved in the
production of smoked meat are tempering or drying, smoking, cooking, and
finally chilling. In general, the pollutants associated with meat smokehouses
include particulate matter (PM), carbon monoxide (CO), volatile organic com-
pounds (VOC), polycyclic aromatic hydrocarbons (PAH), organic acids, acro-
lein, acetaldehyde, formaldehyde, and nitrogen oxides. Acetic acid (followed
by formic, propionic, butyric, and other acids) has been identified as the most
prevalent organic acid present in smoke that is the primary source of emissions
in the smokehouses. Furthermore, heating zones in continuous smokehouses
(and the cooking cycle in batch smokehouses) are not a source of combustion
compounds [e.g., polycyclic aromatic hydrocarbons (PAH)]; they are, however, a
source of odor including small amounts of VOC as a result of the volatilization
of organic compounds contained in the meat or the smoke previously applied to
the meat (AP-42, 2004).
that have been qualitatively identified as potential emissions include organic sul-
phides, disulfides, C-4 to C-7 aldehydes, trimethylamine, C-4 amines, quinoline,
dimethyl pyrazine, other pyrazines, and C-3 to C-6 organic acids. In addition,
lesser amounts of C-4 to C-7 alcohols, ketones, aliphatic hydrocarbons, and aro-
matic compounds are potentially emitted.
Fish processing includes both the canning of fish for human consumption and
the production of fish by-products such as meal and oil. Although smoke and
particulates may be a problem, odors are the most objectionable emissions from
fish processing plants. The fish in the by-products segment is often in a further
state of decomposition, thus causing more odor problems than canning itself. The
fish meal driers form part of the fish by-products segment and are the largest odor
source. In addition, the odorous gases from reduction cookers consist primarily
of hydrogen sulfide (H2S) and trimethylamine [(CH3)3N] but are emitted from this
stage in appreciably smaller volumes than from fish meal driers (AP-42, 2004).
7.6. Baking
The oven exhaust from baking industries releases water vapor, CO2, VOC, and
various combustion products. The VOCs are primarily ethanol produced by the
yeast during the fermentation process. It should be noted, however, that many
individuals find the odor of fresh baked bread very desirable.
8. CONCLUSIONS
How much odor should a community or individual have to tolerate? This is the
primary question that must be answered before any good odor policy can be
developed. Everyone would probably agree that if one person smells a facility
12 E. Psillakis and V. Gekas
one day out of the year, that facility should not be declared a nuisance. But
several hundred people smelling a facility nearly every day throughout the year
would be considered a nuisance. The reality at most sites lies somewhere in the
middle. A definition of nuisance must be established that takes into account
odor intensity and frequency, meaning how bad and how often. An odor nui-
sance also may have to consider a relationship between the number of people
annoyed. Included in this decision also might be the economic impact of odor
control on the food industry.
10. REFERENCES
AP-42 (2004) “Compilation of Air Pollutant Emission Factors, AP-42, Fifth Edition, Volume I:
Stationary Point and Area Sources.” U.S. Environmental Protection Agency. http://www.epa.
gov/ttn/ chief/ap42/index.html
Both, R. (2001) “Directive on odor in ambient air: an established system of odor measurement and
odor regulation in Germany.” Water Science & Technology 44 (9), 119–126.
Bliss, P.J., Schulz, T.J., Senger, T., Kaye, R.B. (1996) “Odor measurement—Factors affecting olfac-
tometry panel performance.” Water Science and Technology, 34 (3–4), 549–556.
Davies, T., Konisky, D.M. (2000) “Environmental Implications of the Foodservice and Food Retail
Industries.” Discussion Paper 00–11, Resources for the Future, Washington DC. http://www.rff.
org [2004. Nov 23]
Environmental Protection Agency, Ireland (2001) “Odor Impacts and Odor Emission Control Mea-
sures for Intensive Agriculture.” Environmental Research R&D REPORT SERIES No. 14.
http://www.epa.ie/pubs/
Federal Protection Act for Ambient Air (1974/1990) Act on the Prevention of Harmful Effects on
the Environment Caused by Air Pollution, Noise, Vibration and Similar Phenomena (Federal
Emission Control Act, Bundes-Immissionschutzgesetz—BImSchG) Federal Ministry for Envi-
ronment, Nature Conservation and Reactor Safety, Bonn (BGBl. I p.880). (available in English).
Frechen, F.-B. (1994) “Odor emissions from wastewater treatment plants—recent German experi-
ences.” Water Science and Technology, 30 (4), 35–46.
Frechen, F.-B. (2000) “Odor measurement policy in Germany.” Water Science and Technology, 41
(6), 17–24.
Frechen, F.-B. (2001) “Regulation and policies.” In Stuetz, R.M., Frechen, F.-B. (eds), Odours in
Wastewater, Treatment: Measurement, Modeling and Control. London: IWA publishing.
Gostelow, P., Parsons, S.A., Stuetz, R.M. (2001) “Odor measurements for sewage treatment works.”
Water Research, 35 (3), 579–597.
Guideline VDI 3883 Part 1 (1997). Effects and Assessment of Odors. Psychometric Assessment of
Odor Annoyance Questionnaires. Dósseldorf (German/English).
Odor Problems in the Food Industry 13
Guideline VDI 3940 (1993). Determination of Odorants in Ambient Air by Field Inspections.
Düsseldorf. (German/English).
Ishaque, M., Bisaillon, J.G., Beaudet, R., Sylvestre, M. (1985) “Degradation of phenolic compounds
by microorganisms indigenous to swine waste.” Agricultural Wastes, 13 (3), 229–235.
ISO 5492:1992 (1992) “Sensory analysis—vocabulary” International Organization for Standardization.
Jacob, T.J.C., Fraser, C., Wang, L., Walker, V., O’Connor, S. (2003) “Psychological evaluation
responses to pleasant and maldor stimulation in human subjects; adaptation, dose response
and gender differences.” International Journal of Psychophysiology, 48 (1), 67–80.
Lancet, D. “Olfaction—The strong scent to success.” Nature 351 (6324), 275–276.
Mahin, T.D. (2001) “Comparison of different approaches used to regulate odors around the world.”
Water Science and Technology, 44 (9), 87–102.
Pearce T.C. (1997) “Computational parallels between the biological olfactory pathway and its analogue.
The Electronic Nose .1. Biological olfaction.” Biosystems, 41 (1), 43–67.
Radon K., Peters A., Praml G., Ehrenstein V., Schulze A., Hehl O., Nowak D. (2004) “Livestock
odors and quality of life of neighboring residents.” Annals of Agricultural and Environmental
Medicine 11 (1), 59–62.
Rosenkranz, H.S., Cunningham, A.R. (2003) “Environmental odors and health hazards.” The
Science of the Total Environment, 313 (1–3), 15–24.
Salter, J. (2000) “The legal context of odor annoyance.” Proceedings of the International Meeting on
Odor Measurement and Modeling, Odor 1, Cranfield University.
Schiffman, S.S., Sattely Miller, E.A., Suggs, M.S., Graham, B.G. (1995) “The effect of environmental
odors, emanating from commercial swine operations on the mood of nearby residents.” Brain
Research Bulletin 37 (4), 369–375.
Skinner, J.A., Lewis, K.A., Bardon, K.S., Tucker, P., Catt, J.A., Chamber, (1997) “An overview of
the environmental impact of agriculture in the U.K.” Journal of Environmental Management,
50 (2), 111–128.
Sohn, J.H., Smith, R., Yoong, E., Leis, J., Galvin G. (2003) “Quantification of odors from piggery
effluent ponds using an electronic nose and an artificial neural network.” Biosystems Engineering
86 (4), 399–410
Stuetz, R.M., Gostelow, P., Burgess, J.E. (2001) “Odor perception.” In: Stuetz, R.M., Frechen, F.-B. (eds),
Odours in Wastewater Treatment: Measurement, Modeling and Control. London, IWA publishing.
Stumn, W., Morgan J.J. (1996) Aquatic Chemistry: Chemical Equilibria and Rates in Natural Waters.
Third Edition. Environmental Science and Technology, New York, John Wiley & Sons, Inc.
Technical Guidance Note H4, Integrated Pollution Prevention and Control (IPPC), Horizontal Guid-
ance for Odor, Part 2 Assessment and Control, Environment Agency, UK, 2002. Available on
www.environment-agency.gov.uk
Technical Instruction on Air Quality Control (1986) (Erste Allgemeine Verwaltungsvorschrift zum
Bundes-Immissionsschutzgesetz – Technische Anleitung zur Reinhaltung der Luft – TA Luft)
Federal Ministry for Environment, Nature Conservation and Reactor Safety, Bonn (GMBl. p. 95).
(available in English).
US Environmental Protection Agency (2001) “Emissions from animal feeding operations, draft.”
EPA Contract No. 68-D6-0011, Emission Standards Division, Office of Air Quality Planning
and Standards, Research Triangle Park, NC.
World Health Organization (1987) “Air quality guidelines for europe.” World Health Organization
Regional Publications, European Series No 23.
World Health Organization (2000) “Air quality guidelines for Europe. Second edition”. World Health
Organization Regional Publications, European Series No 91.
Zald, D.H., Pardo, J.V. (2000) “Functional neuroimaging of the olfactory system in humans.” Inter-
national Journal of Psychophysiology, 36 (2), 1165–1181.
Zhu, J. (2000) “A review of microbiology in swine manure odor control.” Agriculture, Ecosystems &
Environment, 78 (2), 93–106.
2
Odor Measurement
Elefteria Psillakis
1. INTRODUCTION
Predict odor impact in the vicinity of an operation for odor impact assess-
ment purposes.
Provide information on the strength and intensity of odors.
Identify the causes of an odor problem and quantify the scale of odor
emission from a particular source.
Measure/evaluate the performance of an odor control technology imple-
mented by a company.
However, odors are temporal and spatially dimensioned and can be considered
to be one of the most difficult challenges for scientists to investigate. A person’s
response to an odor is highly subjective: different people find different odors
offensive and at different concentrations given that physiologically odor recogni-
tion is associated with the emotional center of brain. Furthermore, some of the
odorous compounds can be detected by the human nose in very low concentra-
tions (e.g., hydrogen sulfide) while others cannot be detected even at very high
concentrations (e.g., methane). This is further complicated by the fact that some
combinations of compounds may be more odorous than the sum of the indi-
vidual gases. Determining the impact area of odorous gases also is very difficult.
As wind direction and speed change, the odor impact area and intensity change.
Depending on specific conditions, odorous gases can travel several meters or
several kilometers. Gas transmission and impact area also depend on the specific
gas. For these reasons, there is no universally accepted method for the quanti-
fication of odors, and odor measurement often has been regarded as an art as
opposed to a science (Gostelow et al., 2001).
15
16 E. Psillakis
2. SAMPLING
In those parts of the sampling equipment coming into contact with the odor-
ant sample, appropriate materials such as PTFE, PET, FEP, Tedlar™, glass, or
stainless steel must be used. Materials such as brass and silicone or natural rub-
ber must not be used in sampling lines and fittings. In general, representative
odorous samples are collected in the field using special purpose atmospheric
sampling bags, whose quality must conform to criteria such as being odor free,
nonadsoprtive, leak-free, reasonably robust, and having sufficient volumetric
capacity. Commonly, sampling bags are made of Tedlar™, FEP, or the low-cost/
single-use Naoplhan NA™. Every new sampling bag should be tested for their
background odor concentration as well as for leakage. Because of their high cost
it is a common practice to reuse these bags. Nonetheless, where materials are
reused, procedures for cleaning and conditioning must be applied as defined in
certified methods. In addition, the interval between sampling and measurement
should not exceed 30 hr, during which time samples must be kept in the dark and
at a temperature below 25°C (EN 13725,2003).
Sampling bags may be filled using either the direct or indirect sampling
techniques. According to the direct sampling, the bag is filled under pressure
by pumping the air sample into the bag. In indirect sampling, the odor sample
is collected using an odor sampling system, having a vacuum pump and a 12-
volt battery built into a sampling drum as shown in Figure 1. After placing a
sampling bag into the sealed sampling vessel, air is pumped out of the sampling
drum, creating a vacuum inside the drum. Sample air is drawn into the bag by
the pressure difference between the inside and outside of the bag. Because of the
Odor Measurement 17
Tedlar Bag
Vacuum
Switch
Exhaust
risk of contamination the direct method is seldom used and the indirect sam-
pling approach is generally recommended.
Typically a point source will be a stack with a known flow rate such as a dis-
charge stack from abattoir or even a vent from a pig shed. Gaseous samples from
these sources can be collected using a sampling train consisting of a probe, a
delivery pipe, and an optional particulate filter before the collection system (EN
13725:2003) such as the one described in the US EPA Method 0030 (1986). It is
important that air velocity, dimensions of the vent, temperature, and humidity are
measured before a sample is taken. Appropriate guidance regarding the selection
of sampling points and velocity measurement and sampling point location can be
found in ISO Method 9096: 1992(E) (Stationary source emissions—Determina-
tion of concentration and mass flow rate of particulate material in gas-carrying
ducts—Manual gravimetric method), ISO 10780: 1994(E) (Stationary source
emissions—Measurement of velocity and volume flow rate of gas streams in
ducts), and Australian Standard AS 4323.1 1995 (Stationary Source Emissions,
Method 1—Selection of Sampling Positions).
Predilution of the sample should be undertaken for samples taken directly
from combustion processes where the air temperature and relative humidity
18 E. Psillakis
Typically an area source will be a water or solid surface such as the water surface
of a slurry storage tank or a cattle feedlot and can be sampled using different
methods such as the portable wind tunnels. Area sources can be distinguished
in sources with a measurable outward airflow and sources that do emit odor but
have no measurable outward flow.
In the early 1970s, Lindvall (1970) introduced an odor emission hood used in the
comparison study of odor strengths from different sources. Later, Lockyer (1984)
designed a wind tunnel system to measure ammonia loses from pastures. Recently,
an improved portable wind tunnel system was developed at the University of New
South Wales (Jiang et al., 1995; Bliss et al., 1995) to measure odors emission rates
from liquid and solid area sources (sewage and industrial wastewater treatment
plants, cattle feedlots, mushroom composting, piggeries, etc). The principle of
the wind tunnel system is that controlled air (recently filtered by activated carbon
through a series of devices) forms a consistent flow over a defined liquid or solid
surface. Convective mass transfer takes place above the surface as odor emission
happens in the natural atmosphere. The odor emissions are then mixed with clean
air and vented out of the hood. A proportion of the mixture is drawn into a sam-
pling bag via Teflon tubing using the sampling vessel. The air velocity used inside
the wind tunnel is 0.3 m/sec, which is the lowest reliable measurable air velocity
directly inside the main section of the wind tunnel. The aerodynamic performance
of the wind tunnel has been validated based on the wind speeds and height where
most complaints occurred. An isometric sketch of a portable wind tunnel sys-
tem is shown at Figure 2. It should be mentioned here that according to the EN
13725:2003 this method is of limited application when the tunnels use unfiltered
atmospheric air as intake, which is not always odor free.
An alternative to wind tunnel systems is the isolation chambers, also called
flux hoods (Klenbusch, 1986; Gholson et al., 1991). According to this method,
during sampling a representative area of the source surface is enclosed by the
isolation flux chamber. Then a controlled flow of clean sweep gas (normally
nitrogen or odor- and hydrocarbon-free air) is released into the chamber and is
used to transport the emission from the surface to be sampled. The performance
of isolation chambers largely depends on the configuration of the enclosure and
operating procedures (Hwang, 1985).
A comparative study on portable wind tunnel system and isolation cham-
ber for the determination of VOCs from real sources (Jiang and Kaye, 1996)
Odor Measurement 19
Mixing Chamber
Main Section
Floating Tubes
(Table 1) reported emission rates that were not in accordance and emphasized
the implications of the Henry’s Law constant for mass transfer processes in the
case of the isolation chamber. The same report concluded that this was due to
the fact that the use of isolation chambers may result in significant underestima-
tions of emission rates for real situations and suggested the use of wind tunnel
systems. Nonetheless, the EN 13725:2003 clearly indicates the need for further
research and standardization.
Table 1. Comparison of Wind Tunnel and Isolation Chamber (Jiang, and Kaye, 1996)
Wind tunnel Isolation chamber
Aerodynamics Parallel and even air flow inside Conditions at the emission
the wind tunnel surface not known
Emissions and air well mixed at the exit Emissions and sweep gas
may not be well mixed
Operating Temperature and relative humidity are Temperature and relative
parameters close to ambient condition during humidity are influenced
sampling by solar heating during
long sampling duration
Air velocity = 0.3 m/s Sweep gas = 5L/min
Stabilization time = 5 min Equilibrium time = 24 min
Sample collection rate = 20 L/min Sample collection rate = 2 L/min
Limitations Sample contamination due to Not for chemicals that have
atmospheric air as intake low Henry’s Law constants
20 E. Psillakis
Although, the use of wind tunnels has been demonstrated for all types of area
sources (Jiang and Kaye, 1996), significant limitations may exist in the cases of
area sources with outward flow. For example, in some instances the placement of
the wind tunnel system may create back pressure, limiting the flow of outward
moving air into the wind tunnel, and leading to an underestimation of odor
emission rates (Jiang and Kaye, 2001). The ideal situation would be to cover the
whole total source (or a large part) with foil. The cover must be left open at some
point in order to allow the air to escape as well as to allow sampling using the
point source apparatus (EN 13725:2003).
Typically building sources, such as chicken and pig sheds, have a number of
openings. Prior to about 10 years ago, little research was undertaken on the
determination of odor emissions from buildings. For building sources, measure-
ments of both odor concentration and air ventilation rate are required. The air
ventilation rate from animal housing is dependent on operational conditions
(e.g., opening or closure of side flaps or shutters) and ambient wind speed and
direction (Jiang and Kaye, 2001).
For animal sheds, odor samples are normally taken from several points
within a shed. Experience indicates that composite sample may be sufficient to
represent a single shed at a particular time. Additional samples can be taken at
different times of the day or week or to understand the fluctuation of the odor
concentration levels within a day or a week. Similarly, sampling may be carried
out for different weeks during the grow-out cycle or for different seasons during
a year or longer.
Emission rates, required for odor impact assessment, are calculates using the
odor concentration measured by olfactometry together with other measured
properties of the emission source and the sampling apparatus. For point sources
the odor emission rate (OER) is calculated using the odor concentration (OC:
ou/m3; see definition in paragraph 3.2) measured by olfactometer and the mea-
sured gas flow rate (Q,m3/s) according to the following equation:
OER = Q × OC
The specific odor emission rate (SOER) may be defined as the quantity (mass)
of odor emitted per unit time from a unit surface area. The quantity of odor
emitted is not determined directly by olfactometry but is calculated from the
concentration of odor (OC: ou/m3) as measured by olfactometry, which is then
multiplied by the volume of air passing through the hood per unit time. The
volume per unit time is calculated from the measured velocity through the wind
Odor Measurement 21
tunnel, which is then multiplied by the known cross-sectional area of the wind
tunnel. Given that A is the area covered by the wind tunnel (m2), then SOER is
calculated by the expression:
Q × OC
SOER =
A
Finally, for building sources the odor emission rate (OER) is calculated
using the odor concentration measured by olfactometer through the door and
window openings. The following equation can be applied and the measured gas
flow rate (Q, m3/s) according to the following equation:
OER = Q × OC
where Q is the gas ventilation rate (m3/s) and OC is the odor concentration mea-
sured by olfactometer (ou/m3).
3. POINTS TO CONSIDER
The odor unit is a difficult unit to define because it relates a physiological effect
to the stimulus that caused it. Odor concentrations derived by threshold olfac-
tometry are dimensionless. The concentrations may be termed threshold odor
numbers (TON) or dilution to threshold (D/T) ratios. Nonetheless, it is becom-
ing increasingly common to envisage odor concentrations as physical concentra-
tions and to express them as odor units per cubic meter (ou/m3). In the United
States, the same measure has been expressed as odor units per cubic foot (ou/ft3).
Numerically, TON, ou/m3, and ou/ft3 are identical, so care must be taken when
the concentration ratios are expressed as a physical concentration.
According to the EN 13725:2003, the European odor unit (ouE) is that
amount of odorant(s) that when evaporated into 1 m3 of neutral gas at standard
conditions elicits a physiological response from a panel (detection threshold)
equivalent to that elicited by one european reference odor mass (EROM) evapo-
rated in 1 m3 of neutral gas at standard conditions. One EROM evaporated into
1 m3 of neutral gas at standard conditions is the mass of substance that will elicit
the D50, which is 50% incidence of effect in physiological response (detection
threshold) assessed by an odor panel in conformity with this standard and has
by definition a concentration of 1 ouE/m3. For n-butanol one EROM is 123 mg.
Evaporated in 1 m3 of neutral gas, at standard conditions, this produces a con-
centration of 0,040 mmole/mole (which is equal to a volume fraction of 40 parts
per billion). There is one relationship between the ouE for the reference odorant
Odor Measurement 23
Table 2. Odorants Associated with Sewage Treatment Works (Abbott, 1993; Bonnin
et al., 1990; Brennan, 1993; Gostelow et al., 2001; Young, 1984)
and that for any mixture of odorants. This relationship is defined only at the D50
physiological response level (detection threshold), where:
1 EROM ≡ 123 mg n-butanol ≡ 1 ouE for the mixture of odorants
This linkage is the basis of traceability of odor units for any odorant to that of
the reference odorant. It effectively expresses odor concentrations in terms of
“n-butanol mass equivalents”. The odor concentration can be assessed only at a
24 E. Psillakis
4. OVERVIEW OF METHODS
There are a number of different methodologies in use for odor analysis. Selection
of a particular method will depend upon the purpose of the measurement, the
frequency of monitoring, sampling location, type of source emission (e.g., point
source), as well as the nature and complexity of the emission.
In general, odor can be “measured” using chemical (analytical) or sensory
methods. It is generally accepted that these categories do not have clear cutoff
points and some assessment methodologies could be considered to fall into more
than one. The analytical techniques include chemical analysis and direct reading
instrumental analysis. Chemical analysis is the indirect assessment involving the
collection of a sample which, when analyzed, will give the concentration of the
various chemical species present. This includes substance-specific wet chemistry
methods, as well as sample collection followed by analysis by means of instruments
such as gas chromatography (GC). Direct reading instrumental analysis provides
information on the concentration of specific chemical species or their concentra-
tions relative to each other. This includes among others portable analyzers (such as
portable GC), the “electronic nose,” as well as colorimetric tubes.
The sensory methods on the other hand relate to human response giving
an assessment of the physiological response to a particular mixture—strength,
quality, characteristics—which provides information on the likely population
response. This is obtained by exposing trained individuals to samples of the
odorous air either in the laboratory or in the field.
5. CHEMICAL ANALYSIS
The detection limits achieved by these techniques are determined by the vol-
ume of gas sampled. Typically limits of detection vary between 0.1 and 0.5 mg/
m3. In order to achieve low detection limits, large volumes of gas need to be sam-
pled; but the sampling flow rate through typical sampling equipment is limited
to about 2 L/min (IPPC H4-Part 2, 2002). Consequently, long sampling times are
required and peaks in concentration will be missed. In addition, the equipment
is fragile and set up can be time consuming.
Gas chromatography (GC) is a widely used analytical technique for the sepa-
ration, identification, and quantification of the components of an odorous air
sample. The use of GC allows chromatographic separation of gaseous and liquid
mixtures into individual components. In general, when a sample mixture is intro-
duced into the heated injection port of a GC, it vaporizes. The gaseous sample
mixes with an inert gas (such as helium) also referred to as carrier gas and passes
through a glass or metal tube (column) that contains an absorbent. Because the
various components of the sample interact with the absorbent (stationary phase)
of column to different degrees, compounds will be released from the tube at dif-
ferent specific times. There is a wide selection of column types from a range of
different manufacturers. Nonetheless, in the case of odorous samples, the choice
of a column can be difficult as there are several options or combinations neces-
sary to obtain the optimum information about an odor’s components. Eventu-
ally the components of the injected sample are separated and exit the column at
different times (called “retention times”). These “elution” times are compared to
those of known compounds (analytical standards), thus allowing to some degree
identification.
Once the compounds are separated, they elute from the column and then
enter a detector. The detector is capable of creating an electronic signal when-
ever the presence of a compound is detected, and in most cases this signal is
linearly related to the concentration of the target analyte in the sample. A variety
of detectors are currently available, including mass spectrometer, flame ioniza-
tion, and photoionization detectors (Hobbs, 2001).
Despite the fact that gas chromatographic analyses result in accurate and
reproducible measurements, however, there are some major limitations. First,
the chemical concentrations corresponding to the odor detection thresholds
26 E. Psillakis
An interesting approach that has been used mainly for the determination of
odor-active compounds in food is gas chromatography-olfactometry (GC-O).
First introduced by Fuller et al. (van Ruth, 2001), GC-O uses the human nose
as a detector and has proven to be a valuable method for monitoring the pres-
ence of an odorant in the effluent of a gas chromatograph. Initially, the GC
Odor Measurement 27
effluent was sniffed and when an odor was perceived a description was given for
each retention time, thus associating odor activity with eluting compounds. The
advantages of coupling GC with olfactometry instead of conventional physi-
cal detectors is that although the latter provide relevant information on volatile
composition, many of them are not as sensitive for odor-active compounds as
the human nose (Ferreira et al., 2003).
Furthermore, GC-O allows differentiation between those volatile compo-
nents with a scent from those that do not have one. Early GC-O devices had seri-
ous problems of reproducibility caused by the discomfort from sniffing hot, dry
effluent gases. In an attempt to minimize the discomfort of sniffing, Dravnieks
and O’Donnell (1971) published a GC-O design in 1971, according to which the
hot GC effluent was combined with humidified air to reduce nasal dehydration.
In general, it is very difficult to judge the sensory relevance of volatiles from
a single GC-O run and the method is limited to the screening of odor-active
volatile compounds, unless the chemical stimuli and the assessors’ responses are
quantified (van Ruth and O’Connor, 2001). It should be kept in mind that in
GC-O, single compounds are assessed, and consequently this method does not
provide information on their behavior in a mixture, although it may indicate
their relevance to odor.
A colorimetric tube is also known as a “length of stain” tube, due to the fact
that the concentration of the chemical being tested produces a color stain in
the analytical material, proportional to the concentration in the air. These tubes
also are called Dräger tubes®, named after the company that introduced them in
1937 (Drägerwerk AG). Colorimetric detector tubes are flame-sealed glass tubes
containing a chemical reagent that reacts with a specific compound or group of
compounds, causing color change. A sample is collected by attaching the detec-
tor tube to a special bellows-type pump that draws a known volume of air with
each stroke. If the target chemical is present, the reagent in the tube changes
color and the length of the color change typically indicates the measured con-
centration. Sampling times are generally short and equipment is portable and
relatively inexpensive, hence they can be used as quick indicators of poor con-
trol to assess a short-term event or for scoping studies. Because they are simple
to use, detector tubes have been used in a broad range of applications. Direct-
reading colorimetric detector tubes also were designed for use in testing workplace
air and for determining compliance with occupational exposure limits, but they
also are used widely to demonstrate compliance with emission limits for spe-
cific substances. Today a considerable number of tubes are available from several
manufacturers covering a wide range of substances, although it is not possible to
differentiate between different chemical species within a generic group. Despite
the simplicity of the method, correct use is essential or results can be highly
misleading. Further, more interference with other chemical species is possible,
28 E. Psillakis
leading to false results. It should be mentioned here that colorimetric tubes are
not suitable for identifying unknown substances, and the correct tube needs to
be selected for a particular situation. Although the limits of detection for this
technique are typically between 0.2 to 1 ppm, detection limit can vary greatly,
depending on the tube used (IPPC H4-Part 2, 2002). According to the manufac-
turers’ literature, the relative standard deviation of detector tubes varies between
5 and 20%. Unused tubes must be stored correctly (some require refrigeration)
and have a finite shelf life, which must be observed.
Portable gold leaf analyzers are frequently used to monitor hydrogen sulfide in
the gas phase. This type of detector utilizes the change in resistance of a gold
film sensor caused by adsorption of H2S molecules, within an output propor-
tional to the H2S concentration. Eventually the gold leaf becomes saturated and
has to be regenerated. A common gold film monitor is the Jerome 631-X H2S
analyzer (Able Instruments and Controls Ltd.), which has a reported detection
limit of the order of 0.003 ppm and can measure up to 0.005 ppm H2S. Sampling
and measurement time depend on the level of sulfides present, but typically it is
less than 1 min.
Paper tape monitors contain a chemically impregnated tape, which when exposed
to a gas sample changes color in direct proportion to the amount of gas present.
A tape is selected that will react with the gas of interest (IPPC H4-Part 2, 2002).
Today a wide range of compounds can be quantified depending on the instru-
ment selected. A particular advantage is that the monitor can be set to sample
at regular intervals, exposing an unreacted section of the tape each time, and
so leaves a permanent record of the concentration of each sample. However,
given that the sampling times can be of the order of minutes rather than sec-
onds, this type of monitor is not recommended when a large number of samples
are required in a short period of time. Furthermore, the equipment is generally
expensive, and although relatively simple to operate, sufficient information relat-
ing to process parameters/activities at the time of measurement must be col-
lected, thus allowing correct interpretation of the concentration data obtained.
Some instruments are affected by the presence of moisture, which limits their use
for stack monitoring.
Odor Measurement 29
Portable GCs also exist and can be used for “fingerprinting,” i.e., to analyze
air samples at the complainant’s location in order to ascertain the identity and
concentration of the main odorous components (Santos and Galceran, 2003).
However, the cost of the instrument and the expertise required for analysis
and subsequent evaluation also limit its use as a “quick check” method for
everyday use.
Portable GCs can be coupled to a variety of detectors (including MS, FID,
PID, etc.) (Santos and Galceran, 2002). It should be mentioned here that these
detectors may be used independently as portable instruments for detecting dif-
ferent groups of compounds in air samples.
Flame ionization detector (FID) is perhaps the most commonly used detector
for odorants, yielding analytical methodologies with a large dynamic range and
limits of detection in the low nanogram range (Hobbs, 2001). The FID detector
is suitable for odorants mostly composed of hydrogen and carbon, and is not
recommended for sulfides, where sensitivity was found to be less satisfactory.
Inside the FID, fuel (H2) and oxidant (O2 in air) are mixed to create and main-
tain a flame. The gaseous samples entering are burned and charged particles are
formed in that combustion process. This creates a current between the detector’s
electrodes, which is measured by the FID.
This device uses ultraviolet light as a means of ionizing an analyte. The ions
produced by this process are collected by electrodes and the current generated
therefore is a measure of the analyte concentration. PID detectors are used to
analyze volatile organic compounds as well as ammonia and hydrogen sulfide.
Given that the PID responds to compounds with a photoionization potential
equal to or less than that of the energy of the source, appropriate setting of the
lamp voltage will exclude direct detection of the major compounds in the air
including water vapor, methane, and carbon dioxide (Hobbs et al., 1995). PID
detectors have a dynamic detecting range of between a few ppb and 10,000 ppm
and offer instant indicative measurements from typically rugged, lightweight,
handheld instrument.
Given that only a small fraction of the analyte molecules are actually ion-
ized in the PID chamber, it is considered to be a nondestructive GC detector.
Therefore, when the inlet of the PID detector is connected to a portable GC,
its exhaust port can be connected to another detector in series such as an FID
or an electron capture detector (ECD). A schematic diagram of a portable GC
equipped with in-series detectors [PID, FID, and dry electrolytic conductivity
30 E. Psillakis
Figure 3. Schematic of PID, FID, and dry electrolytic conductivity detector (DELCD) in series with
a SRI-8610C portable gas chromatograph. (With permission from Koziel et al., 1999).
(DELCD)] is given in Figure 3 (Koziel et al., 1999). In this way data from
different detectors sensitive to different group of analytes can be taken simul-
taneously providing almost real-time analysis and speciation of a wide range
of compounds. It should be mentioned here that the major challenge in this
instrumental setup is to make the design of the ionization chamber and the
downstream connections to the second detector as low volume as possible so
that peaks that have been separated by the GC column do not broaden out
before detection.
6. SENSORY METHODS
Sensory measurements employ the human nose as the odor detector. In this sense,
measurements are directly related to the properties of odors as experienced by
humans and on the relationship between psychological and physical attributes
of odor. The problems of complex mixtures, interactions between components,
and detectability below the threshold of smell become irrelevant as the “total
effect” of the overall odor is measured.
There are many factors other than the properties of the odor sample itself
that may influence the perception of an odor. Principal among these is the vari-
ability in the sense of smell between different observers. This may be overcome
to a certain extent by using a panel of several observers, with the result being
expressed as a measure of the central tendency of the individual results. For
repeatable results, great care must be taken in the presentation of samples to
observers. Factors such as the order in which samples are presented, the envi-
ronment in which the testing takes place, and the flow rate of the carrying gas
stream are all important.
Sensory evaluation techniques can be divided into two categories (Gostelow
et al., 2001):
Subjective evaluation measurements have the advantage of being quick and rela-
tively low cost, given that no special equipment is required. Interpretation of
results is difficult and subjective measurements should be handled with caution
due to the inherent variation in odor perception even for well-trained personnel.
Parameters that may be measured subjectively include odor character, hedonic
tone, and intensity. Indeed, for character and hedonic tone, there are no objective
techniques available with the possible exception of the electronic nose (Laing
et al., 1992; Gostelow et al., 2001).
An example of subjective evaluation measurements is the direct scal-
ing technique, which is actually the oldest technique for measuring sensory
stimuli. According to this method, the odor panelist assigns a number to
the odor-containing sample, relative to the referenced standard. For exam-
ple, an odor sample might be compared to a sample of butanol, a typical
reference standard. This provides a good comparison of odor intensity, but
there is no measure of irritation or how objectionable the odor is (Wise
et al., 2000).
32 E. Psillakis
(hence, the name forced-choice method), using a keyboard. They also indicate
whether their choice was a guess, whether they had an inkling, or whether they
were certain they chose the correct port. Only when the correct port is chosen
and the panelist is certain, the choice is considered to be True.
There are a number of variables involved in olfactometry that will affect measure-
ments. Fatigue from continued exposure to an odor may affect a panelist’s sense of
smell. This phenomenon is called adaptation and may reduce both perceived odor
intensity and perceived odor quality (Dravnieks and Jarke, 1980; Gostelow et al.,
2001). The degree of adaptation will depend on the odor concentration experi-
enced. In general, the weaker the odor concentration of an air sample, the more
does adaptation affect perceived strength. On the other hand, exposure of panel-
ists to strong odors may result in adaptation and affect detection of subsequent
weak odors. Consequently, the presentation schedule of odorous samples may
influence the results. It should be mentioned that in cases where dilutions occur
in a strict order, panelists will begin to expect subsequent samples to be weaker or
stronger and may adjust their responses accordingly. Descending order of presen-
tation may result in olfactory fatigue/adaptation and may obstruct detection of a
weak odor after exposure to a strong odor. In general, presentation of one dilution
series is given either in an ascending or random order of stimuli, thus restraining
adaptation (Dravnieks and Jarke, 1980; Gostelow et al., 2001).
Another beneficial effect of using an ascending schedule of sample presen-
tation is that it also may minimize the effects of adsorption/desorption of mate-
rials commonly leading to sample contamination or alteration. In general, the
use of nonreactive, odor-free materials, minimization of internal surface areas,
and the provision for flushing or easy replacement of flow lines between samples
further reduces problems of adsorption/desorption.
It should be mentioned that the supply of odor-free air is important as any
residual odor in the dilution air would bias the result. The general requirements
for the environment for observations by assessors can be found elsewhere (EN
13725: 2003).
When dynamic olfactometry is used, the flow rates at the sniffing ports can
have a major influence on the reported odor concentration. It should be pointed
out that the airflow is one of the most controversial issues in the effort to reach
international consensus on the standardization of odor measurement.
Other important elements of the test procedure that may influence results
(EN 13725:2003) are the shape/dimensions of the of smelling chamber (olfac-
tometer–nose interface), position of valves of gauge readings on the olfactom-
eter, differences in appearance of sniffing ports where more than one is used,
or even the responses of the operator. It therefore is beneficial if panelists are
isolated from each other, from the olfactometer controls, and from the operator
as shown in Figure 4.
34 E. Psillakis
Figure 4. Representation of a dynamic dilution olfactometer using the ODILE™ of the ODOTEC
Inc. (Montreal, Canada).
An example of a portable device for field sensory measurements that uses the
dilution to threshold approach is the scentometer, originally manufactured
by Barnebey-Cheney Company; it consists of a simple, handheld odor dilu-
tion device used to measure odor concentration in the field. According to this
technique, the person taking measurements breathes through the scentom-
eter. Gases can pass either directly to the nose or pass through an activated
carbon filter. The analyst chooses dilution factor by selecting the size of the
Odor Measurement 35
Dynamic dilution
Yes/no method
A Area
D50 50% of a population that can detect as a sensory stimulus
D/T Dilution to threshold ratio
DELCD Dry electrolytic conductivity
EROM European reference odor mass
ECD EROM European reference odor mass
FID Flame ionization detector
FTIR Fourier transform infrared spectrometry
GC Gas chromatography
GC-O Gas chromatography-olfactometry
LD50 Lethal dose for 50% of a well-defined test population
MS Mass spectometer
OC Odor concentration
OER Odor emission rate
ouE European odor unit
PID Photoionization detector
ppb Parts per billion
ppm Parts per million
ppt Parts per trillion
Q Gas flow or ventilation rate
SOER Specific odor emission rate
TON Threshold odor number
VOC Volatile organic compound
8. REFERENCES
Abbott, J. (1993) “Enclosed wastewater treatment plants-health and safety considerations.” Founda-
tion for Water Research Report FR/W0001.
ASTM Method D1391-78 “Standard method for measurement of odor in atmospheres (dilution
method).” American Society for Testing and Materials (ASTM International) Standards, with-
drawn 1986.
Bockreis, A., Jager, J. (1999) “Odor monitoring by the combination of sensors and neural networks.”
Environmental Modeling and Software, 14, 421–426.
Bonnin, C., Laborie, A., Paillard, H. (1990) “Odor nuisances created by sludge treatment: problems
and solutions.” Water Science and Technology, 22, 65–74.
Bliss, P. J., Schulz, T. J., Senger, T., Kaye, R. B. (1996) “Odor measurement—factors affecting olfac-
tometry panel performance.” Water Science and Technology, 34, 549–556.
Bliss, P.J., Jiang, J.K., Schultz, T.J. (1995) “The development of a sampling technique for determina-
tion of odor emission rate from areal surfaces: II. Mathematical Model.” Journal of Air and
Waste Management Association. 45, 989–994.
Brennan, B. (1993) “Odor nuisance.” Water and Waste Treatment, 36, 30–33.
Dravnieks, A., Jarke, F. (1980) “Odor threshold measurement by dynamic olfactometry: significant
operational variables.” Journal of Air Pollution Control Association, 30, 1284–1289.
38 E. Psillakis
Dravnieks, A., O’Donnell, A. (1971) “Principles and some techniques of high-resolution headspace
analysis.” Journal of Agricultural and Food Chemistry, 19, 1049–1056.
EN 13725(2003). Air Quality-Determination of Odor Concentration by Dynamic Olfactometry. CEN,
Brussels.
Ferreira, V., Pet’ka, J., Aznar, M., Cacho, J. (2003) “Quantitative gas chromatography–olfactometry.
Analytical characteristics of a panel of judges using a simple quantitative scale as gas chroma-
tography detector.” Journal of Chromatography A, 1002, 169–178.
Gholson, A.R., Albritton, J.R., Jayanty, R.K.M., Knoll, J.E., Midgett, M.R. (1991) “Evaluation of
the flux chamber method for measuring volatile organic emissions from quiescent liquid sur-
faces.” Environmental Science and Technology, 21, 519–524.
Gostelow, P., Parsons, S.A. (2000) “Sewage treatment works odor measurements.” Water Science and
Technology, 41(6), 33–40.
Gostelow, P., Parsons, S.A., Stuetz, R.M. (2001) “Odor measurements for sewage treatment works.”
Water Research, 35, 579–597.
Hobbs, P.J., Misselbrook, T.H., Pain, B.F. (1995) “Assessment of odors from livestock wastes by a
photoionisation detector, an electronic nose, olfactometry and gas-chromatography-mass spec-
trometry.” Journal of Agricultural Environmental Research, 60, 137–144.
Hobbs, P. (2001) “Odor analysis by gas chromatography.” In. Stuetz, R.M., Frechen, F.-B. (eds), Odours
in Wastewater Treatment: Measurement, Modeling and Control. London: IWA publishing.
Hwang, S. T. (1985) “Model prediction of volatile emissions: A comparison of several models for pre-
dicting emissions for hazardous waste treatment facilities.” Environmental Progress, 4, 141–144.
Jiang, J.K., Bliss, P.J., Schultz, T.J. (1995) “The development of a sampling technique for determina-
tion of odor emission rate from areal surfaces: I. Aerodynamic performance.” Journal of Air
and Waste Management Association, 45, 917–922.
Jiang, K., Kaye, R. (1996) “Comparison study on portable wind tunnel system and isolation chamber
for determination of VOCs from areal sources.” Water Science and Technology, 34, 583–589.
Jiang, J., Kaye, R. (2001) “Sampling techniques for odor measurement.” In Stuetz, R.M., Frechen,
F.-B. (eds), Odours in Wastewater Treatment: Measurement, Modeling and Control. London:
IWA publishing.
Klenbusch, M. R. (1986) “Measurement of gaseous emission rates from land surfaces using an
emission isolation flux chamber, user’s guide.” EPA/600/8-86/008; US Environmental Protection
Agency.
Koziel, J., Jia, M., Khaled, A., Noah, J., Pawliszyn, J. (1999) “Field air analysis with SPME device.”
Analytica Chimica Acta, 400, 153–162.
IPPC H4-Part2 (2002) Integrated Pollution Prevention and Control (IPPC) Draft: “Horizontal guid-
ance for odor: Part 2—Assessment and control,” Environment Agency. http://www.sepa.org.uk/
pdf/ppc/uktech/ippc_h4_pt2.pdf [2005 Jan 5]
Laing, D.G. Doty, R.L., Breipohl, W. (1992) The Human Sense of Smell. Springer-Verlag Berlin
Heidelberg.
Lindvall. T. (1970) “On sensory evaluation of odorous air pollutant intensities.” Nordisk Hygiejnisk
Tidsskrift, suppl. 2, 1–181.
Lockyer, D.R (1984) “A system for the measurement in the field of losses of ammonia through vola-
tilization.” Journal of the Science of Food and Agriculture, 35, 837–848.
Newby, B.D., McGinley, M.A. (2004) “Ambient odor testing of concentrated animal feeding opera-
tions using field and laboratory olfactometers.” Water Science and Technology, 50(4), 109–114.
Ni, J. Q., Heber, A. J. (2001) “Sampling and measurement of ammonia concentration at animal
facilities— A review.” Proceedings of the American Society of Agricultural Engineers Annual
International Meeting, Sacramento, California (Paper Number: 01–4090).
Pillonel, L., Bosset, J.O., Tabacchi, R. (2002) “Rapid preconcentration and enrichment techniques
for the analysis of food volatile. A review.” Lebensmittel-Wissenschaft und-Technologie, 35(1),
1–14.
Ritter, W.F. (1989) “Odor control of livestock wastes: State-of-the-art in North America.” Journal of
Agricultural Engineering Research, 42, 51–62.
Odor Measurement 39
1. PRECONCENTRATION TECHNIQUES
41
42 E. Psillakis
the outlet stream, either because of saturation within the bed or displacement by
another chemical. Sampling is no longer efficient when this occurs, and as break-
through progresses the sample will be less and less representative of the external
environment (Harper, 2000). Another concern when using sorbent tubes is the
presence of moisture which inhibits and prejudices the profile of adsorbed
species. This can be overcome by using hydrophobic adsorbents, dry purges, and
other techniques (Peng and Batterman, 2000).
Once target compounds are collected on adsorbent, samples can be trans-
ferred to GC for further analysis. This is accomplished by either thermal or liquid
desorption. There are several reports where the ability of absorbents followed
by solvent or thermal desorption and chromatographic analysis to concentrate
volatile organic compounds (VOCs) from air samples has been demonstrated
(Hobbs et al., 1995; Rabaud et al., 2003; Peng and Batterman, 2000). Thermal
desorption is more popular than solvent extraction due to its speed, ease of
desorption, and minimization of artifact formation. It also offers higher sensi-
tivity given that the sample is not diluted as well as higher recoveries for polar
and reactive compounds which can pose problems for whole air samples. None-
theless, sample degradation due to contact with heated transfer line from the
desorber to the GC inlet is possible and carryover from previous samples can be
minimized or even avoided using short-path desorber systems.
Overall, the performance of a sorbent sampling/thermal desorption method
depends on many factors, including the target compounds (e.g., concentration,
species, and mixture), the method (e.g., sorbent selection, procedures for con-
ditioning, desorption, separation, and analysis of VOCs), and environmental
conditions present during sampling (e.g., temperature and humidity) (Peng and
Batterman, 2000).
A more recent trend in the preconcentration of odorous compounds involves
the use of the solid-phase microextraction (SPME) technique. SPME, developed
by Arthur and Pawliszyn in 1990, addressed the need for simple and rapid sample
preparation. The commercially available SPME unit consists of a short length
of narrow diameter fused-silica optical fiber externally coated with a thin film
of a polymeric stationary phase located inside a syringelike SPME holder. The
coated fiber is exposed to the sample where analytes preferentially partition by
adsorption or absorption (depending on the fiber type) from the sample to the
stationary phase and are concentrated. There are two main types of SPME sam-
pling: immersion sampling where the fiber is immersed into the aqueous solution
and headspace sampling where the fiber is exposed to the headspace above the
liquid (or solid) sample (Figure 1). Immersion sampling is widespread in the
SPME approach, but for volatile compounds the headspace mode is preferred as
it results in faster equilibration times and higher selectivity. After sampling for a
well-defined period of time, the fiber is withdrawn and transferred to the heated
injection port of a gas chromatograph (GC) or to a modified high-performance
liquid chromatography (HPLC) rheodyne valve depending on the target analytes
(Eisert and Pawliszyn, 1997).
44 E. Psillakis
(a) (b)
SPME
holder
SPME
fibre
Sample
Figure 1. Schematic representation of the two SPME sampling modes: (a) headspace and (b) immersion
in the case of liquid samples.
GC Gas chromatography
HPLC High-performance liquid chromatography
SPME Solid-phase microextraction
VOCs Volatile organic compounds
Preconcentration Prior to Gas Chromatography 45
3. REFERENCES
Arthur, C.L., Pawliszyn, J. (1990) “Solid-phase microextraction with thermal desorption using silica
optical fibers.” Analytical Chemistry, 62(19), 2145–2148.
Béné, A., Fornage, A., Luisier, J.-L., Pichler, P., Villettaz, J.-C. (2001) “A new method for the rapid
determination of volatile substances: the SPME-direct method: Part I: Apparatus and working
conditions.” Sensors and Actuators B: Chemical, 72(2), 184–187.
Davoli, E., Gangai, M.L., Morselli, L., Tonelli D. (2003) “Characterisation of odorants emissions
from landfills by SPME and GC/MS.” Chemosphere 51(5), 357–368.
Eisert, R., Pawliszyn, J. (1997) “New trends in solid-phase microextraction.” Critical Reviews in
Analytical Chemistry 27(2), 103–135.
Harper, M. (2000) “Sorbent trapping of volatile organic compounds from air.” Journal of Chroma-
tography A, 885(1–2), 129–151.
Hobbs, P., Misselbrook, T. H., Pain, B. F. (1995) “Assessment of odours from livestock wastes by a
photoionization detector, an electronic nose, olfactometry and gas chromatography-mass spec-
trometry.” Journal of Agricultural Engineering Research, 60(2), 137–144.
Hobbs, P. (2001) “Odour analysis by gas chromatography.” In: Stuetz, R.M., Frechen F.-B. (eds),
Odours in Wastewater Treatment. London: IWA Publishing.
Kataoka, H., Lord, H.L., Pawliszyn, J. (2000). “Applications of solid-phase microextraction in food
analysis.” Journal of Chromatography A, 880(1–2), 35–62.
Kataoka, H. (2002) “Automated sample preparation using in-tube solid-phase microextraction and
its application – a review.” Analytical and Bioanalytical Chemistry 373(1–2), 31–45.
Koziel, J., Jia, M. Khaled, A., Noah, J., Pawliszyn, J. (1999) “Field air analysis with SPME device.”
Analytica Chimica Acta, 400(1–3), 153–162.
Peng, C-Y., Batterman, S. (2000) “Performance evaluation of a sorbent tube sampling method using
short path thermal desorption of volatile organic compounds.” Journal of Environmental Moni-
toring, 2(4), 313–324.
Pillonel, L., Bosset, J.O., Tabacchi R. (2002) “Rapid preconcentration and enrichment techniques for
the analysis of food volatile. A review.” Lebensmittel-Wissenschaft und-Technologie, 35(1), 1–14.
Psillakis, E., Mantzavinos, D., Kalogerakis, N. (2003) “Monitoring the sonochemical degradation
ofphthalate esters in water using solid-phase microextraction.” Chemosphere, 54(7), 849–857.
Rabaud, N.E., Ebeler, S.E., Ashbaugh, L.L., Flocchini, R.G. (2003) “Characterization and quanti-
fication of odorous and non-odorous volatile organic compounds near a commercial dairy in
California.” Atmospheric Environment, 37(7), 933–940.
Wardencki, W. (1998) “Problems with the determination of environmental sulphur compounds by
gas chromatography.” Journal of Chromatography A, 793(1), 1–19.
4
The Application of Intelligent
Sensor Array for Air Pollution
Control in the Food Industry
Saverio Mannino, Simona Benedetti, Susanna Buratti,
and Maria Stella Cosio
1. INTRODUCTION
One of the most significant improvements in the food industry during the next
few years is likely to be the development of intelligent sensor array systems that
can give useful information not only about the food quality characteristics but
also on the environment of the food production. For instance, the sensory array
system, often called “electronic nose,” would allow the monitoring of off-odors
and taints, often present at very low levels, or the odor quality of a food from the
raw material to the final product.
47
48 S. Mannino et al.
The sensor array is very far from the human nose, and according to Mielle et al.
(1995) this analytical system is “obviously electronic but not nose.” In fact
the only aspect in common with our odor-sensing organ is its function. There
are striking analogies between the human nose and the electronic nose. Compar-
ing the two is instructive. The human nose uses the lungs to bring the odor to
the epithelium layer; the electronic nose (E-nose) has a pump. The human nose
has mucus, hairs, and membranes to act as filters and concentrators, while the
E-nose has an inlet sampling system that provides sample filtration and condi-
tioning to protect the sensors and enhance selectivity. The human epithelium
contains the olfactory epithelium, which has millions of sensing cells, selected
from 100–200 different genotypes that interact with the odorous molecules in
unique ways. The E-nose has a variety of sensors that interact differently with
the sample. The human receptors convert the chemical responses to electronic
nerve impulses. The unique pattern of nerve impulses is propagated by neurons
through a complex network before reaching the higher brain for interpretation.
Similarly, the chemical sensors in the E-nose react with the sample and produce
The Application of Intelligent Sensor Array 49
electrical signals. A computer reads the unique pattern of signals and interprets
them with some form of “intelligent” pattern classification algorithm.
Several commercial intelligent gas sensor array instruments are now available on
the market that cover a variety of chemical sensor principles, system design, and
data analysis techniques. Operationally, an E-nose is a “sensing system” com-
posed of three parts (Fig. 1): a sampling system (a); an array of chemical gas
sensors (b) producing an array of signals when confronted with a gas, vapor, or
odor; and an appropriate pattern-classification system (c).
The ideal sensors to be integrated in an electronic nose should fulfill the
following criteria (Bartlett et al., 1993): high sensitivity toward chemical com-
pounds, that is, similar to that of the human nose (down to 10−12 g/ml), low sen-
sitivity toward humidity and temperature; medium selectivity, they must respond
to different compounds present in the headspace of the sample; high stability;
high reproducibility and reliability; short reaction and recovery time; robust and
durable; easy calibration; easily processable data output; and small dimensions.
By chemical interaction between odor compounds and the gas sensors the
state of the sensors is altered giving rise to electrical signals that are registered
by the instrument. In this way the signals from the individual sensors repre-
sent a pattern that is unique for the gas mixture measured and is interpreted
by multivariate pattern recognition techniques like, for example, the artificial
neural network. Samples with similar odors generally give rise to similar sensor
response patterns and samples with different odors show differences in their
patterns. The sensors of an electronic nose can respond to both odorous and
odorless volatile compounds.
SENSOR RESPONSES
SINGLE SENSOR CHANGE IN
RESISTANCE
Various kinds of gas sensors are available, but only four technologies are
currently used in commercialized electronic noses: metal oxide semiconductors
(MOS); metal oxide semiconductor field effect transistors (MOSFET); con-
ducting organic polymers (CP); and piezoelectric crystals [bulk acoustic wave
(BAW)]. Others, such as fiberoptic (Dickinson et al., 1996), electrochemical
(Baltruschat et al., 1997), and bimetal sensors, are still in developmental stage
and may be integrated in the next generation of the electronic noses. In all cases
the goal is to create an array of differentially sensitive sensing elements.
Metal oxide semiconductor sensors (MOS) were first used commercially in the
1960s as household gas alarms in Japan under the names of Taguchi (the inven-
tor) or Figaro (the company’s name). These sensors rely on changes of conductiv-
ity induced by the adsorption of gases and subsequent surface reactions (Kohl,
1992). They consist of a ceramic substrate (round or flat) heated by wire and
coated by a metal oxide semiconducting film. The metal oxide coating may be
either of the n-type [mainly tin dioxide, zinc oxide, titanium dioxide, or iron (III)
oxide], which responds to oxidizing compounds, or of the p-type (mainly cobalt
oxide or nickel oxide), which responds to reducing compounds (Mielle, 1996).
The film deposition technique divides each sensor type into thin (6–1000 nm) or
thick (10–300 μm) film MOS sensors. The first one offers a faster response and
significantly higher sensitivity but is much more difficult to manufacture in term
of reproducibility. Therefore, commercially available MOS sensors often are
based on thick film technologies. Due to the high operating temperature (200–
650 °C), the organic volatiles transferred to the surface of the sensors are totally
combusted to carbon dioxide and water, leading to the change in the resistance.
MOS sensors are extremely sensitive to ethanol which blinds them to any other
volatile compound of interest.
Conducting organic polymers (CP) sensors like MOS sensors rely on changes of
resistance by adsorption of gas. These sensors comprise a substrate (such as a fiber-
glass or silicon), a pair of gold-plated electrodes, and a conducting organic polymer
such as polypyrrol, polyaniline, or polythiophene as a sensing element. The polymer
film is deposed by electrochemical deposition between both electrodes previously
fixed to a substrate. When a voltage is passed across the electrodes, a current passes
through the conducting polymer (Amrani et al., 1995). The addition of volatile
compounds to the surface of the sensor alters the electron flow in the system and
therefore the resistance of the sensor. In general, CP sensors show good sensitivities
especially for polar compounds. However, their low operating temperature (<50 °C)
makes them extremely sensitive to moisture. Although such sensors are resistant to
poisoning, they have a lifetime of only about 9–18 months.
Piezoelectric crystal sensors are based on the change of the mass, which may be
measured as a change in resonance frequency. These sensors are made of tiny
disks, usually quartz, lithium niobate (LiNbO3), or lithium tantalite (LiTaO3),
coated with materials such as chromatographic stationary phase, lipids, or any
nonvolatile compounds that are chemically and thermally stable (Guilbault and
Jordan, 1988). When an alternating electrical potential is applied at room tem-
perature, the crystal vibrates at a very stable frequency, defined by its mechanical
properties. Upon exposure to a vapor, the coating adsorbs certain molecules,
which increase the mass of the sensing layer, and hence decrease the resonance
frequency of the crystal. This change may be monitored and related to the vola-
tile present. The crystal may be made to vibrate in a bulk acoustic wave (BAW)
or in a surface acoustic wave (SAW) mode by selecting the appropriate combina-
tion of crystal cut and type of electrode configuration. BAW and SAW sensors
differ in their structure: BAW are three-dimentional waves traveling through the
crystal, while SAW are two-dimentional waves that propagate along the surface
of the crystal at a depth of approximately one wavelength. These devices are
also called “quartz crystal microbalance” (QCM or QMB) because, similar to a
balance, their responses change in proportion to the amount of mass adsorbed.
Since piezoelectric sensors may be coated with an unlimited number of materi-
als, they present the best selectivity. However, the coating technology is not yet
well controlled, which induces poor batch-to-batch reproducibility.
5. DATA PROCESSING
The data processing of the multivariate output data generated by the gas sensor
array signals represents another essential part of the electronic nose concept.
The statistical techniques used are based on commercial or specially designed
52 S. Mannino et al.
Hidden
layer Output
layer w1 Summation
input output
e.g. sensor e.g.conc w2 å
responses
class
w3 Tranfer-function
Weight multiplication
Neuron
7. THE FUTURE
There is a plethora of potential application for the electronic nose today. Odor
control is of increasing importance in our lives, for examples, in cars, trains,
aircraft, inside, and outside of buildings and factories. This environmental appli-
cation area is particularly important because the E-nose can be trained to rec-
ognize hazardous chemicals as well as odors. Furthermore, with respect to the
human nose, the E-nose does not fatigue as easily, can be placed in hazardous
atmospheres, is less costly, and can travel easily into outer space. It also holds the
promise of being much cheaper, smaller, and easier to use and maintain than a
mass spectrometer.
In conclusion, we hope that what has been written could be a contribution
to those in and allied to this field of work.
9. REFERENCES
Amrani, M.E.H., Persaud, K.C., Payne, P.A. (1995) “High-frequency measurements of conducting
polymers : development of a new technique for sensing volatile chemicals.” Meas. Sci. Technol.
6, 1500–1507.
Bachinger T., Martensson P., Mandenius C-F. (in press) Seminars in Food Analysis (vol.2), Chapman &
Hall.
Baltruschat, H., Kamphauser, I., Oelgeklaus, R., Rose, J., Wahlkamp, M. (1997) “Detection of volatile
solvents using potentiodymanic gas sensors,” Analytical Chemistry, 69, 743–748.
Bartlett, P.N., Blair, N. and Gardner, J.W. (1993) Electronic nose. Principles, applications and outlook.
ASIC 15e Colloque, Montpellier, 1993, 478–486.
Beebe K.R., Pell R.J., Seasholtz M.B. (1998) Chemometrics, a practical guide, John Wiley and
Sons, New York.
Benedetti, S., Mannino, S., Sabatini, A.G., Marcazzan, G.L. (2004) “Electronic nose and neural
network use for the classification of honey,” Apidologie 35, 397–402.
Buck, T.M., Allen, F.G., Dalton, M. (1965) “Detection of chemical species by surface effects on metals
and semiconductors,” in T. Bregman and A. Dravnieks (eds.), Surface Effects in Detection, Spartan
Books Inc.
Dickinson, T.A., White, J. Kauer, J.S. and Walt, D.R. (1996) “A chemical detecting system based on
a cross-reactive optical sensor array,” Nature, 382, 697–700.
Dravnieks, A. and Trotter, P.J. (1965) “Polar vapour detection based on thermal modulation of
contact potentials”, J. Sci. Instrum., 42, 624.
Edward, J. S. (1999) EST Public Report. Web site: www.estcal.com
Gardner, J.W. and Bartlett, P.N. (1993) “A brief history of electronic noses”, Sensor and Actuator
B, 18, 211–220.
Gardner, J.W. and Bartlett, P.N. (1992) Pattern recognition in odour sensing. In Sensors and Sensory
systems for an Electronic Nose. Dordrecht: Kluwer Academic Publishers, 212, 161–179.
Gardner, J.W., Bartlett, P.N., Dodd, G.H., Shurmer, H.V. (1990) “The design of an artificial olfactory
system,” in Schild D. (ed.) Chemosensory Information Processing, NATO ASI Series H: Cell
Biology, Vol. 39, Springer, Berlin,1990, 131–173.
Guilbault, G., Jordan, J.M. (1988) “Analytical uses of piezoelectric crystal: a review,” Crit. Rev.
Anal. Chem. 19, 1–28.
Hobbs, P.J., Misselbrook, T.H. and Pain, B.F. (1995) “Assessment of odours from Livestock wastes
by a photoionization detector, an electronic nose, olfactometry and gas chromatography-mass
spectrometry”, J. Agric. Eng. Res. 60, 137–144.
Homan, W.J., Fodisch, H. (in press) Seminars in Food Analysis (vol.2), Chapman & Hall.
Iwanaga S., Sato N., Isegami A., Noro T., Arima H., US pat. 4457161, 1984.
Joreskogand K.G. and Wold H. (eds.) (1982) Systems under indirect observation, Parts I and II,
North Holland, Amsterdam.
Kalman, E-L., Winquist, F. and Lundstrom, I. (1997) “A new pollen detection method based on an
electronic nose“, Atmos. Environ. 31, 1715–1719.
Kohl, D. (1992) “oxidic semiconductor gas sensors”, in Sberveglieri G. (ed.) Gas sensors. Dordrecht:
Kluwer Academic Publishers, 43–88.
Lundstrom, I., Spetz, A., Winquist, F., Ackelid, U. and Sundgren, H. (1990) “Catalytic metals and
field-effect devices useful combination”, Sens. Actuator B, 1, 15–20.
McCarrick, C.W., Ohmer, D.T., Gilliland, L.A., Edwards, P.A. and Mayfield, H.Y. (1996) “Fuel
Identification by Neural Network Analysis of the Response of Vapor-Sensitive Sensor arrays”,
Anal. Chem. 68, 4264–4269.
56 S. Mannino et al.
Meloun M., Militky J., Forina M. (1992) Chemometrics for analytical chemistry. Ellis Horwood,
New York.
Mielle, P. (1996) “Electronic noses: Towards the objective instrumental characterization of food
aroma,” Trends in Food Sci. Technol., 7, 432–438.
Mielle, P., Hivert, B. and Mauvais, G. (1995) “Are gas sensors suitable for on-line monitoring and quan-
tification of volatile compounds ?” In Etievant P. and Schreier P. (eds.) Proceedings of Bioflavour
95 Dijon: INRA editions, 81–84.
Misselbrook, T.H., Hobbs, P.J. and Persaud, K.C. (1997) “Use of an electronic nose to measure odour
concentration following application of cattle slurry to grassland”, J. Agric. Eng. Res. 66, 213–220.
Moncrieff, R.W. (1961) “An instrument for measuring and classifying odours”, J. Appl. Physiol.,16, 742.
NATO Advanced Research Workshop, Reykjavik, Iceland, August 1991, Gardner J.W. and Bartlett
P.N. (eds.), Sensors and Sensory Systems for an Electronic Nose, NATO ASI Series E: Applied
Science, Vol. 212, Kluwer, Dordrecht, 1992.
Persaud, K. and Dodd, G.H. (1982) “Analysis of discrimination mechanisms of the mammalian
olfactory system using a model nose” Nature, 299, 352–355.
Schild, D. (ed.). Chemosensory Information Processing, NATO ASI Series H: Cell Biology, Vol. 39,
Springer, Berlin, 1990.
Stetter J.R., Zaromb S., Findlay M.W. 1984. Sens. Actuator B 6, 269.
Wilkens, W.F. and Hatman, A.D. (1964) “An electronic analog for the olfactory processes”, Ann.
NY Acad. Sci., 116, 608.
5
Electronic-Nose Technology:
Application for Quality
Evaluation in the Fish Industry
Guðrún Ólafsdóttir and Kristberg Kristbergsson
1. INTRODUCTION
The odor of fresh fish is one of the most important quality parameters used to
determine whether fish is acceptable for consumption. The composition of volatile
compounds in fish contributing to the characteristic odors can be determined and
related to quality. Currently there is a need for rapid, automated, and objective
tools for process monitoring and quality assurance of perishable food products.
The possibility of using electronic nose for rapid quality control of fish therefore
is of interest. Knowledge about the composition of the headspace of different fish
products during storage is useful to guide the development of electronic noses for
quality monitoring of fish products. Complicated postmortem processes in the fish
are responsible for the loss of freshness and the onset of spoilage. The spoilage
processes are a combination of physical, chemical, biochemical, and microbio-
logical interactions that are species dependent and additionally various extrinsic
factors such as handling and different storage conditions will further influence the
spoilage pattern and the development of spoilage odors.
The odor of fish can be classified as species-related fresh fish odor, microbial
spoilage odor, oxidized odor, processing odor, and environmentally derived
odor, based on the origin of the volatile compounds (Table 1).
57
58 G. Ólafsdóttir and K. Kristbergsson
600 30
FRESH FISH ALCOHOLS
CONCENTRATION (PPB)
500
400 20
SWEET ESTERS
AROMATICS
300
200 10
SHORT CHAIN
ALCOHOLS
100 5
FRESH FISH CARBONYLS
DIENALS
1 4 8 12 16 20 24 28 32
PUTRID
SWEET
TIME (DAYS)
FRESH
STALE
FLAT
Class of
chemical Examples of Aroma Odor threshold
Fish odor species compounds description in water
(Continued )
60 G. Ólafsdóttir and K. Kristbergsson
Although the importance of cultured fish is increasing the fish trade is largely
based on wild fish that has been harvested by various catching methods. Conse-
quently, the quality of fish as a raw material for food production is very variable.
Therefore, the need for rapid evaluation of fish quality is even greater than for
other food commodities where control of the total food chain is better.
Chemical measurements of total volatile bases and total microbial counts
are used in the industry to verify the freshness of fish according to regulations.
Alternative techniques have been developed to monitor changes occurring
postmortem in fish. Detection of ATP metabolites and physical measurements
evaluating changes in texture, microstructure, electrical properties, and color are
well established and newer technologies based on image analysis and spectro-
scopic methods have shown promising results (Olafsdottir et al., 1997c, 2004;
Di Natale, 2003). However, none of these methods have been widely imple-
mented in the industry. This still leaves sensory assessment as the leading method
for freshness evaluation in the fish industry to give the best description of the
actual consumer preferences. Sensory attributes related to appearance, texture,
smell, color, defects, and handling were considered very important in quality con-
trol according to a survey on the importance of various quality attributes of fish
and detection methods used in the European fish sector (Jørgensen et al., 2003).
The results showed general agreement regarding the need for rapid instrumental
methods to measure the overall concepts freshness and quality of fish. The need to
measure objectively the freshness or quality of perishable food products like fish
is of importance both in process management and in the trade of fish. Remote
buying via the internet has become more common and an objective method to
evaluate the quality would facilitate e-commerce. Additionally, the evaluation of
the freshness of various processed fish products is also of interest for the market-
ability of these products.
Electronic noses have been suggested for various applications related to quality
evaluation of different food such as monitoring freshness and the onset of spoil-
age or bioprocesses of food. Many of these applications are based on detecting
62 G. Ólafsdóttir and K. Kristbergsson
Haddock /freshness, storage time Metaloxide sensors FreshSense Sensory analysis Ólafsson et al. (1992)
Cod /freshness, storage time QMB LibraNose TVB-N Di Natale et al. (2001)
Electrochemical sensors FreshSense Sensory analysis Olafsdottir et al. (2003a)
Cod-fillets/ freshness, storage time QMB LibraNose Di Natale et al. (1996)
Electronic-Nose Technology
Sardine/ freshness, storage time Metalloporphyrins and polymers coated EnQbe Sensory (QIM), color, texture Macagnano et al. (2005)
thickness shear mode resonators:
Capelin / freshness, storage time Electrochemical sensors FreshSense TVB-N / sensory Olafsdottir et al.
(1997a,b, 2000)
Herring / freshness, storage time Electrochemical sensors FreshSense TVB-N / sensory Olafsdottir et al. (1998)
Cod roe / ripening stage Sensory
Redfish / storage time TMA / microbial count / sensory Olafsdottir et al. (2002)
Cod fillets / freshness, storage time TVB-N / microbial count / sensory Olafsdottir et al. (2002)
Herring fillets MOSFET and MOS (Taguchi) Lipid oxidation products Haugen and Undeland
Antioxidants/ sensory analyses (2003)
Tuna Conducting polymers AromaScan Microbial count / sensory Du et al. (2001)
Tuna Conducting polymers e-Nose 4000 Microbial count / sensory Newman et al. (1999)
Trout /freshness, storage time Amperometric sensors/ heating Prototype No reference Schweizer-Berberich
filaments et al. (1994)
Salmon fillets Conducting polymers e-Nose 4000 Sensory Luzuriaga and Balaban.
(1999b)
Salmon fillets / freshness, Conducting polymers AromaScan Microbial count / sensory Du et al. (2002)
storage time
Shrimp / freshness storage time Conducting polymers e-Nose 4000 NH3 - electrode / sensory Luzuriaga and Balaban
(1999a)
Smoked salmon / freshness Metal oxide sensors FishNose Sensory/microbial counts Olafsdottir et al. (2005)
storage time AlphaMOS
Cod, sardine, yellow tail/ TMA/ Semiconductor gas sensor Prototype K value, TVB-N Egashira et al. (1994)
DMA/NH3 detection (TiO2 – In2O3-MgO Ohashi et al. (1991)
63
64 G. Ólafsdóttir and K. Kristbergsson
cellulose acetate matrix and a simple illumination source The TVB-N levels
released from orange roughy and black scabbard, deepwater fish, were moni-
tored by color changes in the sensors.
Figure 2. Quality indicators of fish. Main classes of compounds contributing to microbial spoilage
odor of fish.
Electronic-Nose Technology 65
the main classes of compounds causing spoilage odor have been analyzed by the
electronic nose FreshSense (Olafsdottir et al., 1998, 2002). The results showed
that the electrochemical gas sensors (Dräger, Germany: CO, SO2; City Tech-
nology, Britain: NH3) in the FreshSense had different selectivity and sensitivity
toward the compounds selected from these classes. The CO sensor was sensitive
toward alcohols, aldehydes, and esters; the NH3 sensor detected amines like TMA
and ammonia; and the SO2 sensor detected volatile sulfides. Therefore, the main
classes of spoilage indicator compounds present in the headspace can be estimated
based on the individual sensor responses.
A storage study of haddock fillets kept in cold storage (0–2°C) for 3, 7, 10, and
14 days was performed to obtain more detailed information about the identity and
the level of the most abundant volatile compounds present in the headspace dur-
ing storage of fish (Olafsdottir et al., 2003b). Analyses of the headspace volatiles
were conducted using air pump headspace sampling and collection of volatiles
by a TENAX preconcentration technique and analysis by gas chromatography
(GC-MS). The results were compared to the individual responses of the FreshSense
electronic nose sensors and sensory odor evaluation of the fillets. The responses of
the electronic nose sensors were comparable to the results of the analysis of vola-
tile compounds obtained by the GC-MS detection (Figure 3). The sum of the GC
responses of individual compounds belonging to each class of compounds like
the alcohols, aldehydes, and esters corresponded to the CO sensor responses of
1200
CO
Alcohols, aldehydes
and esters SO2
Response sensors (nA)
NH3
800
arbitrary units
Amines (TMA)
400
Sulfur
compounds
0
0 5 10 15
0 5 10 15
(a) Days (b) Days
Figure 3. Sum of the peak areas of compounds representing the three different classes of compounds
detected by GC in the headspace of haddock fillets during storage in ice (a) and responses of the CO,
SO2, and NH3 sensors toward haddock fillets during storage in ice (b) (from: Olafsdottir, 2003)
66 G. Ólafsdóttir and K. Kristbergsson
the electronic nose. The sum of the amines and the sulfur compounds detected by
GC-MS corresponded to the responses of the NH3 and SO2 sensors, respectively.
The response of the CO sensor was the highest and increased early in the
spoilage process while the responses of the NH3 sensor and the SO2 sensors
increased later in the spoilage process. Because of breakthrough of small polar
molecules on the TENAX traps the technique is not suitable for the quantifi-
cation of compounds like ammonia, hydrogen sulfide, methyl mercaptan, and
ethanol that have been known to be present in abundance in the headspace of
spoiled fish. However, the electrochemical sensors can detect these compounds
and therefore the slopes and shapes of the curves are slightly different.
Increasing concentrations of spoilage indicator compounds with storage time
resulted in the development of characteristic odors and simultaneously increased
responses of the electronic nose sensors were observed. On day 3 the odor of the
fresh fillet was very little or neutral and low responses of the sensors were observed.
The first spoilage odors of the fillets were sweetlike odors that were contributed
by the alcohols which give sweet, solventlike odors in combination with aldehydes
giving sweet, rancidlike odors (day 7). The amines contributed to salted fish or
stock fish odor and in combination with the sulfur compounds, cheesy and foul
odors developed and the fillets became stale on day 10. Finally, esters were ana-
lyzed in high levels on day 14 giving characteristic sweet, fruity odors. When these
sweet odors were mixed with the foul smell of sulfur compounds and ammonia-
like stockfish character of the amines the odor of the fillet becomes TMA/ammo-
nialike and sour/putridlike, signaling the overt spoilage.
It is of interest to compare the composition of the headspace of different
fish products during storage to guide the development of an electronic nose for
quality monitoring of fish products. A similar set of sensors can be used for a
variety of fish species that are stored and processed by different methods. The
FreshSense instrument has been used for freshness monitoring of various spe-
cies of fish like haddock, capelin, redfish, and cod that were handled and stored
under different conditions. A similar trend in the responses of the electronic nose
and the development of volatile compounds was observed for the different fish
species (Olafsdottir et al., 1997a, b, 1998, 2000, 2002; DiNatale et al., 2001).
Raw material is often labeled with days from catch; however, because of the
effect of various extrinsic and intrinsic factors, the information about days
from catch is not always reliable for the determination of quality or freshness
of the raw material. Sensory evaluation of fish fillets is more difficult than sensory
evaluation of the whole fish. Therefore, instrumental techniques to evaluate
the freshness quality of the fillets are of special interest. The most obvious
spoilage signs of the fillets are development of spoilage odor, discoloration,
Electronic-Nose Technology 67
10
9
8
Torry sensory score
7
6
5
4 sensory rejection
Torry = 5.5
3
2
MAY 00 MAY 99 SEP 99
1
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Days
Figure 4. Sensory analysis of haddock fillets from storage studies of whole fish [(May 99 (--); Sept
99 (--)] and fillets [May 00 (--)].
68 G. Ólafsdóttir and K. Kristbergsson
Fish stored as fillets at 0–2°C (May 2000) spoiled faster than whole fish and had
a shelf life of only 8–9 days.
The results of the electronic nose measurements and the microbial and
chemical analysis were mostly in agreement with the sensory analysis (Figures 5
and 6). The traditional microbial analysis (TVC) and electronic nose measure-
ments (CO sensor) showed a similar overall trend in the storage studies of whole
fish (May 1999 and Sept 1999) and fillets (May 2000) (Figure 5) and likewise
the results of TVN chemical measurements and the NH3 electronic nose sensor
showed similar overall trends (Figure 6).
The electronic nose measurements showed that the responses of the sen-
sors were highest for the (May 2000) samples indicating the most rapid spoilage
for fish stored as fillets. The results from the sensory analysis, the TVN (total
volatile bases) measurements, and the NH3 sensor showed that recently spawned
fish from the spring season (May 1999), stored as whole fish, had higher spoilage
rate than fish from the autumn season (Sep 1999). Contradictory to these results
the TVC and CO measurements indicated that the spoilage rate appeared to
be higher in the fall (Sep 1999) than in the spring (May 1999). However, based
on the results of sensory analysis it appears that the spoilage potential of the
microflora might have been greater in the spring. This suggests that the role of
specific spoilage organisms producing offensive odors should be emphasized and
the validity of TVC measurement may be questionable as has been suggested by
other researchers (Gram et al., 2002).
When the end of the shelf life was reached, as indicated by sensory analysis,
the responses of the NH3 and SO2 sensors detecting microbially produced amines
and sulfur compounds start to increase rapidly in all experiments. The conclusions
drawn from these results indicate that spoilage patterns are different depending on
season and fillets spoil faster than whole fish, as was to be expected.
9 900
8 MAY 00 800
Response CO sensor (nA)
MAY 00
7 700
SEP 99
TVC log cfu/g
6 600
5 500 SEP 99
4 MAY 99 400
3 300
MAY 99
2 200
1 100
0 0
0 5 10 15 20 0 5 10 15 20
Days Days
Figure 5. Microbial analysis (TVC) and electronic nose measurements (CO sensor) of haddock fil-
lets from storage studies at 0–2 °C for up to 15 days of whole fish [May 99 (--); Sept 99 (--)] and
fillets [May 00 (--)].
Electronic-Nose Technology 69
70 100
MAY 00
90
50 70
MAY 99
MAY 99 60
40
50
30 40
30
20 SEP 99 20
SEP 99
10 10
0
0 0 5 10 15 20
0 5 10 15 20 Days
Days
Figure 6. Chemical measurements of total volatile bases (TVB-N) and electronic nose measure-
ments (NH3 sensor) of haddock fillets stored at 0–2 °C for up to 15 days, from storage studies of
whole fish [May 99 (--); Sept 99 (--)] and fillets [May 00 (--)].
TPA
1/MAY 00
H2S
15/MAY00
0.5 Whole fish
NH3 Fillets
RT-Meter 6/MAY 00 SO3 15/MAY99
1/SEPT99 MAY 1999 MAY 2000
Torry TVB13/MAY99
4/MAY99 TMA
4/MAY 00
0 13/MAY00
8/MAY00
8/MAY99 11/MAY99 11/MAY00
4/SEP99 log TVC
6/MAY99 logH2S
CO
6/SEP99
−0.5 8/SEP99
11/SEP99
15/SEP99
18/SEP99 Whole fish
13/SEP99
SEP 1999
−1.0 PC1
−0.5 0 0.5 1.0
all data, X-expl: 60%, 18%
Figure 7. PCA showing a biplot of sample scores and variable loadings of all data from three stor-
age studies of haddock as whole fish (May 1999 and Sept 1999) and fillets (May 2000) stored at 0–2
°C for up to 15 days. The samples are labeled with storage days and time of the experiment and the
variable loadings are labeled as follows: E-nose: CO, NH3, SO2, and H2S; chemical: TVB/TMA;
microbial: log H2S, log TVC; texture: Firm /TPA.
70 G. Ólafsdóttir and K. Kristbergsson
samples based on the season. The variable loadings show that the CO sensor and
the microbial counts (log H2S and log TVC) are highly correlated and the posi-
tioning of the H2S, SO2, and NH3 sensors on the upper half of the plot show the
contribution of these sensors to discriminate samples from the spring season in
combination with the texture measurement (Firm). Fish spoils faster in the spring
because the fish is recently spawned and the texture of the muscle is softer and
more gaping than in the fall.
Different spoilage patterns are evident depending on the season indicating
that days of storage is not a good estimate of freshness quality. The E-nose sen-
sor responses appear to give information on the quality of fish related to sensory
scores, microbial growth (TVC), and spoilage indicators like TVB-N.
7. CONCLUSIONS
Rapid measurements of the headspace of fish with an electronic nose has the
potential to detect the freshness or quality stage of the fish, if the sensor array
can detect the respective indicator compounds that truly represent the condition
of the fish. The high sensitivity of the electronic nose sensors toward the very
volatile degradation compounds, like alcohols, sulfur compounds, and amines
signaling overt spoilage suggests that the E-nose may be promising for moni-
toring off odors caused by fish and animal waste and off odors caused by the
processing of fish meal or dried products when the raw material is not of optimal
freshness.
8. ACKNOWLEDGMENTS
The case studies were part of projects financed by The Icelandic Centre for
Research: Accurate predictive models—The effect of temperature fluctuations
on microbial growth and metabolites in fish and The European Commission:
MUSTEC project FAIR98 4076.
9. RECOMMENDED READING
The following papers give a good overview of the application of electronic noses
for monitoring food (Garcia-González and Aparicio, 2002; Harper, 2001; Haugen,
2001; Bartlett et al., 1997; Gardner and Bartlett, 1999; Schaller et al., 1998) and Jurs
et al. (2000) give an excellent overview of data processing techniques for electronic
noses.
Information on the website of The NOSE II 2nd Network on Artificial Olfac-
tory Sensing has the aim to stimulate information exchange between scientists,
manufacturers, and end users in Europe in order to develop synergy in the sensor
Electronic-Nose Technology 71
10. REFERENCES
Ahmed, A., Matches, J. R. (1983) “Alcohol production by fish spoilage bacteria.” Journal of Food
Protein 46, 1055–1059.
Bartlett, H.N., Elliott, J.M., Gardner, J.W. (1997). Electronic noses and their application in the food
industry, Food Technology 51(12): 44–48.
Boothe, A.A.J., Arnold, J.W. (2002). ‘Electronic nose analysis of volatile compounds from poultry
meat samples, fresh and after refrigerated storage’, J. Sci. Food Agric.82, 3, 315–322.
Buttery, R. G., Guadagni, D.G., Ling, L.C., Seifert, R.M., Lipton, W. (1976). Additional volatile
components of cabbage, broccoli and cauliflower. J. Agric. Food Chem., 24 (4), 829–832.
Byrne, L.; Lau, K.T.; Diamond, D. (2003). Development of pH sensitive films for monitoring spoil-
age volatiles released into packaged fish headspace. Irish Journal of Agricultural and Food
Research, 42(1) 119–129.
Deng, Z., Stone, D.C., Thompson, M. (1996). Selective detection of aroma components by acoustic
wave sensors coated with conducting polymer films. Analyst 121,671–679.
Di Natale, C., Brunink, J.A.J., Bungaro, F., Davide, F., d’Amico, A., Paolesse, R., Boschi, T., Faccio,
M., Ferri, G. (1996), ‘Recognition of fish storage time by a metalloporphyrins-coated QMB sen-
sor array’, Meas. Sci. Technol. 7: 1103–1114.
Di Natale, C., Olafsdottir, G., Einarsson, S., Mantini, A., Martinelli, E., Paolesse, R., Falconi, C.,
D’Amico, A., (2001). Comparison and integration of different electronic noses for the evalua-
tion of freshness of cod fish fillets. Sensors and Actuators B77,572–578.
Di Natale, C. (2003), Data Fusion in MUSTEC: Towards the Definition of an Artificial Quality Index. In:
Luten J B, Oehlenschlager J and Olafsdottir G (Eds), Quality of Fish from Catch to Consumer: Label-
ing, Monitoring and Traceability, Wageningen Academic Publishers, The Netherlands, 273–280.
Du, W.X., Kim, J., Huang, T.S., Marshall, M.R., Wei, C.I. (2001), ‘Microbiological, sensory and
electronic nose evaluations of yellowfin tuna under various storage conditions’, J Food Prot. 64
(12): 2027–2036.
Du, W.X., Lin, C.M., Huang, T.S., Kim, J., Marshall, M.R., Wei, C.L. (2002), ‘Potential application
of the electronic nose for quality assessment of salmon fillets under various storage conditions’,
J Food Sci. 67 (1): 307–313.
Egashira, M., Shimizu, Y., Takao, Y. (1994), Fish Freshness detection by semiconductor gas sensors,
Olfaction and Taste XI. Proc. Int. Sym. 11, 715–719.
Egashira, M., Shimizu, Y., Takao, Y. (1990) Trimethylamine sensor based on semiconductive metal-
oxides for detection of fish freshness. Sensors and Actuators B-Chem 1 (1–6): 108–112.
Fazzalari, F.A. (1978). Compilation of Odor and Taste Threshold Value Data. American Society for
Testing and Materials, Philadelphia.
Gacrcía-González, D. L. and Aparicio, R. (2002). Sensors: From biosensors to the electronic nose.
Grasas y Aceites, 53(1) 96–114.
Gardner, J.W., Bartlett, P.N. (1999), Electronic noses. Principles and applications, Oxford University
Press. Oxford, England.
Gram, L., Ravn, L., Rasch, M., Bruhn, J.B., Christensen, A.B., Givskov, M. (2002) Food spoilage
- interactions between food spoilage bacteria. Int. J. Food Microbiology 78 (1–2): 79–97.
Harper, W.J. (2001), Strengths and weaknesses of the electronic nose. In: Rouseff and Cadwallader
(Eds) Headspace Analysis of Food and Flavors: Theory and Practice. Kluwer Academic / Plenium
Publishers, New York pp 59–71.
Haugen, J.E. (2001). Electronic noses in food analysis. In: Rouseff and Cadwallader (Eds). Headspace
Analysis of Food and Flavors: Theory and Practice. Kluwer Academic / Plenium Publishers,
New York pp 43–57.
72 G. Ólafsdóttir and K. Kristbergsson
Haugen, J.E., Undeland, I. (2003), Lipid oxidation in herring fillets (Clupea harengus) during ice
storage measured by a commercial hybrid gas-sensor array system, J Agric. Food Chem. 51, 3,
752–759.
Herbert, R.A., Ellis, J.R., Shewan, J.M. (1975), Isolation and identification of the volatile sulfides pro-
duced during chill-storage of North Sea Cod (Gadus morhua). J.Sci. Food Agric., 26, 1195–1202.
Joffraud, J.J., Leroi, F., Roy, C., Berdague, J.L. (2001), Characterisation of volatile compounds pro-
duced by bacteria isolated from the spoilage flora of cold-smoked salmon, Int. J. Food Micro-
biol. 66, 3, 175–184.
Jørgensen, B.M., Oehlenschläger, J., Olafsdottir, G., Tryggvadóttir, S.V., Careche, M., Heia, K., Nes-
vadba, P., Nunes, M.L., Poli, B.M., Di Natale, C., Pérez-Villarreal, B., Ballo, H., Luten, J.B., Smelt,
A., Denton, W., Bossier, P., Hattula, T., Åkesson, G. (2003), A study of the attitudes of the
European fish sector towards quality monitoring and labeling. In: Luten, J.B., Oehlenschläger,
J. and Olafsdottir, G. (Eds) Quality of Fish from Catch to Consumer: Labeling, Monitoring and
Traceability, Wageningen Academic Publishers, The Netherlands, 57–71.
Jørgensen, L.V., Huss, H.H., Dalgaard, P. (2001). Significance of volatile compounds produced by
spoilage bacteria in vacuum-packed cold-smoked salmon (Salmo salar) analyzed by GC-MS
and multivariate regression. J. Agric. Food Chem. 49, 2376–2381.
Josephson, D.B. (1991). Seafood. In Volatile compounds in foods and beverages, H. Maarse (Ed)
Marcel Dekker, New York, 179–202.
Josephson, D.B., Lindsay, R.C., Stuiber, D.C. (1984), Variations in the occurrences of enzymically derived
volatile aroma compounds in salt- and freshwater fish. J. Agric. Food Chem., 32, 1344–1347.
Josephson, D.B., Lindsay, R., Olafsdottir, G. (1986). Measurement of Volatile Aroma Constituents
as a Means for Following Sensory Deterioration of Fresh Fish and Fishery Products. In Pro-
ceedings of an International Symposium, University of Alaska Sea Grant Program, Anchorage,
Alaska, U.S.A., Kramer DE and Liston J (Ed.), Elsevier Science Publishers B.V., Amsterdam
pp. 27–47.
Jurs, P.C., Bakken, G.A., McClelland, H.E. (2000) Computational Methods for the Analysis of
Chemical Sensor Array Data from Volatile Analytes. Chem. Rev. 100 pp. 2649–2678.
Kamiya, A., Ose, Y. (1984). Study of the odorous compounds produced by putrefaction of foods.
V. Fatty acids, sulfur compounds and amines. Chromat., 292, 383–391.
Kawai, T. (1996). Fish flavor. Critical Reviews in Food Science and Nutrition, 36, 3, 257–298.
Kelleher, S.D., Zall, R.R. (1983). Ethanol accumulation in muscle tissue as a chemical indicator of
fish spoilage. J. Food Biochem., 7, 87–91.
Keshri, G., Voysey, P., Magan, N. (2002). Early detection of spoilage moulds in Keshri bread using
volatile production patterns and quantitative enzyme assays. J. Appl. Microbiol. 92, 165–172.
Lindsay, R.C., Josephson, D.B. and Olafsdottir, G. (1986). Chemical and biochemical indices for
assessing the quality of fish packaged in controlled atmospheres. In Proceedings of an Inter-
national Symposium, University of Alaska Sea Grant Program, Anchorage, Alaska, U.S.A.,
Kramer, D.E. and Liston, J. (Ed.), Elsevier Science Publishers B.V., Amsterdam pp 221–234.
Luzuriaga, D.A., Balaban, M.O. (1999a). Evaluation of odor decomposition in raw and cooked
shrimp: Correlation of electronic nose readings, odor sensory evaluation and ammonia levels.
In: Hurst, W.J. (Ed), Electronic nose and sensor array based system, Design and Applications.
Proceedings of the 5th International Symposium on Olfaction and the Electronic Nose. Tech-
nomic Publication Company, Inc, USA, pp.177–184.
Luzuriaga, D.A., Balaban, M.O. (1999b). Electronic nose odor evaluation of salmon fillets stored
at different temperature. In: Hurst, W.J. (Ed) Electronic nose and sensor array based system,
Design and Applications. Proceedings of the 5th International Symposium on Olfaction and the
Electronic Nose. Technomic Publication Company, Inc, USA, pp.162–169.
Macagnano, A., Careche M., Herrero, A., Paolesse, R., Martinelli, E., Pennazza G., Carmonae, P.,
D’Amico A., Di Natale, C. (2005), A model to predict fish quality from instrumental features.
Sensors and Actuators B 111–112 293–298.
McEntegart, C.M., Penrose, W.R., Strathmann, S., Stetter, J.R. (2000). Detection and discrimination
of coliform bacteria with gas sensor arrays. Sensors and Actuators B 70, 170–176.
Electronic-Nose Technology 73
McGill, A.S., Hardy, T., Burt, R.J. and Gunstone, F. D. (1974). Hept-cis-4-enal and its contribution
to the off-flavor of cold-stored cod. J. Sci. Food Agric., 25, 1477–1481.
Newman, D.J., Luzuriaga, D.A., Balaban, M.O. (1999). Odor and Microbiological Evaluation of
Raw Tuna: Correlation of Sensory and Electronic Nose Data. In: Hurst, W.J. (Ed), Electronic
nose and sensor array based system, Design and Applications. Proceedings of the 5th Interna-
tional Symposium on Olfaction and the Electronic Nose. Technomic Publication Company, Inc,
USA, pp.170–176.
Oehlenschläger, J. (1992). Evaluation of some well established and some underrated indices for the
determination of freshness and/or spoilage of ice stored wet fish’. In: Huss, H.H., Jakobsen, M.,
Liston, J. (Eds), Quality Assurance in the Fish Industry, , Elsevier Science Publisher, Amsterdam,
339–350.
Ohashi, E., Takao, Y., Fujita, T., Shimizu, Y., Egashira, M. (1991). Semiconductive trimethylamine
gas sensor for detecting fish freshness, J. Food Sci. 56(5)1275–1278.
Olafsdottir, G., Martinsdóttir, E., Jónsson, E.H. (1997a). Rapid gas sensor measurements to predict
the freshness of capelin (Mallotus villosus). J. Agric. Food Chem. 45(7) 2654–2659.
Olafsdottir, G., Martinsdóttir, E., Jónsson, E.H. (1997b). Gas sensor and GC measurements of
volatile compounds in capelin (Mallotus villosus). In Seafood from Producer to Consumer,
Integrated Approach to Quality. Eds. Luten, J.B., Börresen, T., Oehlenschläger, J. Amsterdam,
Elsevier, pp 507–520.
Olafsdottir, G., Martinsdóttir, E., Oehlenschläger, J., Dalgaard, P., Jensen, B., Undeland, I., Mackie,
I.M., Henehan, G., Nielsen, J., Nilsen, H. (1997c), Methods to evaluate fish freshness in research
and industry. Trends Food Sci. Technol, 8. 258–265.
Olafsdottir, G., Högnadóttir, Á., Martinsdóttir, E. (1998), Application of gas sensors to evaluate
freshness and spoilage of various sea foods. In: Methods to Determine the Freshness of Fish
in Research and Industry, Proceedings of the Final meeting of the Concerted Action “Evalua-
tion of Fish Freshness” AIR3 CT94 2283. Nantes Nov 12–14, 1997. International Institute of
Refrigeration, pp100–109.
Olafsdottir, G. and Fleurence, J. (1998). Evaluation of fish freshness using volatile compounds -
Classification of volatile compounds in fish. In Methods to Determine the Freshness of Fish
in Research and Industry, Proceedings of the Final meeting of the Concerted Action “Evalua-
tion of Fish Freshness” AIR3 CT94 2283. Nantes Nov 12–14, 1997. International Institute of
Refrigeration, 55–69.
Olafsdottir, G., Högnadóttir, Á., Martinsdóttir, E. and Jónsdóttir, H. (2000). Application of an
Electronic Nose to Predict Total Volatile Bases in Capelin (Mallotus villosus) for Fishmeal
Production, J. Agric. Food Chem. 48 ,6, 2353–2359.
Olafsdottir, G., Li, X., Lauzon, H. L., Jonsdottir, R. (2002). Precision and application of electronic
nose measurements for freshness monitoring of redfish (Sebastes marinus) stored in ice and
modified atmosphere bulk storage. JAFP 11, 3/4, 229–249.
Olafsdottir, G., Di Natale, C., Macagnano, A. (2003a). Measurements of quality of fish by electronic
noses. In: Luten, J.B., Oehlenschläger, J. and Olafsdottir, G. (Eds), Quality of Fish from Catch
to Consumer: Labelling, Monitoring and Traceability, Wageningen Academic Publishers, The
Netherlands, 225–234.
Olafsdottir, G., Jonsdottir, R., Tryggvadottir, S.V., Einarsson, S., Lauzon, H.L., (2003b). Prediction
of freshness quality of haddock fillets using electronic nose, texture and TVN measurements
First Joint Trans-Atlantic Fisheries Technology Conference (TAFT) 33rd WEFTA Meeting
and 48th Atlantic Fisheries Technology Conference, 11–14 June 2003, Reykjavik - Iceland 2003
pp. 148–150. (http://vefur.rf.is/TAFT2003/).
Olafsdottir, G. (2003). Developing rapid olfaction arrays for determining fish quality. In Tothill, I.E.
(Ed) Rapid and on-line instrumentation for food quality assurance. Woodhead Publishing Ltd,
Cambridge, England, pp. 339–360.
Olafsdottir, G., Nesvadba, P., Di Natale, C., Careche, M, Oehlenschläger, J., Tryggvadóttir, S.V.,
Schubring, R., Kroeger, M., Heia, K., Esaiassen, M., Macagnano, A., Jørgensen, B.M. (2004).
Multisensor for fish quality determination. Trends in Food Science and Technology 15, 86–93.
74 G. Ólafsdóttir and K. Kristbergsson
Olafsdottir, G., Chanie, E., Westad, F., Jonsdottir, R., Thalman, C., Bazzo, S., Labreche, S., Marcq,
P., Lundby, F., Haugen, J.E. (2005). Prediction of Microbial and Sensory Quality of Cold
Smoked Atlantic Salmon (Salmo salar) by Electronic Nose. J Food Sci 70 (5) 563–574.
Olafsdottir, G. (2005). Volatile compounds as quality indicators of fish during chilled storage: Evalu-
ation of microbial metbolites by an electronic nose. PhD thesis Faculty of Science, University
of Iceland, 219p.
Olafsdottir, G., Lauzon, H., Martinsdottir, E., Oehlenschläger, J., Kristbérgsson, K. (2006). Eval-
uation of shelf-life of superchilled cod (Gadus morhua) fillets and influence of temperature
fluctuations on microbial and chemical quality indicators. J. Food Sci 71 (2) 97–109.
Ólafsson, R., Martinsdóttir, E., Olafsdottir, G., Sigfússon, Þ. I., Gardner, J. W. (1992), Monitoring
of fish freshness using tin oxide sensors. In: Gardner, J.W. and Bartlett, P. N. (Eds), Sensors
and Sensory Systems for an Electronic Nose, Kluwer Academic Publ. 257–272.
Olsson, J., Börjesson, T., Lundstedt, T., Schnürer, J. (2002), Detection and quantification of ochra-
toxin A and deoxyinvalenol in barley grains by GC-MS and electronic nose. Int. J Food Microb.
72, 203–214.
Pradyumna, K., Namadev, Yair Alroy and Vijay Singh (1998) Sniffing out trouble: Use of an electronic
nose in bioprocesses, Biotechnol. Prog, 14, 75–78.
Saja, R.de, Souto, J., Rodriguez-Mendez, M.L., de Saja, J.A. (1999) Array of lutetium bisphthalocya-
nine sensors for the detection of trimethylamine. Materials Science & Engineering C-Biomimetic
and Supramolecular Systemes 8–9: 565–568.
Schaller, E., Bosse, J.O., Escher, F. (1998) ‘Electronic noses’ and their application to food’, Lebensmittel-
Wissenschaft-und-Technologie, 31 (4) 305–316.
Schnürer, J., Olsson, J. and Börjesson, T. (1999) Fungal volatiles as indicators of food and feeds
spoilage. Fungal Genetics and Biology 27, 209–217.
Schweizer-Berberich, M., Vahinger, S., Göpel, W. (1994) Characterization of food freshness with
sensor array, Sensors and Actuators B, 18–19, 282–290.
Sheldon, R.M., Lindsay, R.C., Libbey, L.M., Morgan, M.E. (1971) Chemical nature of malty flavor
and aroma produced by streptococcus lactis. Applied Microbiol. 22, 263–266.
Storey, r.M., Davis, K., Owen, D., Moore, L. (1984). Rapid approximate estimation of volatiel
amines in fish, J. Food Technol. 19,1–10.
Tryggvadottir, S.V., Olafsdottir, G., 2000. Multisensor for fish: Questionnaire on quality attributes
and control methods -Texture and electronic nose to evaluate fish freshness. Project report for
European Commission (MUSTEC CT-98–4076). RF report 04–00, 36p.
Whitfield, F.B., Shaw, K.J., Tindale, C.R. (1988). 2,6-dibromophenol: The cause of an iodoform-like
off-flavour in some Australian crustacea. J. Sci. Food Agric., 46, 29–42.
Whitfield, F.B., Tindale, C.R. (1984). The role of microbial metabolites in food off-flavours. Food
Technol. Australia, 36, 204–213.
Zeng, K., Tachikawa, H., Zhu, Z.Y., Davidson, V.L. (2000), Amperometric detection of histamine with
a methylamine dehydrogenase polypyrrole-based sensor, Analytical Chemistry (10): 2211–2215.
Zhao, C.Z., Pan, Y.Z., Ma, L.Z., Tang, Z.N., Zhao, G.L., Wang, L.D. (2002), “Assay of fish fresh-
ness using trimethylamine vapor probe based on a sensitive membrane on piezoelectric quartz
crystal,” Sensors and Actuators B-CHEM 81 (2–3): 218–222.
6
Odors Prevention in the Food
Industry
Regina Nabais
1. INTRODUCTION
Taking into account the environmental impact, at the local and regional level, the
control and elimination of the emission of noxious odors in the food-processing
industry is one of that industry’s most difficult problems. The problem normally is
exacerbated not only because of the multiple ways in which it can be approached,
but also because, in spite of many efforts developed, there still is no universally
accepted way to address its solution.
The complexity of the issue of off-odors and the necessary multidisciplinary
expertise entailing a rigorous approach is the reason this chapter is neither exten-
sive nor intended to represent, even at an introductory level, an odor manage-
ment guide for the food industry. The scope is to strengthen the necessity to
reconcile aims, methods, collaborative attitudes, and efforts; to thoroughly out-
line strategies and tactics of scientific, technological, political, and legal inter-
ventions; and to explore the continued improvement and consistency of effective
global and integrated actions that will lessen the general environmental impact,
including odor emissions, of the food-processing industry. Whenever possible,
the main concepts, principles, and fundamentals of a few mentioned suggestions
to odor abatement or control will be mentioned and a list of bibliographic refer-
ences will be provided.
75
76 R. Nabais
protection and the improvement of the quality of the environment must be inte-
grated into the policies of the Union and ensured in accordance with the principle
of sustainable development.” This statement reflects the numerous changes in
European politics owing to environmental problems (Mike Holland et al., 2004;
SEPA and NIEHS, 2002 a, b).
Emissions of foul odors, regardless of their origin, are problems of
extreme complexity, assuming an ever-larger role among environmental issues.
This is especially the case when they have a direct, negative impact upon the
populations near an offending industry, whether they are a simple inconve-
nience or a harmful treat to neighboring communities. An odor is recognized
through sensory perception and subjectively characterized either as agreeable
or disagreeable, or even ambivalently perceived, depending upon the specific
circumstances of how an individual’s olfactory sense is stimulated. The olfac-
tory sense probably has not received the attention it deserves, considering that
it influences the senses of touch and taste, with deep repercussions on psycho-
logical behavior, stemming from its possible memorization and affective and
cultural interpretations.
When the sources of pollution emissions are specific industries, such as
food processing, pharmaceutical, or cosmetics, the supposedly marginal ques-
tions of production processes such as esthetics and odor become important
issues in the production process, because they may be reflected in the price of
the products in the marketplace.
Concerning odors, even those generally considered agreeable, under certain
conditions can become unbearable, depending upon the variability of emission
and reception conditions. This particularity is known by the acronym “FIDOL,”
which stands for the following: (1) frequency: how often an individual is exposed
to odour; (2) intensity: the strength of the odor; (3) duration: the length of a
particular odor event; (4) offensiveness/character: the character relates to the
“hedonic tone” of the odor (meaning “smells like”), which may be pleasant, neutral,
or unpleasant; and (5) location: the type of land, use, and nature of human activities
in the neighborhood of an odor source.
Even if the majority of malodors produced by the food processing industry are
inoffensive, due to low concentrations or reduced toxicity, the industries that
emit them must be prepared to encounter different levels of protest against their
activities:
1. People believe that the odor has a negative effect on their well-being and
health conditions.
2. Environmental and activist political groups often protest against any
activity that may lead to the emission of bad odors.
Odors Prevention in the Food Industry 77
3. The plaintiffs and their lawyers or legal representatives will usually drag
the offending industries through interminable legal battles, extremely
costly in terms of time, money, and resources.
4. The normative legislation at the local, regional, national, and suprana-
tional levels are progressively demanding, regulatory, and restrictive.
5. Other businesses and entities, which complain of the negative or poten-
tially negative impacts on their costumers and their future investments
in the area.
6. Some clients may boycott food products produced by factories that emit
bad odors.
7. The installation of some potentially economically viable activities (e.g.,
commercial ventures, tourism, sporting venues, etc.) would be compro-
mised by locations near odor-emitting industries.
Enterprise Costs
Quality of the Depreciation of the
prevention or Maintenance and enterprise image and Social Costsa
control systems Initial investment operational costs final property value (externalities)
One method to enlist the confidence and trust of the local neighborhood
residents is to offer them a forum in which to explore their apprehensions and
doubts, to offer guided tours of the facility and its production system, and to
include them in both exploring the environmental problems and the quest for
possible solutions. “Science is losing credibility. Conflicts of interest, biased stud-
ies, and secrecy are undermining science’s reputation and its truth-seeking objective
affecting public trust” (Collins, 2000).
Hence, trust and confidence are the main keys to addressing public perception
and understanding. This is not in itself an expensive methodology in monetary
terms, but it demands accomplishing the most difficult tasks human beings can
achieve: cultural changes. Trust enforces the exchange of information and ideas
and increases the likelihood of the community feeling more in control of the solu-
tions for their environmental concerns, instead of promoting overreactions as a
result of exaggerating their fears by useless secrecy or hiding the problems.
Apart from considering process isolation, capable confinement, and good con-
veyor systems, immediately followed by efficient onsite treatments (and despite
the existence of several exceptions) direct interventions in food process and/or
food process control have poor effect on annoying properties of odors. Food
process residues, their manipulation, transfer operations, treatment and final dis-
posal systems or facilities are also sources of malodors emissions.
A food process residue is usually understood to be an incidental organic
material generated by processing agricultural commodities for human or ani-
mal consumption. The term includes food residuals, food coproducts, food-
processing wastes, food-processing sludge, or any other incidental material
whose characteristics are derived from processing agricultural and fishery prod-
ucts. Examples include process wastewater from cleaning slaughter areas, rinsing
carcasses, or conveying food materials; process wastewater treatment sludge; bone;
fruit and vegetable peels; seeds; shells; pits; cheese whey; off-specification food
products; hide; and hair and feathers.
As any other air pollutant, odor sources may be classified as point source or
disperse source; unfortunately, the food industry may present both odor origins.
A rough organization of improving food industry odors sources may be seen in
Table 2.
Thus, odor can be considered as a “primary air pollutant” emitted by some
food industrial processes: raw materials handling, transport, storage, cooking,
concentration, drying, grilling, fermenting, and roasting processes are examples
of sources of odors emissions. If these emissions are potentially annoying to
their neighbors, their sources must be rethought, controlled, and treated sepa-
rately from the fugitive odors of raw materials.
Odors Prevention in the Food Industry 79
Operations and
processess Utilities operation and maintenance Waste handling and disposal
On the other side, in general, liquid wastes produced by food industry have
high contents of organic pollutants including biochemical oxygen demand, ammo-
nia, suspended solids, oils and greases and are commonly acid, biodegradable, and
malodorous (Environmental Protection Agency, 2000). When an unstable organic
waste in liquid, slurry, or sludge starts decomposing, gases are liberated and accu-
mulate in fluid solution phase, until the saturation of the liquid is reached; beyond
this point, smelly vapors are released.
Most of food process treatment systems (ponds, leaching beds, septic tanks,
and trickling filters) are poorly built-up, inefficient, and improperly operated.
Fortunately, there are several highly efficient technologies available for indus-
trial food manufacturing liquid effluents, but the costs are generally unbearable,
considering the narrow marginal profits of these activities. The problem for
80 R. Nabais
Odor emissions in the indoor environment of food industries, affects the indoor
air quality (IAQ). When emissions have negative impacts it is common to observe
a decrease in worker productivity and increases in absenteeism; in extreme con-
ditions, malodorous substances may induce serious illness. Beside these already
mentioned, serious effects, another caution is crucial in IAQ of a food industry
and that is “the taint.” The term taint relates to any foreign smell or taste in
goods, e.g., musky, yeasty, catty, furry, disinfectants, cork, cardboard, hearthy
(geosmin), musty (methylisoborneol), as a result of adsorption or absorption of
volatile compounds, by the commodities materials, throughout the entire pro-
duction line (from raw materials to the final packed products). The causes for
taint occurrences in food products are very diverse, but one of the sources may
be poor indoor air quality.
Anyhow, occurrences like tainting can result in considerable inconveniences
to manufacturers, like lost of production, sales, consumer confidence, damaged
brand names, loss of commercial relationships between suppliers, manufactur-
ers, and retailers, insurance disputes, and in extremely expensive and long legal
battles. The prevention of taint development is vital to any food process, not only
to dignify and preserve product image and stay in business, but also because in
some countries and circumstances (e.g., in United States), if the taint is a result of
intentional or careless manufacturing and cumulatively implies external commerce
depreciation, those legally responsible may be subject to heavy fines and may be
sentenced to up to 3 years in prison (US Food and Drug Administration, 2004).
The three strategies, in order of effectiveness, for reducing pollutants in the
indoor air environment are source control, ventilation, and air cleaning. (1) Source
control eliminates individual sources of pollutants or reduces their emissions, and
is generally the most effective strategy. (2) Ventilation brings outside air indoors.
It can be achieved by opening windows and doors, by turning on local exhaust
fans, or, in some situations, by the use of mechanical ventilation systems, with or
without heat recovery ventilators (air-to-air heat exchangers). However, there are
practical limits to the extent ventilation can be used to reduce airborne pollut-
ants. Costs for heating or cooling incoming air can be significant, and outdoor air
82 R. Nabais
Malodorous emissions are never isolated problems of the food industry envi-
ronment. Indeed, odorant emissions only reflect environmental mismanagement
faults and procedures.
84 R. Nabais
6. Continuous
improvement
5. Managing
review
1. Environmental
principles
2. Planning
3. Implementation
and operation
4. Checking and
corrective actions
indirect steam dryers must be controlled to avoid scorching of the fish (overdry-
ing, losing weight) and amplifying odor emission problems; (2) increasing the
pressure of the presses will improve the recovery of malodorous constitution
solutions; and (3) the size of the holes in the pre-strainer must be increased,
because if the fishery processes poor quality raw materials, these tend to produce
a slurry that clogs up the outlets of the presses, diminishing its efficiency.
When handling material that is date sensitive or perishable, which is the case in
the food industry, several requirements must be foreseen:
The storage spaces must be clean and carefully organized. All points of
the storage must be easily accessible and offer safe conditions for fre-
quent inspections and surveillance procedures; refrigeration units should
be cleaned and maintained on a preset periodic basis.
For refrigerated and frozen products, temperature and moisture is crit-
ical. Given the effects of temperature on the safety conditions of bio-
logical goods and raw materials, alarm devices are necessary to monitor
temperatures.
86 R. Nabais
Figure 2. Residues improperly maintained in open bags, waiting for reutilization or destruction.
Permission by courtesy of Portary Gestão de Resíduos Lda (2004).
two processes. It is during the fermentation process that the Volatile organic
compounds (VOC) are produced. The sponge dough process, which is most
commonly used by commercial bakeries, produces the largest amount of VOC
emissions because the required fermentation time can be 5 hours or more. By
cutting spongeous fermentation period, from 4 hours to just 1, through use of
a preferment concentrate, the bakeries can lower ethanol emissions by half and
more.
In a British bakery, two depanners use a vacuum system to remove bread
from the baking trays and tins. The vacuum is generated in vacuum chambers
by fans driven by standard 15 kW AC motors. With these improved conditions,
having a payback period of 1.3 years, technical and economical advantages were
registered: (1) energy savings: about 135,000 kWh/year; (2) reduction in CO2
emissions: 67,500 kg/year; (3) damage of bread rolls were minimized; and (4)
lower overall maintenance cost was obtained.
Also, to control VOC emissions from commercial and industrial boilers, no
auxiliary equipment is needed; properly maintaining the burner/boiler package
will keep VOC emissions at a minimum. Maintenance services must include keep-
ing the air/fuel ratio at the manufacturer’s specified setting, having the proper air
and fuel pressures at the burner, and maintaining the atomizing air pressure on
oil burners at the correct levels. An improperly maintained boiler/burner pack-
age can result in VOC levels over 100 times the normal levels of specific pro-
cesses, and the energetic efficient decreases.
Another good example in the food industry is deep frying processes. It is
important to choose oil with a high smoke point (the temperature at which it
starts to smoke). When oil smokes, it begins to decompose and, some experts say
many of its unsaturated fatty acid molecules become saturated. Frying oils are
exposed to atmospheric oxygen and moisture at elevated temperatures during
88 R. Nabais
frying. The oil gets deteriorated, as the triglycerides in the oil undergo oxidative,
hydrolytic, and thermal alterations producing more than 400 various chemical
compounds.
Breakdown of the glycerol molecules in the fat creates acrolein, an obnox-
ious-smelling compound that easily inflames. This means that perfectly safe fat,
with a flash point of 260 °C, might burst into flames at a dangerously low tem-
perature after having been used, just once in high temperature.
Avoid adding salt to food before frying, because salt draws moisture to the
food’s surface, which will splatter when the food is added to the hot oil. Salt also
lowers the smoke point and breaks down the oil.
The most common odor management control for wastewater systems is a float-
ing cover made from polyethylene (PE) membrane suspended on cables and/or
supported by floatation devices, with perforated gas collection pipes, weighted
sunken troughs to collect water, sump pumps, and so on. The main function of
such cover is to prevent vector migration of vapors or gases into the atmosphere
and to collect methane gas to be disposed by flaring or other means to reduce
odor pollution. No matter how elaborate installation procedures and the meth-
odologies, efficiency, the puncture of PE membrane, wind, hail, snow, ice vulner-
ability, explosive hazards due to static electricity and accumulation of volatile
gases, vandalism, dependency on electric power supply and reliance on profes-
sional maintenance of electric motors and pumps contribute to a substantial
initial investment and a high operating cost.
However, the best technology in the world or the most costly odor control
system can only be effective in reducing odorous emissions if the odors reach the
control devices. If the containment and ventilation systems are not able to keep
the nuisance smells from escaping, then the money spent on the odor control has
been wasted.
Transfer to air
Release to atmosphere
Dispersion
Annoyance/Nuisance
COMPLAIN ACTION
Raw materials
Extraction and
beneficiation Manufacturing Consumers
process
Residues/Emissions
Environment impact
Environment (Externalities-
SocialCosts)
Some food processing wastes such as tomato and apple pressings, corn cobs
and husks, sweet corn cannery waste, stale bakery products, and spent brewery
mash may be used as feed complements to cattle, helping farmer to reduce
Odors Prevention in the Food Industry 93
Reuse
13%
Housekeeping
24%
Recycle
13%
Material
Substitution
2%
Product
Modification Process
2% Modification
24%
Figure 5. Relative frequency of cleaner technologies principles used for pollution abatement in food
industry. From International Cleaner Production Information Clearinghouse (2004).
overall feeding costs. Also, feeding these recyclable wastes to animals reduces
the volume of solid waste that food processors must dispose of at an approved
sanitary landfill or apply as a soil amendment.
In conclusion, any material that does not become a part of the final products
must be promptly recycled into or off the process, reuse to some other by-product,
or ultimately adequately disposed. The costs of odors treatment can be minimized
by focusing on the previous options and cleaner production.
7.6.1a. Reduce. The next step is to minimize the use of all materials in the
process. This can include reducing errors in batch preparation, optimizing clean-
ing operations to reduce the volume of water used, and turning off equipment
that is not in use.
7.6.1b. Reuse. There are many opportunities to reuse waste products in the
food manufacturing industry. This will reduce the demand for raw materials and
the cost of treatment and disposal. It may be possible to reuse some clean pro-
cess waters in the boiler or install drip trays to catch product that eventually
may return to the process in next batches. To enable biobased materials produc-
tion in the food industry, it is only necessary to obey preestablished rules of
good waste management. An interesting example in this domain is Cargill Dow’s
polylactide (PLA), which is a versatile biopolymer, made from 100% renewable
94 R. Nabais
resources like crop residue (stems, straw, husks, and leaves) from corn or other
crops (Vinka et al., 2003).
7.6.1c. Recycle. Are the wastes identified by your assessment really wastes?
Can some of these be reclaimed through a simple treatment process, which
enables them to be reused on-site. Other by-products that cannot be used on-site
may be recycled off-site. In these cases there may be the potential to sell recy-
clable items and also save by the avoiding disposal costs.
7.6.1d. Treat and Dispose. This option should only be considered after the
other options have been exhausted. Generally these options are typically a cost
to industry. However, it may be essential to consider this as a part of your overall
cleaner production strategy. The costs of treatment can be minimized by focus-
ing on described options, which may be specifically applied to different food
industries and services (UNEP, 2004a,b).
According to the Ontario Ministry of Agriculture (1995), many of those food
wastes are wet and/or decompose readily. Leachates and polluted or contaminated
runoffs may easily develop problems with odors. To avoid such problems, farmers
should tailor handling and storage arrangements to their situation and the waste
or by-product they use. Some general guidelines to minimize odors and water
quality problems include:
Cleaner production aims to prevent pollution, reduce the use of energy, water,
and material resources, and minimize waste, profitably and without reducing
production capacity (Environmental Protection Agency, 2000; ICPIC, 2004;
Pennsylvania State University, 2002; Robert et al., 2002). Strict inventory con-
trol, e.g., minimizing expired shelf-life materials in storage dependences and
avoiding long periods of permanence of the products in the expedition spaces,
is good practice.
Pollution prevention processes are one of the major procedures of cleaner
technologies. Cleaner technology main tasks are:
1. Energy conservation.
2. Water conservation.
3. Maintenance of the equipment in good repair.
4. The cleaning and sanitizing procedure prevents the build up of dirt
and debris. Cleaning and sanitizing must address three basic set points:
exterior facility access, pavements, grounds, and connections to sewers;
internal facility and utilities, including floors, walls, ceilings, and ventila-
tion system; and drains and equipment.
As illustration, one of the important issues for many food industries is oil and
grease residue handling, their removal or treatment systems, and final discharge
or disposal devices; in fact, grease tends to solidify downstream, causing frequent
blockages and/or backups. They also may cling to wastewater ducts and reduce
their flow capacity in the long term. All degreaser equipment only separates oil
and grease onto the surface of the bulk effluent and at best relocates it.
By accidental overflow onto pavements, in many cases it remains as an oily
film that poses a risk for slips and falls. Then, there is the problem with cleaning
and disposal. On the other hand, oily waste cannot be dumped down the drain,
because haul-off can be a very difficult and expensive consequence.
96 R. Nabais
Inspection and maintenance programs embody the idea that some organizing
procedure mechanism must be put into place to ensure that all installations and
equipment must be kept in nonpolluting condition. Measuring resource effi-
ciency is the key to making incremental improvements in performance. Examples
of environmental performance indicators, for companies, are the amount and
the type of energy used, carbon dioxide emissions, water consumption, effluent
discharges, waste management, and environmental incidents registers.
Avoid leaks from gaskets and seals. Spill, splash discharges, and leak preven-
tion is one good tool to prevent odors emissions, while usually also meaning eco-
nomical gains (e.g., these interventions must take part in good practice checklist
procedures of any preventative detailed maintenance plan).
Proper land use planning avoids allowing incompatible land uses in close prox-
imity and is one of the most important tools in odor management. Many odor
problems can be avoided by appropriate placement of new facilities.
The lack of consistent methods to deal with odors prevents authorities from
establishing policies and standards, which would eliminate, or at least minimize,
community odor nuisances [Danish Environmental Protection Agency (DEPA),
2002; New South Wales Department of Environment an Conservation (NSW),
Odors Prevention in the Food Industry 97
2004; UNEP Working Group for Cleaner Production in the Food Industry
(UNEPTIE), 2004a,b; Robert et al., 2002].
According to ZERI (Zero Emissions Research Initiative, of United Nations
University, 1994), in the future industries will be organized into clusters in a way
that waste from one industry becomes input for other industries, forming an inte-
grated system. This simple, genial concept imitates natural ecosystems, where no
real wastes exist. Industrial ecology shares with ZERI by anticipating the environ-
mental consequences of production, consumption, and waste management, inte-
grating overall balanced activities for ecoindustrial parks and product life-cycles,
“taking a systems perspective on the environmental consequences of production
and consumption” (Lifset, 1999; Robèrt et al., 2002).
It also is urgent to ensure compatibility of industrial activities with adjacent
land uses, adequate separation distances to allow for process upsets, and adopt
the principles of waste minimization, cleaner production, and best practice con-
trol technology. These objectives-oriented interventions will minimize or elimi-
nate the release of odors from the site, strictly oriented by the specificity of the
industrial activity (UNEPTIE, 2004a, b).
Avoiding unregulated development zoning serves principally to protect
property owners from the negative externalities (social costs) of new develop-
ments. The zoning plans are based on the following considerations:
Designated land use of the site and the surrounding land under the dis-
trict plan.
Confine location of activities within the site and their orientation in rela-
tion to prevailing winds.
Consider sensitivity uses of the downwind receptors.
Respect buffer distances from the site boundaries to sensitive land uses.
Use screening, such as by earth bunds, shelter belts, or natural topogra-
phy, that functions as obstacles for odors transference.
New facilities should be carefully sited so that prevailing winds do not carry
odors to sensitive neighboring sites. However, no setback distance will work for
every site. Setback requirements are site-specific and should be calculated using
complete information about local conditions. For example, setback calculations
should include the size and type of operation, the facilities covered by the setback
(such as lagoons, fields, and houses), as well as the proposed sites and methods
for applying wastes to land. Local vegetation, prevailing winds, weather pat-
terns, and neighboring land uses also should be considered in the calculations.
An efficient planning option is confining odoros activities to certain districts
(by effective zoning regulations). Examples of sensitivity activities are residen-
tial dwellings, visitor accommodations (hotels, bed and breakfast, guesthouses),
hospitals or nursing homes, schools, churches, holiday and weekend dwellings,
shelters, campsites, caravan parks, and sports facilities.
Figure 6 gives an idea of how an emission source can have a favorable lay-
out previously roughly settled, by considering its location related to the relative
frequency of dominant wind-flow directions.
The setback distance, or buffer zones, of implantation of the facilities must
considerer not only the frequency and intensity and dominant wind, but also the
exposure angle of the facilities (Heber, 1997; Schauberger and Piringer, 1997;
Pennsylvania State University, 2002). Several European countries have established
such setback models, such as the known Austrian guideline that is based on
quantitative odor evaluations and neighbor surveys and is a tool to assess air-
borne emissions resulting from livestock husbandry and the pollution caused by
these emissions.
Odors Prevention in the Food Industry 99
N
100
80
NW NE
60
40
20
W 0 E
SW SE
Figure 6. A hypothetical way to draw a wind pattern relative to frequency and regional vicinities
borders (….), of a well-located emission site related to wind-orientation dominance, spotted in the
center of the winds-rose.
Using geographic and meteorological data, maps can be drawn up to identify lev-
els of exposure and predict odor annoyance impacts on the residents near those
areas.
For each of the currently available management options described in this
chapter, performance criteria should be developed and data collected so that
management improvements can be recommended.
In summary, odor management strategies are shifting their approaches,
from old systems—complaint driven—to the new systems based on anticipation
measures and precautionary inspection systems.
Odors are often low-density gases. Once released into the environment they are
transported by wind and diluted and dispersed by atmospheric turbulence.
According to theoretic common concepts, odorous gases require special consid-
eration and treatment. Warren Spring Laboratory in the United Kingdom (Williams
and Thomson, 1985) suggests the following technique to give rough estimations of
the uncorrected chimney height hu (meters), by one of two alternatives:
hu = (0.1DQ)0.5
where
massive dilution, and simultaneously “virtually increases a fluid stack height” (to
discharge odorous gas stream as far as possible, from the ground level).
Enhanced spatial dilution of odorants, as the technique just described, must
not be considered to be a “miraculous” solution to any air pollution problem,
especially if it is considered to be a nuisance. If pollutant concentrations meet
all emissions regulations and the legal compliance requirements are met, but
the emissions remain a malodorous nuisance to the surrounding neighborhood,
further dilution of the pollutants should be sought.
8. CONCLUSIONS
9. AKNOWLEDGMENTS
The author deeply expresses gratitude to: Doctor Xavier Nicolay, of Institut
Meurice–UBT, for his strong partnership conscience, team responsibility and
consistent coordination, in putting together this “many-hands–multicultures
knitted book”; but, specially, for his kind understanding, restless support, car-
ing spirit, endless patience, attentive readings, helpful and valuable advices about
this chapter content. However, copyrights remain the author’s responsibility for
all the still remaining imprecision or errors.
10. REFERENCES
Anonymous (2002) Cleaner production tools: LCA and beyond. Editorial. Journal of Cleaner
Production 10 403–406.
ASHRAE (2003) HVAC Applications Handbook. (IP). Item 210–8103-03. Publisher: ASHRAE.
ISBN: 1–931862–22–2.
ASHRAE (1999) Ventilation for Acceptable Indoor Air Quality, Standard 62–1999, Atlanta, GA,
American Society of Heating, Refrigeration, and Air Conditioning Engineers.
Collins, R. (2000) CSPI PROJECT “Science Conflicted: Restoring Trust in the National Academy of Sci-
ences”. Science Conflicted: Restoring Trust in The National Academy of Sciences”, Integrity in Science
Project, Center for Science in the Public Interest. Baltimore sun Journal. Editorial of August, 2000.
Danish Environment Protection Agency (DEPA) (2002) “Industrial odour control.” Environmental
Guidelines No. 9 2002. Vejledning fra Miljøstyrelsen Environmental Guidelines no. 9, 2002.
http://www.mst.dk/udgiv/publications/2002/87-7972-297-0/html/helepubl_eng.htm
Odors Prevention in the Food Industry 103
Environment Agency for England and Wales in collaboration with the Scottish Environment Protec-
tion Agency (SEPA) and the Northern Ireland Environment and Heritage Service (NIEHS) (2002a).
Technical Guidance. Note IPPC H4. “Integrated Pollution Prevention and Control – IPPC DRAFT.
Horizontal Guidance for Odour.” Part 1 – Assessment and Control. www.environment-agency.gov.
uk/commondata/105385/ h4_odour_part1.pdf.
Environment Agency for England and Wales in collaboration with the Scottish Environment Protec-
tion Agency (SEPA) and the Northern Ireland Environment and Heritage Service (NIEHS) (2002b)
“Technical Guidance. Note IPPC H4. Integrated Pollution Prevention and Control (IPPC) DRAFT.
Horizontal Guidance for Odour Part 2 – Assessment and Control.” www.environment-agency.gov.
uk/ commondata/ 105385/ h4_odour_part2.pdf.
Environment Protection Authority, Victoria, Australia (2004) “Strategies For Promoting Cleaner
Production In The Food Industry.” Rowan Mackenzie and Scott Hamilton. http://www.uneptie.
org/pc/cp/library/training/howtoCPC/manual_cdrom/CPlinks/pdfs/Hamilton.pdf.
Environmental Protection Agency (2000) “Biosolids and Residuals, Management Fact Sheet, Odor
Control in Biosolids Management,” EPA 832-F-00–067, United States, 2000. http://www.epa.
gov/owmitnet/mtb/odor_control-biosolids.pdf.
Ericsons Naturnomics Company (2004) “Odor Answers. Stack Extenders.” http://www.eric-sons.
com/ne4/odoranswers/stackextenders/index.htm.
Erwin, T.H., Vinka, Rabago, K.R., Glassner, D.A., Gruber, P.A. (2003) “Applications of life cycle
assessment to NatureWorks polylactide (PLA) production. Polymer Degradation and Stability.”
80: 403–419. www.cargilldow.com/corporate/life_cycle/docs/PLA_Article.pdf. 2003.
European Commission Directorate-General Jrc. Joint Research Centre. Institute for Prospec-
tive Technological Studies Technologies for Sustainable Development. European IPPC
Bureau. (2003) “Integrated Pollution Prevention and Control, BREF (Best available tech-
niques REFerence document) on the Slaughterhouses and Animal by-products according to
Article 16(2) of Council Directive 96/61/ EC. 2003.” http://www.umweltdaten.de/nfp-bat-e/
slaughterhousesbref-e.pdf.
Gesamtverband der Deutschen Versicherungswirtschaft e.V. (GDV, German Insurance Association)
(2003) Wild, Y. “Refrigerated containers,” in Chapter 7. Container Handbook. 2003. http://www.
containerhandbuch.de/
Heber, A.J. (1997) “Setbacks for sufficient swine odor dispersion and dilution.” Paper presented at
the Livestock and Environment Symposium, University of Nebraska Cooperative Extension
Service, New World Inn, Columbus, Nebraska, December 10–11, 1997. Modified Dec. 16, 1997.
http://pasture.ecn.purdue.edu/~heber/setba.htm.
Holland, M., Hunt, A., Hurley, F., Watkiss, P. (2004) Draft final methodology paper for “Service
contract for carrying out cost-benefit analysis of air quality related issues, in particular in the
clean air for Europe (CAFE) programme.” 5/ 07/ 04 AEAT/ ED51014/ Methodology Paper
Issue 3. http://europa.eu.int/comm/environment/air/pdf/cba_methodology_issue3.pdf.
Institute for Prospective Technological Studies Technologies for Sustainable Development.
European IPPC Bureau. European Commission Directorate-General Jrc. Joint Research
Centre. 2003. “Integrated Pollution Prevention and Control, BREF (Best available techniques
REFerence document) on the Slaughterhouses and Animal by-products according to Artical
16(2) of Council Directive 96/61/EC.”
International Cleaner Production Information Clearinghouse (ICPIC) (2004) Environmental Man-
agement Center. Technical Case Studies. Manufacturing. “Manufacture of Food Products and
Beverages.” http://www.emcentre.com/unepweb/tec_case/ .
Lifset, R. (1999) United Nations University. Zero Emissions Forum. The Industrial Ecology of
Renewable Resources. ZEF-EN-1999–2. http://www.unu.edu/zef/publications-e/ZEF-EN-1999-
02-D.pdf
Mohunta, D. M. (2004) “Recovery of volatile organic compounds from gaseous streams.” http://
www.ccdcindia.com/index2.php?act=voc&subact=vocpaper. Commercial, Chemical and Devel-
opment Company.
New South Wales Department of Environment and Conservation (NSW) (2004) “Odour control.”
http://www.epa.nsw.gov.au/mao/odourcontrol.htm
104 R. Nabais
Ontario Ministry of Agriculture, Food and Rural Affairs and Ontario Ministry of Environment and
Energy (1995) “The Guide to Agriculture Land Use and Minimum Distance Separation.” http://
www.gov.on.ca/OMAFRA/english/landuse/facts/guidealu.pdf.
Pennsylvania State University in cooperation with the Pennsylvania Department of Agriculture
(2002) Odor Management in Agriculture and Food Processing. Odor Management in Pennsyl-
vania. A Reference Manual. Revised printing June, 2004. http://server.age.psu.edu/fac/Elliott/
EntireOdorManual.pdf.
Pinchin Environmental Newsletter (2004) Ventilation & Emission Control Group. http://www.
pinchin.com/newsletters/air/odourcomplaints.htm.
Portary (2004) Recolha e Transporte de Resíduos Industriais Banais Perigosos. http://www.portaryresiduos.
pa-net.pt/html/gest.htm.
Realff, M.J., Abbas, C. (2003) “Industrial symbiosis. refining the biorefinery.” Journal of Industrial
Ecology. Vol. 7, Issues 3–4.
Robèrt, K.H. (2002) Schmidt-Bleek, B., Aloisi de Larderel, J., Basile, G., Jansen JL, Kuehr R, Price
Thomas P, Suzuki M, Hawken P, Wackernagel M. (2002) “Strategic sustainable development —
selection, design and synergies of applied tools.” J Cleaner Prod.; 10(3): 197–214.
Schauberger, G., Piringer, M. (1997) Guideline to assess the protection distance to avoid annoyance by
odour sensation caused by livestock husbandry. Proceedings of the Fifth International Livestock
Environment Symposium, May 29–31. 1997.
Society of Heating, Refrigerating and Air-Conditioning Engineers (1997) “Handbook of Funda-
mentals.” http://www.containerhandbuch.de/chb_e/ wild/index.html.
Starkey, R. (1998) Centre for Corporate European Environment Agency. (CCEM). Internet Gateway
for Cleaner Production, Pollution Prevention and Sustainable Business. “Environmental man-
agement, Tools for SMEs: A handbook.” Environmental Issues Series. Edited for the EEA.1998.
http://reports.eea.eu.int/GH-14-98-065-EN-C/en/enviissu10.pdf.
Sustainable Agri-Food Production and Consumption Forum (2004) UNEP Publications. United
Nations Environment Programme Division of Technology, Industry and Economics. “Industry
Sector Guide Lines.” Ref. UNEPTIEb, 2004.
Tybor, P.T., Hurst, W.C., Reyknolds, A.E., Schuyler, G.A. (1995) Quality Control. A Model Program
for the Food Industry. University of Georgia College of Agricultural & Environmental Sciences.
Cooperative Extension Service, Extension Food Science. http://pubs.caes.uga.edu/caespubs/
pubcd/b997-w.html.
US Food and Drug Administration (2004) Federal Anti-Tampering Act U.S.C. Title 18 – “Crimes
And Criminal Procedure. Part I – Crimes Chapter 65 - Malicious Mischief. § 1365. Tampering
with consumer products.” http://www.fda.gov/opacom/laws/fedatact.htm.
US Environmental Protection Agency (1995) Technology Transfer Network. Clearinghouse for
Inventories & Emission Factors Compilation of Air Pollutant Emission Factors, “AP-42, Fifth
Edition, Volume I: Stationary Point and Area Sources.” Research Triangle Park, NC. 1995.
UNEP Working Group for Cleaner Production in the Food Industry (2004a) “A Guide to Cleaner
Production in the Food Industry.” http://www.geosp.uq.edu.au/emc/cp/Res/guides/GUIDE1.
HTM to http://www.geosp.uq.edu.au/emc/cp/Res/guides/GUIDE7.HTM.
UNEP Working Group for Cleaner Production in the Food Industry (2004b) “Fact sheets.” http://
www.geosp.uq.edu.au/emc/CP/Res/xfactsheets.htm. The University of Queensland and The
United Nations Environment Programme.
United Kingdom Renderers Association (UKRA) (2004) http://www.ukra.co.uk/index.htm.
United Nations Environment Programme — The UNEP Working Group for Cleaner Production in
the Food Industry (2004) “A Guide to Cleaner Production in the Food Industry.” http://www.
geosp.uq.edu.au/emc/cp/res/yfood_manual.htm. (access on Wednesday, August 25, 2004).
Van Harreveld, A.P. (2001) “From odorant formation to odour nuisance: new definitions for discuss-
ing a complex process.” Water Science & Technology. Vol. 44. No 9. pp 9–15 © IWA.
Williams, M.L., Thompson, N. (1985) “The effects of weather on odour dispersion from livestock
buildings and from fields.” In: Odor Prevention and Control or Organic Sludge and Livestock
Farming. V.C. Nielsen, J.H. Voorburg, P. L’Hermite. Elsevier Applied Science Publishers, New York.
pp. 227–233.
7
Odors Treatment:
Physicochemical Technologies
Regina Nabais
7.1. INTRODUCTION
105
106 R. Nabais
Despite the complexity of the problems of handling off-odors, there are only two
ways of management of their emissions: (1) source control, by selecting appropri-
ate industrial processes able to reduce emissions in the source, and (2) treatment
of the final resultant emissions.
When applied to industrial processing, the goal of cleaner production, is
to avoid pollution generation in the first place,which frequently reduces risks,
cuts costs, and identifies new business opportunities. Cleaner production aims
to reduce waste and inefficiency at the source and also can help to develop the
most efficient ways to operate processes, produce better products, and provide
new services.
Adequate control equipment and effective treatments may be based on cap-
ture recovery (nondestructive technologies) or on destruction. Some treatment
technologies use combinations of both capture recovery and destruction. Crite-
ria for the selection of the best choice for odor control and treatment must be
carefully chosen to assure their effectiveness at sustainable costs.
Even so, whatever may be the selected treatment, it will require high initial
cost to install, efforts to implement, and careful planning to start up and main-
tain, and together with the important capital investments will require careful
criteria that considers heat conservation/recovery. Large airflow rates with low
volatile pollutant concentrations frequently are encountered in food industry
processes, considering VOC and odors emissions.
Regardless of the selected option to deal with off-odors, rigorous, careful,
and strict maintenance programs will need to be designed and implemented,
especially if nonthermal emission control devices are chosen. Usually, all tech-
niques need replacement of costly consumables (i.e., adsorbents, reactants, fuel
etc.) and require energetic provisions to regenerate subsidiary materials (e.g.,
adsorbents) and storage of spare parts.
Consideration also must be given to the legal requirement of using either
the “best available control technology” (the best possible choice of competing
processes for reduction of a specific pollutant given economic, chemical, and
physical constraints), or the “maximum available control technology” (the process
that produces the greatest emissions reduction, regardless of all constraints), or
“reasonable available technology” ( a balanced process that considers equal con-
cerns to abatement technology that is more reasonable, economical, and proven
in practice), or even the “best available technology not entailing excessive costs”
(best at preventing pollution and available in the sense that it is procurable by
the operator of the activity concerned, including a balance between environ-
mental benefit and financial costs) [Integrated Pollution Prevention and Control
(IPPC), 2003].
However, before one starts to identify and choose any suitable technology for
air pollution control, including odor abatement, there are variables and procedures
Odors Treatment: Physicochemical Technologies 107
Table 1. Examples of Odor Control Technological Options for Food Process Industries
Rendering industry
Slaughterhouses
Fermentation
incineration
Fish meal
Animal
Coffee
Boiler
Wet scrubbing. Absorption: moderate to high costs,
three stages often required. Needs careful selection of
Ammonia
Amines,
aerosols
scrubber liquor and usually trials. Not always success-
Amines
Amines
ful, requires regular maintenance and daily tests of
active agent and pH control in some cases.
Thermal oxidizer (direct). Temperatures between
600°C and 1000°C with residence time of 0.3 to 1
second. Doubling of residence time may enable after-
burner temperature to be reduced 20°C to 100°C.
Requires careful design to reduce air volume to a
minimum. Capital cost high and cost of running is
high for large air volumes. Needs further control if
Aldehydes, amines
Aldehydes, amines
sulfur or chlorine present in exhaust gases.
Amines, aerosols
Thermal Oxidizer (catalytic). Temperature 400°C.
Lower temperature operation than direct method,
Amines
Alcohol
intervals. Can be effective but expensive for large air
Amines
Amines
Amines
amines
sive.
gas to be treated.
Destructive methods are those, when properly operated, that are not producers
of secondary organic waste—they destroy the pollutants or contaminants by
oxidation of pollutants or malodorous compounds.
Table 7.2 shows the technologies commonly used to control VOC emissions,
which may be easily adapted to food industry odor emissions reduction and control.
7.3.1. Absorption
Wet scrubbing using packed towers has proved to be the most reliable means
of removing odorous pollutants and odor reduction, handling, for example, a
wide variety of sulfides in inlet concentration. Alkaline (NaOH) scrubbing can
achieve highly efficient removal of hydrogen sulfide (H2S) but the solution pH
must be maintained at a high level to prevent release of the H2S. In most cases
an oxidant such as NaOCl is added to prevent this from occurring. Highly vol-
atile sulfides such as dimethyl sulfide (DMS) and dimethyl disulfide (DMDS)
Odors Treatment: Physicochemical Technologies 111
Mist eliminator
Outlet gas
Scrubbing
liquid
distribution
Packing material
Make up
water or
solution
Inlet gas
7.3.2. Adsorption
Clean
gas
Polluted
outlet
gas
Blower
Adsorbent
material
Figure 2. Sketch of an activated carbon adsorption system. Adapted from Croll-Reynolds Clean Air
Technologies (1996).
Odors Treatment: Physicochemical Technologies 115
7.3.3. Condensation
Condensate
Condensation Ability to cost-effectively The coolant fluid depends on the vapor pressures
recover and reuse con- of the organic contaminants to be isolated;
densed components from Thus, the identity of the VOC pollutants dictates
this process increases the the operating conditions for the control process;
technique’s value; this may be difficult to achieve in gaseous streams
Condensation method composed of complex odors mixtures;
will work with any The relatively low concentration of solvents in air
volatile compound, in requires very low temperatures resulting in high
any concentration, from energy costs (electrical power), and hence, poor
streams with different economics;
flow rates. Recovery by cryogenic condensation will condense
and freeze water vapor from the airstream requir-
ing complex exchanger thawing techniques.
7.3.4. Oxidation
Controls of the
systems and
Performance factors Malfunction conditions usual checklists
Gas
Heat
outlet
exchanger
Gas
inlet
Burner
Oxidation Chamber
which will allow air to pass while serving as a mass to store heat. The media
material selection, size, and shape can vary greatly and substantially impact the
design and utility efficiency of the RTO.
In comparison to both thermal and catalytic oxidation, RTOs have the advan-
tage of the VOC application flexibility and destruction of a thermal oxidizer with
better fuel efficiency than a catalytic oxidizer without the risk of poisoning or fouling
expensive catalyst.Regenerative thermal oxidation systems are the most expensive
thermal oxidizers to build, but savings in auxiliary fuel pays the added initial costs.
It is important to reduce the moisture content of any gas stream requir-
ing incineration (above 400ºC) in order to reduce fuel consumption, because
Supplement fuel
flame
Combustion Chamber
Stack
Bed packing -
ceramic heat
accumulator
media
Blower to
return
exhausted air
Blower of from the stack
Waste Air
Stream Treated Air to the stack
400C 1.00
500C 1.05
600C 1.15
700C 1.36
800C 1.81
850C 2.22
“Catalysis” was a term used by Berzelius in 1835 to describe the ability of a chemi-
cal element to increase the rate of a chemical reaction between other compounds
without it being appreciably consumed by the reaction. Catalysts do this by lower-
ing the amount of chemical energy required for a given reaction to take place.
Catalytic oxidation converts VOC into carbon dioxide and water, as do
other oxidation processes, but it can be planned to avoid by-products requiring
treatments or final disposal. The advantage of this technology, in comparison
with noncatalytic oxidation, is a substantial reduction of process temperature
(315–455°C); hence catalytic systems represent important energy savings.
Typical catalytic oxidizer components include the catalyst housing, blower,
burner, heat exchanger, controls, and stack (see Figure 7.6). Catalytic oxidation
Treated air
Catalyst
Fuel
Catalytic oxidizers are used Low combustion temperature; Inlet and outlet gas tempera-
to treat gas streams that Inadequate residence time; tures;
have organic vapor concen- Poor mixture of gaseous Accumulation of particles,
trations under the lower compounds; followed by emissions opacity
explosive limit Fouling or plugging;
Loss of catalyst activity
Table 12. Major Advantages and Disadvantages of the Catalytic Oxidation Technology
Advantages Disadvantages
Lower fuel costs than thermal Higher capital costs than thermal oxidation;
oxidation; Operates at lower Less fuel is required compared with other thermal systems
temperatures than thermal because they operate at lower temperatures: from 315 ºC to 455ºC;
oxidation Catalytic oxidizers are not effective with high concentrations of
solids or liquid particles;
Frequent cleaning is desired, so the restoration of catalytic
activity can rise to up 90%;
Catalyst must be optimized for the contaminant present in the
off-gas stream (catalytic conversion is species-dependent);
Catalyst must be cleaned regularly (12- to 24-month intervals);
Several compounds influence catalysts performances (e.g.,
fluorine and sulfur);
Require safety precautions;
Catalysts operate typically for less than 3,000 ppm pollutants
concentrations;
Catalysts must be replaced periodically; Catalyst material may
be hazardous and may require disposal as contaminated waste;
Phosphorous, heavy metals (zinc, lead), sulfur compounds,
chlorine, bromine, iodine, fluorine, and any particulate can
result in shortening the life of the catalyst.
7.4. REFERENCES
Air Pollution Control Technology Verification Center (1999) APC. VOHAP. “Control technologies.”
January 28, 1999.
Air Pollution Training Institute (APTI) (2004) Air Pollution Control Technology Series Training
Tool. Control Technology Series. http://www.epa.gov/air/oaqps/eog/utrain.html
Calgon Carbon Corporation (2002)_ ES-OC-TP-005. “Options in odor control. New alternatives
and traditional technologies”, June 2002.
Croll-Reynolds Clean Air Technologies (1996) “Odor control in edible oil processing. Reprinting
of Pollution Engineering”, CRS-11-96.
Crowder, J. (2004) “Carbon adsorption control devices”, in: Air Control Techniques, P C. Air Pollu-
tion Control Technology APTI, 2004. Control Technology Series.
Davis, R. J., Zeiss, R.F. (1998) “Cryogenic condensation: A cost-effective technology for control-
ling VOC emissions”, Proceedings of the 1998 Annual Meeting of the American Filtration and
Separations Society, May 1998, St. Louis, Missouri.
Environmental Protection Agency (1979) Snyder, W.H. Technical Report. “Guideline for fluid mod-
elling of atmospheric diffusion”, EPA-450/4-79-016 (Draft) 1979.
Environmental Protection Agency. (1995) “Compilation of air pollutant emission factors”, AP-42,
Chapter 9: Food and Agricultural Industries. Fifth Edition, Volume I. United States. http://
www.epa.gov/ttn/chief/ap42/
Environmental Protection Agency (2000) Biosolids and residuals, Management fact sheet. “Odor con-
trol in biosolids management.” http://www.epa.gov/owmitnet/mtb/odor_control biosolids.pdf.
Environmental Protection Agency (1993)“Guideline on air quality models” (revised): Appendix
W of 40 CFR Part 51 (supplement b). (AH-FRL-7478-3). RIN 2060-AF01). Revision to the
Guideline on Air Quality Models. http://www.weblakes.com/lakeepa6.html#GAQM.
Environmental Protection Agency (1992)Technical Support Div. Cox, W.M. Technical Report. “Pro-
tocol for determining the best performing model”, PB-93-226082/XAB; EPA-454/R-92/025.
Integrated Pollution Prevention and Control (IPPC) (2003) European Commission Directorate-Gen-
eral Jrc. Joint Research Centre. Institute for Prospective Technological Studies Technologies for
Sustainable Development. European IPPC Bureau. “BREF (Best available techniques. REFerence
document) on the Slaughterhouses and Animal by-products according to Article 16(2) of Council
Directive 96/61/EC.” http://europa.eu.int/comm/environment/ippc/brefs/sa_bref_1103.pdf.
Julie Van Deuren, Teressa Lloyd, Shobha Chhetry, Raycharn Liou, James Peck. Federal Remedia-
tion Technologies Round Table. “Remediation Technologies Screening Matrix and Reference
Guide, Version 4.” http://www.frtr.gov/matrix2/section3/3_14.html.
Muller, K., Meyer-Pittroff, R. (1990) Master Breweries Association of the Americas. “Condensation
and thermal treatment of brew house vapors”, Technical Quarterly. 27(1), 1-4.
Muller, K. (1990) Master Breweries Association of the Americas. “Combustion of obnoxious sub-
stances contained in brewhouse vapors through the boiler firing system.” Technical Quarterly
27(2), 52-55.
New South Wales Department of Environment and Conservation (2004) “Odour control.” http://
www.epa.nsw.gov.au/mao/odourcontrol.htm.
New South Wales Environment Protection Authority (2001) “Approved Methods and Guidance, For
the Modelling and Assessment of Air Pollutants in New South Wales.” http://www.environment.
nsw.gov.au/resources/amgmaap.pdf.
Rafson, H.J. (ed) (1998) Odor and VOC Control Handbook. McGraw-Hill, New York.
US Department of Energy, Office of Environmental Management and Office of Science and Tech-
nology (2001)“Membrane system for the recovery of volatile organic compounds from remedia-
tion off-gases. Summary report.” Industry Programs and TRU and Mixed Waste Focus Area.
http://apps.em.doe.gov/ost/pubs/itsrs/itsr266.pdf.
US Environmental Protection Agency Research Triangle Park. Control Technology Center (1995) Infor-
mation Transfer and Program Integration Division. Office of Air Quality Planning and Standards.
R. A. Zerbonia. J. J. Spivey. S. K. Agarwal, A. S. Damle. C. W. Sanford., “Vapor gas streams, control
Odors Treatment: Physicochemical Technologies 123
technology center, Survey of control technologies for low concentration organic vapor gas streams”,
EPA-456/R-95-003.1995. http://home.triad.rr.com/sanfordconsult/low_vo.pdf.
US Environmental Protection Agency (2002)Air Pollution Control Cost Manual, Sixth edition,
EPA/452/B-02-001. 2002. http://www.epa.gov/ttn/catc/dirl/c_allchs.pdf.
US Environmental Protection Agency (1994) Volatile Organic Compound Removal from Air
Streams by Membranes Separation. Membrane Technology and Research, Inc. United States
Environmental Protection Agency. EPA/540/F-94/503.
US EPA Model (2004) Regulatory Models - ISCST3 | ISCLT3 | ISC-PRIME | AERMOD | CALINE3
| CALINE4 | CAL3QHC | FDM | UAM - downloads. Environmental Protection Agency,“40 CFR
Part 51(AH-FRL-7478-3)”. RIN 2060-AF01. 04/17/03.http://www.weblakes.com/lakeepa3.html.
United States Environmental Protection Agency (1986) Guideline on Air Quality Models (Revised),
EPA-450/2-78-027R, July 1986.
US Army Center for Health Promotion and Preventive Medicine (2005) Introduction to Air Pollution
Control. http://chppm-www.apgea.army.mil/desp/pages/samp_doc/aircontrol_intro/22.2005).
US Army Corps of Engineers. Engineering and Design (2004)Adsorption Design Guide. DG 1110-
1-2. Chapter 3.http://www.usace.army.mil/usace-docs/design-guides/all.htm.
US Army Corps Of Engineers (2001) Engineering and Design–Air Stripping. CECW-E 1110-1-3. 31.
Oct 2001. http://www.usace.army.mil/usace-docs/design-guides/all.htm.
Van Deuren, J., Lloyd, T., Chhetry, S., Liou, R., Peck, J. (2002) Federal Remediation Technologies
Roundtable, “Remediation technologies screening matrix and reference guide, version 4”, http://
www.frtr.gov/matrix2/section3/3-14.html.
Virtuous Cycle (formerly BRM Business) (2004)Major Components of Recuperative Thermal
Oxidizer. http://www.virtuouscycle.co.uk.
8
Odors Treatment: Biological
Technologies
Bram Sercu, João Peixoto, Kristof Demeestere,
Toon van Elst, and Herman Van Langenhove
1. INTRODUCTION
BRAM SERCU, JOÃO PEIXOTO, KRISTOF DEMEESTERE, TOON VAN ELST, AND
HERMAN VAN LANGENHOVE EnVOC Research Group, Department of Organic Chem-
istry, Faculty of Bioscience Engineering, Ghent University, 9000 Gent, Belgium. Department of
Biological Engineering, CEB, University of Minho, 4710–057 Braga, Portugal. Project Research
Gent nv, 9030 Gent, Belgium.
125
126 B. Sercu et al.
Several biological waste gas treatment reactor concepts exist. They can be distin-
guished according to their filter material (organic or inorganic) and the type of
liquid phase (noncontinuous or continuous). Both characteristics influence mass
transfer and the presence or type of biofilm. In most cases, pollutants are first
transferred from the gas phase to the liquid phase and subsequently to the biofilm
(Figure 1). It has been argued, however, that pollutants can directly be transferred
from the gas phase to the biofilm when no water film is present, or that fungal
mycelia, protruding in the gas phase, can directly take up substrates without a dis-
solution step (Engesser and Plaggemeier, 2000). Optimal conditions for the organ-
isms (micro)environment should be provided. Temperature, pH, water activity,
nutrient availability, oxygen concentration, and osmotic potential are important
parameters. In practice, most of these parameters can be controlled in an accept-
able range by reactor choice, operation, and control. However, sometimes waste
gas characteristics like high or fluctuating temperatures and low oxygen concentra-
tions are more difficult to adjust, leading to limitations in the application area of
biological waste gas treatment techniques.
Laboratory studies provide much information about the removal of single
compounds or simple mixtures, but especially for VOCs, often relatively high
Cg C1
Cthr
Figure 1. Pollutant transfer during biological waste gas cleaning (Waweru et al., 2000).
Odors Treatment: Biological Technologies 127
2.1. Biofilters
Biofiltration is often used with a broader meaning, referring to all the waste air
biological technologies. Strictly it means the most used technology about this
matter: the biofilters. They made their appearance in the 1950s in the deodor-
ization of air from wastewater treatment or composting plants. Nowadays they
are aimed not only at air polluted with organic gases, such as VOCs and many
other hydrocarbons, but also ammonia or H2S. They are more efficient with
low molecular weight gases, with high solubility in water and simpler molecular
bonds. Biofiltration has been extensively used, because waste gases in this case
generally contain low concentrations of well biodegradable organic and inor-
ganic compounds.
Basically, a biofilter is a layer of biologically active media (an organic filter
matrix), usually of natural origin. The filter particles are typically soil, compost,
peat, wood chips, tree bark, and heather. Granular activated carbon and plas-
tic material are also used. One kind or several combinations of particles have
been used. The media must provide a large surface area, nutrients and moisture
(around 50% of the media) for the microbial activity, and adsorption/absorption
of the odorous molecules. The microflora for the degradation of odors—mainly
128 B. Sercu et al.
bacteria and fungi—is part of the package. There is no continuous water phase.
For better results, the addition of nutrients containing nitrogen and phosphorus
must be considered, although this will add some cost to the process. The pres-
ence of bulking inerts usually calls for the addition of nutrients, mainly with
high load regimes (Devinny et al., 1999). Adequate porosity (around 0.50) is
essential for low pressure drop (power requirements).
To build a conventional open-bed filter (Figure 2), in the early ages of the
technique, a hole was excavated in the ground (around 1.0 m deep) and filled up
with a bed of the selected media. Nowadays, synthetic material or concrete is
used. Perforated piping or other systems are used for gas distribution under the
bed. The waste air flow, combined with the void fraction, causes the residence
time to be normally between 15 and 60 sec, the time it takes for the odors to be
absorbed and metabolized through the filter. Surface loading rates are about
1.2 m3 m−2 min−1 (Devinny et al., 1999). Impermeableness is desirable to avoid
liquid leaching.
For optimal long-term operation of biofilters, next to controlling the biofilter
moisture content, precautions should be made to prevent acidification if sulfur
or nitrogen-containing compounds are present. This can be accomplished by
buffering, e.g., by adding CaCO3 (Rafson, 1998), or regular replacement of the
filter material (every 1 to 5 years, depending on the loading rate). The latter treat-
ment is also needed to remove other accumulated intermediates or end-products,
to prevent high pressure drops, and to prevent nutrient limitation if nutrients
are not provided during biofilter operation. Indeed, removal efficiencies go from
60% to 100%, depending on the media and the pollutants contaminating the
air. Initial performance is very good but as time goes on problems may occur,
leading to severe degradation of efficiency. Clogging and channeling are likely
to appear.
Modular closed systems are commercially available. They minimize the
surface for installation because they are stacks of trays that can be set in
series or parallel arrangements, or combinations of both. The usual time of
operation, with good removal characteristics, for conventional systems (2 to
Clean air
Filter media
Biotrickling filters are single unit operation reactors (for both capture and
destruction), like biofilters. A packed column is inoculated with microorgan-
isms that attach to the particles. Biofilms grow using nutrients supplied by
the contaminated airstream and by a liquid flow that trickles down the pack-
ing, continuously or periodically. The liquid moving phase and the inorganic
nature of the media particles are the most important differences between
biofilters and biotrickling filters (Figure 3). Unlike biofilters that use natural
materials, most particles for biotrickling filters are built with plastic, steel,
or ceramic material. The simplest of all is the Raschig ring. Many particle
designs have been used. The odor is first transferred from the air to the circu-
lating water. Next it must diffuse to the biofilm. Finally the microorganisms
oxidize the compounds.
In previous studies using a biotrickling filter for VOC removal with Pseu-
domonas putida as the biodegrading bacteria, several packing materials were
checked to try to avoid channeling and clogging to block the process (Peixoto
and Mota, 1998). It was proved that it is very difficult to overcome the men-
tioned problems, even when using 20-mm Raschig rings. The presence of the
four phases (gas–vapor, liquid, biological, and solid) involved in the process
Clean air
Packed
column
Fresh
water
Polluted Purge
Nutrients
air
makes things very hard to deal with. Even with a high surface area and poros-
ity, the bacterial growth reduced significantly those parameters in a short time.
Ultimately the flow used only one last channel (Figure 4), with evident poor effi-
ciency and higher pressure drop, and the process had to stop. Clogging is a direct
result of the bacterial (fungal) growth. Growth means that the microorganisms
are metabolizing the pollutants, as they are meant to. Therefore, it does not seem to
make sense to try to limit the growth to avoid clogging and channeling. A way
to remove the exceeding biomass seems to be the natural answer but very tough to
find and a settling tank would be needed after the reactor. Backwashing and high
shear stress do not seem to solve the problem.
2.3. Biowashers
In biowashers the biomass is suspended in the liquid phase. The waste content
from the air is first washed and next oxidized by the suspended microorganisms.
There may be one or two separate units for absorption and metabolization.
Classically, the airstream is washed in a spray chamber (scrubber; packed
bed scrubber) and then the liquid phase is sent to an activated sludge tank, where
the pollutants are oxidized (Figure 5).
Together, these two units form a bioreactor which is commonly named bio-
scrubber. Instead of the activated sludge tank, the contaminated liquid may be
sent to an airlift reactor (Ritchie and Hill, 1995; Rittmann et al., 2000). This
kind of reactor, also named circulating-bed biofilm reactor, is known for its good
Liquid In Liquid In
Air/Vapor Out Air/Vapor Out
Air/Vapor In Air/Vapor In
Figure 4. Evolution of a biotrickling filter packed with Raschig rings, after colonization by bacteria,
due to clogging and chaneling.
Odors Treatment: Biological Technologies 131
Airlift reactor
Washing
chamber
(scrubber)
Activated sludge
tank Polluted air,
Polluted fresh water
air Air (O2) and nutrients
Figure 5. Schematic of biowashers: to the left, a two unit reactor with scrubber and activated sludge;
to the right, a single unit airlift reactor.
Table 1. Four entry characterization of biological systems for air pollution control
LIQUID PHASE
Circulating Stationary
Biowasher Spray/
Dispersed
MICROBIAL shower
COLUMN TYPE
FLORA Biotrickling Biofilter Packing;
Attached
filter; BPT plates
Supplied with Part of the
water packing media
MINERAL NUTRIENTS
so on, are variables to be considered. With so many variables, there is still a long
way to go before one can be sure of what solution is the best. For the same rea-
son, modeling is also very difficult. For further information about reactor design
and modeling, books of Devinny et al. (1999) and Kennes and Veiga (2001) are
recommended.
Figure 6. Liquid distributor showing the growth of biomass on the lower surface. On the top side,
that occurrence was even stronger.
134 B. Sercu et al.
horizontal surfaces. Basically, the BPT is a pile of parallel circular plates with a
single hole on the border. The plates are placed in such a way that the holes will
alternate (180°) from one to the next plate. In this way, a cascade of liquid will go
downward, changing direction from plate to plate. The gaseous stream follows
the opposite direction, upward. The bacteria attach to their top surface. Figure 7
shows the schematic of the flows and biofilm growth on the plates.
The reactor is a four-module (about 28.8 dm3 each) BPT with 20 plates in
each module. An individual plate surface area (top face) is about 40 195 mm2.
The scratched surfaces of the plates were intended to make the bacterial adhe-
sion easier. Only two or three of the four modules are operated continuously.
The other(s) is kept free and ready to replace any one that reaches saturation
with biomass. In this way the operation can be kept going virtually forever.
The performance is quite stable (the biofilm activity, surface-dependent, is
kept approximately constant) and the constant surface contact area makes it
easy to model and scale-up the process. The total surface area and the space
between plates can be designed for the desired operating time. In theory, the
available surface in a BPT is a tenth of the surface in a biotrickling filter, consid-
ering the same total volume.
The new design proved to ensure a stable operation for longer periods, as well
as high VOC removal (92 % removal for inlet toluene concentration of 10 g m−3
and empty bed residence time, EBRT = 108 sec). It has very good hydrodynamic
performance and operates continuously without problem. In the long term, the
short area is compensated by the steady operation.
The disposal of the newly formed biomass is also much easier than in the
biotrickling filter. Unlike biofilters whose packing has to be rejected after a certain
Air/Vapor Air/Vapor
Liquid Liquid
Figure 7. Simplified schematics of the BPT, with only five plates, to better visualize the directions of
both flows and the attached biofilm on the upper surface of the plates.
Odors Treatment: Biological Technologies 135
Figure 8. Photograph of the bottom plates of the first module showing the biofim growth on the
BPT plates. The huge biofilm does not endanger the permeability of the system.
136 B. Sercu et al.
In the membrane bioreactor concept, one side of the membranes is dry and acts
as a surface for uptake of pollutants from the air flowing along the membranes,
while the other side is kept wet and covered by a biofilm. In Figure 9, a flat mem-
brane bioreactor with a composite membrane is shown, but also other configu-
rations like hollow fiber membranes modules can be applied. Pollutants diffuse
through the membrane and are subsequently degraded by the microorganisms in
the biofilm or in the recirculating aqueous phase. By continuous recirculation of
the aqueous phase, the microbial degradation process can be easily controlled. The
main advantages of membrane bioreactors for waste gas treatment include the
high specific surface area, the ability to prevent clogging, the good reactor control,
the physical separation of gas and biofilm, the low pressure drop, the absence of
channeling, and the independent control of gas and liquid phase (De Bo, 2002).
Potential disadvantages are the high investment costs, the additional mass transfer
resistance caused by the membrane, a decreased biofilm activity as the biofilm
ages, and clumping of hollow fiber membranes at high biofilm growth. The reac-
tor concept, although not implemented in practice yet, has potential to eliminate
VOCs characterized by poor water solubility, by lack of biodegradability, and by
toxicity (Reij et al., 1998). Recently a flat membrane reactor was developed and
applied for the degradation of DMS and toluene as single compounds (De Bo
et al., 2002, 2003). In this case a composite membrane was used, guaranteeing a
Biofilm
Membrane
Substrate
Nutrients
Oxygen
pH-buffers
Mineral
Waste
medium
air
Dense Porous
Figure 9. Scheme of a flat membrane bioreactor for waste gas treatment (De Bo, 2002).
Odors Treatment: Biological Technologies 137
Most of the work has been done about removal of ammonia, reduced sulfur
compounds (either single or as a mixture), and odorous VOCs, especially with
biofiltration. About compounds like amides, indole, scatole, and pyridine no
information was found. However, it can be assumed that when bioreactors can
operate with high odor reduction efficiency, these compounds are also suffi-
ciently degraded.
I II
NH3 NH3 NH4+ NH4+
Nitrosomonas
III
NO2 −
Nitrobacter
NO3−
Figure 10. Main processes taking place during ammonia biofiltration (I: absorption; II: adsorption;
III: nitrification).
138 B. Sercu et al.
Since ammonia has a low Henry’s law constant (H20 °C = 5.6 × 10−4) (Perry
and Green, 1984) and a protonation constant pKa, 20 °C of 9.23 (Weast et al.,
1984), in biofilters it is partly retained by adsorption onto the carrier material
and by absorption into the water fraction of the carrier material. In this context,
Shoda (1991) reports a maximum volumetric NH3 elimination capacity of 15 g
m−3 d−1 in a peat biofilter due to these physical–chemical transfer processes. In
a compost biofilter, Smet et al. (2000) obtained an NH3 adsorption and absorp-
tion capacity, per volume of compost, of 490 g m−3 and 47 g m−3, respectively,
at an NH3 inlet concentration of 159 ppmv and a compost moisture content
of 40%. Next to these physical–chemical processes, nitrification by the autotro-
phic bacteria Nitrosomonas and Nitrobacter is generally considered as the main
microbiological process for the degradation of NH3 (Terasawa et al., 1986; Van
Langenhove et al., 1988; Williams, 1995). More recently, the autotrophic genera
Nitrosospira and Nitrospira also are reported to be responsible for nitrification
(Schramm et al., 2000; Regan et al., 2002).
As a result of these phase transfer and (micro)biological processes, NH3 has
been removed efficiently at concentrations up to 50 ppmv. Using a wood bark
biofilter, Van Langenhove et al. (1988) obtained removal efficiencies of at least
90% at concentrations between 6 ppmv and 17 ppmv and at NH3 mass loading
rates (Bv) up to 58 g m−3 d−1. At similar concentrations (4 ppmv to 16 ppmv),
Weckhuysen et al. (1994) observed elimination efficiencies of 83% or higher at
NH3 mass loading rates between 6.8 g m−3 d−1 and 27.2 g m−3 d−1. In an inoculated
peat biofilter, NH3 elimination capacities (EC) up to 41 g m−3 d−1 are reported
at an inlet concentration of 20 ppmv (Hartikainen et al., 1996). More recently,
removal efficiencies as high as 99.5% were obtained for 100 days in an inoculated
perlite biofilter, at concentrations of 50 ppmv and NH3 loading rates between
8.6 g m−3 d−1 and 21.5 g m−3 d−1 (Joshi et al., 2000).
Due to the sensitivity of nitrifying microorganisms, however, biofiltration of
waste gases containing high ammonia concentrations (above 50 ppmv) has been
reported to be questionable. Don (1985) and Hartikainen et al. (1996) reported the
biofilter removal efficiency for NH3 dropped drastically at waste gas concentrations
exceeding 35 ppmv to 60 ppmv. However, more recently, Liang et al. (2000) could
obtain NH3 removal efficiencies of at least 95% at inlet concentrations between
20 ppmv and 500 ppmv in compost biofilter in which active carbon was added to
reduce compaction and channeling, as well as to increase the reactive surface and
durability of the biofilter. Similarly, removal efficiencies over 90% were achieved
by Kim et al. (2002) at inlet concentrations up to 150 ppmv in a biofilter system
packed with small cubes of polyurethane sponge coated with a powder mixture
of activated carbon and natural zeolite. Kalingan et al. (2004) obtained complete
NH3 removal in a peat biofilter containing inorganic supporting material at a NH3
concentration of 200 ppmv, while Smet et al. (2000) observed no NH3-toxicity in a
compost biofilter at concentrations up to 775 ppmv.
Besides the ammonia input concentration, the mass loading rate of a bio-
filter seems to be critical for an efficient performance. In an inoculated activated
Odors Treatment: Biological Technologies 139
for instance, found almost equal H2S removal efficiencies at pH values between
3.2 and 8.8. Only at pH = 1.6, the removal efficiency decreased to 15%. The high
H2S removal efficiency at pH = 3.2 was attributed to the abundance of acido-
philic sulfur oxidizing bacteria. Also other studies did not report decreased H2S
removal efficiencies at pH values as low as 3 (Wada et al., 1986; Cook et al., 1999)
or even 1.2 (Yang et al., 1994). During biofiltration, the pH will first decrease at
the inlet side of the biofilter, where most of the H2S is oxidized and the low pH
front will consequently move to the deeper parts of the biofilter (Cook et al.,
1999). In general, it should be sufficient to maintain a pH value higher than 3 for
efficient H2S removal. However, it could be useful to maintain neutral pH values
to prevent inhibition of the removal of other compounds present in the waste
gas, corrosion, and increased filter medium degradation. To increase the pH of
the biofilter material, washing can be applied (Yang and Allen, 1994b), although
only small pH increases are usually obtained. Smet et al. (1996b) observed that
regeneration of an acidified biofilter (pH = 4.7) was not possible by trickling
tap water or buffer solution over the bioreactor, because most of the sulfate was
leached as the corresponding sulfate salts and not as sulfuric acid. In addition,
leaching caused washout of essential microbial elements. Alternatively, the use
of more concentrated buffer solutions in combination with a complete mineral
medium or mixing with limestone powder was recommended.
Next to acidification, the accumulation of elemental sulfur and sulfate in the
filter material can potentially inhibit microbial activity. Yang and Allen (1994b)
found the highest concentrations of both compounds at the inlet side of the
biofilter. Elemental sulfur was present because it was formed as an intermedi-
ate during incomplete H2S oxidation after exposure to high H2S concentrations.
By adding increasing amounts of sulfate to different biofilters, Yang and Allen
(1994a) observed that concentrations (mass of S per mass of compost) exceed-
ing 25 mg g−1 were inhibitory for H2S removal, probably due to toxic effects. This
inhibition effect, however, was not confirmed by Jones et al. (2003), for sulfate
concentrations up to 100 mg g−1. In general, it is recommended to evaluate the
expected H2S loading rate before designing a biofilter. If it is assumed that all
sulfur entering a biofilter will ultimately accumulate as sulfate, its cumulative
concentration can be calculated to assess the long-term deactivation of a biofil-
ter, e.g., with a threshold of 25 mg g−1.
Because H2S is very biodegradable, EBRTs can be rather low, e.g., 15 s
(Yang and Allen, 1994a) without affecting the H2S removal efficiency. Possibly
other, less biodegradable or water-soluble compounds present in the waste gas
will determine the lower limit of the EBRT. Next to organic materials like com-
post, peat, or wood bark, different alternative carrier materials were described
for H2S biofiltration, being rockwool, fuyolite, and ceramics (Kim et al., 1998),
a pelletized mixture of pig manure and sawdust (Elías et al., 2000), pellets of
agricultural residues (Elías et al., 2000), porous lava inoculated with Thiobacillus
thiooxidans (Cho et al., 2000), and microorganisms immobilized in Ca-alginate
(Chung et al., 1996a,b, 1997, 1998; Huang et al., 1996; Park et al., 2002).
142 B. Sercu et al.
had to be supplied every two weeks to support growth of the heterotrophic organ-
isms. A number of researchers used microorganisms immobilized in Ca-alginate
to remove mixtures of H2S and NH3, although the performance of these reactors
was somewhat lower than with the more traditional biofilters (Chung et al., 2000,
2001a,b). Possible advantages, however, are increased possibilities for pH control
and removal of metabolic products (elemental sulfur and (NH4)2SO4), as is also
the case with biotrickling filters.
(Weckhuysen et al., 1993; Sercu et al., 2005a). This can lead to a pH decrease,
potentially limiting the removal efficiencies of other compounds in the waste
gas. Next to biofilters, biotrickling filters have been used to remove odorous
VOCs. Again, high removal efficiencies have been obtained for aldehydes, alco-
hols, and volatile fatty acids (Chang and Lu, 2003; Chua et al., 2000; Ibrahim
et al., 2001; Kirchner et al., 1991), even at low EBRT. Kirchner et al (1987), for
instance, showed > 90% removal efficiencies for compounds like aldehydes and
alcohols at 5 ppmv to 40 ppmv influent concentrations and 2.4 s EBRT. Ibrahim
et al. (2001) found 92% and 95% removal efficiencies for 10 ppmv acetaldehyde
and propionaldehyde inlet concentrations, respectively, in a column packed with
immobilized activated sludge beads at EBRT = 12.4 s. At higher influent con-
centrations, the removal of both compounds decreased, however, due to inhibi-
tory effects. For higher influent concentrations, higher EBRT values are needed,
as shown by Chang and Lu (2003). They found nearly complete isopropanol
removal efficiencies in a biotrickling filter, operated between 20 s and 90 s EBRT
time at influent concentrations between 100 ppmv and 500 ppmv. When too high
influent loadings are applied, accumulation of compounds or intermediates can
occur. Chua et al. (2000) found > 99% removal efficiencies for butyric and valeric
acid in a biotrickling filter, at mass loadings between 4.8 g m−3 h−1 and 37.8 g m−3
h−1 (0.05 g m−3 to 0.86 g m−3). However, at loading rates exceeding 32 g m−3 h−1,
the maximal biodegradation capacity was reached and accumulation of volatile
fatty acids in the liquid phase was observed.
From the previous sections, it is clear that most of the components present in
odorous mixtures can be removed efficiently with biological waste gas cleaning
techniques, when properly operated, even at relatively high influent concentra-
tions. Also, in industrial applications treating mixtures of compounds, often high
(odor) removal efficiencies can be obtained. Park et al. (2001), for instance, used
a biotrickling filter packed with ceramics and inoculated with activated sludge to
remove odors at a composting facility. After a 30 d acclimation period, > 95%
removal efficiencies were obtained for NH3 and H2S during about 60 d of opera-
tion. Also, at a biosolids composting facility, Goodwin et al. (2000) found effi-
cient odor removal with a biofilter (> 95%) after about 3 months of operation at
EBRT = 20 s, as determined with olfactometric analyses. Luo (2001) observed >
98% odor reduction with wood bark biofilters treating rendering emissions during
a period of 3 years, at EBRT = 6.8 min. Reducing the EBRT to 1.7 min did not
affect the odor removal efficiencies during the first 3 months of operation. After
22 months, however, the odor removal efficiency was 99.1% at EBRT = 6.8 min
and only 29.7% at EBRT = 1.7 min. This clearly shows that regular filter medium
replacement is necessary, especially when lower EBRT values are used.
When a complex mixture of odorous compounds has to be treated, removal
efficiencies of the single compounds can be smaller than expected. This can be
146 B. Sercu et al.
caused by, e.g., toxic effects of substrates or metabolites. Van Langenhove et al.
(1989a) compared the applicability of a tree bark biofilter for removing odors
from a vegetable processing industry, mainly emitted during the blanching pro-
cess. The main odorous compounds identified were sulfides, isothiocyanates,
nitriles, and aldehydes. In pilot-scale experiments all compounds were removed
with > 95% removal efficiencies at a volumetric loading rate of 200 m3 m−2 h−1.
However, a full-scale biofilter, designed according to the results obtained from
the pilot-scale studies, had lower removal efficiencies after 6 months of opera-
tion (45% to 65% for sulfides). This was found to be caused by the accumulation
of isothiocyanates, which was not observed during the short-term pilot-scale
experiments. For hexanal, Van Langenhove et al. (1989b) observed 85% removal
efficiency in a wood bark biofilter, at 10 ppmv inlet concentration and EBRT =
0.33 min. To simulate emissions from a food processing plant, 40 ppmv SO2 was
added to the waste stream, leading to a drastic decrease of the hexanal removal
efficiency to 40%. Next to toxic effects, preferential degradation of easily biode-
gradable compounds can inhibit the removal of other compounds. Smet et al.
(1997), for instance, found that isobutanal was preferentially degraded before
DMS, in a biofilter inoculated with Hyphomicrobium MS3, when both com-
pounds were simultaneously dosed. This could cause low removal of DMS when
a biofilter is designed too small.
4. CASE STUDIES
4.1. Methodology
In all case studies mentioned in this paragraph, samples have been taken of the
untreated and the treated airflows, in order to determine important parameters.
First, the chemical composition of the airflow was revealed using GC-MS analy-
sis. These data are very useful for the determination of the total chemical load
going to the bioreactor, as well as for improving the working efficiency of it,
being able to indicate the compounds or groups of compounds that are degraded
insufficiently. The second type of analyses used is the determination of the total
odor concentration, using dynamic olfactometry. These data are used to deter-
mine the total odor removal efficiency of the bioreactor, which is the final wanted
effect of the use of a bioreactor in case of odor problems.
4.1.1a. Sampling procedure. The gases were sampled using a method that
involved preconcentration on an adsorbent. This preconcentration step was
carried out at the sampling location. Tenax TA was used as adsorbent. Tenax
TA is a porous polymer based on 2,6-diphenylene oxide. It has been specifically
designed for the trapping of volatiles and semivolatiles (SIS, 2000). The collected
Odors Treatment: Biological Technologies 147
waste gases were cooled at about 4 °C before adsorption. This cooling stage
was used in order to increase the breakthrough volume and in order to separate
excess water vapor. The sampled adsorption tubes were filled with approximately
750 mg Tenax. Sampling rate was 200 ml min−1 and sampled volume varied
between 50 ml and 10 L.
4.1.1b. Analysis. The analysis of the VOCs present in the sample was done
in different steps, including desorption from the adsorbent, separation by gas
chromatographic techniques, quantification by flame ionization detection, and
subsequent identification through mass spectrometry. The desorption step con-
sisted of a thermal desorption. A second preconcentration (cryogenic trapping
of the VOCs) was necessary in order to achieve good chromatographic separa-
tion. The cryogenically concentrated samples were introduced immediately into
the GC by rapid heating of the trap. The different compounds were separated
in a gas chromatograph (Varian 2700) with a 100% polydimethylsiloxane apolar
column (type DB-1, 30 m × 0.53 mm, film thickness 5 μm, J&W Scientific). The
mass spectrometer used was a Finnigan MAT 112 S with an electron impact ion
source and a magnetic sector analyser.
4.1.2. Olfactometry
4.1.2a. Sampling procedure. The gases were sampled using the static
sampling method. In this method, a sample is collected and transferred into
a sampling container (bag). Collecting the sample was done with the “lung
principle,” where the sample bag is placed in a rigid container and the air is
removed from the container using a vacuum pump. The partial vacuum cre-
ated in the container causes the bag to fill with a volume of sample equal to
the volume that was removed from the space around the bag in the rigid con-
tainer. In some sampling points, where a risk of condensation in the sampling
bag existed due to high humidity and high temperatures, a predilution was
applied using dry odor-free nitrogen. Sampling materials used were Teflon™
for tubing and disposable sampling bags made of Nalophan™ film. All samples
were analyzed by the accredited odor laboratory of PRA OdourNet bv (The
Netherlands) within 30 hours after sampling. During transportation, samples
were not exposed to direct sunlight. All measurements of the odor concen-
trations were executed by dynamic olfactometry according to the EN13725
(CEN, 2003).
as a multiple of one European odor unit per cubic meter (ouE m−3) at standard
conditions for olfactometry.
Crushing and extraction plants for vegetable oils often cause considerable emis-
sions of odor, which may cause offense in nearby residential areas. The type of
oil seed processed partially determines the amount and type of odor released.
Only limited information is found in literature on the composition of these
waste gases, on source strength classification, and on possible treatment meth-
ods. Lacoste et al. (1996) performed a quantitative study of odorous compounds
in gas effluents from three rapeseed crushing plants. Olfactometry was used to
determine odor concentrations of gaseous effluents. Chemical analyses revealed
the presence of nitriles, aldehydes, and sulfur compounds like mercaptans in
conditioning and pressing emissions, while hydrogen sulfide and acetaldehyde
were the major odorant compounds in the absorption unit effluents.
An oil crushing and extraction plant, situated in an industrial area in the
northern part of France, mainly processes soybean, sunflower, and rapeseed to
produce vegetable oils. Significantly higher odor emissions occurred when pro-
cessing rapeseed (colza) compared to those associated with other types of oil
seed. Because of the growing number of complaints arising from the surrounding
residential area at distances of more than one kilometre, the plant management
decided to tackle the odor problem. The process of improving the odor situation
in the vicinity of this plant was a process that took several years and included
various types of measurements and interim evaluations. The measurements were
a combination of chemical analyses, olfactometry, and field panels. Chemical
measurements (gas chromatography, combined with mass spectrometry) were
mainly used to get a better understanding of the composition of the different
waste gas streams on the plant. Different gas streams have been sampled to iden-
tify and quantify the VOCs present. In the interpretation, special emphasis was
put on compounds with a low odor threshold. Olfactometric measurements were
used to determine the total amount of odour present in the waste gas stream.
These data were very useful to make a classification of the different sources in
order to set priorities for abatement but also to calculate the total odor abate-
ment efficiency of treatment systems. Field panel measurements determined the
impact of the total odor emission on the vicinity of the plant.
The first step in the processing of the seeds consists of a number of physical
treatments, like cleaning, crushing, heating up to 60°C, pressing, and cooling.
The main odor sources in this treatment are the hot and humid vapors that
Odors Treatment: Biological Technologies 149
arise at the heating stage and during pressing. Also the cooling of the material
before entering the extraction unit, which takes place on the open conveyor belt
between the crushing and the extraction unit, can be considered as an impor-
tant source. The residual oil is then extracted from the flakes with hexane in an
extraction unit. Hexane and oil are separated in a distillation unit. The residual
fraction of the seed is treated in a desolventizer to remove hexane. Before being
vented in the atmosphere, the vapors of the extraction process pass through an
absorption system with mineral oil to recapture hexane. The emissions of the
absorption contain high concentrations of hydrogen sulfide, and thus represent
an important odor source. The extracted residue of the seed is dried and cooled
and sold as livestock feed. Large amounts of fresh air are used in this process
and are emitted to the atmosphere, loaded with odorous components. Olfacto-
metric emission measurements of a selected number of odor sources resulted in
the following emissions (see Table 2).
The vapors of the conditioners were already incinerated in both steam boil-
ers with an odor removal efficiency higher than 90%. The resulting calculated
emissions in European odor units per hour demonstrate the importance of the
emission of the absorption unit on one hand (72%) and of the drying-cooling
unit on the other hand (18%).
Chemical measurements were carried out on the six sampled emission points.
Table 3 gives an overview of the compounds per chemical group. The data in Table 3
are the emitted mass flows per hour for five sampling points (conditioners not
included). The results of the chemical measurements show an important contribu-
tion of mainly organic sulfur compounds and hydrogen sulfide to the total odor
concentration, considering their low odor threshold. Though no data were found
on the odor threshold of the specific nitriles and 4-isothiocyanato-1-butene,
there might be an important influence of these compounds to the global odor
concentration. The high hydrocarbon content in absorption, conveyor belt, and
drying-cooling are mainly caused by the presence of hexane as extraction solvent
(hexane, 2-methylpentane, 3-methylpentane, cyclohexane, methylcyclopentane).
Boiler Drying-
exhaust Presses Absorption Conveyor belt cooling
Figure 11. Scrubber and biofilter (left); fan and silencer (right).
A value between 1000 and 2000 (ouE m-3) can be considered as a normal back-
ground value for the typical “own smell” of a good working biofilter. In some
cases, still lower values are possible (up to 500 ouE m-3).
6. REFERENCES
Busca, G., Pistarino, C. (2003) “Abatement of ammonia and amines from waste gases: a summary.”
16:157.
CEN (2003) EN 13725, Air quality — determination of odour concentration by dynamic olfactom-
etry, European Standard, Brussels, Belgium.
Chang, C.-T., Chen, B.-Y., Shiu, I.-S., Jeng, F.-T. (2004) Biofiltration of trimethylamine-containing
waste gas by entrapped mixed microbial cells, Chemosphere 55:751.
Chang, K., Lu, C. (2003) Biofiltration of isopropyl alcohol in a trickle-bed air biofilter, Biodegrada-
tion 14:9.
Cho, K.-S., Hirai, M., Shoda, M. (1991) Degradation characteristics of hydrogen sulfide, methane-
thiol, dimethyl sulfide and dimethyl disulfide by Thiobacillus thioparus DW44 isolated from peat
biofilter, J. Ferment. Bioeng. 71:384.
Cho, K.-S., Hirai, M., Shoda, M. (1992) Enhanced removability of odorous sulphur-containing
gases by mixed cultures of purified bacteria from peat biofilters, J. Ferment. Bioeng. 73:219.
Cho, K.-S., Ryu, H. W., Lee, N. Y. (2000) Biological deodorization of hydrogen sulfide using porous
lava as a carrier of Thiobacillus thiooxidans, J. Biosci. Bioeng. 90:25.
Chou, M.-S., Shiu, W.-Z. (1997) Bioconversion of methylamine in biofilters, J. Air Waste Manage.
Assoc. 47: 58.
Chua, H., Li, X. Z., Yu, P. F. H., Tam, C. Y., Huang, Y. L., Yang, S. T. (2000) Design and perfor-
mance of a fibrous bed bioreactor for odor treatment, Appl. Biochem. Biotechnol. 84–86:469.
Chung, Y.-C., Huang, C., Liu, C. H., Bai, H. (2001a), Biotreatment of hydrogen sulfide- and ammonia-
containing waste gases by fluidized bed bioreactor, J. Air Waste Manage. Assoc. 51:163.
Chung, Y.-C., Huang, C., Pan, J. R., Tseng, C.-P. (1998) Comparison or autotrophic and mixotrophic
biofilters for H2S removal, J. Environ. Eng. April 1998:362.
Chung, Y.-C., Huang, C. P., Tseng, C.-P. (1996a), Biodegradation of hydrogen sulfide by a laboratory-
scale immobilized Pseudomonas putida CH11 biofilter, Biotechnol. Prog. 12:773.
Chung, Y.-C., Huang, C. P., Tseng, C.-P. (1996b), Operation optimization of Thiobacillus thioparus
CH11 biofilter for hydrogen sulfide removal, J. Biotechnol. 52:31.
Chung, Y.-C., Huang, C. P., Tseng, C.-P. (1997) Removal of hydrogen sulphide by immobilized Thio-
bacillus sp. strain CH11 in a biofilter, J. Chem. Technol. Biotechnol. 69:58.
Chung, Y.-C., Huang, C., Tseng, C.-P. (2001b), Biological elimination of H2S and NH3 from waste
gases by biofilter packed with immobilized heterotrophic bacteria, Chemosphere 43:1043.
Chung, Y.-C., Huang, C. P., Tseng, C.-P., Pan, J. R. (2000) Biotreatment of H2S- and NH3-contain-
ing waste gases by co-immobilized cells biofilter, Chemosphere 41:329.
Chung, Y.-C., Lin, Y.-Y., Tseng, C.-P. (2004) Operational characteristics of effective removal of H2S
and NH3 waste gases by activated carbon biofilter, J. Air Waste Manage. Assoc. 54:450.
Cook, L. L., Gostomski, P. A., Apel, W. A. (1999) Biofiltration of asphalt emissions: full-scale opera-
tion treating off-gases from polymer-modified asphalt production, Environ. Prog. 18 :178.
De Bo, I. (2002) Membrane biofiltration of single-compound waste gas streams, PhD thesis, Ghent
University, Belgium.
De Bo, I., Heyman, J., Vincke, J., Verstraete, W., Van Langenhove, H. (2003) Dimethyl sulfide removal
from synthetic waste gas using a flat poly(dimethylsiloxane)-coated composite membrane biore-
actor, Environ. Sci. Technol. 37:4228.
De Bo, I., Jacobs, P., Van Langenhove, H. (2002) Removal of toluene and trichloroethylene from
waste air in a membrane bioreactor, Environ. Sci. Pollution Res. Special Issue 3:28.
Defoer, N., De Bo, I., Van Langenhove, H., Dewulf, J., Van Elst, T. (2002) Gas chromatography-
mass spectrometry as a tool for estimating odour concentrations of biofilter effluents at aerobic
composting and rendering plants, J. Chromatogr. A 970:259.
Degorce-Dumas, J. R., Kowal, S., Le Cloirec, P. (1997) Microbiological oxidation of hydrogen
sulphide in a biofilter, Can. J. Microbiol. 43:264.
Odors Treatment: Biological Technologies 155
Demeestere, K., Van Langenhove, H., Smet, E. (2002) Regeneration of a compost biofilter degrading
high loads of ammonia by addition of gaseous methanol, J. Air Waste Manage. Assoc. 52:796.
De Roo, K., Van Langenhove, H. (2000) Emissions from animal rendering, in: Proceedings of the 7th
symposium on analytical and environmental problems, Z. Galbacs, ed., Szeged, Hungary, October
2, 2000, p. 38.
Devinny, J. S., Deshusses, M. A., Webster, T. S. (1999) Biofiltration for Air Pollution Control. CRC
Press LLC, Boca Raton.
Devos, M., Patte, F., Rouault, J., Laffort, P., Van Gembert, L. J. (1990) Standardized Human Olfac-
tory Thresholds, Oxford University Press, Oxford, 165 p.
Don, J. A. (1985) The rapid development of biofiltration for the purification of diversified waste gas
streams, VDI Berichte 561:63.
Elías, A., Barona, A., Ríos, F. J., Arreguy, A., Munguira, M., Peñas, J., Sanz, J. L. (2000) Application of
biofiltration to the degradation of hydrogen sulfide in gas effluents, Biodegradation 11:423.
Engesser, K.-H., Plaggemeier, T. (2000) Microbiological aspects of biological waste gas purification,
in: Biotechnology, 2nd edition, Volume 11c. Environmental Processes III, Solid waste and waste gas
treatment, preparation of drinking water, J. Klein, J., and J. Winter, eds., Wiley-VCH, Weinheim,
pp. 275–302.
Gabriel, D., Deshusses, M. A. (2003) Retrofitting existing chemical scrubbers to biotrickling filters
for H2S emission control, Proc. Natl. Acad. Sci. U. S. A. 100:6308.
Goodwin, J. P., Amenta, S. A., Delo, R. C., Del Vecchio, M., Pinnette, J. R., Pytlar, T. S. (2000) Odor
control advances at cocomposting facility, Biocycle January 2000:68.
Hartikainen, T., Ruuskanen, J., Vanhatalo, M., Martikainen, P. J. (1996) Removal of ammonia from
air by a peat biofilter, Environ. Technol. 17:45.
Heller, W., Schwager, H. (1996) Luftreinigung durch Kompost-Biofiltration, Agrarforschung 3:520.
Huang, C. P., Chung, Y. C., Hsu, B. M. (1996) Hydrogen sulfide removal by immobilized autotrophic
and heterotrophic bacteria in the bioreactors, Biotechnology Techniques 10:595.
Ibrahim, M. A., Mizuno, H., Yasuda, Y., Fukunaga, K., Nakao, K. (2001) Removal of mixtures of
acetaldehyde and propionaldehyde from waste gas in packed column with immobilized activated
sludge gel beads, Biochem. Eng. J. 8:9.
Jones, K., Martinez, A., Rizwan, M., Boswell, J. (2003) Evaluation of sulfur toxicity and media capac-
ity for H2S removal in biofilters packed with both natural and commercial media, in: Proceedings
of the 96th annual Air & Waste Management Association Conference, June 22–26, San Diego, CA.
Joshi, J. A., Hogan, J. A., Cowan, R. M., Strom, P. F., Finstein, M. S. (2000) Biological removal of
gaseous ammonia in biofilters: space travel and earth-based applications, J. Air Waste Manage.
Assoc. 50:1647.
Kalingan, A. E., Liao, C. M., Chen, J. W., Chen, S. C. (2004) Microbial degradation of livestock-
generated ammonia using biofilters at typical ambient temperatures, J. Environ. Sci. Health Part
B-Pestic. Contam. Agric. Wastes 39:85.
Kasakura, T., Tatsukawa, K. (1995) On the scent of a good idea for odour removal, Water Quality
International 2:24.
Kennes, C., Veiga, M.C. (2001) Bioreactors for Waste Gas Treatment, Kluwer Academic Publishers,
Dordrecht.
Kiared, K., Wu, G., Beerli, M., Rothenbühler, M., Heitz, M. (1997) Application of biofiltration to
the control of VOC emissions, Environ. Technol. 18:55.
Kim, N. J., Hirai, M., Shoda, M. (1998) Comparison of organic and inorganic carriers in removal of
hydrogen sulfide in biofilters, Environ. Technol. 19:1233.
Kim, H., Xie, Q., Kim, Y. J., Chung, J. S. (2002) Biofiltration of ammonia gas with sponge cubes
coated with mixtures of activated carbon and zeolite, Environ. Technol. 23:839.
Kirchner, K., Hauk, G., Rehm, H. J. (1987) Exhaust gas purification using immobilized monocul-
tures (biocatalysts), Appl. Microbiol. Biotechnol. 26:579.
Kirchner, K., Gossen, C. A., Rehm, H. J. (1991) Purification of exhaust air containing organic pol-
lutants in a trickle-bed bioreactor, Appl. Microbiol. Biotechnol. 35:396.
156 B. Sercu et al.
Kosteltz, A. M., Finkelstein, A., Sears, G. (1996) What are the ‘real opportunities’ in biological gas
cleaning for North America?, in: Proceedings of the 89th Annual Meeting & Exhibition of Air &
Waste Management Association, A&WMA, Pittsburgh, PA; 96-RA87B.02.
Lacoste, F., Bosque, F., Brenne, E. (1996) Monitoring atmospheric emissions from industrial colza
trituration plants, Oleagineux Corps Gras Lipides 3:369.
Liang, Y. K., Quan, X., Chen, J. W., Chung, J. S., Sung, J. Y., Chen, S., Xue, D. M., Zhao, Y. Z.
(2000) Long-term results of ammonia removal and transformation by biofiltration, J. Hazard.
Mater. 80:259.
Luo, J. (2001) A pilot-scale study on biofilters for controlling animal rendering process odours, Wat.
Sci. Technol. 44:277.
Mohseni, M., Allen, D. G. (1999) Transient performance of biofilters treating mixtures of hydro-
philic and hydrophobic volatile organic compounds, J. Air Waste Manage. Assoc. 49:1434.
Mohseni, M., Allen, D. G. (2000) Biofiltration of mixtures of hydrophilic and hydrophobic volatile
organic compounds”, Chem. Eng. Sci. 55:1545.
Oberthür, R., Vossen, F. (2001) Intrinsic odour emission from rendering and the odour history of a ren-
dering plant, in: Proceedings of the 1st IWA International Conference on Odour and VOCs: Mea-
surement, Regulation and Control Techniques, The University of NSW, Sydney, Australia, 25 – 29th
March, pp. 131–138.
Otten, L., Afzal, M. T., Mainville, D. M. (2004) Biofiltration of odours: laboratory studies using
butyric acid, Adv. Environ. Res. 8:397.
Park, S.-J., Hirai, M., Shoda, M. (1993) Treatment of exhaust gases from a night soil treatment plant
by a combined deodorization system of activated carbon fabric reactor and peat biofilter inocu-
lated with Thiobacillus thioparus DW44, J. Ferment. Bioeng. 76:423.
Park, S.-J., Nam, S.-I., Choi, E.-S. (2001) Removal of odor emitted from composting facilities using
a porous ceramic biofilter, Wat. Sci. Technol. 44:301.
Park, D. H., Cha, J. M., Ryu, H. W., Lee, G. W., Yu, E. Y., Rhee, J. I., Park, J. J., Kim, S. W., Lee, I.
W., Joe, Y. I., Ryu, Y. W., Hur, B. K., Park, J. K., Park, K. (2002) Hydrogen sulfide removal utiliz-
ing immobilized Thiobacillus sp IW with Ca-alginate bead, Biochem. Eng. J. 11:167.
Peixoto, J., Mota, M. (1997) Volatile organic compounds evaporation chamber for the simulation of
gas effluents in laboratory research, Biotechnology Techniques 11:1.
Peixoto, J., Mota, M. (1998) Biodegradation of toluene in a trickling filter, Bioproc. Eng. 19:393.
Perry, R. H., Green, D. (1984) Perry’s Chemical Engineers’ Handbook, 6th ed., McGraw-Hill Interna-
tional Editions. Chemical Engineering Series, New York, 2234 p.
Rafson, H. J. (1998) Odor and VOC Control Handbook . McGraw-Hill, New York.
Regan, J. M., Harrington, G. W., Noguera, D. R. (2002) Ammonia- and nitrite-oxidizing bacterial
communities in a pilot-scale chloraminated drinking water distribution system, Appl. Environ.
Microbiol. 68:73.
Reij, M. W., Keurentjes, J. T. F., Hartmans, S. (1998) Membrane bioreactors for waste gas treatment,
J. Biotechnol. 59:155.
Ritchie, B. J., Hill, G. A. (1995) Biodegradation of phenol polluted air using an external loop airlift
bioreactor, J. Chem. Technol. Biotechnol. 62:339.
Rittmann, B. E., Schwarz, A. O., Sáez P. B. (2000) Biofilms applied to hazardous waste treatment, in:
Biofilms II, J. Bryers, ed., John Wiley & Sons, Inc., New York, pp. 207–234.
Ruokojärvi, A., Ruuskanen, J., Martikainen, P. J., Olkkonen, M. (2001) Oxidation of gas mixtures
containing dimethyl sulfide, hydrogen sulfide, and methanethiol using a two-stage biotrickling
filter, J. Air Waste Manage. Assoc. 51:11.
Schramm, A., De Beer, D., Gieseke, A., Amann, R. (2000) Microenvironments and distribution of
nitrifying bacteria in a membrane-bound biofilm, Environ. Microbiol. 2:680.
Sercu, B., Demeestere, K., Baillieul, H., Verstraete, W., Van Langenhove, H. (2005a), Degradation of
isobutanal at high loading rates in a compost biofilter, J. Air Waste Manage. Assoc. 55:1217.
Sercu, B., Núñez, D., Van Langenhove, H., Aroca, G., Verstraete, W. (2005b), Operational and
microbiological aspects of a two-stage biotrickling filter removing hydrogen sulfide and dimethyl
sulfide, Biotechnol. Bioeng. 90:259.
Odors Treatment: Biological Technologies 157
Sheridan, B. A., Curran, T. P., Dodd, V. A. (2003) Biofiltration of n-butyric acid for the control of
odour, Bioresource Technol. 89:199.
Shoda, M. (1991) Methods for the biological treatment of exhaust gases, in: Biological Degradation
of Wastes, A. M. Martin, ed., Elsevier Science Publishers, New York, pp. 31–46.
SIS, (2000) Scientific Instruments Services: adsorbent resins, general listing and description of seven
resins; http://www.sisweb.com/index/referenc/resin2.htm
Smet, E. (1995) Evaluatie werking biofilter – De Kleijn (Maaseik), Unpublished Report, Ghent University.
Smet, E., Chasaya, G., Van Langenhove, H., and Verstraete, W. (1996a) The effect of inoculation
and the type of carrier material used on the biofiltration of methyl sulphides, Appl. Microbiol.
Biotechnol. 45:293.
Smet, E., Lens, P., and Van Langenhove, H. (1998) Treatment of waste gases contaminated with
odorous sulphur compounds, Crit. Rev. Environ. Sci. Technol. 28:89.
Smet, E., Van Langenhove, H., and Verstraete, W. (1996b) Long-term stability of a biofilter treating
dimethyl sulphide, Appl. Microbiol. Biotechnol. 46:191.
Smet, E., Van Langenhove, H., Maes, K. (2000) Abatement of high concentrated ammonia loaded
waste gases in compost biofilters, Water Air Soil Poll. 119:177.
Smet, E., Van Langenhove, H., and Verstraete, W. (1997) Isobutyraldehyde as a competitor of the
dimethyl sulfide degrading activity in biofilters, Biodegradation 8:53.
Smits, M. C. J., Hoek, A. P., Osinga, B., Ottengraf, S. P. P., and Wijngaard, M. H. (1995) Removal of
ammonia with odour from composting facility waste gas with a bio-trickling filter, Meded. Fac.
Landbouww. Gent 60:2313.
Tang, H.-M., Hwang, S.-J., and Hwang, S.-C. (1996) Waste gas treatment in biofilters, J. Air Waste
Manage. Assoc. 46:349.
Terasawa, M., Hirai, M., and Kubota, H. (1986) Soil deodorization systems, Biocycle 27:28.
Van Elst, T., and Van Langenhove, H. (2001) Case study: odour emission at a vegetable oil plant in
France, in: Proceedings of the 1st IWA International Conference on Odour and VOCs: Measure-
ment, Regulation and Control Techniques, The University of NSW, Sydney, Australia, 25–29th
March, pp. 91–398.
Van Groenestijn, J. W., and Hesselink, P. G. M. (1993) Biotechniques for air pollution control,
Biodegradation 4:283.
Van Langenhove, H., Bendinger, B., Oberthür, R., and Schamp, W. (1992) Organic sulfur com-
pounds: persistant odourants in the biological treatment of complex waste gases, in: Biotech-
niques for Air Pollution Abatement and Odour Control Policies, A. J. Dragt, and J. van Dam, Eds.,
Elsevier, Amsterdam, pp. 309–313.
Van Langenhove, H., Cornelis, C., Schamp, N., and Wuyts, E. (1989a) Biofiltration in a vegetables
processing industry, Meded. Fac. Landbouww. Gent 54:1471.
Van Langenhove, H., Lootens, A., and Schamp, N. (1988) Elimination of ammonia from pigsty
ventilation air by wood bark biofiltration, Meded. Fac. Landbouww. Gent 53:1963.
Van Langenhove, H., Lootens, A., and Schamp, N. (1989b) Inhibitory effects of SO2 on biofiltration
of aldehydes, Water Air Soil Poll. 47:81.
Van Lith, C., Leson, G., and Michelsen, R. (1997) Evaluating design options for biofilters, J. Air
Waste Manage. Assoc. 47:37.
VDI (1996) Emissionsminderung, Anlagen zur Verwertung und Beseitigung von Tierkörperteilen und
tierischen Erzeugnissen, Düsseldorf, Verein Deutscher Ingenieure
Wada, A., Shoda, M., Kubota, H., Kobayashi, T., Katayama-Fujimura, J., and Kuraishi, H. (1986)
Characteristics of H2S oxidizing bacteria inhabiting a peat biofilter, J. Ferment. Technol. 64: 161.
Wani, A. H., Lau, A. K., and Branion, R. M. R. (1999) Biofiltration control of pulping odors –
hydrogen sulfide: performance, macrokinetics and coexistence effects of organo-sulfur species,
J. Chem. Technol. Biotechnol. 74:9.
Warren, J., Loerh, C., and Fellow A. S. C. E. (1997) Biofiltration: fundamentals, design and opera-
tion principles and applications, J. Environ. Eng. 123:538.
Waweru, M., Herrygers, V., Van Langenhove, W., and Verstraete, W. (2000) Process engineering of
biological waste gas purification, in: Biotechnology, 2nd edition, Volume 11c. Environmental Pro-
158 B. Sercu et al.
cesses III, Solid Waste and Waste Gas Treatment, Preparation of Drinking Water, J. Klein, and
J. Winter, eds., Wiley-VCH, Weinheim, pp. 260–273.
Weast, R. C., Astle, M. J., and Beyer, W. H. (1984) CRC. Handbook of Chemistry and Physics, 64th
ed., CRC Press, Inc., Boca Raton, Florida, 2303 p.
Weckhuysen, B., Vriens, L., and Verachtert, H. (1993) The effect of nutrient supplementation on the
biofiltration removal of butanal in contaminated air, Appl. Microbiol. Biotechnol. 39:395.
Weckhuysen, B., Vriens, L., and Verachtert, H. (1994) Biotreatment of ammonia-containing and
butanal-containing waste gases, Appl. Microbiol. Biotechnol. 42:147.
Williams, T. O., 1995, Odors and VOC emissions control methods, Biocycle 36:49.
Wu, L., Loo, Y.-Y., and Koe, L. C. C. (2001) A pilot study of a biotricking filter for the treatment of
odorous sewage air, Wat. Sci. Technol. 44:295.
Yang, Y., and Allen, E. R. (1994a) Biofiltration control of hydrogen sulfide 1. Design and opera-
tional parameters, J. Air Waste Manage. Assoc. 44:863.
Yang, Y., and Allen, E. R. (1994b) Biofiltration control of hydrogen sulfide 2. Kinetics, biofilter
performance, and maintenance, J. Air Waste Manage. Assoc. 44:1315.
Yang, Y., Togna, A. P., and Skladany, G. (1994) Treatment of CS2 and H2S vapors by biofiltration, in:
Proceedings of the 49th Purdue Industrial Waste Conference, Purdue University, pp. 449–456.
Yani, M., Hiral, M., and Shoda, M. (1998) Ammonia gas removal characteristics using biofilter with
activated carbon fiber as a carrier, Environ. Technol. 19:709.
Zhang, L., Hirai, M., and Shoda, M. (1991) Removal characteristics of dimethyl sulfide, methane-
thiol and hydrogen sulfide by Hyphomicrobium sp. I55 isolated from peat biofilter, J. Ferment.
Bioeng. 72:392.
Index
Absorption, 30, 41, 43, 81, 88, 89t, 108t, Biofiltration, 108t, 127, 137–142
109, 110t, 111f, 117, 127, 130–131, Biomass, 80t, 84, 130–131, 133f, 134–135,
138–139, 148–150 139–140
Activated carbon, 34, 42, 111–113, 114t, Bioscrubber, 130–131, 139
127, 138, 142 Biotrickling filter, 129–130
Adaptation, 2, 33 Biowasher, 127, 130–132
Adsorbent, 42–43, 112–113, 114f, Biological plate tower (BPT), 127, 131,
146–147 133–135
Adsorbing, 41, 113 Brewery, 92, 116
Adsorption, 28, 32–33, 42–43, 81, 89, 108, Buffer areas, 96–100
111–113, 114f, 146–147 Building Sources, 20–21
Advertisements, 84 Bulk acoustic wave, 50–51
Air pollution, 47–54, 102, 106, 109, 112, 132t
Alcohols, 11, 58, 59t, 60, 64–66, 70, 80, Cane sugar processing, 9t, 11
112, 126, 144–145, 150t, 152t Capelin, 62, 63t, 66
Aldehydes, 11, 23t, 59t, 60, 64–66, 80t, Capture-recovery, 106
81, 107, 108t, 112, 121, 126, 144–146, Case study, 64–70
148, 150t, 151, 152t Catalytic metal, 50
Amines, 6, 10–11, 23t, 59t, 62, 64–66, 68, Catalytic oxidation, 119–121
70, 107, 108t, 109, 112, 126, 140 Catalytic oxidizer, 119–121
Ammonia, 6–8, 10, 18, 23t, 25, 29, 62, Catalytic combustion, 152
65–66, 79–80, 97, 109, 127, 137–140, Chemical analysis, 24, 68
142–143 Chemical compounds, 49, 88, 152
Ammonia-like stockfish odor, 64f Chemical gas sensors, 49, 64–65
Animal feeding, 7, 9–10 Cleaner production, 84, 92f, 93–95, 97, 106
Animal rendering, 10, 151–153 Clogging, 42, 128–130, 132–133, 136–137
Area sources, 18, 20 Cluster analysis, 51–52
Artificial neural network, 51–54 Cod, 59t, 62, 63t, 66
Cod roe, 63t
Biodegradation, 54, 80t, 95, 125, 145 Colorimetric Detector Tubes, 27–28
Biofilm, 126f, 136f–137f Condensation, 18, 88–89, 108t, 115–116,
Biofilter, 128f, 129–130, 137f 147, 151
159
160 Index
Volatile organic compound (VOC), Volatiles, 27, 50, 54, 60, 62, 65, 146, 152
7–8, 10–12, 29, 43–44, 87, 107, 108t,
116–117, 119–121, 125, 135 Waste gas, 125–127, 131, 136–141, 143,
Volatile organic sulphur compound 145, 147–148, 150, 152
(VOSC), 143–144, 153 Wind tunnel system, 18–20