WATER POLLUTION CONTROL RESEARCH SERIES
16040 DGH 08/71
CHARACTERIZATION OF
TASTE AND ODORS IN WATER SUPPLIES
U.S. ENVIRONMENTAL PROTECTION AGENCY
-------
WATER POLLUTION COUTROL HESSARCH SERIES
The Water Pollution Control Research Series describes the
results and 'progress in "he control and abatement of
pollution in our Nation's water's. Xhey provide a central
source of information on the research, development, and
demonstration activities in the water* research program
of the Environmental Protection Agency* through inhouse
research and grants and contracts with Federal, State,
and locsl agencies, research institutions, and industrial
organisations.
Inquiries pertaining to Water Pollution Control Research
Kepoi^ts should be directed to the Chief, Publications
Branch (Water), Research Information Division, RSM,
Environmental Protection Agency, Washington, D.C. 20M60.
-------
CHARACTERIZATION OF TASTE AND ODORS IN WATER SUPPLIES
by
Ralph P. Collins
The Biological Sciences Group
Regulatory Biology Section
University of Connecticut
Storrs, Connecticut 06268
for the
Office of Research and Monitoring
ENVIRONMENTAL PROTECTION AGENCY
Project //16040 DGH
August 1971
For sale by the Superintendent ol Documents, U.S. Government Printing Office, Washington, D.C., 2W02 - Price 35 cents
-------
EPA Review Notice
This report has been reviewed by the Environmental Protection
Agency and approved for publication. Approval does not
signify that the contents necessarily reflect the views and
policies of the Environmental Protection Agency nor does
mention of trade names or commercial products constitute
endorsement or recommendation for use.
ii
-------
ABSTRACT
An examination of the odorous constituents produced by microorganisms
which are involved in taste and odor problems in public water supplies
was conducted. The organisms used in the study were the actinoraycete
Streptomyces odorifer, the alga Synura petersenii, and the fungus
Trichoderma viride.
The odorous constituents produced by these organisms were obtained by
steam distillation of the culture medium containing the cells. The
aqueous distillates were extracted with an organic solvent and these
extracts were then reduced to a small volume before analysis.
Analysis of the extracts was made by gas chrotnatography, infrared, mass
and nuclear magnetic spectroscopy. The major odorous constituents pro-
duced by Streptomyces odorifer have been identified and the compounds
are trans-1, 10-dimethyl-trans-9-decalol (geosmin), 2-exo-hydroxy-2-
methylbornane and cadin-4-ene-l-ol. A large number of odorous consti-
tuents were also identified from the culture filtrates Synura
petersenii and heptanal was the one constituent having an odor similar
to that of the organism. The major odorous compound produced by
Trichoderma yiride was 6-pentyl- »C-pyrone.
This report was submitted in fulfillment of project number 16040 DGH.
iii
-------
CONTENTS
Section
I Conclusions
II Recommendations
III Introduction
IV Materials and Methods
V Results
VI Discussion
VII Acknowledgments
VIII References
IX Publications
X Glossary
1
3
5
7
11
13
15
17
19
21
-------
FIGURES
Figure Title Page
1 Structural Formulas of Earthy-Musty Metabolites 14
vi
-------
SECTION I
CONCLUSIONS
1. Satisfactory analytical procedures for the analysis of small
samples of taste and odor constituents have been developed.
2. Large scale cultural and extraction procedures have been developed
for taste and odor microorganisms.
3. The major earthy-musty odor constituents produced by Streptomyces
odorifer have now been identified.
A. The predominant odor constituents produced by Synura petersenii
have been identified.
5. The major odor constituent of Trichoderma viride was identified.
-------
SECTION II
RECOMMENDATIONS
It seems quite apparent that the technology for the identification of
taste and odor constituents is now present in most of the laboratories
engaged in this activity around the country. The techniques developed
have been particularly successful in the elucidation of the earthy-
musty odor complex.
There has not been as much success in identifying the odor constituents
produced by algae. This is not because the analytical procedures are
not satisfactory but rather is due to a lack of information on the
isolation and mass culture of these organisms. It is recommended that
nutritional studies of taste and odor algae be conducted so that these
organisms can be studied by the interested analyst.
While the identification of taste and odor constituents has rapidly
progressed, it is also true that our knowledge of those factors which
lead to the accumulation of these constituents in public water supplies
is sadly lacking. It is recommended that the microbiologists and
chemists cooperate in a study of those conditions which lead to taste
and odors in a natural environment.
The final aim of taste and odor studies is more than an identification
of the compounds responsible for the taste and odors and should lead to
suggestions for their control. It seems clear that this has not hap-
pened and it is recommended that studies which would lead to the control
of taste and odor microorganisms be initiated.
One of the difficulties in the taste and odor field is the subjective
terms used for the description of tastes and odors. It is recommended
that a taste and odor kit be assembled and distributed to water works
operators so that communication between the scientist and the person
involved with a practical taste and odor problem can be enhanced.
-------
SECTION III
INTRODUCTION
The purpose of the present investigation was to isolate and identify
those compounds, produced by selected microorganisms, which give rise
to taste and odor.
The problem of taste and odor in water supplies is an old one and it is
generally conceded that most tastes and odors are microbial in origin.
Only in recent years has there been real progress in the chemical iden-
tification of taste and odor constituents produced by microorganisms.
Most of the effort and progress has centered around the identification
of those compounds which give rise to the earthy-musty odor complex.
Several laboratories have contributed to a solution of this problem
and most of their findings have now been published (1), (2), (3), (4),
and Morris 1966 and Collins, Knaak and Soboslai 1970.
-------
SECTION IV
MATERIALS AND METHODS
This section will be divided into three parts: the first part dealing
with Streptomyces odor if er, the second with Synura petersenii, and the
third with Trichoderma viride.
J3. odorifer, an actinomycete, was originally isolated from the Thames
River in London and it was felt that this organism was responsible for
the earthy odor of salmon that were caught from thatr river.
The organism was grown in potato dextrose (10%) broth containing
0.2 g/liter either fermentation vessels containing 12 liters or large
flasks (2500 ml) containing 700 ml of the medium. After inoculation
the fermentation vessels were placed in a New Brunswick fermentor and
incubated at 25°C for 48 hr. The large flasks were incubated, after
inoculation, at 25°C for 48 hr on a reciprocal shaker.
At the end of the growth period the cells and culture medium were steam
distilled. The steam distillate was freeze-concentrated following the
method of Whitaker (6) . The concentrated material was extracted for
72 hr with petroleum ether (lower boiling fraction) in a liquid-liquid
extractor. The petroleum ether solution was dried, carefully concen-
trated to a final volume of approximately 3 ml and the extract was then
analyzed by gas chromatography (glc) .
The gas chromatograph used was a Hewlett-Packard Model 5750B and was
equipped with a dual flame ionization detector. The column (6 ft stain-
less steel) used 80-100 mesh, acid washed silanized Chromosorb W as the
solid support and this was coated with 10% SE-30. Other operating con-
ditions were: injection port 200°C, collector 200°C; temperature
programmed from 100-285°C at 6°C/min; helium flow rate, 30 ml/min;
hydrogen flow rate, 40 ml/min; air flow rate 500 ml/min; inlet pressure
60 psi; outlet pressure, atmospheric. The instrument was equipped with
an effluent splitter (approximate ratio 1:10) to permit collection of
compounds.
In order to collect sufficient material for infrared and mass spectro-
scopy the compounds of interest were trapped by means of glass U tubes
as they exited from the gas chromatograph. The glass tubes were 12 in.
x 1/16 in. i.d. and were connected to the exit port by means of a
Swagelok fitting. To insure a leak-proof system, one end of the U tube
was passed through a rubber septum inserted into the fitting. The tubes
were cooled in a bath of dry ice and acetone.
Infrared spectra were obtained on a Perkin-Elmer 137 Infracord (KBr
disks) and mass spectra were obtained on an AEI M.S. 12 mass spectrome-
ter, using the gas inlet system, ion source operated at 250°C.
-------
Previous to the present study a detailed examination of the odorous
constituents produced by Synura petersenii has been conducted and a
series of compounds had been identified. In the present work the
volatile compounds were reexamined employing the newer techniques then
available.
Synura petersenii is a member of the Chrysophyta and is found in the
plankton of many lakes and pools. Synura sp are known to produce an
odor in water, when present in small numbers, resembling that of a ripe
cucumber or watermelon. In larger numbers the organism produces a fishy
odor.
The organisms were grown in the defined medium of Provasoli and Pintner
(7) and was maintained at 16°C in a lighted refrigerator (400 ft-c).
In order to obtain enough material for analytical work the organism was
grown in large (25 liter) carboys containing 15 liters of medium. The
contents of three large test tubes containing 300 ml of the medium served
as the inoculum for each large carboy. The carboys were incubated
without agitation in a constant temperature room maintained at 15 + 1°C.
The fluorescent light present in.the room provided sufficient illumina-
tion for good growth (200 ft-c). The incubation period varied between
ten and fourteen days.
The volatile constituents were obtained by steam distillation of the
culture medium followed by liquid-liquid extraction of the aqueous
distillate with ether. The ether fraction was dried, carefully concen-
trated to a small volume and analyzed by glc. The analytical procedures
were essentially the same as those described for S^. odorifer except that
a Carbowax 20M column was used.
Trichoderma viride is a fungus that was isolated from a soil sample
collected near Coventry Lake in Coventry, Connecticut. The organism
was grown in a potato dextrose (10%) liquid medium in quart milk bottles
containing 60 ml of the medium. After inoculation the milk bottles were
placed on their side and incubated at room temperature for three to four
days. At the end of this period the surface of the medium was covered
with the dark green conidia characteristic of this organism.
At the end of the growth period the cells and culture medium were steam
distilled at atmospheric pressure. The aqueous distillate was dried and
extracted with ether. The ether extract was concentrated under reduced
pressure leaving 3.3 g of a yellow oil having a strong coconut-like
aroma. The oil was separated into its components by glc (Hewlett-
Packard Model 7620A) utilizing a flame ionization detector equipped with
an effluent splitter. The operating conditions were the same as those
described for Streptomyces odorifer except that the column used was
Carbowax 20M (19%) plus PDEAS (1%). For further purification of the
compound a column packed with W98 (19%) was used.
-------
Infrared and mass spectra were obtained by the same procedures outlined
for S. odorifer. The (nmr) spectra were obtained on a Varian A-60
nuclear magnetic resonance spectrometer.
-------
SECTION V
EXPERIMENTAL RESULTS
As in the Materials and Methods section the results for each of the
organisms studied will be discussed separately.
In order to determine which compounds gave rise to the earthy or musty
odor in Streptomyces odorifer the odor of each compound was noted as it
exited from the gas chromatograph. Three compounds had a definite
earthy or musty odor. The compound having the shortest retention time
on glc had a definite musty odor and its ir was identical to that of the
published spectrum of 2-exo-hydroxy-2-methylbornane (2), This compound
was the major constituent in the mixture and was identified as
2-exo-hydroxy-2-methylbornane. The second compound had a definite
earthy odor and its ir spectrum corresponded with the published spectrum
of geosmin (1).
The mass spectrum for the suspected 2-exo-hydroxy-2-methylbornane gave
the empirical formula CnH^O with the parent peak at m/e 168 and the
base peak at m/e 95. The fragmentation agreed with that published by
Medsker et al (2). The mass spectrum for the suspected geosmin was
compared with the spectrum published for this compound (1) and the
agreement was excellent. The mass spectrum showed a parent peak at 182
and a base peak at 112.
The third compound was present in very small amounts and it was diffi-
cult to obtain amounts large enough for good analytical work, but this
has been accomplished and the compound identified as cadin-4-ene-l-ol.
The mass spectrum of the compound indicated a C^l^gO compound and it
contained a small parent peak at 222 as well as typical sesquiterpene
alcohol peaks. The ir spectrum was also consistent with the structure
suggested above.
A recent paper by Gerber (8) had described the formation of cadin-4-ene-
l-ol by another species of Streptomyces. An ir spectrum of the compound
isolated from S. odorifer was sent to Dr. Gerber for comparison and the
two spectra were identical. A small amount of the authentic compound
was also obtained and its retention time (glc) was identical with the
compound produced by Streptomyces odorifer. It is concluded, therefore,
that the third odorous constituent produced by J3. odorifer is cadin-4-
ene-l-ol.
As has been previously discussed, the Synura petersenii work was essen-
tially a repetition of earlier work except that more modern instrumenta-
tion was available for analytical purposes. All of the compounds
previously identified by classical procedures were confirmed in the
present study. All of the identified compounds were presented to a
taste panel comprised of laboratory personnel to see which compound or
compounds comprosed a typical Synura-like odor. The only compound which
11
-------
some of the taste panel agreed on as having an odor somewhat like that
of Synura petersenii was n-heptanal. In the author's opinion, while
this compound undoubtedly is part of the Synura odor, there are other
compounds present and as yet unidentified which are responsible for the
typical odor produced by this organism.
The major component of the oil (over 90%) in Trichoderma viride had a
retention time of approximately 24 min. The isolated compound was a
colorless liquid having a strong coconut-like odor.
The infrared spectrum showed two C=0 absorption band's at 1740 and 1725
Cm~l and two strong C=C stretching bands at 1636 and 1553 Cm"-*- suggest-
ing an o^-pyrone ring.
The mass spectrum of the compound showed a parent peak at m/e 166. The
molecular formula was assigned as C^QH-^^02 on the basis of the mass
spectral data and elemental analysis (Found C, 72.84%; H, 8.39%; 0,
18.77%; Calculated, C, 72.9%, H, 8.5%; 0, 19.21%). The mass spectrum
also revealed a base ion at m/e 95 attributable to M-C5~Hn.
The (nmr) showed signals at 00.94 (3H, triplet), 1.4 (6H, complex
multiplet) and 2.48 (2H, triplet) all consistent with an n-pentvl group.
The three °< -pyrone ring protons formed an ABX pattern with $A 5.83,
£B 6.00 and $X 7.19 and JAX = 6.5 and JBX = 9.5 HZ. The ABX
protons represent those attached to positions 5, 3, and 4, respectively.
As an additional proof the compound was hydrogenated over Adams
platinum oxide in ethanol. When the reduction product was subjected to
glc (19% W98 column) two peak areas were revealed with an approximate
peak area ratio of 1:5. The minor peak was identified as n-decanoic
acid (infrared, retention time) and the major peak was identified as
acid
fi -
decalactone (infrared, retention time, odor).
12
-------
SECTION VI
DISCUSSION
It is quite clear that all of the major odorous constituents produced by
Streptomyces odorifer have now been identified. There are, however, two
additional compounds having longer retention times (glc) than the three
discussed above. Both of these compounds have faint but discernable
"woody" odors. A preliminary analysis of these compounds (ir) indicates
that both are sesquiterpene alcohols. An investigation of these
compounds is now being conducted.
The reinvestigation of Synura petersenii failed to unearth any new
compounds and n-heptanal is regarded as the compound primarily responsi-
ble for the characteristic odor produced by this organism.
The major odor compound of Trichoderma viride was identified as 6-pentyl-
°* -pyrone. This is a new fungal metabolic product and the only other
isolation of this compound from a natural source has been from peach
fruits. The compound has interest not only as a new fungal metabolite
but also as a potentially useful new compound in the flavor industry.
A series of techniques have been developed, as described above, which
permit the isolation and characterization of the small amounts of
material normally found in research involving taste and odor. This
does not mean that other techniques are not available which would permit
a more rapid identification, but it does mean that a laboratory with
only moderate facilities can carry out this sort of analysis.
13
-------
2-EXO- HYDROXY-2-METHYLBORNANE
CH-
HO
CH-
TRANS -
1,10-DtMETHYL-TRANS-
9-DECALOL
(GEOSMIN)
'3 CH3
CADIN-4-ENE-I-OL
c*-PENTYL PYRONE
FIGURE I. STRUCTURAL FORMULAS OF EARTHY-MUSTY METABOLITES
14
-------
SECTION VII
ACKNOWLEDGMENTS
The support of A. F. Halim, technician, and R. F. Fairweather of the
Chemistry Department are acknowledged with thanks.
The support and encouragement of Dr. Ronald Webb, the Grant Project
Officer, was also invaluable during the course of this investigation.
15
-------
SECTION VIII
REFERENCES
1. Gerber, N. N., "Geosmin From Microorganisms is Trans-1, 10-Dimethyl-
Trans-9-Decalol," Tetrahedron Letters, No. 25, pp 2971-2974 (1968).
2. Medsker, L. L., Jenkins, D., Thomas, J. F., and Koch, C., "Odorous
Compounds in Natural Waters: 2-Exo-Hydroxy-2-Methylbornane, The
Major Odorous Compound Produced by Several Actinomycetes,"
Environmental Science and Technology, 3, No. 5, pp 476-477 (1969).
3. Rosen, A. A., Safferman, R. S., Mashni, C. I., and Romano, A. H.,
"Identity of Odorous Substance Produced by Strept omy c e s
griseoluteus." Applied Microbiology. 16, No. 1, pp 178-179 (1968).
4. Dougherty, J. D., Campbell, R. D., and Morris, R. L.,
"Actinomycete: Isolation and Identification of Agent Responsible
for Musty Odor," Science. 152, No. 3727, pp 1372-1373 (1966).
5. Collins, R. P., Knaak, L. E., and Soboslai, J. W., "Production of
Geosmin and 2-Exo-Hydroxy-2-Methylbornane by Streptomyces odorifer,"
Lloydia, 33, No. 1, pp 199-200 (1970).
6. Whitaker, D. R., "Purification of Myrothecium Cellulase," Archives
of Biochemistry and Biophysics, 43, No. 1, pp 253-268 (1953).
7. Provasoli, L., and Pintner, I. J., "Ecological Implications of In
Vitro Nutritional Requirements of Algae Flagellates," Annals of the
New York Academy of Science, 56, No. 5, pp 839-851 (1953).
8. Gerber, N. N., "Sesquiterpenoids from Actinomycetes: Cadin-4-Ene-
1-01," Phytochemistry, 10, No. 1, pp 185-189 (1971).
17
-------
SECTION IX
PUBLICATIONS
1. Collins, R. P., Knaak, L. E., and Soboslai, J. W., "Production of
Geosmin and 2-Exo-Hydroxy-2-Methylbornane by Streptomyces odorifer,"
Lloydia. 33, No. 1, pp 199-200 (1970).
2. Collins, R. P., and Halim, A. F., "Characterization of the Major
Aroma Constituent of the Fungus Trichoderma viride (Pers.),"
Journal of Agriculture and Food Chemistry (manuscript in press).
3. Collins, R. P., and Halim, A. F., "The Production of Cadin-4-Ene-
1-01 by Streptomyces odorifer," Lloydia (manuscript in press).
19
-------
SECTION X
GLOSSARY
Actinomycete - A filamentous bacterium. Streptomyces odorifer is an
example.
Chrysophyta - A group of algae characterized by the formation of a
yellow brown pigment. Synura petersenii is an example.
Fungus - A group of microorganisms lacking chlorophyll and generally
filamentous. Trichoderma viride is an example.
ir - Abbreviation for infrared spectroscopy.
glc_ - Abbreviation for gas-liquid chromatography.
nmr - Abbreviation for nuclear magnetic spectroscopy.
PDEAS - Abbreviation for phenyldiethanolamine.
21
-------
1
Accession Number
5
2
.Sni))erf Fn-ld & Group
016C
SELECTED WATER RESOURCES
ABSTRACTS
INPUT TRANSACTION FORM
Organization
THo "Ri r\~\ r\rri /~< a 1 C«~«-i £a-r» y*Q a ^vrMi i^ "P^^^-m 1a4-<^T"\r R-T*^1r-»rTt7 C^s/-t-*-Tr\*-»
University of Connecticut, Storrs, Connecticut
Title
CHARACTERIZATION OF TASTE AND ODORS IN WATER SUPPLIES
1 Q Authors)
Collins ,
Ralph P.
16
21
Project Designation
EPA WQO Project No.
16040 DGH
Note
22
Citation
23
Descriptors (Starred First)
*0dor-producing algae, *Actinomycetes
*Analytical techniques, molds, tastes and odors
25
Identifiers (Starred First)
*Microbially induced odors
27
Abstract
The taste and odor constituents produced by the actinomycete
Streptomyces odorifer, the alga Synura petersenii and the mold Trichoderma
viride were examined.
The odorous constituents were obtained by steam distillation of the
culture medium. The odorous constituents were identified by means of
gas-chromatography, infrared, mass and nuclear magnetic spectroscopy.
The major odorous constituents produced by Streptomyces odorifer have
been identified as trans-1, 10-dimethyl-trans-9-decalol (geosmin),
2-exo-hydroxy-2-methylbornane and cadin-4-ene-l-ol. A large number of
odorous materials were also identified from the culture filtrates Synura
petersenii and heptanal was the one compound having an odor similar to
that of the organism. The major odorous compound produced by Trichoderma
viride was 6-pentyl-°<-pyrone. (Collins--U. Conn.)
Abstractor
Ralph P. Collins
Institution
University of Connecticut
WR;I02 I R E V . JULY 1969)
WRSIC
WATER RESOURCES SCIENTIFIC INFORMATION CENTER
U.S. DEPARTMENT OF THE INTERIOR
WASHINGTON. O. C- 20240
* CPO: !969-359'33«
-------
ENVIRONMENTAL PROTECTION AGENCY
Publications Distribution Section
Route 8, Box 116, Hwy. 70, West
Raleigh, North Carolina 27607
?E AND FEES PAID
ENVIRONMENTAL PROTECTION AGENCY
Official Business
If your address is incorrect, please change on the above label;
tear off; and return to the above address.
If you do not. desire to continue receiving this technical report
series, CHECK HERE D j tear off label, and return it to the
above address.
------- |