PB88-180112
Control of Volatile Organic Contaminants in
Groundwater by In-Well Aeration
(U.S.) Environmental Protection Agency
Cincinnati, OH , "
Mar 88
-------
Regional Center for Environmental Information
US EPA Region III
1650 Arch St.
Philadelphia, PA 19103
-------
PB88-180112
I
-^-'^;^Kv:^-^S'v'. ":''-;" V'^^-?^----":^ V:; < \:-ry-i"EpA/60b/2-8oY020 ' "^VV'^t'
: -^>'l . ';!-. ;; V::?.0 - -: '-v - -; :- ". -;:''. ;,-.; March 1988 . .. . ' ' :^f - ^'-.
' ' ''' '" ' '
i: CONTROL OF VOLATILE ORGANIC CONTAMINANTS
' ": - TM - .-,.''
----- :.-- - ^_ : IN .-'.-- .
.-,;-'.'; . GROUNDWATER BY IN-WELL AERATION
;
.-.. - .
' Judith A. Coyle
Harry J. Borchers, Jr.
North Penn Water Authority
Lansdale, Pennsylvania 19446
. . .
Richard J. Miltner
Drinking Water Research Division
Water Engineering Research Laboratory
Cincinnati, Ohio 45268
Cooperative Agreement CR-S09758
Project Officer
Richard J. Miltner
Drinking Water Research Division
Water Engineering Research Laboratory
Cincinnati, Ohio A526R
WATRR F.NOINEKRING RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIROKMENTAL PROTECTION AGENCY
'CINCINNATI, OHIO 45268
REPRODUCED3Y
INFORMATION SERVICE
.. SPRiNGFlELD.VA. 22161 .
-------
"- Y
pgp^sfe^^f?^::^^
m
f '"'.''.,.
fr .; ' '
f ''-' " '.
('-'
i"""'.
*
r * l"
*
c . '
f. '-
f , -
L
r-
j-
K
!.
i
i ..
J t
t - .
' :'--
>.
1 i
-.V ''V^V;-:.:- '"''.', ~;" '-:' '' .' VV~:',"" ."->'- . v'^.V;, / /.:-. .\-s:--.
!3^E5T^fTฃ7J?>T?[jS'
-'.vv'v;-i':;:v;;
^^^^^^M^^^S^
- "".-.' -' '-. '-'''::.'''"'''''. "j.
."':-.'..--- -'. '. TECHNICAL REPORT DATA . ;?.-. .
"-.-' ; ; -''.. (Kate rcid tnsirjc nous on the reverse before c ample tirtfj . ..''
1. REPORT NO. . . |2. ' . -
- EPA/600/2-88/020 I
4. TITLE AND SUBTITLE - ..
CONTROL OF VOLATILE ORGANIC CONTAMINANTS IN
.GRO'JNDWATER BY IN-WELL AERATION
' ' " ' ' .
7. ALJTiiORlS)
Judith A. Coyle, Harry J. Borchers, Jr., and
Richard J.-Miltner - .
?. PERFORMING ORGANIZATION NAME AN'D AODRESS
North Penn Water Authority -:
200 N.. Chestnut Street . .
Lansdale, PA 19446 . . - '
>2. SPONSORING AGENCY NAME AND ADDRESS
Drinking Water Research Division
Water Engineering Research Laboratory
Office of Research and Development, U.S. EPA
Cincinnati, OH 45268
3. RECIPIENT'S ACCESSION NO.
PQSi- 1 to in
5. REPORT DATE
March 1988
5. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION F.EPCST NO.
10. PROGRA.V ELE.MEr.'T NO.
C104, BNC1A, Task 109 S
:i. CONTRACT/GRANT NO.
CR-809758
13. TYPE OF REPORT AND PERIOD COVERED
Final 2/8ฐ 2/ฐฃ
1ซ. SPGNS'ORTSi'CT ^Ge'dC'V CODE
EPA-600/14
.15. SUPPLEMENTARY NOTES , , . ..-
"% - ' " ' " . . . - ' . '
16. ABSTRACT . . .
~v-At a 0.1 mgd well contaminated with several volatile organic compounds (VOCs) ,
principally trichloroethylene (TCE), several in-well aeration schemes were evaluated
as control technologies. The well was logged by the USCS to define possible zones
of VOC entry. A straddle packer and pump apparatus were utilized to isolate those
zones and define their yield and level of VOC concentration. The technical litera-
ture together with this knowledge of the well were used to design an air lift punp.
Operation of the air lift punp confirmed literature prediction of its low wire-to-
water efficiency. Removal of TCE did not exceed 65 percent. Mass transfer occurred
in the punp's eductor. Air lift pimping coupled with in-well diffused aeration
increased TCE removal to 78 percent. When in-well diffused aeration was used' with
an electric submersible punp, TCE removal averaged 83 percent. In the latter two
schemes, mass transfer occurred utilizing tho uell as a countercurrent stripper.
These technologies are limited by the volume of air that can he transferred to the
well (air-to-water ratios below 12:1) and the cost of compressing air under high
head. Thus, these technologies are not cost-effective compared to packed tower
aeration. They are, however, quickly put on-line, easv to operate, and can serve
as good short-term remedies while above-ground technologies are under design and
construction.
17. \ KEY WORDS A.VD DOCUMENT ANALYSIS
a. DESCRIPTORS b-IDENTIFIERS.'OPEN ENDED TE RMS
in-well aeration, air lift pumping,
diffused aeration, trichloroethylene,
vinyl chloride, cis-l,2-dichloroethylene,
well characterization
'.I. S:ฃT.=.!E'_'T!2N STAT-'"^"1" 19. SECURITY CLASS ITI:n K>;-""':
Unclassified
Release to Public so. SECURITY CLASS ,T/IMPซ.ซI
Unclassified
c. COSATi FiclJ/Cioup
21. NO. Of PAGฃS
m
;:. PRICE
flfok /M-s-
i
' ~ i
;(
-. ' 1
a
1
-
,_
;
;
-i
ft
-;.. p
t ;
* i
HP* Foซm 2220-1 (Riv. i-77) PREVIOUS COITION u OBSOLETE
j 1
I'd
-------
fS^^
<&&^^
.: 'r-^':".^-'^' '.'"-:,, -'/" ' - .'-.- '.:';.''" -V'^r'" '
^v'' '.<:?-. -';,.
:||iif:|ilil|l
DISCLAIMER
= The Information In this document has been funded wholly or In part
by the United States Environmental Protection Agency under assistance
agreement CR-809753 to the North Penn Water Authority. It has been
subject to the Agency's peer and administrative review and has been
approved for publication as an EPA document. Mention of trade nanes or
conmerctal products does not constitute endorsement or reconnendatlon for
use. ' ' . - ' . ' -:.- ,.'-'- . ' -. .'.'.-.''
-------
. . .
> i
FOREWORD
The U.S. Environmental Protection Agency Is charged by Congress with
protecting the Nation's land, air and water systems. Under a mandate of
national environmental laws, the egency strives to formulate and.
Implement actions leading to a compatible balance between human
activities and the ability of natural systems to support and nurture
life. The Clean Water Act, the Safe Drinking Water Act, and the Toxics
Substances Control Act are three of the major congressional Isws that
provide the framework for restoring and maintaining the Integrity o* our
Nation's water, for preserving and enhancing the water we drink, and for
protecting the environment from toxic substances. These laws direct the
EPA to perform research to define our environmental problems, measure the
Impacts, and search for solutions.
The Water Engineering Research Laboratory Is that component of EPA's
Research and Development program concerned with preventing, treating, and
managing municipal and industrial wastewater discharges; establishing
practices to control and remove contaminants from drinking water and to
prevent Its deterioration during storage and distribution; and assessing
the nature and controllability of releases of toxic substances to the
air, water, and land from manufacturing processes and subsequent product
uses. This publication Is one of the products of that research and
provides a vital ccmnunlcatlon link between the researcher and the user
corrmun I ty.
Under the Safe Drinking Water Act, the EPA will regulate
trlchloroethylene, vinyl chloride, cls-l,2-dlchloroethylene and other
volatile organic contaminants found In the nation's groundwater. While
the EPA has stipulated that granular activated carbon adsoprtlon and
packed tower aerators are "best available technologies" for control of
these contaminants In drinking voter, other technologies may be used.
This publication reports on an evaluation of In-well aeration
technologies. Air lift pumping with and without In-well diffused
aeration, and In-well diffused aeration coupled with electric submersible
pumping were evaluated. These technologies were not as cost-effective as
packed tower aeration for long-term control of these contaminants. They
will, however, provide short-term control at low capital cost while above
ground, long-term technologies are under design and construction.
Francis T. Mayo, Director
Water Engineering Research Laboratory
-------
g?gllpf?fl^^
";^A'':-.-^
' ':'; ..-..; .:;.,.'- ... / ABSTRACT -~ " .-" ^ :./",:.-.;.; v.^
.- ' ;,.-'- ' ' ' "':.- ' '.:. .. ". '"'. -' ' '. - " . - ". ' ;:'X;V,V ' -".'"'-'-/I
.....,.., .. . - ^ ........ . ^ . _. ... .-;,-_.-.., (.-J
Several In-well aeration schemes were evaluated as control j
technologies at a 0.1 mgd well contaminated with several, organic ' '...-,. ^
chemicals (VOCs), principally trlchloroethylene (TCE). The well was ' I
logged by the U.S.G.S. to define possible zones of VOC entry. A straddle ' r^
packer and pump apparatus were utilized to Isolate thosa zones and define ; ?
their yield and level of VOC concentration. The te-ihnlcal literature "'''-.'..'1
together with this knowledge of the well were used to Design an air lift .-' ;i
pump. Operation of the air lift pump confirmed the literature prediction . -.^
of Its low wire-to-water efficiency. Removal of TCI" did not exceed 65 -' ; .>
percent. Mass transfer occurred In the pump's eductor. Air lift pumping '- .
coupled with In-well diffused aeration Increased TCE removal to 78 ;
percent. When In-well diffused aeration was used with an electric . r'
submersible pump, TCE removal averaged 83 percent. In the latter two 5
schemes, mass transfer occurred utilizing the well as a countercurrent :'^
stripper. : . . ; ?'
Ofl:-gases of the In-well aeration process were tested and It was '
determined that, depending on the raw water chemical concentration, air
pollution discharge permits may be required to operate an In-well :
aeration treatment system. Dissolved oxygen and pH both Increased with
In-well aeration which may have distribution system corrosion
Implications. Bacteriological testing was Inconclusive. . ' ;
These technologies are limited by the volume of air that can be ;
transferred to the well (air-to-water ratios below 12:1) and the cost of . ]
compressing air under high head. Thus, these .technologies are not :
cost-effective compared to packed tower aeration. They are, however, :
quickly put on-line, easy to operate, and can "serve as good short-term
remedies while above-ground technologies are under design and
construction. _-..-;
This report was submitted In fulfillment of .cooperative agreement
number CR-809758 by thซi North Penn Water Authority under the partial ,
sponsorship of the U.S. Environmental Protection Agency. This report ..
covers a period from February 19S2 to February 1986. Field work was
completed as of March 1985. './
Iv
-------
; K-^'-v-'li.:U"Vv "'/ .'./.'.'"'''" . CONTENTS ,'-; "V . :;: .;:;/'- '<'-. ' -'..v '/-';.:;;-"-;
Foreword Ill , : ,
Abstract ...........;.... Iv. ;' v^
Figures vl - ' .
Tables ..-..'...;...... Ix :- v
Abbreviations x . . . ...
Acknowledgment xl " V - -.
1. Introduction 1 - .
2. . Conclusions %
, -.3. ReconrnendatIons 8 '. .- .-
tf. In-Well Aeration 9
5. Well Characterization '...I.... 12 ' , . |
o. Experimental Design 32
7. Prel Imlnary FootpSece Investigations ^2
8. Air Lift Pump Testing Without a Sparger 61
9. Air Lift Pump Testing With a Sparger 72 ,,
10. Electric Pump Testing 8*1, |
11. Secondary Effects of In-Well Aeration 93 1
12. Cost Estimates 101
References 103 .
Appendices . j-|
A. Surrmary of Equations Used 106 F|
B. Gas Chrcmatographlc Conditions and Quality Control , ||
Program 108 ^
' ' .".. .. ''"-' -M
'.-.: ^.- ., ....:..::/ ;/--.v.-".-;,;.:|
!
:S
:-i
-4
-------
r^S?*^^;r5>i3^^ฃ~'':Tฃ^ฃ^f^1s
:' ''. ;..-.' '."" . :.: ; :; FIGURES '-. ' ;. 'X.:-'' :".-
Number ; ; ' :..'.' .- . "-;'. . . .: Page
. 1 Diagram of Air Lift Pump Operatic .... .............. ' "~ 10
2 . Diagram of wel 1 logging system used during .:
, ; character I zat Ion of wel 1 L-8 .... ................... 17
3 : Profiles produced by U.S.G.S. well logging
7 atwellL-8 ................. . ........ .... ......... . 19
** Straddle Packer System used to characterize . ".:.
well L-8 .......................... . ............... . 21
5 Surface support system for Straddle Packer .......... 22
6 VOC concentrations of Isolated zones during
packer testing ..................................... 25
7 Distribution of total VOCs In well L-8 ......... . ____ 26
8 Well L-8 water level drawdown during purplng test ... 29
9 Well L-8 VOC concentrations during punplng test ..... 30
10 Schematic of air delivery control system .......... . . 33
11 Diagram of air lift purp and sparger footpieces ..... 36
12 Matrix of depth settings for air lift pump and .
: sparger ............. ...... .................... ..... . 37
13 Example of field data collection sheet ............. . 39
I1* Drawing of v-notch weir box ................... ...... 40
15 Diagram of footplece testing equipment ... ..... . ..... ^3
16 Operating air pressure and cost for air lift pump
footplece tests .... ........... ..................... *t6
17 Air lift punplng efficiency for footplece tests
vl
-------
'.-%'"' .'"; .18
.' .:-; ; , v 19;
' - 20
;: 21
> >:y ' 22
'<.. '-. -.. 23
- : '. 2ซป
;. . 25
26.
. 27
28
: 29
c 30
' '"-''.' . 31
32
: 33
3ซป
"J^T ;' v,- :^.:j -;^:' V-C V:V: ^M^^'^ Vr- r^'-' .; ''':'.' '
er . ~;- _ v^ --. , ; J;' "-r^- >:;:-::; .;;.>?-.-"-; -'; .' .'.
.Removal of major contaminants during air lift puntp
footplece tests ....'
Removal of minor contanlnants during air lift pump
Cost for air lift pump and sparger combination
Removal of major contaminants during sparger and
Removal of minor contaminants during sparger and
air lift pump footplece tests
Air lift pump submergence, pressure and efficiency
at three depths In wel 1 L-8 *
Theoretical vs. NPWA air lift pump efficiency data..
Cost to compress air for air lift pump tests at
three depths
VOC removals for air lift pump testing
Raw water VOC concentration at well L-8 during
air 1 1 ft pump test Ing
Pumping efficiency of the air lift punp while
Water flow rates of air lift punp and sparger
tests
Cost to compress air for air lift punrp and sparger
VOC removal for air lift purp at 130 ft while
sparging
VOC removal for air lift punp at 200 ft while
VOC removal for air lift pu-np at 280 ft while
to ' /.'-''- S~- ?."-,.: Si
' 50 " ' ''"' i-v'':'" ',';"l
53 ' ' "' :--; ^''"' \
. 55 .-'.'-..- :-:'"' i
56 "; ?-:;'^'V/v .-'I
-' 57 ": .' ' - '''.-,"'' ^'-''1
63 ' ' .-' : * A"'"
' f i ' " "*
65 i
i
68 '
69 '
. - <
75 "' "ป
77 '""
79 :
80 i
' . 4
- ..' ."J
81 - ri
' ' ; .- ' - !
. .. . i
vll
J-..:^i_l-'A- -i -.^-ij -iiป- '*
-------
Nunber
'--.'-':'--v.35..'.."'; ..-Drawing of electric punp and sparger at well L-8 ...
:.'.* 36 -: ,VOC removal by electric punp with sparger at three
Page
85
88
; 37 ".','_' .VOC removal variation over time with electric punp
: . ; . * '. fino spsrQGr ป*ซ**ป.*
" _ " * """;"-"-"' ' ' 7 ' "' "- - ' - . "'.'"'"'--'- "*-'
38 ." Cost of electric submersible purp and sparger '
-, - . ., .-.... ... ,. -
;39 "-:. Drawing of air samp ling, system ........ ...........
-**0 Air sampling location diagram of well L-8 ........
. .
y *
:'.': '.-'
96.
' 98
'tv
.,.
' "
n
;',!
U
>3
.-fr
8..f
a
1
VIM
-------
^^ ^pg!;:^li^^
f ;
&"
" i'
/''' /_'.'''' .-'.' "- TABLES " ' " .'.., - '
Nurber . . . Page
1 Description of NPWA well L-8 ". .....i. 2
2 Historical VOC data from well L-8 Ik
3 Baseline groundwater quality of the Brunswick
' Formation, Pennsylvania , 15
k Potential water bearing fracture zones and straddle
pecker positions for well L-8 20
5 Well L-8 specific capacity data suunary 2k
6 Well L-8 VOC concentrations for Isolated zones 27
,7 General experimental design 31*
8 VOC removal for air lift pimp footplece testing kS
9 VOC removal for sparger footplece testing 59
10 VOC removal by air lift pump without sparging 67
11 Glen Cove air lift purp operating conditions 70
12 VOC removal by air lift pumping at Glen Cove ........ 71
13 VOC removal by sparging v/Ith air lift puip at 130 ft. 76
14 VOC removal by sparging with air lift pump at 200 ft. 82
15 VOC removal by sparging with air lift purp at 280 ft. 82
16 VOC removal by electric pump with sparger 86
17 Secondary water quality effects of In-well aeration . 91*
18 Results of air sampling of In-well aeration
off-gases 97
ix-.
-------
ABBREVIATIONS
it
t^
t s
1,1-DCE - 1,1-dIchloroethylene
1,1,1-TCA - 1,1,1-trtchloroethane
'c-l,2-DCE - cls-l,2-dlchloroethylene
CCIA - carbon tetrachlorlde
cfpm - cubic feet per minute
CHCL3 - chloroform
DO - dissolved oxygen
ft - foot
GC - gas chromatograph(y)
GC/MS - gas chromatography/mass spectroscopy
gpd - gallons per day
HP - horsepower
HPC - Heterotrophlc Plate Count
In - Inch
n - number of samples tested
NPWA - North Penn V/ater Authority
PCE - tetrachloroethy1ene
%RSD - % relative standard deviation
psl - pounds per square Inch
PVC - polyvlnyl chloride
SD - standard deviation
SP - spontaneous potential
TCE - trlchloroethylene
ug/L - mlcrograms per liter
U.S.G.S. - United States Geological Survey
VC '".- vinyl chloride
VOA - volatile organic analysis
VOC - volatile organic chemical
?
4
i
i
\,i
;l
'i
= fl
'!'"
',
. -...,- , ,. - ...-
-------
i .'-".
:
.ACKNOWLEDGMENT
We wish to acknowledge the following people who have contributed
their efforts to this project: Tucker. Moorshead, Earth Data, Inc., St.
Michaels, MD., for his role In developing the original Idea for this
In-well aeration study and for conducting and Interpreting the well
characterization tests; Dr. I.H. Suffet, Drexel University, Philadelphia,
PA., for helping with experimental design and technical editing; Terry
Gable, Clarence Godshall, and William Smith, NPWA,. for field work;
Charles Hertz, Beth Green Hertz, James Towson, and Ylfen Tsal, NPWA, for.
analytical work; Ralph Walter and Dale Relchenbach, NPWA, for drafting;
Johanna Brown, NPWA, for word processing and editing; and the NPWA Board
of Directors for their support of our research endeavors.
-------
.'.-I -.--'';i^.-/-;>.'-C.;r:''.^r.;^^.H
^i-.^;'-'vvi;H--:;w^O^^ ป
Y.-:* -. , ' : - 1- *: '. ' *'".-. ~ t ** V*.JT
'-..>-j
-M
***$
SECTION 1
INTRODUCTION
OVERVIEW . ' :, , : :;' ,:. . : ;:
"'. Contamination of groundwater with volatile organic compounds (VOCs)
has become cannon throughout the United States. (1,2,3) Southeastern
Pennsylvania and the North Perm Water Authority (NPWA) have not escaped
this problem. In 1979 a large amount of trlchloroethylene (TCE) was
spilled in nearby Col 1egev111 e, PA. The Authority sampled all 3^ of
their operating wells and found eight to be contaminated with TCE and
other VOCs. These wells were shut down, resulting In a loss of
approximately one-third of the system's total pumping capacity.
Since that time NPWA has been actively pursuing various methods of
dealing with the VOC problem. Surface water was purchased from a
neighboring water supplier, a granular activated carbon treatment plant
was Installed at one well, packed tower aeration devices were designed
and this study was carried out to Investigate in-well aeration
techniques.
This report covers the design and operation of various In-well
aeration configurations examined by NPWA during the time period of
January 1982 to May 1985. The configurations Included air-lift punplng
with and without In-well diffused aeration (sparging), as wel 1 as
electric submersible pumping with In-well diffused aeration.
The well selected for this study was Lansdale number 8 (well L-8).
This well was heavily contaminated with VOCs and was being pumped to
waste In an attempt to control the contamination plune. A description of
the well Is given In Table 1. It Is In a mixed resldentlal/comnerclal
area, with homes In close proximity to the well house.
When a well Is found to be contaminated It Is cuuion practice to
pump It to waste, In order to prevent the contamination plume frcm
spreading throughout the aquifer. In-well aeration, by air-lift punplng,..
can treat the water as It Is being pumped. This can reduce the amount of
pollutants being discharged Into the sewer system (as at well L-8) or, In
some cases, may treat the water to potability. Treating the water while
pumping eliminates the need for construction of above-ground treatment
devices. In cases where above-ground treatment systems are necessary,
In-well aeration may provide a good Interim treatment technique while the
permanent system Is being designed and Installed. The in-well aeration
equipment used In this study was easily constructed from materials
already In NPKA stock.
Shortly after the discovery of VOC pollution at NPWA in 1979, a
series of preliminary tests were performed using an air lift pump or an
electric submersible pump with a sparger. No attempt was made to record
-------
^^...'r?.--?":-^--'".V.'.<-.' -' -: -V-;,"%'- "'-r.?-v."V'-..':'!': ;2!'A? v:'.-v "X /^/';v:,';/;-1-'.t;'/.^:;;"lv^r".-";C;A.;;
:-.' '.v:-'-- '-" ''-''' " '' ' ' ' ''' -' ' ' :- ' ' ' ''-" '' ' '<- '- ---.,', ' ' '> ---.-v^1* /.^-.'c^-.. :
:-.v;. air:to .water ratios, or operating conditions; however, even under these ./.i . .Vf.r
f.'vj uncontrolled conditions, good removal efficiencies were seen for TCE /V ;..~'^i"''
;.':..:.'. .-Cabove 80%). \ This provided the Incentive for further study of. In-well,. ' . ' ^
:,.-,-'-' aeration.- ' ''.:. ' :. :-' ' .-'-..-'.-'- -. ",;-. ' .-: .-'.' . -V.r-''-'iJ":.-1%i :-"-' ':
' TABLE 1. DESCRIPTION OF NPWA WELL L-8
\.. Location: West Third Street, Lansdale, PA.
Depth of Well: 292 feet (original), 286 feet (present)
Depth of Casing: 20 feet (8 Inch diameter) .
'_-; Date Drilled: 1923 /...-' : ; :;
Pumping Capacity: 50-70 gpm . ,
VOC PRESENT
HENRY'S LAW CONSTANT Catm - m
vinyl chloride
1,l-dlchloroethylene
cIs-l/2-d!chloroethylene
1,1,I-trlchloroethane
carbon tetrachlorlde
trlchloroethylene
pe rchIoroethy1ene
8.8
0.40
0.070
0.31
1.0
0.30
0.83
The air lift punp used for this study was similar in design to pimps
used by the Lansdale Municipal Authority (predecessor to NPWA) In the
1920's. Compressed air was Introduced Into an open-ended eductor pipe In
the well. The aerated water In the eductor was less dense than the
surrounding water In the well and was therefore forced up the eductor and
out of the well, because of this density gradient. The three test modes
.of In-well aeration for this study Involved purrping by an air lift purp,
air lift punplng in combination with sparging and electric submersible
pumping with a ?->arger. The sparger was simply an open-er
-------
- electric punp tnd then treating It by aeration. :' . ,
\\"i'.;. .:'' Certain sซicondary effects of In-well aeration treatment techniques
.were also erainlned. Off-gases In the well house were tested to determine
whether hazardous conditions were present. The air outside the well
.house, In the adjacent residential area, was also tested for VOCs.
Bacteriological changes as well as corrosion related factors (changes In
pH or dissolved oxygen with aeration) were examined.
-.The first stage of In-well aeration system design In this study was
characterization of .well L-8. This was done by the U.S. Geological
Survey (U.S.G.S.) with a series of well loggings. These tests Included
callper, conductivity, temperature, radiological and brine trace logs
among others. The U.S.G.-S. study determined possible water zones of
'- 'entry. '='--...'... '';;- .;''. - , ,'--.'
..Straddle packers were placed In the well to Isolate the water zones
of entry determined by the U.S.G.S. Each of these zones were analyzed
for VOCs and specific capacity. This knowledge of where pollution was
entering the well was used to determine the depth to locate the aeration
equipment. .(For example, If all the pollution was entering the well above
50 feet, locating the equipment at a depth of 200 ft, attempting to
maximize counter-current operation, may not be cost effective given the
cost of compressing air under that head).
Three different depths for In-well aeration equipment were evaluated
at well L-8, based on the findings of the straddle packer testing and air
lift purp theory. The air lift punp and sparger were operated at each of
these three depths. This provided several types of water and air bubble
movement In the well, Including both counter-current and co-current
stripping.
An electric submersible pump was fixed at one depth In the wel 1
while the sparger was tested at each of the three depths used In air lift
punp tests. The decision of where to place the equipment during electric
pump testing was based on the results of the air lift pump testing.
Parameters measured in the field during In-well aeration testing
Included air pressure, air temperature, air flow rate, water flow rate
and water level In the well. These parameters allowed calculations of
pumping efficiencies and air-to-water ratios. The In-well aeration
systems tested were evaluated based on these findings, as well as VOC
removal and cost.
-------
^-^^.,-^u,.,., - -'^'-..". v;;-^-'.'.-f."'.'"--VA"V^r^^-i^
v''-,^:'':\".''/.'.~;V-. ''"-"'''!v .':,"" : SECTION 2 ': . ''.-.'.:' -'; ' -.'^o'.-;
r';- '"".-/'-';/"-.'' '-*-'-.<:'-.''> '-' - :' -. -"..-CONCLUSIONS- - ". "'. >;.-'" --. V." '. '-",- --"-,: ';v::
r.-'" " _,. - > GENERAL . . ' . '". ''.':,"' ,/ - - ':,"' -" "'.''.:. ' /, ..''"'.'-' '^-^^"."'^'v
',,-';'' ..In-well aeration can be a useful treatment technique for VOC removal
:. on. a short-term emergency basis. Purging with an electric submersible
; : . ,' / pump while using a spargar was particularly well suited to this
:. " application. Many drinking water wells are already equipped with a
.submersible pimp and the addition of the air sparger was completed for
this study In a natter of a few hours with readily available materials.
'''. . For an operating cost of 12$ to 15$ per thousand gallons, VOC removals of
v '' 70% to 95% were achieved (depending on the volatility of the compound).
This method of aeration could be used to keep an Indispensable
. contaminated well In service while a permanent treatment system was
, - '. desIgned and Instal1ed. . .
In-well aeration by air lift punplng (with or without a sparger)
gave somewhat better VOC removal than the electric submersible pump and
sparger for a correspondingly higher cost. It was more complicated to
Install and was, therefore, not as well suited for emergency use (any
existing pimp would have to be removed). Its operating cost and limited
' VOC removal capabilities do not make It an acceptable long-term treatment
technique.
: WELL CHARACTERIZATION . - .
!, The U.S.G.S. well logging Identified two major and 5 minor potential
water zones of entry Into the well. This data was used to determine
; where to place straddle packers for Isolated zone testing. The packer
" testing accounted for 80% of the well's specific capacity, therefore, the
: " remaining 20% of the specific capacity was contributed by zones not
Isolated by packer testing. Differences In VOC concentrations were seen
; for the Isolated writer entry zones. The two most heavily contaminated
zones were also the largest water producing zones. The open borehole
purplng test showed that VOC concentration changed considerably with
; time. Over short-term tests, such as the In-well aeration tests
,i, performed here, large concentration variations could be expected.
: . FOOTPIECE TESTING - -
; The air lift pump and sparger footpleces were tested using large
! '. bubble and small bubble configurations. The small bubble footplece
/ ; caused greater operating pressures for the air lift punp. The pressure
-------
difference was greatest at high air-to-water ratios (5:1 to 12:1). Tlw" : " " / :;i;
most efficient operation of an air lift pump was found to be at a much : f^
Tower air-to-water ratio (1.5:1) where the pressure differences between . :;
the footpleces were very small. If the air lift punp were operated at ' ^
Itsymaxlmum pumping efficiency, there would be little difference In '
operating pressure between the two footpleces tested. If, however, the
pump were operated at a higher air-to-water ratio In order to obtain ''-:-_,
better VOC removal, the small bubble footplece would have a greater - '
/operating pressure and a greater operating cost. .- . . //~V r--. -.- x :.'
.The two air lift pump footpleces tested showed no significant
difference In air lift pump efficiency. The maximum efficiency was found
.-to be 30-35%, which confirmed the literature. There was no difference In '''''
VOC removal brought about by changing from a large bubble to small bubble
air lift pump footplece. Since there was no difference In VOC removal .
between the two air lift pump footpleces, and since the small bubble -
footplece would be potentially more expensive to operate, the small
bubble configuration was abandoned, and the large bubble air lift pump
footplece was used for the rest of the In-well aeration testing. . . >.
Sparging air Into the well decreased the pumping efficiency of the '
air lift pump. The small bubble sparger footplece had a higher operating
pressure, and therefore, a higher operating cost than the large bubble
sparger footplece. There was no difference In VOC removal between the
large and small bubble sparger footpleces. As with the air lift puip, ซ
the small bubble sparger footplece was not used for any further testing I
because of higher operating"pressure with no improvement of VOC removals. *j
In the footplece testing, VOC removals were similar for similar air- ; yS
to-water ratios, regardless of whether air was Introduced by air lift .3
pump or by the air lift pump and sparger combination. Vinyl chloride was
the exception to this as It was better removed by the air lift pump and
sparger combination than by the air lift ptnp alone. VOCs were removed
In an order which would be expected, based on Henry's Law Constant.
AIR LIFT PUMP TESTING WITH AND WITHOUT A SPARGER '
The air lift pump operated most efficiently at 65% submergence. The
. efficiency decreased as submergence increased, confirming the predictions
of the literature. VOC removal was poorer at the 280 ft setting of the
air lift punp than It was at 130 ft or 200 ft. VOC removal percentages'
for the air lift pump alone ranged from 90.^% for vinyl chloride (VC)
(best removed according to Henry's Law) to ^7.4% for c-l,2-dichloro-
ethylene (c-l,2-DCE) (poorest removal). TCE was 85% removed "by air lift
pumping without sparging.
A particular air flow rate setting produced different water flow
rates (and difference air-to-water ratios) on different days. This was
mainly due to water level changes In the well which caused air lift pump . fa
submergence to change which, In turn, changed pumping efficiency. Raw II
water concentrations of VOCs varied widely from one test to the next, as M
well as varying within a particular test. This confirmed the findings of ฃj
the pumping test conducted during well characterization. : 3
. As shown in the footplece testing, the sparger caused the air lift : if
punp efficiency to decrease. Even though the sparger caused the air lift " 'f*
pump efficiency to decrease, the punp efficiency with a sparger was ""'
, r
-------
&^^^
^/;-,- higher, .than If all of the air had been delivered to the air lift; pump
|.'-y,::L/;/- /\;; alone; therefore, In'terms'of purp efficiency Cand cost) It was better to
'a.i ''_ -'-.- :, '-operate an air lift pump and sparger combination, than the air 13ft pump
j ;- . alone. No-VOC removal differences could be seen when the sparger was
I1 .. ; operated at different depths with the air lift pump. This was mainly
;-_ , "-'.'_-. ' !- because of poor reproduclbll Ity of the sparger tests over long periods of
..;-'. * : .time., All measurement systems were checked and the most likely cause of
* ',; , ,'./_ variation was determined to be changing well conditions with time.
i :, ;: ' . .The air lift pump and sparger combination yielded VOC removal
i--"'" -' ""percentages ranging from 98.6% for VC to 65.0% for c-l,2-DCE, with TCE
;-.'. '.,:,'. -; : having .78.3% .removal. A higher air-to-water ratio was obtained by using
j , '"'; the air lift purp and sparger combination, which accounted for the higher
f ; ' - 'VOC removal capabilities. The highest air-to-water ratios obtained were
;-.."' ; 10.6:1 for the air. lift pump and 17:1 for the air lift pump and sparger
V !.:'--'.'.."' : combination. Air-to-water ratio was limited when sparging to the point
, ";'. at which water actually bubbled out of the well head because the entire
;-'. ':.. : ,. well had been turned Into an air lift pump. In a well with a wider
' ' borehole the air-to-water ratio might be higher because water would not
i . be forced out of the well as readily. Also, at well L-& some of the
; . borehole diameter was taken up by test equipment which would not be In
";''- - the wel 1 during regular operational conditions. - . :
Costs of air lift purplng and sparging make this method expensive
; for long-term use; however, It may be useful as an emergency short-term:
treatment system. The electric pump and sparger experiments were
ซ.' designed with this In mind. -
; ELECTRIC PUMP AND SPARGER TEST I NiG . - ;'-
, - .;. An electric-;sa)jrnerstble puip was operated at 200 ft, with sparger
testing being conducted at 130 ft, 200 ft and 280 ft. The best VOC
removal was obtained with the sparger at 130 ft. The 280 ft sparger
depth setting was next best, with the 200 ft depth giving the poorest
; removal. VOC removal findings were consistent with what was expected
. ' from well characterization experiments. Sparging at 130 ft caused
counter-current stripping as water from the most contaminated zone was
pulled past the sparger on its way to the pump. A co-current stripper
would have been created by the 280 ft sparger, with.at least some of the
air. being pulled into the purp before It reached the most heavily
: contaminated zone. With the sparger directly adjacent to the pump, most
or all of the air could have been pulled Into the pump before any
stripping occurred In the well. .
Electric putp and sparger testing showed the same variation over
. . .. . long periods as did the air lift purp and sparger testing. With no
"" -- other part of the experimental system changing, this helped confirm that
::' .. well conditions changing with tir.e CVOC or water zones of entry, not VOC
' concentration) could affect in-well aeration.
. The VOC removals obtained during electric pump and sparger tests
were an average of 82.6% for TCE, 79.9% for c-l,2-DCE and 92.9% for VC.
These removals were better than those achieved by air lift pump ing with
. or without a sparger. The air-to-water ratio used to achieve this better
removal was 8.2:1, which Is lower by half than the maximum air-to-water
-.'.' ratio used In air lift pump and sparger combination tests. For the
-------
\electrlc- pump/and sparger combination,' :a lower
-^achieved better removal than the air lift pump with or without a sparger.
, . Electric pumping with a sparger Is a good emergency treatment
technique If the limited VOC removal capabilities will meet your water
quality goals. It can be Installed very quickly with readily available
materials. It can be used to keep a well In service while the permanent
treatment system Is being designed and installed. . < :
; While well characterization was useful during this project, both for
experimental design and data Interpretation, It would not be necessary In
an emergency.situation. The sparger should not be set directly adjacent
to the pump to avoid having all of the air pulled Into the purp before
any stripping can occur In the well. . . . . ., .'-.-<.
"SECONDARY EFFECTS OF IN-WELL AERATION . ."';" ";'/ ; : / V''i; -7 ^:
All. methods of In-well aeration tested Increased the pH by an
-average of O.*t pH units. Dissolved oxygen (DO) was raised to saturation
.by all of the In-well aeration methods tested. Water entering the weir
box (air and water separator) was bubbly In appearance (actually milky
white when sparging), but all of the bubbles were gone by the time the
water left the weir box. . .
Bacteriological testing of raw and treated water was inconclusive, -
with large variations in bacterial counts masking any trends. The R2A
method provided consistently higher recovery of organisms than the
heterotrophlc plate count.
Air sampling showed that In-well aeration would probably not cause
air quality problems of Industrial hygiene concern, however, It may be
considered an air pollution source, and would require the appropriate
permits for such a source. This would depend on the raw water
concentrations of the contaminants. .
-------
SECTION 3 ;-
RECCf-MENDATIONS
s;: .It.. Is recomnended that, electric pumping with a sparger should be
', used as an emergency treatment technique. The sparger should not be
placed directly adjacent to the pump Intake, as this will cause the
bubbles to enter the pump Inradlately and any VOC stripping effect In the
well borehole will be lost. Air should be added In slowly Increasing
.amounts until the foaming water Is just visible below the well head.
This will .produce the greatest possible air-to-water ratio. An air and.
water separator Is necessary. ,' -: . . ; . .
- . Further air quality testing should be performed to determine the
possible need for air discharge permits. The necessity to treat
; off-^gases from In-well aeration to remove VOCs could negate Its
advantages as a quickly Installed emergency treatment system.
Possible corrosion Implications need to be further considered. The
elevated dissolved oxygen concentration. In the treated water could be a
problem over the long-term but It Is not known whether any additional
treatment would be necessary on a short-term basis.
-------
SECTION
IN-WELL AERATION
HISTORY OF AIR LIFT PUMPING
.The Idea of pumping with the air lift method was first tested by
Carl Emanual Loscher, a German, In 1797. It's first practical use was 50
years later by Cockford, an American. (5) The application was to pump
petroleum from a series of wells In Pennsylvania. It was first patented
.In the U.S. In 1865. During the early 1900's It was widely used In the
water works Industry. The Lansdale Water Company, predecessor of North
Penn Water Authority, used a steam powered air compressor for air lift
pumping during the 1920's and 1930's. After that, the air lift pump was
replaced by more efficient turbine or submersible electric pumps.
THE AIR LIFT PUT-IP . ', - .
The air lift pump functions by bubbling air Into an open-ended
discharge pipe In the well. The water In the discharge pipe (called the
eductor) then has a reduced specific gravity. The lighter water In the
eductor Is forced upward and out of the well by the heavier surrounding
water. See Figure 1.
Early researchers (5) showed that when the air lift pump was
operating at Its maximum efficiency (approximately 35%), the air-to-water
ratio was low (2 or 3 to 1). It was possible to Increase air-to-water
ratios to nearly 16 to 1, however, the pumping efficiency was reduced to
7%. It was surmised from this early work that compromises of either
pumping efficiency or organic removal efficiency would have to be made In
order to successfully use the air lift pump as a volatile organic
chemical treatment device. .
Another consideration when using an air lift pump was that It
required a high percent submergence (% of eductor below purping water
level) In order to work efficiently. The punp was simple to operate with
very few moving parts, therefore, the maintenance cost may end up being
less than tliat of an electric punp.
'IN-WELL DIFFUSED AERATION
Another form of in-well aeration that was examined In this study was
diffused aeration, or sparging. This method has proven useful for
aerating drinking water and waste water. The method Is usually employed
in treatment plant situations. In this study, a plastic sparge line was
dropped Into the well and air was introduced through this line. The
;t
f
j$
ซ r
9-
I:
1
-------
1_E
\ I
,
ESS DENf
MOVE: (.
OUT "OF v
MORE DENSE
WATER s
>E W/
JP g
WELL
^TER .: ;:: -"..;
" '.','' ''"'.
* eoucroR
. WELL
BQC2E HdLE Bซ.
oฐฐ0 ^ AieUFT
o ฐ o pu K/\ P
O O RDQTPiE'lE
O *| o
O
J
/ I
FOOT PIECE
^^
o
)
\ MOM- AERATED
X^ WATER.
Figure 1. Diagram of air lift pump operation.
10
f;;i
-------
t The sparge line was easily moved to various depths In the well. it:
could be placed, directly near pollution entry, zones.. Operating costs
could be adjusted by changing sparger depth, as It was "more expensive, to :
operate deeper In the well. ...'-".. " >/ ;; ^ -.'..-./ > V.^ / :\
The sparger could be set at a variety of depths'with relation to the
pump.;* The actual bubble path would be extremely difficult to determine.
Bubbles.could go up to the well head or be pulled directly Into the pump," f
depending on the relative positions of the pump and sparger. -
-'.''.-" f
WATER QUALITY EFFECTS OF UNWELL AERATION
,The primary water quality effect to be examined In this study was
volatile organic contaminant removal. Aeration In other forms Is a
proven treatment technique for these types of contaminants.
Studies using air lift pumps as sampling devices have shown that a
number of secondary water quality effects were also likely with an
.In-well aeration system (6,7,8). Air lift pumps used In monitoring walls
could raise the pH as much as 1.0 pH unit as compared to a perlstallc
pump or a bailer (8). This was because of bubbles of air In the eductor
stripping dissolved C0_. Tne same study also showed effects on various
metal concentrations. Iron concentrations were reduced by 32% when
pumped by the air lift pump as opposed to the perlstalIc pump. The zinc
concentration was reduced by;: 17%.. Calcium, potassium, magnesium,
manganese end sodium concentrations were not affected (8). Formation of
Iron .oxides was probably the reason for. the loss of Iron. Zinc
adsorption onto the Iron precipitates was cited as the reason for the
' drop In zinc concentration.
Dissolved oxygen Increase during In-well aeration was of concern
because of possible corrosion Implications. The air was Introduced Into
the water at depth, and therefore was under pressure. This pressure may
have caused more oxyqen to be dissolved Into the water than would
normally occur with conventional above-ground aeration systems (9).
ft
11
'i
i
\
i
I
-------
g^^T^
>!>* '
T * ,. ^~ * -, --..ซ* ป ~ J ซ
~ f '''A -~ ^ ~- ^ * c *r ~ j t
I t . ' " v '
t >ป
SECTION 5
WELL CHARACTERIZATION
.'PURPOSE OF CHARACTERIZING THE- WELL
- Well characterization Is helpful when designing a treatment system
for a contaminated well. Knowing where and how contaminants enter the
;well can be used In a variety of ways. In seme cases a section of a well
may be sealed off to prevent pol 1 utant entry while still permitting the
use of the well without treatment. Knowledge of pollutant entry zones
; can help to Interpret sampling results, especially when a small sample
pump Is used which may not be collecting water representative of the
entire well. This could lead to over- or under-estlmatlon of the
contaminant levels. Well characterization can determine the geologic
materials adjacent to the wall. A knowledge of contaminant zones of
entry, together with an understanding of the geology near the well, can
aid In contaminant source location (10). .
, In the case of in-well aeration, effectiveness of the treatment
method could depend on the placement of equipment In relation to
contaminant zones of entry (11,12). Knowing the water and contaminant
entry zones enabled the design of test configurations which would examine
both counter-current and co-current stripping. The well characterization
enabled the Interpretation of test results In relation to the equipment's
proximity to contaminant entry zones.
The well characterization performed at well L-8 was designed to-
answer specific questions relating to the design of the In-well- aeration
equipment and the Interpretation of test results. This was accomplished
through the following: . - . v .
1. Determined as-bullt configuration of the well "''".":'
(I.e., total depth, dlameter(s), casing length)
2. Determined location of water zones of entry.
3. Determined location of VOC zones of entry.
k. Determined specific capacity of entire well and
Individual water zones of entry.
, 5. Determined static water level. . / -. .
6. Determined how the quality of the water entering
the well varied with time.
" , 7. Determined the depths at whicli to place the In-well
aeration equipment. .
The characterization of well L-3 was done In four parts. Historical
data was compiled and-reviewed. A series of well loggings were performed
:: - - "'. ' -; . 12 - - v .-;-; - T: y/
-------
i
I.
-; 'to.determine possible water zones of entry. Straddle packers were used .
''-'. to Isolate, pump and sample the water zones of entry. A pumping test was
} performed to gather further hydrolog leal data. Each of these will be .
discussed In detail In the following pages. ' ,; . .- , ;r
REVIEW OF HISTORICAL DATA - .. v .
.Records of: the North Perm Water Authority were reviewed to compile
as much Information concerning well L-S as possible. Since the well had
been Inventoried by the Pennsylvania Geological Survey during earlier
studies in the area, reports published by the Survey were also reviewed
C13/11*). The driller who constructed the well was contacted and prior
pumpage and water quality data were reviewed (15). .
. Unfortunately, when the well was drilled In 1923 the contractor did
not record .changes In the llthology of the rock penetrated or any
Increases In the yield of the well that occurred at various depths. It
was assumed that, like most other wells In the NPWA system, well
L-8 penetrated layered consolidated rocks of the Brunswick formation and
water entered the open borehole below the casing at various depths. The
actual quantities of water and related qualities were not known and that
was the focus of the well characterization.
Well L-8 was In continuous operation from 1923 until It was taken
out of service In 1979 because of VOC contamination. The well was pumped
continuously to waste (to contain the VOC contamination plume) until this
field work began In February 1983. Table 2 shows VOC concentrations in
well L-8 during the time It was being pumped to waste. The 1979 samples
were analyzed by gas chromatography (GC) at a local contract laboratory.
The March 1981 sample was analyzed by the U.S. EPA at Annapolis,
Maryland, using gas chrarstography/mass spectroscopy (GC/MS). The
December 1981 sample was analyzed by the U.S. EPA at Cincinnati, Ohio,
using GC.
As Table 2 shows, well L-8 was heavily contaminated with a variety
of halogenated volatile organic chemicals. TCE was found In the highest
concentrations, with c-l,2-DCE and PCE also present as major
contaminants. Also found at lower concentrations were VC (a human
carcinogen) and carbon tetrachlorlde (CCm) (an animal carcinogen). Also
present were chloroform (CHCL3), 1,1-dichloroethylfcne (1,1-DCE),
1,2-dlchloroethane (1,2-DCA), and 1,1,1-trlchloroetnane (1,1,1-TCA), all
of which are suspected carcinogens (16).
Well L-8 Is located In a heavily populated, mixed residential and
commercial area of Lansdale, PA. Possible sources of the VOC
contamination were numerous.. The most likely source, however, was a
commercial facility directly across the street from the well. VOC
degreaslng agents were known to have been used at the facility In the
past, and In the early 1970's, the building was destroyed by a fire which
caused several explosions. Hundreds of thousands of gallons of water
were flushed over ruptured storage drxms, which is suspected to have
contaminated the groundwater in the area.
Natural groundwater quality Is controlled by the chemical nature of
the rocks In the aquifer and the residence time of water In the aquifer.
The natural quality of ground water in the Brunswick formation is
represented In Table 3. The table condenses data from 28 wells In the
1
ij
']
'l
13
*'
i- \
-------
BSK
1 ฃ
m-
1 '-"
P|
fc
t
P'
1
S-
1
||
jjp-
&-'
!ฃV
IS
f7
|f?
HCS
|f;:
ll
if:'
t,-
$i
$f
Ifi
IV'
ฃ-.'
fe-
,.
s, .
ฃJ
^;
'!* ~
'>,
f?;.
!'
|:
^
I-
*?.
|,
ฃ*.
*u
ฃ'
* -.
ฃ
F^
sl;.
f-"."
!<
f-
;*V.
ฃ'
1"
Plv
tf \
gift'.
gfc
"*C" "i."*--" IIL> ;J7T"T'l f^Tf^r- <- -y. ' 1 ' I 1 |- 1 n| i mi iiniiiiปii H
"' - '- . ' .- , ' 'V' '."' ' -' ' * - ' ' J- ' ~'-
""-'.- ';-"; \ *.-;'* <,'.'.''.';.;.''.'. '' ;' v-.r '_. -i'_.i". v_;;^, v
, ' .' " -*.'t ''f ~- *" ' -~* s." "V" ')" . - ^' ' .".*'.'- x "- * :r.' -. '-
---'': ^ " . ." ' -'.-' .''"X , - ' r -,' /_' " '.' ' ' : *--j "~ '" ', ' "
. ;: .:ri > i >j. Brunswick fbrnet Ion wh I'ch"
.'; ;,; - ' - l- . ' ' ,',v -'
*"" /.;'*'-.;-' "v;:f,*^i.;
. ''"'.>'t,.r:\ Civ/'. ",'.'
, V-.;'" ',. '."; " ', ^:/' "-^
;\7 '>--."' ^ -'.;". ';'.3
/:V ''\.'.'/ -"7X-?,
.''' ' "'.'. ~'~ -'"\;::> >'
.,- \
are not felt to be Influenced by
-. >,;{5^; phenomenon and would represent natural water quality
:- r?1 ' NPWA records of natural water quality at
'.:'"':--s -V: - ^ ;.;'., -.., TABLE' 2. HISTORICAL voc
'.''.>..:''. ' '
'':' ''- '-? - ''' '".- - , - ; ' ..-.
: "; PARAMETERS
'-' CHLOROFORM :
BROMODICHLC^WeTHANE
DI BROMXHLOROI^THANE
BROMOFORM
' DICHLOROIODCtCTHANE
. VINYL CHLORIDE
. 1,1-DICHLORQETHYLENE
. .1,1-DICHLOROETHANE
; . CIS-1,2-DICKLOROETHYLENE
i 1,2-DICHLORCETHANE
J 1, 1, 1-TRICKLOROETHANE
I ' CARBON TETRACHLORIDE
1, 2-DICHLORCPROPANE
" . TRICHLOROETHYLENE
1, 1, 2-TRICHi.OROETHANE
1 , 1 , 1 , 2-TETRACHLOROETHANE
TETRACHLORCCTHYLENE
' 1, 1, 2, 2-TETRACHLOROETHANE
' . CHLOROSENZEfrS
1, 2-DIBROMQ-3-CHLOROPROPANE
BENZENE
I TOLUENE
] ETHYLBENZENE
i . BROMOBENZErS
! ISOPROPYLBENZENE
| M-XYLENE
! STYRENE
i 0- + P-XYLENE
I ' N-PROPYLBENZENE
1 ' ' 0-CHLOROTOLUeฃ
; . P-CHLOROTOLUENE
[ . M-DICHLOROKN2ENE
1 0-DICHLOROBENZENE
|. P-DICHLORO^NZENE
{ .
ALL VALUES ARE IN ug/L
NA - NOT ANALYZED
LT - LESS THAN
i tO - NONE DETECTED, LOV^R
1 ' ' - -'' .
1 - . " .
* -
i . ' '
N .-..''
'\ '-. ' ' ' ' '
. '\ ' ' . ".' "
>.->"~^S^' -': ''' - " '"' ;;. ,
, DEC 1981
(GC)
---V-8;9 '
LT 0.2
LT 0.5
LT 1.0
LT 1.0
35. '
6.7
LT 0.2
400.
0.43
7.0
0.58
LT 0.2
650.
LT 0.5
LT 0.2
230.
LT 0.5
LT 0.5
LT 5.0
LT 0.5
LT 0.5
LT 0.5
LT 0.5
LT 0.5
LT 0.2
LT 0.5
LT 0.2
LT 0.5
LT 0.5
LT 0.5
LT 0.5
' -. LT Q.5/
LT 0.5
'"^"'1 '*'ฃ*-'' 'il' '-'
'R,V";'.-'.'*"^V ^
";'!K>>'W ?
cul tural
j*i^~ XX -S ' ^4
.' 'V-fr^ijJ'''^
-'kr^'vl-?;?^
=,;,.-. .,;-.:^-|
In the Brunswick. '-:.":-"..-iv''v'V*i
Well 8 are also shown. - ;- :? -\C ฃ^
DATA FROM WELL L-8 ""'; '"'"' .* '''.'!'^-^i
MAR 1981
(GC/MS)
. 0.5 *:
ND ...
0.4
-. ND '-:
. - ND
, 23.
8.3 .
ND
240.
ND
7.9
ND
ND
290.
ND
ND
120.
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND.
ND
ND
ND
ND
ND
ND
ND -
AUG 1979
CGC)
NA
:', NA ..
. NA
NA
NA
NA
NA
NA
: NA
NA
NA
NA ;
NA
410.
NA
NA
' . NA
NA
NA
- NA
NA
NA
NA
NA .
NA
NA
NA
NA
. NA
NA
NA
NA
NA ,'
NA
SEP 1979
CGC)
;VNA'V; :
NA
NA
- ' NA. '
KA - ''
NA
NA
NA
NA
NA
NA
NA.
, NA
702.
NA
NA
KA
NA
NA
NA
MA
NA
KA
NA
NA
NA
NA
KA
NA
. NA
NA
NA
NA
NA
' ' '-'.*
- '. ' ' ' ' 'i
' '"--'-- ':->
'" '' - . "" 'V
; V.',''-'?7:--]
. . , :. ... <
" "*"," 'J , -, "^
' *; ''- - *. '" -J
' ':^'-"v"|
: .'' ." ^
" ' " ". -j
' ' .'$
' * '.-.'-.
: -- .-" ' '
" -7
:"ซ
.'' . .' ' "'
_- i
- "' - ":
.
DETECTION LIMIT UNKNOWN '. . -'. ' . ;
1 A
. ' , .
,., . ' -
'-.'.
' .'-.'' '"
' * -
I"-.'- ". .. -:". . '-
J '. ' - ' ;
. , ":
' '-' . "i
-' '; '.. '-'!
-------
r-'-l^vfe-:*-
.t.-u ,>*?.'..;,>ฅ . -.-,
jffv:^^
i <:'.*4ฅ,ฃy.ฃZ
';*".X - -? ?V: ,- ' ;
ft%?SJSWงSvVi\^;
T^fe;,';-;tv:":>-;>::-.* :'^>x*..'"r,: ;
"V' - -* ' TI "' "' ---
' ~f" - ' '!> >^ -' "^pป- ' 'TARI P ^C
.^'1*' ~- ?'~' '~-<. ''':'^''. '.'^- ~~'- i, '"
'.':' -~::.'
''* '-*''. !; * '"' V-*"^."* * '"?'.'*'ป
'- ."' , .
... ., _- ^ _^.
"'-cC.^'
'-/C"'^'-:'~
-~W ;';';-".i'"'\-f r
:'"..v."^;;'.--r:-<--'-\--
BASELINE GROUNDWATER QUALITY OF THE
-;;^&S
BRUNSWICK
1;-.; -\;- :;--v/; :;,.: <-M:"V;>' -:--: ^ :^ ^ :-.-,. :>FORhwr I ON, /PENNSYLVANIA ; - . > -:-;.;;.
J -:'->-'.-"''i';;v ;i:v
*-' .--"' ;>' ..
t
r; ' "" . -'".'';" '' .-'"^ / : ' 'rf" ' "
*:',' ' ' ' . ' ' '
t ' -NUMBER
f ~
[
t
i
i
i
f .
f ' -, . ' * -
i . - '
AFTER R.E.
OB
CONSTITUENT WELL SAMPLES RANGE
. PH
Specific
Conductance
Nitrate (as NO-")
Chloride
Sulfate
Sodium
Iron
Manganese
Hardness(as CaCO,)
Bicarbonate
Total Dissolved
Solids
.
27
25
27
27
27
22
28
19
29
29
25
5.2 -
\
172. -
>
0. 10-
1.0 -
6.3 -
3.2 -
0.01 -
0.01 -
32. -
37. -
66. -
WRIGHT
(17)
^
MEDIAN
8.1
959
M.
31.
370.
22.0
1.6
0.38
500.
298.
732
7.2
381*.
6.5
8.5
31.
13.
0.15
0.01
100.
163.
'..
263.
i
UNITS
Standard
Units
Micro
mho/cm
mg/1
mg/1
mg/1
mg/1
mg/1
mg/1
mg/1
mg/1
mg/1
;ซ.o.ii^^^/^^3
K', -., i- ^V,-^7 T.-.--'* l-^-.T',.- - 3-' ----'ir, v
,^."^' , ^v^^V.'-j^^r 'V?>r'v.j5
' : ^r" : -,: '-1^:"^ป"^';>": -:.v^1-:Y^
~" J i.
NPWA
RECORDS"
OF L-8
7.7
f
-
V
39.
38.
18.
0.05
0.01
155. "'".:
'.- ' .' ' -
375.
" Average concentrations of samples collected at v-vell L-8 from 1970
_ - to 197.9 ..'.. ..-. .-. ..;.;. . .;"' .--.- ; ' . --..,-'.,'.- -..:.
Generally, ground\vater quality In the formation Is hard to very
hard, slightly alkaline and moderately mineralized. Groundwater at well
L-8 Is similar but has somewhat higher than average concentrations of
chlorides, hardness and total dissolved solids. ,
GEOPHYSICAL WELL LOGGING ; .
S*jbsurface llthologles, well bore features and borehole fluid
characteristics can be determined using borehole geophysical logging.
Borehole geophysics can measure the electrical, physical, chemical and
acoustical properties of adjacent geologic materials and fluids both In
the borehole and adjacent formations C10,18,19). .
15
fev.
-------
Borehole geophysical surveys were undertaken In well L-B with the
.following objectives: , . .- .;..--;.-.
Vv 1. To .locate-potential water-bearing fracture zones. ,. :.
"\" . 2. To locate relatively smooth borehole segments for the
X placement of Inflatable packers In order to Isolate the
;; > . : = water-bearing fractures zones.
; .;." 3. To Identify 1Ithologle changes (slltstones, sandstones, etc.)
; ,.; -k. To determine well depth, changes In well diameter and casing
'.'-"- -"length.": .. . ' -: '; . ' '-.- - . '- .._ .- .
f., The geophysical logs employed In thJs study Included; callper,
garnna, spontaneous potential (SP), resistivity, fluid conductivity,
temperature, and brine tracing. In each Instance a sensing probe was
lowered to the bottom of the well on a cable, then the probe was slowly
retracted as data was collected at the surface (see Figure 2).
Continuous readings were recorded on an analog strip chart recorder. The
logging at well L-8 was performed by the U.S.G.S. Information concerning
borehole geophysical logging procedures can be found In the literature
(10,18).
Borehole diameter and roughness were measured using callper logging
equipment. The hole diameter was measured by four callper arms which
were held against the borehole by springs. The arms opened and closed as
they moved up the borehole, drawing a picture of the physical dimensions
of the wel1.
Resistivity logs were employed to aid In the location of bed
boundaries and general changes in 1Ithology, which were associated with
water-bearing fracture zones. Resistivity logs measured the resistance
of the flow of an electric current from one point on the probe to
another. Generally consolidated rock had a high resistance because of
Its low porosity. V/hen the rock became more porous and contained a more
conductive fluid the borehole log showed a lower resistivity.
Spontaneous potential logs recorded the natural voltage developed In
the borehole due to dissimilar fluids contained In the pore spaces of the
formation and In the borehole itself. SP logs aided In delineating
formation boundaries and certain llthologles Csuch as shale and
sandstone). SP and resistivity logs were run simultaneously using the
same probe and the resultant log was called an electric log.
The measurement of natural garrma rays emitted from the formation was
used In conjunction with electric logging to differentiate strata.
Sedimentary rock, clays and shales all had a unique gamma signature.
Places where different strata intersect were possible water ^producing
;zones.
Conductivity logs measured the relative conductivity (Inverse of
'resistance) of the borehole fluid. This logging technique did not
'measure the specific conductance of the fluid because temperature
compensation was not provided. Conductivity logs were used to identify
differences In water quality due to fracture openings or breaks in the
'casing. Conductivity logs could also locate stagnant water In the
borehole and water that was moving from one zone to another because of
'differences In head.
'i Temperature logs were used to measure the temperature of the water
16
-------
p
f
fฃ
!
'!
! il
'."I
;---/ /-./ ':"
/
y////\
/
/
/
lA
ป
t>
4
J
'$/////
>v
:/
L___
Figure 2. Diagram of well logging system used during
characterization of well L-8.
17
X
t-.--'X
Li
-------
;Imnedlately adjacent to the sensor. 'Thermal gradients In 'the borehole
were examined-'to locate potential fracture zones. ''Temperature logs can
? "also bemused to Identify water that has recently entered the aquifer as
\naturar, or artificial recharge. _ . ..- - -."- /."A '..-,-- '---. -.; -..- : :;-'.:; -"/': .-.
-:. :'..'-.-:":'Brine' treeing was used to measure the direction and velocity of the
borehole fluid at selected depths In the well. A slog of a saline
solution was released In the borehole at a particular depth, then.It was
tracked with the fluid conductivity probe as It moved up and down the
, well. (20) ..--' - ; :' ';, .;, " -,.-.- ;-,:/ ^
,; ; Well L-8 was logged on February 17, 1982 by the U.S. Geological
-""''Survey's Pennsylvania District Office. Reduced profiles of the callper,
.natural gamna, resistivity and SP logs are shown In Figure 3. Potential
. water ..producing fracture zones are marked on the figure along with the
positions of the straddle packer system when It was subsequently
employed. ' ''-';."-'..".-" ^' :'-. ..--. . . . -:-'.'' : ' ''.'. -:,'.-; ..'.-.;
The well logqlng data was cohplled and plotted by hydrogeologlcal
consultants to North Penn Water Authority. (21) The consultant's report
Indicated that no one single well log was sufficient to characterize the
well. Instead, the log plots were compared by depth and when fracture
Indications were present on several of the logs at a given depth, that
depth was chosen as a possible water zone of entry.
.The logs Indicated that the casing In wall L-8 extended from the
pump motor base to approximately 20 feet below the ground surface. The
callper log detected major borehole enlargements associated with zones 3
and 6. Other enlargements wers encountered at zones 1,2,4,5, and 7.
The major enlargement at zone 3 was associated with a peak In gamna
activity Indicating a probable change from a sandy llthology to a low
permeability slltstone or mudstone. The resistivity log Indicated a
relatively low resistance at this zone which may have been related to an
Increased volume of water In a water producing fracture zone. Zone 6 was
associated with a marked decrease In natural garrma radiation, possibly
because of an Increase in the sand content of the associated bed. Zones
4, 5, and 7 also appeared to be associated with relatively lower gamna
signatures.
The overall llthology of the sedimentary rock sequence penetrated by
L-8 was fairly uniform with low permeability siltstone and mudstone
apparently dominant. Beds with Increases In sand content appeared to
occur between 73 and 92 ft, 151 and 170 ft, 182 and 195 ft, and 254 and
263 ft. - . ' - : -.-. "
The fluid conductivity log (not shown due to difficulty In
transcription from the original plot) Indicated a major change In fluid
characteristics at 50 to 60 ft and ..itnor changes at 172 ft, 188 ft, and
264 ft. The temperature log (also not shown) Indicated a major change In
fluid temperature at 56 ft, a minor change at 186 ft, and a very subtle
change at 244 ft. The brine tracing logs showed a major discontinuity at
approximately 70 ft arid again at 136 ft. A minor deflection was
Indicated at 180 ft. . .
Based on the results of the geophysical logs, seven potential water
zones of entry and Intervals for straddle packer testing were Identified.
These Intervals are shown on Figure 3 and in Table 4.
18
-------
:w!ฅป&^^^
Vfiซ^^^^
<ฃ>
DEPTH .
.FEET. PACKER FRACTURE.. .CALIPER
BETTING. ZONE, --.LOG
SPONTANEOUS::
'-'-^'--'':''''''"
Figure 3. Profiles produced by U.S.G.S. well logging of well L-8. ;(
-------
VTABLE k. POTENTIAL WATER BEARING FRACTURE ZONES AND STRADDLE PACKER
';;'; -/.-;' --. -.. : .POSITIONS FOR WELL L-8 ; . .. \- :..<
CFEET BELOW-CASING TOP) -
. . . Straddle
Potential
Fracture Zone
Fracture Zone Location
:'! :"'' '".
".?- - ,." " :
-'3 '.'-
4 ;
5 - . '
.6
7.
22
-;' ,-53
72
126
156
188
260
- 32
- 61
- 86
- 130
-161
- 196
- 261
ft
ft
ft
ft
ft
ft
ft
Packer Position
Center of Upper Center of Lower
Packer
28
47
70
121
152
182
252
ft
ft
ft
ft
ft
ft
ft
Packer
47
66
89
140
171
201
271
ft . -
ft -'-
ft :
ft
ft
ft
ft
STRADDLE PACKER TESTS
. Discrete zone borehole testing utilizing packer equipment has been
developed and used by petroleum engineers, engineering geologists and
. hydrogeologlsts for many years. By far the greatest application of
packers for borehole testing has been by the petroleum Industry and
engineering geologists for fracture permeability evaluation at dam and
foundation sites. The use of packer technology In hydrogeologlc studies
at North Penn Water Authority has been reported by Suffet, et al (11).
Shuter and Pemberton (22) described early U.S. Geological Survey systems
along with a system which they developed for borehole testing and
sampling. Cherry (23) describes a portable system developed for sampling
discrete zones in wells and Koopman, et al (24) describes a system of
Inflatable packers used In multiple-zone testing of water wells.
Straddle packer tests can provide a variety of information In wells
having multiple screen or water entry points. Because discrete sections
of the borehole can be Isolated when straddle packers are Inflated, the
tests can be used to determine the following:
1. Head difference and relative elevations of the isolated
zone and zones above and below the packer system.
2. The quality of water punped from the isolated zone.
3- The specific capacity and drawdown characteristics of
the Isolated zones.
The equipment used In this study was fabricated by Earth Data, Inc.,
St. Michaels, MD. A diagram of the In-well portion of the system
appears In Figure 4 and the surface support system is shown In Figure 5.
The procedure for straddle packer testing In well L-8 was as
follows: .
1. Measured lengths of lift pipe and added sections to reach
the desired depths.
2. Lowered sample tubing and wire into well with lift pipe.
. 20 '
{" ".'-
i. - ; '-'
.' - >
il :*--X
-------
v^F^-*3~f*'r&r-rr3&&F?;
"'/f-T.'V.-* '.'^^'X-w
r^^^ro^^^v^TsWf^.
^>sagtfa^'j&ffH^'"a''g^^
OPENING FOR
WATER LEVEL
MEASUREMENT
PU,kAP 5RACK.ET-
TO^RFACE
fn^ LIFT Pipe
V2BTU6E FOR WATER
LEVEL
BELOW LOWฃR
PACKER
UPPER PACKER
AOJUSTA6LE
D16CHARGE PIPE
'/a." TUBE TO IW PLATE
LOWER PACKER
Figure 4. Straddle Packer System used during characterization
of well L-8 to pump isolated zones.
21
I
-------
SOURCE
PUM9
CONTROL BOX
CABLE REEL
Figure 5." Surface support system for,straddle packer apparatus.
'/z. IN. TUBI
FOR WAT
LEVEL READIN65'
&S&J-
?*ฃ&&?
-------
'Cut 1/2 In tubing; measured water levels In the lower, middle : .
_/," and upper zones with packer deflated. . . '. : :',...".
. *f. .Determined hydrostatic: pressure because of submergence of each \. - . .^
. -." ."i. packer. ' . " _ - -- - ../- - :.' ;':';. '. ? v->'
5. Determined the necessary packer Inflation pressure by adding v "
:. '.;'';' .18 psl to the calculated hydrostatic pressure. ' '
'.,6. Started punp with packers deflated trying to maintain a .
' constant rate of discharge. " . . " . _: ; 5
. - . 7. Measured water levels and pimping rate. Collected samples - ; , ' i','"-^
for volatile organic analysis CVOA) just prior to the end " . j
: of the step. - .: ""; j
'.'. 8. Inflated upper packer Cthe resulting Interval was referred .! -'--,- ]
, . to as the lower composite test). . ... ^ ;; ; -1
: 9. Measured water level changes and pumping rate. ' ' V" . : -y
10. Collected VGA sample just prior to the end of the step. " ;. j
11. Deflated upper packer and Inflated lower packer Creferred ., .1
to as upper composite test). . . . "]
12. Measured water levels and pumping rate. Collected VGA * . .-t'"\
samples just prior to the end of the step. j
13. Inflated upper packer and left the lower packer Inflated .- :]
Creferred to as the Isolated test). . -'
14. Measured water levels and purplng rate. Collected water - ... j
samples for VGA. : ;
15. Stopped punplng, deflated packers, spliced 1/2 In tubing and ;
. moved the packers to the next position. . " . \
Zones 1 'and 7 dewatered rapidly during the testing, therefore only r
limited measurements were made and no samples were collected at these two
depths. Drawdown In the other zones ranged frcm 1.49 to 38.34 ft, :
Indicating a wide range of response to purplng. The head differences '']
between particular zones under non-ptuplng conditions ranged between 0.12 .;
and 38.24 ft. This Indicated the general effectiveness of the hydraulic {
seal created by the packers and the small potential for movemer.t between j
portions of the borehole. j
The specific capacity of the composite segments and the Isolated :
fracture zones were calculated and compared with an open borehole :
specific capacity Csee Table 5). The specific capacity data shows that
zone 3 contributed the majority of the well's yield at 37%. Zones 4, 2, j
6, and 5 contributed 22%, 15%, 6%, and 3% to the yield of well L-8, j
respectively. ]
The average open hole specific capacity was 4.75 gpm/ft. The sun of >
the specific capacities of each Isolated zone totals 3.82 gpm/ft. J
Therefore approximately 19% of the total specific capacity of the well j
was contributed by portions of the borehole not Isolated during straddle j
packer testing.
Figure 6 and Table 6 show results for each VOC Individually. A bar ,
graph showing the sun of the various VOC concentrations for each Interval
tested appears in Figure 7. Vinyl chloride data was not obtained during -j.
packer testing. The figures graphically demonstrate the general . .;
distribution of contamination In the well borehole. . '
23
-------
^
y"
f .
: . -
TABLE 5. WELL L-8 SPECIFIC
SPECIFIC CAPACITY (cjpm/ft of
PACKER UPPER
ZONE LOCATIONS COMPOSITE
1 28-47 0
2 47-66 0.91
3 70-89 2.31
4 121 - 140 3.24 .
5 152 - 171 3.63 .
6 182 - 201, 4.62
7 252 - 271 -
. ' '.'''." '''. -: ' ' '';' .' ^'.' V-:.'.- ..'''' ^-TV'.^ L
" . .' : .''.."'.?.''''..'. i -.' '. : ' '' '''.'-, .;^-.-;v'v-v;;t;V''.'
." ' .'.." .'!.> >" . ' '- '..- -.';::,.-'- :' .,ซ.'
.. . . ''''.'', '.' .- ' '.'.". -.'... .':t. ...._.=. . ,;,';;'., ''.""'.";''''{:
. .'' ' .'.. '. : ".;.' "';'.'.-''-.' *" v '.;.'".'.'... '"-' '..": '".'>'.' 'v>' ''y;'r;r*j4'-
.','. ' ': ''.' '' :' ''>'<' V :.' !"'...'' ''V ''.' '-^ --';',''' '''V.i''v;.:'<'''-,:v'J'
CAPACITY DATA SUMMARY .''''/' ; '" -^ ':": ' ' .' "-'' "'^^^^J^
drawdown) ' :: % OF OPEN HOLE SPECIFIC CAPACITY P'..?^- 'v'v^':
LOWER ' OPEN '.-.' ' .-." "':'-.-^-:^--:^':>.
COMPOSITE ISOLATED HOLE "UPPER LOWER -ISOLATED... .1 : ;;.' '.tv t^. ^^"'
- ' '"- ฐ ' ":' -"."' - : --'.-'- -:/:-:^V:-'';:W|^
4.78. 0.79 5.16 . 17.6 ' 92.6 ',"; -.15.3 ' :. . :/'\' %.>-^'v;V:^;
2.73 1.58 4.29 '54.0 64.0 ," 37.0 ?.'v /:v^;!;V^
; '. '-:-.': . -....:.>.;:, ; -. ,:^--^::^^,
' 2.86 . ' . 1.03 , . 4.62 70.0 '; 62.0''' '/22.0 -.'/ , .**.. '^^-^^ 1-
.' 1.03 '.-'. 0.15 ''I ,^.^ .[' 7k.O' ^':---- 2\.0 .^^ ^.O' .^.''^^f^^^fr^-
0.22 -'. :. ' 0.2?" - '.'ป.75-ti; 97.0 "j: ,'/'5.0; ' yf:' '.6.b/"V^.-^^^^i^
. . -. . . ' ' v .v; ...'v"."--.".- -,.'.:
. . . ' - -.,-!,.; ...V."'-. ' ._;
}' Average of previous four values . . : . V - . '."'-/; -.?v ^ '>.;;./
i , "' ' - -v>' < ' '- "V". """."'' "" i' '*
. : ,_,..,,( ,J .^ ; (- '' ;.^ , * y ,'ซ / '..-_,
. ' ' . ''.'< . '. * ''>'',..< './ ''-ซ' -" ' ' . ' '" >'v, -" 'J*1 *y *,.;,'; f * .
v " , -..-.-.. , ... ,*v '11 . .i-r / .-.'' r ., \-
-.-.- ' . - ' ' ; -- -..-,-. :'<.: .:--./:;;" .-./ V-;1.; :-i,^:^.jt-v
''.. "-.>'. '. ' ' ' ' '. :'':-': ''.,"" '-.""."""-.'-i' .'','" '-.'. "./\N-;;V; L^:':^s.-';r:;v<^^'/^
. . . ..''-,' .'' "'.' ' ' '''. :''',-' '..''.' ''.':..;,"'"-''. \' -i i- 'V "''^''';^' ''!NV.:;':;^::'';^';i:;Vv;KvJ*^-'^^
''c V"."- ' - ' .'. "'" ': - '.'. . ' ' ' ''.''. /' '-'"' !'v- .':.' ''. ' -'"'.' ''''", -'.:'' -.':7V"ov '-^/S^i'^'- '-^.Ci
- .' , . . "'.. .'.'. - ' ^ .*'.. \. ' .'. '',''" .V ,.;':. ,VK'.\,ji I/ V- v'" '' ' ' " .T'--'','.>'';i-^)^;'-f'i;^;'ii
; : ... : ' V " ''',,';.''," ' . ; '. " - v ''.-.:'"''' '"'"::''=.- !'i:S?''''V:fy'',.:V'^^.'-',. ^.^'^jf^&K$:'$
i -*
i
f
-;
j
]
{'
f
^
-------
K,
':!'--:'"''"^C"':600'V; "''.' " ". ' " "'" ".' " ' '" i f^f^' .>-;;!"ฃ$
- 700 -
600-
500-
: 4OO-
3OO-
id' 200 H
S -'-
z 100-
22-
8 2ฐ-
18-
10-
14-
12-
10-
a-
e-
'l * - A
' ' 2H
1.1 -DCE
1.1.1-TCA
0 20 4O 60 BO 100 120 14O 16O 18O 2OO
DEPTH OF ISOLATED ZONE (TCEO
Figure. 6. VOC concentrations for isolated -zones during
. " packer testing of well L-8. "''.
25
n
.!
. !v1
- *'*?
; ;'
rl
* i
-------
1.3
v
0.8 H
fg OJJ
be
&l 0.6 ^
tฃ
0.5 H
8 ฐ-4H
0.3-
0.2-
0.1 -
0
.-' -.-' '".' .'" 'ป':'''
'''., v
r1
/
/
/
/
/
/
/
A
^
\
\
A
\
\
\
\
V
\
\
s.
\
\
v
V
0
1
"1
1
J
1
/
1
f
/
/
1
/
/
I
/
/
f
/
\)
\
i
\
i *
N
\
\
\
\
\
\
\
\
\
\
\
V
\
'/
%
'/
v/
/ /
1 1
If
/ I
/ f
it
f/
J
71
'
1
1
1
1
1
1
1
/
I
f
I
I
/
/
f
/
fj
fj
X
\
\
\
\
\
\
\
\
V
v
\
V.
y
v
^
'//
//
'//
7j
'/t
//
f/j
V
i
/i
/j
'/
/I
'/,
ri
I
fj
/
/
1
I
/
/
/
/
/
/
/
-
\
\
\
\
\
\
L \
I
V
\
/
/
J
/
/
'/
''/
/
^
\
\
V
\
1
Upper Composite
T
4
ZONEf
Composite
-I 1 T
5 6 7
Isolated Zone
,:f
rl
-I
Figure 7. Distribution of total VOCs in \tell L-8 as
determined by straddle packer testing.
26
-------
TABLE 6. WELL L-8 VOC CONCENTRATIONS FOR ISOLATED ZONE
ZONE ft .
DEPTH (ft)
1, 1-DCA
c-l>2-DCE
1,1,1-TCA ...
ccL4
TCE
PCE
TOTAL VOC
% OF WELLS
YIELD
' 2 ' '
47-66
2.1
105
8.1
0.5
363
,58
536
15%
CONCENTRATION,
' :' -3' '" : 4 -
70-87 121-140
6.1
289
10.5
0.5
710
234
1250
37%
21.3
. 212
.16.7
3.3
483
118
857
.
22%
ug/L
. 5 . ' ;
152-171
20.6
^ 182
16
3.0
398
102
724
3% "
f, '-A' ."'': "
182-201
6.2
256* .
4.0 .
0.5
466 :
33 .. ; ." ' ..
,' 765 ':
, 6%' V. '' .-.:
The Isolated sample of zone 2 snowed that this was the least
contaminated of the zones Identified In the well logging, (see Figure
7). The upper composite sample was less contaminated than the Isolated
sample, Indicating that cleaner water must be entering the well above
zone 2. Since zone 1 did not produce any water, there must be a water
production zone between zones 1 and 2 which was missed by the well
logging. The lower composite sample was considerably higher In VOCs
than the Isolated zone sample, showing that the water below zone 2 was
more heavily contaminated than that at or above zone 2.
Zone 3 results showed a dramatic contrast with zone 2. The Isolated
sample at this zone showed It to be the most heavily contaminated water
producing zone In the well. This zone also produced the most water of
the zones tested, at over one-third of the well's total capacity. The
major contaminants (TCE, c-l,2-DCE and PCE) were all found In the highest
concentration In zone 3. It was Interesting to note that the minor
contaminants did not reach their highest concentration until zone 4.
(see Figure 6). The upper and lower composite samples at this zone
confirmed that the water coming from above and below was of lesser VOC
concentration than the water produced at zone 3. '
Zone 4 was the zone of highest concentration for 1,1-DCE, 1,1,1-TCA
and CCL4. The upper composite sample confirmed the presence of the more
heavily contaminated water from zone 3, and the lower composite showed
the presence of less contaminated water below zone 4.
The zone 5 upper composite sample confirmed the presence of the more
highly contaminated \vater coming from zone 4. The lower composite,
however, showed less contaminated water coming from below zone 5. Since
the Isolated sample from zone 6 was actually higher In VOC concentration
than the Isolated sarple from zone 5, It can be concluded that more water
(at a lower concentration than zone 5) must have entered the wel1 between
zone 5 and zone 6. This may account fcr some of the missing well
capacity discussed earlier.
The zcc-e 6 upper composite sample confirmed the presence of lower
concentration water above the zone. The lower composite sample was equal
27
-------
' 'In concentration to the Isolated sample. This, together with very
: - .'similar specific capacity data for the Isolated and lower zone (0.27 and
0.22), Indicated that there were no more water producing zones below zone
. - :.-.:-;.-,'6.'; '.; - .-'' . . . .-..- - .
The well characterization of well L-8 has Indicated that casing,
grouting or packing off a portion of the well would not be an effective
"-.' means of Improving the quality of the well, because the contamination was
entering at many different points. A well of this type would require
. treatment other than 1 solat ten of the contaminant entry zone.
'..-.> . The Information gathered by this well characterization was used In
designing the In-well aeration experiments to be performed. Further
explanation of how this Information was used can be found In Section *ป.
PUMPING TEST ' : . '-' -: .;~: '.
ซ?-) . Traditional methods of well characterization often Include pumping
|;j - tests. Pumping tests can be used to determine well efficiency and
jฃ;"'j long-term expected yield. Water quality samples taken during pumping
tests can be used to establish long-term trends In quality as well as
provide a representative sanple of nearby aquifer conditions.
. A pimping test was performed at well L-8 to further characterize the
well and provide Information about the nature of VOC contamination In the
vicinity of the well. It was desirable to know how the VOC concentration
would change with time, because each In-well aeration test took one full
day and was, therefore, performed from a resting (non-pumping) condition
of the wel1.
The test was conducted over a period of 5 days beginning on
September 20, 1982. The pumping rate, was held constant at 50 gpm and
jj.-. standard pumping'-test procedures were followed (25,26,27). Data from the
test are found In Figures 8 and 9. Results of the test Indicated that
the aquifer surrounding the well was not uniform. The average
permeabilities of the aquifer appeared to Increase, then decrease
away from the well. This could be seen by the changes In slope of the
drawdown curve (Figure 8). The starting water level was 21.5 ft below
the top of the casing. This dropped 15.1 ft, to 36.6 ft after 91* hours
of pumping at 50 gom.
During the purplng test, samples were collected periodically for
volatile organic analysis (see Figure 9). The major contaminants showed
|_: similar trends of decreasing within the first 0.2 hours of punplng then
J*V gradually Increasing for tho remainder of the test. TCE snowed the
greatest change, dropping from 720 ug/L to 410 ug/L after 0.2 hours, then
steadily rising to 1,190 ug/L by the end of the test.
The nature of the change In the VOC concentration suggested that the
ฃ= j sources of contamination were relatively near the well. The decline In
concentration during the early period of drawdown Indicated that the cone
of Influence spread Initially to areas that were not severely
contaminated, thus causing a dilution of the contaminants that enter L-8
under natural or non-pumping flow conditions. After 0.2 hours the
contaminant concentration began to rise due to the spread of the cone of
Influence Into the contaminated areas. After approxlnately 30 hours of
pimping, the area contributing contamination.was completely engulfed by
the cone of Influence and the rate of Increase In contaminant
28
-------
/
;- '-,. v v'-', :/ ._'_:_;;
I V-
z
0
Q
I
J
hi
Id
I
o
,'-1
-2
-3
-4
-5
-6
-7
-8
-9
-10
-11
-12
-13
-H
-15
-16
RATE = 50 GPM
ElAPSED TIME = 94 MRS
TOTAL DRAWDOWN = 15.1 FT
i
2
I
4
LN TIME (0 TO 94 HOURS)
Figure 8. Well L-8 water level dra^vdcjwn during pumping test
phase of well characterization.
N.
29
-------
re^Wp?^?f-?^gg^^
\
fl
I
ii
H
.a/
5
I
o
o
0
a
0
D
v
0
E
H
0
Z
0
0
LN TIME (0 TO 94 HOURS)
Figure 9. VQC concentration changes with time during pumping
test ef well L-8.
30
-------
^^"concentration began to lessen.
>;!.>: .This purplng test Indicated that large variations In . raw water:
..!. '^concentrations could be expected over apy single day of In-well aeration ...'.%.'.-'-.'
;;: testing. Because of this, It was necessary to monitor raw water ^ , \,^;
'^.concentration carefully during all phases of the testing. ;, .. ! '/c
.' ;, ;. -;"/'; -; ^ ='^-: /': 'V" \'1-
.*.- A review of historical data for North Perm Water Authority's well
-.< L-8 showed It to be Contaminated with a variety of VOCs, six of which -. :
; were found at high enough concentrations to be used for this study. ;.;
'"'.," These compounds.of Interest were TCE, PCE, VC, c-l,2-DCE, 1,1,1-TCA and '- ' "|
CCL*t. When the contamination was discovered the total VOC concentration ^<
.-'.: was approximately 1,000 ug/L. . : '. '.'i
"--'.: The Inorganic quality of well L-8 was typical of the Brunswick *:.-;
..formation. It was hard and slightly alkaline, with higher than average :'
chloride levels. -';' , :''.--.'_-'" -*!
. Well logging, performed by the U.S.G.S., Identified seven potential . ';. "
'..water zones of entry Into the well. This Information was used to '.'- /x.
". determine where to place straddle packers used to collect VOC and water . .
quality data for each zone.. . . ':J
The straddle packer testing of the seven zones Identified by the ':
U.S.G.S. accounted for approximately 81% of the well's total specific . H.
capacity; .therefore, the remaining 19% of the specific capacity was ; ..
contributed by zones not Identified during the well logging. In <:
addition, two zones Identified as potential water producing zones
dewatered almost Irmediately during packer testing.
VOC concentrations were different for each of the zones tested. The
two most heavily contaminated zones were also the two largest water
producing zones. These zones were located at 70 to 89 ft (zone 3) and at
121 to 140 ft (zone 4). Zone 3 contained the highest concentrations of
the major contaminants (TCE, c-l,2-DCE and PCE) and zone 4 showed the
highest concentrations of the minor contaminants (1,1-DCE, 1,1,1-TCA and . '.
CCL*O. The Information gained during the packer testing was used to
determine the depths at which to place the In-well aeration equipment.
The open borehole pimping test showed that VOC concentrations at ;
well L-8 changed considerably with time, and that the changes were .
especially significant, because the in-v.-ell aeration testing would all be
performed during this early part of the punp curve. Because of this;
very frequent raw water sampling was designed Into the In-well aeration .:
testing. . . . ,
31
.
-------
5Sฃซea3Sปฃs3SSE^s
SECTION 6
EXPERIMENTAL DESIGN
,the experimental design for this project was developed based on
Independent and dependent variables. The Independent variables were
experimental variables that were set In the field. The dependent
variables were data which were obtained by setting the independent
variables. The dependent variables were, In some cases, calculated (e.g.
horsepower) and In other cases were measures of quality changes (water or
air) which occurred as a result of varying the Independent parameters.
Table 7 lists these Independent and dependent variables.
Air flow rate was an Independent variable. The air delivery system
used to control and measure air flow rates throughout the In-well
aeration experiments Is shown schematically In Figure 10. The system
was mounted on a 4 ft x 6 ft plywood board and was located on the well
house wal 1 approximately *t ft from the well head.
As can be seen In Figure 10, the air delivery system was divided
Into two air flow pathways. Side A was used for the air lift purp and
Side B was for the sparger.
The source of compressed air was a gasoline powered Schram Air
Compressor capable of delivering 100 ps* of air to the In-well aeration
system. This was sufficient for the depths at which work v/as performed
at well L-8, but wells of greater depth would require a more powerful
compressor.
A pressure gauge (Gl) measured the delivery air pressure from the
compressor. Each pressure gauge was mounted with a test tee to accept a
calibrated gauge by way of a quick-connect fitting. All gauges were
checked against the calibrated test gauge at the start of each testing
day, as well as periodically during testing. Fol lowing gauge Gl was a
thermometer (Tl) used to measure the temperature of the Incoming
compressed air. Next there were two filters In series (Fl ฃ F2) to
remove oil and water vapor and partlculates. Fl was a Deltech 810 series
filter and F2 was a Fine AIre glass fiber coalescing filter. Another
pressure gauge (G2) followed the filters. Gl and G2 were used to monitor
pressure drop over the f'lters to observe any possible filter clogging.
After gauge G2 the air delivery system split Into side A (for the air
lift purp) and side B (for the air sparger). Ball valves (BV-A and BV-B)
were used to turn either side on and off. The ball valves were followed
by pressure reducing valves (P-A and P-B) to control the amount of air
pressure at the rotometers. Three Brooks rotometers were used. Rl-A and
Rl-B were both calibrated from zero to sixty cubic feet per minute
(cfpm). Rotcmeter R2-A'was calibrated from zero to six cfpm, and was
used only during the lower flow rate portions of air lift pump efficiency
32
-------
[.V ';j^ '".'' '.''.-''.
ฃ & V ' ' ' '
ฅ '
i . . ...... '
^ _.....- . . ,,:. ;:.-. .. _-.._.;. . : ',' '
'"^ ' ' . ' ' ''' .''
. , . - * . . . -
> . ' ' '
',.'. . : . . .
'.' .''' . . " ' 61 .
I- ฃWi 'PI
(.'. . L-TSC to
. AIR AIR y* ^ T <*
r < '.*"" coMpRe&cc i ' 5
' U.
i .'
iv... . . '
(jO
i. : w .. . LE6END
6 - PRE55URE GAUGE
." T - THERMOMETER
1 ',-. P - PRE6. REDOCIN6 VALVE
>:','., .'.. ' R ~ ROTO -METER
! .. ' .- f ' - FILTER
: . . . 6v - GATE VALVE
!. BY - 6ALL VALVF
[''..'_ -' .."' X -PET COCK
f'"'' ' -
' ' '''..
' 1 ' '
>... .
!' Figure 10. Schematic drawing of air delivery
|:| ' /;. '.' _ . . ' '_ .--.-..
r i ' .
;i" - . " .' ' .
t'1' '''...' , ' . ' ' '
1 ' , ' ' ' ' " ' '
i'1''1 ' ' ' ' ' V ' '..''
JT ''.' ' . .
M. .--..'' ' ' "' '
'*'l . .
K _.". ' ' ' . "" .. ' ' '' ..." . ','.' ' -"' ' ; :'.. ' ": '
H ''. ' ' ' .. '''''.'..
^Ju'l-:fr*v^i^s ?.^>i^jz^ ^.^J^JuAjJsii^a^;j^iia^*i^i^
. ':..,. ' ' :..'-' ' : , '. ". ...'; ''
' '; ' . ' '" '. ' '':", i'' ' ' ''"/' : .' ''-'' .'.' r'V' :' '.- 'V'X-- .
;''' ;-.' , ...:--/' '-.''.' '"?""':' '\;'\'/' '.'.^;'S;'is
- * - ^ * - ' ' . "- . ''- ....(,'.'. .'-.-..',"'' 't '-.'," ','.'. .'*'*,. '.-
. " . . . . - . . . -. .. ' v ., . 7 - ' > ' ,''-.' $ . ", -
.. .'..'.'. '' ''..'.'-;..::: " ':":.'' - ''' '!'' '^>;"--y-f-':^'-
' 6|-A T-A ' ' '62-A , :'-^. :..' .'...;: -. . /.>. '' .- :;;-f'! -'.'
ft> d\ - (f\ ' '" ' ' -''' ' -' " ' "
V YG/A 1* .' - A"? TOAlRUFT PUMP; .. > '.:-.. :T-.j^
RI-A-T .rjR2.A.. ' . ;,'-':::; ,1;v;
11 '.;-'', / . .' . ... ."; '"''''
^ifx I ___T.. . ; .- ,, ; ' . , ;-_;', -,..;... -./'.;.;'. .:.;;v...
' C> ' ' ' '.' ;' . ''. ''"' '"' . -". ''*:'': '' ''.-. '^'"'-":
f2 to* / '.:' ',''; '"'.'''' .^"': ,-:V': ' ;:^;;v-
ซ ฉ yrW-A . . ..:. '.:/.' ,. '-;,. -;/
UJ'J_/ '. 6I-6T-&' '.' ..'6Z-B': '-.'/;'. :./' ';i ' . .. ,,'; ;.'v'^"V;Hl'
C \,iซv.R T T H< AiR. T0 WAKfaER.1 -'-,'< ;V.'-,
*^ jiJLxV D ^/-* wf ^/ ' " ' '" T . ,
>v i GVB ' - . ;-.. '- ' :. , ' .
^w . 1^^^ " ^^ ' ' ' ' ' i ' ' . , '''.
. ^fU , '-Ri-B... ^,.:.'-'': ::"'':-! '.^.'v-! HvV
p.Q N^ ' . .- . " ' . " '.'.' . .' .. . ' .;- i V''.'> i
: -.-'.''' '".. ' ;:';. ", - '. '._ . .
' ' ; '. ! ' .'. ''. ' . : ' ', . ". ' -\ '-.-' ;
- . t i- '. " '-.'. ."'"".
. - ' .'.'.''''' ' :''; -. ' ' ''-. -' :- '.'-..- .;:''.'.". !;'' :'v-
'.'''. ' ' ' . " . '.."' "'"l' ' ,-
' '. '. ''...."'' ' ' ' ''' . '. ' .'': '''-,' '"'^
.' .. .- :-. :.-.' ;>;.' ;, ' ''.''.:' ' ';. -'"' ... '" '!..".'' -i'-.'
' ' ' .'..'-,'.' ' '." '"''
. .',' ' ' . ' . ' . ,.-.-.;.. ? .' ' '-. - _ '- i' ' - V; '. '-
''-. :\.'-':'i- ':'.''."'' . ,''-: -'-,. ^;-.;.:'i''^y-^-'^-&*'
' ' ' . - ' ' - . ' *'''.*.' ' ; *.-." -' '' "..'.*' ' 4'*. , t-"*"*''* . ". '' *
ซ - , * ' '. ' ' ' ' - ' T . ' ' '. . * ' ':, l '' j ' I '' ^ '
control system for in-well aeration testing.. :' ; ; . ' -.' :>'"
:'- ".''".. : .".'..'"'''' ."..- v;V'; '''.''^''-^'.: : r':i'..'. . '.';''.'''. .'?;'-'.'.'!'i>'
~ - ' ' ' " '"-',-, -'>\ ,'.' '''.*.''''.'-. -V '.:'.^.'
'. ' ..'''"'' "' .- '' ,' '."'' '' '.".' .'"'! '''"';
. - ' ,'-., --./.' ' - s ', . '.'- ' '' ' ' ~- "
'. ..' /'. '' '-' -/.-' - ' . / .."'.; ' ' .."'.-.. ,.,, '/'>>..'
' ..' ,. -.-..''. ...:.'.'"- : -..:.;:- ' ,.'. ;..';'-..;;-;''.' r:
,.' . '.' . - . .'....'.. -:.. ,'..' ... -.;r,'K' .(,
.,,.-,' . ;. ... ,',.'"-,. . . - ,
,.:.\ .' . . '.' '-'-.; .-/ i. ' ' -:V. ; *>-.'. >, .;'. :'.
- ' ' '"'-'','. "' ' '' " . ; - '''''. -; '." ''* !'^'1 '
- '.''. i - "' ..-,'. '"'* .' - ' ^ '. . *:~ . ' i-'.
'. -.-.-' ..' - ' ' .,'". - .;- ; ' I,'',; v..'1.; t' \r- ''.
. ' ' . . ' o .."..' ,..'. " v . * ' .' ,f -'.>->..*
.* . . . ,..'--.. ..*-.' - .' ;> -A .' , t i' J ^lj.Vi.<. .' .. * -3 4.'~l> . ' t,>. .r. ~t~ . >v.*. < .s ซป. JlWti'.^t'. .J'7
Hw
II
\.t
s
'J,\
f-'.'i
~)s
K.* ^
w
e, :[:
-'?,'-
v.).?
* i'M
'N
$
'"'*1
'>'
i>
I/:
1 " -
i;^
I'--
!|
,1
.: i.
'V
''.$
i
j
^>*vi
v^
'}*j'5
v-,3
^
,*jv^ln
^il
i^s
!'"'/
\fy
'si
ss
'<*
i'fi
'"T"!
M
-------
fcv
TABLE 7. GENERAL EXPERIMENTAL DESIGN
-INDEPENDENT VARIABLES
VARIABLE RANGE
Air flow rate . 0-116 cfpm
Footplece Configuration large/small
:. , ,.;..;_..; . , . bubble
Depth cf Sparger 130 ft,200 ft,
280 ft
Depth of Air Lift Pump 130 ft, 200 ft
280 ft
Depth of Electric Pump 200 ft
Pumping Water Level
ft - 70 ft
DEPENDENT VARIABLES
VARIABLE RANGE
Water.flow rate . 20-100cfpm
Submergence 50-75%
Water Horsepower 0.5-1.7HP
Air Horsepower 0-10HP
(adlabatlc)
Required Com- 0-15HP
pressor HP
Estimated wire-to-water
efficiency 0.2-^0% .
Cost to Compress
Air
Raw Water Quality organic
Cprlmary) removal
Treated Wtr Quality organic
; (primary) removal
Raw Water QualIty pH/DO bact
(secondary)
Treated Wtr Quality pH/DO bact
(secondary)
Air Quality
OrganIcs
In air at
wel1 head,
weir box G
20 ft from
weir box
testing. Another pressure gauge followed the rotcmeter on each side
(Gl-A and Gl-B) to determine air pressure at the rotcmeters. These
gauges were followed bv thermometers (T-A and T-B) to measure air
temperatures at the rotometers. Rl-A, R2-A, Rl-B, Gl-A, Gl-B, T-A and
T-B were used to measure the Independent variable of air flow rate. Air
pressure, and temperature at the rotometers were necessary to correct air
flow rate for deviations from the rotometer calibration temperature and
pressure (see Appendix A). Next, there were gate valves to precisely
control the air flow rate through the rotometers (GV-A and GV-B).
Finally pressure gauges G2-A and G2-B were used to measure the actual sir
delivery -pressure to the well. After these gauges, flexible rubber
compressor hoses were used to connect the air delivery system to the
equipment In the well by way of quick-connect fittings.
One independent variable was the footplece configuration for the air
lift pimp and the air sparger. Early research (5) had shown that the
footplece configuration of the air lift pump could be varied considerably
without altering the pumping efficiency. The experimental design
34
-------
K f
v '.'- -.
I-:, -%\
i Ek-^..--^-!...!
described here was derived to test whether altering the footplece of the
air lift pump could change VOC removal efficiency. Varying bubble size
In diffused aeration affects the organic chemical removal efficiency C28)
and^ bubble size Is Important In rrodel Ing aeration treatment systems.
In fact, the need to know Cor at least predict) bubble size made It
.difficult to model an In-well aeration system. The bubbles could not be
measured, and the flow of the water within the well was unknown. ;
Bubbles may have coalesced,, been pulled Into the pump or even heve left
the borehole through fractures before reaching the puop. Because of
this, the data gathered during In-well aeration testing Is site-specific,
and Is not useful to predict specific behavior In another system.
The two air lift pump footpleces chosen for this study provided two
different bubble sizes. It was hypothesized that smaller bubbles would
provide a greater, air-to-water surface Interface area which could,
therefore, Increase organic removal efficiency. It was known that the
device to .reduce bubble size would also Increase head loss and be more
expensive to operate. One design Introduced air Into the footplece
through a simple open pipe configuration Csee Figure 11). This open
pipe footplece was to produce large bubbles as compared to the second
footplece, which Introduced air through a diffusing device. The device
used was a Pearl comb air dlffuser, manufactured by FMC Corporation for
use In aeration basins. It resembled .an aquarium stone, producing very
small bubbles. The air sparger footplece was also tested. The first
sparger configuration was a simple, open PVC pipe to produce., large
bubbles Csee Figure 11). The second configuration tested had a
Pearlconb aerator attached at the base of the sparger.
The next Independent variables were the depths at which the air lift
pump and sparger were set In.the well. Three depths were chosen, based
on the VOC and water zones of entry determined during the well
characterization procedures Cdlscussed In the previous section).
The first depth chosen was 130 feet. All depths were measured from the
90ฐ bend In the eductor pipe at the wel'. head, to the top of the air lift
pump, electric purp or sparger footplece. The 130 foot depth was
selected because, based on the historic purplng water level at well L-8,
It was likely to provide 65% submergence for the air lift purp, which was
reported In the 1iterature C5, 29) as the most efficient submergence for
operating an air lift.-pump. This depth was also slightly belav the most
heavily contaminated zone of water entry Into the well. The second depth
chosen was 200 feet. This setting was below all of the major VOC and
water zones of entry. The third depth setting was 280 feet, which was 10
feet above the bottom of the well. The 280 ft depth was chosen to
maximize aeration contact time in the eductor when air lift pumping, and
to maximize countercurrent stripping by the sparger In the well borehole.
The 200 ft depth would do the same to a lesser extent, and because the
cost to compress air increased with depth, the 200 ft setting would be
less expensive than 280 ft. Operating the air lift purp at depths
greater than 130 ft would Increase costs due to submergence greater than
the optimum of 65%. Testing would show whether the Increased costs would
pay off In Increased VOC removal.
A 12-member matrix of equipment configurations was devised for the
three chosen depth settings Csee Figure 12). The entire matrix was
examined for the air lift pump combinations. After examining air lift
35
I $
u
fi!
1 'YJ
-------
; AIR "LIFT PUMP
.-:.'" OPEN PIPE
(LAR6E BU65LE)
AIR LIFT PUMP
. PEARLCCM5
(6MALL6UBBLE)
SPARGER
OPฃM PIPE
PEARlCOMB
(6MALL 5UBBLE)
a
:l
H
f
Figure 11. Diagram of air lift pump and sparger configurations
for footpiece testing.
" ''' =;. 36 ' -'' ". ..- ' . '.:.'.. .'::
-------
ip^^^.'-^^y'y^^''^'^
*, t. .;..-. . . /. v.- -
'.;;' \ - *.:-./--- . -'.v ;
f.- .-,--..> - ฃ,.;-. -.. -,v ../
f-i ' - I'-- *-. ' -..-'.:.: -. -.'.
'i '". - ' ... '''.. '
~ . ".'-.
* - ' . ..."'
1
^PU\
<
^.&?Ai&;^-wvvg
'-.. ..'/.-:T.- -:-' ''--.
.'> .; '.'-,'/ ;- -/ ' " V-~ "
'.'."". ."' f" ;-. : ;;-';
> ...
f*f>
5
-150
ft-
-2OO
ft
i
iฃ^.'*.* *&%&%%
' *-' , *"'*' '" '.'-'- '' .'. :-." '-'
'- *'; .~'\ vv-T/;''V/-*
o-'l
ง-. &
.
'ARfaER
1
o
o
Q
-150
ft " ;
-130
ft
-200
ft
m^m^ ^mm^^f&^^m^^mgm
' l"r .;*';- --v-*'. ""' -P-
' " ' * '."?? , ' '
o"
0ฐ
' <
)
> J
J
>
-130
ft
-200
ft-.
-200
ft
- *' "' - ',"" .s"''^.-1 .;'-<;'---*' -*;" - '- f^'v
-; -i -ij^ i : ^ ,\~'" "T. -- * --" - ^ :"*^ r : _,, "wv/' -- " '; -- 1' .*.-i-;
.,-' ".;' '-'... "-"X-^"17' -"*.*: - -:'-::-^
ฐo
o
<
o
o
o .
4
1
5
i;?^?^-?--':-,!
-zoo \3
ft ...fl
1 ."<
. \\
- 11
-2&0 |'4
ft
<
>
-2&O
r*
4
o
o
o
-ISO
ft
-260
<
>
*ฐ
-zoo
ft
-z&o
o
0ฐ<
>
-260
r i
,-tt ' H . Vt ' vt
-J
1
Figure 12. Matrix of depth settings for air lift pump and sparger
during in-well aeration testing at well L-8.
37
H
i-'"-' X1 -
-------
'f:j -. . punp results the electric pimp testing matrix was extracted as a subset
':;..' of the air lift punp matrix. "
C ; . . ' Another Independent variable was the pumping water level. Under set
; . conditions, the water level while purplng will vary tremendously during
1 ; the course of the year. This Is caused by seasonal rainfall conditions
|. '.. -.'; giving the aquifer differing static water levels (the water level In the
!.-.. well when not purplng). On a small scale, when the pump Is first turned
;,-,." on, the water . levs! will drop from the static level to the pumping level.
! . . . - At well L-8 this takes at least 90 hours (see results of pumping test at
.;,-.".: " L-8, Section 5). Because of resources the aeration tests were not rxai
\,'\ this long In each configuration, so the pumping water level was changing
I with drawdown during the testing.
i ' In earlier tests the water level was measured by dropping an
s-"- electrical conductance probe down a 1/2 In plastic pipe In the well.
|.'. When the probe encountered water, a circuit war. completed and a
ฃ-.. .. . deflection was observed on a conductivity scale at the surface. The
:"- water level was measured directly from the length of line In the well at
'e the point where the deflection occurred. This method was found to be
1: faulty when air was being Introduced Into the well through the sparger.
'.' In certain configurations (sparger adjacent to or below the probe line
r pipe), the plastic probe line Itself acted like an air lift purp eductor
' and water was air lifted up through the probe line pipe. To correct this
. error. In later tests, a 3/8 In plastic line was dropped down the well to
; a known depth (275 ft). A small air compressor was used to Introduce
Just enough air Into this plastic line to displace the water In the line.
This air was carefully controlled so as to be negligible In comparison
v/lth the amount of air flowing during aeration testing. The water level
was then calculated from the amount of pressure required to Just displace
the water In the plastic line (see Appendix A). -
The Independent variables thus were all direct measurements made In
the field. An example of the data collection log may be seen In Figure
13.
'. The first dependent variable was water flow rate. This was measured
by passing the well discharge from the eductor through a v-notch weir
box. The box Is shown In Figure l*t. It was made of stainless steel and
: was located outside of the well house. The v-notch was 90 degrees. An
: observation well was attached to the side of the weir box. The well was
; an 8 In pipe set up vertically with a metal plate welded to the bottom.
A smaller pipe allowed water to flow from the box to the well. In this
way, the water level In the observation well was the same as the level at
; the v-notch. A water level recorder was set up on the observation well
to provide a continuous record of water level at the v-notch weir. A
transparent Tygon tube was attached to the observation well with an
: adjacent measuring tape to obtain direct readings of water level at the
v-notch. Water flow rate calculations from a 90ฐ v-notch are given In
Appendix A.
The weir box served a second purpose, aside from water flow rate
measurement . It was also the gas and water separator during aeration
experiments. The water discharged was milky white with bubbles during
experimentation when the air sparger was being used. The open weir box
: provided a place for the bubbles, as well as the VOCs to be released to
; .. the atmosphere. The water was clear of bubbles before It reached the
; 38
-------
1 ""'"'' \ji \ - " ' "
Cir'-'-" '"''' ' "''""' '""
.'...' Time
:-i
i
i
:::. . ' 8^
.. . /:Z5
' . /'30
PM.
J;40
1:56
PM.
io fiM-
PM
' 2;S5
' PM
. 2i50
PM
:r . . .-..- 3-^0
' . ,-:'' RM.
- 3^7
'-.. . .. PM.
Av .' Date'. . io- 18-64 Ai .' . .
FVtos.
at.
eoto-
mefer
23
75
75
75
75
75
75
75
76
75
Ton?.
for
Id
&9'
9/ซ
94"
94'
94ฐ
95"
95ฐ
96'
96ฐ
fob-
a
35
36
36
35
35
35
35
35
35
down
well
35
31
34
34
34
34
34
M
34
ftoks-
Zber
75
75
75
75
75
75
75
75
75
Temp.
Air
Ib
G3ฐ
90'
9aฐ
92ฐ
93"
93ฐ
Q4*
O/3^
94'
TEDT * 23
Rote-
fa
35
35
35
35
35
35
35
2,5
35
Prett>.
down
wel \
5b
34
33
33
33
33.
33
33
33
33
&
.^^
f^X
ป^r
/\f^/^
^^
^X
r^r
'^
. -: WeafHer f cloodyCool-
HZ0
welt
%tl
^^^
$/'
^x
&x
<&/
<&/
%$
^x
T^X^
f.tter
7/5
//5
//5
/tf
;/5
H5
115
115
115
Temp.
beVorc
85^
/06-
/06a
X06'
/or
/07'
/ofl-
.W
j/^rtc
I\J\
filter
//r
//4
. //d
//r
//4
IM.
//4~*
://4
//4^
5TAT/C.
^a-S'if
STAT/C
23.64
AT
\ '
rฐewซ
)
r
(AIR
\TdSl"
\ '
Ui/? re?
1 ': '., . :. \ ''..-; Figure 13 '. Field data collection sheet for in-well, aeration testing at well L-8.
.1 ' - ' -''. ':''.' .
' . . '' '* "' * ..^'
-- ...".'.' ..' 'A
!"( !.
' "'v ':'','. '.'?'. -.;;>'
... .:' :;'; ';/>;:V"
. ..., " i' .'.. "'.'_** '
' ; '."' .- ' -.' -''.'"''
/;.-:l':'';'.'.-'\-''.^'-^''?:v'
..'/;*'*"' '.:/>. :.'i?f.-r'.
?,';;> f,.v'";'. , V'^^V
'','' "'.'':'"''' '"'''j''.v''0"''-:
''J:":;'4^:!".'-;^;^;^V
..:'>' /?-"-f'-::v!;:L' ''
:iS
vt:
s-
^;i
1
i
.ซฅ
^
1
*v.'
J-'<
;Sป
;'ซ
1
V
1
' ,i
1
y^\
;> '
S
^
-'
&
%
'I
I
V
?'!
-------
P!T
1
?l
i
END OF,.'..< vV ,
'EDOCTOR '''^.'-..
AIM '' '.;''v'':
WEIR 6OX
Figure 14. V-notch weir box used as air-water separator and to measureiwater flow
rate. Treated samples were collected at the v-notch, water spilled
into the lower box, then went to waste. , .'" . . .'. <
t>';
I .T
"'^
'&
*i
.'^
>;
V-
'i
-------
P-w^pi usnyvG
!L.^SSilw
v-notch (VOC sample collection point).
.... The dependent variables of submergence, horsepower, efficiency and
''cost'were calculated from the Independent variables. See Appendix A. .
.". Primary and secondary water quality parameters were determined by
: samplIng.and analyses.:- During air lift pump testing the raw water sample
was collected from a small Cl HP), submersible pump which was located at
the .entrance to the air lift pump footplece. When the air sparger was
not being used, rew water could be collected from this point at any time.
When the air sparger was being used, the raw water sample had to be taken
before the sparger was turned on, or the sample col 1 ected at this point
would be aerated and not representative of the raw water. Each time the
sparger air flow, rate was changed, the sparger was first turned off to
allow the air to clear from the well. The well wss declared free of air
from the sparger when the raw water dissolved oxygen concentration
returned to background conditions defined at the start of testing, before
any air had been Introduced to the sparger. It typically took 45 minutes
to one hour to clear the well between sparger air flow rate changes.
. The primary water quality parameters examined were the six volatile
organic compounds found In well L-8 In high enough concentrations to be
statistically compared In raw and aerated samples. The samples were
collected and analyzed according to EPA method 502.1 (30), which Is a
purge and trap gas chromatographic procedure using the halogen
specificity of the Hall Electrolytic Conductivity Detector (HECD). The
gas chrcmatograph used was a Varlan 4600 with a Vista 401 data system.
The purge and trap apparatus was a Teknar LSC-2 with the Tekmar ALS
autosampler. The quality control program was developed based on Method
502.1 and the literature (31). Gas chromatographic conditions and
quality control procedures are given In Appendix D.
The secondary water quality parameters were pH, dissolved oxygen
(DO) and bacteriological quality. The pH and DO measurements were
made In the field, Itrmediately after the sample was collected. The pH
was measured using a Corning portable pH meter, Model 4. Dissolved
oxygen was measured with an Orion DO probe, model 97-08-00. The
bacteriological samples were analyzed by both the heterotrophlc plate
count and the R2A agar plate count methods (32).
The treated water quality, both for primary and secondary
parameters, showed water quality after In-well aeration treatment under
the conditions represented by the Independent variables. The treated
water samples were collected from the v-notch of the weir box. The
analytical methodology was the same as discussed above for the raw water
samples.
The final dependent variable was air quality. Air samples were
taken In the well house, over the weir box and 20 ft from the weir box.
Two methods were used for examining the air. The first Kas to pull a
known amount of air through a tenax/sllica gel/charcoal trap using an
Industrial hygiene sampling pump. The trap was capped and returned to
the laboratory where It was desorbed by the purge and trap device Into
the gas chromatograph for analysis. Two traps were used In series to
monitor for sample breakthrough from the first trap. The second method
used to monitor the air was by the on-slte analysis using an HNU portable-.
photolonlzatlon detector. Both of these methods are described
In more detail In Section 12.
41
-------
/\
.";.'.'.' SECTION 7;
PRELIMINARY FOOTPIECE INVESTIGATIONS
AIR LIFT PIMP FOOTPIECE TESTS - PROCEDURES
-. Footplece configurations for both the air lift pump and the In-well
diffused aerator (sparger) were Independent variables as discussed In
Section 6. The preliminary Investigations tested two types of footpleces
for the air lift pump and two types of footpleces for the sparger. These
experiments were also used to develop operating techniques and gain
experience In using the air lift punp, both with and without a sparger.
Although air lift pumping had been used at North Perm Water Authority In
the past, personnel who were familiar with air lift pump operation had
retired. .
The tests run on the air lift pump, without the sparger, were also
used to develop air lift pumping efficiency curves. The air lift pump
and sparger were both set at a 130 ft depth In the well. All equipment
depths were measured from the top of the air lift pump or sparger
footplece to the center of the 90ฐ bend of the eductor (water discharge
pipe) at the v/el 1 head.
The a!r lift punp footplece used for preliminary testing was a
combination of the two footpleces to be tested. This design v/as used so
that the air lift pump would not have to be removed from the well to
change footpleces. Installing and removing an air lift pump was
comparable to performing similar operations for an electric submersible
pump In terms of labor and time required. Figure 15 Illustrates the air
lift punp design used for the preliminary footplece testing. The air
lift purp footplece was made of 4 In diameter steel pipe. The steel pipe
was tapped to accept 3/4 in fittings for the electric submersible sample
pump, the air feed line ending with an air diffuser, and the air feed
line which ended as a simple open pipe. The air lift footplece was
attached to the 2-1/2 In steel eductor pipe with reducing fittings. The
air lift punp was then lowered Into the well by attaching 10 ft lengths
of 2-1/2 In steel pipe for the eductor and 10 ft lengths of 3/4 in PVC
pipe for the sample pump discharge, diffuser air line (small bubble
footplece) and the open pipe air line (large bubble footplece).
A 1/2 In PVC pipe was lowered Into the well to allow a water level
measuring probe to be used. This pipe prevented the water level probe
wire from becoming entangled with the rest of the equipment In the well.
*!.'.' Each of the PVC pipes were color coded and labelled at the well head
|l j so they could be distinguished. A color code diagram was drawn and kept
|> ; at the wel1 head for reference. The color coding Is shown In Figure 15-
i The footpleces were chosen to provide different bubble sizes for
I ' . . 42
K I ' .
-------
VฃSF w**ซ.W '?i^^^ f^^.^-^s^^'^s^sS^SSB&^^S
..*..._ ซrt. *.r*rri*m<^^^^ฅซ20SR^^^ ^ J ^ _^ -,.._. ..,-^.j- .-. ---'.":",..' .' "ป-'*'' : ;' ^ ' |-
'-.- ri . '*.*, ' '.''': ' ' , " ~"- '' . ฃ- * *'*-* * ." _' ' -^-- .*i-'" ซ ";'.*ป -' ' *" ~ 'i il^--/"*'-," V ' ~- ~ *ป'-*"' s' . - * ป i
; ; f "^ป * ซTr \* . - - ' ''( >"ป-.; I--*- - -ซ -- , ./ - '- 4. . ' . '*.'. -".'' -J < , ; : .'-".' -.*; .' * I ' -- \
"-'.'- ;^,"D/'.' w ; "' --''" '"- "": "z.''- -UJ V- - '. ..-:': '' v^.vK^' ?'--'<"''. .:7-.-|
i... ud uj .-.....- - < ^ -. 5 :-'- ';v-.-''- .,vv, .,-;:;>: 1
: -,2
"p
"fL
' _^
'Cf
'. CL
tL
, UJ
(D
O
lil
^
tY
lU
(-
2
^
^
'8f
_ x
a
u
2
4
tf
s
"2
a.
i
^
rt>
r
i
5
_-^
ill
C
a
o
EOUCT
X
Csl
O
iu'
Q.
^v
' .O
H
_J
CL
/ UJ\
OtD
CitO
-JtQ
^ o
UJ
CL
. . IN
ti.
J
111
UJ
in
* y
!'*''
. m-*". *.
-i*;-
V-'-*
U
._
"jtฃ
'-:. '..; ':''.'. ;.'" |
'- ''- ' ' -.":'-'. :.""'' ''"""1^i;./. i
^ - ' - '. ' '. . - ' " '.." .
-. ' '-
' ' :' - '-.' - '' *'""-; :
'" -''- ' ". >''"'" '"'. ' ; ".'.'. ;-
"'".''-- : - ^ ' .--' -T'- *
"* ' ' ' .
:';' W " ".''';.''. \
' i
' . i
i
r
II
. . .'.' - . 1
' - : ' ': -' / 2
'"' -- y:^'.l
. . . .
. ' ' ' . -1
. - 1
gure 15. Diagram of in-well aeration equipment used . . |
during air lift pump and sparger footpiece . . 1
testing. j
43
-------
\
Vi.
/~"':-x-
aeration. The-tests were to determine whether changing bubble'size would
: significantly affect VOC removal or the cost to compress air because of
Increased hand loss provided by the diffused aerator. Based on the
literature, changes In air lift pumping efficiency (amount of water
pumped for amount of air added) were not expected with changing air lift
.. punp.footpleces.
'.'=_.-' . 'It Is important to make a semantic distinction between a ''test" and
a "run". Four tests were performed during preliminary footplece studies.
A.test was a series of runs, with one depth configuration of In-well
" aeration equipment per test. For each run, air flow rates were altered. -
For example, for a particular test the air lift pump and sparger were set
at 130 ft In the well. These depth settings were rปt changed during the
test. For each run during the test, a different amount of air was
delivered to the air lift punp and/or sparger. Also, during certain of
these preliminary tests the footplece configurations were different for
different runs within a test.
Preliminary Test ftl will be described In detail, as It Is
descriptive of the procedure used In other tests. The air lift pump was
operated without the sparger. Air was first delivered to the large
bubble air lift pump footplece. The range of air delivery rotometer
settings were from 5 to 116 cfpm. The same air flow rates were then used
with the small bubble footplece.
The portable air compressor was started and the output xvas connected
to the air delivery system (described In Section 6, see Figure 10). The
first air flow rate used was 6 cfpm. The rotometer was adjusted to 6
cfpm and the air lift pump was allowed to run for 5 minutes. After 5
minutes, a series of readings were taken of the water level, water flow
rate, and the various gauges and thermometers. If the rotometer setting
was drifting, It was adjusted at the 5 minute reading. The air lift pump
then ran for 10 more minutes to give a total run time of 15 minutes. At
the end of the 15 minute run the readings were taken again and VGA
samples were collected. This 15 minute run procedure was repeated for
26,40,58,78,96 and 116 cfpm rotometer settings. The entire test
consisted of 14 runs; seven with the large bubble air lift punp footplece
and seven with the small bubble air lift pump footplece. The basic
procedure of using 15 minute run times with readings at 5 and 15 minutes
and sampling at 15 minutes was used throughout the rest of the In-well
aeration testing at well L-8.
During the first air lift punp footplece test It was discovered that
the water level at the v-notch of the weir box was difficult to determine
directly. Also, the original rotometers for measuring air delivery were
difficult to adjust at low flow rates. Because of these difficulties,
the air lift punp efficiency data for the first test were questionable.
The weir box was modified to include a direct reading tube and water
lievel recorder (see Section 6 for details). A new rotometer, calibrated
from 0 to 6 cfpm, was Installed for use for the very low air delivery
rates. Preliminary Test %l was then repeated, using these modlfIcatlons.
The modifications were adopted for use during the rest of the In-well
aeration tests at well L-8.
After examining the air lift pump operating data, It was decided
that the cost to operate the air lift pump could be reduced by enlarging
the size of the air delivery line from 3/4 In to 1 In. A third air lift
n
$
I
I
44
./U.
-------
pump'preliminary test was run using the new, larger air delivery line.
The large bubble footplece was used for this test. ., -.: . f
AIR LIFT PUMP FOOTPIECE TESTS - RESULTS AM> DISCUSSION . :-X;, ^
T In terms of pump operating parameters, the air lift pump footplece
tests confirmed what had been predicted frcm the literature. Figure 16
Illustrates changes In air lift pump operating air pressure and cost to
compress air for the three preliminary air lift pump tests. Higher
operating pressures resulted In higher operating costs. Costs shown
here were only the costs to compress air. They do not Include costs of
disinfection, repumplng after a clear well, operation and maintenance,
etc. (see Appendix A). As expected, the operating pressure of the air
lift pump equipped with the restricted opening, small bubble footplece
had the highest operating pressure and therefore the highest operating -
cost. The maximum air pressure for this configuration was 83 psl with
a maximum cost of 29.5$ per thousand gallons. The maximum operating
pressure for the large bubble air lift pump footplece with a 3/4 In air
line was 68 psl, which produced a cost of 25.7$ per thousand gallons.
The large bubble air lift pump footplece, with the 1 !n air line, had a
maximum operating pressure of 54 psi with a cost of 20.4$ per thousand
gallons to compress air. Pressure differences were most apparent at the
higher air del Ivery rates where friction loss was greater. Since the
most efficient operation of the air lift pump was at a lower air flow
rate where operating pressure differences were smaller, the choice of
footplece configuration was not found to be critical In terms of cost
when designing an air lift pump.
Figure 17 Illustrates the results of the preliminary air lift pump
tests in terms of pumping efficiency (wire-to-water efficiency). The
shape of the pumping efficiency curve followed that predicted frcm the
literature (5,29). The curve sharply rose to a maximum pumping
efficiency of 30-35%, then gradually fell as more air was added. The
maximum efficiency coincided with an air-to-water ratio range of 1.1:1
through 1.4:1. The shape of the curve was because of slippage In the
eductor at higher air-to-water ratios. (At higher than optimum
air-to-water ratios, the excess air moves faster up the eductor than the
air/water mixture Itself. This phenomenon Is defined as slippage.) The
depth setting of 130 ft for these preliminary tests was designed to
provide 65% submergence for the air lift pump, which was the theoretical
optimum submergence In terms of air lift pumping efficiency (5,29).
Because of this, the pumping efficiency maximum shown In Figure 17 was
the best efficiency which could be expected for the air lift pump used In
these studies. The 30-35% efficiency was consistent with observed air
lift pump efficiencies In the literature.
The data frcm the air lift pump footplece test also confirmed the
literature wherein the two footplece configurations examined showed very
little difference In air lift pump efficiency (see Figure 17). Changing
from the 3/4 in to 1 In air delivery line did cause a slight Improvenent
In air lift pump efficiency.
The VOC removal results for the air lift pump footplece tests are
shown In Table 8 and Figures 18 and 19. The figures represent data from
both repeats of the air lift pump footplece tests using the 3/4 In air
45
-------
t -
/ r-
"."^-'. , ^'-'*.- *-' ,^'i-.- VA'; >''i i^^^
"'v;-'Uv;\-:.^^ซ
H
I
.1
e
y
i
I
M
I
0
0
I
y
o
BO
60-
70-
60-
6O-
4O-
30
28-
26-
24-
22-
20-
18-
16-
14-
12-
10-
8-
6-
4-
2-
O
OPERATING AIR PRESSURE (pal)
In LAROE BUBBLE
COST TO COMPRESS AIR (CENTS/THOUSAND OAL)
3/4 In SMALL BUBBLE,!
LAROE BUBBLE
1 In LAROE BUBBLE
I T I l I I \ I
2O 40 60 BO
100 120
il
V" *
AIR DELIVERY TO PUMP - cfpm
fi
Figure 16. Operating air pressure and cost to compress air
for air lift pump footpiece tests during
preliminary investigations.
46
7a5T^5
-------
|'S^ta^SS^5^S^3^!^^B2B.^5iS5B
r ;;/-; ;;.:. '-...- '.;ฃ,**. T
(^h^^f^:f}^---^
(/- ^ ..'- ,'-; .;' . ;_.^- ..''--.*'
N;:^V"^<:V::^':;Y;30"-
^;-;S::-^Vk:;:;'f-'
r.-.-;:-;-,iHr;|- ;:.
'" "' -' ->"'' "'" 25-
.; .:. '- " -.-;.--> :**..
V../-V /'-.-- .0 .
;-y:-v ซ:. z
-:- . ,- : . - ..::u ..
0 9n
t ' ' L Z0"
i .. --.- -.- S1-- .
.- ,'/ -..--.It ' .
!' ';:---;.rul - '
K
; ; ,- | 15-
2 ,o-
U
K
".-', ' *
': - - . s-
! '
V " - t : 3/4 in SMALL BUBBLE v -
.'-'- '- " ',- ..'"" ' - '- ...- .''. '':. .. '. '.-:";:':;..
D ft ; :-VD/ 3/4 in URGE BUBBLE
.... If . .- , . . ;. ;. .-,. :- ; ''..;.:'- ;-.-'; ' ' '''-' '
A :^;i in LARGE BUBBLE : :;
- " " " "* '" .. ' ''" '''. ---.-' , - - = ' -i, -.'-' ,
' ' . - -. ' '' ' .-,.>'.,'' -- : ,-."'"""..-.- .,-' , '- ." _ ."". " _".,,
n- " :.: ;-: ';\v;^"^y.v^7:v^''-'v:-=;-'-."-;:^^',":^
'" ^/\':'W'<'-?^'!'^^-:.$M
. D ':'.. .\ ";. .-.. -,.:-. . ';': - - .. -.: -.;-
- -...-.- :. '.:.- , ; ; . '..'- ,:, -:;...-
f . ' ' - -- - ' . '. ".;' : - :" '. '- '.'. '' " "'.-".
. .._. .:'.'. -X' --.' . ";....' - . "-.- -.".--.- .: ~. : -'
-..'. V '-.:-
. .. ... D^ ' - - - - . -..' ... _
" .- 5A " . - " . ''-" .
' :".-'ซ ''-, : :-:;.: '..
B A -
v .
..'.' :'.' D ป. . ;; :;..;
" ' n ' .-' -
' ''. / " ' ,."" ' . -.-'. * '
9'
. . . . .- * " . . .
, . '-- -. .'' .-..'' n..- :. -A
:' ." "-'' -: .-.'/- - ':''.,; *
'. -~ ' '
- j i i r i T i i i ^ n i
) 20 40 60 80 100 12
H
AIR DELIVERY TO PUMP - cfpm
Figure 17. Air lift pump wire to water efficiency for
footpiece testing. ,
JV-'X
47
-------
If
I-
TABLE 8. VOC REMOVAL DATA FOR AIR LIFT PUMP FOOTPIECE TESTING (AVERAGE OF ALL
' LARGE AND SMALL BUBBLE AIR LIFT PUMP FOOTPIFCE TESTS)
AIR-TO-WATER
RATIO
AVE %RSD N
1.2 9.3 5
2.4 6.3 4
3.5 5.1 5
4.9 0.0 2
7.0 1.4 2
8.7 1.1 2
11.1 5.1 5
% REMOVALS
TCE
AVE %RSD
33.1 19
40.4 3.7
48.5 5.8
52.6 4.7
58.2 6.9
65.1 0.65
70.8 3.3
FOR VOCs
PCE
AVE %RSD
40.4 14
52.2 4.1
02.0 4.7
66.2 1.6
71.4 5.7
c-l,2-DCE
AVE %RSD
26.5 33 -
26.7 2.1
35.5 7.3
37.2 16
43.3 9.5
1,1,1-TCA
AVE %RSD
48.3 19
56.7 2.1
1,1-DCE
VC
76.7 0.74 48.7 . 0.60
81.1 2.8 55.2 7.5
AVE %RSD AVE %RSD
57.6 11 54.8 22 ',
67.7 2.1 72.1 0.81
64.9 9.9 74.7 5.2 76.6 4.6 .
73.8 1.4 . 80.8 0.5 82.6 3.3
78.9 5.9 85.5 0.1 .86.8 2.0
82.0 1.0 87.0 2.4 89.3 0.80
85.3 2.4 89.9 1.2 93.6 3.8
-------
CO
ao
40
30
ao
10
100
BO
o
7O
00
BO
4C
30
ao
10
o
3/4 in LARGE BUBBLE FOOTPIECE a
3/4 in LARGE BUBBLE FOOTPIECE (DUP) *
. 1 in LARGE BUBBLE FOOTPIECE *
3/4 in SMALL BUBBLE FOOTPIECE ^
3/4 in SMALL BUBBLE FOOTPIECE (DUP) x
O1.2 DOE REMOVAL.
a 7 .
AIR TO WATER RATIO
11
1
R
.8
Ii
'?;H
$
Pi
!l
Figure 18. Removal of major contaminants from well L-8.
during air lift pump footpiece tests.
49
-------
100
ro-
ao-
to -
70-
M
ao-
ao-
10-
ao -
mo -
70 -
o -
eo -
ao -
ao -
10 -
1.1.1 TOA REXOVAU
9
1.1 DOE REMOVAL.
VINYL CHLORIDE REMOVAL.
s r
AIR TO WATER RATIO
11
Figure 19. Removal of minor contaminants from veil L-8
during air lift pump footpiece tests. For legend,
see Figure 18. . -
50
-------
,_-^_^_^ ^.^
'-'>' - '.. .,' ,' ' '''..'..-'.- ""'".I"-"-}
line as well as data from the test using the 1 In air line. Visual . '''"'ฃ''''' 1
examination of the results Indicated that none of the configurations '' j
tested were"significantly better or worse In terms of volatile organic V'-l
chemical removal efficiencies with a ฑ10% experimental error. It also 1
demonstrated good reproduc 1 b111ty where duplicate tests wore conducted .'.\j
and lent confidence to the procedures. , .; -' \?
The .air-to-water ratios plotted on the x-axes of these figures are . }
averages of'ratios obtained during the tests for:the same air rotometer . . ':;..-:]
setting. A problem with air lift punp field work was that the same .j
.rotometer setting yielded different water flow rates at different times. . . 4
This could be caused by a number of factors, but the principal cause was ]
the air lift punp submergence changed constantly as the water level was :.
drawn down during pumping. Since well water level could not be . '''}
controlled, It was decided to keep the rotometer settings constant for '',
each run from test to test. ...*-.- .J
The air-to-water ratio column of Table 8 shows tne deviations which -i
occurred for rotometer settings during the preliminary air lift punp ->i
tests. The percent relative standard deviation was the highest for the . .j
lowest air-to-water ratio. At these very low air deliveries the <
air/water mixture surged from the eductor at 5 to 10 second Intervals ]
causing an uneven water flow. The uneven flow made It difficult to '\
accurately measure water flow rate, which would account for the greater -;"1
variability In air-to-water ratio at low air delivery rates. These low J
air delivery rates produced air-to-water ratios below 2 to 3 which Is the j
theoretical optimum for air lift punp operation.
If the air-to-water ratio was the only factor affecting organic ;
removal efficiencies Cand not air lift punp footplece configurations),
the percent relative standard deviations for the removal efficiencies ;
should be similar to those for the air-to-water ratios. In most cases In . :
Table 8, this seems to be true, but for several ratios the data were '.
derived from only two data points, which diminished confidence In these
conclusions. The data for the lowest air-to-water ratio was consistently
higher In percent relative standard deviation for the VOC removals than
for the air-to-water r.tlo, I.e. the VOC removal varied more than the
air-to-water ratio at a particular rotometer setting. For this data,
air-to-water ratio was probably not the only factor affecting organic .
removal. Because of the surging water flow at the lowest air-to-water ' - '.
ratio, VOC removals may have been affected by a different, Inconsistent
type of air/water Interface In the eductor and the weir box.
Both the volatile organic chemical removal data and the air lift ;
punp efficiency data Indicated that there was no reason to attempt to
create smaller bubbles with the air lift punp footplece. It was not :
possible to see what was happening In the well during testing, but It can
be assured that, at that oepth In the well, In the relatively small '
eductor pipe, the bubbles rapidly coalesced. Davis and Weldner C5) .
observed coalescing of small bubbles in their laboratory tests enploylng
glass eductors. For most of the distance the air traveled In the eductor .
pipe, there likely was no difference between bubbles Introduced by the
large bubble or small bubble air lift punp footplece. Because of this,
and the higher operating cost of the small bubble footplece, the large !
bubble open pipe air lift punp footplece configuration was used for the
rest of the In-well aeration tests. The 1 In air feed line remained In
' --si . - ''. ' ;
-------
"use because of Its lower operating pressure. . .-'.....,;:,.,"~ \-'-;:-.:?%'~l:'^.:..-*:-%:v>
SPARGER FOOTPIECE TESTS - PROCEDURES / -,. ;-.:;: : ''^ .- :^^- V :i V >:
two 3A In air sparge lines were lowered Into the well to a depth of '
130 ft. The air lift pump remained In the well at a depth of 130 ft. In ..
contrast to Installing the air lift pump, Installing an air lift sparger
:was a simple task. The spargers were lowered Into the well by attaching' .
10 ft lengths of 3A In PVC pipe at the surface. One sparger ended In a
dlffusslon device to produce small bubbles, while the other was left as
an open pipe to produce larger bubbles. - -
Preliminary Test ง1 will be described. It was typical of the other
tests performed. The air lift pump was operated at a rotorneter setting -:
of 25 cfprn. This air delivery rate was found to be the most efficient In
the a*r.lift punp footplece tests, producing the highest water flow rate
per unit of air Introduced Into the air lift pump. ; v . -
The first run of each of the sparger footplece tests was done with
the air lift pump alone; no air was being added by the sparger. Next/ I
several runs were performed with Increasing amounts of air added through , i
the sparger. Finally the air lift punp and sparger were both turned up I
to their maximum air delivery rates for one run. This final step was I
done In order to try to get the highest possible air-to-water ratio J .
without regard to air lift pumping efficiency. : . |
During air sparging It was not possible to collect a raw water . .3
sample, because the sample pump would have collected water which was 1
aerated by the sparger (see Figure 15 for sample purp location). In |
order to collect raw water samples between runs, the sparger had to be ]
shut down, then the well was allowed to clear of bubbles while the air ซ
lift puip was still running. The effluent of the sample purp was watched J
until It was clear to the eye, then the raw water was sampled and the ;
next 15 minute sparging run was started. A dissolved oxygen probe was :
used to verify the clearing of the well Csee Section 6) The clearing ]
process took approximately 45 minutes. During the 15 minute runs, ]
readings of the various gauges and thermometers, as well as water flow ":
rates and well water levels, were taken at 5 and 15 minutes Into the run. '
As with the air lift purp footplece tests, any necessary adjustments to . ;
air delivery rates were made at the 5 minute reading. Treated VOA ;
samples were collected at the v-notch weir at 15 minutes. ''
SPARGER FOOTPIECE TESTS - RESULTS AM) DISCUSSION *
'' " *
Sparging air Into the well borehole had an effect on air lift
punplng which can be seen In Figure 20. As air was added through the ::
sparger, the amount of water punped gradually declined. This could be :
easily explained by re-examining the operating principle of the air lift '
purp. Introducing air Into the eductor of the air lift purp created a i
density difference between the water Inside and outside the eductor. !
This density difference was the driving force which lifted the water up \
and out of the well. When the sparger was In use, air was being added ;
both Inside and outside the eductor. The density differential was
reduced, which reduced the driving force of the air lift purp. In Figure i
20, the air lift purp was operated at a constant air delivery rate of 25
j
i
52 '':
{' v'
-------
Ill
z
2
U
a.
Q
U
Q.
2
D
Q.
U
I
100
80-
80-
70-
60-
50
GPU LOSS WHH INCREASING AIR ID SPARGER
i i i
8
i i r
10
12
AIR TO WATER RATIO
D LARGE BUBBLE + SMALL BUBBLE
:ง
fel
a
^
%
1
' Figure 20. The amount of water pumped by the air lift pump
decreased as air was added through the sparger.
-------
'
cfpm and the amount of water pumped dropped (from.78 to 68 gpm) as the "
'air lift pump driving force was reduced with addition of air through the
sparger. '.. ; : . - -.. ' . .. .: .. v . :\ . .; ;-'>-. ' :
.. :The effect of sparging on air lift pump efficiency could not be
determined during the early tests because of inaccurate water level
measurements. Water level wss needed to calculate the pump horsepower
(see Appendix A). When the sparger was operating, the plastic pipe used
for the water level probe was filled with aerated water which operated
like an air lift pump Itself, causing artificially high water level
readings. This was not discovered until later testing when water was
actually pumped out of the well through the water level probe pipe. The "."; *.;|
problem was corrected In later tests and the air pressure method of ^
obtaining water levels was used. A description of this procedure can be
found In Section 6.
'; Figure 21 shows the cost to compress air for the sparger footplece
testing. The costs for the air lift pump In combination with the large
bubble or small bubble spargers were nearly Identical, ranging from kt to
13$ per thousand gallons at air delivery rates of 28 to 75 cfpm. The
operating pressures were only slightly different for the two sparger
footpleces. The small bubble sparger footplece ranged from 31* to **3 pslg
while the large bubble sparger footplece required 33.5 to *ป1 pslg
operating pressure. These small pressure differences between sparger . |':
footpleces translated Into very little difference between cost to . <"
compress air for the two types of spargers. ''.-[
The cost to compress air was similar for the air lift pump alone j
versus the air lift pump and sparger combination. The 3/k In air line (
air lift pump with a large bubble footplece cost b$ to 1H per thousand j
gallons (air delivery of 26 to 77 cfpm). This compared with *ป$ to 13* \
per thousand gallons for the air lift pump and sparger cont>lnatlon at \
similar air delivery rates (28 to 75 cfpm). Cost differences for the air
lift pump footpleces became more apparent at air deliveries above 75
cfpm. The sparger was not operated at these higher air delivery rates
because, at rates above 75 cfpm, water was blown out of the well
borehole. This occurred because the sparger was delivering enough air to
turn the entire well into an air lift pump, with the borehole acting as
an eductor.
The VOC removal results for sparger footplece testing are shown In
Figures 22 and 23. The figures compare removals for small bubble and
large bubble spargers, as well as the earlier results for small bubble
and large butble air lift pump footpieces.
Apparently, there was no significant difference In organic removal
efficiencies between the two types of spargers tested. The same
conclusion was made earlier for the air lift purp footpleces. The
figures also show not only were the small and large bubble sparger VOC
removals similar to each other, but VOC removal by the air lift
pump-sparger combination was similar to VOC removal by the air lift purp
alone. In addition, the VOC removal at the highest air-to-water ratios,
representing the runs where the air delivery to the air lift pump was
raised to its maximum, agree with removal obtained by keeping the air
lift pump at a constant air delivery rate (25 cfpm) and adding increased
alr.to the sparger. In short, the preliminary footplece tests indicated
that for a given air-to-water ratio at a given depth In the well It did
.
-i
-------
'^^^'fr^^^^-^^S^S^LSSS^^'S^^?.
0
j
u
0
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
COST FDR SPARGER AND AIR LIFT PUMP COMBINATION
D URGE BUBBLE SPARGER
SMALL BUBBlf SPARGER
\ I III I I I
0 20 40 60 80
AIR DELIVERY TO SPARGER AND PUMP- cfpm
Figure 21. Cost to compress air for air lift pump and
sparger combination during footpiece testing.
55
-,r-
i
-------
" s'.-* -'r '-
M
J- ป
fc
m
M
1
*H
Si
"ซ'_(
Rtf
ซi:ป
-
TO-
4O
SO
ao
10
CO-
8
BO -
ao
10
eo -
o -
TO -
ao -
00 -
so -
ao -
10 -
TCE
a
KEYl AIR Urr PUMP ALONK SMALL PUBTO g
AIR UPT PUMP ALONE LAROZ BUปBLg.
AIR urr a* LAROS euseuc RPAROK
AIR UPT & SMALL. BUBBLE
fl
C-1.2-DCE
PCE
4. a a
AIR TO WATER RATIO
to
ia
Figure 22. Removal of major contasinants during sparger and
air lift pump footpiece testing.
56
-------
p.'-'-v.."';--"-'--: ';"r-' .V; >'.:-. 100 -p -
^."^J^'^^'io^. 1,1,1-1
"7D--
00-
ao
10
too
4O -
. SO -
ao -
10 -
100
BO -
BO -
TO -
so -
BO -
4O -
SO -
ao -
10 -
o
-TCA
e
AIR UPT PUMP ALONE 8MAU. BUSBUS A
AIR LIFT PUMP ALONE LAROE BUBCBLE *
,1,1-DCE
AIR UPT fit LAROE BUBOUE QPAROff a
AIR UPT ซt SMALL. BUBBLE BPAROE -f
vc
4. a 10 12
AIR TO WATER RATIO
Figure 23. Removal of minor contaminants during sparger
and air lift pomp footpiece testing.
57
-------
not matter what configuration of In-well aeration equipment was used, the
VOC removals were the same. This held true for all of the compounds
except, vinyl chloride. For vinyl chloride, although the sparger
footpleces both produced the same curves, and the air lift pump
footpleces both produced the same curves, the air lift punp by Itself did
not seem to remove the vinyl chloride as well as the air lift punp and
sparger combination. Vinyl chloride has the highest Henry's Law
partition constant of all of the compounds found In well L-8, and
therefore, was the most easily removed. Perhaps .because of.this It was
more sens I Uve to In-well aeration configuration changes. - ...
.Table 9 Is a sunnmary of the air lift pump and sparger footplece
testing removal data, In terms of percent relative standard deviation
(%RSD). A similar Interpretation was presented for the air lift punp
footplece tests without the sparger In Table 8. As was mentioned In the
air lift punp footplece testing discussion, a given air lift punp
rotometer setting did not always give the same water flow rate each time.
This was because of changes such as static and pumping water level of the
well, which affected the operation of ths air lift punp. The air-to-
water ratio coluin of Table 9 showed this variation. The VOC removal
percentages showed how removal varied at these different air-to-water
ratios. The removals could be expected to va^y at least the same amount
as the air-to-water ratios, (as seen by the percent relative standard
deviation). If the In-well aeration configuration affected removal, a
much higher %RSD for VOC removals would be expected than that observed
for the air-to-water ratio. Table 9 did not show major differences
between the %RSD for air-to-water ratios and the %RSDs for removals. The
exception was In several of the vinyl chloride points, where the graph in
Figure 23 Indicated a difference In removal between the air lift punp
alone and the air lift punp with a sparger. Table 9 showed a greater
variability In air-to-water ratio and VOC removal for lov.er air delivery
rates (also seen In Table 8 for the air lift punp without a sparger).
This could be caused by variable flow patterns when less air was being
Introduced into the well borehole.
CONCLUSIONS FOR FOOTPIECE TESTING , '.
Results of the footplece testing for the air lift punp Indicated
that the small bubble footpiece caused greater operating pressurethan
the large bubble footplece, as expected. The operating pressure
difference was greatest at the highest air-to-water ratios (5:1 to 12:1).
In terms of pumping efficiency, the most efficient operation of the air
lift punp was found to be at a lower air-to-water ratio (1.5:1), where
the operating pressure difference between the punp footpieces was small.
There was very little difference In maxlmun punp efficiency between the
large and srnal 1 bubble air lift punp. The maximum efficiency was found
to be 30-35%, which confirmed the literature. .
.There was no difference in VOC removal between the two air lift punp
footpleces. At the maximun obtainable air-to-water ratio of 11.1 tb-4>
VX removals ranged from 55.2% for c-l,2-DCE (lowest Henry's Law
constant) to 93.6% for vinyl chloride (highest Henry's Law constant).
i The cost to compress air ranged from 2<;/l,000 gallons at an alr-to-
v/ater ratio of 1.2:1, to 29.5^, 25.7$, and 20.k$ per 1,000 gallons for
58
-------
!
1 r~* "-'" ";/
I K
t: --t;
N, ;^
'' '
' ^
TABLE 9. VOC REMOVAL DATA FOR SPARGER FOOTPIECE TESTING (AVERAGE OF ALL LARGE 8.SMALL BUBBLE
SPARGER FOOTPIECE TESTS)
AIR TO WATER
RATIO
AVE
1.8
2.9
3.2
3.8
5.5
7.4
8.6
12
%RSD N
39 4
1
1
5.3 4
0.0 2
1
1.2 2
2.5 4
% REMOVALS FOR VOC
TCE PCE
AVE %RSD AVE
35.3 9.2
39.0 - :
43.5
47.3 4.5
58.6 1.1
' 66.4
66.4 1.1 .
75.6 3.5
43.3
54.1
56.5
61.7
75.1
79.5
. 79.7.
84.4
C-1.2-DCE . 1,1,-TCA 1,1-DCE VC :
%RSD AVE
11.1 25.2
28.6
- . 34.8
3.6 34.6-
0.10 40.6
50.7
0.52 53.0
3.9 59.2
%RSD AVE
8.5 53.6
- 54.5
- '61.1..
2.8. 66.0
0.40 75.6
80.2
2.3 72.8
6.5 85.5
%RSD AVE
10.5 63.0
70.0
- . 75.0
. 3.7 76.9
1.3 79.1
- 80.5
.1.5 72.8
1.0 87.0
%RSD AVE
9.3 , 57.9
.'.'.'- .79.3
' - i. 85.6
2.3 86.4
0.60 98.2
- 97.9
3.6 . 97.5
2.9 95.3
%RSD'
. 22.1-
' -.-.
-."
.10.5
0.53;
' - '
0.79
.. 3.5
-------
E2Sฃ
'" the small bubble - 3A In air line, large bubble - 3A In air line and
large bubble - 1 In air line air lift pump footpleces, respectively. The
. cost differences were greatest at the highest air-to-water ratios, where
the operating pressure differences were the greatest. ' ..
.''' VOC removal efficiency was greatest at the highest air-to-water
ratios and the air lift pump efficiency was best at a low air-to-water
ratio; therefore, an air lift pump would have to be operated in an
Inefficient way In order to use the pump.for VOC treatment. Operation at
the highest air-to-water ratios caused the small bubble air lift pump
footplece to . be considerably more expensive than the large bubble
footplece. Since there was no difference In VOC removal between the two
air lift pump footpleces, and the small bubble footplece was more
expensive to operate, the large bubble air lift fcotplece with a 1 In air
delivery line was chosen for use during the rest of the In-well aeration
testing. . :
Sparging air Into the well decreased the efficiency of the air lift
pump. This was consistent with air lift pump operating theory. There
was very little difference In operating pressure and cost to compress air
between the large bubble and small bubble sparger footplece. The cost to
compress air for the air lift pump and sparger combination ranged from ^ .
to 1U$ per 1,000 gal for air-to-water ratios of 2:1 to 9:1. These costs
were similar to the air lift pump without the sparger at the same air-to-
water ratios. . ;
There was no observable difference In VOC removal between the large
and small bubble sparger. The large bubble sparger was chosen for use
throughout the rest of the In-well aeration testing.
It was foundvthat adding air to the well through the sparger could
turn the entire well into an air lift pump and cause water to be lifted
up and out of the well borehole and onto the punphouse floor. The
air-to-water ratios were limited when sparging to the point at which the
most air could be added without blowing water out of the well borehole.
In the footplece testing, VOC removals were similar for similar air-
to-water ratios, regardless of whether air was Introduced by the air lift
pump alone or by the air' lift pump and sparger combination. Vinyl
chloride, with the highest Henry's Law constant, was the exception to
this. It was better removed by the air lift pump and sparger combination
than the air lift pump alone.
60
tj-^'i^ii
-------
V." :./ SECTION 8 . " '
":-: '- AIR LIFT PUMP TESTING WITHOUT SPARGING
AIR LIFT. PUMP TESTS WITHOUT SPARGING - PROCEDURES . . .
1 . The air lift pump was tested without the sparger (In-well diffused
aerator) at each of the three depths chosen for study. The depths (130
ft, 200 ft, and 280 ft) were chosen based on water and VOC zones of entry
as determined during the wel1 characterization stage of this study. The
130 ft and 280 ft tests were run In duplicate. The air lift pump
testing, without a sparger, Is represented In the first colum of the
experimental matrix In Figure 12. The large bubble footplece and 1 In
air line were used as a result of preliminary testing (see Section 7 for
details). One of the tests at 130 ft was performed with the 3A In air
line.
As with the preliminary footplece experiments, each air lift pump
test consisted of a series of 15 minute runs using various air flow
rates. The air flow rates ranged from the maximum air delivery rate
possible to the lowest air delivery rate which would pump water. The
maximum air delivery was obtained by attaching both the air lift pump and
sparger air lines (side A and B of the air delivery system shown In
Figure 10) to the air lift pump. The lowest air flow rates were measured
by the small air rotometer (0 to 6 cfpm Instead of 0 to 60 cfpm), which
was placed on the A side of the air delivery system for this purpose.
The actual procedure followed for each run within the air lift pump
tests was the same as that used during the air lift purp footplece
testing. At the beginning of a testing day, the static water level
(water level before any pumping occurred) was obtained. The air lift
pump was turned on at the maximum air delivery rate. The pump was
allowed to operate until the weir box was full. When the weir box was
full, the timer was started to measure the first 15 minute run. After
five minutes, readings were taker, of all the valves, gauges, and
thermometers, as well as the water flow rate and pumping water level In
the well. Any necessary adjustments to rotcmeters or pressure valves
were made at the five minute reading. It was very rare that the
rotcmeters or pressure reducing valves required adjustment during a run.
'At the end of the 15 minute run time the readings were taken again and
VOC samples were collected from the raw water sample tap and the v-notch
of the weir box. Once all of the readings and samples were obtained the
hext air flow rate was selected at the rotometer and the 15 minute run
clock, was started again. This procedure was repeated until all desired
;alr Flow rates had been tested. Note that when using the air lift punp
without a sparger It was possible to move directly from one run to the
61
-------
r'-?,^^^'VSS;^YCr^~^^.^-'-^''T^
-*;.'" ' ~-~* _'_ ปi--rn-c~--->Mi;^.-ซna*=r^,i i,^ . /-''-.- -^ - ~ ; - "'-V,~~ L ']_'"." ~*"~ T7T~7 ' :"~1' 3"
'''.''.'-.: ซ-; Y ' . -''- ':-' C''.ซ.ป- .'''--;. .'- .- '--"- ' ' .'."--. '.'-'..-".':'." V-'..- .''' . ''..-- '"''.' ':.'-' ^v^..^ Fa
. '-; -.. ;..,-::--.. -.'.; "-';' :'''.'/..':'_' ' ."..'" '''" Yv' <"'. '. ' ': . . - '- '-,. ; ' ซ- >'--'.-KS
next. When a sparger was used It was necessary to provide a US-minute
waiting time between runs to allow the well to clear of bubbles so that
an air-free raw water sample could be collected. "_
AIR LIFT PUMP TESTS WITHOUT SPARGING - RESULTS AND DISCUSSION . .' .. > . ', fs
Air 1 Iftpunp operational results are show In Figure 2U. The graphs . 5;|
show air lift punp operating pressur&, submergence and wire-to-water '_-'K|
efficiency for tests run at each of the three selected depths In the .'--ฃ$
The submergence of the air lift pump determines the required '
operating pressure. If the air lift punp submergence Is too low,
operating air pressure will be low, but there will be greater loss
because of air slippage so larger volunes of air will be required. If - ,
the submergence Is very large, air slippage losses and the volume of air
required will be smaller, but operating pressure will be very high. Air ; '|
slippage, losses and high operating pressures cause loss In pumping
efficiency. . hf
At well L-8, the greatest air lift punp efficiency was predicted to
occur at 65% submergence. Figure 2U shows that did occur. The
submergence obtained at the 130 ft depth setting was the closest to 65%,
and the wire-to-water efficiency was the highest at this depth setting.
The wire-to-water efficiency decreased with Increasing percent
submergence which agrees with the literature (5,29). Figure 25 is a
theoretical punping efficiency curve adapted from Ivens (29). The
submergence obtained for the three air lift punp testing depths at well
L-8 have been plotted as points on this curve. The data from this
testing closely matches the theoretical curve, showing that the 130 ft
depth provided the most efficient air lift pump. Operation of the air
lift punp at 280 ft (far below the point of optimum submergence) was
difficult. Debris and oily sludge-like material were pumped up from the
bottom of the wel1. I '
Figure 2k shows the results of duplicate testing at the 130 ft and i\
280 ft depths. The squares, plus signs and diamonds show the results of |.?:
the air lift punp with a 3A In air delivery line. All of the other . ;1
tests were performed with a 1 In air delivery 'ine. The smaller air line :x-
caused a noticeably higher operating pressure at 130 ft, but this *3
difference was not apparent at 280 ft. This figure also shows that even ;j
with a different size air line, reproducIb111ty of the air lift punp I-
operational parameters was very good. . ']'
The cost to compress air for the air lift punp tests Is shown In . \...
Figure 26. The alr-to^-water ratios tested ranged from 1.3 through 10.6 j^
to 1. As expected, the maximum costs were seen at the highest air-to- i
water ratios, where operating pressures were highest. With the air lift
punp at 280 ft, the maximum cost to compress air was 28.U$ per thousand jX
gallons. With the punp at 200 ft the maxlmun cost as 25ฃ per thousand |.:^
gallons and at 130 ft the maximum cost was 20.UC per thousand gallons. |;
In comparison, It was estimated that the total cost for packed tower f^
aeration would be similar to these In-well aeration costs, but the packed
tower costs included amortization of capital, operation, maintenance and
repumplng to the distribution system (33). Since packed tower aeration
also achieves better VOC removal than air lift punping, these tests
. 62 .
,,.-:.
-------
11O
fr
a
ปi
u
n
ti
Cti
W
u
o
H
en
to
CO
100 -
TO-
CO -
7O -
30-
ao -
10 -
110
100 -
TO -
ao -
ao -
*o -
so -
ao -
10 -
110
100 -
oo -
ao -
7O -
O-
bO -
4O -
3O -
ao -
10 -
130 FEET
200 FEET
- . SUBMERGENCE
-0-0-" ^PRESSURE
EFFICIENCY
SUBMERGENCE
280 FEET
EFFICIENCY
B* OPERATING
PRESSURE,
PSIG
i eo ao
ROTOMBTTER BSTTIHO
' SUป!ERGฃNCE,
PERCENT
100
120
*ป PUMPING
EFFICIENCY,
PERCENT
Figure 24. Submergence, operating pressure and pumping
efficiency for air lift pump tests at 3 depths
in well L-8.
63
^
'::-3 :^m
.* - *)
-
\-
,;,i
. I si
-------
50
40-
30-
Ul
0
L
L
y
K
20-
10-
^^
20
280 FT
ADAPTED FROW IVENS"(29)
I i i T i i i r
40 60 80
% SUBMERGENCE
100 120
|
I
-:S
1
Figure 25. Theoretical air lift pump efficiency curve with
points showing a comparison to NFWA test data.
. 64
F\ '
I-.-' V
__ ' . _;_ " _' _ฑ-__ " ' -" " ~-' i
-------
t, ,^ ' ฃf~~* *ฐ^ '*" *" v ~ '-^i "" ^^ *~^ 1**1
"- - ' - l r
^ ซ ** * r * ' ~-* " -* * , I?'
f ^ ปv ^ I <- ^ *- *ฃ
f ซ s $f
!' > f^
; l P
30-i
28-
26-
Ojl
AT "
Z 22-
0
j 20-
0 18-
D
Z 16-
' " ' Ml
' D 14-
o
. - * 1 7
\
p 10-
z
ui 8-
o
; '; .:"' ;- '6-
; -. ; ' 4-
-/'; ".-'-.'; -V.2-
'--. '''.' .- ' n.
*
^ = AIR UFT PUMP AT 280 F * +
i
i
+ 'ซ AIR UFT PUMP AT 200 FT
D = AIR UFT PUMP AT 130 FT _ D
D
V ' ;'- . ='".- '.'"' .';.; ' _--.'.'.' -.!'.: ;
".'- '. . ' "' ' ' ,''''
6 ' '"' ' '
.v . - - _ - .
t ' '' . . .
o ' '
U -'':'..''''
Q 4.
. '''"' .
'-..*+ - -' '.-.';-:
i* Dฐ' " .' : '-' ' -' ''!:- :'->
DD.: " . ' =-':--;--;-: '''> '--:'-.V
4
!'i
II
fjj
! 4
M
i i
>~i
r|
i
y
' a
".-'" "- "v- p
' '- '...-. - ll
' : '' ' " - " 1
'""*ป
' I ?
'- ' - : ll
wi
' . . H
r
;
' 3-
- J-
,...,.', :'"-.|
. .;-..;- ,:'-'|
':-"-- ''' i
: Vj
4 6
AIR TO WATER RATIO
8
10
Figure 26.. Cost to compress air for air lift pump testing
at three depths in well L-8. --, -.
\
65
-------
v -.;
Vv;
A; -J
J*::."':'".-" ':"-'-''*'' ' './'-:-'-_-;-.':--'--:: .--..- '-.--...-;.." . -.'ป:-/:'' -:-.... : ;j;---^',^v'^;'";'
t _ ' V showed that packed tower aeration would be the preferred aeration process . : i '_-.'
!"''. --''-';.-..'for VbC treatment over air lift pumping. V . \.~. ''."';:-> "r ; - V \K^
r ' v ;" ."''." The VOC removal results for air lift pumping without a sparger are pi
'. shown In Table 10 and Figure 27. Three VOCs were graphed to evaluate the '__'.. |.,f
': - removal capabilities of the air lift pump. These compounds were TCE, " : I'd
>. .. c-l,2-DCE and VC. The air lift pump foptplece testing Indicated that the
:. .; VOCs found In well L-8 were removed in an order which could be predicted
;J by Henry's Law constants. Rather than evaluate each compound found In .
' " " the well, these three compounds were chosen because they represent the ' [.|
;'.. . major contaminant -(TCE) as well as vinyl chloride (highest Henry's Law
; : --constant) and c-l,2-DCE (lowest Henry's Law constant).
| . : The results of air lift pump testing have been analyzed similarly
to the results of the air lift pump and sparger preliminary footplece
:..'_ - testing. VOC removal percentages could be expected to vary from
" test to test by at least as much as the air-to-water ratio. If some
- other factor were affecting VOC removal, the VOC removal percentages -
... would vary more than the variation of the air-to-water ratios. Table 10
' shows the variation, %RSD, of the air-to-water ratios and VOC removal
percentages for each compound studied In well L-8. The %RSDs for the
,'.-. air-to-water ratios are, In most cases, lower than the %RSDs for VOC
; : . . removal percentages. This would Indicate that some factor, other than
' ' the variation In air-to-water ratios, was causing the variation in VOC
1 removal percentages. Since the air lift purp was operated at different
, ^ depths for these tests, depth of the pump In the well was one possible
k\ x, cause of the VOC removal variations.
Figure 27 shows the removal curves for TCE, c-l,2-DCE, and VC. In
Vi . .'. each case, It was difficult to visually distinguish a difference In VOC
"t\ : removal at the various depths. It might have been expected that the 280
'' ft depth would have provided the best removal, because the air and water
contact time was the longest. Since no VOC removal advantage was
obtained by operating the air lift pump at greater depths, and the
greater depths were more expensive In terms of operation, an air lift
pump treatment system should be designed to operate at the maximum
" . pumping efficiency depth setting.
Raw water concentrations for the three VOCs discussed above are
i, shown In Figure 28. Because of the nature of Henry's Lav/, it Is
_._ ' generally accepted that In aeration, removal efficiency Is not related to
raw water concentration. This figure shows that the raw water
concentration varied widely from one test to the next. Notice also, that
. the raw water concentrations change a great deal from run to run within a
I single test. This Is why a raw water sample was taken for every treated
I . water sample.
" The raw water variation from test to test may also Indicate that ||
. . other conditions within the well were changing. If the concentrations 1
/ varied, perhaps the zones of VOC entry Into the well also changed. This -\
n could be a factor, other than depth of the air lift punrp, which afvected f
VOC removal. This possible effect of well conditions changing with time ||.
will be discussed further when sparger results are shown. j;j|
Variation in the analytical method for VOCs was 10% or less. Some H
:of the %RSDs for VOC removal were less than 10% which could be from the y
analytical method. At several air-to-water ratios, however, the i;i
/ variation was greater than 10% which could not be explained by variation t'/j
i 66
3
:j
i . .; . , . -. I
V.
-------
\
....: \:.'..
p. /
TABLE-10V VOC REMOVAL BY AIR LIFT PUMP WITHOUT SPARGING (AVERAGE OF ALL THREE PUMP DEPTHS) /j:. : ,;''i.
AIR TO WATER
RATIO
AVE %RSD N
1.3 .3.5 3
1.7
2.5
2.9
3.5
5.3
- 1
6.1 2
- 1
2.7 3
- 1
9.4
1
10.6 5.66
% REMOVAL OF VOCs
ttSi"
TCE
AVE %RSD
27.7 21.1
41.9
33.7
PCE
c-l>2-DCE 1,1,1-TCA ' 1,1-DCE
vc
AVE %RSD AVE %RSD AVE %RSD AVE %RSD AVE %RSD
35.3 8.76 18.1 26.8 . 41.4 21.6 55.0 15.3 43.8 11.5.
41.8 - 24.5 - 44.6 - 52.6 - - -
35.6 - 18.6-49.2 - 54.7 - 52.1 .- :
56.5 - 34.9 . - . 60.9 - .72.8 - 60.0 -:
-^
..'-':.:."!
50.0
46.2 9.20 60.5 6.06 33.3. 4.68 59.9 4.09 71.9 ' 3.69 73.2 .4.
48.8 - 59.4 - .30.7 ' - 70.3 -. '72.9 \- ;, 74.8 ; J-
59.5 ;- 75.2 - 46.4 - 75.3 - ' 84.6 - 100. -
64.7 7.15 ; 76.7 3.04 47.4 11.8 i'.1 81.8 :} 3.57; 83.8 8,90 90.4'.':/.7.
'; . ''" :'>'' " .'.' '" ''-.' '.' '.'-.?...'. /''? *'-':'', <'.'i
'- .'- V" '-;. ' I..-.''.* ;- '' '' ' '. ";'' :V;' ". * ป'.'. '-l^f
' ':".- '' '" '""!-''''^,''- ' '::-l ''"''':. :.^'^'-"k^A-^->V;^
ffซaJปry^t<^^ai"""^lซr|i^'i.i'''*'ili'i''rv'^;'i^-*<''J^*fyi''.' '-Tfrf*^^1'*"*?VT'i*i*r"f''""'*rrrwftป!r'rrfrT?i:*^T-'i*taT^51
'ฃฃ' ':> 'ซ-:^^i$ffi^''*^ ;-\^^?^"^Wf^
-------
r.*? T*-1,~7'T*!C'"7'ir%1'"ซ"? > _T*-*ฃlTr-K"aTrr -
V.
TO -
40 -
ao -
io-
CU
ปo -
70 -
BO-
ao
10
100
BO
TO -
o -
BO -
40 -
ao -
ao -
10 -
o
TMtOHUOftOC1>nUCNB RBMOVAU
1 .aDIOHUDROGCTHYUCNEC REMOVAL.
=-Pump @ 130 feet
= Pump @ 130 feet (Dup Test)
= Pump @ 200 feet .
= Pump @ 230 feet
= Pump @ 280 feet (Dup Test)
VINYL. CHLORIDE REMOVAL.
10
"Si
n
AIR TO WATER RATIO
Figure 27. Percent renoval of major contaminants by the
air lift pump at three different depth settings
in well L-8.
68
V -'!
Pi
-'1 -
-------
1.*
1.1
. 1
o.e
O.T
OJt
o.a
0.1
aoo
100
so
10 -
280 Feet
200 Feet
130 Feet
130 FeetTTxro)
TOE IN MAW WATOR
280 Fej* (Dup)
01.2DOE iN |4A*T WAFER
VO IN RAW WATER
130 Feet
130 Feet (Dup)
^
10
AIR TO WAPER rtAHO
Figure 28. Well L-8 raw.water VOC concentration during
air lift pump testing at three de^chs.
69
-------
during analysis.
AIR LIFT PUMP TESTING AT GLEN COVE, NY
Air lift purpjng was also evaluated for VOC control at Glen Cove,
New York In another U.S. EPA funded project (34). The air lift punp
evaluation was only a minor portion of that project; adjacent wells were
used for evaluation of other control technologies. ., x
In the Glen Cove project, a shallow well was selected for
Installation of an. air lift pump.. The design was based on 70%
submergence and water delivery of 50 gpm. The footplece configuration
was an open-ended pipe; the air line was located concentrically within
the eductor. Air and water flow rates were measured by similar
procedures to those employed at NPWA. A blower was used rather than an
air compressor.
There were several factors, however, that Interfered with proper
.evaluation of air lift pumping. Although designed for 70% submergence,
those conditions were rarely realized. Seventy percent assumed no
drawdown resulting from operation of nearby wells, however nearby wells
were In frequent operation. Pumping water level was measured only once.
It revealed 58% submergence as a result of adjacent well operation.
Although designed for an air-to-water ratio near 3.3 to give optimum
efficiency, It was operated at ratios of 11 or greater. Finally, optimum
design conditions suggest an eductor size of 2 to 3 In Inside diameter.
Allowing for the concentric air line, the 4 In eductor that was used
resulted In an equivalent Inside diameter of approximately 3.8 In. A
larger diameter In cotrbinatlon with larger air flow would promote
slippage. Slippage and reduced submergence would promote decreased
efficiency. The operating conditions for the one day when pumping water
level was measured, are given In Table.11.
TABLE 11. GLEN COVE AIR LIFT PUMP OPERATING CONDITIONS (34)
Water flow, gpn LT or = 44
Alr-to-Water Ratio GT or = 11
Lift, ft 12.7
Submergence, ft 17.75
Percent submergence 58
Wtre-to-Water Efficiency (a) 3.7
Eductor diameter, inch (b) 3.8
Cost to Compress Air $/1000 gal (c) 4.2
Date: g i ven In:
a Assuming air .compressor of type utilized by NPWA
b 4-lnch 1.0. less nominal 1-Inch concentric air line
c Assuming 6^/kw-hr
LT: Less Than
GT: Greater Than
70
-------
:-:,;,- The very low efficiency at higher than optimum alr-to-water ratio Is ง
;consistent--with NPWA air lift performance. The relatively low cost to . . 1;
.compress air at that air-to-water ratio results from operating In a ;.':> |
shal.low well; the compression ratio at Glen Cove was small relative to |
compressing air under 130 ft or more head at NPWA. ' ! <' . -|;
.TCE, PCE, c-l,2-DCE and 1,1,1-TCA were measured In the well outside . ^
the eductor and effluent from the weir box/separator. Results for the f!
principal VOCs are given In Table 12. The data are consistent with those j!
developed atNPWA In that air 1 Ift pumping never achieved removals that '\
would make the technology comparable to packed tower aeration. . ' . - - . ' '" i;
-.'--.- ''. -' . - -. -: -' ' - . ' " ' .- -.' ' " ' '' '""" ' : :;- -'I
TABLE 12. VOC REMOVAL BY AIR LIFF PUMPU-JG AT GLEN COVE (34) VY I
NUMBER
TESTS
2
13 :
8
6
' AIR TO
WATER RATIO
11
15-20
21-27
34-51
RANGE PERCENT REMOVAL WATE* FLOW CONTACT
TCE
58 ฑ 2
54 ฑ 10
54 * 3
52 . ฑ 3
c-l,2-DCE GPM a
38 ฑ 2 44
33 ฑ 14
27 ฑ 6
31 ฑ 13
TIME, SEC
24
a Data only reported for selected alr-to-water ratios
CONCLUSIONS FOR AIR LIFT PUMP TESTING WITHOUT SPARGING
The operating efficiency of the air lift punp was greatest when the
punp submergence was near 65% Cat the 130 ft depth setting). This was
the submergence which the literature predicted would produce the greatest
pumping efficiency. The air lift pump efficiency decreased as the
submergence increased, which was also described In the literature.
VOC removals for the air lift pump without a sparger ranged from
maxlmums of 90.4% for VC Cbest removed according to Henry's Law Constant)
to 47.4% for c-l,2-DCE (lowest Henry's Law Constant). The best TCE
removal obtained was 65%.
VOC removal was poorer et the 280 ft setting of the air lift pump
than It was for the 130 ft or 200 ft settings, which were similar to each
other in terms of VOC removal. Pump operation was difficult at the
deepest setting.
Raw water VOC concentrations varied widely from one test day to the
next, as well as varying within a particular test. This confirmed the
predictions derived from the pump test conducted during well
characterization and emphasized the need for frequent raw water sampling.
The cost to compress air for the air lift pump was found to be
similar to the total costs (air, operation and maintenance) of a packed
tower aeration system. It was concluded that In-well aeration by air
lift pumping alone would not be ah adequate substitute technology for
packed tower aeration. Air lift purp testing performed at Glen. Cove, NY,
under a separate U.S. EPA funded project, confirmed this observation.
71
-------
; SECTION 9
AIR LIFT PUMP TESTING WITH A SPARGER
AIR LIFT PUMP TESTS WITH SPARGING - PROCEDURES
-: - The air lift . ptnp was tested with the sparger In various
; combinations of depths, as shown by the experimental matrix In Figure 12.
; ' . .-...: The air lift pump was the same one used In air lift pump testing without
<-'.'. - a sparger. The sparger was made of 10 ft sections of 3/4 In plastic pipe,
1 . . connected with fittings to reach the desired depth. The sparger did not
;.'. . have any type of attachment at the footplece, as was determined In the
; Preliminary Footplece Testing (Section 7).
"-.' The testing procedure established during the preliminary footplece
; testing and air lift puip testing was used for all of the air lift pump
- and sparger combination testing. The static water level was recorded,
then the air lift pump was operated at a constant air delivery rate of 20
cfpm. This rate was chosen because it produced an efficient air lift
pump In terms of punplng Cover, though it was not the most efficient In
terms of VOC removal). The weir box was allowed to fill, so that water
flow rate measurements could begin. Next, the sparger was turned on to
:.';- the first selected flow rate. The tasts were conducted as a series of
. fifteen minute runs. For these tests, the A side of the control board
was used for the air lift punp, and the B side of the control board was
used for the sparger. (See Figure 10 for a diagram of the air control
. . board). After each run using a sparger, air delivery to the sparger was
. stopped, and the well was allowed to clear of air. During the well
" clearing time, the raw water tap was sampled and monitored for DO
:.'..' concentration. When the DO value reached what It had been before
' sparging was started, a raw water sample was collected, and the next
sparger flow rate was started to begin the next 15 minute run. An
average time to clear the well of air between runs was 45 minutes. This
clearing was an Important procedural difference between running tests of
the air lift pump with the sparger as opposed to without the sparger.
When a sparger was not used, the only air being Introduced Into the well
was being Injected directly Into the air lift purp, and therefore the air
i ' would not affect the Integrity of a raw water sample. While the sparger
v/as operating the raw water sample punp was surrounded by aerated water.
Collecting the raw water while sparging would result In a sample full of
; i bubbles, which Is unacceptable forr~VOA sampling. It was Important to
collect raw water VOC samples for each run, because preliminary tests "had
/ ' ' shown that raw water VOC concentrations varied a great deal, especially
When the well was first turned on. The air lift pump could not be turned
-'...'. bn and left on for the nunnber of days that it might take to allow VOC
-''.'..:..''': ..".- .72 '
-------
A
\ ~~
doncentratIonsto level off (see pumping test results, Figure 9), because ' .
the portable air compressor which we used was too loud to leave operating '
overnight In a residential area. . . '.
,= The operation of the sparger caused problems with the method
"originally chosen to measure the water level. A 1/2 In plastic pipe had :
been lowered Into the well where an electrical resistance water level
probe was used to measure the water level In the plastic pipe. This
design worked well as long as the sparger was not operating. When the
sparger was operating In some configurations, the air from the sparger
entered the plastic pipe for the water level probe and actually turned
the probe pipe Into an air lift pump eductor. This water level
measurement problem made it Impossible to determine air lift pumping
efficiency while the sparger was operating and several tests were
repeated In order to obtain pumping efficiency data. The water level
measurement procedure had to be changed to an air pressure method, which - .
Is described In full In Section 6, Experimental Design.
Air lift pump operational problems occurred at the 280 ft depth
setting, so these tests were not repeated. Operating the air lift pump
this close to the bottom of the well caused an oily sludge to be pumped
out at the beginning of each testing cycle. Water flow rates were
Inconsistent. No pumping efficiency data was obtained at this depth.
AIR LIFT PUMP TESTING WITH SPARGING - RESULTS AND DISCUSSION
Figure 29 shows the wire-to-water efficiency of the air lift pump
with and without sparging. It Is Important, when looking at this figure,
to realize that the curves for air lift pump efficiency with and without
a sparger are shaped for fundamentally different reasons. The curves for
the air lift pump without a sparger were created by adding Increasing
amounts of air directly to the air lift pump footplece, thus Increasing
the air-to-water ratio for In-well aeration. As discussed earlier, an
air lift pump has a point of maximum efficiency, at which the minimum
amount of slippage will occur In the eductor pipe. Operation with
air-to-water ratios greater than the maximum efficiency will cause a drop
In the efficiency curve.
The curves showing the effect of the sparger on air lift pump
efficiency were obtained by operating the air lift pump at a constant air L
delivery rate (20 cfpm) while increasing air to the sparger. The f;
efficiency of the air lift pump was observed to have dropped as the I
sparger air delivery rate was Increased (which Increased the total I-
air-to-water ratio of the In-well aeration system). This drop In j*
efficiency of the pump was caused by decreasing the water density \
differential across the pump footplece, which provided the driving force . };
for operation of the pump. Even though the wire-to-water efficiency of ', [;
the pump decreased while sparging, It still remained higher than when the \
air lift pump was operated at Increasing air delivery rates, without a f.
sparger (as examined In terms of total air-to-water ratio). |-
Figure 30 shows amounts of water pumped with and without sparging, fc
with the air lift pump set at 130 ft and 200 ft, respectively. The air fc
lift pump without a sparger pumped more water than the air lift pump with [ป
a sparger at a given total air-to-water ratio. I,
Even though less water was punped when the air lift pump was T
73
;
-------
f--
**
r
"
i
^w
35-
30-
E
5 ---
- - ' ' z "-
i.':. . .<:.; . y , .20-
"' ":- ' :"':'.' '?:' ":- '-"''
. .- -.-. '-. ฃ 18-
. ' '..,' k. .
'. ' , H '
i.-: .. .-; ' ' \ 10-
'"'"'"-' 8'-
.']-.'' .- 35-
30-
K
I 2a"
z
y 20-
0'
..' V. ฃ 18-
-,- y
:' " - '' '.. 1ฐ-
.'.-..'."
0-
; ' '"-.
y.';--: ... -''-::'\
r .\ ' '....:
i 'i ""X '
L_Jiii2ik. ' ' '- ""'', ';. "
~ " ' ' "
"" y '" *
" ,i~
< "
*
-
AIR UFT PUMP AT 130 FT
n NO SPARGER 8-10-83 = a
D
n SPARGER AT 130 FT 4-3-84 = +
. , . SPARGER AT 200 FT 4-3-84 = ซ
.n + ; -.'.. -.SPARGER AT 200 FT 3-18-85 = A
% ' ;-' : " SPARGER AT 280 FT 3-19-85 = x
^iป " . . .-",-'
- ;...ปn .-_ _ -.-',-::=>' ' '':.-..'...':.- -''^.::'' ;'"
' ' ' ' '- '.'-:' ".'.'. . ' "
' - T ; * ' - . - . . ,-
- . . . - - v _ . ". ''-'"
Q '"' -""'-;"' '
' ' '.-'.-'' .-.'.;; 'n ' . + : ',': ': :;- ,-,'-." '
. , ' V ' '" : ':.'." ,'.. a ." '. . ' ':'...-..'. .' '-'
AIR UFT PUMP AT 2OO FT
D ' - ' . ' ' . ' ' .':;-.
n - '.. ' . . .
- - .''-
a * " .'..'"'
-'. ' .
v "
'a ฐa . '. . . :'
Da 5
- a --.-.' .-'.- . .
4^ 4 '
D + * *
a ' + * A ' ซ
.''.'-. . a
) 2 4 6 8 10 12 14 16 1
AIR TO WATER RATIO
Figure 29. Air lift pump efficiency curves showing the
effect of operating the air lift pump while
sparging.
74
-'v ' ..'.''.'. -,..' :
"' - ' '! . ' ,.".-' :.',.''-. , - ' jf a ' <
f *, ^ i
~
*
"
' . ' : '
"''. " " ' '',
:.-'.-' - .' .' ''' ' '
. ..' _ . :. .;;-- -
-'i, - '. -- ' "
'''.
- ; - " : ' . - ' '
'' ,' - .
'
* .
8
' ..
.
.
." .
-- 'v'- -; - -: '--";_;: ."
i
1
^
'i
2
-,
^
J
t
"*!
1
i
"*i!
'^1
y
ti.
ji
ฃ
?
. l.
'
!'
\.
i
\
i
'^~
|
|-
Hj
J .
1
;-
''
I"
\
1
';.
I
_-
|
'? -
J
-------
fl'ift
i H '*
f V ' .' "
I - ' " '
f- ' . '. - - - .,:
**.-."' ."'.',
I". ~ .
:-' "" ''-.
i" .
: ~~
' '
^
*-
''.' '.'
',:-"
'
..;.f'tC'^V''
'-""'." ':''. '?'-'?
' *' ' . . ""'-
' -;
- . :;'.'' " .
..- ' :-'. ..' .
-
-." -'
. - f\
c
:E
I~
v
Q
1"
I
T
*
I
E
H
\
>
' t. .
\
''''' ''.^-^-''.' ''' 'v-:;.^'.vV;S--' V-I;;,^;;^';;.;^ ':'
\''": .'.'' ' ,"" . ..'.!"" r'.' "'-"-*':. . " ' '"': ,'"';'J '^' ''''
.ปv' - "' "-.''"-"' V . - '. "' ' ' '""r- ', - '. * . -''
*- .*'.-' ' . ~ . >. " ' . . .ซ.-"",
' ''.*''.-''.''' ''.'."--' ' " "" '' . - "v" '":. - '' ' .'.'-' '-
< 1O l ....: .
100-
oo-
80-
7O-
60-
6O-
4O-
30-
20-
10-
1 m
1OO-
90-
8O-
70-
60-
80-
40-
30-
20-
1O-
0 -
" ; ' '' ' ' '.-"' '. : -.' '':: AIR UFT
nfff^ - O . ' - ป .
" ' n ' '
' '.' + . :- + ''. -. '-' .::'- . + '
a - . .- ' '
. - .,.
''.'"- ' . .'-'.-' :
;' '' "V-' 'r-'^'-- *>":' '/ -'-, :"C;'-.?-.- '' *-'f:^^^.~^ ||
~".- rX.;.'^- ', :':'. ':' :' ' ">''-'' -'- *:-'-~- '-.'f : - - S
" " ' . ?''.- . * ' - : ''*- ' ^~" "*" \ - "'' ' ''"*'' '''"JCS
.ซ * *- -" " .""' " " ' ^' '' s . ~* i '^-^ " '_" ' . ..'^ * -. f"--
.'," '" ''>." ' - ' ' ,'" '** "' '""'*'' -'' ' , "-^ " 5*?
' - * ;/.-. ; *;.'.ซ," "''-.',-*'>. "_ :l - - . 'v-*-ป -^
"'-''.- '- '.. V. ':".v- '-' -&
PUMP AT 130 FT
-..: " . ' />-;< .'' - .-.*
'. 4. '
.'. : '.--"', :
." ' - ". .
. NO SPARGER 8-10-83 = a
D -, .' .-.
SPARGER AT 130 FT
;. SPARGER AT 200 FT
; . ' '.*.
a SPARGER AT 200 FT
' - SPARGER AT 200 FT
o -....'
iff*1 ฐ ฐ a AIR UFT
' . : O
D
ซ
0 + +
X
A * *
X '
A . .
n S ป
. *
a ' ..'-.;/
a . ' . -
0 24 6 8 10 12
AIR TO WATER RATIO
4-3-84 = +
4-3-84 =. '*
3-18-85 = A
3-19-85 = x
PUMP AT 200 FT
;
'' ' .' *
* ' '
.' ' . I-
.'" " " . '' "-"' t!,
j- _ ; . ; ._ ป:'-
-' ' i--'
- "' . ' " --'!- ^
' ^"" ".' -'-'' ;';l
' . - 5 '
-.'-' - ' ^ฅ
' >: " ฃ
' " &i
' "' ':[',.. I
'f-
' ' . fe.
1 i/'.
ซv
;.:;/ ; 1
" I*
{
I
" $
i;.
l
i.
1)
1
1
- i
s
- 1
1
"'"'.;|
; ' .-. :--; I
' I
14 ie 18 .....' |
" . ' . 1?!
;'' Figure 30, Rate of water punped by the air lift pump with .'.'' :.
X
! X
* ' . X
i . . :\
F ,'- X '
{?.',,-- l-loj^
,
'k
.;t>v ^ _
and without a sparger.
- 75
j. . . . - '-.'.' "f '
. . - , . '. ' " . '..-->*
" .
-------
- \ '
operated with a sparger, the air lift putp efficiency was greater with
the sparger than without. This was because, for any air-to-water ratio,
only a portion of the total air being delivered was used by the air lift
pump to move water, so only a portion of the total air delivered was used
In the air lift pump efficiency calculations. In terms of pumping
efficiency, for a given amount of air to be delivered, It was better to
use the air lift pump and sparger combination than the air lift pump
alone.
Even though less water was pimped for the same air-to-water ratio
while air lifting and sparging, as opposed to air lift pumping alone, the
cost to compress air for the combined operation was less than for putting
all of the air Into the air lift pimp. See Figure 31 for the costs to
compress air for the various air lift pump and sparger configurations and
compare this with Figure 26 which showed the costs of air lift pumping
without a sparger. At the same air-to-water ratios It was less expensive
to operate the air lift pimp-sparger combination than the air lift pimp
alone; however, the combination system allowed a larger amount of air to
be delIvered. At these high air delivery rates, the costs exceeded 3(K
per 1,000 gal. Since this was only the cost to compress air, that makes
this method cost more than a packed tower aeration system (34), which
achieves better VOC removal.
Table 13 shows the results for testing the sparger with the air lift
pimp at 130 ft. The % removals shown are averages for all three sparger
depths combined (130 ft, 200 ft, and 280 ft). These removals are shown
next to their respective air-to-water ratios.
TABLE 13. VOC REMOVAL WITH AIR LIFT PUMP AT 130 FT
(AVERAGE OF ALL 3 SPARGER DEPTHS)
AIR
AVE
2.2
3.C
3.2
4.2
5.7
7.3
8.6
12.
TO WATER
%RSD
9.4
1.7
0
6.3
3.3
3.8
1.2
4.8
N
5
2
2
4
5
it
2
5
TCE
% REMOVAL
AVE
33.0
42.6
39.6
49.3
55.1
61.4
66.4
78.3
%RSD
19.6
8.45
9.71
5.83
11.0
18.2
1.05
7.98
c-l,2-DCE
% REMOVAL
AVE
24.6
28.7
29.2
34.6
39.8
45.2
53.0
65.0
%RSD
11.1
0.35
19.2
2.70
14.8
21.6
2.26
13.1
VINYL CHLORIDE
% REMOVAL
AVE
62.3
79.0
77.8
91.2
96.0
97.6
97.5
98.6
%RSD
9.93
0.38
10.0
4.87
2.97
0.79
0.82
0.40
As with the previous testing, air-to-water ratio varies, even at the same
76
-------
40
'so.
SO
an
90
1O
1C
.'.-. '
4O
SB'
so
aa
ao
IB
10
B '
4O <
SO -1
ao -
ao -
IB -
10 -
AW UPT PUMP AT 13O PT
BPAROOt AT aSO PT
6PAROSR AT 2OO PT
8PAMOCR AT ISO PT
AIR UPT PUMP AT 2OO PT
AIM UFT PUMP AT 28O PT
o a
ซ 10 12 1-* 10 ia
AIM TO WATEK RATIO
Figure 31. Cost to compress air for air .lift pump and
. . . sparger testing at well L-8.
77
-------
I ;^\r ;-; rotometer. setting, because of changes In pump efficiency caused by _;.".;
i/r'.V 'varying well water level; The air-to-water ratio variation and the VOC . ".'; '.'.-.;
| -, .. removal variations are shown as % RSD. If nothing but the air to water .
i v . :.; V ratio was changing, the %RSDs for the VOC removals should be similar to
T. , the %RSDs for the air-to-water ratios. In Table 13 the %RSDs for the VOC
p. ; .removals are greater for the removal of TCE and c-l,2-DCE than are the
; , .! %RSDs .for the air-to-water ratio. This Indicates that for these '": :'H
i .'.-.',' compounds, there was some other cause of variation besides the variation 3
i - . In atr-to-water ratio. Since these results are compiled from testing ||
J. -.-'.. '-.--- with the sparger at three different depths, .the sparger depths might have ||
:'" : been assumed to be the cause of the variation. An examination of VOC ji
, : : removal data Csee Figure 32) shows there was no pattern associated with ^j
;-':"' depth of the sparger. The variation appeared to be. caused by poor . R
,'-.-. duplicate testing whan there was long period of time between tests. The . . Pt
; . " . ,. footplece testing had shown very good reproduclbll Ity of repeat tests fj
;/' when they were run within a few days of each other. All air and water
flow rate measurement equipment was checked for proper calibration and 3
. t the VOC analytical quality control data was evaluated, yet no changes had
; " .. :' occurred In any of these over time. It was known that raw water
concentration and well water levels were, changing. These could be
- Indications that VOC or water zones of entry might also have changed.
; . While It Is generally accepted that VOC removal Is Independent of raw .
: water concentration, water entry zone changes might affect the
reproduclbllIty of In-v/ell aeration over time.
Figure 32 Illustrates that the vinyl chloride data did not show as
large a variation as the TCE or c-l,2-DCE. Table 13 shows the %RSDs for
vinyl chloride removal to be similar to or less than the %R5Ds for
air-to-water ratios. This Indicated that nothing other than air-to-water
ratio was causing variation IT VC removal (unlike TCE and c-l,2-DCE).
Vlr.yl chloride had the Mghe5,t Henry's Law constant of the compounds
; tested and was very rapidly removed to greater than 90%. The variation
decreased as percent removal Increased.
Tables 14 and 15 show the VOC removal data for tests with the air
lift pump at 200 ft and 280 ft respectively. These results show the
same variations In %RSDs as the results for the air lift pump at 130 ft.
The tests at 200 ft were conducted over the longest period of time and
they show the greatest variation. Figure 33 illustrates testing with the
: air lift pump at 200 ft and Figure 34 shows the 280 ft air lift punp
tests. No pattern existed which would suggest a difference In VOC
removals caused by the depth of the sparger. Instead, changing
' . geological well conditions over time were the most likely sources of the
variation. ''.'
The maximum air-to-water ratios which could be achieved varied from
12:1 with the air lift punp at 130 ft, to 17:1 with the air lift at 200
ft; to 16:1 at 280 ft. The air-to-water ratio was limited by the point
at-which water was lifted out of the well bore hole by the foaming caused
by the sparger. This maximum air delivery point was Influenced mainly by
: ..-. the static water level of the well. When the water level was closer to
the surface, the water blew out of the borehole more quickly.
lw
-------
\
^^^-"^^^^^'^'^^^''^^'^''^^l^^^^t^^^Ji^^ ^ i ^TvT^? ^T ^Qfgft&l -^-P J- ^-^WK- 'ป * ^!t#. j^|5F^C jsP^'pf!!
I i
r ~^ * ., ^ ~ ~~ ' 1
' , - %
" " 'fl
'- - J
<- i on , * A
1 WW
0 -
0 -
ro-
o -
ao -
4O -
ao -
10 -
4OA m
.-'-.- ' ' %fw
. .- - < '." . .eo -
;- /: ; -; ' ;. -.-: 70-
- ''-: ' .-"' ao -
:.-.-; . . ฃ 40-
Q ao "
' ' ^
: - 2 ao -
H 10 -
n
' tl 1OO M
1
; . S *o-
/ . ' ' ' ' *9 -
ro -
! - - . . '
'.' ' - ' : ." ฐฐ"
. . oo -
r " ". ' . "" -ปฐ-
: :..-'..-. --. . -. ' ao -
' .. .-_' " . ao--
- -.- .-.''.'' 10 -
:- .- ' . o -
TCE REMOVAL A
H
^
ft 9
9
* *
* 0ฐ *
6 x
A
C-1.2-DCE RQ50VAL - . ^ - ,Vf.'
:.:'. -V:.;-:'-; '/;;0'':s;r;;Vv.::v-;-v:'-:^
; "w: "v *'. ' *" ' .'-''*
a @ ' ' -
ง A A '
x ....... . . - .
''.-..'' . '
. ' . ' " - " ' ,
A Q O S B
*s
' * * --
B S
. 9 ." ,;..-- . .
+ x SPARGER AT 130 FT 3/17/83 o
* SPARGER AT 130 FT 3/23/83 .*
. - SPARGER AT 200 FT 6/22/83 *
'-' - '-" ' ;-' .
. SPARGER AT 200 FT 4/10-84 A
VC REMOVAL SPARGES AT 280 FT 8/3/83 x
N
fl
i ^
/i
H
! -j
1 A
U
j^
4
.. : -a
' ' >.'-'" ' i,"7*
:.;:.-;;-y;';^ J
"..'.' v; ''". fl
.. - " i-;
"'' f'l
'..-'..- - SH
- . ;;- ' - |1
ง 3
p| '-7
' ll
i.^
i ^
'- |!'i
1
' - ' 1"^
-. -: '' -: ;'. -ll
1''
\~\
- : ฃ
.."- ' ' :''' ''A
' 4
RATIO
Figure 32. .VOC removal data for air lift pumping at 130
. feet while sparging at three depths.
79
-------
i
Pu
o -
ao-
ro -
ao-
ao -
4O-
ao -
ao -
10 -
ao -
ao -
TO -
ao -
'BO -
4O -
ao -
ao -
10 -
eo -
eo -
TO -
ao -
BO -
4O -
ao -
ao -
10 -
0 -
TCE REMOVAL
o B a K
^ *
*
T x
X X
a
* .x
x
C-1.2-DCE. REMOVAL ' ' .. , , -. , .... .
''.'' ;.,'"-'*' " n: .'-, w :';'. '-T ' ',' x :
- '.-''/ - ' ':.': * -n. ฐ T .*'V ' A ."'x '' .. .'.-' :
" ^ . . * '
- ' D -"'
'"'*-- K x -: ' ... ; '-.-"
T a ''''.-'..:.-.'." ' '.''
I x x
v . .-'' " -.. ' -- : ;, -.; -: .' -' - _; ' - '
u B a *T * S *
jj -xK x.
SPARGER AT 130 FT 8/19/83 n
A x SPARGER AT 130 FT 4/3/84 ซ.
B .. SPARGER AT 200 FT 8/12/83 * '
. SPARGER AT 200 FT 4/3/84 A
-.'" SPARGER AT 200 FT 3/18/85 x
VC REMOVAL ... SPARGER AT 200 FT 3/19/85 *
. -'
a a 10 12
AIR-TO WATER RATIO
ia
ia
Figure 33. VCX3 removal data for air lift pumping at 200 feet
while sparging at three depths.
i"*
1$
If
i -?
;.'i
ri
80
Ki:
-------
'v *-
TO-
1O-
O
PU
ao-
10 -
TO-
ao-
10-
TCE REMOVAL
a
B *
fi
C-1.2-DCE REMOVAL
fi
ง
SPARGER AT 130 FT 9/14/83
SPARGER AT 200 FT 9/29/83
SPARGER AT 200 FT. 12/1/83
SPARGER AT 280 FT 10/6/83
VC REMOVAL
o 10 . ia
A1K TO-WATER RATIO
1*
Figure 34. VOC removal data for air lift pumping at 285
- ,-.. ''..'.'' feet while sparging at three depths.
r i
i i
K, "J
fcr1 ^
'- '">
I* "^
t '^
f 3
I _t
-------
" -^ TABLE I/*. VOC REMOVAL WITH AIR LIFT PUW AT 200 FT,
'.'' K^-y'^V .(AVERAGE OF ALL 3 SPARGER DEPTHS)
':'. .....< '--' ':-. ' , . -.-;. '- -.--".- ' ' .'.--
; "AIR TO WATER
0 AVE
1.8
2.5
3.5
5.2
. 6.0
- 7.2
8.9
10. .
12.
15.
!7.
%RSD
7.9
-3.7>
6.2
6.3
V3.4
-2.3
7.7
1.2
1.9
0.69
1.2
N
' 3
:3
3
6
3
-3
5
3
3
2
2
TCE
. : % .REMOVAL
." AVE %RSD
39.6 . 3.04
; .29.7 19.1
50.0 18.4
59.4 26.7
66.5 12.4
75.8 10.7
71.5 20.6
65.2 .20.3
72.5 15.4
77.8 4.57
84.2 2.67
c-l,2-DCE ,
% REMOVAL ?.
.AVE %RSO .
24.7 7.49
26.9 17.5
37.0 23.6 ;
45.7 26.9
50.8 13.0
62.1 16.7
59.0 23.5
56.1 18.3 :
60.8 18.0
68.8 9.45
75.2 6.18
VINYL CHLORIDE
. % REMOVAL .-. ':
AVE
58.7
66.0
86.6
93.1
92.8
98.0
94.1
94.7
95.5
96.0
98.6
%RSD
9.11 ,
5.58
8.28
6.01
5.48
0.87
4.43
3.59
3.07
2.24
1.37
-
TABLE
15.
: VOC REMOVAL WITH
(AVERAGE OF ALL
AIR LIFT PUMP AT
3 SPARGER DEPTHS)
280 FT
T,
AIR TO WATER
AVE
. 2.1
3.7
5.4
7.2
7.8
8.7
11.
16.
%RSD
8.4
1;3
3.3
2.0
0.0
6.4
3.3
N
4
3
4
.*:''
2
3
3
1
TCE
% REMOVAL
AVE %RSD
23.4 62.1
34.5 17.7
55.7 6.19
61.4 7..64
67.9 4.71
73.1 12.2
72.3 13.4
88.4
C-1,2-DCE
% RB1QVAL
AVE %RSD
13.4 70.3
23.7 36.9
41.6 13.2
51.5 8.82 r.
58.2 2.66
62.8 11.0
63.5 12.9
83.0
VINYL
CHLORIDE .
% REMOVAL
AVE
42.7
83.6
92.0
95.7
97.6
97.4
97.6
99.7
%RSD
59.1
6.00
6.75
1.95 ;.-.
0.36
0.90
1.87
_
CONCLUSIONS FOR AIR LIR PUMP TESTING WITH A SPARGER
Operation of the sparger while air lift puiping caused the air lift
purp effIcIency to decrease. This was predicted by air lift purp
operational theory, because the density difference across the air lift
purp eductor Cthe oorp's driving force) was reduced by sparging. Even
though the sparger caused the air lift pump efficiency to decrease, the
pump efficiency with a sparger was higher than If all of the air had been
delivered to the air lift pump alone; therefore, In terms of pump
efficiency (and cost), It was better to operate an air lift punp and
sparger combination than to operate an air lift ptnp alone.
The maximum VOC removal percentages for the air lift purp and
sparger combination were 98.6% for VC, ..65.0% for c-l,2-DCE and 78.3% for
TCE. The air lift punp and sparger combination was able to produce a
higher alr-tb-water ratio (17:1) than the air lift pump alone (I1:!).
: ..' ':-' . ' 82 ,': '.. ''.. " .-' -"' -: .'
'i;.S
,v
'r?- *-
-------
r.
t;
-This accounted for:the higher VOC removal capabilities of the "combined
;i system> . When .sparging, the alr-to-water ratio was limited to the point
. :"at which -adding more air caused water to foam out.of the top of the well.
1-vhead.-;.In a well-.v .which, would make It suitable for an emergency treatment
technique. The following experiments using an electric submersible, pump
.and air sparger were designed with the goal oF Investigating a
short-term, emergency VOC treatment technology, -
83
' !m
-------
;.. , .;. SECTION 10
ELECTRIC PUMP TESTING WITH A SPARGER
ELECTRIC PUMP TESTING WITH SPARGER - PROCEDURES '"'' -
For this portion of In-well aeration testing, an electric
submersible punp was used In corb I nation with a sparger. Since the air
lift punp testing with a sparger did not Indicate that In-well aeration
would be an effective long-term treatment option for VOCs, this portion
of the study was designed with a short-term emergency treatment method as
a goal. In many cases, electric submersible pumps are already In place
In drinking water wells. The sparger which was used for this experiment
was of the simple, open-pipe, large bubble design used In air-lift
pump/sparger combine ".Ion testing. One Inch plastic pipe was used Instead
of 3A In, to reduce air delivery costs caused by higher pressures In the
smaller pipe. The electric pump was operated at one fixed depth (200 ft)
which was below the zones of significant water yield (see Section 5). The
sparger was tested at three depths (130 ft, 200 ft, and 280 ft). Figure
35 shows the electric submersible punp and sparger combinations used.
These were the same equipment depths examined In the air lift testing
portion of this study. The depths were chosen on the basis of well
characterizations discussed In Section 5. Installing and moving the
sparger was a very easy task and could be performed In a matter of
minutes, which was consistent with the goal of developing an emergency
treatment procedure.
As with the air lift pump testing, the electric pump experiments
were broken Into tests which consisted of several fifteen minute runs.
Each test was conducted with a single equipment configuration (I.e.,
electric pump at 200 ft with the sparger at 130 ft). Each air flow rate
within a test was called a run. Runs were fifteen minutes long. The
static water level was recorded before the start of each test. Next, the
electric submersible punp was turned on and the weir box was allowed to
fill. (See Section 6 for a discussion of the weir box and Its role In
water flow rate measurements). A raw water sample was collected from the
electric submersible pump at the well head. As noted In Figure 35, no
small electric sample punp was used, as was necessary for air lift punp
testing. In the electric punp testing (as well as the air lift punp
testing) raw water samples could not be collected while the sparger was
In operation. Because of this, the wel 1 had to be cleared of air between
each sample run so that a raw v/ater sample could be obtained for each air
flow rate tested. To clear the well, air delivery to the sparger was
stopped, and the submersible punp continued to operate. The punp
effluent was monitored for dissolved oxygen at the raw sample location.
84
-------
-T.NJ.
&t
n~~
f .'i
$.
M-
f -" RAW WAT ฃR 6AMPLE
P LOCATION
AT ISO FT. ,200 FT.
OR..Z80FT,
ฐo
o
00
o
-ELECT 1C 'PUMP
AT 2OOFT-
Figure 35. Electric pump and sparger configuration vised for
testing at well L-8. .
85
-------
a'
When'the dissolved oxygen of the water being discharged'returned to the
concentration at which It was before aeration, the well was declared
-clear of air. The raw water sample was collected, the next air flow rate
; for the sparger was started and the fifteen minute run began. Readings
of all of the air delIvery equipment, water flow rate, and well water
'level were obtained at five and f1fteen ml nutes Into each run. The
treated sampled was collected at the fifteen minute point from the water
discharging at the y-notch of the weir box. The water flow rate used for
most of the electric pump testing was 100 gpm. As with the air lift pump
testing, the limiting factor for the amount of air which could be added
through the sparger was water blowing out of the borehole at the well
surface. The water flow rate was lowered for later testing, to achieve a
higher air-to-water ratio. This effort was only partly successful
because the pumping water level was closer to the surface when less water
was pumped, so less air could be added before the water came out of the
'borehole. / .-- v - - . :.-,, "'.' .'.'...'-..-'.';'!.;'-?--'-. --.-'.." .--..' '.'"- -.' ", ".:'
ELECTRIC PUMP TESTING WITH SPARGER - RESULTS AND DISCUSSION '
.'-.:' The VOC removal results for the electric pump testing with a sparger
are shown In Table 16. Three contaminants (TCE, c-l,2-DCE, and VC) were
used to evaluate these tests. The maximum air-to-water ratio which could
be achieved before water blew out of the borehole was 8.2. Maximum VOC
removals were 82.6%, 79.9%, and 92.9% for TCE, c-l,2-:DCE and VC,
respectively. These removals were better than those achieved by the air
lift pump systems at the same air-to-water ratio. The variabilities of
VOC removal were generally greater than air-to-water ratios at the same
rotometer settings, Indicating that varying the depth of the sparger did
affect VOC removals.
. - TABLE 16. VOC % REMOVALS BY ELECTRIC PUMP WITH SPARGER
(AVERAGE OF ALL 3 SPARGER DEPTHS)
: % REMOVALS OF VOCs
AIR
AVE
0.00
0.80
1.5
2.2
3.0
'3.7
;.4.5
; 5.2
5.9
6.4
7.4
: 8.2
TO WATER
%RSD
0.00
0.00
0.00
9.2
0.0
0.0
4.4
3.6.
0.0
1.9
0.67
0.61
RATIO
N
6
3
2
5
2
2
6
5
4
3
2
2
TCE
AVE
14.7
18.5
32.9
49.8
53.7
62.0
68.4
72.9
70.5
77.3
79.4
82.6
c-l,2-DCE
%RSD
68.3
44.2
42.9
27.8
15.3
10.7
15.3
13.8
11.7
7.57
5.10
0.420
AVE
8.57
10.4
21.2
30.3
30.0
38.2
51.6
59.1 .
52.2
66.5
74.3
79.9
%RSD
97.0
33.6
65.6
55.9
43.7
26.2
26.4
19.7
18.5
15.1
3.90
0.500
VINYL
AVE
17.5
48.9
78.3
85.8
92.7
91.0
91.5
92.7
92.2
92.2
94.5
92.9
CHLORIDE
%RSD
60.3
8.13
9.45
6.00
1.29
4.23
5.00
1.92
3..3S
. 1.85
0.00
0.650
86
-------
t!;
,> - ": Figure 36 shows VOC removal results with theelectrlc pump at 200 ft
and the sparger set at 130 ft, 200 ft or 280 ft. The results Indicated
that the best removal was obtained with the sparger set at 130 ft. These
findings were consistent with what could be expected In terms of the well
characterization discussed In Section 5. It was determined that most of
the VOC contamination was entering the well In the two zones between 106
ft and 1*ซ0 ft. These zones also produced nearly 60% of the water
entering the well. Very little water (less than 10% of the well's yield)
was entering below 200 ft, therefore, most of the water movement In the
well was down from 106-140 ft toward the pump at 200 ft. With the
sparger at 130 ft, a counter-current stripper could have been created
with the air being delivered just at the bottom of the most contaminated
zone. This would be an efficient way to strip the VOCs, and Figure 36
shews the best TCE removal with the sparger at 130 ft. When the sparger
was set at 280 ft, this created a co-current stripper through a very
small portion of the total yield of the well. A good deal of the air
could have been pulled Into the pump before It reached the more
contaminated water entry zones. When the sparger was adjacent to the
pump at 200 ft perhaps a portion of the air was pulled Into the pump
before any stripping could occur In the well. In an emergency situation
a well characterization would not need to be done. The sparger should be
placed above the pump or below the pump but these tests Indicated that
the sparger should not be placed directly next to the pump.
In order to look at the reproduclbi1ity of electric punp and sparger
treatment, the test of the sparger at 130 ft was repeated In October,
1984, five months after the original test at 130 ft. The results were
vastly different, as can be seen In Figure 37. The weir box and air
delivery systems were all recalibrated to determine If any air or water
flow rates were Incorrect. No Inconsistencies were found during the
recalIbratlon. The quality control program for sampling and analysis v/as
stir* !n place and gave no Indication of any analytical problem. It was
decided to do two more tests with the sparger at 130 ft, two days In a
row. These results for TCE are shown as the January 23 and 24, 1985
curves In Figure 37. As can be seen In the graph, the results of tests
run two days In a row closely matched each other, they also matched the
second test of the sparger at 130 ft, conducted In October, 1984.
Because no cause for these variations could be found In the measuring
procedures, It was determined that variations In VOC removals for the
same In-well aeration configurations could have been caused by well
changes over long periods of time. Conclusions based on the curves In
Figure 36 are considered to be valid because those three tests were
performed within one month of each other. It would have been Impossible
to determine exactly what was changing In the well without repeating the
well characterization Clogging and packer testing) described In Section
5. It was known that the raw water VOC concentration and static water
level of the well changed with time. Perhaps water entry zones or VOC
entry zones also changed with time. Whatever caused the changes, It was
clear that In-well aeration, both with the air lift punp and electric
pump with a sparger, provided variable results over a long period of
time at NPWA. The same may be the case In other locations, depending on
the hydrogeology.
It was difficult to see any difference In vinyl chloride removal for
87
-------
Ej
H
H
TO
aO
ao
to
3O -
100
ao -
so -
TO -
ao -
ao -
4O -
ao -
ao -
10 -
o
. TOB ซ RBMOVAI-
F-AROER 1SO FT
apAROER aoo FT o/a*/a<ป
af*AROER aao FT. 10/3/0*
D
O 1.9 DOE X REMOVAL.
VO X REMOVAL.
AIR TO WATER RATIO
Figure 36. VOC removal by electric submersible pump and
sparger combination, with sparger set at three
different depths. Electric pump at 200 ft.
-------
'rS-rr^
1OO
7O -
00-
30 -
ao-
10-ii
100
a
iv
oo -
ao -
ro -
40
SO
2O
1O
103
BO
O
7O
eo
BO <
4O
SO
ao
10
o <
TOE a REMOVAL-
-1.2DOC SS MCMOVAU
* a
VC K R1CMOVAU
o a 4
Am^TO-WATtm RATIO
0 5/24/84 "10/24/84' * 1/23/85 Al/24/85
Figure 37. Results of duplicate testing with electric
submersible pump and sparger showing variation
over five months time. Electric pump-at 200 ft
and sparger at 130 ft.
89
X:
:':- --x'
-------
*it
the four repeat tests performed with the sparger at 130 ft Csee Figure
-37). .As with the rest of the In-well aeration tests, vinyl chloride was
the most easily treated, with the removal curve rising very sharply, then
.leveling out Into a plateau. .'.. _ 'i ' ;.--
In Figure 37, the tests conducted in January, 1985 were performed at
60 gpm, while the two earlier tests use a flow rate of 100 gpm. While It
was possible to Increase the maximum air to water ratio from 5.9 to 8.3,
the amount of air which could be added was still limited by water coming
out of the borehole at the surface of the well. For all compounds, with
the exception of c-l,2-DCE, It did not appear that raising the air-to-
water ratio beyond that achieved In this testing would cause any further
removal. This Is said because the removal curves have reached the
plateau, which Is normally seen In aeration curves, where adding much
larger amounts of air produces very little further removal. In Figure
37, It can be seen that c-l,2-DCE did not reach that plateau. This
compound has the lowest Henry's Law constant of all the compounds
examined. If some way could be found to increase the air-to-water ratio,
more removal could be expected for this compound.
Costs for the electric purp and sparger combination are shown In
Figure 3?. The cost to compress air was calculated and 3$ per thousand
gallons was added as an estimated cost for electricity to operate the
punp. As with previous cost discussions, these numbers do not include
operation and maintenance or amortization of capital expenses.
: The most costly electric punp and sparger combination tested wฃ.-
wlth the punp at 200 ft and the sparger at 280 ft. The cost approached
15$ per thousand gallons with an air to water ratio of 6 to 1. Very
little difference In cost could be seen between the 130 ft and 200 ft
sparger depths. The maximum cost for these depths was around 12ฃ per
thousand gallons. The cost of this method, together with the ease and
rapidity of Installation, showed that electric pumping with sparging was
a viable emergency treatment technique for VOC removal, providing the
limited removal capabilities would meet the necessary effluent
concentration requirements.
CONCLUSIONS FOR ELECTRIC PUMP AND SPARGER TESTING
VOC removal differences were seen when the sparger was operated at
different depths relative to the electric submersible pump which was set
at a depth of 200 ft. The VOC removal findings were consistent with the
well characterization findings. Sparging at 130 ft provided the best VOC
removal. This depth would have produced a counter-current stripping
effect as water from the most contaminated zone in the well was pulled
down past the sparger on Its way to the purp. With, the sparger at 280
ft, a co-current stripper was produced as both bubbles and water
travelled up toward the purp with some or all of the air being pulled
into the purp before It reached the most contaminated zone. When the
sparger was adjacent to the punp, at 200 ft, most of the air was likely
to have been pulled directly Into the purp, before any stripping could
occur In the well.
The maximum VOC removals obtained by electric purplng with a sparger
were 82.6% for TCE, 79.9% for c-l,2-DCE and 92.9% for VC, at a maximum
.air-to-water ratio of 8.2:1. These removals were better than for the air
90 -'..
-------
If
2
C
<
G
0
C
0
0!
It
a
i
2
It
C
10-
r 12-
11-
10-
-
ป 9-
i
1 Q_
'' ' 7-
i
fป
6-
r 0
4-
3-
2-
1-
l
, ' " - - ' '- -" '.,,-"' l '.-."
"- "".-""ป'- ' . ' - " ' - :"..'"
/^':':^J^r^y^-'/-f^ji
'.--" . " ' -. ','''*"--- CH - * ' . ' '"
' ' '"" ;' J "-'.'. ," " "' ' . '. '"<-->.
-,"'-,' "i - ','.--' ','.'", ."'.".
-.. : - ' , '' /.:..=v-.;.; , o:',V;-' n -;';.': -. .;.'; ':;-v
. ' "'.-'' *.- '.. .
"'." .'':-':' -; ^:,'xv- ft '.':'."'" '". : :' '; -:' /
'' ' ' A ' ' ' .'''..-".''-
-.-'-. .-.* ..-. n '.. .;. _.--, ;.''
' '-'. . ' ' V'- ' :;'n'- '.' ': ".;>-i"' '':' ' : 'V^V":> ' . '
i .' ' ' ' ' . '
'" " . - '
D
* SPARGER 130 FT 10/24/84 = D
t .''.'
SPARGER 200 F 9/24/84 = +
SPARGER 280 FT 10/3/84 = ^
it i i i i i
) 2 4 6 fi
AIR TO WATER RATIO
Figure 38. Costs to operate electric pump and sparger for
in-well aeration at well L-8. Electric pump
at 200 ft. .
91
-------
'<;:'" 11ft pump systems at the sane air-to-water ratio; however, a higher air-
/^^to-Water ratio was obtained with the air lift pump (with or without a
''Ji.--:sparger).. '/-,- ,'.- . ......- -. ' '.'..,.- -- '-';.,.- . :.'". '-':.-.'
\'.':'.; The electric purp with a sparger would be a good emergency treatment
;T technique If Its limited VOC removal capabilities will meet the necessary
.water quality goals. It was easy to Install and Inexpensive to operate
(up to 15^ per thousand gallons for air and electricity). While well
..". .characterization was helpful for experimental design and data
: Interpretation, It would not be necessary In an emergency situation. The
sparger should, not be set directly adjacent to. the pump. In-well
; aeration by the electric submersible pump.and sparger combination can be
''', .used to keep a contaminated well In service while permanent treatment
, systems are being designed and Installed. ,: .... ; .
92
-------
**"
*ป
1
" ฃ *
""~x
SECTION 11
SECONDARY EFFECTS OF IN-WELL AERATION
GENERAL
Part of the scope of this project was to examine secondary effects . ... _....
of In-well: aeration. Dissolved oxygen, pH and temperature were ^ ;5
determined In order, to discuss corrosion Implications. Bacteriological . -C?j
tests were conducted using both the heterotrophlc plate count (HPC) and : *?
the R2A plate count procedures. In addition, air samples were analyzed. ' *,;
at the well head, at the weir box, and at the sidewalk In front of the "-?
well house. These studies were not the major emphasis of this project , ;
.and they will be presented In a brief fashion. . ... " . ... ~ ..'-?-'-'. v.
DISSOLVED OXYGEN, pH AND TEMPERATURE .,. : -;: I .v"'.. :'"'-'.'"{ y; '",vr|
Dissolved oxygen concentration and pH of drinking watซr In the ''
.distribution system may have an effect oh corrosion of water mains. ".!
High DO and/or low pH can cause water to become aggressive, thus :,:
shortening the useful life of various piping materials. .';':
Dissolved oxygen and pH samples were collected In BOD bottles at the
raw water sample tap and at the v-notch of the weir box. The samples - :,
were analyzed Inmediately In the field for DO, pH and temperature. DO ....-.:
was measured using an Orion Research Model 97-08 oxygen electrode. The :
pH and temperature were obtained with a Corning Model *t portable :
pH/temperature meter. -
The results of the pH, DO and temperature testing are shown In Table ';
17. The mean values for all of the samplt-s collected at the raw tap and . ?
weir box (treated) have been calculated. The table also shows the i " ".'*
standard deviation (SD), number of samples tested (n), and the %RSD. The ..--"
range of values obtained throughout the testing Is also shown. . ' ">
In over sixty runs, the value of pH was found to Increase with i . '.-.'
In-well aeration 100% of the time. The average raw water pH was 7.2 and - ".
the average treated water pH was 7.6, yielding an average Increase of 0.4 - '[
pH units. The variation In pH was higher for the treated water (%RSD = . . ' Ji
2.75) than for the raw water (%RSD = 1.62). This Increased variation In ;: .^
treated water was due to the fact that these numbers represented pH /
values from the entire range of air-to-water ratios used during In-well ^
aeration testing. The raw water could be expected to show less ...;_]
variability. In very general terms, an Increase In pH, as seen here, ; |
could help prevent corrosion In the distribution system by pushing the 5
Langelier Index to a more positive value. - .:;
The average dissolved oxygen concentration for the raw water was . .. . :.'
/--' .:''.-.. 93 .""-.-.:' .-:'.' '".'.. .:"'..... ;" :: ::=
-------
2^?^M;^^^^
TABLE 17. SECONDARY WATER QUALITY EFFECTS OF IN-WELL AERATION
Raw
Treated
, Ave
7.2
7.2
SD
ฑ 0.12
ฑ 0.21
n.
63
65
%RSD
1.62
2.75
Range -:
7.0-7.5 ' - '.'; ---.'."".' '.
7.1-8.0 ; ' "..-;.
Dissolved Oxygen-mg/L '
Raw 2.17*1.15 78 53.0 0.9-7.5
Treated 10.8 ฑ 0.52 78 4.78 9.9-13.0
Temperature - ฐC . '''".:"-.
Raw 15.8 i 1.28 69 8.03 13.0-18.4
.Treated . . 15.1 * 1.27 72 8.40 12.7-17.3
. Bacterlologlcal-counts/ml
. Method Ave SD n %RSD Range
Raw HPC 73 * 61 21 84 2-243
Treated HPC 71 ฑ 79 21 110 1-282
Ave % Saturation
22%
100%
Raw
Treated
R2A
R2A
638 ฑ 540 12
647 i 430 12
85
66
130-2100
70-1370
2.17 ppm over 78 runs. The average dissolved oxygen concentration at the
v-notch of the weir box was 10.78 ppm. With average water temperatures
of 15.8 and 15.1 at the raw and weir sample collection points, the
dissolved oxygen concentrations obtained represent an Increase In
dissolved oxygen saturation from 22% to 100%. This increase In DO to
100% saturation occurred for the entire range of ^lr-to-water ratios
tested. No bubbles were present In the weir box by the time the water
reached the v-notch. Water that Is saturated with dissolved oxygen has
the potential to be more aggressive than a lower DO water. This could be
especially harmful In systems with unllned, Iron pipe. Since the
conclusions of this study show the best use for In-well aeration Is on a
short-term basis for emergencies., no treatment to reduce oxygen levels
should be necessary.
The bacteriological testing during In-well aeration experiments was
Inconclusive. The results In Table 17 showed that both the heterotrophlc
plate count and the R2A agar plate count methods revealed no difference
In bacteriological quality of the water before and after In-well
aeration. This was to be expected, because the samples were collected
Immediately after treatment, and any bacteria present would not have had
time to be affected by the changes In conditions. This should be
Investigated further If the water enters a distribution system where the
bacteria have more time to contact the water with higher pH and dissolved
oxygen at saturation.
The bacteriological testing did show a significant difference In
recovery of bacteria between the two methods used. The counts of
bacteria per mi 11II Her were approximately 9 times higher with the R2A
method than for the HPC.
N
94
-------
'.;'Further study Is needed to detennlne the effects of In-well aeration .- : ':
on corrosion and bacteria In a water distribution system. Dissolved ',
oxygen and pH, both factors which can Influence corrosion end \ ,
bacteriological growth, are clearly affected by In-well aeration. - ./ . "
AIR SAMPLING . ', '>;' ' . -;: '. . / ;. ;- -'"'. '- ' .">"''.. .; "":. ':'~. ' "'',;_'. J''^' /'%
.-'. Air samples were collected during four In-well aeration tests to - ,.:
determine possible exposure hazards In the area around the well house.
One method- used was an adaptation of GC purge and trap techniques "
C35,36).. A second survey was done using a portable photoIonIzatIon .
detector. The air tests were Intended to be a preliminary estimate of
air concentrations to determine whether further consideration of air . ;
"exposure would be necessary. . . ..- . , . - ',-'.
Trap and Pesorb Method - Results and Discussion -'...' . . ; ;. :
Volatile organic chemical traps from the GC purge arid trap sample
concentrator were used to collect air samples. The traps were packed '.-
with Tenax, silica gel and charcoal. To clean the traps In preparation
for testing, they were baked for one hour In the purge and trap
apparatus. After the bake cycle they were desorbed Into the GC for
analysis to confirm that no organlcs were remaining on the trap. Once
cooled, the traps were removed from the purge and trap device and the
ends were Itimedlately sealed with parafln film. One sealed trap was
carried Into the field and returned to the lab without breaking the seal,
to act as a field blank.
The configuration of the air sampling system Is shown In Figure 39.
Two of the laboratory prepared traps wore hooked together with fittings.
This assembly was performed In the field immediately before the sample
was collected. Two traps were used In series In order to provide a
primary analytical trap and back-up trap to test for breakthrough of the
primary trap.
The traps were connected to an Industrial hygiene sampling pump.
The outlet of the pump was directed Into a bubble flow mettr to measure
the air flow. This was necessary because earlier laboratory tests showed
that each combination of traps produced a si Ightly different flow rate
for one particular setting on the sample pump. .
The system was calibrated by purging an aqueous standard onto the
Tenax/slllca gel/charcoal trap and desorblng as for a regular GC analysis
of water, however, Instead of calculating as ug/L of compound in the
aqueous sample, the GC was calibrated to the actual nanograms of material
adsorbed by the trap. When the traps from the field were desorbed, the
results were reported In nanograms. The field measured air flow rates of
sample collection were then used to calculate aJr concentrations. An
example calculation to determine volatile organic chemical concentration
1n .aIr Is shown be1ow: , ' ..
1. Determine cubic meters of air collected from field flow 1
data which is measured In liters/minute. I
- .;' -.'-.. ' ' ' ;"-.-, .';:. I
' J' .. - - ' 95 ' ' ."'-.'-/'.!
-------
m
w
BUBBLE
FLOW
METER
SAMPLE PUMP
Figure 39. Drawing of air sampling system vised to test off-gases
of in-well aeration.
96
^^&^li^aliSl^^ฃ^^
-------
collection time (mln) x flow rate (L/mln) = Liters air
sampled
Liters air sampled = cubic meters of air sampled
1,000 L/m3
2. Determine mlcrogram/cubic meter concentration In air.
.. ng VOC col 1 ected / cubic meters of air sampled = . . ."-;''
; . 'V 1,000 ng/mlcrogram mlcrograms VOC/cublc meter air
;;. : 3. Convert to parts per billion VOC In air . . ,1 '.
. --.." ppm- (0.02445 cubic meters air/mole air) x .;/>'
. (x ug/cublc meter VOC In air) .
:';.; molecule weight of VOC, g/mole
V ppb = ppm x 1,000 .;..: ._.:.-.-,',. ' :. . ;
..:' Sampling locations are diagramed In Figure 40. For this method,
.'sampl es were collected directly at the well head, above the treated water.
sample location of the weir box and at the sidewalk In front of the
pumphouse. The samples were collected while operating the electric pump
at 200 ft In the well and with the sparger delivering the maxlnnum
possible air at 130 ft.
The results of the air sampling by trap and desorb method are shown
In Table 18. In both samplings, concentrations were very high at the
well head, lower at the weir box and nearly zero at the sidewalk In front
of the pumphouse. There were several problems with the method. Each of
the back-up traps showed some contamination. (1% to 5% of the main trap
concentrations) by the major compounds. In the second sampling, the
field blank was contaminated (see Table 18). Even with these problems,
It can be seen that the VOC concentrations In air may be reason for
concern, and would require further study by a standard method for air
analysis. The Pennsylvania Interim Air Standards (37) for TCE and PCE
are 1200 ppb. The well head sampling exceeds or nears these limits In
both cases. The VC Interim limit is 2.4 ppb, 1,1-DCE is 37 ppb and CCL4
Is 12 ppb. All of these are exceeded at the well head by this method of
TABLE 18. RESULTS OF AIR SAMPLING BY TRAP AND DESORB METHOD
FIRST SAMPLE COLLECTION
Air Flow
Col lection Time
VC (ppb air)
1,1-DCE "
c-l,2-DCE "
'CHCL3 "
11,1,1-TCA "
CCL4 "
TCE "
PCE "
Well Head
21.70 mL/mln
30 min
810
200
GT3,600
0
120
14
GT4, 200
GT1,900
Weir Box
15.06 mL/mln
30 mln
0
5.3
55
0 ' .
5.3
0
130
13
Sidewalk
19.40 mL/mln
30 mln
0
0
0
0
0
0
1.0
i 0
Blank
no al r
sampled
0
0
0
0
0
0
0
0
i(continued)
97
fe^&a^^
-------
SECOND SAMPLE COLLECTION
. '* v. , . Well Head
Air Flow
Collection Time
VC (ppb air)
1> 1-DCE " .
C-1,2-DCE "
CHCL3 H
1,1,1-TCA "
cent '".:
TCE "
PCE "
20.02 mL/mln
15 mln
21
1,900
" - - 0
160
'- 35 .
GTS, 700
700
Weir Box
15.62 mL/mln
15 mln
2.7
0
11
0
7.5
0
.57
Sidewalk
19.36 mL/mln
15 mln
0
0
3.3 .;
. o ../:
0
0 -.''.
8.2
1.0
':'-. -.:>' .
Blank
no air
sampled
:.. 0
10
0
1.5
0
12
0
,GT = Greater Than . . '-'
0 = Less than 1.0 ppb In air ."''; , -"''":
sampling and analysis. Air discharge permits and/or air VOC control
measures might be needed for wells as heavily contaminated as this one.
PhotoIonIzatIon Detector Method - Results and Discussion
.'-'- Another type of air survey was completed using an HNU Model P1101
PhotoIonIzatIon Analyzer. This Is a portable Instrument which samples
air and Instantaneously reads total VOC concentration In parts per
million as benzene. The air was tested at the well head, above the weir
box and at the sidewalk In front of the well house, Just as with the trap
end desorb method. In addition, the air was tested two feet directly
above the well head and at a workbench In the well house approximately 8
ft from the well head. The latter two locations were selected to see If
workers who were In the well house would be exposed to dangerous levels
of VOCs In the air.
The HNU analyzer survey vies done while operating the air lift pump
at 200 ft with the sparger at 200 ft. Air was tested with the sparger
off, then with the sparger blowing the maximum amount of air C50 cfpm).
The results show no response on the analyzer at any of the sample
locations when the sparger was not operating. The lower detection limit
of the Instrument Is 0.2 ppm as benzene. With the sparger operating,
there was no response at the work bench, weir box or the sidewalk sample
sites. The weir head gave a response of 6.2 ppm and the sample two feet
directly above the well showed a response of 1.8 ppm. Since workers
would generally not be In the Immediate vicinity of the well head for
long periods of time, the operation of this In-well aeration system would
not pose a hazard to workers. In spite of these results, however, It Is
reccmnended the.t the well house be well ventilated when using an-In-well
diffused aeracor to avoid build-up of the VOCs In the air. These samples
were collected with well house windows and doors closed to try to
simulate a "worst-case" situation.
98
'^N '
-------
I
:rffi
WELL HEAD
WORK
BENCH
WEIR BOX
V-NOTCH
6IPEW
ALK
3rd STREET
Figure 40. Air sampling locations for in-well aeration
off-gas testing at well L-8.
: 99 '.''.
&***&.
-\
-------
$ _'
f'
fc i
:^ '.;-:'' All, methcxls of: In-well aeration tested Increase the pH of the well
"water by an average of O.1*. pH units. .Dissolved oxygen was raised to
/saturation by all of the In-well aeration methods tested. The oxygen
Increase could cause water . treated by this method to become more
corrosive. This should not be a problem for short-term use In an
, .emergency situation. ",..-.: . . .' '-. - .''; :; , :
'Water entering the weir box after sparging was milky white In
appearance. By the time It left the separator It was clear, causing no
'problems with aesthetics. . '. ,_ . ;..":. .:' ': .- '':
V; : The bacteriological testing of raw and treated water was
Inconclusive, with large variations In bacterial counts masking any
trends. The R2A method provided consistently higher recovery of
; organisms than the heterotrophlc plate count. . .
Air sampling showed that In-well aeration would probably not cause
air quality problems of Industrial hygiene concern, however, the well may
be considered an air pollution source, and would require any appropriate
permits for such a source. This would depend on the raw water
concentrations of the contaminants and on air flow rates used. It would
be prudent to ventilate a well house with an In-well diffused aerator In
'operation.' - -.- .; . . - ' : -::.;-- ;. -."-".' .'}.-.,," '"..-.'.. '
100
0eu&ir^^
-------
, SECTION 12
COST ESTIMATES
FIRST SCENARIO ; . ; ..'."-. . :
In this case, costs have been computed for an In-well aeration
system to be used at a well with an electric submersible pump In place.
The well is 500 ft deep with a static water level of 25 ft and a pumping
water level of 100 ft. The water flow rate Is 70 gpm. A sparger will be
added to the well for use with the existing submersible pump. The
sparger will be set at 175 ft. The pump v;Ill remain at 200 ft. The
system will be In service for approximately three months.
The VOC concentrations are 10 ug/L of TCE and 250 ug/L of 1,1,1-TCA.
It would require 50% removal for TCE and 20% removal for 1,1,1-TCA to.
bring the VOC concentrations below the limits. In-well aeration would
produce the necessary removals. .
Cost to rent air compressor (3 months)
1 In PVC plpe/flttlngs/valves
Fuel tank (250 gals)
Air filter
Well vent
Installation labor
Operating Costs -
Oil and Filters
Gasoline (2 gal/hr)
Labor
Electricity (3^/1000 gal/wtr ptnped)
Electric Pimp and Sparger
Total Cost-3 months operation
Cost per 1,000 gallons of water treated
SECOM) SCENARIO
$ M92
200
500
175
. 300
2,000
$ 7,267
$ 500
4,368
5,460
281
$10,609
$17,876
$ 1.91
'In this case vie desire to bring an old well back Into service.
There Is no punp presently In the well. The well Is 300 ft deep with a
static water level of 45 ft and a pumping water level of 70 ft. The
water flow rate Is 70 gpm. An air lift purp and sparger will be
Installed. The air lift purp footplece will be at 200 ft to yield 65%
submergence. The sparger will be at 250 ft. The system will be in
operation for three months.
101
-------
: .>-..': ./'.' .'The well has a TCฃ concentration of 20 ug/L Cneed 75% removal), a
,;-.; PCE concentration of 5 ug/L and a VC concentration of 5 ug/L Cneed 80%
-.'.-/"''-''":. removal).' The air lift pump and sparger, combination should provide the
.-.',' necessary VOC removals. . .,.-.'. ;; : .. ;.'.---''.' '-;/>..: .-.
Cost to rent air compressor , $ 4,092
/ '.:."". _2i In steel pipe Ceductor) : 390
Purp footplece Cmaterlal 6 fabrication) 200
... . . ;Alr line Cl In) forpump . 150
-'."-" . ' 1 In sparger pipe . ,250
" . . . Fuel tank C250 gal) - /" 500
:',.' ; Air filters . "..-'--. 175.
.- . ^ Well vent . 300
. .Crane rental Cto Install punp eductor) 500
. : Installation labor 3,000
/';.'' - ' $ 9,557
- " Operating Costs -
< Oil and filters $ 500'
Gasoline C2 ga^./hr) 4,368
Labor . ' . . ; . 5,^60
' ' - $10,328
Air Lift and Sparger -
Total cost - 3 months operation $19,885
Cost per 1,000 gallons of water treated $ 2.12
GAS AND WATER SEPARATOR ' ; .
If the well being treated does not have a previously existing
chlorine retention tank, one will have to be Installed to serve as the
gas and water separator. Costs shown here do not reflect the costs of
additional chlorine, which might be necessary because of, stripping as the
bubbles are released.
Concrete tank (1,500 gal) $ 2,000
Punp Cto system) 1,000
Electrical control equipment 1,500 .
Installation 2,000
. - $ 6,500
DISCUSSION . .
The total cost of an electric submersible pump with a sparger was
less than an air lift punp and sparger, as long as the electric
submersible punp was previously In the well. These costs did not show
purchase or rental of an electric submersible punp.
In-well aeration may not be a good treatment option If a gas-water
separator Is not already present. A chlorine retention tank would
probably be Included In the construction of the permanent VOC treatment
facility. The time Involved In Installing the tank would probably negate
Its usefulness as an emergency treatment measure.
102
-------
X
REFERENCES
Woodhall, Richard S., "Groundwater Contamination In Connecticut",
AWWA Annual Conference Proceedings, Atlanta, GA, 1980.
Massachusetts, Ccmnonweal th of, "Chemical Contamination"; Dept. of
. :V Environmental Quality Engineering, Working Paper for the Water
. ". Quality Task Force of the Special Legislative Ccrmlssion On Water
.;; Supply, September 1979. v : ? ^ . . ,i3 :.
3. , Cl Ine, Gary C., Thomas J. Lane, and Mario Saldamando, "Packed :
Column Aeration for Trlchloroethylene Removal at Scottsdale, .
"v 'o. Arizona", AWWA Annual Conference Proceedings, Washington, D.C.,
;'..;-,. 1985.;.._ _ :-,.;,. ;.... -' r .; ' . : :\^.,-; \: ,_:. ^ . ^.-:f, ^-,' C- ^
4^ Cumnlns, Michael D., "Removal of Etyhlene Dlbromlde CEDB) from
'-'-' Contaminated Groundwater by Packed Colum Air Stripping", U.S.
:" ;EPA Off Ice of Drinking Water Report, August 198
-------
f. : - '1 11. Suffet; I.H. (Mel), Jacob Gibs, Judith A. Coyle, Robert S. Chrobak,:
H ' .:. : -. '-.- ."and Thomas L. Yohe, JAWWA, Vol. 77, No. 1, 1985, pp. 65-72. V;
. ' ''.'.- '" -. 12. Coyle, Judith A., Harry J. Borchers, Jr., and Richard J. Mlltner,
' . ."'-.. .:- "Control of Volatile Organic Chemicals In Groundwater By In-Well
': : Aeration", AWWA Annual Conference Proceed I ngs, Washington, D.C.,
; '/' '", 1985, pp. 1101-1114. . ,. '. .-.,.. . ;;-:\ -; "'?: ..;
:"''. 13. Rlma, D.R., "Groundwater Resources of the Lansdale Area,
'-;-. ... Pennsylvania", Comnonwealth of Pennsylvania, Dept. of Internal
.Affairs, Topography and Geologic Survey, Progress Report No. 146,
( '' - ' '; 1955. .-- ..,.-.;... .. ' -; /: . , .'.
14. Longwlll, S.M. and C.R. Wood, "Groundwater Resources of the .
'-..- Brunswick Formation In Montgomery and Bucks Counties, Pennsylvania",
- .'-. Comnonwealth of Pennsylvania, Dept. of Internal Affairs, Bureau of
: . 'Topography and Geological Survey, No. 7, 1965. , - .
15. Itorth Perm Water Authority, Well L-8 Permanent File, Driller's
;''.' . ..,- Report, 1923. . ^..-;. - .-'.'
16. Federal Register, 40CFR141, Vol. 50, No. 219, 11/13/85, p. 46880.
/ . 17. Wright, R.E. Associates, Inc., "Special Groundwater Study of the
Middle Delaware River Basin, Study Area II", Vol. 2, Delaware River
Comnlsslon, Trenton, N.J., 1982.
18. Johnson, A.I., "Geophysical Logging of Boreholes for Hydrologlc
Studies", U.S. Geological Survey Open File Report, Hydrologlc
Laboratories, Denver, CO, 1963.
19. Keys, W.S., "Borehole Geophysics as Applied to Groundwater'',
Geological Survey of Canada, Economic Geology Report No. 26, 1967.
: 20. Papadopulls, S.S., J.D. Bredenhoeft, and H.H. Cooper, Jr., "On The
: Analysis of Slug Test Data", Water Resource Research, Vol. 9, No. 4,
. / !. - ',- . .-''-. 1972. - .'....:.-
;"'. 21. Moorshead, Tucker, "Hydrogeologlcal Investigation of Well L-8,
Lansdale, PA", Project Report to NPWA, Earth Data, Inc., St.
- Michaels, MD, 1983.
22. Shuter, E. and R.R. Pemberton, Inflatable Straddle Packers and
1 " Associated Equipment for Hydraulic Fracturing and Hydrol.oglc
Testing, U.S.G.S. Water Resources Investigation 78-55, 1978.
'',', 23. Cherry, R.M., "A Portable Sampler for Collecting Water Samples
from Specific Zones In Unusual or Screened Wells", Geological
Survey Research, (U.S.G.S. Professional Paper 522), 1966.
104
-------
37.: Pennsylvania Interim Operating Guidance For Air Toxic Substances:
--^.' New and Modified Sources, l27.12CaX2>:9, Revised 9-27-85.
j-
38.* Kent, Robert T., edi> Kent's Mechanical Engineers' Handbook,
El eventh Ed 111 on, John Wl1 ey g Son, New York, 1948.
-39." Seelye, Design, Vol. 1, 2nd ed., John Wiley S Sons, NY, 1953.
*J _
105
-------
APPENDIX A
SUWARY OF EQUATIONS USED
fe:
1. AIR' DELIVERY - Equation used to adjust for differences from
calibration temperature and pressure for air rotometers (supplied by
Brooks Instrument Company): . . ..'...: .
rotometers
Qact = Q
!Pcal
. Where - Qact = Air Discharge, SCFH ...
. -, Qroto = Rotcmeter Reading "'.'
" - Pread = Pressure at Rotometer + lk.7 psl (absolute!)
''>; :. ,;':, Peal = 75 psl + 14.7 psl = 89.7 psl
'. .Tread = ฐF Air at Rotcmeter + H60ฐF ' /.'.'
'.; -.":''.--.- Teal = 70ฐF + A60ฐF = 530ฐF . .'.' . '\ , .'
2. SUBMERGENCE - percent of air lift pimp eductor which was below the
pumping water level (reference 29}: '-..-.
S = L - h x 100
Where - S = % Submergence
L' L = Distance from footplece
.'". . to discharge pipe
'-. . h = Distance from pumping
. water level to discharge
pipe
3. V/IRE-TO-WATER EFFICIENCY - derived fran water horsepower and required
compressor horsepower (reference 38):
HP =0
w -*w
h x 8.3U5 Ib/qal .= 0 . h
33,000 39^5
Where - HP = Water Horsepower
0 w = Water Flow Rate (gpm)
h = Distance from pumping water level
discharge pjpe
to
x 0.01511 x P.
29 -
Where - HP = Adlabatlc A!r'HorsepOwer
PI = Absolute Pressure of Free Air
" P9 = Absolute Pressure of Air Supplied to
Q7 = SCFM Air Supplied to Pump
106
Pump
-------
;.:'-.,. :" '-.v.'i:":-'"---i-t'' Horsepower-for. Electric "/" ,;.'"ป\ r-"C5'..-ฃ-ฃ>'f
.',, /.V-Vip'' ^~';'-).:'.-V'G Motor' ""'.-.:>~.^^-;~~~^^J,'";:-lf^:'^ฃ^fฃ^:^
'-',^-; - - c.*V /'.-' ^^ -."' -=- 'Compressor Ef f I c I ertcy ;:.' [i-;':' ^'.^^''^-''^
A.-':;'1 - /XX'* E ;=' Motor Efficiency - '-.-'-. V-v-e:V ^&^;^t:'Z&
Efficiency.
r?'-1?--."--"-'1'^ "'-'."'' '"-.'.'". ".:.-.;'/>" ;'.'/-'-vv';--V-'v?S<:;,^''U,r' ;-^';,;--'?-rni-':-:v.: s ;'^
. Where - E = Estimated WIre-to^ater: : . -,. ' :-; "0 i'';|
:;;:,: :',\;;-<;-,- ^:ฃ*';-w%['-":' .:'.-:"-'"..'.'-'.-v.-.V Efficiency; ^.^^^//:-:-4i->^';^j^^^
:':--''V ; "'';'f'-;- --'-^V. :":-^" .:"i :''.':'.'.' " -.'"". HP."'='Water .Horsepower ''. '..{"_"/'''/ :":%".'i^'^jj
;.,.';.'.;;. -'-'-'^i -;.;. ["il"-:";':.;--'''vX:',.''V.'"'.; ' ; .Hpc = Required Compressor . v'.';-;.^-':l^^"^>-ป^;'-|
:/',,-.":;;;-' -i-:-VT%^:./;.> 7 ^.'.:/^v:v ;-';.;/,;'- -,- ; >:>. .. ^Horsepower -/ '.;; '""/ vV^->;-C-5:--:Sv^^_;|
U. .- ATR-TO-WATFR RATIO- " :; ' ' '"':-" ' '' " '" .":'. -"'"?; {'.^:/iT/^-S ;:-'"^\^^^>^^ฎ:?^^i;.^i'^
-E . .. = HP
,-: -j 'H -;V- =
,v.: /;;;=:'.: V*" "; .
"
x 7.^8 gal/ft3
gal/mln
' '
-,'^.:-/'- ;t:; r - Where -r V. = Vo1uiป of Air '--. '^;^''-: V-:'r:;. -:^ V'iv^;
J'-.^::-'--^":--:-:^ '":'-,.^:- Voluns of Water ' ; - '.-";>-;"/'-: 1: ^,:'" ^V^v-'-:^
' '"*-- '' '"'.'.-.'""':'...';' .'<-'- W'.'.:' , ..: -' ,./-. ...'.'.:. ^ . _ "''..-.-/'X ' -. ':"..'",'"' ' -\.-'' .'-".'<'t.
:5.; WATER LEVEL MEASUREMENT USING AN AIR LINE - feet of water' In the air
line Is calculated from the pressure required to remove the water from
ithe line, then substracted from the length of the air line: . -
L = FV/
w
0.^335
WL = LA, - L
AL w
Where - L = Feet of water In air line
PW = Pressure required to move
water out of air line
WL = Water Level
L., = Length of Air Line
'
6. .WATER FLOW RATE - as calculated from a 90ฐ triangular v-notch weir
(reference 39): . .
: /;. , Q>cH5/2
7. % REMOVAL OF VOC =
C^ - C,. x 100%
Where - Q = Water Flow Rate CftVsec)
H = Height of Water Above Weir
(ft)
c = ^86.2 for weir box used
Where - CR = Raw water concentration
Cp CT = Treated water concentration
K I ' . . - - . ^.
8. COST TO COMPRESS AIR - calculated at 6* per kilowatt hour
Cost.t/1000' gal = 6<=/KWHr x 0,7*t6 KW/HP x HP x 1,000 gal
x 60 mln/hr
w
Where - HP = water horsepower
0. = water flow rate (gal/mln)
107
w
-------
APPENDIX B
GAS CHROMATOGRAPHIC COUDITIONS Ahฉ
QUALITY CONTROL PROGRAM
; . All yo.latlle organic analyses were performed using EPA Method 502.1
for the gas chromatograph (GC). The analyses were done at the North Penn
Water Authority Laboratory using a Varlan 4600 GC with a Varlan Vista 401
Chromatography Data System. The samples were concentrated and Introduced
Into the GC by way of a Tekmar LSC-2 Purge and Trap equipped with a
Tekmar ALSAutosampler. The trap was packed with Tenax, silica gel and
charcoal.
The GC colum used was 8-foot, glass, 1% SP-1000 on Carbopack-B
(60/80 mesh). The purge and carrier gases vrerc helium. The detector was
a Hall Electrolytic Conductivity Detector (HECD). The temperature
program sequence was 45ฐC Isothermal for three minutes, program at 8ฐC
per minute to 220ฐC> then hold at 220ฐC for five minutes.
The NPWA Laboratory Is certified by the Comronwealth of Pennsylvania
to perform trlhalomethane analyses under the Safe Drinking Water Act.
Method 502.1 is used for the ThM as well as the VOC analyses. The
quality control program used with this method Is approved by the
Carmonwealth and has undergone In-house Inspection of record keeping,
equipment, analysts and procedures.
In addition to the Safe Drinking Water Act requirements, several
quality control procedures were developed specifically for this study.
Once every three months, quality control samples were obtained from U.S.
EPA-EMSL for volatile organic chemicals. The samples were provided, at
two levels of concentrations, and had nine compounds In each sample.
Every day one of these samples was analyzed, with high or low range
concentrations being examined on alternate days. The results for each
day were compared to the true values provided by U.S. EPA-EMSL before any
samples were analyzed. If error was less than 10%, samples were
analyzed. If the error was greater than 10%, the cause of error was
determined and corrected before samples were analyzed.
All VOC samples were collected In duplicate In the field. At least
10% of these field duplicates were analyzed. The percent difference for
each compound was determined by dividing the difference between the
duplicates by the average of the duplicates, then multiplying by 100.
The percent differences were wel 1 within 20% for the compounds found In
well L-8 at high concentrations, and within 25% to 30% for the minor
contaminants. . .
In order to determine the GC retention time stability, and thus be
more confident In the VOC identifications, the retention times for each
compound In the-.dally calibration standard were" recorded. These data
--. 108
"^
-------
-.->! ป s-^ir*. f 'A. J
-.' ป .A. , * t " -**1" -^ *,*/',.V*>t *"* * "*3^-
t {<'' ป. i ' * * J- i- ~ - ป~ * ^ ^ ,2 *
'f4 , ""were compiled on a monthly basis and the average, standard deviation, ' " ;'
t~ - '^percent relative standard deviation and retention time relative to: " "'
ซ chloroform were calculated for each compound In the method.: Percent **
relatlye ;st'andard deviations were generally less than 0.5%. ' If the - ^^
^average.retention time changed from the previous month, the new retention
' * time was Inserted Into the GC method In the data system. -,*
X
\
109
\
S
-------
------- |