PB91-154617
PESTICIDE ASSESSMENT GUIDELINES • SUBDIVISION F
HAZARD EVALUATION: HUMAN AND DOMESTIC ANIMALS
ADDENDUM 10 - NEUROTOXICITY SERIES 81, 82, AND 83
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, DC
MARCH 1991
U.S. DEPARTMENT OF COMMERCE
National Technical Information Service
-------
PB91-154617
March 1991
EPA 540/09-91-123
PB 91-154617
PESTICIDE ASSESSMENT GUIDELINES
SUBDIVISION F
HAZARD EVALUATION:
HUMAN AND DOMESTIC ANIMALS
ADDENDUM 10
NEUROTOXICITY
SERIES 81, 82, AND 83
Prepared by:
William F. Sette, Ph.D.
Health Effects Division
Office of Pesticide Programs
U.S. Environmental Protection Agency
-------
FOREWORD
These new and revised neurotoxicity guidelines are intended
to replace and supplement the set of neurotoxicity guidelines
originally published in the 1982 Pesticide Assessment Guidelines,
Subdivision F, Hazard Evaluation: Human and Domestic Animals;
EPA-540/9-82-05; October 1982; National Technical Information
Service, Springfield, VA 22161. These guidelines have been
written in coordination with upcoming proposed revisions to the
Toxicology Data Requirements of Part 158 of Title 40 of the Code
of Federal Regulations (40 CFR 158). These guidelines have
undergone extensive review within the Agency, public comment, and
review by the FIFRA Scientific Advisory Panel.
ACKNOWLEDGMENTS
The development and revision of these guidelines was due to
the dedicated efforts of many people inside and outside EPA, but
predominantly the members of 2 workgroups. Their efforts are here
noted and gratefully acknowledged.
The following people participated in the revision of these
guidelines as members of the ad hoc Workgroup :Kevin Crofton,
Karl Jensen, Tina Levine, Robert MacPhail, Suzanne McMaster,
Virginia Moser, Stephanie Padilla, D. Cooper Rees, Lawrence
Reiter, Hugh Tilson, and William Sette, chairman.
The Developmental Neurotoxicity Guideline Workgroup
included: Zoltan Annau, Angela Auletta, Marlissa Campbell, Neil
Chernoff, Kevin Crofton, Lynda Erinoff, Earl Grey, Karl Jensen,
Carole Kimmel, Tina Levine, Robert MacPhail, Suzanne McMaster,
C.J. Nelson, Myron Ottley, D. Cooper Rees, Larry Reiter, Jennifer
Seed, William Sette, Mark Stanton, Hugh Tilson, Hal Zenick, and
Elaine Francis as chairwoman.
-------
Subdivision F
Neurotoxicitv
Table of Contents
Foreword, Acknowledgments 1
Table of Contents 2
Series 81-7, 82-6
Delayed Neurotoxicity of Organophosphorus Substances
Following Acute and 28 Day Exposures 3
Series 81-8, 82-7, 83-1
Neurotoxicity Screening Battery 13
Appendix 1: Guideline for Assaying Glial
Fibrillary Acidic Protein 28
Series 83-6
Developmental Neurotoxicity Study 32
Series 85
Schedule Controlled Operant Behavior 49
Peripheral Nerve Function.. 55
-------
DELAYED NEUROTOXICITY
OF ORGANOPHOSPHORUS SUBSTANCES
FOLLOWING ACUTE AND 28 DAY EXPOSURES
(a) tPurpose. In the assessment of organophosphorus
substances , studies of delayed neurotoxicity using the adult hen
as the test animal and including behavioral observation of gait,
histopathological assessment of brain, peripheral nerve, and spinal
cord, and neurochemical assessment of inhibition of
acetylcholinesterase(AchE) and neurotoxic esterase (NTE) are needed
to identify and characterize these potential effects.
This guideline now requires an acute dosing regimen in hens in
combination with assays of neurotoxic esterase (and
acetylcholinesterase) to screen for this effect. Use of data on
the inhibition of NTE in conjunction with, behavioral and
pathological data offers a number of advantages . It is important
to recognize that many acute studies can provide equivocal evidence
of behavioral or pathological effects. Some trixylenyl phosphates
(Johnson, 1975), for example, are negative, or at best equivocal,
after acute exposures, yet clearly cause OPIDN after repeated
exposures. The continuous, rather than descriptive, nature of NTE
data and the fact that considerable inhibition is generally
required to produce OPIDN, will help to more convincingly conclude
that a substance is negative, based solely on an acute study.
Conversely, NTE data can also provide a better indication of
potential delayed neurotoxicity, i.e. if the behavioral and
histopathological data after an acute exposure are equivocal, and
the NTE inhibition is significant, then further study is
appropriate.
The revision of the 90 day study to a 28 day study is based on
the idea that these shorter exposures offer savings in animals and
cost and because 28 days are often closer to the duration of
exposure applicators may experience than 90 days. In some cases,
further study may be required to resolve data that are difficult to
interpret clearly, or to establish more refined dose response
relations, or to assess the particular use patterns of the
substance.
(b) Definitions. (I) Organophosphorus induced delayed
neurotoxicity (OPIDN) is a neurological syndrome in which limb
weakness and upper motor neuron spasticity are the predominant
clinical signs; distal axonopathy of peripheral nerve and spinal
cord are the correlative pathological signs; and inhibtion and
aging of neurotoxic esterase in neural tissues are the correlative
biochemical effects. Clinical signs and pathology first appear
between 1 and 2 weeks following exposures that typically inhibit
and subsequently age neurotoxic esterase.
(2) Neuropathy target esterase (NTE) or neurotoxic
esterase is a membrane-bound protein that hydrolyzes phenyl
valerate. The inhibition and "aging" of the phosphorylated NTE,
i.e., the covalent binding of the OP to the enzyme, is highly
correlated with the initiation of OPIDN. Not all O-Ps that inhibit
-------
Delayed Neuro
NTE cause OPIDN, but all O-Ps that cause OPIDN inhibit NTE.
(3) NTE activity is operationally defined as the phenyl
valerate hydrolytic activity resistant to paraoxon but sensitive to
mipafox or neuropathic O-P ester inhibition.
(c) Principle of the test method. The test sequence consists
of acute and 28 day exposure studies. Any significant effects on
behavior(delayed effects), histopathology, or inhibition of NTE in
the acute study are sufficient cause to conduct the 28 day study.
The test substance is administered orally to domestic hens that in
some cases have been protected from acute cholinergic effects. The
animals are observed for at least 21 days for gait changes and
other signs. Neurochemical examination of selected neural tissues
is undertaken on some animals at some time(s) after exposure.
Histopathology of brain, spinal cord, and peripheral nerve are
performed at the termination of 21 day observation periods.
If the results of the acute study are completely negative, that is,
there are no delayed behavioral effects, and no histopathological
effects, and no significant NTE inhibition, then the 28 day study
is not required. Otherwise, the 28 day study should be conducted.
In the 28 day study, 3 exposure levels are used to describe the
dose response curve sufficiently to estimate a reference dose.
(d) Test Procedures. (1) Animal selection. The adult domestic
laying hen (Callus gallus domesticus), aged 8 to 14 months, is
recommended. Standard size breeds and strains should be employed.
Healthy young adult hens free from interfering viral diseases and
medication and without abnormalities of gait should be acclimatized
to the laboratory conditions for at least 5 days prior to
randomization and assignment to treatment and control groups.
(2) Housing and feeding conditions. Cages or enclosures
which are large enough to permit free mobility of the hens and easy
observation of gait should be used. Where the lighting is
artificial, the sequence should be 12 hours light, 12 hours dark.
Appropriate diets should be administered as well as an unlimited
supply of drinking water. The hens should be weighed weekly. Any
moribund hens should be removed and sacrificed.
(3) Route of Administration. Dosage of test substance
should normally be by the oral route, preferably by gavage. Liquids
may be given neat or dissolved in an appropriate vehicle such as
corn oil; solids should be dissolved if at all possible since large
doses of solids in gelatin capsules may significantly impair
absorption. Dermal exposures may be the most significant route of
exposure for applicators and for non-food uses and there may be
important differences in toxicity by this route. Conduct of these
studies by this route may be appropriate and should be considered.
(4) Study Design. (i) General. An important
consideration for the design of these studies is prediction of
activity based on the structure of the material and the published
literature. Some materials, e.g. phosphinates, are known to inhibit
-------
Delayed Neuro
NTE, but not to be capable of aging and, thus, are not expected to
cause OPIDN. Many materials have structural features that will
permit inhibition and aging, i.e. ester linkages, and are of
potential concern. Published data are available for many materials
and may be very useful for many aspects of the design and
interpretation of these studies.
(ii) Dose levels and selection. For the acute
study, a single exposure group is required. The acute dose level
should be chosen to maximize the amount of material given to the
hens, particularly in cases where some activity is expected. For
the 28 day study, at least 3 exposure groups are required in
addition to the vehicle control group. Ideally, the data should be
sufficient to produce a dose-effect curve. We strongly encourage
the use of equally spaced doses and a rationale for dose selection
that will maximally support detection of dose-effect relations.
The rationale for dose selection chosen by the investigator should
be explicitly stated. The following guidance for dose selection is
somewhat complex and is not intended to be rigidly followed.
(A) Acute Study. Selection of the dose level
for the acute study may be based on a limit dose, or lethal doses
and other available data, e.g.,on NTE inhibition.
(1) Levels of test substances greater than
2 g/kg need not be tested.
(2) Lethal Doses. Either an LD50 or an
approximate lethal dose (ALD) in the hen may be used to determine
the acute high dose. If a hen LD50 has been established, then this,
of course, may be used, although some verification may be prudent.
If the rat LD50 is known, it may serve as the starting point of
estimation. A preliminary lethality study in unprotected hens may
be conducted to estimate the acute high dose. A variety of test
methodologies may be used to estimate the unprotected lethal dose
of test materials. Of course, the method of estimate of the lethal
dose may influence the subsequent dosage -estimates.
From the preliminary data, if cholinergic signs were seen very soon
after dosing, prophylaxis using atropine may be appropriate.
Atropine (20 mg/kg, s.c., up to every 2 hours) should be used to
prevent death from acute cholinergic effects.
(B) 28 Dav Study. (1) Levels of test substances
greater than 1 g/kg need not be tested.
(2) High dose. The high dose selected
should be estimated to be sufficient to cause OPIDN or be a maximum
tolerated dose based on the acute data, but not result in an
incidence of fatalities that would prevent a meaningful evaluation
of the data.
(3) Low dose. The low dose should be
estimated to be a minimum effect level, e.g., an ED10, or
alternatively, a no effect level.
(4) The intermediate dose level should be
equally spaced between the high and low doses.
(5) Intermediate responses in NTE i.e.,
-------
Delayed Neuro
greater than 15% and less than 70%, can be crudely extrapolated as
if the dose-response were a simple first order relationship. That
is, if a certain dose caused 50% inhibition, then twice that dose
might cause 75%. Such extrapolation is very crude but can be useful
in giving some guidance for dose estimation.
(iii) Numbers of animals. Exposure groups should be
large enough to provide six survivors for both behavioral
observations and histopathology- At least 3 hens are required for
determination of NTE in each dose or control group and at each time
point.
(iv) Control Groups. A positive control group of at
least six hens treated with a known delayed neurotoxicant, such as
Triorthocresyl phosphate (TOCP), is required for both acute and 28
day studies. This group may be a concurrent or historical control
group.(This should also include at least 3 hens assessed for
biochemical measurements.) Periodic re-determinations of the
sensitivity of the assays is suggested, for historical control
data, i.e., when some essential element of the test conduct by the
performing laboratory has changed. A concurrent control group
sufficient to provide 6 survivors for histopathology and 3 hens for
NTE measurement are treated in a manner identical to the treated
groups, except that administration of the test substance is
omitted. When protective agents are used, all members of the dose
groups and vehicle controls should receive the same treatment.
(5) Study Conduct. (i) Biochemical measurements. (A)NTE
Assay. The test method is a differential assay of the ability of
neural tissue, following 0-P exposure, to selectively hydrolyze a
phenyl valerate substrate. The principle of the assay is: first,
to determine the amount of hydrolysis that occurs in the presence
of a non-neurotoxic inhibitor, paraoxon, (a), which is intended to
occupy irrelevant sites; Second, to determine the activity in the
presence of paraoxon and a known neuropathic inhibitor, mipafox,
(b). NTE activity is the difference between (a) and (b), that is,
the proportion of activity inhibited only by mipafox. Thus, the
"mipafox site" is already occupied following exposure to a
neuropathic O-P ester and the activity of (b) is therefore reduced.
(1) Three hens from each group should be
sacrificed at 48 hours after the last dose. Depending on the
duration of acute signs as an indication of the disposition of the
test material, the time for sacrifice for NTE and AchE assessment
may be chosen at a different time to optimize detection of effects.
Both the brain and spinal cord should be prepared for assay of NTE.
Perform duplicate assays of NTE in brain and spinal cord of three
birds from each group and control group.
(2) Materials. This assay requires
paraoxon (diethyl 4-nitrophenyl phosphate), mipafox (N, N' -
diisopropylphosphorodiamido fluoridate), and phenyl valerate. They
all can be obtained commercially.
(3) The assay has four stages:
Preparation of tissue; differential preincubation; hydrolysis of
-------
Delayed Neuro
substrate; and measurement of product. The quotations that follow
are from Johnson (1977) as corrected or modified in Johnson (1982).
His is the best known method for conduct of this assay. Other
acceptable methods have been used. They primarily involve minor
technical modification (Sprague et al. 1981; Soliman et al. 1982).
(a) Preparation of tissue. "the
whole brain (is) removed and cooled in ice-cold buffer (50 mM
Tris/0.2 mM EDTA adjusted to pH 8.0 at 25° with HCL) . Meninges and
blood vessels are rapidly removed and the brain is blotted dry,
weighed, and homogenized thoroughly in ice-cold buffer(at a volume
of at least 1:30, w/v), using a high-speed rotating perspex pestle
with not more than 0.25 mm difference in diameter between pestle
and tube."
(b) Differential preincubation.
"Paired samples of homogenate (equivalent to about 6.0 mg tissue)
are pre-incubated in Tris/EDTA buffer pH 8 at 37° for exactly 20
minutes with paraoxon (40 to 100 uM) plus either (a) buffer or (b)
mipafox (50uM) in a final volume of 2 ml."
(c) Hydrolysis of substrate. "After
preincubation, dispersion (2ml) of phenyl valerate is added and the
incubation is continued for exactly 15 minutes. The dispersion is
prepared by adding a solution of Triton X-100 (0.03 percent in
water) (30 vol) to a solution of phenyl valerate (15 or 20 mg/ml)
in redistilled dimethylformamide (1 vol) and mixing thoroughly (by
swirling): other solvents give less satisfactory dispersions.
Reaction is stopped by adding 2 ml of sodium dodecyl sulphate (1-2%
w/v) in buffer containing 4-aminoantipyrine (otherwise known as 4-
aminophenazone) (0.25 percent)."
(d) Measurement of product. This
assay is based on the colorimetric determination of liberated
phenol.
(1) "The coupling of phenol
liberated in the assay with the aminoantipyrine may be performed at
any convenient time after quenching the enzyme: 1 ml of K3Fe(CN)6
(0.4 percent in water) is added and the stable red colour is read
at 490 nm."
(2) "A nontissue blank, kept to
10 percent of the paraoxon tube value by maintaining the substrate
phenol fee, should be included in each group of assay tubes.
Typical control absorbance values would be 0.8 for paraoxon, 0.35
for paraoxon and mipafox and 0.07 for the blank. Colour
development takes (1-2 min) in solutions stopped with sodium
dodecyl sulphate. The extinction coefficient of phenol under these
conditions is 15,600 at a wavelength of 490 nm. NTE activity is
represented by the difference in absorbance obtained from samples
incubated under conditions (a) and (b) respectively."
(3) "Under standard conditions
NTE hydrolyzes about 2400 nanomoles of substrate/min/g of cortex,
550 for spinal cord, and 100 for sciatic nerve."
(B) AChE measures . Assay of
-------
Delayed Neuro
acetylcholinesterase in the brains of the same birds (Johnson and
Russell, 1975; or Ellman et al. , 1961) shall also be performed.
The level of AChE inhibition is a useful index of lethal potency
and the ratio of lethal potency to NTE inhibitory potency can be
useful for subsequent dose selection.
(ii) 21 Day observation. All remaining hens should
be carefully observed at least once daily for a period of at least
21 days and signs of toxicity recorded, including the time of
onset, degree, and duration. Observations should include, but not
be limited to, behavioral abnormality, locomotor ataxia, and
paralysis. At least twice a week the hens should be taken outside
the cages and subjected to a period of forced motor activity, such
as ladder climbing, in order to enhance the observation of minimal
responses. A rating scale of at least four levels should be used to
grade ataxia, e.g. Roberts et al. (1983).
(iii) Necropsy and Histopathology. Gross necropsies
are recommended for all survivors and should include observation of
the appearance of the brain and spinal cord. All animals shall be
prepared for microscopic examination. Tissues shall be fixed by
whole body perfusion, with a fixative appropriate for the embedding
media. Sections should include medulla oblongata, spinal cord, and
peripheral nerves. The spinal cord sections should be taken from
the rostral cervical, the midthoracic, and the lumbo-sacral
regions. Section of the proximal regions of both of the tibial
nerves and their branches should be taken. Sections should be
stained with appropriate myelin and axon-specific stains.
For 28 day studies, a stepwise examination of tissue samples is
recommended. In such a stepwise examination, sections from the high
dose group are first compared with those of the control group. If
no neuropathological alterations are observed in samples from the
high dose group, subsequent analysis is not required. If
neuropathological alterations are observed in samples from the high
dose group, samples from the intermediate and low dose groups are
then examined sequentially.
(e) Data reporting and evaluation. (1) Test report. In
addition to any other applicable reporting requirements, the final
test report must include the following information:
(i) Toxic response data by group with a description
of clinical signs; the criteria for the grading system for ataxia
and any other scales should be defined.
(ii) For each animal, time of death during the study
or whether it survived to termination.
(iii) The day of the first occurrence of each
abnormal sign and its subsequent course including its degree.
(iv) Body weight data.
(v) Necropsy findings for each animal, including a
description of the appearance of the brain and the spinal cord.
(vi) Biochemical data for each animal assessed,
including absorbance values for each animal tested, and blank
8
-------
Delayed Neuro
sample data.
(vii) A detailed description of all
histopathological findings.
(viii) Statistical treatment of results, where
appropriate.
(2) Treatment of results. Data may be summarized in
tabular form, showing for each test group the number of animals at
the start of the test, the number of animals showing lesions or
effects, the types of lesions or effects and the percentage of
animals displaying each type of lesion or effect.
(3) Evaluation of results. The findings of these delayed
neurotoxicity studies should be evaluated in terms of the incidence
and severity of behavioral, neurochemical, and histopathological
effects and of any other observed effects in the treated and
control groups, as well as any information known or available to
the authors, such as published studies. For a variety of results
seen, further studies may be necessary to characterize these
effects.
(f) References. For additional background information on this
test guideline the following references should be consulted:
Caroldi, S., Lotti, M. "Neurotoxic Esterase in Peripheral Nerve:
Assay Inhibition, and Rate of Resynthesis." Toxicology and Applied
Pharmacology. 62, 498-501 (1982).
Davis, C.S. and Richardson, R.J. Organophosphorus compounds. In:
Experimental and Clinical Neurotoxicology. P.S. Spencer and H.H.
Schaumberg, Eds., Williams and Wilkins, Baltimore, pp. 527-544.
(1980).
Ellman G.L., Courtney, K.D., Andres, V., and Featherstone, R.M. "A
new and rapid colorimetric determination of acetylcholinesterase
activity. Biochem. Pharmacol. 7:88-95.(1961)
Johnson, C.D. and Russell, R.L. "A rapid, simple, radiometric assay
for cholinesterase, suitable for multiple determinations. Anal.
Biochem. 64: 229-238 (1975).
Johnson, M.K. "Organophosphorus esters causing delayed neurotoxic
effects: Mechanism of action and structure/activity studies",
Archives of Toxicology 34:259-288.(1975a)
Johnson, M.K. "The delayed neuropathy caused by some
Organophosphorus esters:Mechanism and challenge". Crit.Rev.Toxicol
3:289-316.(1975b)
Johnson, M.K. "Improved Assay of Neurotoxic Esterase for Screening
Organophosphates for Delayed Neurotoxicity Potential," Archives of
Toxicology. 37, 113-115 (1977).
-------
Delayed Neuro
Johnson, M.K. "Delayed neurotoxicity tests of organophosphorus
esters: a proposed protocol integrating neuropathy target
esterase(NTE) assays with behaviour and histopathology tests to
obtain more information more quickly from fewer animals,"
Proceedings of the International Conference on Environmental
Hazards of Agrochemicals in Developing Countries. Alexandria,
November 8-12, 1983; Volume I, pp. 474-493.
Johnson, M.K. "The target for initiation of delayed neurotoxicity
by organophosphorus esters: biochemical studies and toxicological
applications", E.Hodgson, J.R. Bend,R.M.Philpot, eds., Reviews in
Biochem. Toxicol. 4,141-212. Elsevier, New York(1982)
Johnson, M.K., Richardson, R.J. "Biochemical Endpoints:
Neurotoxic Esterase Assay." Neurotoxicology, 4(2):311-320 (1983).
Kayyali, U.S., Moore, T., Randall, J.C. and Richardson, R.J.
"Neurotoxic Esterase Assay: Corrected wavelength and Extinction
Coefficient. The Toxicologist . 6:1 #292, 73 (1989)
Roberts, N.L., Fairley, C. , and Phillips, C. Screening acute
delayed and subchronic neurotoxicity studies in the hen:
Measurements and evaluations of clinical signs following
administration of TOCP. Neurotoxicology . 4, 263-270.
Soliman, S.A., Linder, R. , Farmer, J., Curley, A. "Species
Susceptibility to Delayed Toxic Neuropathy in relation to in vivo
inhibition of Neurotoxic Esterase by Neurotoxic Organophosphorus
Ester," Journal of Toxicology and Environmental Health. 9, 189-197
(1982).
Sprague, G.L., Sandvik, L.L., Bickford, A. A. "Time course for
neurotoxic esterase activity in hens given multiple diisopropyl
f luorophosphate injections, " Neurotoxicology. 2, 523-532 (1981).
10
-------
Delayed Neuro
TABLE 1
STUDY DESIGN
# OF HENS NTE/TIME BEHAVIOR/PATHOLOGY
ACUTE STUDY HOURS
POSITIVE CONTROLS 9 3/48 6
VEHICLE CONTROLS 9 3/48 6
DOSE 9 3/48 6
28 DAY STUDY HOURS
POSITIVE CONTROLS 9 3/48 6
VEHICLE CONTROLS 9 3/48 6
HIGH DOSE 9 3/48 6
LOW DOSE 9 3/48 6
MID DOSE 9 3/48 6
11
-------
Delayed Neuro
NOTES
1. Substances appropriate for testing are uncharged esters,
thioesters, or anhydrides of organophosphoric, organophosphonic, or
organophosphoramidic acids or of the related phosphorothioic,
phosphonothioic, or phosphorthioamidic acids.
2. Two commenters questioned the validity or usefulness of NTE
measurements.
The Agency believes that this assay will be both valid and
useful and that this is the consensus of the scientific community
. Only 2 of roughly 30 commenters questioned the addition of this
assay. The SAP clearly endorsed this approach. The published
literature amply demonstrates that inhibition of this protein is
necessary but not sufficient for the initiation of OPIDN, and that
it is highly correlated with the other signs of OPIDN, i.e. gait
changes, and central-peripheral distal axonopathy. The assay itself
was reviewed by the SAP in 1987 and has since been reviewed again
after minor revision both in ORD and by a number of reviewers as
well as the SAP- We believe that this assay possesses therefore,
concurrent, predictive, and content validity. While inhibition of
NTE is not, per se, an adverse effect, it is not being used as the
sole basis of such assertions.
12
-------
NEUROTOXICITY SCREENING BATTERY1
(a) Purpose. In the assessment and evaluation of the
potential human health effects of chemical substances, it is
appropriate to test for neurotoxic effects. This neurotoxicity
screening battery consists of a functional observational battery,
motor activity, and neuropathology. The functional observational
battery consists of non-invasive procedures designed to detect
gross functional deficits in animals and to better quantify
behavioral or neurological effects detected in other studies. The
motor activity test uses an automated device that measures the
level of activity of an individual animal . The
neuropathological techniques are designed to provide data to
detect and characterize histopathological changes in the central
and peripheral nervous system. This battery is designed to be
used in conjunction with general toxicity studies and changes
should be evaluated in the context of both the concordance
between functional neurological and neuropatholgical effects, and
with respect to any other toxicological effects seen This test
battery is not intended to provide a complete evaluation of
neurotoxicity, and additional functional and morphological
evaluation may be necessary to assess completely the neurotoxic
potential of a chemical.
(b) Definitions.
(1) Neurotoxicity is any adverse effect on the
structure or function of the nervous system related to exposure
to a chemical substance.
(2) A toxic effect is an adverse change in the
structure or function of an experimental animal as a result of
exposure to a chemical substance.
(3) Motor activity is any movement of the experimental
animal.
(c) Principle of the test method. The test substance is
administered to several groups of experimental animals, one dose
being used per group. The animals are observed under carefully
standardized conditions with sufficient frequency to ensure the
detection and quantification of behavioral and/or neurologic
abnormalities, if present. Various functions that could be
affected by neurotoxicants are assessed during each observation
period. Measurements of motor activity of individual animals are
made in an automated device. The animals are perfused and tissue
samples from the nervous system are prepared for microscopic
examination. The exposure levels at which significant neurotoxic
effects are produced are compared to one another and to those
levels that produce other toxic effects.
(d) Test procedures. (1) Animal selection, (i) Species. In
general, the laboratory rat should be used. Under some
circumstances, other species, such as the mouse or the dog, may
be more appropriate, although not all of the battery may be
13
-------
Neuro Screen
adaptable to other species.
(ii) Age. Young adults (at least 42 days old for
rats) shall be used.
(iii) Sex. (A) Both males and females shall be
used.
(B) Females shall be nulliparous and
nonpregnant.
(2) Number of animals. At least ten males and ten
females shall be used in each dose and control group for
behavioral testing . At least five males and five females shall
be used in each dose and control group for terminal
neuropathology. If interim neuropathological evaluations are
planned, the number shall be increased by the number of animals
scheduled to be perfused before the end of the study. Animals
shall be randomly assigned to treatment and control groups.
(3) Control groups, (i) A concurrent (vehicle) control
group is required. Subjects shall be treated in the same way as
for an exposure group except that administration of the test
substance is omitted. If the vehicle used has known or potential
toxic properties, both untreated or saline treated and vehicle
control groups are required.
(ii) Positive control data from the laboratory
performing the testing shall provide evidence of the ability of
the observational methods used to detect major neurotoxic
endpoints including limb weakness or paralysis (e.g., repeated
exposure to acrylamide),tremor (e.g., pp'DDT), and autonomic
signs (e.g.,carbaryl). Positive control data are also required to
demonstrate the sensitivity and reliability of the activity-
measuring device and testing procedures. These data should
demonstrate the ability to detect chemically induced increases
and decreases in activity- Positive control groups exhibiting
central nervous system pathology and peripheral nervous system
pathology are also required. Separate groups for peripheral and
central neuropathology are acceptable (e.g., acrylamide and
trimethyl tin). Positive control data shall be collected at the
time of the test study unless the laboratory can demonstrate the
adequacy of historical data for this purpose, i.e., by the
approach outlined in this guideline .
(4) Dose level and dose selection . At least 3 doses
shall be used in addition to the vehicle control group. Ideally,
the data should be sufficient to produce a dose-effect curve. We
strongly encourage the use of equally spaced doses and a
rationale for dose selection that will maximally support
detection of dose-effect relations.
For acute studies, dose selection may be made relative to
the establishment of a benchmark dose (BD) . That is, doses may
be specified as successive fractions, e.g. 1/2, 1/4, of the BD.
The BD itself may be estimated as the highest non-lethal dose as
determined in a preliminary range-finding lethality study. A
variety of test methodologies may be used for this purpose, and
14
-------
Neuro Screen
the method chosen may influence subsequent dose selection. The
goal is to use a dose level that is sufficient to be judged a
limit dose, or clearly toxic.
(i) Acute Studies. The high dose need not be
greater than 2 g/Kg. Otherwise, the high dose shall result in
significant neurotoxic effects or other clearly toxic effects,
but not result in an indicence of fatalities that would preclude
a meaningful evaluation of the data. This dose may be estimated
by a benchmark dose procedure as described above, with the middle
and low dose levels chosen as fractions of the benchmark dose.
The lowest dose shall produce minimal effect, e.g. an ED10, or
alternatively, no effects.
(ii) Subchronic (and Chronic) Studies. The high
dose need not be greater than Ig/Kg. Otherwise, the high dose
level shall result in significant neurotoxic effects or other
clearly toxic effects, but not produce an incidence of fatalities
that would prevent a meaningful evaluation of the data. The
middle and low doses should be fractions of the high dose. The
lowest dose shall produce minimal effects, e.g. an ED10, or
alternatively, no effects.
(5) Route of exposure. Selection of route may be based
on several criteria including, the most likely route of human
exposure, bioavailability, the likelihood of observing effects,
practical difficulties, and the likelihood of producing non-
specific effects. For many materials, it should be recognized
that more than one route of exposure may be important and that
these criteria may conflict with one another. In order to save
resources, initially only one route is being required for
screening for neurotoxicity. The route that best meets these
criteria should be selected. Dietary feeding will generally be
acceptable for repeated exposures studies.
(6) Combined protocol. The tests described in this
screening battery may be combined with any other toxicity study,
as long as none of the requirements of either are violated by the
combination.
(7) Study conduct, (i) Time of testing. All animals
shall be weighed on each test day and at least weekly during the
exposure period.
(A) Acute Studies. At a minimum, for acute
studies observations and activity testing shall be made before
the initiation of exposure, at the estimated time of peak effect
within 8 hours of dosing, and at 7 and 14 days after dosing.
Estimation of time(s) of peak effect may be made by dosing pairs
of rats across a range of doses and making regular observations
of gait and arousal .
(B) Subchronic (and Chronic) Studies8, in a
subchronic study, at a minimum, observations and activity
measurements shall be made before the initiation of exposure and,
before the daily exposure, or for feeding studies at the same
time of day, during the 4th, 8th, and 13th weeks of exposure. in
15
-------
Neuro Screen
chronic studies, at a minimum, observations and activity
measurements shall be made before the initiation of exposure and
before the daily exposure, or for feeding studies at the same
time of day, every 3 months.
(ii)Functional Observational Battery. (A) General
Conduct. All animals in a given study shall be observed
carefully by trained observers who are unaware of the animal's
treatment, using standardized procedures to minimize observer
variability. Where possible, it is advisable that the same
observer be used to evaluate the animals in a given study. If
this is not possible, some demonstration of inter-observer
reliability is required. The animals shall be removed from the
home cage to a standard arena for observation. Effort should be
made to ensure that variations in the test conditions are minimal
and are not systematically related to treatment. Among the
variables that can affect behavior are sound level, temperature,
humidity, lighting, odors, time of day , and environmental
distractions. Explicit, operationally defined scales for each
measure of the battery are to be used. The development of
objective quantitative measures of the observational end-points
specified is encouraged. Examples of observational procedures
using defined protocols may be found in Irwin (1968), Gad (1982),
and Moser et al. (1988). The functional observational battery
shall include a thorough description of the subject's appearance,
behavior, and functional integrity. This shall be assessed
through: observations in the home cage; while the rat is moving
freely in an open field; and through manipulative tests. Testing
should proceed from the least to the most interactive with the
subject. Scoring criteria, or explicitly defined scales, shall be
developed for those measures which involve subjective ranking.
(B) List of measures. The functional
observational battery shall include the following list of
measures.
(1) Assessment of signs of autonomic
function, including but not limited to:
a) ranking of the degree of
lacrimation and salivation, with a range of severity scores from
none to severe;
b) presence or absence of
piloerection and exophthalmus;
c) ranking or count of urination
and defecation, including polyuria and diarrhea. This is most
easily conducted during the open field assessment.
d) pupillary function such as
constriction of the pupil in response to light or a measure of
pupil size;
e) degree of palpebral closure,
e.g., ptosis.
(2) Description, incidence, and severity
of any convulsions, tremors, or abnormal motor movements, both in
16
-------
Neuro Screen
the home cage and the open field.
(3) Ranking of the subject's reactivity
to general stimuli such as removal from the cage or handling,
with a range of severity scores from no reaction to
hyperreactivity-
(4) Ranking of the subject's arousal
level or state of alertness during observations of the
unperturbed subject in the open fifQl<^ with a range of severity
scores from coma to hyperalertness .
(5) Descriptions and incidence of
posture and gait abnormalities observed in the home cage and open
field.
(6) Ranking of any gait abnormalities,
with a range of severity scores from none to severe.
(7) Forelimb and hindlimb grip strength
measured using an objective procedure, e.g. that described by
Meyer et al. (1979).
(8) Quantitative measure of landing foot
splay11; the procedure described by Edwards and Parker (1977) is
recommended.
(9) Sensorimotor responses to stimuli of
different modalities will be used to detect gross sensory
deficits. Pain perception may be assessed by a ranking or measure
of the reaction to a tail-pinch, tail-flick, or hot-plate. The
response to a sudden sound, e.g., click or snap, may be used to
assess audition.
(10) Body weight.
(11) Description and incidence of any
unusual or abnormal behaviors, excessive or repetitive actions
(stereotypies), emaciation, dehydration, hypotonia or hypertonia,
altered fur appearance, red or crusty deposits around the eyes,
nose, or mouth, and any other observations that may facilitate
interpretation of the data.
(C) Additional measures. Other measures may
also be included and the development and validation of new tests
is encouraged. Further information on the neurobehavioral
integrity of the subject may be provided by:
(1) Count of rearing activity on the
open field;
(2) Ranking of righting ability;
(3) Body temperature;
(4) Excessive or spontaneous
vocalizations;
(5) Alterations in rate and ease of
respiration, e.g., rales or dyspnea;
(6) Sensorimotor responses to visual or
proprioceptive stimuli.
(iii) Motor activity. Motor activity shall be
monitored by an automated activity recording apparatus. The
device used must be capable of detecting both increases and
17
-------
Neuro Screen
decreases in activity, i.e., baseline activity as measured by the
device must not be so low as to preclude detection of decreases
nor so high as to preclude detection of increases in activity.
Each device shall be tested by standard procedures to ensure, to
the extent possible, reliability of operation across devices and
across days for any one device. In addition, treatment groups
must be balanced across devices. Each animal shall be tested
individually- The test session shall be long enough for motor
activity to approach asymptotic levels by the last 20 percent of
the session for non-treated control animals. All sessions shall
have the same duration. Treatment groups shall be counter-
balanced across test times (See endnote 6). Effort should be
made to ensure that variations in the test conditions are minimal
and are not systematically related to treatment. Among the
variables which can affect motor activity are sound level, size
and shape of the test cage, temperature, relative humidity,
lighting conditions, odors, use of the home cage or a novel test
cage, and environmental distractions.
(iv) Neuropatholoqy: Collection, Processing and
Examination of Tissue Samples. To provide for adequate
sampling as well as optimal preservation of cellular integrity
for the detection of neuropathological alterations, tissue shall
be prepared for histological analysis using in situ perfusion and
paraffin and/or plastic embedding procedures. Paraffin embedding
is acceptable for tissue samples from the central nervous system.
Plastic embedding of tissue samples from the central nervous
system is encouraged, when feasible. Plastic embedding is
required for tissue samples from the peripheral nervous
system .
Subject to professional judgment and the type of
neuropathological alterations observed, it is recommended that
additional methods, such as Bodian's or Bielchowsky's silver
methods, and/or GFAP immunohistochemistry be used in conjunction
with more standard stains to determine the lowest dose level at
which neuropathological alterations are observed. When such
special stains indicate evidence of structural alterations it is
recommended that the GFAP radioimmunoassay also be performed,
particularly when additional animals are available for use in the
radioimmunoassay ^See Appendix 1, Guideline for GFAP
radioimmunoassay)
fA) Fixation and Processing of Tissue. The
nervous system shall be fixed by in situ perfusion with an
appropriate aldehyde fixative. Detailed descriptions of
vascular perfusions may be found in Zeman and Innes (1963) , Hayat
(1970), Spencer and Schaumburg (1980), and Palay and Chan Palay
(1974) . Any gross abnormalities should be noted. Tissue
samples taken shall adequately represent all major regions of the
nervous system. Detailed dissection procedures are described in
chapter 50 of Spencer and Schaumburg (1980) and in Palay and Chan
Palay (1974) . The tissue samples should be postfixed and
18
-------
Neuro Screen
processed according to standardized published histological
protocols such as AFIP (1968), WHO (1986), Spencer and
Schaumburg (1980), Bennet et al. (1976), Di Sant Agnese and De
Mesy Jensen (1984), or Fender (1985). Tissue blocks and
slides shall be appropriately identified when stored.
Histological sections shall be stained for hematoxylin and
eosin (H&E), or a comparable stain according to standard
published protocols such as AFIP(1968), Ralis et al. (1973), or
Bennet et al. (1976) . 15
(B) Qualitative Examination .
Representative histological sections from the tissue samples
shall be examined
microscopically by an appropriately trained pathologist
for evidence of neuropathological alterations. The
nervous system should be thoroughly examined for evidence
of any treatment-related neuropathological alterations.
Particular attention should be paid to regions known to be
sensitive to neurotoxic insult or those regions likely to be
affected based on the results of functional tests. Such
treatment-related neuropathological alterations should be clearly
distinguished from artifacts resulting from influences other than
exposure to the test substance. Guidance for both regions to be
examined and the types of neuropathological alterations that
typically result from toxicant exposure can be found in WHO
(1986). A stepwise examination of tissue samples is recommended.
In such a stepwise examination, sections from the high dose group
are first compared with those of the control group. If no
neuropathological alterations are observed in samples from the
high dose group, subsequent analysis is not required. If
neuropathological alterations are observed in samples from the
high dose group, samples from the intermediate and low dose
groups are then examined sequentially-
(C) Subjective Diagnosis16. If any evidence
of neuropathological alterations is found in the qualitative
examination, then a subjective diagnosis will be performed
for the purpose of evaluating dose-response relationships.
All regions of the nervous system exhibiting any evidence
of neuropathological changes shall be included in this
analysis. Sections from all dose groups from each region
will be coded and examined in randomized order without
knowledge of the code. The frequency of each type and
severity of each lesion will be recorded. After all samples
from all dose groups including all regions have been rated,
the code will be broken and statistical analysis performed
to evaluate dose-response relationships. For each type of
dose-related lesion observed, examples of different degrees
of severity shall be described. Photomicrographs of typical
examples of treatment-related regions are recommended to augment
these descriptions. These examples will also serve
to illustrate a rating scale, such as 1+, 2+, and 3+ for
19
-------
Neuro Screen
the degree of severity ranging from very slight to very
extensive.
(e) Data reporting and evaluation. The final test report
must include the following information.
(1) Description of equipment and test methods. A
description of the general design of the experiment and any
equipment used should be provided. This should include a short
justification explaining any decisions involving professional
j udgment.
(i) A detailed description of the procedures used
to standardize observations, including the arena and scoring
criteria. Procedures for calibrating and assuring the equivalence
of activity devices and balancing treatment groups should also be
described.
(ii) Positive control data from the laboratory
performing the test that demonstrate the sensitivity of the
procedures being used. Historical data may be used if all
essential aspects of the experimental protocol are the same.
Historical control data can be critical in the interpretation of
study findings. We encourage submission of such data to
facilitate the rapid and complete review of the significance of
effects seen.
(2) Results. The following information must be arranged
by test group dose level.
(i) In tabular form, data for each animal must be
provided showing:
(A) Its identification number;
(B) Its body weight and score on each sign at
each observation time, the time and cause of death (if
appropriate), total session activity counts, and intra-session
subtotals for each day measured.
(ii) Summary data for each group must include:
(A) The number of animals at the start of the
test ;
(B) The number of animals showing each
observation score at each observation time;
(C) The mean and standard deviation for each
continuous endpoint at each observation time;
(D) Results of statistical analyses for each
measure, where appropriate.
(iii) All neuropathological observations shall
be recorded and arranged by test groups. This data may be
presented in the following recommended format:
(A) Description of lesions for each animal.
For each animal, data must be submitted showing its
identification (animal number, sex, treatment, dose,
duration), a list of structures examined as well as the
location(s), nature, frequency, and severity of lesion(s).
Inclusion of photomicrographs is strongly recommended for
20
-------
Neuro Screen
demonstrating typical examples of the type and severity of the
neuropathological alterations observed is strongly recommended.
Any diagnoses derived from neurological signs and lesions
including naturally occurring diseases or conditions, shall be
recorded. .
(B) Counts and incidence of neuropathological
alterations by test group. Data shall be tabulated to show:
(1) The number of animals used in each
group and the number of animals in which any lesion was found.
(2) The number of animals affected by
each different type of lesion, the locations, frequency, and
average grade of each type of lesion.
(3) Evaluation of data. The findings from the screening
battery should be evaluated in the context of preceding and/or
concurrent toxicity studies and any correlated functional and
histopathological findings. The evaluation shall include the
relationship between the doses of the test substance and the
presence or absence, incidence and severity, of any neurotoxic
effects. The evaluation should include appropriate statistical
analyses, for example, parametric tests for continuous data and
non-parametric tests for the remainder. Choice of analyses
should consider tests appropriate to the experimental design,
including repeated measures. There may be many acceptable ways to
analyze data. Statistical analysis comparing total activity
counts of treatment vs control animals at each measured time must
be made and supplied. The report must include dose-effect curves
for observations, motor activity expressed as activity counts,
and any gross necropsy findings and lesions observed.
(f) References. For additional background information on
this test guideline the following references should be consulted:
AFIP. Manual of Histoloaic Staining Methods New York: McGraw
Hill, 1968.
Bennet, H.S., Wyrick, A.D., Lee, S.W., McNeil, J.H. "Science and
art in the preparing tissues embedded in plastic for light
microscopy, with special reference to glycol methacrylate, glass
knives and simple stains" Stain Technology 51: 71-97 (1976) .
Di Sant Agnese, P.A., De Mesy Jensen, K. "Dibasic staining of
large epoxy sections and application to surgical pathology"
American Journal of Clinical pathology 81: 25-29 (1984) .
Edwards, P.M., Parker V.H. "A simple, sensitive and objective
method for early assessment of acrylamide neuropathy in rats,"
Toxicology and Applied Pharmacology. 40: 589-591 (1977) .
Finger, F.W. "Measuring behavioral activity," Methods in
Psvchobiology Vol. 2 Ed. R.D. Myers . New York: Academic Press.
21
-------
Neuro Screen
pp.1-19 (1972).
Gad, S. A neuromuscular screen for use in industrial toxicology -
J.Toxicol. Environ. Health 9:691-704 (1982)
Hayat, M.A. "Volume 1. Biological applications." Principles and
Techniques of Electron Microscopy- New York, Van Nostrand
Reinhold. (1970).
Irwin, S. "Comprehensive observational assessment: la. A
systematic quantitative procedure for assessing the behavioral
physiological state of the mouse," Psychopharmacoloqia. 13: 222-
257 (1968).
Kinnard, E.J. and Watzman, N. "Techniques utilized in the
evaluation of psychotropic drugs on animals activity." Journal of
Pharmaceutical Sciences. 55: 995-1012 (1966).
Meyer, O.A., Tilson, H.A., Byrd, W.C., and Riley, W.T. A method
for the routine assessment of fore- and hindlimb grip strength of
rats and mice. Neurobehav. Toxicol. 1:233-236 (1979)
Moser V.C., Me Cormick J.P-, Creason J.P., and MacPhail R.C.
Comparison of chlordimeform and carbaryl using a functional
observational battery. Fund. Appl. Toxicol. 11:189-206 (1988).
Palay, S.L., Chan Palay, V. Cerebellar Cortex: Cytology and
Organization New York: Springer Verlag. (1974).
Pender, M.P. "A simple method for high resolution light
microscopy of nervous tissue" Journal of Neuroscience Methods 15:
213-218 (1985).
Ralis, H.M., Beesley, R.A., Ralis, Z.A. Techniques in
Neurohistology London: Butterworths. (1973).
Reiter, L.W. "Use of activity measures in behavioral toxicology,"
Environmental Health Perspectives. 26: 9-20 (1978).
Reiter, L.W. and MacPhail, R.C. "Motor Activity: A survey of
methods with potential use in toxicity testing," Neurobehavorial
Toxicology. 1: Suppl. 1, 53-66 (1979).
Robbins, T.W. "A critique of the methods available for the
measurement of spontaneous motor activity," Handbook of
Psvchopharmacology. Vol 7. Eds. Iversen, L.L., Iverson, D.S.,
Snyder, S.H. New York: Plenum Press, pp. 37-82 (1977).
Spencer, P.S., Schaumburg, H.H. (eds) Experimental and Clinical
Neurotoxicology Baltimore: Williams and Wilkins (1980) .
22
-------
Neuro Screen
WHO, Principles and Methods for the Assessment of
Neurotoxicity Associated with Exposure to Chemicals
(Environmental Health Criteria 60) World Health
Organizations Publications Center USA, Albany, New York
(1986)
Zeman, W., Innes, J.R. Craigie's Neuroanatomy of the Rat New
York: Academic Press. (1963).
NOTES
1. This version of these Neurotoxicity Test Guidelines represents
a joint effort of the Offices of Pesticide Programs (OPP) and the
Office of Toxic Substances (OTS), in cooperation with many
scientists in the Office of Research and Development (ORD), to
develop a common set of guidelines for their testing
requirements. This OPTS version grew out of the set of guidelines
developed and eventually published by OTS (50 FR 39397 9/27/85;
amended at 52 FR 19082, 5/20/87). The revisions were initiated by
an ad hoc Workgroup of scientists from these 3 offices. They were
presented for review by the Scientific Advisory Panel of OPP and
made available for public comment for 2 months. Over 30 groups
and individuals submitted comments. Many sections of these
neurotoxicity guidelines have been revised to take account of
these comments. The rationale for some of these general revisions
is provided here. What earlier were separate guidelines for the
functional observational battery, motor activity, and
neuropathology have now been combined into this single guideline
both for efficiency and because they were designed to be used
together.
2. The Agency recognizes that tests of motor acitivity alone do
not provide a complete evaluation of the effects of a chemical on
the nervous system. However, the automated test of motor activity
will provide an objective assessment of neurobehavioral function,
as well as the only measurement of habituation, which is an
indication of the organism's ability to adapt to its environment.
3. The power calculations to determine group size for motor
activity have been deleted. Group sizes of ten/sex will be
sufficient for well designed and executed studies. Poorly
designed and/or executed studies may be judged invalid. The
original intent was to allow for flexibility in the use of
devices with different operating characteristics (larger variance
in measures necessitate larger group sizes). Comments were mostly
negative, based on concerns about perceived uncertainty of the
adequacy of a sample size until a study was complete.
23
-------
Neuro Screen
4. Positive control data need only be generated approximately
once every year as long as most conditions remain the same. Each
observer should conduct such studies as part of their training.
The untreated control data from training studies could then be
submitted as part of the laboratory's historical control data.
5. The adequacy of the high dose for hazard identification is
one of the most critical issues of experimental design. The high
dose criteria in these guidelines now include both a limit dose
and " significant neurotoxic effects or other clearly toxic
effects".
These revised criteria are very similar to those for
systemic studies and are intended by this to facilitate combined
studies.
We now also include some guidance for dose selection in
acute studies based on a benchmark dose criterion and fractions
thereof.
This is intended to provide an example of one operational means
of establishing a set of acute doses and to reduce the number of
animals used for estimates of lethal doses.
These 3 dose studies are intended to both identify effects
of exposure and to estimate dose levels without adverse effect.
We have added language to encourage greater emphasis on obtaining
dose response data, e.g., equally spaced doses and lesser
emphasis on a low dose totally without effect. This was done for
2 reasons. First, the presence or absence of dose-related changes
can be critical in the evaluation of effects of exposure. Second,
various methods using an ED10 as the basis for estimating
reference doses are increasingly discussed by many authors.
We strongly urge sponsors to seek guidance from the Agency
before initiating their studies and to provide a rationale for
dose selection in these studies.
Several commenters were concerned about the interpretation
of effects seen at levels where other significant toxicity was
present and questioned the need for or efficiency of testing at
such doses. First, in OPP, the 90 day Neurotoxicity Study is
intended as a screen prior to inclusion of neurotoxicity tests in
chronic studies. In OTS, subchronic non-oncogenicity studies are
generally considered sufficient for evaluating chronic toxicity.
Thus the doses of a 90 day neurotoxicity study for either office
should be maximized to encourage the identification of chronic
effects.
Further, concurrent toxicity does not obviate the need to
identify other kinds of effects that may be more important in
different situations, either for other groups or after different
exposure regimens. Ultimately, all of the targets of a toxicant
may be important for identifying affected individuals or be the
critical effect under a variety of exposure conditions or in
different groups.
24
-------
Neuro Screen
6. The term benchmark dose as defined here is for the purpose of
dose selection and should not be confused with the use of this
term for the purpose of establishing a common dose such as an
ED10 for extrapolation in risk assessment.
7. The time of peak effect refers to the time within
approximately 8 hours after dosing and was intended to help
maximize detection of effects of acute exposure.
8. For repeated exposure studies using routes other than through
the diet, the intent of testing before the daily dose is to
minimize the impact of that day's dose. We recognize that for
some materials, residual material from preceding dose or doses
may be the source of observed effects, but these are part of the
effects of concern. In some studies, examination of animals
following exposure may help to further describe the duration of
such effects.
9. Not all rats must be tested in one day, but time of testing
should be balanced across groups, and for any other potential
confounds, e.g., sex.
10. Measures of reactivity refer to the subject's reaction to
some external stimulus, e.g., removal from the cage or handling,
while arousal or state of alertness refers to the behavior of the
undisturbed subject observed in the open field. This is often
described more technically as the distinction between respondent
and operant behaviors.
11. Landing foot splay and grip strength do not measure the same
function. These tests are viewed as complementary, 'and having
both will aid in the interpretation of data.
12. The goal of the procedures outlined for the preparation and
processing of tissue samples is to optimally preserve tissue
morphology for microscopic examination. The higher resolution
obtainable in plastic embedded tissue is considered to optimize
the detection of a number of types of lesions, particularly in
the peripheral nervous system. In contrast, paraffin-embedded
material is more amenable to sampling large regions of the
nervous system and is considered optimal for a variety of special
stains that may be useful in characterizing neuropathological
alterations. Several organizations felt the requirement for
separate animals for plastic and paraffin-embedding of tissue
samples was excessive. Furthermore, commentors presented views
that differences in plastic and paraffin techniques did not
25
-------
Neuro Screen
require the use of separate animals and that in an appropriate
tiered evaluation scheme both plastic and paraffin-embedding
could be used with tissue samples from the same animals. The
requirement for separate animals is now eliminated, with paraffin
being acceptable and plastic being recommended.
13. The higher resolution obtainable in plastic embedded tissue
is considered to optimize the detection of a number of types of
lesions, particularly in the peripheral nervous system. Paraffin
embedded material is more amenable to sampling large regions of
the nervous system and is considered optimal for a variety of
special stains that may be useful in characterizing
neuropathological alterations.
14. Although, EPA believes that the GFAP assay has been shown to
be sensitive to the neurotoxic effects of agents in both the
adult and developing nervous systems, this assay has been deleted
at this time. However, the Agency will continue using this assay
experimentally and encourages others to do so, as well, in an
effort to obtain additional validation of its use as a means to
assess the neurotoxic potential of agents. In addition, if GFAP
immunohistochemistry is used as a special stain in the
neuropathology segment of the testing protocol and evidence of a
glial response to toxicant injury is observed, application of the
radioimmunoassay is encouraged in order to provide objective,
quantitative dose-response data.
15. The Agency received some comments that the list of specified
regions of the nervous system to be examined was inadequate,
while others felt it was too detailed. Moreover, comments were
received that argued that the list of potential types of
neuropathological alterations also was too restricted. These
lists were intended to serve as guidance. Since they appear to
be subject to misinterpretation, the requirement for a thorough
examination of the nervous system for any evidence of
neuropathological alteration is now explicitly stated. In
addition, a list of all structures examined is required in the
final report. The requirement for examination of more than one
section per region, however, has now been deleted.
16. The purpose of the semi-quantitative analysis is to evaluate
the relationship between the incidence and severity of the
neuropathological alterations and the exposure. Since the rating
scale is by necessity subjective, it is necessary to ensure that
any bias resulting from the previous qualitative examination of
the tissue samples is minimized. Several organizations commented
26
-------
Neuro Screen
that blind evaluation of tissue sections should not be required.
In the initial qualitative analysis, in which the types of
lesions and regions affected are first identified, blind
evaluation is not required. However, in the semi-quantitative
analysis in which the dose-response relationship is evaluated, it
is imperative that the evaluation be as objective as possible.
Moreover, since the semi-quantitative analysis focuses on a
limited number of regions for lesions previously described in the
qualitative analysis, blind reading is required to ensure
objectivity. Thus, it is required that the subjective rating of
the severity and incidence be performed without knowledge of
treatment.
17. The data and analyses supplied in the report must be
evaluated by Agency risk assessors. Thus, the report must be
sufficiently detailed for the Agency to evaluate the quality of
the study. Since no list of regions to be examined is outlined
in the guideline, a list of regions examined must be supplied
with the report. Similarly, an adequate description of lesions
observed must be supplied also. The Agency received comments
that the requirement of photomicrographs to document
neuropathological alterations was extremely costly. The Agency
has decided to recommend, rather than require, the use of
photomicrographs to aid in the description of typical examples of
treatment-related lesions.
27
-------
Neuro Screen
APPENDIX 1
GUIDELINE FOR ASSAYING GLIAL FIBRILLARY ACIDIC PROTEIN
(a) Purpose. Chemical-induced injury of the nervous system
is associated with astrocytic hypertrophy at the site of damage
(see O'Callaghan, 1988). Assays of glial fibrillary acid protein
(GFAP), the major intermediate filament protein of astrocytes,
can be used to document this response. To date, a diverse
variety of chemical insults known to be injurious to the central
nervous system have been shown to increase GFAP- Moreover,
increases in GFAP can be seen at dosages below those necessary to
produce cytopathology as determined by routine Nissl stains
(standard neuropathology). Thus, it appears that assays of GFAP
represent a sensitive approach for documenting the existence and
location of chemical-induced injury of the central nervous
system.
(b) Principle of the test method. This guideline will
describe the conduct of a radioimmunoassay for measurement of the
amount of GFAP in the brain of exposed and control animals. It
is based on modifications (O'Callaghan & Miller 1985, O'Callaghan
1987, O'Callaghan and Miller, 1988) of the dot-immunobinding
procedure described by Jahn et al. (1984). Briefly, samples are
assayed for total protein, diluted in dot-immunobinding buffer,
and applied to nitrocellulose sheets. The spotted sheets are
then fixed, blocked, washed, and incubated in anti-GFAP and
[1251] Protein A. Bound protein A is then quantified by gamma
spectrometry. In lieu of purified protein standards, standard
curves are constructed from dilution of a single control sample.
By comparing the immunoreactivity of individual samples (both
control and treated groups) with that of the sample used to
generate the standard curve, the relative immunoreactivity of
each sample is obtained. The immunoreactivity of the control
groups is normalized to 100% and all data are expressed as a
percentage of control. This biochemical test is intended to be
used in conjunction with behavioral and pathological studies as
part of the screening battery that includes the functional
observational battery, motor activity and histopathology.
rc^ Test procedure. (I) Animal selection, (i) Species and
strain. Test shall be performed in the species being used in
other tests for neurotoxicity. This will generally be the
laboratory rat.
(ii^ Aae. Based on the other concurrent testing
young adult rats shall be used.
(iii) Number of animals. At least 6 animals per
dose shall be used.
<-2l Materials: [1251] Protein A (2-10 uCi/ug) , Antisera
to GFAP, Nitrocellulose paper (0.1 or 0.2 urn pore size), a sample
28
-------
Neuro Screen
application template (optional, e.g. Minifold II, Schleicher &
Schuell, Keene NH), plastic incubation trays.
(i) Study conduct, fi) Tissue Preparation. Animals are
killed by decapitation 72 hours after the last dose. The brain
is excised from the skull. The following six regions are then
dissected freehand: cerebellum, cerebral cortex, hippocampus,
striatum, thalamus/hypothalamus, and the rest of the brain.
Each region is then weighed and homogenized in 10 volumes of
hot (70-90 degrees C) 1% (w/v) sodium dodecyl sulfate (SDS).
Homogenization is best achieved through sonic disruption. A
motor driven pestle inserted into a tissue grinding vessel
is a suitable alternative. The homogenized samples can then
be stored frozen at -70 C for at least 4 years without loss
of GFAP content.
(ii) Total Protein Assay. Aliquots of the tissue
samples are assayed for total protein using the method of Smith
et. al. (1985). This assay is available in kit form (Pierce
Chemical Company, Rockford, IL).
(iii) Sample Preparation. Dilute tissue samples
in sample buffer (120 mM KC1, 20 mM NaCl, 2 mM NaHC03), 2 mM
MgC12), 5 mM Hepes, pH 7.4, 0.7% Triton X-100) to a final
concentration of 0.25 mg total protein per ml (5 ug/20 ul) .
(iv) Preparation of Standard Curve. Dilute a
single control sample in sample buffer to give at least five
standards, between 1 and 10 ug total protein per 20 ul. The
suggested values of total protein per 20 ul sample buffer are:
1.25, 2.50, 3.25, 5.0, 6.25, 7.5, 8.75, and 10.0 ug.
(v) Preparation of Nitrocellulose Sheets.
Nitrocellulose sheets of 0.1 or 0.2 micron pore size are rinsed
by immersion in distilled water for 5 minutes and then air dried.
(vi) Sample Application. Samples can be spotted
onto the nitrocellulose sheets free-hand or with the aid of a
template. For free-hand application, draw a grid of squares
approximately 2 cm by 2 cm on the nitrocellulose sheets
using a soft pencil. Spot 5-10 ul portions to the center of
each square for a total sample volume of 20 ul. For
template aided sample application a washerless microliter
capacity sample application manifold is used. Position the
nitrocellulose sheet in the sample application device as
recommended by the manufacturer and spot a 20 ul sample in
one application. Do not wet the nitrocellulose or any
support elements prior to sample application. Do not apply
vacuum during or after sample application. After spotting
samples (using either method), let the sheets air dry. The
sheets can be stored at room temperature for several days
after sample application.
(vii) Standard Incubation Conditions. These
conditions have been described by Jahn et al. (1984). All steps
are carried out at room temperature on a flat shaking platform
(one complete excursion every 2-3 seconds). For best results do
29
-------
Neuro Screen
not use rocking or orbital shakers. Perform the following
steps in enough solution to cover the nitrocellulose sheets
to a depth of 1 cm.
(A) Incubate 20 minutes in fixer (25% v/v
isopropanol, 10% v/v acetic acid).
(B) Discard fixer, wash several times in
deionized water to eliminate the fixer, and then incubate for 5
minutes in Tris-buffered saline (TBS, 200 mM NaCl, 60 mM Tris-HCl
pH 7.4).
(C) Discard TBS and incubate 1 hour in
blocking solution (0.5% gelatin (w/v) in TBS.
(D) Discard blocking solution and incubate
for 2 hours in antibody solution (anti-GFAP antiserum diluted to
the desired dilution in blocking solution containing 0.1% Triton
X-100). Serum antibovine GFAP, which cross reacts with GFAP from
rodents and humans, can be obtained commercially (e.g. Dako
Corp.) and used at a dilution of 1:500.
(E) Discard antibody solution, wash in 4
changes of TBS for 5 minutes each time. Then wash in TBS for 10
minutes.
(F) Discard TBS and incubate in blocking
solution for 30 minutes.
(G) Discard blocking solution and incubate
for 1 hour in Protein A solution ([1251]-labeled Protein A
diluted in blocking solution containing 0.1% Triton X-100,
sufficient to produce 2000 cpm per 10 ul of protein A solution).
(H) Remove protein A solution (it can be
reused once). Wash in 0.1% Triton X-100 in TBS (TBSTX) for 5
minutes, 4 times. Then wash in TBSTX for 2-3 hours for 4
additional times. An overnight wash in a larger volume can be
used to replace the last 4 washes.
(I) Hang sheets up to dry, cut out squares or
spots and count radioactivity in a gamma- counter.
(viii) Expression of data. Compare radioactivity
counts for samples obtained from control and treated animals with
counts obtained from the standard curve. By comparing the
immunoreactivity (counts) of each sample with that of the
standard curve, the relative amount of GFAP in each sample
can be determined and expressed as a percent of control.
(d) Data Reporting and Evaluation. (1) Test Report. The
final test report shall include the following information:
(i) Body weight and brain region weights at time
of sacrifice for each subject tested.
(ii) Indication of whether each subject survived
to sacrifice or time of death.
(iii) Data from control animals and blank samples.
(iv) Statistical evaluation of results.
(2) Evaluation of Results, (i) Results shall be
evaluated in terms of the extent of change in the amount of GFAP
30
-------
Neuro Screen
as a function of treatment and dose. GFAP assays (of any brain
region) from 6 samples typically result in a standard error of
the mean of +/- 5%- Chemical-induced increase of GFAP of 115% of
control are likely to be statistically significant.
(ii) Results of this assay shall be compared to
and evaluated with behavioral and histopathological data.
(&} References. For additional background information on
this test guideline the following references should be consulted:
Brock, T.O. and O'Callaghan, J.P- 1987. Quantitative changes
in the synaptic vesicle proteins, synapsin I and p38 and the
astrocyte specific protein, glial fibrillary acidic protein,
are associated with chemical-induced injury to the rat
central nervous system. J. Neurosci. 7:931-942
Jahn, R. , Schiebler, W. Greengard, P. 1984. A quantitative
dot-immunobinding assay for protein using nitrocellulose
membrane filters. Proc. Natl. Acad. Sci. U.S.A. 81:1684-
1687.
O'Callaghan, J. P. 1988. Neurotypic and gliotypic protein as
biochemical markers of neurotoxicity. Neurotoxicol. Teratol.
10:445-452.
O'Callaghan, J. P. and Miller, D. B. 1988. Acute exposure of
the neonatal rat to triethyltin results in persistent
changes in neurotypic and gliotypic proteins. J. Pharmacol. Exp.
Ther. 244:368-378.
O'Callaghan, J. P. and Miller, D. B. 1985. Cerebellar
hypoplasia in the Gunn rat is associated with quantitative
changes in neurotypic and gliotypic proteins. J. Pharmacol. EXP.
Ther. 234:522-532.
Smith, P.K., Krohn, R.I., Hermanson, G.T., Mallia, A. K.
Gartner, F. H., Provenzano, M.D., Fujimoto, E. K. , Goeke,
N.M. Olson, B.J., Klenk, D.C. 1985. Measurement of protein
using bicinchoninic acid. Annal. Biochem. 150:76-85
31
-------
DEVELOPMENTAL NEUROTOXICITY STUDY1
(a) Purpose. In the assessment and evaluation of the toxic
characteristics of a chemical substance or mixture ("test
substance"), determination of the potential for developmental
neurotoxicity is important. This study is designed to develop
data on the potential functional and morphological hazards to the
nervous system which may arise in the offspring from exposure of
the mother during pregnancy and lactation.
(b) Principle of the test method. The test substance is
administered to several groups of pregnant animals during
gestation and early lactation, one dose level being used per
group. Offspring are randomly selected from within litters for
neurotoxicity evaluation. The evaluation includes observations
to detect gross neurologic and behavioral abnormalities,
determination of motor activity, response to auditory startle,
assessment of learning, neuropathological evaluation, and brain
weights. This protocol may be used as a separate study, as a
follow-up to a standard developmental toxicity and/or adult
neurotoxicity study, or as part of a two-generation reproduction
study, with assessment of the offspring conducted on the F2
generation.
(c) Test procedure. (1) Animal selection, (i) Species and
strain. Testing should be performed in the rat. Because of its
differences in timing of developmental events compared to strains
that are more commonly tested in other developmental and
reproductive toxicity studies, it is preferred that the Fischer
344 strain not be used. If a sponsor wishes to use the Fischer
344 rat or a mammalian species other than the rat, ample
justification/reasoning for this selection must be provided.
(ii) Age. Young adult (nulliparous females)
animals shall be used.
(iii) Sex. Pregnant female animals shall be used
at each dose level.
(iv) Number of animals. (A) The objective is for a
sufficient number of pregnant rats to be exposed to the test
substance to ensure that an adequate number of offspring are
produced for neurotoxicity evaluation. At least 20 litters are
recommended at each dose level. For behavioral tests, one female
and one male pup per litter shall be randomly selected and
assigned to one of the tests.
(B) On postnatal day 4, the size of each
litter should be adjusted by eliminating extra pups by random
selection to yield, as nearly as possible, 4 male and 4 females
per litter. Whenever the number of pups of either sex prevents
having four of each sex per litter, partial adjustment (for
example, 5 males and 3 females) is permitted. Testing is not
appropriate for litters of less than 7 pups. Elimination of
runts only is not appropriate. Individual pups should be
32
-------
Devel Neuro
identified uniquely after standardization of litters. A method
that may be used for identification can be found in Adams et al•
(1985) .
(C) Assignment of animals for behavioral
tests, brain weights, and neuropatholoaical evaluations. After
standardization of litters, one male and one female from each
litter shall be randomly assigned to one of the following tests:
motor activity; auditory startle; and learning and memory, in
weanling and adult animals. On postnatal day 11, either one male
or one female pup from each litter (total of 10 males and 10
females/dose group) shall be sacrificed. Brain weights shall be
measured in all of these pups and, of these pups, 6/sex/dose
shall be selected for neuropathological evaluation. At the
termination of the study, either one male or one female from each
litter (total of 10 males and 10 females/dose group) shall be
sacrificed and brain weights shall be measured. An additional
group of 6 animals/sex/dose group (one male or one female per
litter) shall be sacrificed at the termination of the study for
neuropathological evaluation.
(2) Control groups. A concurrent control group(s) is
(are) required. This group shall be a sham-treated group or, if
a vehicle is used in administering the test substance, a vehicle
control group. The vehicle shall neither be developmentally
toxic nor have effects on reproduction. Animals in the control
group(s) shall be handled in an identical manner to test group
animals.
(3) Dose levels and dose selection, (i) At least 3
dose levels of the test substance plus a control group (vehicle
control, if a vehicle is used) shall be used. ,
(ii) If the test substance has been shown to be
developmentally toxic either in a standard developmental toxicity
study or in a pilot study, the highest dose level shall be the
maximum dose which will not induce in utero or neonatal death or
malformations sufficient to preclude a meaningful evaluation of
neurotoxicity .
(iii) If a standard developmental toxicity study
has not been conducted, the highest dose level, unless limited by
the physico-chemical nature or biological properties of the
substance, shall induce some overt maternal toxicity, but shall
not result in a reduction in weight gain exceeding 20% during
gestation and lactation.
(iv) The lowest dose should not produce any
grossly observable evidence of either maternal or developmental
neurotoxicity.
(v) The intermediate dose(s) shall be equally
spaced between the highest and lowest doses used.
(4) Dosing period . Day 0 of gestation is the day on
which a vaginal plug and/or sperm are observed. The dosing
period shall cover the period from day 6 of gestation through day
10 postnatally. Dosing should not occur on the day of
33
-------
Devei Neuro
parturition in those animals who have not completely delivered
their offspring.
(5) Administration of the test substance. The test
substance or ve"hicle shall be administered orally- Other routes
of administration may be acceptable, on a-case-by-case basis,
with ample justification/reasoning for this selection . The
test substance or vehicle shall be administered at the same time
each day. The animals shall be weighed periodically and the
dosage to be administered based on the most recent weight
determination.
(6) Observation of dams, (i) A gross examination of the
dams shall be made at least once each day before daily treatment.
The animals shall be observed by trained technicians, who are
unaware of the animal's treatment, using standardized procedures
to maximize inter-observer reliability. Where possible, it is
advisable that the same observer be used to evaluate the animals
in a given study. If this is not possible, some demonstration of
inter-observer reliability is required.
(ii) During the treatment and observation
periods, observations shall include:
(A) Assessment of signs of autonomic
function, including but not limited to:
(1) ranking of the degree of
lacrimation and salivation, with a range of severity scores from
none to severe;
(2) presence or absence of piloerection
and exophthalmus;
(3) ranking or count of urination and
defecation, including polyuria and diarrhea;
(4) pupillary function such as
constriction of the pupil in response to light or a measure of
pupil size;
(5) degree of palpebral closure, e.g.,
ptosis.
(B) Description, incidence, and severity of
any convulsions, tremors, or abnormal movements.
(C) Description and incidence of posture and
gait abnormalities.
(D) Description and incidence of any unusual
or abnormal behaviors, excessive or repetitive actions
(stereotypies), emaciation, dehydration, hypotonia or hypertonia,
altered fur appearance, red or crusty deposits around the eyes,
nose, or mouth, and any other observations that may facilitate
interpretation of the data.
(iii) Signs of toxicity shall be recorded as they
are observed, including the time of onset, degree, and duration.
(iv) Animals shall be weighed at least weekly and
on the day of delivery and postnatal days 11 and 21 (weaning);
such weights shall be recorded.
(v) The day of delivery of litters shall be
34
-------
Devel Neuro
recorded and considered as postnatal day 0.
(7) Study conduct, (i) Observation of offspring.
(A) All offspring shall be examined cage-side
at least daily for gross signs of mortality or morbidity.
(B) All offspring shall be examined outside
the cage for gross signs of toxicity whenever they are weighed or
removed from their cages for behavioral testing. The offspring
shall be observed by trained technicians, who are unaware of the
animals' treatment, using standardized procedures to maximize
inter-observer reliability. Where possible, it is advisable that
the same observer be used to evaluate the animals in a given
study. If this is not possible, some demonstration of inter-
observer reliability is required. At a minimum, the end points
outlined in paragraph (6) (ii) shall be monitored as appropriate
for the developmental stage being observed.
(C) Any gross signs of toxicity in the
offspring shall be recorded as they are observed, including the
time of onset, degree, and duration.
(ii) Developmental landmarks. Live pups shall be
counted and each pup within a litter shall be weighed
individually at birth or soon thereafter, and on postnatal days
4, 11, 17, 21 and at least once every two weeks thereafter. The
age of vaginal opening and preputial separation shall be
determined. General procedures for these determinations may be
found in Adams et al. (1985), and Korenbrot et al.
(1977),respectively.
(iii) Motor activity. Motor activity shall be
monitored specifically on postnatal days 13, 17, 21, and 60 (+2
days) . Motor activity must be monitored by an automated
activity recording apparatus. The device must be capable of
detecting both increases and decreases in activity, (i.e.,
baseline activity as measured by the device must not be so low as
to preclude detection of decreases nor so high as to preclude
detection of increases in activity). Each device shall be tested
by standard procedures to ensure, to the extent possible,
reliability of operation across devices and across days for any
one device. In addition, treatment groups must be balanced
across devices. Each animal shall be tested individually. The
test session shall be long enough for motor activity to approach
asymptotic levels by the last 20 percent of the session for non-
treated control animals. All sessions shall have the same
duration. Treatment groups shall be counter-balanced across test
times. Animals' activity counts shall be collected in equal
time periods of no greater than 10 minutes duration. Efforts
shall be made to ensure that variations in the test conditions
are minimal and are not systematically related to treatment.
Among the variables that can affect motor activity are sound
level, size and. shape of the test cage, temperature, relative
humidity, light conditions, odors, use of home cage or novel test
cage, and environmental distractions. Additional information on
35
-------
Devel Neuro
the conduct of a motor activity study may be obtained in the
Office of Pesticides and Toxic Substances Neurotoxicity Screening
Battery Guideline.
(iv) Auditory startle test. An auditory startle
habituation test shall be performed on the offspring on days 22
and 60+2. Details on the conduct of this testing may be
obtained in Adams et al. (1985). In performing the auditory
startle task, the mean response amplitude on each block of 10
trials (5 blocks of 10 trials per session on each day of testing)
shall be made. While use of pre-pulse inhibition is not a
requirement, it is highly recommended. Details on the conduct of
this test may be obtained from Ison (1984).
(v) Learning and memory tests. A test of
associative learning and memory shall be conducted around the
time of weaning (postnatal day 21-24) and at adulthood (postnatal
day 60 + 2). The same or separate test(s) may be used at these
two stages of development. Some flexibility is allowed in the
choice of test(s) for learning and memory in weanling and adult
rats. However, the test(s) must be designed so as to fulfill two
criteria. First, learning must be assessed either as a change
across several repeated learning trials or sessions, or, in tests
involving a single trial, with reference to a condition that
controls for non-associative effects of the training experience.
Second, the test(s) shall include some measure of memory (short-
term or long-term) in addition to original learning
(acquisition), but note that this measure of memory cannot be
reported in the absence of a measure of acquisition obtained from
the same test. If the test(s) of learning and memory reveal(s)
an effect of the test compound, it may be in the best interest of
the sponsor to conduct additional tests to rule out alternative
interpretations based on alterations in sensory, motivational,
and/or motor capacities. In addition to the above two criteria,
it is recommended that the test of learning and memory be chosen
on the basis of its demonstrated sensitivity to the class of
compound under investigation, if such information is available in
the literature. In the absence of such information, examples of
tests that could be made to meet the above criteria include:
delayed-matching-to-position, as described for the adult rat in
Bushnell (1988) and for the infant rat in Green and Stanton
(1988, Experiment 2); olfactory conditioning, as described in
Kucharski and Spear (1984, Experiment 3); and acquisition and
retention of schedule-controlled behavior, e.g., Cory-Slechta et
al., 1983, and Campbell and Haroutunian, 1981. Additional tests
for weanling rats are described in Spear and Campbell (1978) and
Krasnegor et al. (1986), and for adult rats in Miller and
Eckerman (1986). 6
(iv) Neuropatholoav . Neuropathological
evaluation shall be conducted on animals on postnatal day 11 and
at the termination of the study. At 11 days of age, one male or
female pup shall be removed from each litter such that equal
36
-------
Devel Neuro
numbers of male and female offspring are removed from all litters
combined. Of these, 6 male and 6 female pups will be sacrificed
for neuropathological analysis. The pups will be killed by
exposure to carbon dioxide and immediately thereafter the brains
shall be removed, weighed, and immersion fixed in an appropriate
aldehyde fixative. The remaining animals will be sacrificed in a
similar manner and immediately thereafter their brains removed
and weighed. At the termination of the study, one male or one
female from each litter will be killed by exposure to carbon
dioxide and immediately thereafter the brain shall be removed and
weighed. In addition, 6 animals/sex/dose group (one male or
female per litter) shall be sacrificed at the termination of the
study for neuropathological evaluation. Neuropathological
analysis of animals sacrificed at the termination of the study
shall be performed in accordance with the Office of Pesticides
and Toxic Substances Neurotoxicity Screening Battery.
Neuropathological evaluation of animals sacrificed on postnatal
day 11 and at termination of the study shall include a
qualitative analysis and semi-quantitative analysis as well as
simple morphometrics.
(A) Fixation and Processing of Tissue
Samples for Postnatal Day 11 Animals. Immediately following
removal, the brain shall be weighed and immersion fixed in an
appropriate aldehyde fixative. The brains should be postfixed
and processed according to standardized published histological
protocols a such as the AFIP (1968), Spencer and Schaumburg
(1980), Di Sant Agnese and De Mesy Jensen (1984) , or Fender
(1985) . Paraffin embedding is acceptable but plastic embedding
is preferred and recommended. Tissue blocks and slides shall be
appropriately identified when stored. Histological sections
shall be stained for hematoxylin and eosin, or a similar stain
according to standard published protocols such as AFIP (1968),
Ralis et al. (1973), or Bennet et al. (1976). For animals
sacrificed at the termination of the study, methods for fixation
and processing of tissue samples are provided in the section
"Fixation and Processing of Tissue Samples" in the OPTS
Neurotoxicity Screening Battery.
(B) Qualitative Analysis. The purposes of
the qualitative examination are: one, to identify regions within
the nervous system exhibiting evidence of neuropathological
alterations; two, to identify types of neuropathological
alterations resulting from exposure to the test substance; and
three, to determine the range of severity of the
neuropathological alterations. Representative histological
sections from the tissue samples shall be examined
microscopically by an appropriately trained pathologist for
evidence of neuropathological alterations. The following
stepwise procedure is recommended for the qualitative analysis.
First, sections from the high dose group are compared with those
of the control group. If no evidence of neuropathological
37
-------
Devel Neuro
alterations are found in animals of the high dose group, no
further analysis is required. If evidence of neuropathological
alterations are found in the high dose group, then animals from
the intermediate and low dose group are examined. Subject to
professional judgement and the kind of neuropathological
alterations observed, it is recommended that additional methods
such as Bodian's or Bielchowsky's silver methods and/or
immunohistochemistry for glial fibrillary acid protein be used in
conjunction with more standard stains to determine the lowest
dose level at which neuropathological alterations are observed .
Evaluation of postnatal day 11 pups is described in sections (1)
and (2) below. For animals sacrificed at the termination of the
study, the regions to be examined and the types of alterations
that shall be assessed are identified in the section "Qualitative
Examination" in the OPTS Neurotoxicity Screening Battery.
(1) Regions to be Examined. The brains
should be examined for any evidence of treatment-related
neuropathological alterations and adequate samples should be
taken from all major brain regions [e.g., olfactory bulbs,
cerebral cortex, hippocampus, basal ganglia, thalamus,
hypothalamus, midbrain (tectum, tegmentum, and cerebral
peduncles), brainstem and cerebellum] to insure a thorough
examination.
(2) Types of Alterations. Guidance for
neuropathological examination for indications of developmental
insult to the brain can be found in Friede (1975) and Suzuki
(1980) . In addition to more typical kinds of cellular
alterations (e.g., neuronal vacuolation, degeneration, necrosis)
and tissue changes (e.g., astrocytic proliferation, leukocytic
infiltration,•and cystic formation) particular emphasis should be
paid to structural changes indicative of developmental insult
including but not restricted to:
a) gross changes in the size or
shape of brain regions such as alterations in the size of the
cerebral hemispheres or the normal pattern of foliation of the
cerebellum;
b) the death of neuronal
precursors, abnormal proliferation, or abnormal migration, as
indicated by pyknotic cells or ectopic neurons, or gross
alterations in regions with active proliferative and migratory
zones, alterations in transient developmental structures [e.g.,
the external germinal zone of the cerebellum, see Miale and
Sidman (1961) for discussion];
c) abnormal differentiation, while
more apparent with special stains, may also be indicated by
shrunken and malformed cell bodies;
d) evidence of hydrocephalus, in
particular enlargement of the ventricles, stenosis of the
cerebral aqueduct and general thinning of the cerebral
hemispheres.
38
-------
Devel Neuro
(C) Subjective Diagnosis. if any evidence
of neuropathological alterations is found in the qualitative
examination, then a subjective diagnosis will be performed for
the purpose of evaluating dose-response relationships. All
regions of the brain exhibiting any evidence of neuropathological
changes shall be included in this analysis. Sections of each
region from all dose groups will be coded as to treatment and
examined in randomized order. The frequency of each type and the
severity of each lesion will be recorded. After all sections
from all dose groups including all regions have been rated, the
code will be broken and statistical analyses performed to
evaluate dose-response relationships. For each type of dose-
related lesion observed, examples of different ranges of severity
shall be described. The examples will serve to illustrate a
rating scale, such as 1+, 2+, and 3+ for the degree of severity
ranging from very slight to very extensive.
(D) Simple Morphometric Analysis. Since the
disruption of developmental processes is sometimes more clearly
reflected in the rate or extent of growth of particular brain
regions, some form of morphometric analysis shall be performed on
postnatal day 11 and at the termination of the study to assess
the structural development of the brain. At a minimum, this
would consist of a reliable estimate of the thickness of major
layers at representative locations within the neocortex,
hippocampus and cerebellum. For guidance on such measurements
see Rodier and Gramann (1971).
(e) Data collection, reporting, and evaluation. The
following specific information shall be reported:
(!)• Description of test system and test methods. A
description of the general design of the experiment should be
provided. This shall include:
(i) A detailed description of the procedures used
to standardize observations and procedures as well as operational
definitions for scoring observations.
(ii) Positive control data from the laboratory
performing the test that demonstrate the sensitivity of the
procedures being used. These data do not have to be from studies
using prenatal exposures. However, the laboratory must
demonstrate competence in evaluating effects in neonatal animals
perinatally exposed to chemicals and establish test norms for the
appropriate age group.
(iii) Procedures for calibrating and ensuring the
equivalence of devices and the balancing of treatment groups in
testing procedures.
(iv) A short justification explaining any
decisions involving professional judgment.
(2) Results. The following information must be
arranged by each treatment and control group:
(i) In tabular form, data for each animal must be
39
-------
Devel Neuro
provided showing:
(A) Its identification number and the litter
from which it came.
(B) Its body weight and score on each
developmental landmark at each observation time; total session
activity counts and intrasession subtotals on each day measured;
auditory startle response amplitude per session and intrasession
amplitudes on each day measured; appropriate data for each
repeated trial (or session) showing acquisition and retention
scores on the test(s) of learning and memory on each day
measured; time and cause of death (if appropriate), any
neurological signs observed, a list of structures examined as
well as the location(s), nature, frequency, and extent of
lesion(s); and brain weights. Inclusion of photomicrographs
demonstrating typical examples of the type and extent of the
neuropathological alterations observed is recommended. Any
diagnoses derived from neurological signs and lesions, including
naturally-occurring diseases or conditions, should also be
recorded.
(ii) Summary data for each treatment and control
group must include:
(A) The number of animals at the start of the
test.
(B) The body weights of the dams during
gestation and lactation.
(C) Litter size and mean weight at birth.
(D) The number of animals showing each
abnormal sign at each observation time.
(E) The percentage of animals showing each
abnormal sign at each observation time.
(F) The mean and standard deviation for each
continuous end point at each observation time. These will
include body weight, motor activity counts, auditory startle
responses, performance in learning and memory test(s), regional
brain weights and whole brain weights (both absolute and
relative).
(G) The number of animals in which any lesion
was found.
(H) The number of animals affected by each
different type of lesion, the location, frequency and average
grade of each type of lesion for each animal.
(I) The values of all morphometric
measurements made for each animal listed by treatment group.
(3) Evaluation of data. An evaluation of test results
must be made. The evaluation shall include the relationship
between the doses of the test substance and the presence or
absence, incidence, and extent of any neurotoxic effect. The
evaluation shall include appropriate statistical analyses. The
choice of analyses shall consider tests appropriate to the
experimental design and needed adjustments for multiple
40
-------
Devel Neuro
comparisons. The evaluation shall include the relationship, if
any, between observed neuropathological and behavioral
alterations.
(e) References. For additional background information on
this test guideline the following references should be consulted:
Adams, J. , Buelke-Sam, J. , Kimmel, C.A., Nelson, C.J., Reiter,
L.W., Sobotka, T.J., Tilson, H.A., Nelson, B.K. Collaborative
behavioral teratology study: Protocol design and testing
procedure. Neurobehavioral Toxicology and Teratology. 7:579-586
(1985) .
Bennet, H. S., Wyrick, A. D., Lee, S.W., McNeil, J.H. Science
and art in the preparing tissues embedded in plastic for light
microscopy, with special reference to glycol methacrylate, glass
knives and simple stains. Stain Technology 51:71-97 (1976).
Bushnell, P.J. Effects of delay, intertrial interval, delay
behavior and trimethyltin on spatial delayed response in rats.
Neurotoxicologv and Teratology 10:237-244 (1988).
Campbell, B.A., Haroutunian, V. Effects of age on long-term
memory: Retention of fixed interval responding. Journal of
Gerontology 36:338-341 (1981).
Cory-Slechta, D.A., Weiss, B., Cox, C. Delayed behavioral
toxicity of lead with increasing exposure concentration.
Toxicology and Applied Pharmacology 71:342-352 (1983).
Di Sant Agnese, P. A., De Mesy Jensen, K. Dibasic staining of
large epoxy sections and application to surgical pathology-
American Journal of Clinical Pathology 81:25-29 (1984).
Friede, R. L. Developmental Neuropathology. New York: Springer
Verlag (1975).
Green, R.J., Stanton, M.E. Differential ontogeny of working
memory and reference memory in the rat. Behavioral Neuroscience
103:98-105 (1989).
Ison, J.R. Reflex modification as an objective test for sensory
processing following toxicant exposure. Neurobehavioral
Toxicology and Teratology 6:437-445 (1984).
Korenbrot, C.C., Huhtaniemi, I.T., Weiner, R.W. Preputial
separation as an external sign of pubertal development in the
male rat. Biology of Reproduction 17:298-303 (1977).
Krasnegor, N. A., Blass, E. M., Hofer, M. A., Smotherman, W. p.
(eds.) Perinatal Development: A Psychobiological Perspective.
41
-------
Devel Neuro
Orlando: Academic Press (1986).
Kucharski, D., Spear, N. E. Conditioning of aversion to an odor
paired with peripheral shock in the developing rat.
Developmental Psychobioloay 17:465-479 (1984).
Luna, L. G. Manual of Histologic Staining Methods of the Armed
Forces Institute of Pathology. (Third Edition) New York: McGraw
Hill (1968).
Miale, I. E. , Sidman, R. An autoradiographic analysis of
histogenesis in the mouse cerebellum. Experimental Neurology
4:277-296 (1961).
Miller, D. B., Eckerman, D. A. Learning and memory measures. In;
Neurobehavioral Toxicology. Z. Annau (ed). Baltimore: Johns
Hopkins University Press, pp. 94-149 (1986).
Pender, M. P. A simple method for high resolution light
microscopy of nervous tissue. Journal of Neuroscience Methods
15:213-218 (1985)
Ralis, H. M. , Beesley, R. A., Ralis, Z. A. Techniques in
Neurohistology. London: Butterworths (1973)
Rodier, P. M. , Gramann, W. J. Morphologic effects of
interference with cell proliferation in the early fetal period.
Neurobehavioral Toxicology 1:128-135 (1971).
Spear, N. E., Campbell, B. A. (eds.) Ontogeny of Learning and
Memory. New Jersey: Erlbaum (1979).
Spencer, P. S., Schaumburg, H. H. (eds.) Experimental and
Clinical Neurotoxicology. Baltimore: Williams and Wilkins (1980)
Suzuki, K. Special vulnerabilities of the developing nervous
system. In: Experimental and Clinical Neurotoxicology. P. S.
Spencer and H. H. Schaumburg (eds.) Baltimore: Williams and
Wilkins, pp. 48-61 (1980).
US Environmental Protection Agency- Office of Pesticides and
Toxic Substances Neurotoxicity Screening Battery (1990).
42
-------
Devel Neuro
NOTES
Response to Public and SAP Comment
on the
Developmental Neurotoxicity Study
EPA's Work Group on Developmental Neurotoxicology would like
to acknowledge and thank the individuals and organizations that
provided comment on the "Developmental Neurotoxicity Study"
protocol. The Work Group has reviewed all of the comments and
taken them into consideration in revising the protocol.
Responding to every individual comment is beyond the scope of
this effort. Therefore, response will be limited to those
comments that were raised by more than one individual or
organization and which significantly impact the design of the
study.
1. COMMENT: The developmental neurotoxicity protocol is too
complex and should be restricted to those agents for which there
is sufficient -justification to undergo such testing. A simpler,
"tier 1" test should be developed that could be used more
routinely-
EPA RESPONSE; The Agency had a number of discussions on this
issue. At first, the Agency considered development of a two-tier
approach within the confines of a single study design. That is,
measurements' would have been carried out periodically during
postnatal development. The "tier 1" component would have been
the study carried until postnatal day 24 and the "tier 2"
component would have been an extension of "tier 1" into
adulthood. Whether or not the "tier 2" component would have been
conducted would have depended on the analysis of the data up to
postnatal day 24. This proposal was presented at several
scientific meetings. However, it was criticized by the public
for several reasons. First, concerns were raised that
assessments only through the time of weaning may not be
sufficiently sensitive to detect all potential developmental
neurotoxicants. That is, unless assessments were carried out
into adulthood, there would be a possibility that some potential
developmental neurotoxicants would not be identified. Second,
there was a strong sentiment among scientists from industry and
contract laboratories that the assessments from tier 1 would not
be completed in time for a decision to be made as to whether or
not to proceed with the tier 2. In light of these concerns, the
Agency has decided to publish the protocol as a single test to be
conducted in its entirety, as it had been proposed. In the
meantime, the Agency will be considering more feasible approaches
43
-------
Devel Neuro
for development as a "screen" or "tier 1" type of protocol and
encourages the development of screening methodologies by
scientists working in this area.
2.COMMENT: Reference to the highest dose level as inducing
"some overt maternal toxicity" was considered to be excessive.
Some suggested that the highest dose level should be below the
"threshold" for "minimal maternal toxicity."
EPA RESPONSE: The Agency disagrees. The protocol further
qualifies that the highest dose "shall not result in a reduction
in weight gain exceeding 20% during gestation and lactation" and
that it "will not induce in utero or neonatal death or
malformations sufficient to preclude a meaningful evaluation of
neurotoxicity." This represents as minimal a toxic level as one
could require in order to ensure that the agent has been
adequately tested across an appropriate range of dose levels. The
Agency is satisfied with this requirement in the guideline and
does not believe a change is necessary-
3.COMMENT; The specified duration of dosing, that is day 6 of
gestation through day 21 postnatally, is excessive. Dosing of
the dams postnatally should not be required because of potential
effects on maternal behavior, milk production, or milk let-down,
or sequestration of the agent in the milk with transfer to the
pup, any of which may alter maternal-neonatal interactions.
Furthermore, it was noted that pups would be undergoing
observations -and testing while potentially being exposed to the
agent via the milk; thus, alterations in these measurements may
be due to pharmacologic action of the agent rather than to a
neurotoxic effect. It would not be possible to distinguish these
effects.
EPA RESPONSE: After careful consideration of all of the issues,
the Agency has revised the protocol so that the period of
exposure is now from day 6 of gestation through day 10
postnatally- The rationale behind this revision was as follows:
First, the Agency strongly believes that dosing should
continue into the postnatal period for several reasons. These
include: 1) several major events that occur prenatally in the
nervous system of the human are still going through critical
stages postnatally in the rat, and 2) exposure to the still
developing organism may occur when agents are transferred from
the mother to the offspring via the milk. An alternaltive
postnatal exposure route that has been suggested is direct dosing
of neonates, but the Agency believes that more work is needed to
develop better methods before this should be adopted.
Second, although the Agency feels strongly that dosing
44
-------
Devel Neuro
should continue into the postnatal period for the reasons
identified above, it recognizes the potential confounding factors
with postnatal dosing that were raised in the public comments and
has, therefore, adopted a compromise position. The Agency has
modified the duration of exposure so that it ceases on postnatal
day 10. The Agency chose this time for cessation of exposure for
the following reasons: 1) while the revised period of exposure
does not cover the entire period of lactation, it should still be
sufficient to detect any potential effect that may be caused
through exposure via breast milk, 2) any agent with a reasonable
half-life would, theoretically, be eliminated sufficiently by the
time testing begins (day 13 for motor activity) and would not
significantly influence the results, and 3) while the nervous
system is still undergoing some development beyond this time and,
thus, effects on these events may be missed, the majority of the
critical periods for CNS development have occurred by this time,
most notably, proliferation of neuronal precursors in the
cerebellum and hippocampus. Thus, it was felt that dosing through
postnatal day 10 would maximize the detection of most
developmental neurotoxic effects while minimizing the potential
for pharmacologic influences of the agent on the outcome of
functional evaluations.
4. COMMENT: The requirement for oral dosing by intubation is too
rigid. Greater flexibility should be allowed for exposure via
other routes.
EPA RESPONSE; The protocol has been revised to read: "The test
substance or 'vehicle shall be administered orally. Other routes
of administration may be acceptable, on a case-by-case basis,
with ample justification/reasoning for this selection." The
Agency, however, recognizes that conduct via other routes of
exposure may necessitate modifications of the protocol because of
potential problems with postnatal (lactational) exposure, that
may be considered too drastic, and, thus, unacceptable. EPA is
conducting research in this area and encourages others to do so,
as well, in order to address the complications that may arise in
studies conducted via routes other than oral.
5.COMMENT: The frequency of monitoring motor activity is
excessive; furthermore, inclusion of two preweaning evaluations
will separate the pups from the mother for lengths of time that
may be detrimental to the pups.
EPA RESPONSE: In the proposal, EPA specified monitoring of motor
activity on days 13, 17, 21, and day 60 (+ 2). These days were
selected because they represent critical periods of motor
development. Testing over a number of days provides the assessor
45
-------
Devel Neuro
with information regarding the developmental pattern of motor
activity. Interpretation of data from just a single day of
testing would be much more difficult to interpret and less
meaningful than more complete data on the ontogeny of motor
activity and within-session habituation. Furthermore, the Agency
does not have any reason to believe that handling of the pups, or
separation from the mother for the length of time needed to carry
out the motor activity testing, will result in any adverse
effects on the pups; treated and control pups will be handled in
the same manner to avoid any bias in the data.
6.COMMENT: A number of comments were made regarding the
section on neuropathology. A suggestion was made to include
simple morphometrics. In some cases, certain aspects of the
Agency's procedures related to neuropathological evaluation were
questioned. These included: 1) conducting neuropathology on
animals sacrificed on postnatal day 4, 2) the number of animals
included for neuropathologic evaluation, 3) plastic and paraffin
embedding of tissue samples, 4) the qualitative examination, 5)
the list of specified regions of the nervous system to be
examined, 6) the semi-quantitative analysis (subjective
diagnosis), and 7) the reporting of results.
EPA RESPONSE: The Agency has agreed to adopt the suggestion of
including simple morphometrics. Detection of disruption of the
development of the nervous system is the major purpose of this
test. Compounds may alter nervous system development in a
variety of ways including altering the rate and extent of growth
of the nervous system. Alterations of this type are not always
accompanied by gross neuropathological alterations. Thus,
assessment of the extent of potential change in the normal state
of development should be included. The approach will be to make
simple measurements in regions known to be undergoing extensive
growth at the time of sacrifice.
The Agency has formulated the following response to the
questions raised regarding the aforementioned procedures:
1) The protocol has been revised to include
neuropathological examination of animals sacrificed on postnatal
day 11, the day after dosing ends, rather than postnatal day 4.
This revision is based on several advantages to assessing the
later time point. First, cumulative injury should be more
apparent if exposure continues for an additional 6 days
postnatally- Second, by postnatal day 11, several brain regions
are approaching maximal proliferative activity (e.g.,
cerebellum). Third, a greater number of brain regions undergo
further development during the additional 6 days, thus, the
likelihood of revealing potential vulnerability is increased.
Furthermore, brains of postnatal day 11 pups are more amenable to
routine neuropathological analysis.
46
-------
Devel Neuro
2) The requirement of 6 male and 6 female animals per dose
group is to ensure an adequate sample size for statistical
analysis of incidence and severity of neuropathological
alterations as well as the morphometric analysis.
3) The goal of the procedures outlined for the preparation
and processing of tissue samples is to optimally preserve tissue
morphology for microscopic examination. The higher resolution
obtainable in plastic embedded tissue is considered to optimize
the detection of a number of types of lesions, particularly in
the peripheral nervous system. In contrast, paraffin-embedded
material is more amenable to sampling large regions of the
nervous system and is considered optimal for a variety of special
stains that may be useful in characterizing neuropathological
alterations. Several organizations felt the requirement for
separate animals for plastic and paraffin-embedding of tissue
samples was excessive. Furthermore, commentors presented views
that differences in plastic and paraffin techniques did not
require the use of separate animals and that in an appropriate
tiered evaluation scheme both plastic and paraffin-embedding
could be used with tissue samples from the same animals. The
requirement for separate animals is now eliminated, with paraffin
being acceptable and plastic being recommended.
4) Many regions of the developing brain have been
demonstrated to be sensitive to neurotoxic insult. The purpose
of the qualitative examination is to identify regions within the
developing brain that exhibit evidence of neuropathological
alterations and to identify types of .neuropathological
alterations that result from exposure to the test substance. A
stepwise evaluation is recommended since, if evidence of
neuropathological alterations is not observed at the high dose
level, additional processing of tissue samples is not required.
5) The Agency received some comments that the list of
specified regions of the nervous system to be examined was
inadequate, while others felt it was too detailed. Moreover,
comments were received that argued that the list of potential
types of neuropathological alterations also was too restricted.
These lists were intended to serve as guidance. Since they
appear to be subject to misinterpretation, the requirement for a
thorough examination of the nervous system for any evidence of
neuropathological alteration is now explicitly stated. However,
it is noted that examination of the developing brain is
particularly difficult and not usually the subject of routine
neuropathological analysis. Therefore, a list of regions and the
types of major alterations to be evaluated are included. In
addition, a list of all structures examined is required in the
final report. The requirement for examination of more than one
section per region, however, has now been deleted.
6) The purpose of the semi-quantitative analysis is to
evaluate the relationship between the incidence and severity of
the neuropathological alterations and the exposure. Since the
47
-------
Devel Neuro
rating scale is by necessity subjective, it is necessary to
ensure that any bias resulting from the previous qualitative
examination of the tissue samples is minimized. Several
organizations commented that blind evaluation of tissue sections
should not be required. In the initial qualitative analysis, in
which the types of lesions and regions affected are first
identified, blind evaluation is not required. However, in the
semi-quantitative analysis in which the dose-response
relationship is evaluated, it is imperative that the evaluation
be as objective as possible. Moreover, since the semi-
quantitative analysis focuses on a limited number of regions for
lesions previously described in the qualitative analysis, blind
reading is required to ensure objectivity. Thus, it is required
that the subjective rating of the severity and incidence be
performed without knowledge of treatment.
7) The data and analyses supplied in the report must be
evaluated by Agency risk assessors. Thus, the report must be
sufficiently detailed for the Agency to evaluate the quality of
the study. Since no list of regions to be examined is outlined
in the guideline, a list of regions examined must be supplied
with the report. Similarly, an adequate description of lesions
observed must be supplied also. The Agency received comments
that the requirement of photomicrographs to document
neuropathological alterations was extremely costly. The Agency
has decided to recommend, rather than require, the use of
photomicrographs to aid in the description of typical examples of
treatment-related lesions.
7. COMMENT: Additional validation is needed before the glial
fibrillary acidic protein (GFAP) radioimmunoassay should be
included as part of the battery of tests in the developmental
neurotoxicity study. Use of special stains should be at the
discretion of the pathologist conducting the study.
EPA RESPONSE: Although, EPA believes that the GFAP assay has
been shown to be sensitive to the neurotoxic effects of agents in
both the adult and developing nervous systems, this assay has
been deleted at this time. However, the Agency will continue
using this assay experimentally and encourages others to do so,
as well, in an effort to obtain additional validation of its use
as a means to assess the neurotoxic potential of agents. In
addition, if GFAP immunohistochemistry is used as a special stain
in the neuropathology segment of the testing protocol and
evidence of a glial response to toxicant injury is observed,
application of the radioimmunoassay is encouraged in order to
provide objective, quantitative dose-response data.
48
-------
SCHEDULE-CONTROLLED OPERANT BEHAVIOR 1
(a) Purpose. In the assessment and evaluation of the
potential human health effects of substances, it may be necessary
to test for functional neurotoxic effects. Substances that have
been observed to produce neurotoxic signs in other toxicity
studies (e.g., CNS depression or stimulation), as well as
substances with a structural similarity to neurotoxicants
affecting performance, learning, or memory may be appropriate to
evaluate with this test. This guideline defines procedures for
conducting studies of schedule-controlled operant behavior,
one way of evaluating the rate and pattern of a class of learned
behavior (Dews, 1972; NAS 1975, 1977, 1982). Our purpose is to
evaluate the effects of acute and repeated exposures on the rate
and pattern of responding under schedules of reinforcement. Any
observed effects should be evaluated in the context of both the
concordance between functional neurological and neuropathological
effects and with respect to any other toxicological effects seen.
Operant behavior tests may be also used to evaluate many other
aspects of behavior (Laties, 1978). Additional tests may be
necessary to completely assess the effects of any substance on
learning, memory, or behavioral performance.
(b) Definitions.
(1) Neurotoxicitv. Neurotoxicity is any adverse effect
on the structure or function of the nervous system related to
exposure to a chemical substance.
(2) Behavioral toxicitv is any adverse change in the
functioning of the organism with respect to its environment in
relation to exposure to a chemical substance.
(3) Operant. operant behavior, operant conditioning.
An operant is a class of behavioral responses which changes or
operates on the environment in the same way. Operant behavior is
further distinguished as behavior which is modified by its
consequences. Operant conditioning is the experimental procedure
used to modify some class of behavior by reinforcement or
punishment.
(4) Schedule of reinforcement. A schedule of
reinforcement specifies the relation between behavioral responses
and the delivery of reinforcers, such as food or water (Ferster
and Skinner, 1957). For example, a fixed ratio (FR) schedule
requires a fixed number of responses to produce a reinforcer
(e.g., FR 30). Under a fixed interval (FI) schedule, the first
response after a fixed period of time is reinforced (e.g., FI 5
minutes).
(c) Principle of the test method. Experimental animals are
trained to perform under a schedule of reinforcement and
measurements of their operant behavior are made. Several doses
of the test substance are then administered according to the
experimental design (between groups or within subjects) and the
duration of exposure (acute or repeated). Measurements of the
49
-------
Neuro Behav
operant behavior are repeated. For use of this test to study
learning, animals may be trained following exposure. A
descriptive and statistical evaluation of the data is made to
evaluate the nature and extent of any changes in behavior in
relation to exposures to the test substance. Comparisons are
made between any exposures that influence the behavior and
exposures that have neuropathological effects or effects on other
targets of the chemical.
(d) Test Procedures.
(1) Experimental design. These test procedures may be
used to evaluate the behavior of experimental animals receiving
either acute or repeated exposures. For acute exposure studies,
either within-subject or between groups experimental designs may
be used. For repeated exposure studies, between groups designs
should be used, but within subject comparisons (pre-exposure and
post-exposure) are recommended and encouraged.
(2) Animal selection.
(i) Species. For most studies the laboratory
mouse or rat is recommended. Standard strains should be used.
Under some circumstances other species may be recommended.
(ii) Age. Experimental animals should be young
adults. Rats or mice should be at least 14 and 6 weeks old,
respectively, prior to exposure.
(iii) Sex. Approximately equal numbers of male
and female animals are required for each dose level and control
group. Virgin females should be used.
(iv) Experimental history. Animals should be
experimentally and chemically naive.
(3) Number of animals. Six to twelve animals should be
exposed to each level of the test substance and/or control
procedure.
(4) Control groups.
(i) A concurrent control group or control
session(s) (according to the design of the study) are required.
For control groups, subjects shall be treated in the same way as
for an exposure group except that administration of the test
substance is omitted.
(ii) Positive control data from the laboratory
performing the testing shall provide evidence that the
experimental procedures are sensitive to substances known to
affect operant behavior. Both increases and decreases in
response rate should be demonstrated. Data based on acute
exposures will be adequate. Data shall be collected according to
the same experimental design as that proposed for the test
substance. Positive control data shall be collected at the time
of the test study unless the laboratory can demonstrate the
adequacy of historical data for this purpose, i.e., by the
approach outlined in this guideline .
(5) Dose levels and dose selection. At least 3 doses
50
-------
Neuro Behav
shall be used in addition to the vehicle control group (or
sessions for within subject studies). Ideally, the data should
be sufficient to produce a dose-effect curve. We strongly
encourage the use of equally spaced doses and a rationale for
dose selection that will maximally support detection of dose-
effect relations.
(i) Acute studies. The high dose need not be
greater than 2 g/Kg. Otherwise, the high dose shall result in
significant neurotoxic effects or other clearly toxic effects,
but not result in an incidence of fatalities that would preclude
a meaningful evaluation of the data. The middle and low doses
should be fractions of the high dose. The lowest dose shall
produce minimal effects, e.g., an ED10, or alternatively, no
effects.
(ii) Subchronic (and Chronic) Studies. The high
dose need not be greater than Ig/Kg. Otherwise, the high dose
shall result in significant neurotoxic effects or other clearly
toxic effects, but not produce an incidence of fatalities that
would prevent a meaningful evaluation of the data. The middle and
low doses should be fractions of the high dose. The loweset dose
shall produce minimal effects, e.g an ED10, or alternatively, no
effects.
(6) Route of Exposure. Selection of route may be based
on several criteria including, the most likely route of human
exposure, bioavailability, the likelihood of observing effects,
practical difficulties, and the likelihood of producing non-
specific effects. For many materials, it should be recognized
that more than one route of exposure may be important and that
these criteria may conflict with one another. The route that
best meets these criteria should be selected. Dietary feeding
will be generally be acceptable for repeated exposure studies.
(7) Combined protocol. The tests described in this
screening battery may be combined with any other toxicity study,
as long as none of the requirements of either are violated by the
combination.
(8) Study conduct, (i) Apparatus. Behavioral responses
and the delivery of reinforcers shall be controlled and monitored
by automated equipment located so that its operation does not
provide unintended cues or otherwise interfere with the ongoing
behavior. Individual chambers should be sound attenuated to
prevent disruptions of behavior by external noise. The response
manipulanda, feeders, and any stimulus devices should be tested
before each session; these devices should periodically be
calibrated.
(ii) Chamber assignment. Concurrent treatment
groups should be balanced across chambers. Each subject should
be tested in the chamber to which it is initially assigned.
(iii) Schedule of food availability. (A) If a non-
preferred positive reinforcer is used, all subjects should be
placed on a schedule of food availability until they reach a
51
-------
Neuro Behav
fixed percentage e.g., 80 to 90 percent, of their ad libitum body
weight, or kept at a fixed weight and fed after each session.
(B) Subjects must be trained until they
display demonstrable stability in performance across days prior
to exposure. One simple and useful criterion is a minimum number
of sessions on the schedule and no systematic trend during the 5
days before exposure.
(iv) Time, frequency, and duration of testing.
(A) Time of testing. All experimental
animals should be tested at the same time of day and with respect
to the time of exposure. For acute studies, testing should be
performed when effects are estimated to peak, which may be
estimated from data on the functional observational battery,
motor activity, or from pilot studies. For subchronic studies,
subjects should be tested prior to daily exposure in order to
assess cumulative effects.
(B) Frequency of testing. The maintenance of
stable operant behavior normally will require regular and
frequent (e.g., 5 days a week) testing sessions. Animals should
be weighed on each test day-
(C) Duration of testing. Experimental
sessions should be long enough to reasonably see the effects of
exposure, but brief enough to be practical. Under most
circumstances, a session length of 30-40 minutes should be
adequate.
(v) Schedule selection. The schedule of
reinforcement chosen should generate response rates that may
increase or decrease as a function of exposure. Many schedules
of reinforcement can do this: a single schedule maintaining a
moderate response rate; fixed-interval schedules, which engender
a variety of response rates in each interval; or multiple
schedules, where different components may maintain high and low
response rates.
(e) Data reporting and evaluation. The final test report
must include the following information.
(1) Description of equipment and test methods, (i) A
description of the experimental chambers, programming equipment,
data collection devices, and environmental test conditions should
be provided. Procedures for calibrating devices should also be
described.
(ii) A description of the experimental design
including procedures for balancing treatment groups, and the
stability criterion should be provided.
(iii) Positive control data from the laboratory
performing the test that demonstrates the sensitivity of the
schedule used should be provided. Historical data may be used if
all essential aspects of the experimental protocol are the same.
Historical control data can be critical in the interpretation of
study findings. We encourage submission of such data to
52
-------
Neuro Behav
facilitate the rapid and complete review of the significance of
effects seen.
(2) Results, (i) Data for each animal should be
arranged by test group in tabular form including the animal
identification number, body weight, pre-exposure rate and
patterns of responding, changes in response rate and patterns
produced by the chemical, and group data for the same variables,
including standard measures of central tendency and variability
e.g, means and standard deviations, and results of statistical
analyses.
(3) Evaluation of data. (i) The findings should be
evaluated in the context of preceding and/or concurrent toxicity
studies and any correlated functional and histopathological
findings. The evaluation shall include the relationship between
the doses of the test substance and the incidence and magnitude
of any observed effects, i.e. dose-effect curves for any effects
seen.
(ii) The evaluation should include appropriate
statistical analyses. Choice of analyses should consider tests
appropriate to the experimental design, including repeated
measures. There may be many acceptable ways to analyze data.
(iii) Any known citations from the open literature
related to the interpretation of the neurotoxicity of the test
material shall also be included.
(f) References. For additional background information on
this test guideline the following references should be consulted.
Dews, P.B. "Assessing the Effects of Drugs", In Methods in
Psychobiology. Vol. 2, Ed., R.D. Myers (New York: Academic Press,
1972) 83-124.
Ferster, C.B. Skinner, B.F. Schedules of Reinforcement.
(New York: Appleton-Century-Crofts, 1957).
Laties, V.G. "How Operant Conditioning can Contribute to
Behavioral Toxicology". Environmental Health Perspectives. 28:29-
35 (1978).
National Academy of Science. Principles for Evaluating
Chemicals in the Environment. (Washington, DC: National Academy
of Sciences, 1975).
National Academy of Science. Principles and Procedures for
Evaluating the Toxicitv of Household Substances. (Washington,
DC: National Academy of Sciences, 1977) .
National Academy of Science. "Strategies to determine needs
and priorities for toxicitv testing." Appendix 3B. Reference
Protocol Guidelines For Neurobehavioral Toxicity Tests. 2:123-129
(1982) .
53
-------
Neuro Behav
NOTES
1. This guideline has only been modified for clarity of prose and
to make generic changes to conform to other guideline revisions
in sections such as dose selection and route of administration.
Otherwise, it is essentially identical to the guideline
previously published by OTS in the Federal Register and codified
in 40 CFR 798.6500
2. Positive control data need only be generated roughly once
every year as long as most conditions remain the same.
54
-------
PERIPHERAL NERVE FUNCTION1
(a) Purpose. In the assessment and evaluation of the potential
human health effects of substances, it may be necessary to test for
neurophysiological effects. Substances that have been shown to
produce peripheral neuropathy in other neurotoxicity studies (or
other neuropathological changes in peripheral nerves), as well as
substances with a structural similarity to those causing such
effects, may be appropriate to evaluate with this test. This
guideline defines procedures for evaluating certain aspects of the
neurophysiological functioning of peripheral nerves. Our purpose is
to evaluate the effects of exposures on the velocity and amplitude
of conduction of peripheral nerves. Any observed effects should be
evaluated in the context of both the concordance between functional
neurological and neuropathological effects and with respect to any
other toxicological effects seen. Additional tests may be
necessary to completely assess the neurophysiological effects of
any substance.
(b) Definitions.
(1) Neurotoxicity. Neurotoxicity is any adverse effect
on the structure or function of the nervous system related to
exposure to a chemical substance.
(2) Conduction velocity is the speed at which the
compound nerve action potential traverses a nerve.
(3) Amplitude is the voltage excursion recorded during
the process of recording the compound nerve action potential. It
is an indirect measure of the number of axons firing.
(c) Principle of the test method. The test substanc'e is
administered to several groups of experimental animals, one dose
being used per group. The peripheral nerve conduction velocity and
amplitude are assessed using electrophysiological techniques. The
exposure levels at which significant neurotoxic effects are
produced are compared to one another and to those levels that cause
neuropathological effects and/or other toxic effects.
(d) Test Procedures.
(1) Animal selection, (i) Species and strain. Testing
should be performed on a laboratory rodent unless such factors as
the comparative metabolism of the chemical or species sensitivity
to the toxic effects of the test substance, as evidenced by the
results of other studies, dictate otherwise. All animals should
have been laboratory-reared to ensure consistency of diet and
environmental conditions across groups and should be of the same
strain and from the same supplier. If this is not possible, groups
shall be balanced to ensure that differences are not systemically
related to treatment.
(ii) Age and weight. Young adult animals (at least
60 days for rats) must be used. Age (± 15 days for rats must not
55
-------
Periph Neuro
vary across groups. Weights should be within ± 10 percent of the
mean.
(iii) Sex. Either (or both) sex(es) may be used.
Sex must not vary across groups.
(2) Number of animals^ 20 animals should be exposed to
each level of the test substance and/or control procedure. The goal
is to be able to detect a 10 percent change from normal conduction
velocity at the 5 percent level with 90 percent power.
(3) Control groups, (i) A concurrent control group is
required. For control groups, subjects shall be treated in the same
way as for an exposure group except that administration of the test
substance is omitted.
(ii) Positive control data from the laboratory
performing the testing shall provide evidence that the experimental
procedures are sensitive to substances known to affect peripheral
nerve function. Positive control data shall be collected at the
time of the test study unless the laboratory can demonstrate the
adequacy of historical data for this purpose, i.e., by the approach
outlined in this guideline .
(4) Dose levels and dose selection. At least 3 doses
shall be used in addition to the vehicle control group. Ideally,
the data should be sufficient to produce a dose-effect curve. We
strongly encourage the use of equally spaced doses and a rationale
for dose selection that will maximally support detection of dose-
effect relations.
(i) Acute studies. The high dose need not be greater
than 2 g/Kg. Otherwise, the high dose shall result in significant
neurotoxic effects or other clearly toxic effects, but not result
in an incidence of fatalities that would preclude a meaningful
evaluation of the data. The middle and low doses should be
fractions of the high dose. The lowest dose shall produce minimal
effects, e.g., an ED10, or alternatively, no effects.
(ii) Subchronic (and Chronic) Studies. The high dose
need not be greater than Ig/Kg. Otherwise, the high dose shall
result in significant neurotoxic effects or other clearly toxic
effects, but not produce an incidence of fatalities that would
prevent a meaningful evaluation of the data. The middle and low
doses should be fractions of the high dose. The loweset dose shall
produce minimal effects, e.g an ED10, or alternatively, no effects.
(5) Route of administration. Selection of route may be
based on several criteria including, the most likely route of human
exposure, bioavailability, the likelihood of observing effects,
practical difficulties, and the likelihood of producing non-
specific effects. For many materials, it should be recognized that
more than one route of exposure may be important and that these
criteria may conflict with one another. The route that best meets
these criteria should be selected. Dietary feeding will be
generally be acceptable for repeated exposure studies.
(6) Combined protocol. The test described in this
guideline may be combined with any other toxicity study, as long as
56
-------
Periph Neuro
none of the requirements of either are violated by the combination.
(7) Study conduct, (i) Choice of nervefs). The nerve
conduction velocity test must separately asses the properties of
both sensory and motor nerve axons. Either a hind limb (e.g.,
tibial) or tail (e.g., ventral caudal) nerve must be chosen.
Response amplitude may be measured in a mixed nerve.
(ii) Preparation. (A) In vivo testing of
anesthetized animals is required. A barbiturate anesthetic is
appropriate. Care should be taken to ensure that all animals are
administered an equivalent dosage and that the dosage is not
excessive. If dissection is used, extreme caution must be observed
to avoid damage to either the nerve or the immediate vascular
supply.
(B) Both core and nerve temperature must be
monitored and kept constant (±0.5 °C) during the study. Monitoring
of skin temperature is adequate if it can be demonstrated that the
skin temperature reflects the nerve temperature in the preparation
under use. Skin temperature should be monitored with a needle
thermistor at a constant site, the midpoint of the nerve segment to
be tested.
(C) Electrodes. (1) Choice of Electrodes.
Electrodes for stimulation and recording may be made of any
conventional electrode material, such as stainless steel, although
electrodes made of non-polarizing materials are preferable. If
surface electrodes are used, care must be taken to ensure that good
electrical contact is achieved between the electrode and the tissue
surface. Following each application, all electrodes must be
thoroughly cleaned.
(2) Electrode placement. Electrode
placement must be constant with respect to anatomical landmarks
across animals (e.g., a fixed number of millimeters (mm) from the
base of the tail). Distances between electrodes used to calculate
conduction velocity must be measurable to ±0.5 mm. The recording
electrodes should be as far from the stimulating electrodes as
possible. A 40 mm separation is adequate in the caudal tail nerve
of the rat.
(3) Recording conditions. The animal
should be grounded at about the midpoint between the nearest
stimulating and recording electrodes. With the preamplifier set at
its maximal band width, the stimulus artifact should have returned
to baseline before any neural response to be used in the analysis
is recorded.
(D) The electrical stimulator must be isolated from
ground. Biphasic or balanced pair stimuli to reduce polarization
effects are acceptable. A constant current stimulator is preferred
(and required for polarized electrodes) and should operate from
about 10 uA to about 10 mA. If a constant voltage stimulator is
used, it should operate to 250V. All equipment shall be calibrated
with respect to time, voltage, and temperature.
57
-------
Periph Neuro
(E) The recording environment should be enclosed in
a Faraday cage unless electromagnetic field pickup can be shown to
be more than 1.5 times the amplifier baseline noise, under
recording conditions. The recording output should be amplified
sufficiently to render the compound action potentially easily
measurable with an oscilloscope. The amplifier should pass signals
between 2.0 Hz and 4 kHz without more than a 3dB decrement. The
preamplifier must be capacitatively coupled or, if direct coupled
to the first stages, must be able to tolerate any DC potentials
which the electrode-preparation interface produces, and operate
without significant current leakage through the recording
electrodes.
(F) A hard copy must be available for all waveforms
or averaged waveforms from which measurements are derived, and for
all control recording required by this standard. Hard copies must
include a time and voltage calibration signal.
(iii) Procedure. (A) General. Stimulation should occur at
an inter-stimulus interval significantly below the relative
refractory period for the nerve under study. Stimulus intensity
should be increased gradually until the response amplitude no
longer increases. At this point the "maximal" stimulus current is
determined. An intensity 25-50 percent (a fixed value in a given
study) above the maximal intensity so determined should be used for
determining response peak latency and response amplitude. Response
peak latency may be read off the oscilloscope following single
sweeps or determined by an average of a fired number of responsors.
The baseline-to-peak height technique (Daube, 1980) is acceptable
for determination of the nerve compound action potential amplitude,
but in this case, at least 16 responses must be averaged.
(B) Motor nerve. Motor conduction velocity may be
measured from a mixed nerve by recording the muscle action
potential which follows the compound action potential of the nerve.
The stimulus intensity should be adjusted so that the amplitude of
the muscle action potential is supra-maximal. Measurement of the
latency from stimulation to the onset of the compound muscle action
potential gives a measure of the conduction time of the motor nerve
fibers. To calculate the conduction velocity, the nerve must be
stimulated sequentially in two places each with the same cathode-
anode distance, and with the cathode located toward the recording
electrode. The cathode to cathode distance between the two sets of
stimulating electrodes should be divided by the difference between
the two latencies of muscle action potential in order to obtain
conduction velocity. Placement of electrodes shall be described;
site of nerve stimulation may differ from point of entry through
skin.
(C) Sensory nerve. The somatosensory evoked
potential may be used to determine the sensory nerve conduction
velocity in a mixed nerve. The cathode should be placed proximally
at the two stimulation locations with the same cathode-anode
distances. The recording electrodes are placed on the skull. The
58
-------
Periph Neuro
conduction velocity is calculated by dividing the distance between
the two stimulating cathodes by the difference between the two
latencies of the largest primary peak of the somatosensory evoked
potential. Between 64 and 123 responses should be averaged. The
stimulation frequency should be about 0.5 Hz. Stimulus intensity
should be the same as that used for determining the motor
conduction velocity. Should the peak of the somatosensory response
be so broad that it cannot be replicated with an accuracy of less
than 5 percent of the latency difference observed, then a point on
the rising phase of the potential should be chosen, e.g., at a
voltage that is 50 percent of the peak voltage. Alternatively, the
sensory nerve conduction velocity can be obtained from a purely
sensory nerve or from stimulation of the dorsal rootlets of a mixed
nerve, using two recording electrode pairs.
(e) Data collection, reporting and evaluation. The final test
report must include the following information.
(1) Description of equipment and test methods. A
description of the experimental chambers, programming equipment,
data collection devices, and environmental test conditions should
be provided.
(i) A description of the experimental design
including procedures for balancing treatment groups should be
provided.
(ii) Positive control data from the laboratory
performing the test which demonstrate the sensitivity of the
procedure being used should be provided. Historical data may be
used if all essential aspects of the experimental protocol are the
same. Historical control data can be critical in the interpretation
of study findings. We encourage submission of such data to
facilitate the rapid and complete review of the significance of
effects seen.
(iii) Hard copies of waveforms from which
measurements were made as well as control recordings should be
included.
(iv) Voltage and time calibration referable to the
standards of the Bureau of Standards or to other standards of
accuracy sufficient for the measurements used should be included.
(v) Data demonstrating that nerve temperature was
maintained constant throughout the recording period should also be
included.
(2) Results. Data for each animal should be arranged by
test group in tabular form including the animal identification
number, body weight, nerve conduction velocity, and amplitude.
Group summary data should also be reported, including standard
measures of central tendency and variability, e.g., means and
standard deviations, and results of statistical analyses.
(3) Evaluation of data. (i) The findings should be
evaluated in the context of preceding and/or concurrent toxicity
studies and any correlated functional and histopathological
59
-------
Periph Neuro
findings. The evaluation shall include the relationship between the
doses of the test substance and the incidence and magnitude of any
observed effects, i.e. dose-effect curves for any effects seen.
(ii) The evaluation should include appropriate
statistical analyses. Choice of analyses should consider tests
appropriate to the experimental design, including repeated
measures. There may be many acceptable ways to analyze data.
(iii) Any known citations from the open literature
related to the interpretation of the neurotoxicity of the test
material shall also be included.
(f) References. For additional background information on this
test guideline the following references should be consulted:
Aminoff, M.J. (Ed.). Electrodiagnosis in Clinical Neurology.
(New York: Churchill Livingstone, 1980).
Daube, J. "Nerve Conduction Studies," Electrodiagnosis in
Clinical Neurology. Ed. M.J. Aminoff (New York: Churchill
Livingstone, 1980). pp. 229-264.
Glatt, A.F., H.N. Talaat and W.P. Koella "Testing of
peripheral nerve function in chronic experiments in rats,"
Pharmacology and Therapeutics, 5:539-534 (1979).
Johnson, E.W. Practical Electromyography. (Baltimore:
Williams and Wilkins, 1980).
NOTES
1. This guideline has only been modified for clarity of prose and
to make generic changes to conform to other guideline revisions in
sections such as dose selection and route of administration.
Otherwise, it is essentially identical to the guideline previously
published by OTS in the Federal Register and codified in 40 CFR
798.6850.
2. Positive control data need only be generated roughly once every
year as long as most conditions remain the same.
60
-------
61
Assumptions
Oil price
C02 price (fresh)
CO2 price (recycled)
C02 utilization
Well & Equipment Cost
Workovers & Maintenance Cost
S15.00/BBL
$ 0.80/MCP
$ 0.40/MCF
10MCF/BBL
$ 1.00/BBL
$ 1.50/BBL
Based on these assumptions, the following simplified
economics can be generated.
Simple Economics of C02 Flooding ($/BBL)
Case I
Gross Revenue
Less Royalties & Severance
Net Revenue
$15.00
$2.50
$12.50
Case II
$15.00
$ 2.50
$12.50
C02 Related Operating Cost
C02 $ 8.00* $ 6.60**
Wells/Equipment $ 1.00 $ 1.00
Workovers/Maintenance $ 1.50 $ 1.50
Subtotal $10.50 $ 9.10
Operating Profit $ 2.00/BBL $ 3.40/BBL
* 100% fresh C02
** 65% fresh C02, 35% Recycled C02
A project with C02 and operating costs similar to these could be
justified in today's environment with relatively low downside risk and
good upside potential.
FUTURE POTENTIAL
The underlying domestic resource base which could be responsive to CO?
flooding is in excess of 100 billion barrels of residual oil. The
ultimate size of the potential is bounded more by economic criteria and
competition from other EOR techniques than by technical limits.
Carbon dioxide has proven itself to be a highly versatile oil recovery
agent, applicable to low permeability carbonate and dipping sandstones,
useful as a substitute or enhancing agent of a waterflood or displacing
agent of tertiary oil and may be used in either a miscible or immiscible
mode. Generalized technical criteria for determining whether a reservoir
is amenable to C02 flooding are of limited value and have lead to many
potential C02 projects being overlooked. Individual reservoir analysis
is required to establish oil displacement and efficiency. However, the
following values have been established as desirable reservoir properties
for miscible C02 flooding.
-------
62
Oil gravity greater than 25°API
Oil viscosity less than 12 cp.
Depth greater than 2500 feet
Low vertical to horizontal permeability
Multiple isolated and continuous pay intervals
Reservoir dip to promote gravity stable displacement.
It has been demonstrated here and by the continuing interest in EOR that
some C02 flooding can exist in today's environment. It is impossible
to determine when oil prices will return to the point that C02 flooding
will supply major quantities of oil. However, with 100 billion barrels
of oil at stake, the future of C02 flooding is real.
WHAT IS THE NEED FOR ADDITIONAL SOURCES OP CARBON DIOXIDE
The existing sources of C02 can be divided into two categories, 1)
natural sources of C02, and 2) manmade (industrial or by-product). The
estimated reserves for the natural sources are listed below:
Natural Sources of C02
SOURCE RESERVES
1) Sheep Mountain, S.E. Colorado 1 TCP
2) Bravo Dome, N.E. Mew Mexico 6-12 TCP
3) McElmo Dome, S. W. Colorado 10 TCP
4) Jackson Dome, S.H. Mississippi 1- 3 TCP
5) LaBarge-Big Piney, S.W. Wyoming 20 TCP
6) Slanter-Brownfield, Central Utah 4 TCP
Total 42-50 TCP
In many cases the determination of reserves is not applicable to manmade
C02 as the quantity of C02 available is manufactured rather than
produced from natural sources. The following are examples of major
sources of manmade C02.
Sources of Manmade C02
Gas processing plants, eg. Val Verde Basin, Texas
Fertilizer plants, eg. Enid, Oklahoma
Ammonia Plant, Skillington, Louisiana
Coal gasification, eg. Great Plains, N. Dakota
Refinery hydrogenation units, California, Texas, Louisiana
The National Petroleum Council study of enhanced oil recovery estimated
that 5.5 billion barrels of oil could be economically recovered at $30.00
per barrel using current CO2 technology.
In their view, higher oil prices ($50.00 per barrel) would add 2.2
billion and advanced technology would add less than a billion as
tabulated below:
-------
63
NPC Estimates of Economic Oil
Recovery from CC>2 Flooding
(Billions of Barrels)
Nominal Crude
Oil Price Current Advanced
(S/BBL) Technology Technology
30 5.5 6.1
40 7.0 7.8
50 7.7 8.5
Similar studies have estimated the ultimate economic recovery from C02
floods at $30 per barrel to approach 10 billion barrels of oil. The
amount of C02 required to recover this 5.5 billion to 10 billion
barrels of oil would be between 55 TCP and 100 TCP. Comparing this to
the known natural C02 reserves would indicate that between 5 TCP and 50
TCP of manmade C02 would be required at a $ 30 oil price.
SUMMARY
1) Limited C02 flooding can exist and develop under the current
S15-18/BBL environment, however, for large scale C02 flooding to
exist, oil prices will have to rise considerably.
2) The long term outlook for C02 flooding is bright due to the
relative large quantities of additional oil which could be produced
from existing reservoirs and the inevitable long term rise of oil
prices.
3) The known reserves of natural C02 will be insufficient to supply
the long term needs of the C02 industry. Additional sources of
manmade C02 will have to be developed.
BIBLIOGRAPHY
Gill, T. E., "Ten Years of Handling CO2 for SACROC Unit," paper number
SPE 11162, presented in New Orleans, Louisiana, September 26-29, 1982.
Kuuskraa, V. A., "An EOR Status Report on Carbon Dioxide and Nitrogen
Flooding," presented at the Gas EOR Technology and Economics conference,
Houston, Texas, October 27-28, 1986.
National Petroleum Council, "Enhanced Oil Recovery," published, June 1984.
Stalder, J.L., "Responding to Fluctuations in Economic Climate during
Precommitment Design Efforts for a Major EOR Project," presented at the
Gas EOR Technology and Economics Conference, Houston, Texas, October
27-28, 1986.
U. S. Department of Energy, "Target Reservoirs for C02 Miscible
Flooding," published October 1981.
-------
64
-------
Preceding page blank
CURRENT STATUS OF THE USE OF CO2 FOR ENHANCED OIL RECOVERY
J. J. Taber
New Mexico Petroleum Recovery Research Center [A Division of New Mexico Institute
of Mining and Technology], Socorro, New Mexico
ABSTRACT
Carbon dioxide flooding is probably the fastest growing enhanced oil recovery
method in use today with about 100 projects underway or planned. There
are both technical and economic reasons for this steady growth. Large volumes
of CC>2 are now available in some areas at high purity and at pressures needed
for efficient oil recovery. Existing pipelines are already capable of de-
livering large quantities of C02 to reservoirs which respond well to C02
flooding. If pressure requirements are met, the displacement of oil by
CC-2 can be efficient, at least in those areas of the oil reservoirs swept
by CC<2. Although C02 bypasses some oil, and breaks through early at the
production wells, oil is produced effectively for a very long period. For-
tunately, the produced C02 can be separated and recycled efficiently to
achieve good ultimate recovery. In general, as CC*2 floods in the Permian
Basin of west Texas and eastern New Mexico mature, it appears that the net
oil recovery will be even better than predicted. The technology of C(>2
flooding is still evolving, and the economics depend strongly on crude oil
prices. The current status of CC>2 flooding is described with reference
to specific field results. Information on all of the major C(>2 floods in
the United States is given, and the projects are located on maps.
INTRODUCTION
At the 1985 and 1986 International Energy Agency (IEA) Workshops on EOR
in Tokyo, Japan, and Hannover, Germany, it was reported that gas injection
was one of the faster growing enhanced recovery methods, and that C02 flooding
was becoming the most important gas injection method in the United States. *•»2
These facts are still true today. However, even though the number of gas
injection projects is increasing steadily, the rate of increase in C02 flooding
is leveling off for the first time since 1980. Fig. 1 shows the trends
for all of the active enhanced recovery projects in the United States, and
Fig. 2 shows the gas injection projects. It can be seen from Fig. 2 that
the number of C02 projects is still increasing at a rate of about eight
projects per year even though the recent drop in oil prices will probably
slow this rate of increase. However, the most recent Oil and Gas Journal
survey on EOR reports that 42 new C02 projects are planned to start before
the end of 19883, so the increase in oil recovery from C02 floods will certainly
continue until well into the next century.
Preceding page blank
-------
66
-------
67
The economic and technical reasons for the growth in C02 flooding have been
reported by a number of authors.**-ll ^t present, reservoir characteristics
and the availability of C(>2 favor three general areas in the United States
which are shown in Fig. 3. These are the Permian Basin of west Texas and
eastern New Mexico, Mississippi and the Gulf Coast Area, and the Wyoming-
Colorado-Utah area.
COg PRODUCING AREAS
CC-2 FLOOD AREAS
PIPELINES
Figure 3. General Locations of C(>2 Flooding Areas
and Natural Sources in the united States.
Current Sources of CO? for Enhanced Oil Recovery
The most widely publicized of the active CC>2 flooding areas has been the
Permian Basin because of the availability of CC>2 within reasonable pipeline
distance of the oil reservoirs. Recently, however, the La Barge-Big Piney
area of western Wyoming has received attention. It has been reported that
there are 20-25 trillion cubic feet (TCF) of C02 in the Wyoming sources.
This approximately equals the combined sources of Sheep Mountain, Colorado
(about 2 TCF), the McElmo Dome of southwest Colorado (about 10 TCF), and
the Bravo Dome of northeastern New Mexico (also containing 10 TCF) of C02
reserves. Thus, in the Rocky Mountain area, there are in excess of 45 TCF
of C02 available from natural sources. While the Wyoming C02 resource con-
sists of approximately 657. C02 plus methane and other hydrocarbons of low
molecular weight, the Permian Basin sources (see Fig. 4) produce an injection
gas which has a much higher concentration of C02- The McElmo Dome and Sheep
Mountain sources in southern Colorado contain small amounts of hydrocarbons,
whereas the Bravo Dome reservoir in northeastern New Mexico contains C02
of more than 997. purity.
-------
68
Figure 4. C02 Pipelines Which Supply the Permian Basin.
Fig. 4 illustrates three pipelines which deliver C(>2 to the carbonate reser-
voirs of the Permian Basin. Because these pipelines deliver CC>2 to the
oilfields at a reasonable cost, half of the 53 miscible projects listed
in Table 1-M are in this Permian Basin area. The current capacities of
the pipelines are as follows: Sheep Mountain (completed in 1983) 500 million
cubic feet (MMCF) per day; the Cortez Pipeline (completed in 1984) 650 MMCF
per day, and is capable of almost one billion cubic feet (BCF) per day;
the Bravo Pipeline (completed in 1985) from the Bravo Dome to the Permian
Basin is capable of delivering from 400-700 MMCF per day. The National
Petroleum Council (NPC)8 forecasts that the Rocky Mountain and Permian Basin
areas would require about three BCF of C0£ per day to reach the enhanced
recovery target of 500,000 to 600,000 barrels of oil per day by the year
2000. These forecasts appear to be quite reasonable; the current CO2 pipeline
capacity will be close to three BCF per day when the Wyoming-Colorado and
other pipelines which are now in the final planning or construction stages
are completed. The Exxon-Chevron pipeline (200 MMCF/D) which supplies the
Rangely, Colorado, C0£ flood is illustrated in Fig. 4a.
In addition to the pipelines which supply C(>2 for EOR in the Rocky Mountain
and Permian Basin areas, Fig. 5 shows the Choctaw Pipeline which is, or
will be servicing, several C(>2 floods in Mississippi and Louisiana.^2 The
Jackson Dome C(>2 source, northeast of Jackson, Mississippi, has several
deep reservoirs (14,000-16,000 feet) which contain 6 TCF of high-purity
C02- The reservoirs closest to the Jackson Dome igneous intrusion contain
C02 of 99* purity. The purity falls off (more light hydrocarbons are present)
at distances greater than 25 miles northeast of the Jackson Dome. The Miss-
issippi section of the Choctaw Pipeline (Fig. 5) is completed and supplying
C02 at high pressures to the expanding Little Creek CC<2 flood. Until it
crosses the Mississippi River, it is a 20 inch, carbon steel pipeline; from
there to Weeks Island, it is a 10 inch line, and this section should be
completed by the fall of 1987. In addition to the ongoing C(>2 floods at
-------
69
Little Creek and Weeks Island, Shell is planning C(>2 recovery projects for
the Mallalieu, Olive, and White Castle fields shown on Fig. 5.^2
B
EVANSTON
CO2 SOURCE
SHUTE CREEK PLANT
C02 DEUVERY POINT
ROCK SPRINGS
FERTILIZER PLANT
96 MILE 10"
SLURRY
VERNAL
CO, P/L •+-
126 MILE 16"
"EBuauao
RANGELY RELD
D
• GRAND JUNCTION
Figure 4a. C02 Supply for Rangely Field in Colorado.
(Courtesy of S.L. Walker, Chevron, USA, Inc.)
JACKSON DOME
COj SOURCE
CHOCTAW
PIPELINE
Mollolieu
Little Creek
Figure 5. The Choctaw C(>2 Pipeline for Enhanced Oil
Recovery Projects in Mississippi and Louisiana.
(After Reference 12.)
A few C02 floods are supplied by C02 from industrial sources. For example,
East Velraa and Northeast Purdy (project nos. 6 and 25 in Table 1-M) are
supplied by a fertilizer plant, while Lick Creek (project no. 7 in Table
1-IM) receives C02 from an ammonia plant.
-------
Table 1-M. Miscible C02 Projects in the USA
(Fields A-N)
Project
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Plaid HUM
Alvord South Field
Bay St. Elaine '
Crossett
Dilllnger Ranch
Dollarhide
East Velma
Farnsworth, North
Ford Gerald Ina
Gar be r
Granny1* Creek
Greater Aneth
Kurten
Level land Unit
Level land (Hint Test)
Little Creek Pilot
Little Creek Field
Little Knife
Haljamar Pilot
MCA Unit
HcElmo Creek Unit
HcElroy
Head Strawn
Means (San Andres)
North Coles Levee
North Cowden Unit
Northeast Purdy
State
Tex.
La.
Tex.
Uyo.
Tex.
Okla.
Tex.
Tex.
Okie.
W.Va.
Utah
Tex.
Tex.
Tex.
Miss.
Hiss.
N.Dak.
N.N.
Utah
Tex.
Tex.
Tex.
Calif.
Tex.
Okla.
County
Wise
Terre bonne
Crane & Upton
Campbell
Andrews
Stephens
Ochiltree
Reeves &
Culberson
Garfleld
Clay
San Juan
Brazos
Hock ley
Hockley
Lincoln & Pike
Lincoln & Pike
Billings
Lea
San Juan
Upton
Jones
Andrews
Kern
EC tor
Garvin
Operator
Mitchell Energy
Texaco
Shell Western E&P
Tenneco
Unocal
Arco
Dorchester
Enhanced Co.
Conoco
Arco
Columbia Gas '
Transmission
Superior
Chevron
Amoco
Amoco
Shell
Shell Western E&P
Gulf
Conoco
Mobil
Southland Royalty
Union
Exxon
Arco
Amoco
Cities Service
Start
date
1980
1/81
4/72
10/80
5/85
1983
6/80
2/81
10/81
6/76
1982
8/81
3/73
8/78
2/74
12/85
1/81
5/83
2/85
2/81
12/64
11/83
6/81
2/79
9/82
Area,
Number
Nells
acres Prod.
•••••••Ml
2,291
9
1,500
600
6,183
1,472
3,850
80
7
13,357
672
13
1.5
31
8.200
5
5
13,440
640
43
6,700
70
12
8.320
245
2
23
20
62
8
198
9
4
140
5
2
1
3
110
4
4
170
38
3
248
8
2
106
InJ.
mmmmmmi
10
i
11
10
43
6
123
4
1
21
4
6
4
1
40
1
1
100
20
4
176
3
6
102
Pay zone
Caddo
8000-Foot
Devonian
Hinnelusa Sand
Devonian
Sims
Marmaton "B"
Delaware
Crews
Pocono Big Injun
Aneth
Woodbine
San Andres
San Andres
Tuscaloosa
Lower Tuscaloosa
Madison Canyon
Zone D
Grayburg/
San Andres
Ismay Desert
Creek
San Andres
Strawn
San Andres
Stevens
Grayburg
Springer
Permea-
Porosity blllty
Lithology
Congl.
S
Trlpolltlc
chert
S
Dolo./
Trlpollte
S
LS
S
S
S
LS
S
Dolo.
Dolo.
S
S
L/Dolo.
S/LS
LS
Dolo.
S
LS
S
Dolo/LS
S
Z
12.8
32.9
22.0
13
13.5
17
12
23.0
19.0
16
10
12.0
11.5
11.8
23
23.0
18
11.0
14.0
11.6
11
9.0
19.5
10.0
13.0
md.
55
1,480
5
10-100
17
70
41
64
12
7
0
4
4
33
33
22
18
5
2
12
20
9
5
44
Abbreviations: JS - Just started
HF - Half finished
NC • Near completion
Term. • Terminated
TETT " Too early to tell
Prom. • Promising
Succ. • Successful
Disc. • Discouraging
-------
Table 1-M. Mlscible C02 Projects in the USA (cont'd.)
(Fields A-N -- data continued)
Project
No. Field Nane
1
2
3
4
5
6
7
8
9
10
11
12
13
1*
IS
16
17
18
19
20
21
22
23
24
25
Alvord South Field
Bay St. Elaine
Crossett
Oil linger Ranch
Dollarhlde
East VelM
Farnsworth, North
Ford Cera Id In*
Garber
Cranny '• Creek
Greater Aneth
Kurten
U veil and Unit
Level land (Hint Teat)
Little Creek Pilot
Little Creek Field
Little Knife
Haljamar Pilot
HCA Unit
NcElmo Creek Unit
HcElroy
Mead Strawn
Means (San Andres)
North Coles Levee
North Couden Unit
Northeast Purdy
Reservoir oil
Depth API Viscosity
ft. Gravity cp 9 F
5,700
7.400
5.300
9.000
8.000
6.500
2.680
1.900
2,000
5.750
8.300
4.900
4.900
10.700
10.640
9,800
3.665
5,600
3.850
4,475
4,300
9,000
4,300
9,400
44.0
36.0
44.0
37.0
40.0
26.0
39
40.0
44.0
45.0
42.0
38.0
30.0
30.0
39
38.0
43.0
36.0
41.0
31.0
41
29.0
36.0
34.0
38.0
0.39
0.67
0.36
0.86
0.44
2.50
1.61
1.40
1
3.14
0.47
0.40
2.30
2.30
0.40
0.40
0.20
0.80
0.50
2.30
1.30
6.00
0.45
1.67
1.20
154
170
106
230
122
131
83
100
75
135
230
105
105
248
248
240
90
125
86
135
97
235
94
148
Residual oil
Previous saturation X Project
Prodn. Start End maturity
WF
Prim.
Gas InJ'n
Prim/HP
Prlm/WF
Prlra/WF
Prim.
Prln/WF
WF
Prl«/WF
Prla/HF
Prim.
WF
WF
Prl»/WF
WF
Prln.
Prln/WF
Prln/WF
WF
Prln.
WF
Prln.
WF
WF
60.0
20.0
34.0
35.0
56
30.0
43
40.0
74.0
43.0
21
21.0
50.0
39
34.0
46
46.0
45.0 HF
5.0 NC
22.0 HF
HF
22.0 JS
JS
HF
JS
16.0 NC
Tern.
JS
HF
HF
Tern.
2.0 JS
NC
JS
Tern.
Tern.
JS
25.8 HF
JS
40.0 NC
Total Enhanced
prodn. prodn.
bo/d bo/d
680
7
2.000
120
1.900
395
35
160
47
9
3,300
50
5.600
150
22
3,500
200
7
2.000
60
395
120
21
6
3,300
SO
950
Project
eval.
Pron.
Disc.
Succ.
TETT
Succ.
Succ.
TETT
TETT
Succ.
Pron.
TETT
Succ.
Disc.
TETT
TETT
Profit
No
Yes
TETT
No
Yes
TETT
TETT
No
TBTT
-------
Table 1-M. Miscible C02 Projects in the USA (cont'd.)
(Fields P-W)
Project
No. Field Name.
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
Paradls (No. 8)
Par ad is
Plttsburg
Quarantine Bay
Rangely
Rankln
Raymond
Rock Creek
Rose City North
Sable
SACROC Unit
Seminole
Sho-Vel-Tum
Slaughter Estate
Slaughter (Fraeier
Unit)
Slaughter - (Central
Mallet Unit)
S. Bishop Ranch (9200')
S. Bishop Ranch (9400')
Tinsley
Twofreds
University Waddell
Vacuum
Uasson (Denver Unit)
Wasson (ODC Unit)
Weeks Island Field
Welch
Wellman
West Sussex Unit
State
La.
La.
Tex.
La.
Colo.
Tex.
Hont.
W.Va.
Tex.
Tex.
Tex.
Tex.
Ok la.
Tex.
Tex.
Tex.
Wyo.
Wyo.
Hiss.
Tex.
Tex.
N.H.
Tex.
Tex.
La.
Tex.
Tex.
Wyo.
County
St. Charles
St. Charles
Camp
Plaquemlnes
Rio Blanco
Harris
Sheridan
Roane
Orange
Yoakum
Scurry
Gaines
Stephens
Hockley
Hockley
Hockley
Campbell
Campbell
Yazoo
Loving, Ward
Reeve*
Crane
Lea
Yoakum
Yoakum
New Iberia
Dawson
Terry
Johnson
Start
Operator d.ile
Texaco ' 2/82
Texaco 2/82
Chevron 6/85
Chevron 10/81
Chevron 10/86
Petromac Inc. 1/81
Santa Fe Energy 8/83
Pennzoil 11/76
Highland Resource 4/81
Arco 3/84
Chevron 1/72
Amerada Hess 4/83
Arco 9/82
Amoco 11/72
Amoco 12/84
Amoco 12/84
Grace Petroleum 1/82
Grace Petroleum 1/82
Pennzoil 11/81
t HNG Fossil Fuels 1/74
Chevron 5/83
Phillips 2/81
Shell Western EiP 4/83
Amoco 12/84
Shell Western E&P 1979
Cities Service 2/82
Union Texas Petro 7/82
Conoco 12/82
Number
Area. Wells
acres Prod. Inj.
320
347
43
57
20,000
80
685
20
800
825
49,900
15,700
1.100
12
1,600
5,700
640
1.280
1,338
4,392
920
4,900
20,000
7,800
B
2,675
1,400
10
7
12
4
4
360
6
2
2
9
31
887
328
65
2
64
325
5
5
21
42
50
237
840
316
2
129
29
3
3
4
1
1
360
1
1
6
5
11
379
133
43
6
37
73
2
4
33
13
97
280
250
1
132
2
1
Pay zone
No. 8
Lower 9000-Foot
Pittsburg
4 Sand Reservoir
Weber
Ycgua
Nlsku
Big Injun
llack.be try
San Andres
Canyon Reef
San Andres
Sims
San Andres
San Andres
San Andres
Hinnclusa
Hinnelusa
Perry
Delaware
Devonian
San Andres
San Andres
San Andres
S RES B
San Andres
Wolf camp
Shannon
Porosity
Lithology 7.
S
S
Limey Sand
S
S
S
LS
S
S
Dolo.
LS
LS
S
Dolo.
Dolo/LS
Dolo/LS
S
S
S
S
Dolo.
Dolo.
Dolo.
Dolo/LS
S
LS
LS
S
27.0
26.0
11.0
30.0
15
27.0
8.2
22
37.0
8.4
3.9
13
16.0
10.5
10.0
13.0
16.0
15.0
26.4
19.5
12.0
11.7
12.0
9.0
26.0
9.3
4.2
19.5
Permea-
bility
md.
795
770
2
100-1.000
20
300
13
20
4,500
2
19
70
4
4
7
50
150
49
32
14
11
8
5
1,800
9
100
121
K)
-------
Table 1-M. Miscible C02 Projects in the USA (cont'd.)
(Fields P-W -- data continued)
Project
No.
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
Field Name
Paradls (No. 8)
Paradis
Pittsburg
Quarantine Bay
Rangely
Rank in
Raymond
Rock Creek
Rose City North
Sable
SACROC Unit
Semlnole
Sho-Vel-Tum
Slaughter Estate
Slaughter (Frailer
Unit)
Slaughter - (Central
Mallet Unit)
S. Bishop Ranch (9200'
S. Bishop Ranch (9400*
Tinsley
Tvofreds
University Waddell
Vacuum
Wasson (Denver Unit)
Wasson (ODC Unit)
Weeks Island Field
Welch
Wellman
West Sussex Unit
Depth
Reservoir oil
API Viscosity
ft. Gravity cp
8,600
10.400
8.000
8.120
6,000
7.900
7,900
2,000
8,200
5,200
6.700
5.300
6.200
4,950
4,950
4,950
) 9,200
) 9,400
4.800
4.900
8.500
4,500
5,200
5.100
12.760
4,890
9,800
3.040
39.0
37.0
41.0
32.0
34.0
37.0
40.0
43
37.0
32.0
41.0
35
25.0
27.0
31.0
31.0
35.0
34.0
39.0
36.0
43.0
38.0
33.0
32.0
32.0
34.0
43.5
38.0
0.40
0.50
0.99
0.60
0.40
3.20
2.00
1.46
0.35
1.70
3.30
2.00
1.40
1.40
1.14
1.50
1.50
0.45
1
1.30
1.30
0.50
2.15
0.54
1.70
9 r
192
205
205
183
160
192
178
73
180
107
130
105
115
105
105
105
220
180
175
105
140
101
105
110
225
96
151
93
Previous
Prodn.
Prim.
Prim.
WF
Prim.
WF
WF
Prim.
Prim/
WF
WF
Prim/WF
Prlm/WF
WF
WF
WF
WF
WF
WF
WF
WF
WF
Prim.
WF
WF
WF
WF
WF
Residual oil
saturation Z
Start
3.0
62.0
55.0
36.0
55.0
50.0
55.0
54
59.0
61.0
42.0
65.0
65.0
71.0
70.0
40.0
45.0
22.0
30.0
35.0
End
2.0
48.0
25.0
35.0
25.0
42.0
38.0
50.0
27.0
2.0
18.0
10.0
Project
maturity
HF
HF
NC
HF
JS
JS
JS
NC
HF
JS
NC
JS
NC
JS
JS
JS
JS
HF
HF
JS
JS
JS
JS
NC
HF
JS
Term.
Total
prodn.
bold
400
575
237
85
30.000
80
86
460
750
43.863
41.800
2.500
28
3,000
7.500
400
892
2,000
12.400
46.000
13.000
160
3.100
7,000
79
Enhanced
prodn.
bo/d
400
575
105
85
80
46
460
16.000
750
28
400
892
70
80
1.200
160
300
79
Project
eval.
Prom.
Prom.
Succ.
Prom.
TETT
Prom.
Prom.
Prom.
TETT
Succ.
Prom.
Prom.
Succ.
Prom.
Prom.
Prom:
TBTT
TETT
Succ.
Prom.
Succ.
Succ.
TETT
Profit
mmmmmmm
No
No
Yes
No
TETT
No
Yes
No
TETT
Yes
Yes
No
Yes
TETT
Yes
No
No
TETT
-------
Table 1-IM. Immiscible C(>2 Projects in the USA
Project
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
Field
Bayou Sale
Cote Blanche Bay
West
East Coyota-Hualde
Dome Unit*
Heidelberg
Ufltte
Lake Barre
Lick Creek
State
La.
La.
Calif.
Miss.
La.
La.
Ark.
Magnet Withers BH&S STex.
State Tract*
Hagnet Withers
Fierce Estates BtC
Manvel
Manvel
Paradis
Pewit t Ranch
Plcfcett Ridge
Pierce Ranch
Plttsburg
Plymouth
Sho-Vel-Tuai
South Marsh Island
Block 6
Talco
Thompson
West Columbia
West Delta Block 109
West Delta Block 109
West Delta Block 109
West Delta Block 109
Wilmington
Wilmington
Withers North
Wither* North
Yates
Tex.
Tex.
Tex.
La.
Tex.
Tex.
Tex.
Tex.
Tex.
Ok la.
La.
Tex.
Tex.
Tex.
La.
La.
La.
La.
Calif.
Calif.
Tex.
Tex.
Tex.
County
St. Mary
St. Mary
Orange
Jasper
Jefferson
Terre bonne
Bradley-Union
Wharton
Wharton
'
Brazorla
Bracorla
St. Charles
Titus
Wharton
Wharton
Camp
San Patrlclo
Stevens
Offshore
Franklin
Fort Bend
Brazoria
Offshore
Offshore
Offshore
Offshore
Los Angeles
Los Angeles
Wharton
Wharton
Pecos & Crockett
Operator
Texaco
Texaco
Unocal
Chevron
Texaco
Texaco
Phillips
Texaco
Texaco
Texaco
Texaco
Texaco
Exxon
Texaco
Texaco
Chevron
Texaco
Texaco
Texaco
Texaco
Texaco
Texaco
Texaco
Texaco
Texaco
Texaco
Champ 1 in
Champ 1 In
Texaco
Texaco
Marathon Oil
Start
date
5/84
3/84
6/82
12/83
8/84
3/84
1/76
10/83
7/83
11/83
10/82
3/84
6/83
5/83
1/83
11/83
10/83
5/83
6/85
5/82
1/83
6/83
6/85
6/85
6/85
6/85
2/84
3/81
3/83
5/83
11/85
Area,
Number
Wells
acres Prodn. Inj .
564
55
766
40
271
1.164
1,640
1,224
500
128
43
110
726
480
120
380
120
82
240
100
33
48
74
75
78
156
41
454
768
14,300
5
3
92
1
5
12
39
2
1
3
2
2
1
1
1
1
10
3
895
2
3
21
2
4
13
2
1
1
1
1
1
12
4
13
Pay zone
St. Mary
14 Sand
1.2,3 Anaheim
But aw
8900-Foot
Upper M
Meakln
Magnet Withers
Oligocene
Oakville
Main Pay
Paluxy
Pickett Ridge
Pierce Ranch
Sub-Clarksville
Main Greta
Deese
Rob E-S
Paluxy
Frio
PTSD
10200-Foot
10800-Foot
13100-Foot
12500-Foot
Tar
Tar
Withers N.
C-Sand
Grayburg/
San Andres
Lithology
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
Dolo.
Porosity
*
31.0
29.0
26.0
25.0
27.0
25.0
30.3
23.0
23.0
30.0
30.0
28.0
24.0
30.0
31.8
23.0
31.0
13.0
29.6
25.0
27.0
30.0
27.0
30.0
29.7
27.0
24.0
24.0
25.0
25.0
17.0
Permea-
bility
md.
500
322
400
74
250
139
1,200
1.700
1,700
1,000
400
1,033
1000-1500
1.200
534
460
350
100
323
388
100-1,000
560
205
1,900
68
1.032
465
465
1.050
400
175
-------
Table 1-IM. Immiscible C02 Projects in the USA (cont'd.)
Project
No.
1
2
Reservoir oil
Field
Bayou Sale
Cote Blanche Bay
Depth API Viscosity
ft. Gravity cp 9 F
10,000 34.0 0.4
8,000 32.0 1.3
194
184
Previous
Prodn.
Prim.
WF
Residual oil
saturation Z
Start
50.0
50.0
End
45.0
45.0
Project
maturity
JS
JS
Total
prodn .
bo/d
2.000
80
Enhanced
prodn.
bo/d
30
10
Project
eval.
Succ.
Prom.
Profit
Yes
Yes
West
3 East Coyote-Hualde
Dome Units
4 Heidelberg
5 Lafitte
6 Lake Barre
7 Lick Creek
8 Magnet Wither* BH&S
State Tracts
9 Magnet Withers
Pierce Estates B&C
10 Manvel
11 Manvel
12 Paradls
13 Pewitt Ranch
14 Plckett Ridge
15 Pierce Ranch
16 Plttsburg
17 Plymouth
18 Sho-Vel-Tun
19 South Marsh Island
Block 6
20 Talco
21 Thompson
22 West Columbia
23 West Delta Block
24 West Delta Block
25 West Delta Block
26 West Delta Block
27 Wilmington
28 Wilmington
29 Withers North
30 Withers North
31 Yates
3300-460 23.0
12
130 WF
Tern.
1,600
Disc.
No
IS
>d
109
109
109
109
5,060
8,900
13.000
1100-170
5,500
5.500
5.000
4.000
10.200
4,500
4,600
4,900
3,800
4.650
5.530
11,200
3.785
5,100
2.600
10,000
10,500
11,325
12.000
2.500
2,500
5,250
5,320
1100-170
20.0
34.0
33.0
17.0
26.0
26.0
26.7
25.0
38.0
19.0
25.0
24.4
14.0
23.3
24.0
34.0
23.0
25.2
30.0
37.3
36.5
37.0
34.1
14.0
14.0
25.7
25.3
30.0
15
0.7
0.4
160
2.3
2.3
7.2
4.4
0.5
30
2.5
4.57
2.200
3.19
18
0.3
25
2.7
8
0.26
0.28
0.3
0.25
283
283
2.45
2.9
5.5
150
185
236
118
154
154
149
149
195
160
138
155
120
150
129
208
147
120
116
194
200
210
218
123
123
145
147
82
Prim.
Prim.
WF
Prim.
Gas inj'n
Gas Inj'n
Prim.
Prim.
WF
Prim.
Prim.
Prim.
Prim.
Prim..
Prim.
Prim.
Prim.
Prim.
Prim.
Prim.
Prim.
Prim.
Prim.
WF
WF
Prim.
WF
Gas Inj'n
50.0
50.0
55.0
35.0
35.0
45.0
42-65
50.0
29.0
48.0
31.5
49.0
50.0
50.0
42-65
36-48
50.0
50.0
50.0
50.0
51.0
51.0
35.0
32.0
45.0
45.0
46.0
31.0
34.0
20.0
Variable
45.0
28.0
18.0
20.0
40.0
45.0
45.0
Variable
10-20
45.0
45.0
45.0
45.0
30,0
30.0
32.0
30.0
HF
JS
JS
NC
JS
JS
JS
JS
JS
JS
HF
HF
JS
Term.
Term.
JS
Tern.
JS
JS
JS
JS
JS
JS
JS
HF
HF
Term.
JS
400
650
450
563
200
550
127
250
112
429
ISO
12
220
462
93
290
3.000
2.200
500
1.600
375
170
300
98
97.341
30
50
400
5
17
30
15
1
8
100
3
9
10
300
10
375
170
4
4
Succ.
Succ.
Succ.
Succ.
Succ.
Succ."
Succ.
Disc.
No
Prom.
Succ.
Disc.
Prom.
Prom.
Disc.
Succ.
Succ.
Disc.
TETT
Succ.
Prom.
Disc.
TETT
Yea
Yes
Yes
Yes
Yes
Yes
Ye«
No
No
Yes
Yes
No
Yes
Yes
Yes
Yes
es
Yes
TETT
No
Yes
No
TETT
-------
76
C02 FLOODS IN THE UNITED STATES OF AMERICA
Tables 1-M (miscible) and 1-IM (immiscible) list the best known C02 floods
in the United States. The data in Table 1-M and 1-IM have been drawn from
many sources. Most information has come from the 1986 Oil and Gas Journal
survey,^ or from a recent compilation of C0£ field and laboratory projects. *••*
Original literature references to many of the field projects are given in
reference 13. Most of the 84 projects in the combined table are producing
oil today by (X>2 injection. A few terminated C02 projects are still listed
because they have been cited so often in the literature that they are useful
for comparison purposes.
The locations of the 85 C02 floods are given in Figs. 6, 7, and 8. The
numbers inside the circles correspond to the sequence number of each project
in Tables I-M and I-IM, which are arranged alphabetically by field. Thus,
any C02-EOR project can be found quickly by field name in the tables and
then located on one of the maps. Fig. 6 gives the location for all miscible
(X>2 floods except those in the Permian Basin of New Mexico and Texas. These
are located more precisely (note that the counties are shown) in the detailed
map of that area, Fig. 7. The immiscible floods are located on Fig. 8 which
shows that they are concentrated primarily in the Gulf Coast area because
of the nature of the reservoirs, or in California because the oil is too
heavy for a miscible flood (see following sections). A study of the general
map (Fig. 3) along with the actual project locations in Figs. 5-8 shows
that C02 flooding, except for the immiscible individual well projects, is
developing fastest in those areas which have a good natural source of C02-
Figure 6. Locations of Miscible C02 Floods in the
United States.
-------
77
Figure 7. Locations of Miscible C02 Floods in the
Permian Basin of New Mexico and West Texas.
Figure 8. Locations of Immiscible C02 Projects.
-------
78
Miscible vs. Immiscible Projects
The combined Table 1 is divided into 1-M and 1-IM to draw a distinction
between the miscible (M) and immiscible (IM) types of projects. The differ-
ence between miscible displacement and immiscible displacement by C02 or
other soluble gases has been considered in several publications.l'1^"16
The usual way to distinguish between the two types of projects is to observe
the oil recovery at different pressures in a slim tube test. The general
shape of the oil recovery curve by C(>2 in a slim tube displacement is given
in Fig. 9. As discussed by many authors, true, first-contact misciblity
between C0£ and common crude oils is never achieved. However, excellent
oil recovery is obtained from the idealized porous medium of a slim tube
by multiple contact miscibility, as long as the pressure is high enough.
This pressure at which excellent recovery is obtained, and beyond which,
only insignificant increases occur with added pressure, is called the minimum
miscibility pressure or MMP. Normally this occurs at about 95% oil recovery.
Obviously, there is a large region (shown in Fig. 9) where oil recovery
is significant (and usually much greater than waterflood recovery), but
still in the immiscible region. Because of this potential for significant
recovery, there has been an increasing interest in immiscible C02 floods
in the past year, even though the authors of the NPC report did not include
immiscible floods in their projections.
100
7S
o
cj
UJ
25
J
MISCIBLE DISPLACEMENT
/ . 'IMMISCIBLE DISPLACEMENT
INERT GAS DRIVE
TEST PRESSURE
Figure 9. The General Effect of Pressure on Oil
Recovery by C02 in Slim Tube Tests.
Although data will not be available from most fields, it is instructive
to try to assign an oil recovery value which corresponds to a pressure which
is a reasonable fraction of the MMP required for the maximum C02 recovery.
Fig. 10 is an attempt to illustrate this graphically. Some slim tube experi-
ments from the literature15*17»18 have been replotted in Fig. 10 so that
the oil recovery is expressed as a fraction of the pressure required for
optimum recovery at the MMP. For most of the curves, the oil recovery at
the lower pressures (i.e., at the lower percentages of the MMP) is a long
extrapolation. It is assumed that the curves must approach the origin with
the general shapes indicated because even a completely immiscible gas drive
will give an oil recovery of 157. or more, depending on the oil viscosity.
The oil recovery values from three immiscible displacements of the Retlaw
(Canada) crude oil are also plotted in Fig. 10. The MMP for the Retlaw
-------
79
crude oil was not observed, but estimated by extrapolating the data of Sigmund
et al.18
too
80
ec
UJ
o
o
LJ
60
40
20
RETLAW (MANNEVILLE)
(UMP. 2850 (Mil)
FARMSWORTH
(MMP . 4100)
FORD ZONE
(MMP • 3350)
I
I
I
0 20 40 60 BO
FRACTION OF MMP REOUrRED (*/.)
IOO
Figure 10.
Oil Recovered from Slim-Tube Tests Performed
at Various Fractions of the Minimum Miscibility
Pressure (MMP). (Calculated from Data in
References 15, 17, and 18.)
If one assumes that immiscible C(>2 flooding will be carried out at reservoir
pressures which are equal to 757. of the pressures for the MMP, Fig. 10 indi-
cates that the oil recovery should range from 47-83% of the recovery value
predicted from the slim tube tests in the laboratory. However, only 95%
of the oil is usually recovered in slim tube tests when 100% of the pressure
required for MMP is used. Therefore, if oil recovery is compared to the
standard NPC model8 for miscible flooding, the recovery percentages in Fig.
10 should be divided by 0.95. This would mean that the immiscible oil re-
covery figures should range from 50-87% of the quantities predicted by the
miscible NPC model for those immiscible floods carried out at 75% of the
pressure required for the maximum recovery at the MMP. Current field results
for immiscible projects indicate that recovery should be at least that good,
for there is overlap between the oil recovery percentages observed for mis-
cible and immiscible floods. In general, the immiscible oil recovery often
appears to be better than predicted by simulation methods which assume that
the additional recovery is caused only by oil swelling and viscosity reduction
from the dissolved
Fig. 11 shows the recoveries which should be expected from the different
types of porous media at a range of pressures which spans the immiscible
and miscible regions. Note that close to 100% oil recovery can be expected
from miscible displacement from a slim tube. However, the oil recoveries
drop to only 5-10% at the lower range of immiscible C02 floods in the field
at those lower pressures.20 It appears that the slope of the oil recovery
versus pressure curve is not as steep for the field projects as it is for
the slim tube experiments. Also, as mentioned before, the recovery from
some of the immiscible field projects has been better than originally
anticipated.
-------
80
too
90
0. 80
a
111
5 so
§
a 40
20
10
0
SLIM TUBE
( Moljomor Oil)
BEREA SANDSTONE
(Moljamor Oil )
HETEROGENEOUS ROCKS.
(Estimated)
FIELD RECOVERY
RANGE
500
750 1000 1250 ISOO
co2 FLOODING PRESSURE . PSIG
1750
2000
Figure 11,
Effect of Pressure on Oil Recovery by
(After Reference 1 and 20.)
Miscible CO? Floods
When most petroleum engineers discuss the major CC>2 field projects in the
United States, they are referring to miscible C(>2 displacements, which have
the highest potential recovery. It is well known from screening criteria
publications that C(>2-miscible field projects are limited by the depth of
the formation and the average molecular weight of the crude oil.^»8,21-23
In general, reservoirs deeper than 2,000 feet, which contain oils lighter
than 25° API, are considered candidates for C(>2 flooding. Fig. 12 shows
that the pressure required for miscibility (the MMP) increases markedly
with the API gravity of the oil, especially at higher temperatures. The
MMP required for a given oil increases with depth because the reservoir
temperature goes up with depth and the MMP increases with temperature.
Fig. 12 gives that relationship between the MMP and temperature for oils
ranging between 22-50° API gravity.24-28 Fortunately, the pressure required
to fracture a reservoir also increases with depth because of the heavier
overburden. Fig. 13 shows that the pressure available for injection (to
avoid parting the reservoir) increases much faster with depth than the pres-
sure that is required for the MMP at the greater depths.28 Note that the
fracture pressure and the MMP, for the 40° API oil shown, intersect at just
under 2,000 feet. Therefore, miscible C02 projects are rarely found at
depths shallower than 2,000 feet, and Table 1-M lists only one such project,
Garber at 1,900 feet. The rest of them are distributed at various deeper
depths. Most of the miscible projects are arranged by increasing depth
in Table 2 which shows that the projects range from the aforementioned 1,900
feet, to the Weeks Island project with a depth of 12,760 feet. It is clear
that all of the miscible projects lie within the "window" of Fig. 13, which
is an easy way to make a quick screen of a formation if only the depth and
API gravity of the crude oil are known. The gravity-temperature relationship
for other oils can be cross-plotted easily from Fig. 12 to Fig. 13 as needed.
-------
81
sooo
4000
3000
2000
1000
- CORRELATION OF DATA GATHERED FROM
HOLM end JOSENDAL. M. SUVA AND
THE NATIONAL PETROLEUM
COUNCIL
80 KXD IZO 140 160
TEMPERATURE. *F
ISO
200
Figure 12. Variation of Minimum Miscibility Pressure
with Temperature and Oil Composition
(from Data and Correlations of Holm and
Josendal,15 the National Petroleum Council,&
and M.K. Silva.24'27) (After Reference 28.)
, z j 4 9 • 7 a 9
DEPTH IN THOUSANDS OF FEET
10 II
Figure 13.
Increase in Minimum Miscibility Pressure
(MMP) and Fracture Pressure with Depth
for Permian Basin Reservoirs.
(After Reference 28.)
-------
82
Table 2. C02 Miscible Projects Arranged by Increasing Depth
Field
State
Reservoir oil
Depth API
ft. Gravity cp @°F
Garber
Ford Geraldine Unit
West Sussex Unit
Maljamar MCA unit
McElroy
Means (San Andres)
North Cowden Unit
Vacuum
Tinsley
Welch
Level land
Twofreds
Slaughter (Estate)
Slaughter (Frazier)
Slaughter (Cent. Mallet)
Wasson (ODC Unit)
Sable
Wasson (Denver)
Crossett
Seminole
McElmo Creek Unit
Alvord South Field
Greater Aneth
Rangely
Sho-Vel-Tum
SACROC Unit
Bay St. Elaine
Rank in
Raymond
Pittsburg
Dollarhide
Quarantine Bay
Rose City North
Kurten
University Waddell
Paradis
North Coles Levee
South Bishop Ranch
Northwest Purdy
South Bishop Ranch
Wellman
Paradis
Little Creek Field
Weeks Island
Ok la.
Tex.
Wyo.
N.M.
Tex.
Tex.
Tex.
N.M.
Miss.
Tex.
Tex.
Tex.
Tex.
Tex.
Tex.
Tex.
Tex.
Tex.
Tex.
Tex.
Utah
Tex.
Utah
Col.
Okla.
Tex.
La.
Tex.
Mont.
Tex.
Tex.
La.
Tex.
Tex.
Tex.
La.
Calif.
Wyo.
Okla.
Wyo.
Tex.
La.
Miss.
La.
1,900
2,680
3,040
3,665
3,850
A, 300
A, 300
A, 500
A, 800
A, 890
A, 900
A, 900
A, 950
A, 950
A, 950
5,100
5,200
5,200
5,300
5,300
5,600
5,700
5,750
6,000
6,200
6,700
7.AOO
7,900
7,900
8,000
8,000
8,120
8,200
8,300
8,500
8,600
9,000
9,200
9,400
9,400
9,800
10,400
10,640
12,760
AA.O
AO.O
38.0
36.0
31.0
29.0
3A.O
38.0
39.0
3A.O
30.0
36.0
27.0
31.0
31.0
32.0
32.0
33.0
AA.O
35.0
41.0
44.0
42.0
34.0
25.0
41.0
36.0
37.0
40.0
41.0
40.0
32.0
37.0
38.0
43.0
39.0
36.0
35.0
38.0
34.0
43.5
37.0
38.0
32.0
1
1.4
1.7
0.8
2.3
6
1.67
1
1.5
2.15
2.3
1.5
2
l.A
l.A
1.3
1.A6
1.3
0.36
1.70
0.5
0.39
O.A7
3.3
0.35
0.667
0.6
O.A
0.44
0.99
2
0.4
0.45
0.4
0.45
1.14
1.2
0.54
0.5
0.4
1
100
83
93
90
86
97
94
101
175
96
105
105
105
105
105
110
107
105
106
105
125
154
135
160
115
130
170
192
178
205
122
183
180
230
140
192
235
220
148
180
151
205
248
225
-------
83
Table 3 shove the same miscible projects, but this time arranged in order
of decreasing API gravity. It is not an accident that the shallowest project
(Garber) contains one of the lightest oils in order to fit within the limits
of Fig. 13. The projects in Table 3 fall into three almost-equal groups:
those fields which have rather light crude oils of 40-44° API, oils of inter-
mediate gravities between 36-39° API, and those ranging from 25-35° API.
Therefore, Table 3 shows that almost two-thirds of the miscible C02 floods
in the United States are carried out in reservoirs which have oils lighter
than 35° API.
Immiscible CO? Projects
The immiscible C02 projects listed in Table 1-IM are immiscible presumably
because they are either too shallow or the crude oil is too heavy to meet
the MMP criteria for miscible displacement as shown in Figs. 9, 11, 12,
and 13. The immiscible C(>2 projects have been arranged in order of decreasing
API gravity in Table 4. An examination of this table indicates that some
of these projects should meet the criteria for miscible flooding, i.e.,
their combination of depth and API gravity fall within the window of Fig.
13. The fact that they are reported by the operator as being immiscible
floods indicates that sufficient pressures were not available or were not
utilized to carry out the normally preferred miscible C(>2 displacement.
It must also be emphasized, however, that many of the immiscible floods
in Table I-IM are not typical, long-term C02 displacements, similar to a
waterfloods, but are well-stimulation or cyclic huff 'n1 puff techniques
(see below). Immiscible projects which are regular C02~drive projects seem
to be working very well, e.g., the Lick Creek and Wilmington CC>2 floods.19,29
Cyclic Or CO? Huff 'n' Puff Methods
Traditionally, most enhanced recovery methods involve the injection of a
solvent or chemical which drives the oil from the reservoir into a production
well. Therefore at least two wells are needed; the large CC>2 floods in
the Permian Basin (see Table I-M) often have hundreds of wells; for example,
SACROC (No. 36) has 379 injection wells and 887 production wells. These
projects utilize repeating injection-production well patterns to develop
the large and continuous, near-horizontal reservoirs found in that area.
On the other hand, in the highly faulted, saltdome-intruded reservoirs of
the Gulf Coast, large horizontal reservoirs are the exception, and repeated,
multiwell patterns are not possible in many cases. For these single-well
reservoirs, C(>2 can be used to recover oil by the huff 'n1 puff method.30-33
This CC>2 huff 'n' puff operation is similar to the routine steam stimulation
technique used in the heavy oil reservoirs of California. A specific volume
of C02 is injected into the production well (normally in 1-2 days time),
and then the well is shut in to permit the C(>2 to dissolve into the oil.
This "soak" period may last for 3-6 weeks, during which time the C(>2 swells
the oil and reduces its viscosity. The well is then put back on production.
If the treatment is successful, the production rate will be higher than
before the C02 injection, and it will be sustained for some time. Additional
cycles may be performed as long as production increases are observed. Because
of the big reduction in viscosity when C02 is dissolved in heavy crudes,
-------
84
Table 3. C02 Miscible Projects in Order of Decreasing °API Gravity
Field
Garber
Alvord South Field
Crossett
Wellman
University Waddell
Greater Aneth
Pittsburg
SACROC Unit
McElmo Creek Unit
Ford Geraldine Unit
Raymond
Dollarhide
Tinsley
Paradis
Kurten
Northwest Purdy
West Sussex Unit
Vacuum
Little Creek Field
Rose City North
Rank in
Paradis
North Coles Levee
Maljamar MCA unit
Twofreds
Bay St. Elaine
Seminole
South Bishop Ranch
Rangely
North Cowden Unit
Welch
South Bishop Ranch
Wasson
Wasson
Sable
Quarantine Bay
Weeks Island
Slaughter
Slaughter
McElroy
Level land
Means (San Andres)
Slaughter
Sho-Vel-Tum
State
Okla.
Tex.
Tex.
Tex.
Tex.
Utah
Tex.
Tex.
Utah
Tex.
Mont.
Tex.
Miss.
La.
Tex.
Okla.
Wyo.
N.M.
Miss.
Tex.
Tex.
La.
Calif.
N.M.
Tex.
La.
Tex.
Wyo.
Col.
Tex.
Tex.
Wyo.
Tex.
Tex.
Tex.
La.
La.
Tex.
Tex.
Tex.
Tex.
Tex.
Tex.
Okla.
neservoir oix --
API Depth
Gravity ft. cp @°F
44.0
44.0
44.0
43.5
43.0
42.0
41.0
41.0
41.0
40.0
40.0
40.0
39.0
39.0
38.0
38.0
38.0
38.0
38.0
37.0
37.0
37.0
36.0
36.0
36.0
36.0
35.0
35.0
34.0
34.0
34.0
34.0
33.0
32.0
32.0
32.0
32.0
31.0
31.0
31.0
30.0
29.0
27.0
25.0
1,900
5,700
5,300
9,800
8,500
5,750
8,000
6,700
5,600
2,680
7,900
8,000
4,800
8,600
8,300
9,400
3,040
4,500
10,640
8,200
7,900
10,400
9,000
3,665
4,900
7,400
5,300
. 9,200
6,000
4,890
9,400
5,200
5,100
5,200
8,120
12,760
4,950
4,950
3,850
4,900
4,300
4,950
6,200
1
0.39
0.36
0.54
0.45
0.47
0.35
0.5
1.4
0.4
0.44
1.5
0.4
0.4
1.2
1.7
1
0.4
2
0.6
0.5
0.45
0.8
1.5
0.667
1.70
1.14
1.67
2.15
1.3
1.3
1.46
0.99
0.50
1.4
1.4
2.3
2.3
6
2
3.3
100
154
106
151
140
135
205
130
125
83
178
122
175
192
230
148
93
101
248
180
192
205
235
90
105
170
105
220
160
94
96
180
105
110
107
183
225
105
105
86
105
97
105
115
-------
85
Table 4. C02 Immiscible Projects in Order of Decreasing °API Gravity
Field
Paradis
West Delta Block 109
West Delta Block 109
West Delta Block 109
West Delta Block 109
Lafitte
Bayou Sale
South Marsh Island Block 6
Lake Barre
Cote Blanche Bay West
Yates
West Columbia
Manvel
Magnet Withers Pierce Estates B&C
Magnet Withers BH&S State Tracts J.
Withers North
Withers North
Thompson
Pickett Ridge
Manvel
Pierce Ranch
Sho-Vel-Tum
Plymouth
Talco
East Coyote-Hualde Dome Units
Heidelberg
Pewitt Ranch
Lick Creek
Wilmington (1981)
Wilmington (1984)
Pittsburg
State
La.
La.
La.
La.
La.
La.
La.
La.
La.
La.
Tex.
Tex.
Tex.
Tex.
Tex.
Tex.
Tex.
Tex.
Tex.
Tex.
Tex.
Ok la.
Tex.
Tex.
Calif
Miss.
Tex.
Ark.
Calif.
Calif.
Tex.
API
Gravit
38.0
37.3
37.0
36.5
34.1
34. 0
34.0
34.0
33.0
32.0
30.0
30.0
26.7
26.0
26.0
25.7
25.3
25.2
25.0
25.0
24.4
24.0
23.3
23.0
23.0.
20.0
19.0
17.0
14.0.
14.0.
14.0
Kes<
Depth
y ft.
10,200
10,000
11,325
10,500
12,000
8,900
10,000
11,200
13,000
8,000
1100-1700
2,600
5,000
5,500
5,500
5,250
5,320
5,100
4,600
4,000
4,900
5,530
4,650
3,785
3300-4600
5,060
4,500
1100-1700
2,500
2,500
3,800
ervoir DIJ.-
cp
0.5
0.26
0.3
0.28
0.25
0.7
0.4
0.3
0.4
1.3
5.5
8
7.2
2.3
2.3
2.45
2.9
2.7
2.5
4.4
4.57
18
3.19
25
12
15
30
160
283
283
2,200
@°F
195
194
210
200
218
185
194
208
236
184
82
116
149
154
154
145
147
120
138
149
155
129
150
147
130
150
160
118
123
123
120
-------
86
the method was considered as an alternative to the steam huff 'n' puff cycles
in the California fields. Recently, encouraging laboratory and field results
indicate that the huff 'n' puff method may also provide a good way to utilize
CC>2 flooding in the medium gravity oils of the faulted reservoirs of the
Gulf Coast.'0»32,33 Monger has shown that the C02 requirements and favorable
economics may be similar to the results observed in large horizontal reser-
voirs.32 Several field examples are described in the references.30'32'33
OIL RECOVERY FROM C02 FIELD PROJECTS
The largest C02 floods in the United States are listed in order of field
size in Table 5. C0£ flooding produces about 70,000 barrels a day of enhanced
recovery oil in the United States, and that amount is increasing steadily.
Note that the total daily oil production in Table 5 is somewhat larger;
in some cases it may be difficult for operators to separate the regular
secondary recovery production from the incremental oil which can be assigned
exclusively to C02 flooding. For this and other reasons, the data on incre-
mental oil from C02 injection are missing in the tables in some cases.
There are no real surprises in the oil production figures, i.e. most of
the expected recoveries fall within the ranges discussed earlier and illus-
trated in the shaded area for field recoveries in Fig. 11.
Table 5. Twenty Largest C02 Miscible Projects in the USA
'Arranged by Decreasing Order of Field Size
Field Name
SACROC Unit
Wasson (Denver Unit)
Rangely
Seminole
McElmo Creek Unit
Greater Aneth
Northwest Purdy
Little Creek
Wasson (ODC Unit)
Means (San Andres)
Dollarhide
Slaughter - (Central
Mallet Unit)
Vacuum
Twofreds
Ford Geraldine
Welch
Alvord South Field
Slaughter (Frazier
Unit)
Crossett
Farasvorth, North
Wellman
State
Tex.
Tex.
Col.
Tex.
Utah
Utah
Okla.
Miss.
Tex.
Tex.
Tex.
Tex.
N.M.
Tex.
Tex.
Tex.
Tex.
Tex.
Tex.
Tex.
Tex.
Start
date
1/72
4/83
7/86
4/83
2/85
1982
9/82
12/85
12/84
11/83
5/85
12/84
2/81
111 It
2/81
2/82
1980
12/84
it/72
6/80
7/82
Area,
acres
49,900
20,000
20.000
15,700
13,440
13,357
8,320
8,200
7,800
6.700
6.183
5.700
4.900
4,392
3,850
2,675
2,291
1,600
1,500
1,472
1,400
Number Total
Wells Project prodn.
Prodn. Inj. maturity bo/d
887
840
360
328
170
140
106
110
316
248
62
237
42
198
129
245
64
23
8
29
379
280
360
133
100
21
102
40
250
176
43
97
33
123
132
10
37
11
6
2
NC
JS
JS
JS
JS
NC
JS
JS
JS
JS
JS
JS
HF
JS
HF
HF
JS
HF
HF
JS
43,863
46,000
30,000
41,800
5,600
3.500
3.300
13,000
1,900
7,500
12,400
892
395
3,100
680
3,000
2,000
7,000
Enhanced
prodn .
bo/d
16,000
1.200
950
3,300
80
892
395
300
200
2,000
-------
87
CO? Breakthrough Into Production Wells
Experience from almost all of the C(>2 floods in the United States shows
that most of the oil recovery occurs after C0£ breakthrough. Indeed, some
of the early floods and pilots may have been stopped too early because the
unwanted early breakthrough frightened the operators. Now, until better
mobility control is perfected, early breakthrough is accepted calmly as
an inescapable part of C<>2 flooding. Fig. 14, which shows the oil and C(>2
production from a single well of the recent West Sussex Unit pilot in Wyoming,
illustrates this normal, early C(>2 breakthrough and shows that the flush
oil production comes along with produced CC>2 shortly thereafter.34 Fig.
14 shows that C02 breakthrough occurred only one month from the initiation
of C(>2 injection. It also shows that the oil production continued to increase
for a few months after the C(>2 injection was terminated. In this case,
the CC>2-continuous slug was approximately 30% of the reservoir pore volume.
It seems clear that the oil which had been mobilized by the CC>2 was moved
to the producing well by the waterflood which commenced immediately upon
completion of the C(>2 injection. Fig. 15 shows the cumulative C0.2 produced
and oil production for the whole pilot. Again, it showed that the good
oil kick came after C02 breakthrough, and the increase in oil production
continued for a long period after the C(>2 injection was completed.
™. a
o .
w o
100
10
YEA
eo2 /"i
MCF/D^1 V'
!WN
OIL
B/O^v^
R: 1982
*
^
r v
C02 INJECTION STOPPED
(Nmi0 Wol
-------
88
Continuous vs. Water Alternating with CO? Injection
In the West Sussex pilot, the C(>2 was injected continuously until the requi-
site amount had been pumped into the reservoir, i.e., water alternating
with gas (WAG) for mobility control was not practiced, and the operator
did not consider C(>2 breakthrough a problem. The controversy over whether
to WAG or not to WAG will continue until many more reservoirs have been
flooded by both methods. Table 6 will show that the three projects with
the highest net oil recovery (Weeks Island, Crossett, and Little Creek)
are also projects which have not used the WAG method for mobility control.
However, some engineers will point out quickly that this is not a fair compari-
son. The Weeks Island flood is a gravity-stabilized displacement down-dip,
and it would therefore be expected to yield much higher recovery than hori-
zontal floods. The Little Creek early pilot, from which a record 46% of
the residual oil was recovered from a watered-out reservoir, used a very
large quantity of C(>2, 160% of the hydrocarbon pore volume (HCPV) compared
to the common 30-40% HCPV for most field projects. Therefore, it also holds
the record for the highest C(>2 requirement: 26 MCF of CC>2 injected per
barrel of incremental oil produced. The Crossett C02 flood appears to be
on the way to very high ultimate recovery, no doubt because it is an enhanced
secondary flood in a formation which was too tight for prior waterflooding.
Therefore, the oil saturation in the reservoir was very high (compared to
tertiary C(>2 floods) when C02 injection started. In addition, Crossett
was not plagued with early C02 breakthrough, perhaps because of asphaltene
precipitation and/or multiphase flow which provided added mobility control.35
Oil Recovery Observed and Predicted
The oil recovery and CC>2 requirements for several C(>2 floods are listed
in Table 6. Note that oil recoveries may range from 10-60% of the remairiing
oil in place. As mentioned earlier, the enhanced secondary floods show
up very well, especially Crossett. This is not surprising because CC*2 flooding
is basically a vaporizing gas drive method; if this method is used in a
non-waterflooded reservoir, very high oil recoveries can be expected. For
example, recovery of 60% of original oil in place is expected from University
Block 31 which started with methane injection, switched to flue gas, and
finally to nitrogen flooding.36'41
As more experience is gained in CC<2 flooding, the general optimism appears
to be increasing. For example, published predictions by the same author2^»^^
of oil recovery from simulations and other engineering calculations have
changed from 1983 to 1986 as follows:
Expected Oil Recovery
(% ROIP)
Predicted in 198342 Predicted in 198629
Kelly Snyder (SACROC) 11 21
Crossett / 25 55
Twofreds 27 16
Lick Creek 17 24
-------
89
Table 6. Oil Recovery and C(>2 Requirements for C(>2 Floods
C02-011 Ratios (MCF/BO)
Observed or Expected
Oil Recovery
(IOOIP) (IROIP)
Purchased Recycled Total
SECONDARY
Kelly Snyder (SACROC) 10 21
Crossett 44 55
Twofreds 13 15
TERTIARY
H.E. Purdy 8 12
East Velna 9 15
Little Creek Pilot 18 46
GRAVITY STABLE
Paradis 8 10
Weeks Island - 60
IMMISCIBLE
Lick Creek 16 24
4-5
4-5
12
2
2
14
4-5
7
10
12
6-7
6-7
26
11-12
10-11
General Expected Range
(Tertiary) 8-15 15-30
A-8
3-7
7-14
(After Reference 27 and other sources.)
Note that three of the four estimates went up markedly during the past three
years. (I do not know why the Twofreds went down.) Indeed, engineers have
almost doubled their oil recovery estimate for SACROC and Crossett since
1983. Although "hard copy" references are not readily available, company
personnel continue to suggest that the current production figures from the
major C02 floods often are better than their prior engineering predictions.
Seminole, Garber, Slaughter, and Wasson are mentioned as projects which
continue to look better and better as more oil production history is accumu-
lated.
It seems clear that the long-term prediction in the NPC report of more than
five billion barrels of additional oil by miscible flooding (for which C02
will be the major contributor) should be met. A recent article, which in-
cludes an update of the miscible flooding results, points out that the present
production rate of about 70,000 B/D indicates that the NPC prediction is
right on target.^
ECONOMICS
The optimism which has been growing because of the good field response from
C02 injection has been tempered in recent months by the precipitous drop
in oil prices. Estimated costs for C02 are shown in Table 7 for five of
the C02 floods that were listed in Table 6. It is presumed that this table
-------
90
was prepared before the prices had dropped to their lowest values. Table
7 shows that CC>2 costs range from about $4.50 per barrel to less than $9.00
per barrel of incremental oil recovered by the injected CC>2. Therefore
the table indicates that it would not be possible to continue floods economi-
cally if the price of oil drops much below $10.00 a barrel for those CC-2
floods which consume the most C02- However, most of the long-term purchase
contracts for C02 contain a clause which allows the C02 price to drop with
a decline in oil prices. Therefore, some of the costs in Table 7 can go
down when oil prices are lower. Indeed, Marvin Katz has been quoted as
saying that C02 can be continued economically with oil prices ranging from
as low as $3.00 to $12.00 per barrel as long as the pipelines and the distri-
bution and wellhead equipment are already in place in the oilfields.^ The
oil companies have invested more than two billion dollars in the pipeline
supply system for the Permian Basin alone, and as long as any profit margin
can be maintained from the produced oil, it is presumed that the C(>2 will
keep flowing through the pipelines and into the injection wells.
Table 7- Sample C02 Costs for EOR Projects
Project
• SECONDARY
Kelly Snyder
Crossett
• TERTIARY
H.E. Purdy/
East Velna
• IMMISCIBLE
Lick Creek
Purchased
C02
(S/BO)
4.90
3.30
4.40
6.20
Recycled
C02
($/BO)
0.60
1.30*
0.80
1.90
Total C02
Costs
($/BO)
5.50
4.60
5.20
8.10
Assumingi
Purchased C02 2.
-------
91
Some of the individual states are considering tax incentives (such as for-
giving a portion of the production taxes) in order to encourage the initiation
of new C(>2 field projects. If this is done, many legislators assume that
the increased economic gain from the higher oil production will far outweigh
the loss of income from the direct taxes on the oil produced. The New Mexico
Research and Development Institute sponsored a recent study to determine
if such tax incentives would increase oil production by CC-2 flooding.**
The results of this study show that some tax relief should have a very posi-
tive impact on oil recovery. The study used the methodology of the NPC
report8 to determine the potential of C(>2 flooding for 97 reservoirs in
New Mexico. Fig. 16 shows the increased oil production from C(>2 injection
that is predicted for oil prices between $20 and $32 per barrel with the
present tax structure. Note that the increased oil production is small
unless the oil price rises above $24/BBL. However, if proper tax incentives
are provided, such as the forgiving of all production taxes until the project
recovers the front end costs (i.e., the break even point or incentive to
payback on Fig. 17), substantial new reserves, increased oil production,
and other benefits to the State will result with oil prices between $20
and $30/BBL. Fig. 17 shows that the increased oil production by CC>2 flooding
with tax "incentives to payback" at $24/BBL almost equals the increased
CC>2 production at $28/BBL without the incentives. If oil prices rise to
$32/BBL, the report concludes that the incentives are not needed to encourage
new C(>2 projects.^
70
g «M
PROJECTED
CONVENTIONAL
DECLINE
1SB5 1990 199S 2000 20O5 2010
10% RATE OF RETURN
70
eo •
50
40
X
PROJECTED
CONVENTIONAL
OCCLINC
1885 1990 1995 2000 200$ 2010
15% RATE OF RETURN
Figure 16. Estimated New Mexico Total Production for
Conventional and C02 Flooding Techniques
as a Function of Oil Price with Current
Taxes. (Reproduced from Reference 45.)
-------
92
80
70 -
60 -
SO -
40 -
20 J
\
\
1985 1990 1195 2000 2005 2010
OIL PRICE S24/8BL
10V. RATE OF RETURN
70
60 •
50 -
40 -
30 •
1985 1990 1995 2000 2005 2010
OIL PRICE S28/BBL
10V. RATE OF RETURN
IECSHO
— PHOJCCTtO CONV. »«OOuCIION
TOTAk MOOUCTIOPI V/COl. ASSUMING
^^-^^ CURftCNT TAKES
.-..-- IHCCHTIVt 10 P«Ti»C«
___ — mctxrivc ron Lire
Figure 17. Potential Effects on Total Oil Production
Due to C(>2 Flooding in new Mexico for Three
Tax Structures at $24 and $28 per Barrel.
(Reproduced from Reference 45.)
There have been investigations of the possibility of obtaining C02 from
power plants in the United States, but with present oil prices, the costs
appear to be too' high unless a significant "acid rain credit" could be given
to some newer processes which are being considered for removing C(>2 from
power plant stack gases. Argonne National Laboratories has been investigating
the possibility of burning coal in pure oxygen so that the stack gas stream
would be almost pure C(>2 plus the acid rain components, all of which could
be injected into the ground for good oil recovery. A recent review of past
and ongoing flue gas injection projects indicates that the additional S(>2
and NOX should not be serious problems for the reservoir, and the corrosion
problems can be managed-*6 (see next section). The quantities of relatively
pure C02 which could be produced from power plants are very large compared
to the volume of the reservoirs which can use CC>2 effectively in the central
and eastern part of the United States. If idle pipelines could be reversed
to carry low cost C(>2 from the industrial Ohio Valley area to the Gulf Coast
oil fields, it should provide a big boost to EOR by C02 flooding. Studies
indicate that this is feasible.46
PRACTICAL FIELD PROBLEMS
There are many more operating problems associated with a C02 flood than
with a straight waterflood for secondary recovery. Problems which are most
often cited are: the early breakthrough of C02 plus the continued production
of C02 throughout the life of the project; the large volumes of C02 which
-------
93
must be separated from the produced oil and gas, and in most cases, recycled;
the lower-than-anticipated injection rates which have been experienced in
many of the floods; and the hardware problems such as increased corrosion
of tubular goods, swelling or deterioration of the elastimers used in gaskets,
packers, etc. Although these problems can be serious, all of them seem
to be manageable as long as the planning and engineering are done carefully.
We have already discussed the fact that early breakthrough is now considered
a normal part of C(>2 floods, and the chemical engineers are making great
strides in efficient techniques for separating and recycling the produced
C02.47'48 Corrosion and problems which may arise from the injection of
acid gases into reservoirs are treated briefly in separate sections.
Acid Gas Corrosion Problems
Corrosion has been a problem in oil fields ever since the operators in Penn-
sylvania started to convert their wooden tanks and pipes to iron and steel
equipment many decades ago. Pure oil is not corrosive, but any combination
of water with oxygen, or an "acid gas" compound such as S(>2, 803, H2S, NOX,
or C02 will normally corrode ordinary carbon steel. Therefore, corrosion
engineers expect that C(>2 floods will be more troublesome than waterfloods,
but technical problems should have technical solutions, and the corrosion
problems with CC>2 are being solved. Corrosion along with other operating
considerations are addressed in several references to CC<2 field projects.49-55
Flue gases manufactured from methane for injection into reservoirs, and
CO2 obtained from power plants can be much more acidic than the pure C02
from the natural sources in the United States. However, flue gas has been
injected into oil reservoirs for about 40 years, and the corrosion problems
have been dealt with successfully. Table 8 lists seven flue gas projects
along with the corrosion control method used.-*" Except for the one started
in 1924, all were successful even though corrosion was recognized as a problem.
A well-documented field trial where corrosion was controlled and monitored
very carefully was Amoco1s Slaughter Estate Pilot which utilized an acid
gas composed of 727. CC>2 and 287. H2S.54 This mixture was injected success-
fully for three years with no mishaps except for an occasional shutdown
by the automatic safety equipment. Because of the safety concerns, more
attention was given to corrosion monitoring and control than in an ordinary
CC>2 flood, but the documentation should be very helpful for anyone concerned
with acid gas corrosion. The corrosion monitoring system can be summarized
by the following list of devices and methods which is reproduced from reference
36:
Corrosimeter® Probes. These devices measured metal loss electrically
by the change in resistance across a test probe inserted in the pipe.
These probes permitted continuous monitoring of the corrosion rate.
Corrosion Coupons. These mild steel (SAE-1010) rods were inserted
into the pipe and removed after specified times to determine metal
loss.
Corrosion Test Nipples. Short sections of the same pipe material
were cut in half and welded back together with flanges attached so
the test sections could be inserted in the CC^-l^S flow line at sen-
sitive points, such as at low points in the pipe at injection wellheads.
-------
94
Ultrasonic Metal-Thickness-Measuring Devices. These instruments were
used to measure the piping thickness at various points in the com-
pressor, dehydrator, and gas heater units.
Hydrogen Probes. Corrosion was measured by determining the pressure
of H£ which built up inside a hollow probe inserted into the pipe.
The pressure comes from molecular hydrogen (H£) which formed from
the atomic hydrogen ions (IT1") that are formed by corrosion at the
probe surface and then diffused through the probe wall. Because the
H£ could not escape, the pressure increased in proportion to the corro-
sion outside the probe. These hydrogen probes were reported as the
most valuable monitoring devices because the H2 pressures could be
read daily and the rate of pressure increase gave a quick indication
of change in corrosive environment, including the effectiveness of
corrosion inhibitors.
Table 8. Examples of Flue Gas Injection Projects in the United States.
Starting
Date
1924
Field Name
or Location
Texai
Succenful?
No
Corrosion
Control
Not controlled
Breakthrough of
S(>2 or N(>2
Observed?
No
1949 Elk Basin, Wyoming
19S9 Louisiana
1966 University Block
(1949 for 31, Texas
methane)
1966 Neale, Louisiana
1977 Hawkins, Texas
1977 East Blnger,
Oklahoma
satisfactorily
Yes Asmonla Injection No
Yes Catalytic conversion No
of NOX
Yes Addition of NH^OH; No
recycling flue gas
around burner tips
to reduce NOX
Yes Catalytic reduction No
and excellent
dehydration
Yes Catalytic reduction, No
corrosion inhibitors,
and dehydration
Yes Not reported No
After Reference 36.
Concerns Related to Interaction of Acid Gases with Reservoir Rock
If Argonne's method for burning coal in oxygen should be adopted by electric
utilities, large quantities of CC>2 would become available at low cost.
If the stack gas C02 is used directly for EOR, questions about the interaction
of the reservoir rock and acid rain components in the C(>2 stream will arise.
A recent study36 concludes that an oil reservoir should be an excellent
scrubber to remove the acid rain compounds from stack gases. Even though
hundreds of billions of cubic feet of flue gas have been injected in oil
recovery projects in the United States, Table 8 shows that no NOX or SOX
has been observed at the production wells. However, most of the flue gas
-------
95
projects In the United States have injected lower percentages of the acid
components because the operators have taken steps to reduce the NOX concen-
tration to control corrosion prior to the gas injection, and the S(>2 content
is very low when sweet methane is used as fuel. Extremely high concentrations
of S02 might pose a problem in carbonate reservoirs because laboratory flow
experiments with 157. S(>2 in C(>2 have gradually plugged limestone cores.^6
However, no plugging, even with pure S02, was observed when the gas was
flowed through Berea sandstone. Therefore, lab experiments should be carried
out before a large project with acidic flue gas is started. Since the S(>2
concentration in the Argonne-process stack gas would be somewhat less than
17. (depending on the coal), it is assumed that many oil reservoirs in the
United States could utilize the untreated gas effectively, as long as it
was dehydrated enough to control the corrosion.
CONCLUSIONS
C02 flooding for enhanced oil recovery in the United States is working,
and it works well in either the secondary or the tertiary recovery modes,
as miscible or immiscible floods, and with cyclic or continuous C02 injection.
As more experience is gained from existing floods, the indications are that
oil recoveries will be higher than predicted originally. C02 from natural
sources is available for many of the reservoirs in west Texas and eastern
New Mexico, for many other reservoirs in the Rocky Mountain area, and for
some reservoirs in Mississippi and Louisiana. Other C02 sources will cer-
tainly be developed, depending upon the predicted price of oil over the
long-term.
The present CC>2 floods will continue to operate with no letup at today's
prices (March 1987), but one must wait for indications of higher oil prices
before a huge number of new CC>2 floods will be initiated. However, if the
present prices hold temporarily, and rise above $20.00 eventually, projections
indicate that sufficient C02 floods will be underway during the next decade
to ensure that the NPC report predictions of 500,000 BBLS/day will be exceeded
easily by the time the peak production from C02 flooding is reached in the
year 2005.
REFERENCES
1. Taber, J.J.: "Enhanced Oil Recovery by Gas Miscible Flooding," presented
at the Enhanced Oil Recovery Symposium of the International Energy
Agency Collaborative Research Program, Tokyo, Japan, Oct. 9, 1985.
2. Taber, J.J.: "Carbon Dioxide Field Projects in the United States of
America," presented at the Enhanced Oil Recovery Symposium of the Inter-
national Energy Agency Collaborative Research Program, Hannover, Germany,
Sept. 16-19, 1986. (Parts of this paper are reproduced here.)
3. Leonard, J.: "Increased Rate of EOR Brightens Outlook," Oil & Gas J.
(April 14, 1986) 84, No. 16, 71-101.
4. Taber, J.J.: "Technical and Economic Criteria for Selecting Methods
and Materials for Enhanced Oil Recovery (or Why C02 Fills the Bill
-------
96
in the Permian Basin of New Mexico and Texas)," presented to the Enhanced
Oil Recovery Committee of the Interstate Oil Compact Commission, Santa
Fe, NM, Dec. 4, 1984.
5. Taber, J.J. and Martin, F.D.: "Technical and Economic Criteria for
Selecting Methods and Materials for Enhanced Oil Recovery (or Why C02
Fills the Bill in the Permian Basin of New Mexico and Texas)," presented
at the 1985 Southwestern Petroleum Short Course, Lubbock, TX, April
23-25.
6. Smith, L.R.: "Overview of C©2 Flood and Supply and Supply Source Activity
for Enhanced Oil Recovery in the Permian Basin," Bull.. IOCC (1983)
1-5.
7. Stalkup, Fred I.: Miscible Displacement, Monograph Series, SPE, Dallas,
New York (1983).
8. Bailey, R.E. et al.: Enhanced Oil Recovery, National Petroleum Council,
Industry Advisory Committee to the U.S. Secretary of Energy, Washington,
D.C. (1984).
9. Cobb, L.B. and Goodrich, J.H., Principal Investigators "Target Reservoirs
for C02 Miscible Flooding, Task II: Summary of Available Reservoir
and Geological Data," final report, Contract No. DE-AC21-79MC08341,
U.S. DOE (1982).
10. Taber, J.J.: "Oil and Gas Production and Research in New Mexico," From
Sundaggers to Space Exploration, New Mexico Academy of Science and
New Mexico Sigma Xi Chapters and Clubs, Albuquerque, NM (Feb. 1986)
132-149.
11. Hagar, R.: "Permian Basin C02 Floods Mushroom; More Scheduled," Oil
& Gas J. (July 8, 1985) 83_, No. 127, 17-22.
12. Seba, R.D.: "Shell's EOR Program in Louisiana and Mississippi," presented
to the Enhanced Recovery Committee of the Interstate Oil Compact Com-
mission, Salt Lake City, Utah, Dec. 8, 1986.
13. Pande, N.: "A Comprehensive Study of the Use of Carbon Dioxide for
Enhanced Oil Recovery, Part 1: C02 Field Project Study, Part 2: C02
Literature Survey," New Mexico Petroleum Recovery Research Center,
Socorro, NM (1985).
14. Holm, L.W.: "Miscibility and Miscible Displacement," J. Pet. Tech.
(Aug. 1986) 817-818.
15. Holm, L.W. and Josendal, V.A.: "Effect of Oil Composition on Miscible-
Type Displacement by Carbon Dioxide," Soc. Pet. Eng. J. (Feb. 1982)
87-98.
16. Lee, J.L.: "Effectiveness of Carbon Dioxide Displacement Under Miscible
and Immiscible Conditions," Research Report RR 40, Petroleum Recovery
Institute, Calgary, Canada (March 1979).
-------
97
17. Orr, F.M., Jr. et al.: "C02 For EOR," ChemTECH (1983) 13. No. 8, 482-487-
18. Sigmund, P.M. et. al.: "A Laboratory and Computer Model Evaluation
of Immiscible Carbon Dioxide Flooding in a Low-Temperature Reservoir,"
paper SPE/DOE 12703 presented at the 1984 Third SPE/DOE Symposium on
EOR, Tulsa, OK, April 16-18.
19. Mayer, E.H. et. al.: "An Analysis of Heavy Oil Immiscible Core Flood
Data," paper SPE/DOE 14901 presented at the 1986 Fifth SPE/DOE Symposium
on EOR, Tulsa, OK, April 20-23.
20. Orr, F.M., Jr. and Taber, J.J.: "Displacement of Oil by Carbon Dioxide,"
final report, Contract No. AS19-80BC10331, U.S. DOE (May 1981).
21. Taber, J.J. and Martin, F.D.: "Technical Screening Guides for the
Enhanced Recovery of Oil," paper SPE 12069 presented at the 1983 SPE
Annual Technical Conference and Exhibition, San Francisco, CA, Oct.
5-8.
22. Taber, J.J.: "Enhanced Recovery Methods for Heavy and Light Oils,"
Heavy Versus Light Oils; Technical Issues and Economic Considerations,
R. El Mallakh (ed.), The International Research Center for Energy and
Economic Development, Boulder, CO (1984) 221-249.
23. Geffen, T.M.: "Oil Production to Expect from Known Technology." Oil
& Gas J. (May 7, 1973) 71, No. 19, 66-76.
24. Silva, M.K. and Orr, F.M., Jr.: "Effect of Oil Composition on Minimum
Miscibility Pressure - Part 1: Solubility of Hydrocarbons in Dense
C02," paper SPE 14149 presented at the 1985 SPE Annual Technical Con-
ference and Exhibition, Las Vegas, NV, Sept. 22-25; in press, 1987.
25. Orr, P.M., Jr. and Silva, M.K.: "Effect of Oil Composition on Minimum
Miscibility Pressure - Part 2: Correlation," paper SPE 14150 presented
at the 1985 SPE Annual Technical Conference and Exhibition, Las Vegas,
NV, Sept. 22-25; in press, 1987.
26. Silva, M.K., Taber, J.J., and Orr, F.M., Jr.: "Pressures Required for
Miscible Displacement of Crude Oils by C02," presented at the 1985
International Energy Agency Workshop on EOR, Tokyo, Japan, Oct. 5.
27. Silva, M.K.: personal communication, Socorro, NM (Dec. 1985).
28. Heller, J.P. and Taber, J.J.: "Influence of Reservoir Depth on Enhanced
Oil Recovery by O>2 Flooding," paper SPE 15001 presented at the 1986
SPE Permian Basin Oil and Gas Recovery Conference, Midland, TX, March
13-14.
29. Kuuskraa, V.A.: "The Status and Potential of Enhanced Oil Recovery,"
paper SPE/DOE 14951 presented at the 1986 SPE/DOE Fifth Symposium on
EOR, Tulsa, OK, April 16-18.
30. Welch, D.H. and Graham, J.H., Jr.: "A New Low Cost Enhanced Oil Recovery
Method for Louisiana," presented to the Enhanced Recovery Committee
-------
98
of the Interstate Oil Compact Commission, Salt Lake City, Utah, Dec.
8, 1986.
31. Begin, R. and Krueger, D.A.: "Computer Simulation of Recovery of Heavy
Crude Oil Using Carbon Dioxide Drive or Huff 'n' Puff," U.S. DOE Final
Report DOE/BC/10640-21 (Nov. 1983).
32. Monger, T.G. and Coma, J.M.: "A Laboratory and Field Evaluation of
the C02 Huff 'n1 Puff Process for Light Oil Recovery," paper SPE 15501
presented at the 1986 Annual Technical Conference and Exhibition, New
Orleans, Oct. 5-8.
33. Palmer, F.S. et al.: "Design and Implementation of Immiscible Carbon
Dioxide Displacement Projects (O>2 Huff 'n' Puff) in South Louisiana,"
paper SPE 15497 presented at the 1986 SPE Annual Technical Conference
and Exhibition, New Orleans, Oct. 5-8.
34. Hoiland, R.C. et.al.: "Case History of a Successful Rocky Mountain
Pilot C02 Flood," paper SPE/DOE 14939 presented at the 1986 Fifth SPE/DOE
Symposium on EOR, Tulsa, OK, April 20-23.
35. Pontious, S.B. and Tham, M.J.: "North Cross (Devonian) Unit C02 Flood,"
J. Pet. Tech. (Dec. 1978) 1706-1714.
36. Taber, J.J.: "Fate of Small Concentrations of SO2, NOX, and 02 When
Injected with C02 into Oil Reservoirs," Report No. ANL/CNSV-50, Argonne
National Laboratory, Argonne, IL (June 1985).
37. Caraway, G.E. and Lowrey, L.L.: "Generating Flue Gas for Injection
Releases Sales Gas," Oil & Gas J. (July 28, 1975) ^3_, No. 30, 126-132.
38. Hardy, J.H.-and Robertson, N.: "Miscible Displacement by High-Pressure
Gas at Block 31," Petr. Eng. (Nov. 1975) 47. 24-28.
39. Herbeck, E.R. and Blanton, J.R.: "Ten Years of Miscible Displacement
in Block 31 Field," J. Pet. Tech. (June 1961) 543-549.
40. Warner, H.R. et al.: "University Block 31 Field Study, Part 1: Middle
Devonian Reservoirs History Match," J. Pet. Tech. (Aug. 1979) 962.
41. Warner, H.R. et al.: "University Block 31 Field Study, Part 2: Reservoir
and Gas Plant Performance Predictions," J. Pet. Tech. (Aug. 1979) 971.
42. Kuuskraa, V.A.: "Current and Future Economics of Enhanced Oil Recovery,"
presented at the 1983 Symposium on EOR for the Independent Oil Producer,
Institute for the Study of Earth and Man, Southern Methodist University,
Dallas, Nov. 9-10.
43. Robl, F.W., Emanuel, A.S., and Van Meter, O.E., Jr.: "The 1984 National
Petroleum Council Estimate of Potential EOR for Miscible Processes,"
J. Pet. Tech. (Aug. 1986) 875-822.
44. Katz, M.I "Oil Price Seen Lagging EOR Threshold," Oil & Gas J. (April
28, 1986) 84, No. 17., 38, 40. ~
-------
99
45. Brashear, J.P. et al. and the Interstate Oil Compact Commission: "The
Potential of Enhanced Oil Recovery by Carbon Dioxide Flooding in New
Mexico," Report No. NMRDI 2-74-4806, New Mexico Research and Development
Institute, Santa Fe, Dec. 1986.
46. Ford, Bacon and Davis, Inc.: "Technical and Cost Evaluation of the
Use of Idle Pipelines for Reverse Carbon Dioxide Service," Report No.
ANL/CNSV-TM-159, Argonne National Laboratory, Argonne, IL, Feb. 1985.
47. "Amoco Starts Up C02 Recovery Plant in Big West Texas Field," Oil &
Gas J. (Sept. 9, 1985) _8_3, No. 36, 80.
48. Ormiston, R.M. and Luce, M.C.: "Surface Processing of Carbon Dioxide
in EOR Projects," J. Pet. Tech. (Aug. 1986) 823-828.
49. Hansen, P.W.: "A C02 Tertiary Recovery Pilot, Little Creek Field,
Mississippi," paper SPE 6747 presented at the 1977 SPE Annual Technical
Conference and Exhibition, Denver, CO, Oct. 9-12.
50. Newton, L.E., Jr. and McClay, R.A.: "Corrosion and Operation Problems,
C02 Project, Sacroc Unit," paper SPE 6391 presented at the 1977 SPE
Permian Basin Oil and Gas Recovery Conference, Midland, TX, March 10-11.
51. Palmer, F.S., Nute, A.J., and Peterson, R.L.: "Implementation of a
Gravity Stable, Miscible C02 Flood in the 8000-Foot Sand, Bay St. Elaine
Field," paper SPE 10160 presented at the 1981 SPE Annual Technical
Conference and Exhibition, San Antonio, TX, Oct. 5-7.
52. Macon, R.S.: "Design and Operation of the Levelland Unit C02 Injection
Facility," paper SPE 8410 presented at the 1979 SPE Annual Technical
Conference and Exhibition, Las Vegas, Sept. 23-26.
53. Johnston, J.W.: "A Review of the Willard (San Andres) Unit C02 Injection
Project," paper SPE 6388 presented at the 1977 SPE Permian Basin Oil
and Gas Recovery Conference, Midland, TX, March 10-11.
54. Adams, G.H. and Rowe, H.G.: "Slaughter Estate Unit C02 Pilot - Surface
and Downhole Equipment Construction and Operation in the Presence of
Hydrogen Sulfide Gas," J. Pet. Tech. (June 1981) 1065-74.
55. Frey, R.P.: "Operating Practices in the North Cross C02 Flood," Proc.,
22nd Annual Southwestern Petroleum Short Course, Lubbock, TX (1975)
165-68.
ACKNOWLEDGMENTS
The author thanks the following for their valuable contributions to this
effort: Jessica McKinnis for drafting the figures, Guadalupe Williams for
preparation of the tables, Paula Bradley for expert editing, Thomas Taber
for preparation of the maps with C02 project locations, Stanley Walker,
Chevron, U.S.A., Inc., for the map in Fig. 4-a, and Janet Golding for suf-
fering through the many drafts of the manuscript.
-------
100
-------
Preceding page blank 101
DESCRIPTION OF A PLANNED CO2 RECOVERY PROJECT IN WYOMING
Richard J. Tupper and Tom Fuller
Wyodak Resources Development Corp., Rapid City, South Dakota
ABSTRACT
Wyodak Resources Development Corp. (Wyodak) has completed studies to determine
the market for and the economics of extracting CO^ gas from the stack gases of
a coal-fired electric generating station located in the Powder River Basin near
Osage, Wyoming. The results of these studies indicate that there is a market
for C02 to be used for enhanced oil recovery. This market is, however,
dependent on the price of crude oil, the cost to produce COg, and the distance
of the CC>2 supply from the oil field. Wyodak is prepared to construct a C02
plant if contracts can be obtained for its production. If a plant is
constructed it will be one of the first plants to be constructed to extract
C02 gases from a coal-fired electric generating station and could provide a
vast new economical source of C02 to be used for enhanced oil recovery.
INTRODUCTION
Wyodak is the oldest, continuous operating coal mine in the Powder River
Basin. Wyodak is located 5 miles east of Gillette, Wyoming, and has an
annual production of approximately 3 million tons of coal per year.
Wyodak's parent company, Black Hills Corporation (BHC), is a diversified
corporation consisting of an electric utility, a coal mining company, a dry
bulk trucking company, and an oil and gas operating company. The electric
utility generates, transmits, and distributes electric energy in the Black
Hills of South Dakota and a portion of northeastern Wyoming.
BHC has five coal-fired generating stations located within its service area
that provide generating capacity to supply its customers' needs. The stack
gases of a coal-fired generating station is one of the most plentiful sour-
ces of C02 gas. The potential users of the C02 gas are the oil fields
located in the Powder River Basin within a radius of approximately 75 miles
of the coal-fired generating stations. The extraction of C02 gases from
the stack gases of a coal-fired power plant, if done economically, would be
a very reliable source of C02 for enhanced oil recovery. A C02 plant would
also be a new source of income for Wyodak and its parent company in utili-
zation of a product that is presently going to waste.
MARKETING STUDY
Wyodak commissioned Stone and Webster Engineering Corporation to do a
marketing study to determine the market potential for C02 gas in the Powder
River Basin area and within a 75-mile radius of BHC power plants. This
Preceding page blank
-------
102
study concluded that there was approximately 1.2 billion barrels of oil
which could be recovered by the C02 enhanced oil recovery method. This
recovery would require a demand for 300 million standard cubic feet per day
OOOMMscfd) of C02 gas for the next twenty years. Refer to Figure 1 for
location of potential sources and users of
There are two main sources of CO? gas—natural deposits of C02, and C02
available from industrial plant flue gas. The largest single source of
industrial plant flue gas is the flue gas from coal-fired electric power
plants.
The feasibility of an enhanced oil project is determined by the price of
crude oil, the price of the C02 available, and the distance of the C02
source to the oil field. The marketing study indicated that, for a project
to be feasible from BHC power plants in Wyoming, the cost of producing C02
would need to be less than $2.00 per Mscf. The feasibility of using C02
depends on the price of oil—with each oil field having its own economic
conditions of price of C02 versus the price of crude oil— to make C02
economical.
FEASIBILITY STUDY
Wyodak, in its search to determine if C02 could be economically captured
from the stack gases of BHC's power plants, hired Pritchard Corporation to
do a feasibility study to determine the cost of constructing a plant to
extract C02 from BHC's Osage coal-fired power plant near Osage, Wyoming.
This power plant consists of 3-10 MW stoker-fired units and was picked
because it is a base-loaded plant located close to the oil fields. The
Osage Plant also has a record of over 90* availability. With three units,
an uninterrupted supply of C02 could be assured.
The Pritchard Study surveyed the different methods available to extract C02
from stack gases. They then proceeded to calculate the construction and
operation costs of a proposed C02 plant at Osage. Pritchard1s recommen-
dation was to construct a C02 plant of approximately 1,000 tons per day C02
capacity using the Dow Chemical GAS/SPEC FS-1 solvent and FT-2 technology.
This process is based on Dow's proven GAS/SPEC FT-1 technology for removing
C02 from flue gas plus Dow's new KT-2 (caustic) technology for removing S02
from a flue gas. These two processes are combined into their FT-2 process.
The projected sale price of the C02 at 2,000 psi delivered at Osage was
estimated in the range of $1.50 per Mcft. The total cost of the plant was
estimated to be in the $20 million range. These prices are based on 1985
costs and economic conditions. Before a project is started these costs
would need to be updated to reflect the new tax laws and cost of capital
for construction.
ADDITIONAL STUDY
Wyodak's interest and desire to look at other ways to extract CO? from
power plant stack gases has led them to work with Alan Wolsky of Argonne
-------
103
National Laboratories on a new method of obtaining C02 from stack gases.
This method was originally described by Wolsky at the last C02 conference
at the Asilomar Conference Center in Pacific Grove, California, on
February 11-13, 1985. In this process, a portion of the flue gas flow--
which consists of C02, H20, and other gases--is recycled to the furnace.
The recycled flue gas is mixed with pure oxygen to provide firing charac-
teristics similar to air.
C02 (and H20)
1
Air |
k »• i
i
f Oxidizing ^
02 Mixture w
f 1
FURNACE
1
1
Product ^
' 1
C02
NO,
N2
H20
Wolsky ran a small test-run using this method at Battelle-Cplumbus
Laboratories. No work had been done on an actual coal-fired boiler. To
further Wolsky's work, Wyodak and Black Hills Corporation have recently
completed a test using this method on a stoker-fired boiler at BHC's
Service Center in Rapid City, South Dakota. The purpose of this test was
to determine the feasibility of installing a C02 recycle system-on commer-
cially sized stoker-fired or pulverized-fired utility boilers.
Summary of Results
The tests showed that it is possible to run the heating boiler in
a flue gas recycle mode and to achieve increases in the carbon
dioxide levels in the flue gas. Recycle operation did not have
noticeably adverse affects on boiler. The equipment for the test
consisted of a Keewanee fire-tube boiler with Canton (Detroit Stoker)
stokers. It is rated at 2.2 million Btu's and is fired on Wyodak
sub-bituminous coal. The coal is a low sulfur coal with approximately
8,000 Btu's per pound. The boiler is equipped with forced draft fan and
induced draft fan. The unit normally operates with balanced draft. It
provides 15 psi, 230 degree Fahrenheit hot water to the heating system.
Equipment Modifications
The major modification to the system consisted of installing the bypass
ductwork to provide the flue gas recirculation capabilities. The
bypass duct was 12 inch round, insulated duct. It had provisions for
the installation of the sparger near the connection to the flue gas
duct. It also had slide gates to allow normal, air-fired conditions
-------
104
as well as oxygen-flue gas firing. A damper was installed in the flue
gas duct after the bypass connection. The damper provided a means of
varying the recycle percentage.
Oxygen Equipment
The oxygen supply equipment was provided by Linde Division of Union
Carbide Corporation. This equipment included a LOX storage tank,
evaporators, safety valves, and flow control equipment. Argonne, Linde,
and Black Hills Power personnel reviewed the project from the safety
standpoint. All piping was installed according to Linde recommendations.
This included pickling the one-inch copper supply lines, silver soldering
all fittings, and purging the system with the nitrogen before each
operation. Interlocks were installed to shut off the oxygen due to
parameter excursions such as high bypass duct temperature, no bypass
flow, high excess oxygen, and high stoker temperature.
Boiler Operation
The boiler operation had three phases, normal air fired, transition to
oxygen-flue gas, and oxygen-flue gas test. The boiler was started
under air fired conditions. The slide gate at the Force Draft Fan was
opened to the atmosphere, the slide gate at the sparger was closed and
the damper was fully opened. The transition to oxygen enriched was
relatively simple because the coals in the fuel bed would "hold the
fire" during the transition. Recycle was started by closing the two
slide gates and partially closing the damper. The oxygen feed was
started and gradually increased to the test level. Forced Draft and
Induced Draft fan settings were then adjusted to get the least negative
wind box pressure.
Observations of the fuel bed seemed to indicate that the size of the
fuel bed was a governing factor on efficient combustion. This was
indicated by high level of CO readings when the bed was at a larger
than normal level even though excess oxygen was observed in the flue
gas. When the bed burned into the stoker slot (grate), excessive grate
temperatures were sometimes observed. However, CO levels were very low
and flame conditions were very good with a small bed. The operator
observed that when firing with the small bed (03 enriched) the
clinkering appeared to be less than during normal operations. Also,
the stack was visually cleaner. The boiler required more operator
attention during Og enriched firing.
Conclusions of Additional Study
The tests provided some very useful information concerning the possible
retrofit of an existing unit to C0£ recirculation.
1. The transition from air to 03 enriched firing did not
seem to be difficult. It was accomplished several
times during the tests without any problems at all.
-------
105
2. No extraordinary training was required for the operators
to feel comfortable operating the boiler under recycle
conditions.
3. Gas leakage, both air into the system and flue gas out,
will require significant attention with any commercial
unit. Apparently, minor leaks can limit the purity of the
C02 in the recycled gas. Existing stoker-fired boilers
that were not designed for leak tight boiler settings would
be difficult to retrofit. Pulverized coal boilers, especially
those with pressured furnaces, would probably be good
candidates for retrofit. These types of boilers are already
designed with relatively tight settings. Moderate modifica-
tions would be required to obtain a suitable leak-free system.
Blanketing the coal bunker with 00^ will probably improve the
CC>2 levels in the recycle gas significantly.
4. Safety considerations in a commercial unit should address the
hazards of handling the pure oxygen. Safety interlocks that
would prevent explosions in the boiler from high 03 levels
would be one item. Prevention of pure oxygen leaks into the
plant would be addressed as in any other industrial facility.
The other major safety consideration would concern the fact
that Carbon Dioxide and Carbon Monoxide are not life supporting
atmospheres. Permanent monitoring and alarms would probably
be required to prevent accumulations of high concentrations
in the plant. Special precautions would also be required for
the entry to enclosed areas such as coal bunkers.
Economical consideration will dictate which method of CC"2 recovery Wyodak
would recommend using. The Dow Chemical FT-2 process or the Argonne oxygen
process. The timing for construction of a C02 plant will ultimately depend
on the price of crude oil.
-------
POTENTIAL SOURCES AND USERS OF C02
MONTANA
GILLETTE
OIL FIELD
OIL FIELD
OIL FIELD
IQ
C
..—J-—'Efv
• ***^^^Jfc ^^^
PLANT
OIL FIELD
OIL FIELD
KIRK PLANT
BEN FRENCH
WYOMING
*••••••••••
SOUTH DAKOTA
-------
107
RECOVERING CO2 FROM STATIONARY COMBUSTOR&
A BONUS FOR ENHANCED OIL RECOVERY AND THE ENVIRONMENT
Alan M. Wolsky and Caroline Brooks
Energy and Environmental Systems Division,
Argonne National Laboratory, Argonne, Illinois
ABSTRACT
Argonne National Laboratory is conducting research on a new approach to recovering
carbon dioxide from stationary combustors. This research is aimed at providing the
private sector with the information it needs to decide whether the approach can
contribute to future supplies of carbon dioxide for enhanced oil recovery. The approach
includes the simultaneous recovery of other gaseous combustion products, such as oxides
of sulfur and oxides of nitrogen. The product stream, essentially all carbon dioxide,
could be used for enhanced oil recovery. The approach also may find application where
strict air pollution controls are mandated.
INTRODUCTION
Enhanced recovery of oil by carbon dioxide flooding is one of the fastest growing oil
production methods in use today, with the possibility that 500,000 barrels per day could
be produced by this method by the year 2000. According to the National Petroleum
Council, oil production by CO^ miscible flooding will surpass thermal recovery in 20
years. Currently, carbon dioxide sources fall into two broad categories: natural
deposits (or plants at which carbon dioxide is already recovered and vented) or plants
with carbon dioxide present in dilute vented streams. Argonne National Laboratory has
concentrated its research into carbon dioxide recovery on the second category,
specifically on recovery from power-plant flue gases.
The conventional approach to post-combustion recovery of carbon dioxide is to separate
it from stack gas, which is expensive because carbon dioxide is a relatively small fraction
of the stack gas (about 15% by volume), and because the stack gas includes various
molecular species that interfere with carbon dioxide recovery. Conventional techniques
use monoethanolamine (MEA) or hot potassium carbonate systems, or variations of such
systems, to absorb carbon dioxide from stack gas. The absorbing material is then
regenerated and the carbon dioxide is driven off and recovered as a gas. Difficulties
with the conventional approach include the expense of boiler duty for regeneration, the
need to reduce the concentration of oxygen in the flue gas, and the fact that sulfur
oxides (SOX) poison the solvent. The last consideration is very important when
considering carbon dioxide recovery from combustion of heavy oil or coal.3
-------
108
The concept being investigated by Argonne permits total recovery of the carbon dioxide
and other gaseous combustion products (SOX and nitrogen oxides, NOX). In this approach
the procedure is as follows:
• Incoming air is separated into O2 and N2 streams by a facility adjacent
to the combustor. The nitrogen is immediately returned to the
atmosphere or, in a fortunate case, sold as a by-product.
• The oxygen stream is mixed with carbon dioxide and other inert gases
from a recycled flue-gas stream.
• This mixture (which is 70% CO2 and 30% O2 by volume) is used instead
of air to burn the fuel.
• After heat exchange is complete, the resulting gas (about 91% CO2,
7% H2O, and 2% O2 by volume, with small quantities of SOX, NOX, etc.)
is divided into a product stream and a recycled stream.
• Finally, the product and recycled streams are further conditioned as
desired (water could be removed, leaving a stream that is 95% CO2, 3-
4% O2, and 1-2% SOX and NOX by volume; carbon dioxide with liquid
water may corrode pipelines, while "dry" carbon dioxide will not).
Since this approach involves no gaseous emissions after combustion, a combustor
operated this way would need no air-pollution control equipment, either conventional or
of an advanced type.
STATUS OF RESEARCH
Argonne has conducted five research projects to prove this concept. These projects,
described in detail in Refs. 4-8, are listed and discussed briefly here:
• An Argonne-designed experiment, performed by Battelle Columbus
Division, to obtain sufficient experimental data to identify relative
differences between coal-air and coal-CO2-O2 flames.4
• Computer modeling to simulate the heat transfer that results from
burning coal in a mixture of CO2 and O2 rather than in air.**
• Experimental testing using wet recycle, at the 2-million-Btu/h scale.6
• Evaluation, with the help of a detailed furnace computer model, of the
impact of using CO2-O2 or CO2-H2-O2 mixtures as an oxidizer (instead
of air) on the thermal performance of a coal-fired boiler.®
• Experimental testing using wet and dry recycle, at the 10-million-Btu/h
scale.7
-------
109
Evaluation of Firing Pulverized Coal in a CO2-O2 Atmosphere
This project compared the performance of a test combustor firing coal in a CO2-O2
mixture to performance when firing in air, and it provided a basis for inferences about
the effect of this substitution in larger combustors. Specifically, this work:
• Fired coal in air and in CO2-O2 at three mixture ratios,
• Measured the important combustion parameters for both coal-air and
coal-CO2-O2 combustion,
• Determined the mixture of carbon dioxide and oxygen that provides a
flame with a total radiant heat flux similar to that of a coal flame
burning in air, and
• Estimated the effect of the combustion atmosphere composition on
combustion efficiency, emissions, deposits, and other items of interest to
boiler designers.
This multiple-test experiment, designed by Argonne, was conducted by Battelle Columbus
Division at its combustion facilities. A water-jacketed, refractory-lined cylindrical
furnace (2 ft x 7 ft) was used, with tubes simulating a superheater placed downstream.
The furnace was fired with pulverized coal (about 400,000 Btu/h). A baseline test used
air as the combustion atmosphere; other tests used various mixtures of carbon dioxide
and oxygen.
These tests resulted in important evidence of the technical feasibility of the Argonne
approach, providing data for comparing coal-CO2-C>2 firing with coal-air firing. Results
indicated that the process, firing coal in a large utility boiler in an atmosphere of
recycled flue gas and added oxygen, is technically feasible. This conclusion was based on
the similarities between firing coal-air and coal-CC^-C^ in regard to combustion
characteristics, radiant heat transfer, and emissions.
Model of Furnace Heat Transfer for Combustion in CO2-O2 Atmospheres5
During the Battelle Columbus test, Argonne developed a one-dimensional model of heat
transfer from a cylindrical combustor. The model simulates the heat transfer from
fossil-fuel combustion when air or a CO2-O2 mixture is used as the oxidant. The coal
feed rate, combustor dimensions, and other model parameters are the same as those
specified by Argonne and used in the Battelle tests described above.
The Argonne model effectively predicted heat transfer in the coal-air burn and in the
three CO2-O2 burns. These findings lend further credence to the experimental results
cited above and to the general feasibility of burning coal in CO2-O2 rather than air.
-------
110
The small-scale tests conducted by Battelle suggest that combustion of coal with
mixtures having CO2-to-O2 molar ratios between 2.23 and 2.42 yields heat transfer and
combustion characteristics similar to those seen in air. These tests were conducted using
mixtures of pure CO2 and O2 to look at the feasibility of the fundamental process. The
next steps in the development process were to consider some of the practical aspects of
the process, including:
• Evaluating, at a realistic scale, the practical feasibility of converting a
furnace system from coal-air combustion to coal combustion in a mixture
of oxygen and recycled flue gases.
• Identifying the ratio of recycle gas to input oxygen that is needed to
achieve heat transfer performance similar to that of coal-air
combustion, and quantifying any changes in important parameters (such
as burnout and flame stability) that might affect overall system
performance.
• Providing a basis for scaling experimental results up to larger
commercial, utility-scale equipment.
The Argonne approach raised crucial questions for research: Will fuel (particularly coal)
burn normally in mixtures of carbon dioxide and oxygen or in mixtures of carbon dioxide,
oxygen, and water? If it will, will normal heat transfer take place with such a burn?
What practical problems will be encountered when retrofitting Argonne's new method to
an existing furnace being operated by its usual staff? To answer these concerns, Argonne
directed a project with the Black Hills Power and Light Company in Rapid City,
South Dakota.
Tests to Recover CO2 at the Black Hills Power and Light Company6
A 2.2-million-Btu/h, coal-fired, stoker-fed boiler was retrofitted for wet-recycle CO2
recovery by the staff of Black Hills Corp., the owners and operators of the furnace. Two
related modifications — sealing the brickwork supporting the boiler and blanketing the
coal bunker with carbon dioxide ~ were beyond the scope of this retrofit and test,
although they would be necessary for practical operation of a stoker furnace retrofitted
for recovery of carbon dioxide. Linde Division of Union Carbide provided oxygen and the
associated plumbing, and Argonne provided instrumentation and staff to monitor the
tests.
The modified utility boiler was instrumented to examine the feasibility of producing and
recovering carbon dioxide by burning coal in oxygen and recycled flue gas in a utility
environment. The tests demonstrated that the boiler can be operated in the oxygen-
blown/flue-gas-recirculation mode without any noticeable effects on coal combustion,
heat delivery to the water, or the coal-feed or ash handling systems.
Pretest calculations showed that a feasible set of operating parameters for a CO -
producing combustor system (tightly sealed against air infiltration and containing no
-------
Ill
more than about 5% O2 [dry basis] at the furnace exit) would be a flue-gas recycling
ratio between 0.6 and 0.7 and an oxygen feed rate of 1.17 g-moles per g-atom of carbon,
yielding an exhaust gas composition (wet basis) of approximately 46.9% CO2, 50.6% H2O,
and 2.5% O2 (dry basis). However, because air leaked into the test combustor and the
flue-gas handling system, the highest carbon dioxide concentration achieved in the
exhaust gas was 48.5% (dry basis). Major sources of in-leakage were the furnace
brickwork, the gas-handling system, and the coal-feed and ash-extraction systems.
o
Two-Dimensional Modeling of Fossil-Fueled Power Plant Behavior0
A comprehensive analytical study investigated how the thermal performance of a utility
boiler is affected when air is replaced by mixtures as the oxidizer. The study was
'performed using an Energy and Environmental Research Corporation heat transfer and
combustion zone model that incorporates state-of-the-art methods for predicting the
performance of fossil-fuel-fired boiler furnaces.
The model is based on local heat and mass balances solved for various arrangements of
furnace zones. Radiative heat exchange between all furnace zones, which is the
dominant mode of heat transfer in the radiant section of a boiler, is accurately simulated
by use of Monte Carlo calculation techniques. The model requires specification of
certain input data, including a description of the furnace flow distribution and the
distribution of wall deposits, which are considered to be the key parameters. The boiler
selected for the performance study was a tangentially fired coal combustor.
The study indicated that optimal CO2-O2 or CO2-O2*H2O molar ratios exist at which a
particular boiler can be operated with these mixtures in a way that performance changes
are minimal compared to the air operation for which the boiler was designed. These
ratios were later found to be compatible with the experimental results cited below. The
main criterion for determination of the optimal molar ratios was achievement of heat
transfer efficiencies (for the dry- or wet-recycle process) that are, at full load, the same
as for air operation.
Pilot Tests to Simulate a Typical Utility Boiler Fired with Pulverized Coal7
While tests were underway at the Black Hills plant, pilot-scale experiments were being
conducted by the Energy and Environmental Research Corporation at its Tower Furnace
facility. The tests were.conducted at a scale of 10 million Btu/h with the facility
configured to simulate both the geometry and thermal environment (temperature-time
history) of a typical utility boiler fired with pulverized coal. The Tower Furnace, which
is fired by a single, variable-swirl coal burner, has multiple access ports for sampling and
observation, incorporates many features characteristic of full-scale boilers (such as a
simulated superheater section, a tubular air heater, and fly ash removal), and is equipped
with a full complement of measurement and control instrumentation.
The base program was conducted with a series of trials to establish the optimal flue gas
and oxygen mixture that would produce performance matching conventional combustion
-------
112
in air. The baseline air condition was characterized with a subbituminous coal from
Wyoming. A range of recycle gas/oxygen mixtures were tested and the resulting
performance compared with the baseline air case. During this basic test program, the
flue gas was wet — that is, the water vapor from combustion was not removed. A
comprehensive series of measurements were made to quantify (1) carbon burnout,
(2) flame stability, (3) heat transfer performance in both the radiant and convective
furnace section, and (4) slagging, fouling, and ash deposition throughout the system.
In addition to this base program, a series of optional tests were conducted to more fully
characterize this recycle gas/oxygen combustion process and extend the evaluation to a
wider range of operating conditions. These additional tests included:
• Evaluation of a highly volatile, bituminous western coal,
• Evaluation of reduced load operation with the recycle gas/oxygen
combustion process,
• Modification of the furnace system to accommodate evaluation of dry
recycle gas,
• Detailed in-furnace measurements to fully characterize selected
conditions, and
• Two-dimensional heat transfer modeling of the furnace performance to
provide a tool for extrapolation and evaluation of data, as well as a link
to other experimental and theoretical studies.
This program demonstrated that pulverized coal can be burned satisfactorily in mixtures
of pure oxygen and recycled flue gases, under conditions representative of utility
boilers. Optimal flue gas recycle ratios were found for which performance changes were
minimal compared to operation on air. For the wet-recycle process, where flue gases are
recycled without drying, the optimal recycle ratio was found to be about 3.25. For the
dry-recycle system, where a substantial fraction of the flue gas moisture had been
removed, the corresponding optimal recycle ratio was found to be 2.6.
For both recycle conditions, measurements showed the heat transfer to decrease with
increasing recycle ratio, with heat transfer to the cooled water-wall panels showing a
slighter stronger dependency on the ratio than did heat transfer to the hot refractory
walls. Although the scale of the experimental system is still small (10 million Btu/h)
compared to full-scale utility boilers, care was taken to simulate overall heat transfer
characteristics, and the heat transfer results are believed to be a favorable indication of
the potential for full-scale application.
Other performance parameters — such as flame stability, carbon burnout, and slagging
and fouling tendencies — were found to undergo minimal changes for optimal recycle
conditions, compared to baseline operation in air. However, NOX and SO emissions were
found to be quite sensitive to the recycle process. Surprisingly, the emissions of NO
were reduced by about 70% under optimal dry-recycle conditions and by about 80% for
-------
113
wet recycle, compared to baseline values recorded for air. Emissions also decreased in
direct proportion to the recycle ratio applied. This behavior is believed to be due to
incineration (or reburning) of recycled NOX as it passes through the main combustion
zone, and it is a characteristic expected to recur in full-scale applications.
Emissions of SOX also were substantially reduced under recycle conditions. For wet-
recycle operation, this result is believed to stem from a relatively inefficient participate
removal device, which allows fly ash to be recycled through the furnace, providing an
opportunity for enhanced use of the inherent alkali material. Under dry-recycle
operation, the water removal device acts like a wet scrubber, also increasing the use of
alkali material in the fly ash.
In operating the furnace system, few problems were experienced. It was necessary to
prevent air ingress into the system that would reduce the purity of the carbon dioxide
product. In addition, it was necessary to install additional fan capacity to handle the
required volume of hot, recycled flue gases. Air in-leakage limited the carbon dioxide
concentration in the flue gas to 94% in the test furnace, an acceptable concentration.
CONCLUSION
Overall, the results of the studies described above, along with results of the other work
Argonne has undertaken concerning the recovery of carbon dioxide from stationary
combustors, indicates that the process may be applied successfully as a retrofit to a wide
range of utility boiler and furnace systems.
REFERENCES
1. Taber, J.J., Need, Potential and Status of CO2 for Enhanced Oil Recovery, in
Recovering Carbon Dioxide from Man-Made Sources (Proceedings of a Workshop
Held in Pacific Grove, California, February 11-13, 1985), Argonne National
Laboratory Report ANL/CNSV-TM-166, p. 11 (Oct. 1985).
2. Enhanced Oil Recovery, National Petroleum Council, Washington, D.C. (1984).
3. Wolsky, A.M., A New Method of CO2 Recovery, Proc. 79th Annual Meeting of the
Air Pollution Control Assn., Minneapolis (June 1986).
4. Weller, A.E., et al., Experimental Evaluation of Firing Pulverized Coal in a CQ^/^2
Atmosphere, prepared by Battelle Columbus Division, Argonne National Laboratory
Report ANL/CNSV-TM-168 (Oct. 1985).
5. Berry, G., N. Reddy, and A. Wolsky, Computer Simulation of Furnace Heat Transfer
for Coal Combustion in CO2/O2 Atmospheres, Argonne National Laboratory Report
ANL/CNSV-55 (June 1986).
-------
114
6. Kumar, R., et al., Tests to Produce and Recover Carbon Dioxide by Burning Coal in
Oxygen and Recycled Flue Gas: Black Hills Power and Light Company, Customer
Service Center Boiler No. 2, Rapid City, South Dakota, Argonne National
Laboratory Report ANL/CNSV-61 (Dec. 1987).
7. Abele, A.R., et al., An Experimental Program to Test the Feasibility of Obtaining
Normal Performance from Combustors Using Oxygen and Recycled Gas Instead of
Air, prepared by Energy and Environmental Research Corp., Argonne National
Laboratory Report ANL/CNSV-TM-204 (Dec. 1987).
8. Richter, W., W. Li, and R. Payne, Two-Dimensional Modeling of Fossil-Fueled
Power Plant Behavior When Using CO^-Op or ^-^2~^2~^2 Mixtures, Instead of Air,
to Support Combustion, prepared by Energy and Environmental Research Corp.,
Argonne National Laboratory Report ANL/CNSV-TM-187 (June 1987).
-------
115
CONFERENCE AGENDA
March 19, 1987
9:00 a.m. Introduction and Welcome
Alan Wolsky, Argonne National Laboratory
9:15 a.m. Current Status of the Use of CO2 for Enhanced Oil Recovery
Joseph. Taber, New Mexico Petroleum Research Center
10:00 a.m. Break
10:15 a.m. A New Method for Recovering CO2 from Stationary Combustors
Alan Wolsky, Argonne National Laboratory
11:00 a.m. Description of a Planned CO2 Recovery Project in Wyoming
Dick Tapper, Wyodak Resources
11:45 a.m. Lunch
1:00 p.m. Description of a North Sea Enhanced Oil Recovery Project
Ray Park, Oil and Petrochemical Consultant (U.K.)
1:45 p.m. Potential CO2 Sources, Costs and Risks
William B. Johnson Jr., Big Three Industries
2:30 p.m. Break
2:45 p.m. Potential Need for Man-Made CO2 in Enhanced Oil Recovery
Tom Shepard, Production Operators, Inc.
3:30 p.m. Discussion Session
4:00 p.m. Adjournment
March 20, 1987
9:00 a.m. Introduction
Alan Wolsky, Argonne National Laboratory
9:15 a.m. Environmental Issues of Coal Combustion
Charles Hakkarinen, Electric Power Research Institute
10:00 a.m. A Perspective on the Greenhouse Effect and CO2 Flue Gas
Recovery for EOR
Ralph Rotty, t/niversity of New Orleans
10:45 a.m. Break
11:00 a.m. Description of the.Test Results of the Argonne Coal
Oxygen Process
Roy Payne, Energy and Environmental Research Corp.
11:45 a.m. Discussion Session, Lunch, and Concluding Remarks
Alan Wolsky, Argonne National Laboratory
1:30 p.m. Adjournment
-------
CONFERENCE PARTICIPANTS
116
T. M. Allen
Pritchard Corp.
8205 West 108 Terrace
Overland Park, KS 66210
John M. Angelovich
Pacific Power & Light Co.
920 SW 6th Avenue
Portland, OR 97204
Fred P. Apffel, Vice President
Flexivol, Inc.
13135 Champions Drive, #200
Houston, TX 77069
Bruce Joseph Bolduc
Chem Systems, Inc.
303 South Broadway
Tarrytown, NY 10591
Caroline Brooks
Energy & Environmental Systems Div.
Argonne National Laboratory
Argonne, IL 60439-4815
William Mark Campbell
Sun Exploration & Production Co.
P.O. Box 830936
Richardson, TX 75083-0936
Clifford H. (Buddy) Collen
Alpine Operating Co.
P.O. Box 50235
Amarillo, TX 79109
Jerome F. Collins
Office of Industrial Programs
U.S. Department of Energy
1000 Independence Avenue, SW
Washington, DC 20585
Kevin Corbley
Enhanced Recovery Week
1401 Wilson Boulevard
Arlington, VA 22209
Dave A. Craig
Mobil Producing Texas and New Mexico
#9 Greenway Plaza, Suite 2700
Houston, TX 77046
George K. Crane
Southern California Edison Co.
P.O. Box 800
Rose mead, CA 91770
Bruce Cranford
Office of Industrial Programs
U.S. Department of Energy
1000 Independence Avenue, SW
Washington, DC 20585
G. A. Cremer
Shell Development Co.
P.O. Box 4452
Houston, TX 77210
Tom Fuller
Wyodak Resources Development Corp.
625 Ninth Street
Rapid City, SD 57709
Jerome L. Glazer
Air Products - Separex Div.
P.O. Box 538
Allentown, PA 18105
Charles Hakkarinen
Electric Power Research Institute
3412 Hillview Avenue
Palo Alto, CA 94303
-------
117
Robert Heigold
Delta Project, Inc.
P.O. Box 5244, Station "A"
Calgary, Alberta
Canada T2H 2N7
Stan Hemmeline
Mobil Producing Texas and New Mexico
#9 Greenway Plaza, Suite 2700
Houston, TX 77046
William B. Johnson, Jr.
Big Three Industrial Gas, Inc.
P.O. Box 3047
Houston, TX 77253
H. Sho Koboyashi
Union Carbide Corp.
Old Saw Mill Road
Tarrytown, NY 10591
Larry D. Long
Exxon Company, U.S.A.
Box 1600
Midland, TX 79702
Edward P. Lynch, Consulting Engineer
c/o Energy & Environmental Systems
Div.
Argonne National Laboratory
Argonne,IL 60439-4815
John M. McNeill
NCI Membrane Systems, Inc.
4676 Admiralty Way No. 602
Marina Del Rey, CA 90292
Kent B. Me Reynolds
Dow Chemical U.S.A.
1691 North Swede Road
Midland, MI 48674
Joneil R. Olds
Amoco Production Co.
P.O. Box 800
Denver, CO 80201
Yoram S. Papir
Chevron
575 Market Street
San Francisco, CA 94105
Raymond Scott Park
Oil and Petrochemical Consultant
Charter House
Lord Montgomery Way
Portsmouth PO1 2SU
United Kingdom
Roy Payne
Energy and Environmental Research
Corp.
#18 Mason
Irvine, CA 92718
Nelson B. Peterson
Hudson Engineering Corp.
P.O. Box 218218
Houston, TX 77218
Fred A. Pettersen
Chevron Research Co.
576 Standard Avenue
Richmond, CA 94802
Brad Petzold
Enron Gas Processing Co.
110 North Marienfeld
Midland, TX 79701
Kenneth A. Pritchard
International Permeation, Inc.
P.O. Box 5244, Station "A"
Calgary, Alberta
Canada T2H 2N7
Ralph M. Rotty
University of New Orleans
Lakefront Street
New Orleans, LA 70148
Chuck P. St. Laurent
Shell Oil Co.
200 North Dairy Ashford Road
Houston, TX 77079
-------
118
Bill Saulmon
Shell Oil Co.
P.O. Box 2463
Houston, TX 77001
Thomas H. Shepard
Production Operators, Inc.
11302 Tanner Road
Houston, TX 77041
Patrick L. Simiskey
Dow Chemical Co., B-1605
Freeport, TX 77541
Frederick (Tom) Sparrow
Purdue University
c/o Energy & Environmental Systems
Div.
Argonne National Laboratory
Argonne, IL 60439-4815
Herbert W. Spencer
Joy Manufacturing Co.
4565 Colorado Boulevard
Los Angeles, CA 90039
Wallace A. Stanberry
Transpetco, Inc.
625 Market Street, Suite 200
Shreveport, LA 71101
Ken M. Stern
Chem Systems, Inc.
303 South Broadway
Tarrytown, NY 10591
Joseph J. Taber
New Mexico Petroleum Research Center
c/o Energy & Environmental Systems
Div.
Argonne National Laboratory
Argonne, IL 60439-4815
Ram Tarakad
The M. W. Kellogg Co.
3 Greenway Plaza East
Houston, TX 77046
Rod Taylor
Pacific Power & Light Co.
P.O. Box 720
Casper, WY 82602
Richard J. Tupper
Wyodak Resources Development Corp.
625 Ninth Street
Rapid City, SD 57701
Dale H. Vander Wai
Liquid Carbonic Corp.
135 South La Salle Street
Chicago, IL 60603-4282
Donald E. Wain
Pacific Power & Light Co.
1591 Tank Farm Road
Glenrock, WY 82637
F. Brian Walter
United Engineers
700 South Ash Street
Denver, CO 80217
Bill R. Wiggins, Jr.
Carbon Dioxide Associates, Inc.
P.O. Box 463
Houston, TX 77001
Robert A. Wojnarowski
Koch Process Systems, Inc.
20 Walkup Drive
Westborough, MA 01581-5003
Alan M. Wolsky
Energy & Environmental Systems Div.
Argonne National Laboratory
Argonne, IL 60439-4815
------- |