United States Solid Waste And EPA 51O-K-92-802
Environmental Protection Emergency Response November 1988
Agency 5403W
&EPA Tank Corrosion Study
^Printed on Recycled Paper
-------
-------
FINAL REPORT
TANK CORROSION STUDY
EPA ASSISTANCE ID NO. X-813761-01-0
By
JAMES H. PIM, P.E.
JOHN M. SEARING
SUFFOLK COUNTY DEPARTMENT OF HEALTH SERVICES
15 HORSEBLOCK PLACE
FARMINGVILLE, LONG ISLAND, NEW YORK, 11728
NOVEMBER 1988
For
OFFICE OF UNDERGROUND STORAGE TANKS
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
401 M STREET, S.W.
WASHINGTON, D.C. 20460
ATT: MR. DAVID O'BRIEN
-------
COUNTY OF SUFFOLK
PATRICK G. HALPIN
SUFFOLK COUNTY EXECUTIVE
EPA 510-K-92-802
DEPARTMENT OF HEALTH SERVICES
DAVID HARRIS. M.D., M.P.H.
COMMISSIONER
November 17, 1988
Mr. David O'Brien
Office .of Underground Storage Tanks
U.S. Environmental Protection Agency
401 M Street, S.W.
Washington, D.C. 20460
i
Dear Mr. O'Brien:
In accordance with EPA Assistance ID No. X-813761-01-0
enclosed please find Suffolk County's final report on the Tank
Corrosion Study.
Very truly yours,
David Harris, M.D., M.P.H.
Commi s s i one r
DH/lst
Enclosure
Printed on Recycled Paper
225 KABRO DRIVE EAST
HAUPTAUGE. N.Y. 1178«
«t 01348-2000
-------
TABLE OF CONTENTS
INTRODUCTION
2
SUMMARY
BACKROUND
tmv 4
GEOLOGY AND HYDROLOGY
SOILS
DEVELOPMENT PATTERN
TANK REGULATION ........... ....... ' ..................... L
TANK POPULATION .............. .' ' ..................... '
••••<............ .......... y
PROCEDURE ........ . ...................
CONTENTS OF TANKS ............ ...... ...................... 1U
CAUSE OF PERFORATIONS ...... ............................ 77
TANK WALL THICKNESS ....... ____ ......................... f ,
AVERAGE SIZE OF PERFORATION . ! '. '. ......................... 7o
AVERAGE THICKNESS BY TANK VOLUME . . . .' ! ................... VA
Perforated Tanks ................ ......... ' ...... fT
Non-perforated Tanks ......... ................ " ...... 7?
AGE OF TANKS .......... ...................... Jl
VOLUME OF TANKS ...... .............................. }%
LOCATION OF PERFORATIONS .!.'!.'!.'!."!!! .................... T?
GROUNDWATER LEVEL . . . . ..... .................. }7.
BACKFILL CONDITIONS . . ____ .' .............................. H
PERFORATIONS VERSUS LEAKAGE .'.'.'.'.'. ...... ' ................ on
Fuel Oil ........ . .............. ; ......... ' .......... *"
Gasoline .............. ..... .................. ^
AGE VERSUS VOLUME - Perforated taAks * ! .................. ' o V
VOLUME VERSUS AGE ................ .....'.'.'.'.'.'.'.'.'.'. ..... -" 22
Tanks without perforations (greater than' '4666' .......
gallons ) .........................
Ta?£nnW±t??Ut Perf°rations (less than and equal 'to
4000 gallons) .......................... ^ 23
OBSERVATIONS ................................
•**•••••••*•• ^i J
FUEL OIL TANKS ..... . ................
General Statistics ...... ........ ........................ ^ ,
Analysis ....... ...... ...... ............... ............... TO
CONCLUSIONS .........................
-------
APPENDIX A
Exerpt Portions of Article XII
APPENDIX B
38
38
39
Suffolk County Tank Removal Standard .'.'.'.' 39
APPENDIX C
Tank Corrosion Study Inspection Sheet
40
40
APPENDIX D
Non-corrodible Tanks . „ 4
END OF TABLE
41
ii
-------
LIST OF ILLUSTRATIONS
Figure 1 - Tank Contents 35
Figure 2 - Age of Tanks 35
Figure 3 - All Gasoline 36
Figure 4 - All Fuel Oil !!.'!.'.'.'!.'!!! 36
Figure 5 - Average Plate Thickness 37
Figure 6 - Cause of Perforation 37
111
-------
50,000 Gal. 100,000 Gal. or more No idea .
Approximate Buria'l "Depth Below
•burierL,Tank D^'lection; total length:__
Cminus)to tank top:.
A. tank diameter:
total length:
(minus)to tank top:
B. tank diameter:
Diameter*.A_i.:
(minus) Diameter B_._;
Deflection :
Description of Tnnk raTF.RIOK Corrosion;
Point Corrosion: nominal Jiald moderate severe
General Corrosion: nominal mild moderate severe
Description of Tank INTERIOR Corrosion: ;
mild moderate severe
BURIED
top:
i
i
'
'
n
ii
i ii
REMOVED
Point Corrosion: nominal_
General Corrosion: • nominal mild moderate_ ^severe
-------
FTNAT. RFPORT1
TANK CORROSION STUDY NOVEMBER 1988
INTRODUCTION
This is the final report for the United States Environmental
Protection Agency of the TANK_CORROSI.ON STUDY performed in
Suffolk County, New York by the Suffolk CounTy~Department of
Health Services. This report covers the observations made on 500
underground tanks spanning the time period from February 24 1987
to September 1, 1988. This report is the summation of four
interim reports plus observations and conclusions. The first
interim report was issued on July 31, 1987 and covered the first
1UU tanks. The second was issued on November 10, 1987 and
covered 200 tanks (including the first 100). The third interim
report was issued on February 8, 1988 and covered exempt heatinq
oil tanks. The fourth was issued on May 2, 1988 and: covered 320
The study was conceived as a means of .gathering information about
old buried steel tanks and the nature of corrosion that plaques
them, by closely observing them as they are removed from the
ground for disposal. Suffolk County was chosen for the studv
because a large number of tanks are being removed in a relatively
short time to meet the requirements of a local tank replacement
ordinance - Article XII of the Suffolk County Sanita-rv Code
(Appendix A) . J
The tanks involved in the study varied from 175 gallons to 50,000
gallons and from 2 years old to 70 years old. All but 18
?£rl,f,££?^ S?hS *?*?> °f- Petroleum Product. All of the 500 tanks
included in the statistics were plain welded steel tanks. Durinq
the Period of the study there were also 12 tanks other than plain
examined. The results of these were not included
wh
?owar.rf ?h 2 b"t instead were covered in a seperate section
toward the end of the report (Appendix D) .
-------
FINAL REPORT
T.ANK CORROSION STUDY
NOVEMBER 1988
SUMMARY
Five hundred plain steel tanks plus twelve corrosion protected
tanks were removed from the ground over an eighteen month period
in Suffolk County, Long Island, New York. They were examined
carefully before disposal to, gather statistics on the nature and
extent of corrosion that had attacked them. Information was
gathered on the number, type, location, arid size of perforations;
the general interior and exterior corrosion condition; soil,
backfill, and groundwater conditions; the presence of leaked
product; and tank statistics such as volume, plate thickness,
location, product, age, etc;. The statistics were compiled and
compared, observations made and conclusions developed.
The major conclusions can be summarized as follows:
1) Size is more important than age in predicting tank
failure;
2) In general, small tanks are much more likely to perforate
than large tanks due to the thinner walls found in smalle'r
tanks;
3) Compared to external corrosion, internal corrosion is
insignificant; - *
4) Fuel oil tanks are just as susceptible to perforation as
gasoline tanks of the same size; • > •
5) Existing tanks are in worse shape than is demonstrated bv
testing; > • •
6) Tanks do not always leak immediately upon perforation,
-------
FINAL -REPORT NOVEMBER IQfifl
TANK CORROSION STUDY NOVEMBER 1988
BACKROUND
GEOGRAPHY
Suffolk County is located in southeastern New York State and
encompasses the eastern two-thirds of Long Island it is
bordered on the west by Nassau County. The other three sides are
bounded by bodies of water: Long Island Sound to the north, Block
Island Sound to the east, and the Atlantic Ocean to the south.
The county has a land area of approximately 885 square miles. It
is 86 miles in length and varies in width (on the main body)
between 12 and 20 miles. The eastern end of the county is split
in two forks which are seperated by the Peconic Bay System
There are 5 significant islands on the east end which are also
under the county's jurisdiction.
The geographical features of Suffolk County are a result of the
last ice age, which ended some 12,000 years ago. Two lines of
terminal moraine hills were formed during this period. They
reach a maximum height of 400 feet above sea level and traverse
the length of the county. A moderately flat land surface (called
an outwash plain) forms most of the southern area of the county
This plain terminates at off-shore barrier beaches that are
seperated from the mainland by shallow bays. The north shore is
characterized by headlands that have been eroded away into
steeply vertical bluffs that reach almost 100 feet high in some
places. There are also several harbors and wetland areas alona
this shore. y
The updated 1988 planning data indicated a population of
approximately 1.37 million people with an additional transient
seasonal population of approximately 200,000 people.
The county's land use (of approximately 566,000 acres) is broken
down as follows: residential (25%), commercial (3%), industrial
(2%), transportation (8%), institutional (6%), recreational and
open space (14%), agricultural (9%), and vacant (33%). This is
based on 1981 figures and remains fairly accurate according to
latest planning estimates.
GEOLOGY AND HYDROLOGY-
The Upper Glacial Aquifer compromises the uppermost layer of the
land surface in the county. This layer consists of glacial
material, which itself consists mostly of sand. The whole layer
is composed of sand, gravel, clay, silt, organic mud, peat, loam,
and shells. The gravel ranges in size from pebbles to boulders,
and the sand from fine to very coarse. This composition creates
a^ filter-like effect allowing any liquid to percolate easily from
tr_e surface all the way down to the water table.
-------
FINAL REPORT NOVEMBER 1988
TANK CORROSION STUDY
All of Suffolk County's water supply comes from below its
surface. For this reason, the ynited States Environmental
Protection Agency declared the groundwater of Suffolk as a sole-
source aquifer. This means Suffolk is dependent on its
groundwater and the recharge capabilities of the ground to
maintain it's water supply.
The average rainfall is approximately 45 inches per year. For
the main body of the county, approximately 48 % of the
precipitation is lost to evaporation, 1.4 % is lost as direct
run-off, and the remaining 50.6 % is recharged. The water table
in the county is always above sea level and tends to vary
seasonally (by up to several feet in some areas). The water
table ranges between 0 and 110 feet above sea level, while the
land elevation varies between approximately 0 and 300 feet above
sea level.
SOILS
According to the US Soil Conservation Service1 there are 10
major soil associations in the county, depending on location and
relation to the glacial moraines and plains. Among these, there
are 18 soil series and 67 mapping units. The series are a more
specific classification of the soil associations and the mapping
units are a direct soil label. These all contain glacial sands.
The pH ranges from approximately 3.5 to 6.5, with .most soils in
the 4.5 to 5.5 range. This is more acidic than the average
United States soil. The corrosivity ranges from low to high
depending on soil type, location, and soil characteristics.2
The permeability ranges from < 0.2 inches per hour to > 6.3
inches per hour. The available moisture capacity of the soil
ranges from 0.01 inches of water per inch of soil (very low /
dry) to 0.2 inches of water per inch of soil (high / moist).
It has been previously established1'2that Suffolk County soil
corrosivity ranges from low to high (this is the entire range of
corrosivity - low, moderate, high). This rating is based on
several factors: drainage class and texture, acidity, resistivity
(field), and conductivity (saturated). Soil reaction (pH)
correlates poorly with corrosion potential and is not included in
the rating. But there is a notable exception - a pH of less than
4.0 almost always indicates a high corrosion potential.
.Resistivity values range from less than 2,000 ohm-cm for high
corrosivity potential to greater than 5,000 ohm-cm for low
potential. Suffolk County soil encompasses this entire range
with resistivity readings varying from 35 ohm-cm in tidal
locations to approximately 120,000 ohm-cm in typical dry, sandy
locations.
-------
FINAL REPORTNOVEMBER 19RR
TANK CORROSION STUDY wuvhMBER 19 88
The following is a description of each of the soil series.1
Carver Soils - Generally a coarse textured sandy loam. It is
excessively drained with a very low available moisture capacity
and rapid permeability throughout. The soil reaction (degree of
acidity or alkalinity) is strongly acid to very strongly acid fpH
range of 4. 5-5. 5). v
--Land - Generally a loam and sand mix associated with
Carver and Plymouth soils. It has a low moisture capacity and
follows most other characteristics of the mentioned series.
Haven Loam - A medium textured loam. It .is well drained with a
moderate to high available moisture capacity and varying
permeability (moderate in the upper layers and rapid in the lower
layers). The soil reaction is strongly acid to very strongly
acid .
Made Land - This type of land consists of many materials
including rubble, soil, non-organic material, and non-soil
material. Its characteristics are variable.
Montauk Soils - Generally a fine sandy loam and silt loam. It is
a medium to moderately coarse textured soil that is moderately -
well to well drained. It has a moderate to high available
moisture capacity with varying permeability (moderate to
moderately-rapid in the upper layers and moderately slow in the
lower layers). The soil reaction is strongly acid to very
strongly acid throughout.
Muck ~ Poorly drained organic soil. This type of soil is usually
located near wetlands or in areas of high water table.
PlYIB2Hth_LoamY_San.d - A coarse textured loamy sand. It is
excessively drained with a low to very low available moisture
capacity and varying permeability (rapid in the upper layers and
moderate in the lower layers). The soil reaction is stronqly
acid to very .strongly acid.
Riverhead & Haven Soils - Generally a medium to moderately coarse
textured loam or sandy loam.. They are well drained with a
moderate to high available moisture capacity and rapid to very
rapid permeability. The soil reaction is strongly acid to very
strongly acid.
Riverhead Sandy Loam - Similar characteristics as the above.
Tidal Marsh - Wet, sandy areas near bays and tidal creeks. They
are poorly drained areas.
Urban ~ This type of land has variable characteristics. It has
already been developed and the -soil characteristics modified.
-------
SUFFOLK COUNTY DEPARTMENT OF HEALTH SERVICES
FIELD INSPECTION FORM
FIRST DRAFT
Inspection Date
Time Start '
Time End
Time Spent Min,
Facility ID; Tank No. Prop.Tax No.
Facility Name:
Facility Address;
Type of Facility: Gas Sta. Car dealer Commercial___ Residential_
Industrial Other (describe)
Type of Tank: Plain steel Asphalt- or coal tar coated steel
STIP3 Other induced current cat.prot.steel .Imposed current cat.
prot.steel Bufhide Other fiberglas coated steel _,- Owen---Corning
*
fiberglas Xerxes fiiberglas Other fiberglas Other material
(describe) :. ,
Contents (Observed) : Gasoline ; Fuel oil if 2 Fuel oil *4__ Fuel
oil f|5 Fuel oil «6 Kerosene Diesel Gasahol (% mixture)
Jet fuel Av.gas Solverits_n__ (Describe)
Waste oil_;_ Other oil (Describe)
Other material (Describe)
Dimensions: D'iameter (x) ' Diameter (y) - * Length . '
Width ' Height ' Volume cu' , gal. (Calculated from
dimensions)
Date Installed: Present age
"•"""'""•'" " """"' ••-•-""-""•• L- »
End plate thickness " (mic,) " (Ultrasonic)
Wall thickness " (Ultrasonic)
-------
- 2 -
Holes ; Yes _ _ No _ Total No. observed _ Leak confirmed before
removal _ Dia. largest___ _ " Dia. smallest _ " Average dia. _ " (Es
Hole locations Bottom below fill___ Bottom below gage hole
Elsewhere along bottom _ __ ?tultiple along bottom _____ _
Top around fittings _ Eleshere on top _ Along groundwater line
' On side _ On end _ Multiple on sides and/or ends _
Cause of holes; Point corrosion internal _ Point corrosion external
General' Corrpsion internal _ General corrosion external _ Combination
internal and external corrosion _ Weld failure _ Mechanical damage
internal _ Mechanical damage external _
Sludge: Volume _ gal. Wt. _ . _ 8
Exterior coating: Yes _ No _ Completely intact _ Minor flaws _
Many failed areas _ Completely failed _ _ * Rcirainini;
coating: Yes _ No _ Fiberglas " lining Other
Completely intact _ Minor flaws _ Many failed areas _ . Completely
failed
Natural Soil Conditions: Clean sand "or gravel _ Clay _ Bog Loam
Sand w/some cTay _ Other _ _ (describe) _ _ _
Can't tell _
Backfill Conditions: Clean sand or gravel _ Same as natural soil
Concrete, asphalt, stones or rubble against tank__ Other
(describe) _ _ _ '
" Groundwater level; Always in contact with tank _ Sometimes in contact
with tank _ Never in contact, with tank _
Groundwater Quality :• Salinity (if near shore) _ PH Floating
i •^^^••—••••^
product in observation wells __ In excavation dissolved product _
Closest estimate of 'total product leaked: Ogal 10 Gal. 50 Gal.
100 Gal, _ 500 Gal _ 1,000 Gzl. _ 2,000 Gal. 5,000 Gal. 10,000 Ga!
-------
FINAL REPORT
TANK CORROSION STUDY
NOVEMBER 1988
The United States Soil Conservation Service has developed a
corrosivity rating2 based on several factors. For the soils
encountered in this study, the following ratings apply?
Carver Series
Haven Series
Montauk Series
Plymouth Series
Riverhead Series -
Low Corrosivity
Low Corrosivity
Low Corrosivity
Low Corrosivity
Low Corrosivity
These soils generally have an average soil acidity of less than
8 meg/100 gram and are well to excessively drained medium to
cnnnSS textured soils. Their resistivity is greater than
5000 ohm-cm and conductivity (of saturated extract) is less than
0.3 mmhos/cm.
Other soils such as tidal marsh and muck have a lower resistivity
and will tend to be placed in a higher corrosion category. These
soils typically have a high or fluctuating water table.
The initial corrosivity of the soil series leads to the premise
that Suffolk County soils are generally of low corrosivity,- all 4
determining factors considered.
DEVELOPMENT PATTERN
Prior to World War II, Suffolk County was primarily an
agricultural area with scattered small villages. After the war,
the wave of population that spread out from New York City
engulfing"neighboring Nassau County, rolled into western Suffolk
and spread eastward. Currently the western two-thirds of the
county are solidly developed and agriculture is restricted to the
eastern third, even though Suffolk still ranks as the number one
agricultural county in the state in production. It is an
affluent area, ranking amonig the highest in the country for
family income. The population is well-educated and therefore
especially environmentally concerned.
TANK REGULATION
This combination of factors: education, wealth, population
density, and restricted water supply has led Suffolk to take a
position of leadership in the field of environmental regulation.
It led the Suffolk County Board of Health in 1979 to pass
sweeping restrictions on the storage and handling of toxic
materials including underground petroleum storage. Until that
time, nearly all underground tanks installed in the county were
constructed of plain steel. Since that time, only non-corrodible
tanks such as fiberglass or cathodically protected tanks have
been installed.
-------
FINAL REPORT NOVEMBER 1988
TANK CORROSION STUDY
In addition to establishing new construction standards, the law
required replacement of all existing plain steel tanks by
January 1, 1990 with non-corrodible single- or double-walled
tanks.
The removal and replacement effort has been progressing steadily
for several years now and is beginning to accelerate as the due
date draws near. It was recognized some time ago that the very
large number of old steel tanks being removed constituted a
valuable source of information on tank corrosion.
This study was designed to take advantage of that resource and to
obtain practical information on the nature of tank corrosion that
might be useful in developing regulations in other areas.
TANK POPULATION
The tanks that were examined in this study constituted as random
a sample as was available on Long Island, being composed of every
tank of any type that was removed during the period of the study.
The reasons for removal were not documented, but included: 1)
compliance with the replacement requirement of the law (Appendix
A); 2) business expansion requiring greater storage capacity; 3)
new construction requiring removal to eliminate obstructions; 4)
test failure; and 5) change of business. The only major group of
tanks conspicuously missing from the study is the very small
heating tanks. Though there are some in the study, there are not
many because the law still does not require the replacement of
heating tanks or even testing of those under 1100 gallons.
Another group that could certainly be considered as
underrepresented would be the farm tanks. Because of the
difficulties in managing and carrying out a regulatory program in
the farming areas, less enforcement effort has been applied to
farmers and therefore fewer replacements have been made than in
the commercial and industrial areas.
There is one other factor that must have had some effect on the
true randomness of the studied sample forcing the results to the
conservative side. The tank regulatory program has been in
effect since 1980 and by the time the study started, about 1800
tanks had already been removed. This naturally resulted in the
removal of many of the worst tanks before the study began.
Tank testing statistics support this contention. By the time of
the beginning of the study, over 6000 tanks had been tested in
the county. The annual test failure rate had declined steadily
from about 15 % in 1981, to about 2 %, apparently indicating
that the tanks most likely to be leaking were being removed first
and were already gone by the time the study began.
-------
FINAL REPORT : NOVEMBER 1988
TANK CORROSION STUDY
PROCEDURE
In Suffolk County, by regulation, when tanks are removed or
abandoned, the Health Department must be notified beforehand.
The attached sheet (Appendix B) describes the required
abandonment procedure. •;
t
Nearly every tank that was abandon during the study period was
examined and is included |in the statistics. Two Health
Department sanitarians, Janet iSwords and Tom Nanos, were assigned
to accomplish this task.
An inspection information sheet was filled out by the inspector
for each tank during the inspection, a copy of which is attached
(Appendix C). The information was then computerized to create a
usable data base.
While a tank was being excavated, the process was observed and
notes taken on the condition pf. the backfill and any evidence of
leakage or spillage.
When the tank was removed, it was set on the surface near the
excavation and thoroughly cleaned of clinging dirt and scale by
the inspector and then very [carefully inspected (visually) for
any evidence of perforations. } If one was found, it was examined
closely and a determination |made as to the type of corrosion,
internal or external. Information was taken as to the size,
location, and number of perforations and any unusual conditions
described. The tank was measured for original dimensions and the
plate thickness measured with an ultrasonic tester.The ends were
then cut out of the tank by 'the contractor using a pneumatic
cutting device and the interior was cleaned of all remaining
sludge.
After the tank was cleaned and vented reasonably well, the
sanitarian inspected the intlerior from the end holes using a
light. Perforations could be ^een as points of light in the dark
interior, and these were checked against the information from the
outside. The condition of interior corrosion was described and
any unusual conditions "noted.
Warning
Tank abandonment is dangerous! During the time of this study,
there were three explosions of tanks and three fires which
fortunately resulted in only two cases of minor injury. All
accidents were caused by the extreme carelessness of the
individuals working on the tanks. They resulted from the use of
torches or abrasive cutting t|ools on tanks where they should not
have been used.
-------
FINAL REPORT NOVEMBER 1988
TANK CORROSION STUDY
FINDINGS
GENERAL STATISTICS
-Number of tanks inspected 500
-Number of facilities inspected 199
-Number of facilities inspected with
perforated tanks 84
-Number of tanks with perforations 143
-Percent of tanks with perforations 28.6 %
-Average number of tanks removed
per facility 2.51
-Average number of tanks with perforations
removed per facility .72
Making a generalization of the above numbers yields the
following:
1) For every 11 tanks removed, one would expect to find
approximately 3 tanks with perforations.
2) For every 2 facilities inspected that had perforated
tanks, one would expect to find approximately 3 tanks with
perforations.
-Number of Perforated Tanks that Showed
Evidence of Having Leaked Product 83
-Percent of Perforated Tanks that Showed
Evidence of Having Leaked Product 58.0 %
-Percent of Total Tanks that Showed
Evidence of Having Leaked Product 16.6 %
Total volume of the 500 tanks equalled 2,216,650 gallons. Of
these, the volume of those which were perforated was 315,525
gallons, or 13.6 % of the total volume.
10
-------
FINAL REPORT
TANK CORROSION STUDY
NOVEMBER 1988
CONTENTS OF TANKS
Material
Gasoline
•#2 Fuel Oil
Waste Oil
Diesel Fuel
Solvents
#4 Fuel Oil
Kerosene
Motor Oil
Waste Water/Oil
Aviation Fuel
Transmission Oil
#6 Fuel Oil
Caustic Soda
Jet Fuel - JP5
Sodium Hypochlor
Other Oil
Unknown
All
Tanks
233
128
43
33
14
11
11
8
4
3
3
2
2
2
. 1
1
!_
500
% of
All Tanks
(n/500) !
46.6 %!
25.6 %
8.6 %
6.6 %
2.8 %:
2.2 %
2.2 %
1.6 %
0.8 %
0.6 %
0.6 %
0.4 %
0.4 %
0.4 %
0.2 %
0.2%
0.2 %
100.0 %
Perf.
Tanks
73
39
7
11
6
1
3
1
0
0
1
0
0
0
1
0
0_
143
% of
Perf.Tanks
(n/143)
51.0 %
27.3 %
4.9 %
7.7 %
4.2 %
0.7 %
2.1 %
0.7 %
0 %
0 %
0.7 %
0 %
0 %
0 %
0.7 %
0 %
0 %
100.0 %
% of Tanks
Perforated
According
to Material
31.3 % .
30.5 %
16.3 %
33.3 %
42.9 %
9.1 %
27.3 %
12 ..5 %
0 %
0 %
33.3 %
0 %
0 %
0 %
100.0 %
0 %
0 %
NOTE: Upon further investigation, two of the tanks that were.
listed in previous reports as having contained aviation fuel were
reclassified as having contained jet fuel.
CAUSE OF PERFORATIONS
GENERAL
External Corrosion
Internal Corrosion
Combination Internal/External
Weld Failure
External Mechanical Damage <
(n)
108 =
9 =
21 =
4 =
143
(n/143)
75.5 %
6.3 %
14.7 %
2.8 %
0.7%
100.0 %
Types of External and Internal Corrosion Causing Perforations
(Note; This information is;extracted from above)
External
Point
General
Internal
Point
General
105 =
3 =
2 =
73.4 %
2.1 %
4.9 %
1.4 %
11
-------
FINAL REPORT
TANK CORROSION STUDY
NOVEMBER 1988
TANK WALL THICKNESS
Average Wall Thickness (inches)
Endplate
(Ultrasonic)
1) Avg. Thick.
(Perf.Tks.): .1764
# of records: 65
Range - Low: .0982
High: .2814
Endplate
(mic)
1747
89
0770
2812
Bottom
Plate
.1761
93
.1090
.2984
Top
Plate
.1778
60
.0780
.2852
2) Avg. Thick.
(Non-perf.Tks.): .2296
# of records: 156
Range - Low: .0981
High: .4672
,2248
205
1001
4^33
,2160
223
0921
,3920
,2203
151
,0927
,4150
3) Avg. Thick.
(All Tks.): .2139
# of records: 221
Range - Low: .0981
High: .4672
,2096
294
,0770
,4733
2043
316
2043
3920
,2082
211
,2082
,4150
Since not all tanks had readings taken and those which did, did
not necessarily have both ultrasonic and micrometer readings, one
or the other was used to obtain the following results. This
yielded a greater number of records for use in determining the
endplate thickness. All ultrasonic readings were used as the
primary readings, and micrometer readings used only on tanks
where the ultrasonic meter was not used.
-Item 1 total records for endplate thickness equals 99. Average
endplate thickness equals .1736 inches.
-Item 2 total records for endplate thickness equals 237. Average
endplate thickness equals .2253 inches.
-Item 3 total records for endplate thickness equals 336. Average
endplate thickness equals .2101 inches.
12
-------
FINAL REPORT
TANK CORROSION STUDY
NOVEMBER .1988
AVERAGE SIZE OF PERFORATION
-The average hole size for 143 tanks was .36 inches.
-The average hole size for >131 tanks on which perforation size
was measured was .39 inches..
-The average size for the largest holes was .55 inches based on
121 tanks. The large hole size rariged from .02 inches to 5.0
inches (see note).
-The average size for the smallest holes was .11 inches based on
93 tanks. The small hole size ranged from .02 inches to .40
inches.
Note: A total of twelve tanks were not included in the above
calculations for.the following reasons: A) Three (3) tanks were
not included because they [had perforations that were all in
excess of 20 inches (20"r 26", and 58"), and would not have
yielded a true representation of perforation size; B) Nine (9)
tanks did not have perfoation size measured. Additionally, if a
tank had only one perforation;, it was listed as the largest hole.
Of 143 perforated tanks, 99 had more than 1 perforation. The
average number of perforations for the population of perforated
tanks was 7 perfs./tank. The average number "of perforations for
the 99 tanks that had more than one perforation was 10
perfs./tank.
13
-------
FINAL REPORT
TANK CORROSION STUDY
NOVEMBER 1988
AVERAGE THICKNESS BY TANK VOLUME
Perforated Tanks
Volume
185
275
315
500 •
550
1000
1500
2000
3000
4000
5000
12000
Endplate
.1308
.1157
.1250
.0770
.1548
.1887
.1934
.1725
.1804
.1937
.2481
n/a
Average Thickness (inches)
Bottom Wall
.1341
.1244
.1250
n/a
.1589
.1720
.1884
.1736
.1789
.1747
.2596
n/a
n/a - measurement not available
Top'Wall
n/a
n/a
n/a
.0780
.1559
.1731
.1820
.1730
.1787
.1738
.2595
n/a
Non-perforated Tanks
Average Thickness (inches)
Volume Endplate
275 .1185
500 .1788
550 .1556
575 .1994
1000 .1887
1100 .1340
1500 .2142
2000 .1913
2500 .2406
3000 1937
3500 .3205
4000 .1937
5000 .2604
6000 ,2621
7500 .3039
8000 .2554
10000 2752
12000 .2388
15000 .3312
20000 .3116
25000 . .4068
30000 .3503
n/a - measurement not available
Bottom WaJ1
.1150
.1779
.1640
.1929
.1758
.1355
.2055
.1890
.2149
.1875
.2343
.1832
.2528
.2618
.2747
.2598
.2660
.2582
.3054
.2604
.3556
.3434
Top Wall
.1212
.1749
.1690
n/a
.1858
.. 1344
.2584
.1894
.2125
.1936
.2539
.1825
.2541
.2623
.2703
.26"63
.2771
.2535
.3062
.2810
.3544
n/a
14
-------
FINAL REPORT
TANK CORROSION STUDY
NOVEMBER 1988
AGE OF TANKS
Age in Years
70
60
57
55
50
48
47
46
44
43
41
40
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
3
2
UNKNOWN
Totals
All Tanks
1
1
1
1
1
5
1
1
12
7
2
4
2
1
3
1
2
6
5
14
3
17
14
22
10
9
26
10
9
23
16
15
25
28
18
16
19
8
7
30
7
3
5
1
1 :
87
500
Perforated Tanks
1
0
1
1
1
3
0
0
1
3
0
1
1
0
0
0
1
0
0
7
0
3
6
16
3
1
8
1
3
6
9
7
9
13
4
6
2
1
3
2
0
1
0
0
0
18
143
% of Tanks
Perforated
100.0 %
0.0 %
100.0 %
100.0 %
100.0 %
60.0 %
0.0 %
0.0 %
8.3 %
42.9 %
0.0 %
25.0 %
50.0 %
0.0 %
0.0 %
0.0 %
50.0 %
0.0 %
0.0 %
50.0 %
0.0 %
17.6 %
42.9 %
72.7 %
30.0 %
11.1 .%;
30.8 %
10.0 %
33.3 %
26.1 %
56.3 %
46.7 %
36.0 %
46.4 %
22.2 %
37.5 %
10.5 %
12.5 %
42.9 %
6.7 %
0.0 %
33.3 %
0.0 %
0.0 % .
0.0 %
20.7 %
15
-------
FINAL REPORT . NOVEMBER 19RR
TANK CORROSION STUDY NOVEMBER 1988
In most cases, tanks of unknown age were very old but there was
no way of determining exact age.
Average age of all tanks was 21.8 years old (excluding 87 of
unknown age). v y
Average age of perforated tanks was 23.4 years old (excludina 18
of unknown age). y
NOTE: Twelve tanks which were included in Interim Report 4 were
moved to another database because they were not 'true'
unprotected steel tanks - that is: they were either
fiberglass, fiberglass-coated, or cathodically protected. None
of them had perforations. This accounts for the discrepancy in
the eight year old age category and the loss of the six year old
age category. J
VOLUME OF TANKS
Perforated Tanks % Perf.
0 0.0 %
1 100.0 %
5 27.8 %
1 100.0 %
1 50.0 %
13 22.4 %
0 0.0 %
33 51.6 %
0 0.0 %
2 25.0 %
35 47.9 %
0 0.0 %
21 35.6 %
0 . 0.0 %
25 38.5 %
5 14.7 %
0 0.0 %
0 0.0 %
0 0.0 %
0 0.0 %
1 33.3 %
0 0.0 %
0 0.0 %
0 0.0 %
0 0.0 %
P_ 0.0 %
143
Average volume of tanks was 4433.3 gallons.
Average volume of perforated tanks was 2206.5 gallons.
NOTE: Two of the tanks moved/to the other ' non-corrodible '
database were 500 gallons, accounting for the discrepancy in this
category as compared to Interim IV.
16
Volume (gals)
175
185
275
315
500
550
575
1000
1100
1500
2000
2500
3000
3500
4000
5000
6000
7500
8000
10000
12000
15000
20000
25000
30000
50000
Totals
All Tanks
1
1
18
1
2
58
1
64
1
8
73
5
59
1
65
34
12
5
12
51
3
8
5
8
2
2
500 -
-------
FINAL REPORT
TANK CORROSION STUDY
NOVEMBER 1988
LOCATION OF PERFORATIONS ;
% Perf.
20.3 %
18.9 %
17.5 %
16.1 %
14.0 %
4.9 %
4.2 %
2.1 %
1.4 %
0.7 %
100.0 %
i
NOTErl) The location designated as 'Not Listed1 corresponds to
the tank with the external mechanical damage. It was damaged in
more than one location. : "
2) The 'Elsewhere on Bottom1 category represents a hole which was
not below either the fill or the gage holes.
Location
Single on Side
Multiple on Bottom
Single on End
Multiple on Sides
and/or Ends
Single Elsewhere
on Bottom
Single or Multiple
Below Gage Hole
Top
Along Groundwater
Line
Single or Multiple
Below Fill
Not Listed
Perf . Tanks
29
27
25
23
20
7
6
3
2
1
143
GROUNDWATER LEVEL
Tank always in
groundwater
Tank sometimes in
groundwater
Tank never in
groundwater
Groundwater level
unknown
All Tanks
55
29
379
37
% of % of
All Tanks Perf: Perf. Tanks
(n/500) Tanks (n/143)
11.0 % 19 13.3 %
5.8 % 17 11.9 '%
75.8 % 95 66.4 %
.7.4 % 12 8.4 %
Percentage of tanks that were always in groundwater and were
perforated is 34.5 % (19/55). '
Percentage of tanks that were ^sometimes in groundwater that were
perforated is 58.6 % (17/29).
Percentage of tanks that were never in groundwater that were
perforated is 25.1 % (95/379).
Percentage of tanks that groundwater level was unknown and were
perforated is 32.4 % (12/37).
Note: The 'Sometimes in Groundwater' category includes those
tanks which are subject to tidal or groundwater fluctuation and
the tank bottom conditions vary between wet and dry.
17
-------
FINAL .REPORT
TANK CORROSION STUDY
NOVEMBER 1988
OBSERVED SOIL CONDITIONS
The soil conditions listed below were the categories us~d bv the
inspectors to best describe the soil conditions at the tank sit"
This is not the backfill material, but the soil conditions thlt
would have been found prior to tank installation. The table on
page 2.0 compares this same soil list to the backfill condition
found at the perforated tank sites.
Number of tanks per soil condition
All Tanks
1)Clean sand/gravel,
clay, and loam 159
2)Clean sand/gravel 116
3)Clean sand/gravel,
and loam 107
4)Clean sand/gravel,
and clay 29
5)Clay and loam 23
6)Clean sand/gravel,
and bog 13
7 ) Bog and. loam 8
8)Clay and bog 7
9)Clean sand/gravel,
clay,and bog 6
10)Sand w/clay 5
11)Loam • 4
12)Bog and sand w/clay 3
13)Bog 2 ,
14)Clean sand/gravel,
bog, and loam 2
15)Clean sand/gravel,
clay, bog, and loam 1
16)Clay and sand w/clay 1
17)Clay, bog, and loam 1
18)Clay 1
• 9)Unknown 12
500"
% of
All Tanks
(ri/500)
31.8 %
23.2 %
21.4 %
5.8 %
4.6 %
2.6 %
1.6 %
1.4 %
1.2 %
1.0 %
0.8 %
0.6 %
0.4 %
0.4 %
0.2 %
0.2 %
0.2 %
0.2 %
2.4 %
Perf.
Tanks
44
20
33
6
11
6
3
1
4
3
1
2
2
1
1
1
0
3
T43
.% of
Perf. Tanks
(n/1431
30.8 %
14.0 %
23.1 %
4.2 %
7.7 %
4.2
2.1
0
7 %
2.8 %
2.1 %
0.7 %
1.4 %
1.4 %
0.7 %
0.7 %
0.7 %
0.7 %
0.0 %
2.1 %
18
-------
FINAL REPORT
TANK CORROSION STUDY
NOVEMBER 1988
Number of perforated tanks per original soil condition in type of
backfill and in level of groundwater
L
The purpose of this list is to compare the backfill conditions to
the original soil conditions and the groundwater conditions to
the original soil conditions. Regarding the backfill category,
'Clean' refers to clean sand, gravel, or stone brought in to
backfill the tank; 'Same as Orig' refers to the backfill being
the same as original soil; and 'Rubble' indicates the presence of
a foreign material. Regarding the groundwater category, 'Alw1
indicates the tank is always in groundwater; 'Some' refers to the
tank sometimes being in groundwater; 'Nvr' refers to the tank
never being in groundwater; and ' Unk' refers to an unknown
condition.
Note: The number of perforated tanks is only additive in each
category (ie: backfill or groundwater) and not across each row.
1)Clean sand/gravel,
clay, and loam
2)Clean sand/gravel,
and loam
3)Clean sand/gravel
4)Clay and loam •
5)Clean sand/gravel,
and clay
6)Clean sand/gravel,
and bog
7)Clean sand/gravel,
clay,and bog
8)Bog and loam
9)Sand w/clay
10)Bog and sand w/clay
11)Bog
12)Loam
13)Clay and bog
14)Clean sand/gravel,
bog, and loam
15)Clean sand/gravel,
clay, bog, and loam
16)Clay and sand w/clay
17)Clay, bog, and loam
18)Unkno\
-------
FINAL REPORT
TANK CORROSION STUDY
NOVEMBER 1988
BACKFILL CONDITIONS
Clean Backfill
Backfill with
rubble
Unknown
All Tanks
317
167
16
% of
All Tanks
(n/500)
63.4 %
33.4 %
3.2 %
% of
Perf. Perf. Tanks
Tanks (n/143)
80 55.9 %
60 42.0 %
3 2.1 %
Percentage of tanks in clean backfill that were perforated is
25.2 % (80/317) ,
Percentage of tanks in rubble backfill that were perforated is
35.9 % (60/167) .
Percentage of tanks in unknown backfill that were perforated is
18.8 % (3/16).
PERFORATIONS VERSUS LEAKAGE
Fuel Oil
-Total number fuel oil tanks (f2 - #6) with perforations was 40.
-Total number of fuel oil tanks with perforations that showed
evidence of leakage was 31.
-Therefore, 77.5 % of all perforated fuel oil tanks showed
evidence of leakage. This is 37.3 % of all tanks that showed
evidence of leakage.
Gasoline
-Total number of gasoline tanks with perforations was 73.
-Total number of gasoline tanks with perforations that showed
evidence of leakage was 37.
-Therefore, 50.7 % of all perforated gasoline tanks showed
evidence of leakage. This is 44.6 % of all tanks that showed
evidence of leakage.
SLUDGE VOLUME
The remaining average sludge volume in all tanks was 33.3
gallons.
The remaining average sludge volume in perforated tanks was 34.6
gallons.
20
-------
FINAL REPORT
TANK CORROSION STUDY
NOVEMBER 1988
AGE VERSUS VOLUME - Perforated tanks
Age in Years
70
57
55
50
48
44
43
40
37
33
30
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
8-
UNKNOWN
Number of tanks by Volume
1
1
1
1
1
1
3
1
1
1
1
1
1
1
1
13
1
1
4
2
1
1
1
2
4
2
2
3
1
2
5
1
3
1
' 1
1
1
1
3
3
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
500
1000
1500
315
185
4000
1000
2000
2000
3000
275
2000
3000
1000
4000
2000
550
550
1000
5000
275
1000
1000
550
4000
550
4000
1000
5000
550
3000
1000
2000
1000
4000
1000
2000
1000
275
2000
2
2
1
2
1
1
3
2
1
2
1
2
1
3
3
1
3
1
1
2
1
3
1
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
550
1000
3000
4000
2000
5000
4000
1000
3000
4000
2000
2000
1000
2000
1000
4000
3000
3000
2000
2000
3000
550
4000
1
1
2
1
4
1
2
2
2
2
7
1
X
X
X
X
X
X
X
X
X
X
X
X
1500
5000
3000
4000
4000
3000
3000
3000
2000
4000
1000
12000
Total
143
Average age of perforated tanks was 23.4 years.
21 years. '
Average volume of perforated tanks was 2206.5 gallons.
volume was 2000 gallons.
Median age was
Median
21
-------
FINAL REPORT
TANK CORROSION STUDY
NOVEMBER 1988
VOLUME VERSUS AGE
Tanks without perforations(greater than 4000 gallons)
Volume (gals)
5000
6000
7500
8000
10000
12000
15000
20000
25000
30000
50000
Number of Tanks by Age in Years
3X13
1 X 11
3 X 14
1 X 21
1 X 24
1 X 32
3 X 10
3 X 16
2 X Unk.
1 X 18
1 X Unk.
2 X 10
1 X 19
3 X Unk.
1
1
X 7
X 12
3 X 17
3 X 26
1 X 30
13 X Unk,
1 X 13
2 X 14
1
1
X 29
X Unk,
1 X 37
2 X 17
1 X Unk.
2 X 13
2 X 35
1 X 12
1 X 16
1 X 22
2 X 25
2 X Unk.
1 X 14
1 X 18
1 X 30
1 X 16
1 X 21
1 X 44
1 X 25
1 X 31
'3 X 40
2 X 22
Total
4 X 20
7
1
X 23
X 28
1 X 15
1 X 27
2 X 44
1 X 18
3 X 23
6X9
4 X 13 .
1 X 18
2 X 27
1 X 43
10 X 10
1 X 15
2 X 20
1 X 29
1 X 44
1 X 26
1 X 46
1 X Unk.
3 X 44
136
Average age of tanks (in the above category) was 20.7 years
(excluding 24 unknown).
Total volume of tanks was 1,464,500 gallons.
Average volume of tanks was 10,768 gallons.
22
-------
FINAL REPORT
TANK CORROSION STUDY
Tanks without perforations(less than and equal to 4000 gallons)
Volume fqals)
175
. 275
500
550
575
1000
1100
1500
2000
2500
3000
1 4 Unk.
1X2
2 X 16
1 X 25
1 X 10
1X3
1 X ,13
3 X 16
2 X 23
5 X 28
1 X 48
1 X Unk.
2 X 7
2 X 11 ..
1 X 16 ..
2 X 25
2 X 31
1 X 4,4
8 X Unk.
1 X 29;
1 X 17
2 X; 32
1 X- 10
1 X 15-
2 X 19
3 X 24 ,
5 X: 2,8 •:
1 X 35
I^r *7
*». /
1 X, 7
2 X^ 12
1 X 16
3 Xi 20
2 Xj 27
1 X 31
1 X 43
1X9
1 X 20
5 X Unk.
1 X 30
1X8
1 X 14
2 X 21
1 X 24
1 X 43
14 X Unk.
1X8
, 2 X 13
1 X 1-7
1 X 27
1 X 41
1 X 48
1 X 18
1 X 36
2 X 12
1 X 16
4 X 20
1 X 25
2 X 30
1 X 41
. 2 X 32
1 X 10
1 X 14
1 X 17
2 X 22
2 X 28
1 X 32
..IX 44 .-..'
1 X 10
1 X 23
6 X 10
2 X 15
1 X 22
1 X 27
1 X 47
1 X 10
1 X .1-4
2 X 20
1 X 30
1 X 43
1 X 60
'••.'' k
1 X 22
2 X 13
3 X . 1.7
1 X 23
2 X' 26
1 X 34
5 X Unk.
2 X Unk.
1 X 11
3 X 15
4 X 19
1 X 23
1 X .30 . .
1 X. 33
6 X Unk.
3500 1 x 23
23
-------
FINAL REPORT .
TANK CORROSION STUDY
NOVEMBER 1988
Volume (gals)
4000
Number of Tanks by Age in Years
2 X 10
1 X 14
3 X 17
2 X 21
3 X 24
1 X 31
1 X 12
2
1
2
15
18
6 X
3 X
X 22
X 27
X 44
2 X 13
2 X 16
X 20
X 23
X 28
1
2
1
5 X Unk,
Total
221
Average age of tanks in the above category) was "21.2 years
(excluding 47 unknown).
Total volume of tanks was 434,625 gallons.
Average volume of tanks was 1,967 gallons.
24
-------
FINAL REPORT
TANK CORROSION STUDY
NOVEMBER 1988
OBSERVATIONS
observati°ns ate 'based on ths inspection of 500
chnrr fr+l removed. In [several instances, the data have
changed from the last interim report due to the removal of 12
tanks from the database which ^ere placed in a seperate section -
we-re bett^r classified as "non-corrodible"
fib-glass-coated steel, or
Every piece of data was re-examined to ensure accuracy and
2o«?™^X,?Xty- ^ e*amPle^ the unknown age of a tank was
determined or an odd gallon amount was verified, etc. For cases
where the age was still unknown, the tanks were either buried for
aw 1g e and the land°wner had changed hands many times , - were
JSS1*0*!1 °£ Pr°Perties' °r it, was unknown to the current owner
that they had ever existed (ie: they were found when removing
other tanks ) . 3
1) More than one-quarter (28.6i%) of the tanks removed had
perforations in them. Although the overall percentage is lower
than previous interim reports, lit is reasonably consistent with
tnose findings. The percentage of perforated tanks from the
previous reports were: Interim 1 - 36 % Interim 2 - 11 s %
Interim 4-30.9%. jj.j *,
A calculation was performed in an attempt to determine the
prediction capability of a polynomial regression based on the
least squares method. This was^ based on the first 100 tanks. it
was determined that the prediction capability (ie: the ability to
predict the number of perforated tanks that would be found based
on the first 100 tanks removed); was not accurate and could not be
applied. A simple straight line percentage relationship based on
the number of failures in 500 tanks appeared as usable for
prediction as any other method.
2) Only 58 % of the perforated [tanks showed evidence of leakage.
This is 16.6 % of all tanks. Assuming the tanks that actually
leak would show positive on a tank test, almost twice as many
tanks have holes in them than can be detected by testing (ie:
16.6 % of all tanks showed evidence of leakage, while 28.6 % of
all tanks actually had perforations).
°"? ,143 Perforated tanks, 38 had tank tests associated with them
within the two years prior to removal, but it was not necessarily
the reason for removal. Of those tanks, 29 passed the test and 9
failed the test. Of the 29 that passed, 11 showed evidence of
leakage (37.9 %), while of the 9 that failed, 8 showed evidence
of leakage (88.9 %).
3) There is a strong relationship between wall thickness and
perforations. Page 12 details; the average wall thicknesses for
25
-------
FINAL REPORT NOVEMBER 1988
TANK CORROSION STUDY
perforated tanks, non-perforated tanks, and all tanks. The wall
thicknessesjmeasured were as close to original thickness as
possible ( ie : taken where the wall appeared in the best
condition).
The numbers on page 12 and the numbers on page 14 lead to the
formulation of a correlation between wall thickness and
perforation. The chart on page 14 that lists wall thickness by
volume for both perforated and non-perforated tanks best shows
the relationship. By comparing the two charts for the same
volumes, it can be seen that the perforated tanks have thinner
walls in most instances (refer to Figure 5, Average Plate
Thickness). These thin wall tanks also correspond to those of
small volume (ie: less than 5000 gallons). Therefore the now
obvious can be stated - The smaller tanks are more susceptible to
perforation because they are made of thinner material, while the
larger tanks are not as likely to fail because they are made with
thicker material.
Of all the perforated tanks, only 5 had both endplate and wall
thicknesses greater than 0.20 inches. One of these was only 8
years old but had failed from weld failure, which could happen to
a tank of any size or age. Of the remaining 4 tanks, -the average
age was 24 years, slightly above the average age of 23.4 years
for all of the other perforated tanks, ie: those with plate
thicknesses less than 0.20 inches. These all failed at their
thinnest wall.
4) The percentage of perforated tanks which held gasoline, #2
fuel oil, diesel fuel, and kerosene, respectively, was close to
the percentage of all tanks that were perforated (refer to Figure
1, Tank Contents). Of 233 gasoline tanks, 31.3 % were
perforated; of 128 #2 fuel oil tanks, 30.5 % were perforated; of
33 diesel tanks, 33.3 % were perforated; and of 11 kerosene
tanks, 27.3 % were perforated. This seems to indicate that no
specific product is responsible for causing a greater percentage
of perforations in tanks.
5) There is no clear correlation between tank age and
perforations(refer to Figure 2, Age of Tanks). The range of ages
for perforated tanks was between 8 and 70 years, with
perforations scattered throughout. The average tank age was 21.8
years (excluding 87 unknown) . The average perforated tank age
was 23.4 (excluding 18 unknown). The tanks with the greatest
population of perforated tanks were the 26 year old tanks, with
72.7 % of the tanks having perforations (this excludes age
categories with only a few tanks in the category).
6) Tanks 5000 gallons and smaller accounted for 99.3 % of all
perforated tanks. Tanks 4000 gallons and smaller accounted for
95.8 % of all perforated tanks. Tank volumes ranged from 175
gallons to 50,000 gallons. Only one tank larger than 5000
gallons had perforations. The average tank volume for all tanks
was 4433.3 gallons and the average perforated tank volume was
26
-------
FINAL REPORT
TANK CORROSION STUDY
NOVEMBER 1988
less?. °° gallons and less' 63.6 % were 2000 gallons and
evidence of leakage (refer to Figure 4, All Fuel Si! Tanks) d
Jl?-r£orateA 9asolins tanks, 50.7 % showed evidence of
o^1S t."as 44'1 * of a11 Perforated tanks thlt showed
of leakage (refer to Figure 3, All Gasoline Tanks)
and
very small tanks have the thinnest walls. category oince the
?nnXn41 f^1 Oil tanks ^#2' *4' and #6 oils), 57 were greater than
4000 gallons in volume. Only 3 of those 57 t 5 3 %f h a H
perforations (All were 5000 Igallons ) . of the 84 tanks' 4000
gallons and less, 37 (44.0 %) were perforated. T?tal
of perforation for all 141 tanks was 40, or 28.4 %
*
9reater tha" 4000 gallons in
27
-------
FINAL REPORT NOVEMBER
TANK CORROSION STUDY
gallons and less, 72 (40.0 %) were perforated. Total occurrence
of perforation was 31.3 %.
14) The greatest number of perforations, 75.5 %, was caused by
external corrosion (73.4 % point corrosion and 2.1 % general
corrosion). Only 6.3 % of the tank perforations were caused by
internal corrosion (4.9 % point corrosion and 1.4 % general
corrosion). A combination of internal and external corrosion
caused 14.7 % of the perforations, meaning that in most of these
cases there were too many holes present to determine which caused
perforations first, or that varying stages of both types pf
corrosion did no allow for an accurate determination.
The conclusion that significant internal corrosion does not occur
should not be drawn from these numbers. A more appropriate
conclusion is that perforations occur, much more frequently from
the outside than from the inside.
15) Internal and external corrosion were observed in tanks. Both
general corrosion and point corrosion were classified as either
nominal ,( <25 % corrosion), mild ( >25 % & <50 %), moderate
( >50 % & <75 %), ~or severe (>75 %) in each category. Since it
is already known that the external corrosion "did the most damage,
internal corrosion will be examined below.
There were 17 tanks with severe general internal corrosion and 48
tanks with moderate general internal corrosion. Of these, the
number of tanks with perforations in each category was 13
(76.5 %) and 19 (39.6 %) respectively. There were 10 tanks with
severe point internal corrosion and 9 tanks with moderate point
internal corrosion. Of these, the number of tanks with
perforations in each category was 7"(70.0 %) and 9 (100. %),
respectively. Of the severe general internal corrosion category,
1 failed from general corrosion, 1 failed from point corrosion, 7
failed from the combination of internal and external, and 4
failed from other causes. Of the moderate^enera^ internal
corrosion category, 6 failed from the combination of internal and
external corrosion, and 13 failed from other causes. Of the
severe_£oi1nt internal corrosion category, all 7 failed from the
combination of internal and external corrosion. Of the moderate
pjoijit internal corrosion category, 1 failed from internal point
corrosion, 3 failed from the combination of internal and external
corrosion, artd 5 failed from other causes. If the failure
mechanisms that are not internal or external corrosion related
are discounted the numbers of tanks with perforations for this
observation is reduced by 25 tanks. These adjusted numbers of
perforated tanks for the categories severe general, moderate
general, severe point, and moderate point (all internal) become:
9 (52.9 %), 6 (12.5 %), 7 (70.0 %), and 4 (44.4 %).
16) Other than the thinnest tank wall, there is no good indicator
to predict where perforations will occur. The three locations
of perforations occurring most frequently were: 1) Side (20.3 %);
2) Multiple on Bottom (18.9 %); and 3) End (17.5 %).
28
-------
FINAL REPORT
TANK CORROSION STUDY
NOVEMBER 1988
Approximately 6.3 % of the perforations occurred either below the
fill (1.4 %) or below the gage ihole (4.9 %).
17) The greatest percentage of perforated tanks were found in
changing groundwater conditions where the tank was sometimes in
groundwater and sometimes not. This was 58.6 % of the tanks in
that category, suggesting that changing water conditions may
accelerate the corrosion of tanks. The percentage of tanks
always in groundwater and never in groundwater were much lower,
with 34.5 % and 25.1 % of the., tanks perforated, respectively.
The percentage of perforated tanks in unknown groundwater level
was 32.4%.
18) Backfill conditions appear £0 have an effect on the frequency
of perforation. It appears that tanks with non-uniform backfill
perforate more frequently than those in a clean backfill. The
percentage of perforated tanks [with rubble in the backfill .was
35.9 % while the percentage of perforated tanks in clean backfill
was 25.2 %. Rubble is taken toimean any miscellaneous material
not related to native soil or clean backfill. Examples of rubble
found include: concrete, asphal;t, rock, wood, paper, scrap metal,
brick, and shells.
19) The 143 perforated tanks we|re located in a variety of soil
conditions, most of which werej composed of well drained sand or
sandy loam. They were found in 12 of the 18 series of soils
which make up Suffolk County. Approximately one dozen tanks were
located in a mixture or on the; borderline of two soil series.
The twelve soil series were composed of 6 of the 10 major soil
associations found in Suffolk. ; See the soils section on page 5
for a discussion of soil classification. The chart below details
the number of tanks and soil series where-they were, located
according to the soil maps iof the USDA Soil Conservation
Service.1
No. of Tanks Corrosivity Soil Series
1)
2)
3)
4)
5)
6)
1
2
2
2
2
3
Moderate to
High,
Low
Low
Low
Low to
High!
I
High,
Cut & Fill Land
Tidal Marsh
Carver & Plymouth Sand
Carver & Plymouth Sand
Plymouth Loamy Sand
Montauk Soil
Riverhead & Haven Soil
Tidal Marsh
Made Land
Tidal Marsh
29
-------
FINAL REPORT
TANK CORROSION STUDY
NOVEMBER 1988
No. of Tanks Corrosivity Soil Series
7)
8)
9)
10)
11)
12)
13)
3
11
13
15
19
31
39
143
Low to
High
Low
Low
Low
Low
Varies
Low
Plymouth Loamy Sand
Muck
Riverhead Sandy Loam
Haven Loam
Plymouth Loamy Sand
Cut & Fill Land
Urban
Riverhead & Haven Soil
30
-------
FINAL REPORT NOVEMBER 1988
TANK CORROSION STUDY
FUEL OIL TANKS
General Statistics
Of the 500 tanks removed in this study, 141 were fuel oil tanks
of all kinds (ie: #2, #4, and t6). Of these 141 tanks, 125
qualified as exempt tanks under the federal definition, that is,
they were strictly for on-premises consumption. . Therefore, 25%
of the tanks removed in this study were exempt fuel oil tanks.
Of the 125 exempt tanks, 40 had perforations. This is 32% of the
exempt tanks. It is interesting to note that none of the
non-exempt fuel oil tanks had jperforations.
The following data cover the 40 tanks that had perforations.
Volume Distribution
Tank Size # Qf Tks. with Perfs.
275 : 4
550 3
1000 7
1500 2
2000 16
3000 3
4000 2
5000 _3
40
Of the 40 perforated tanks, 39 contained #2 fuel oil. Only 1
contained #4 fuel oil. That tank was 5000 gallons.
Age Distribution
Tank Age # of Tks. with Perfs.
10 1
16 1
17 2
18 2
20 3
23 1
25 1
26 12
27 4
30 3
37 1
43 2
44 1
55 i
Unk. _5
40
31
-------
FINAL REPORT NOVEMBER 1988
TANK CORROSION STUDY
Analysis
Twenty-nine of the perforated fuel oil tanks failed from exterior
corrosion (72,5%), two suffered weld failure (5%), six failed
from the combination of internal and external corrosion (15%),
and three succumbed to internal corrosion (7.5%). Only 9 of the
tanks (22.5%) were in contact with the groundwater some or all of
the time. Regarding leakage, 31 of the 40 tanks showed some
evidence of leakage (77.5%). Many of the tanks (27/40 - 67.5%)
had multiple holes. The average number of holes per perforated
tank was 7. Of the.27 tanks with multiple holes the average
number of holes was 9. The average age of these perforated tanks
was 26.7 years old, and the average volume was 191B.8 gallons.
All except 5 tanks had moderate or severe general external
corrosion and all but 8 had moderate or severe point external
corrosion. In contrast, only 6 tanks had moderate or severe
general internal corrosion and only 2 had moderate or severe
point internal corrosion.
32
-------
FINAL REPORT
TANK' CORROSION STUDY
NOVEMBER 1988
CONCLUSIONS
1) Size is more important than!age in predicting tank failure.
The age of perforated tanks studied spread widely between 8 and
70 years, but the volume of perforated tanks was almost always
less than 5000 gallons.
2) In general, small tanks arejmuch more likely to perforate than
large tanks due to the thinner5tank walls found in smaller tanks.
A major dividing line seems to be at the 5000 gallon level
because most tanks of that siize and above are constructed•of
1/4" or thicker steel plate,; while most tanks less than that
volume are made from lower gauge steel. With steel of 1/4" or
greater, there is a reduction in the number of, perforations to
nearly zero, in the surveyed tanks. One should not go so far,
however, as to use the statistics in.this report to claim that
there is no need to address, tanks greater than 5000 gallons,
since a very large number of the higher volume tanks had already
been removed by the beginning of the study. These most certainly
included a significant number with leaks that would have turned
up in the survey had they still been in place.
3) Compared to external corrosion, internal corrosion is
insignificant. However, once external corrosion is eliminated,
internal c.orrosion becomes a Ivery important consideration and
should be controlled.
4) Fuel oil tanks are just as I susceptible to perforation as
gasoline tanks of the same size. If the two groups are compared
as a whole, fuel oil tanks iare even more susceptible than
gasoline tanks since they are generally of much smaller size.
The study produced no evidence that the contents of tanks ( ie:
gasoline or fuel oil) significantly affected the rate of
perforation of the tanks.
5) Existing tanks are in worse shape than is demonstrated by
testing. Testing, even if .totally successful and accurate only
can locate tanks that are actively leaking product. The study
proved that tanks can rust through completely long before they
begin to leak product. In fact, the number of tanks found to
have holes vas nearly twice the number that showed evidence of
having leaked.
6) Tanks do not always leak immediately upon perforation. As
stated in the preceding paragraph, only a little over half the
tanks with holes actually showed evidence of leakage. It was
frequently observed that the corrosion products were still
tightly adhered to tanks at the points of corrosion and had to be
forcibly knocked off before |the holes were revealed. At sights
such as these, product had not yet succeeded in seeping through
the plug of corrosion products. There was no way of determining
through this study at what point a corrosion hole would finally
turn into a leak.
33,
-------
FINAL REPORT NOVEMBER 1988
TANK CORROSION STUDY
7) Tank testing can not be relied upon to locate tanks with
holes. Even if functioning accurately, tank testing can only
locate tanks that are leaking and there are many tanks with holes
that are not yet leaking.
8) The study, though not conclusive, throws some doubt on the
reliability of tank testing. There were 29 tanks found with
perforations that had passed a tightness test within the two
years prior to removal. Of these, 11 actually had leaked into
the soil and therefore should not have passed. Clear conclusions
cannot be drawn from this however because of unknown factors.
The tanks could have started leaking after testing was completed
or seepage could have been too slow to be detected by testing.
The database regarding this particular subject is too small to
provide reliable information, however as an indication, the
numbers are perhaps sufficient to suggest the need for further
investigation.
9) On tanks constructed of plates of more than one thickness,
perforations can usually be expected .to occur first in the
thinnest plates regardless of where they are located on the
tanks.
10) Non-uniform backfill increases somewhat the likelihood of
tank perforations, but the rate is not dramatically different
than that for uniform backfill.
11) The findings of this study are conservative and should be
applicable elsewhere. Because Suffolk County soils fall
generally in a low corrosivity classification and because many of
the worst tanks had already been removed before the study began,
it can be reasonably assumed that the occurrence of tank
corrosion and perforation at most other locations in this country
can be expected to be at least as bad as that indicated by the
statistics in this report.
In addition, only perforations that were large enough to be
easily observed visually were recorded. There were undoubtedly
other, smaller perforations that went undetected that could only
have been found by careful air testing and soaping of the tanks .
This would have been a tedious task that was beyond the scope of
the study. Therfore the actual number of perforated tanks was
certainly larger than the number observed.
34
-------
Figure 1 - Tank Contents
See chart, page 11
I
6
HJ
TANK; CONTENTS
ALL TANK VS. PERFORATED TANKS
33
11
X'\
FINAL REPORT
TANK CORROSION STUDY
36
EZ1 ALL TANKS
A - GASOLINE
B - ALL FUEL OIL
C - ALL WASTE OIL
D - DIESEL FUEL
E - SOLVENTS
F - OTHER MATERIAL
Figure 2 - Age of Tanks
See chart, page- 15
P
b
£
ID
I
180
170-
160-
150
140 -
130-
120 -
110-
100-
90-
80-
70-
60 -
50 -
40 -
30 -
20 -
10 -
0
18
PERFORATED TANKS
AGE OF TANKS
ALL VS. PERFORATED
134-
/.
60
48
24-
UMK
4- 3
11-20
21-30
31-40
41-50
51-70
ALL TANKJ
AGE
35
PERFORATED TANKS
-------
Figure 3 - All Gasoline
See chart, page 20
FINAL REPORT
TANK CORROSION STUDY
ALL GASOLENE TANKS
NON.PERF. VS. PERF. AND LEAKAGE
PERF.-MO LEAKAGE (15.5%)
NON-PERFORATED (68.7%) \
PERF.-LEAKAGE (15.3%)
Figure 4 - All Fuel Oil
See chart, page 20
ALL FUEL OIL TANKS
NON.PERF. VS. PERF. AND LEAKAGE
PERF.-NO LEAKAGE (6.4%)
PERF.-LEAKAGE (22.0%)
MOH-PERFORATEO (71,6%)
36
-------
Figure 5 - Average Plate
Thickness
See chart,
page 14
AVERAGE PLATE THICKNESS
PERF.TKS. VS. HON-PERF. TKS.
FINAL REPORT ,
TANK CORROSION STUDY
0.26-
0.24-
0.22 -
0.2 -
04 O
.18
0.16 -
0.1 4--
0.12 -
:'
0.1 -
0.08 -
0.06 -
0.0 4--
0.02 -
n -
'
/
s
s
/
j
/
x
X
S
V
X
V
X
>
x
"t
*
v
'*"
•£
'f
£
/
/
/
/
^
/
^
••
f
/
'
\
\
\
\
\
X
S
x
X
^
Y
X
T"
/
£
/
^
/
/
/
V
y
/
/
'
/
/_
i
— »
x
\
\
\
X
\
X
\
X
X
x
\
V
\
>
\
x
\
X
'V
x
X
x
X
X
X
x
x
X
'It
X
A
^
<*
5?
'A
y.
"v.
X
¥
X
X
X
X
'f'
•J
V
$
'f
?
'
S
/
/
/
/
/
f
/
f
/
"^ X
s^
v/
vx
vx
s
\ f
/
/
/
'/
\£
\ /
x
v X
XX
X
f
—^
\
X
\
\.
\
^
\
\
\
\
s
\
\
\
••^
\
N
v
^
X
x
X
X
x
X
X
X
x
X
1
ff
"k
s
x
^
:!
' x!
x
< '•<
< y
v
<[ x
:<'
t. X
X
X
X
'x
< >
< X
*• (*
v1 '••
•4
< ;•
A
/
/
>
''
/
/
/
/
/
/
/
X
X
x /
\
\
\
\
\
\J^
\
v
^,
\
y %
x ":•*
\^
X
\>
\ ')
\ >
\
\ >
S
\
\
\
\ )
\ :
X '
\ *
\
\
•-.
$
V
s X
X
ex
v
c| V
' X!
X
X
X'
•X
< >
X
'1 X
^ '*'
y
^ i«
v
'
' X
f'v-
' S
f \
,y
'• \
' \
( \
t1 \
> s
•
x
^
\
\
\
\
\
\
s
N
\
. '•
' \
V
t>TI
H
•• X
' x
INTERNAL (6.3X)
37
-------
FINAL REPORT NOVEMBER 1988
TANK CORROSION STUDY
APPENDIX A
Exerpt Portions of Article XII
of the
Suffolk County Sanitary Cocle
38
-------
Section 1210. Underground Storage Facilities
a. New Storage Facilities
1. All new storage facilities used or to be used for the
underground storage of toxic or hazardous materials shall be
designed and constructed in a manner which will, in the
opinion of the commissioner, provide the maximum reasonable
protection available against leakage or spillage from the
facility due to corrosion, breakage, structural failure, or
other means. Double-walled or equivalent facilities are
required for all toxic or hazardous materials except those
with a specific gravity of less than one and which are only
slightly soluble in water such as oils and gasoline. For
these floatable materials, acceptable designs for tank con-
struction include cathodically protected steel; glass fibre
reinforced plastic; steel clad with glass fiber reinforced
plastic; double-walled steel or plastic; or other equivalent
design approved by the commissioner.
2. Approval of design by the commissioner is required
before installation, and the determination of equivalency or
adequacy lies with the commissioner.
3. Design, construction, fabrication, and installation
of new underground storage facilities shall be in accordance
with regulations and standards as they may be adopted by the
commissioner under this article from time to time.
4. A new storage facility for all facilities not
previously covered by this section is one for which
construction actually begins on or after November 1, 1982?
subject however to the exemptions contained in Section
1208(a). !
5. It shall be unlawful for any person to sell for use in
Suffolk County, install, use, put into service or maintain
the existence of any new underground storage facility or part
thereof after November 1, 1982, if said new storage facility
or part thereof fails to conform to all of the provisions of
subsections (1),- (2), and (3) above, and all regulations and
standards promulgated thereunder; subject however to the
exemptions contained in Section 1208(a).
b. Existing Storage Facilities
1. An existing underground storage facility is one for
which construction 'actually begins prior to November 1,
1982.
-------
2. It shall be unlawful for any person to substantially
modify or cause the substantial modification of any existing
underground storage facility or part thereof without
complying with the provisions of subdivision (a) above and
all regulations and standards promulgated thereunder.
3. It shall be unlawful to use, or maintain the existence
of any existing underground storage facility beyond
January 1, 1990, which is intended for use with toxic or
hazardous materials with a specific gravity of less than one
and which are only slightly soluble in water such as oils and
gasoline, without modifying said storage facility so as to
comply with all of the provisions of subdivision (a) above
and all regulations and standards promulgated thereunder.
4. It shall be unlawful to use or maintain thS existence
of any existing underground storage facility beyond
January 1, 1987, which is intended for use with any toxic or
hazardous materials other than those with a specific gravity
of less than one and which are only slightly soluble in water
such as oils and gasoline, without modifying said storage
facility so as to comply with all of the provisions of
subdivision (a) above and all regulations and standards
promulgated thereunder.
c. Abandonment
1. It shall be unlawful for any person to use or maintain
the existence of an abandoned underground storage facility or
part thereof.
2. It shall be unlawful for anyone to sell or transfer to
another an improperly abandoned underground storage facility
or land containing an improperly abandoned underground
storage facility if there exists any reasonable evidence of
the existence of such a facility, unless the purchasing party
has been made fully aware of the presence of such facility or
evidence.
3. It shall be unlawful for any person t9 repair, alter or
prepare for use any abandoned storage facility without first
obtaining a permit to construct from the commissioner.
4. It shall be unlawful for the owner or other person in
possession or control of any real property, building or place
or vehicle to fail to immediately empty of all toxic or
hazardous materials and to completely fill with sand or
concrete or permanently remove an abandoned storage facility
or part thereof within ninety (90) days of the discovery
thereof on or in said real property, building or place
pursuant to the provisions of subdivision (h) below unless
approval is granted by the commissioner to do otherwise.
5. For the purposes of this section, an abandoned storage
facility or part thereof means one which has remained out of
service for two (2) years 'or more, or which has been declared
by the owner to be abandoned.
6. For the purposes of this section, out of service
means substantially empty, meaning five (5%) percent or- less
-------
filled; or not in use, meaning [no regular filling or drawing;
or not being maintained, meaning lacking adherence to the
requirements of this article; or uncontrolled, meaning not
attended or secured; or any combination thereof.
7. For the purposes of this section, discovery means
either actual discovery or knowledge of the existence of the
abandoned storage facility or part thereof or possession of
sufficient knowledge of the facts and circumstances involved
• so that the existence of the abandoned storage facility or
part thereof should have been discovered or known of.
d. Testing and Inspection
1. All existing underground [storage facilities or parts
thereof which do not meet the construction standards in
subdivision (a) above, must be tested and inspected in
accordance with the schedule set forth below. It shall be
unlawful for any existing underground storage facility' owner,
operator or lessee to fail to test his tanks and file an
acceptable certificate of test .completion with the
commissioner in accordance with the following schedule:,
TESTING SCHEDULE FOR EXISTING UNDERGROUND TANKS
AGE OF SYSTEH BY 1980 , f
(in years) ( .
1-4 5-9 10 - 14 15 - 19 20 or more
"1980
1981 ' X
1982 X
1983 X X
1984 X
1985 X X X
1986 X
ALL TANKS COVERED BY SECTION 1208(b) BY VIRTUE
OF THE 1986 AMENDMENT SHALL BE INITIALLY TESTED IN 1986
IF THE TANK IS TEN (10) YEARS OR OLDER, AND/OR ALL TANKS
SHALL BE TESTED ON THE^R TENTH ANNIVERSARY AND
EVERY FIVE (5) YEARS THEREAFTER UNTIL PERMANENTLY CLOSED.
FULL COMPLIANCE FOR ALL FACILITIES EXCEPT THOSE
DESCRIBED IN 1210(b)(3) .
1987 X X
1988 X
1989
1990 FULL COMPLIANCE FOR ALL FACILITIES ,
-------
2. If for any reason testing satisfactory to the
commissioner cannot be performed, the tank must be removed
from service or brought up to the standards of subsection (a)
by the first scheduled test date.
3. The Final Test of the National Fire Protection
Association (NFPA), Recommended Practice No. 329 or other
test of equivalent or superior accuracy as approved by the
commissioner must be used to comply with the testing and
inspection requirement of Section 1210(d)(l).
4. Any test and inspection as required by this subdivision
shall be performed by a person whose qualifications are
acceptable to the commissioner, pursuant to Department
standards, for performing such tests. Certificates of test
completion containing the results of such tests as performed
shall be prepared by the tester and shall be filed with the
commissioner within thirty (30) days after completion of the
testing of the storage facility. No certificate of test
completion shall be acceptable to the commissioner to
indicate satisfactory compliance with the testing
requirements of this subdivision if the qualifications of the
tester have not been accepted by the commissioner prior to
the test. No certificate of test completion shall be
acceptable to the commissioner, pursuant to Department
standards, if the test and inspection were not performed in
accordance with subsection (3) of this subdivision and in
accordance with any regulations and standards which may be
promulgated pursuant thereto. •
5. The Certificate of Test Completion shall be filed on a
form provided by the commissioner and a copy of such form,
completed, shall be kept by the storage facility owner,
operator or lessee and by the tester for a period of not less
than five (5) years from the date of its issuance. It shall
be unlawful for the storage facility owner, operator or
lessee and for the tester thereof to fail to keep a copy of '
the Certificate of Test Completion for the required five (5)
year period.
6. Certificates of Test Completion shall contain a legally
authorized form notice to the effect that false statements
made knowingly therein are punishable pursuant to Section
210.45 of the Penal Law.
7. A Certificate of Test Completion not properly completed
and/or not subscribed by the tester shall not be acceptable
to the commissioner.
e. General Provisions and Requirements
1. When an underground storage facility or part thereof is
found to be leaking, the portion containing the leak must be
immediately emptied of all contents therein and removed frcm
service. It shall be unlawful to cause or permit a leaking
-------
underground storage facility or part thereof to remain in
service or to continue to retain its toxic or hazardous
contents after the owner, operator or lessee of. said storage
facility or part thereof knows or should have known of the
existence of the leak therein.
2. It shall bfe unlawful for any person to repair or to
permit the repair, in place, of any underground storage
facility or part thereof which has leaked or has otherwise
failed, for the purpose of reusing said storage facility,
* '
i. such repair will result in the storage facility or
part thereof complying, with the requirements of
subdivision (a) above and all regulations and standards
promulgated thereunder; and unless
ii. such repair occurs pursuant to plans therefor
previously submitted to and approved by the commissioner.
3. It shall be unlawful for any person to replace or cause
the replacement of any underground storage facility or part
thereof for any reason if the replacement facility does not
meet the requirements of subdivision (a) above and all
regulations and standards promulgated thereunder.
4. It shall be unlawful for any person to use, maintain,
or put into service any underground storage facility or part
thereof without first complying with the testing and
inspection requirements of Subdivision (d) above and
regulations and standards promulgated thereunder.
-------
FINAL REPORT
TANK CORROSION STUDY NOVEMBER 1988
APPENDIX B
Suffolk County Tank Removal Standard
39
-------
-------
SUFFOLK COUNTY DEPARTMENT OF HEALTH SERVICES
BUREAU OF HAZARDOUS MATERIALS
STANDARDS FOR THE ADMINISTRATION
OF ARTICLE 12
OF THE SUFFOLK COUNTY SANITARY CODE
'REMOVAL OF UNDERGROUND STORAGE FACILITIES FROM SERVICE"
-------
1.0 General
Revised 1/26/88
,,«HO , ' u 1210Jl provides for two method3 °f decommissioning
underground tanks: either complete removal, or abandonment in place if local
town ordnances permit. Complete removal is preferred since a tank left buried
creates a major obstacle for future construction.
abandonmefli of an underground hazardous/loxic
rT7 eflort must be made to determine
r t, f ^l** 1S removed* *"» «« usually be accomplished by
of the tank and the bottom of the excavation. For tanks abandoned in
l°rm8 we Ismust be tartalted to the satisfaction of the Department or a
test must ave been performed
t« ,h I'3 ^P6/,?01^31100 of faciiity removal or abandonment must be provided
to the Bureau of Hazardous Materials. Notification must include but is not UmTted
to a minimum of forty eight hours verbal notification prior ic ^ ^scheduled
decommissioning to schedule the required inspection.
C°mply With the fortv eight nour notice requirement will
he °Wnef °f ^ Other person to Possession or^ontrol
to ln H J' WM rem°Ved or abandoned without notice, to a fine of
rpmlli ? ? °rder t0 reslore ^ l**<* to its' condition at the time of tank
removal or abandonment so that an appropriate inspection may be undertaken to
determine whether pollution was created by the storage facility. UnaeUaken to
2.Q Procedures for r*«*
nn.. J'1 Pr*>r l° remova1' t"16 tan* «""' be pumped as empty as possible usina
portable pump to scavenge the bottom or all liquid that can be removed
2'
elcavation «»« remain open and the bottom left
*****
and
or an
-------
2.3 After eicavation of the tank is completed, holes, at least 48" in diameter
must be cut in each end of the tank to allow quick venting and to facilitate sludge
removal and internal inspection.
2.3.1 Care must be taken in cutting holes to avoid sparks that could
ignite flammable fumes. Only non-sparking pneumatic tools may be used.
2.3.2 All sludge must be removed from the tank by shoveling and
Crushing and be collected in properly labeled drums for removal as a toiic
waste. The tank interior shall have all major scale knocked loose and be in a
"brush clean" condition before removal from the site. Proper protective
breathing appartatus and protective clothing should be worn since the
sludge is likely to contain lead and other toxic materials.
2.3.3 The sludge may not be removed from the site until a Health
Department inspector has noted the volume of sludge and inspected the
tank.
2.4 The tank can not be removed from the site until it has been inspected by
a Health Department inspector or otherwise released by the Health Department.
~'",
2.5 All piping such as fill and vapor recovery lines must be removed to at
least 12" below grade.
3.0 Procedure for Taut Abandonment
3.1 Prior to abandonment., the tank must be pumped as empty as possible
using a portable pump to scavenge the bottom of all liquids that can be removed.
In addition, the tank must be cleaned and freed of all residual toiic/hazardous
materials.
3.1.1 All waste must be removed from the tank and be collected in
properly labeled drums for removal as a toxic waste. Proper protective
breathing appartatus and protective clothing should be worn since the
sludge is likely to contain lead and other toxic materials.
3.1.2 The sludge and waste may not be removed from the site until a
Health Department inspector has noted the volume of sludge and inspected
the tank.
3.2 After the tank has been inspected and determined to be clean by a
Health Department inspector, it must be completely filled with a clean inert
material such as sand or concrete.
-------
pipiD8 SUCh as fm and vapor re(»very lines must be removed to at
™«,K3/* In ^f abaence of a satsisfactory systems tightness test within ihe last 6
months, groundwater monitoring wells must be installed.
i • • * ' '
n7n« Th'1 7f,U!,mUSt be 4" diameter Schedule 40 PVC with a slot size of
.020 . The slotted portion of the wells must extend 5 ft. above and below
the groundwater elevation. Each well must be brought to grade and ail
covers at grade must be liquid tight and labeled monitoring well A
n™1?^ °f lw,°,wells • one "Pstrdam and one downstream of groundwater
flow at the tank location, must be installed.
•4-2vGr°ucndwaier monitoring wells must be sampled and analyzed as
* Environiaental amservation
-------
FINAL REPORT NOVEMBER 1988
TANK CORROSION STUDY
APPENDIX C
Tank Corrosion Study Inspection Sheet
40
-------
SUFFOLK COUNTY DEPARTMENT OF HEALTH SERVICES
FIELD INSPECTION FORM
FIRST PRATT
Inspection Date
Time Start _..jn
Time End
Time Spent
. Hin.
Facility 10;
Tanlc No.
Prop.Tax No.
Facility Name;
Facility Address;
Type off Facility; Gas Sta. Car- dealer__ Commercial^ Residential.
Industrial other__t describe!.
Type of Tank; Plain steel__ As|phalt or coal tar coated sta<*l—
STIP3_ Other induced current cat.prot.steel Imposed current cat.
>Prot.steel_ Bufhide_ Other fiberglas coated stael__ Overcoming
£iberglasj_ Xerxes iiberglas__ Other fiberglas_ Other material___
(describe) • _J
Contents (Observed) : Gasoline^ Fuel oil J2_ Fuel oil 84__ Fuel
-oil!5_ Fuel oil IG_ Keros0ne_ Diesel_ Gasahol_ (% mixture)
Jet fuel Av.gas Solvents^' (Describe) .——_ .
Waste oilj_ Other oil (Describe) ———.
Other material (Describe) .
Dimensions; D'iameter CO * Diameter (y)
Width ' H«ioht Volume
dimensions)
Date Installed; . Tresent age
gal.(Calculated from
End plate thickness
Wall thickness
* (mic,)
(Ultrasonic)
(Ultrasonic)
-------
- 2 -
Holes.: *es__ No_ Total No. observed _ Leak confirmed before
removal. - Dia. largest - _» Dia. smallest. _ - Average dia. _ ;• CES
Hole locations Bottom below fill_ Dottom bclow gago hole__
Elsewhere along bottom Multiple along bottom __ _
Top around fittings _ Eleshere on top _ , Along groundwater line__
On side — On end__ Multiple on sides and/or ends _
Cause of holes: Point corrosion internal__ Point corrosion external^ '
General. Corrosion internal^ General corrosion external^ Combination
internal and external corrosion - Weld failure^ Mechanical
interaal__ Mechanical damage external
Sludge: Volume _ gal. wt. _ . _ I
Exterior coating: Yes_ No__ Completely intact__ Minor flaws__
Many failed areas_ Completely failed^ _ . Korrellnlnc
coating; Yes_ NO_ Fiberglas ' lining^ Other__
__
Completely intact_ Minor flaws^ Many failed areas_. Completely
failed^ _ jSRemaininc
Natural Soil Conditions: Clean sand or gravel__ Clay__ Bog_ Loam_
Sand w/some clay _ Otlier _ (describe) _
Carr't tell _ __
Backfill Conditions: Clean sand: or gravel_ Same as natural soil_
Concrete, asphalt, stones or rubble against tank _ Other _
C dss cribe ) ____m_^ _
Groundwater level: Aiways in"cont.ct-wt«Tt«k_ Someti.es in contact
with tank _ Never in contact with tank
Groundwater Quality:. Salinity ,if near sTore) _ pH__ Floatina
product in observation wells__ m excavation dissolved product _
Closest estimate of'total -proHn-t ^.V^d: Ogal__ 10 Cal._ S0 Gal.___
10.0.- Gal,_ 500 Cal__ 1,OOC»G,1._ 2,000 Gal. _ 5/000 Cal._ 10, 000 Gal
-------
• 50,000 Gal. 100,000 Gal./or more . No idea___.
Approximate Burial "Depth Below Grade
Tank D£31-ec£ion: total length:—
Cminua ) to tank top : .
A. tank diameter:
total length:
(minus) to tank top:.
B. tank diameter:
Diameter.X.,:.
(minus 1 Diameter n_._;'
Deflection :'
BURIED
REMOVED
teseription of Tnnk FyTTP.TOn Corrosion;
Point Corrosion: nomlnal^jidld^^noderute r.cvei«
General Corrosion: nominal mild moderate jsevere
Description of Tank INTERIOR Corrosion;
Point Corrosion; nominal mild moderate severe
General Corrosion: nominal mild iroderate severe1
-------
FINAL REPORT
TANK CORROSION STUDY
NOVEMBER 1988
APPENDIX D
Non-corrodible Tanks
41
-------
FINAL REPORT
TANK CORROSION STUDY
NOVEMBER -198-8
During the course of the study!, twelve non-corrodible tanks were
removed from service for a variety of reasons. Though these
were not included in the statistics, an investigation similar to
that performed on each of the 500 unprotected steel tanks was
performed on each of these. A synopsis of that investigation
follows.
Of the 12 non-corrodible tanks removed, two were fiberglass, two
were fiberglass coated steeL, and the remaining eight were some
type of cathodically protected tanks.. None of these had
perforations.
Fiberglass
The two tanks in this categqry were both single-walled tanks.
One was 8 years old and the other was 10 years old. For
discussion here, the 8 year iold will be tank A and the 10 year
old will be tank B. Tank A contained gasoline and tank B
contained caustic soda. Tanjk A was 6000 gallons and B was 4000
gallons.
Although no holes were immediately evident on either tank, tank A
showed evidence of leakage and! had product oozing from its ribs.
This could have been due to leaky fittings or gasket failure
causing product to leak ar.ound the outside of the tank.
However, the tank had failed three tank tightness tests prior to
removal. This tank was sometimes in groundwater and had rubble
in the hole. Tank B had several areas where crystallization had
begun on the outside of the tank. The areas were soft when
pressure was applied, indicating that the fiberglass was
beginning to breakdown under the effects of the caustic. This
tank was never in groundwater and was in a natural (sand/clay)
backfill. The interior resin ;liner was intact and appeared to
have held up well, but the exterior of the tank was badly
deteriorated, apparently from caustic that had leaked down around
the outside surface of the tanjc.
Fiberglass Coated Steel
Two tanks of this type were removed. Tank A was 16 years old and
tank B was 12 years old. Tank A was 1000 gallons, tank B was
2000 gallons. Both were in good shape with 100 % of the exterior
coating (fiberglass) intactJ They were both in natural soil
backfill (sand, clay, loam) [and buried approximately 18 inches
below the surface. Both contained #2 fuel oil. The tank
interiors were also in good shape with very little corrosion (in
both the general and point category). Tank A had thickness
measurements taken: the steel was .180 inches and the fiberglass
coat was .128 inches. Neither tank was in contact with the
groundwater and there was no evidence of leakage.
Cathodically Protected Tanks
42
-------
FINAL REPORT NOVEMBER 1989
TANK CORROSION STUDY
Eight cathodically protected tanks were, removed. Four were
STIP-3 type tanks and four were other type.
Of the four STIP-3 type tanks, two were 8 years old and two were
9 years old. The 8 year old tanks were used as oil-water
seperators. Both were 2000 gallons and had average wall
thicknesses as follows: endplate - .1790 inches; top wall - .1773
inches; bottom wall - .1753 inches. Both had minor flaws in the
exte'rior coating, but 90 % was still intact. They were
backfilled in natural backfill (sand, clay, loam) and were never
in contact with the groundwater. No evidence of leakage was
found. The tanks did not have perforations. The only corrosion
problem was the interior of the tanks. Both were very scaly and
were classified as having moderate general corrosion. In the
area of the exterior flaws, the coating was brittle and peeled
off readily. It appeared as though corrosion had begun. There
may have been some installation damage to the tanks.
The 9 year old tanks were used to hold a solvent material. These
tanks were also backfilled with natural soil and were never in
contact with groundwater. The coating was completely intact and
there was only minor, corrosion associated with the interior of
the tank.
None of these four tanks had perforations.
Of the four 'other' types, two were made by one manufacturer (A)
and two by another manufacturer (B). None of these tanks were of
the STIP-3 type.
Manufacturer A's tanks were 5 and 6 years old. Both contained
#2 fuel oil, and had a capacity of 10,000 gallons and 2,000
gallons respectively. Both were in a natural backfill and never
in contact with the groundwater. Neither leaked any product.
While corrosion was only nominal, it was apparent that the
coating was bubbling and beginning to peel. The coating was very
thin and corrosion was beginning to take place under many areas
of the bubbled coating.
Manufacturer B's tanks appear to have been custom built. They
were of unknown age. Both contained chemicals used in the
plastics industry and were 10,000 gallons.and 20,000 gallons in
capacity. Each of the tanks had baffles on the interior. The
10K tank had its baffles reinforced while the 20K tank did not.
Each tank had two anodes attached by wires to three locations on
each tank. The exterior coating was 90 % intact on the 2OK tank
and 80 % intact oh the 10K tank. The 10K tank had some severe
exterior pitting on the bottom of one endplate. No leakage was
evident on either tank.
43
-------
FINAL REPORT
TANK CORROSION STUDY
NOVEMBER 1988
BIBLIOGRAPHY
1. USDA - Soil Conservation Service, Soil Survey of Suffolk
County, New York, Issued April 1975 (No publication number).
2. USDA - Soil Conservation Service, Publication 430-VI-NSH,
Pages 603-38 and 603-39, July 1983 (no publication title).
44
------- |