Note: Descriptions are shown in the official language in which they were submitted.
1.~5~3~Sl
FIBLD OF THE INVENTION
_
This invention relates to radioactive logging methods.
More particularly, it relates -to the use of signals indirectly
related to oxygen nucleus concentrations to determine the frac~
tional content of water in a porous petroleum reservoir rock
containing oil, water, and gas.
BACKGROUND OF THE INVENTION
Economic feasibility of methods for secondary and ;
textiary recovery of petroleum often depends on accurate
measurement of the quantity and location of reservoir water in a
! formation after previous recovery processes have been completed.
Such measurements are desirably carried out in "old" wells, i.e.,
in wells used to produce the formation. Reasons: (i) accuracy
is increased; and (ii) costs are decreased; the process of drill-
ing a new well displaces some fraction of the formation fluids
originally in the formation away from the hole, and it is -- -
desirable to evaluate the potential recovery from a reservoir
without incurring the expenses of drilling a new well.
I In my Patent No. 3,817,328 for "~eutrons Absorption and
- 20 Oxygen Log for Measuring Qil Content of Formations", June 18, ` - -
1974, assigned to the assignee of this application, I describe a
method for accurately determining the oil content of a reservoir
containing both mobile oil and a significant gas saturation. The
first step was the recording of the response of both thermal-
neutron-decay-time log and a neutron-activated-oxygen log to a
formation penetrated by a well bore. A purposeful change was
then made in the oil saturation in a given region of the
formation surrounding the well bore by injecting fluid under
sufficient pressure to displace the connate fluids. The change
should remove substantially all the oil or remove as much oil as ~;~
could be displaced by a proposed flooding technique. The com-
bination of the thermal-neutron-decay-time log and the oxygen
~l~59651
log was 'hen run again to record the response of the same given
region. The difference in the oil conten~t around the well bore
was determined from the differencec, between the two sets of logs.
My method may be somewhat limited, however, by the
requirement that the oxygen activation log be calibrated at least
to the extent that changes in log readings be proportional to
changes in the oxygen content of the reservoir fluids with a
predetermined single constant of proportionality. The responses
from logging tools currently available are unduly influe~ced--in
some applications--by the pipe, cement and liquids in the well
bore; experience has shown that calibration valid at all depths
in the well is difficult (if not impossible) to achieve in such
applications.
Also many oil reservoirs, e.g., the SACROC field in
the West Texas region of the United States,~present additional ~-
problems in attempting to measure water saturation in and around
existing wells; while the production brine may be saline enough
in such reservoirs that conventional methods could work, the
brine in the formation is often significantly different between
producing intervals. Reason: it has been common practice for
secondary recovery purposes to inject brine with salinity
different from that in the virgin reservoir. At a later time
when an accurate measurement of formation water is desired, the
salinity of the water in the formatlon--a function of the degree
of flushlng of the injected water--is different from interval to
interval within the reservoir.
Conventional techniques have also provided several
logs which serve as useful fluid indicators (including forma-
tion water) which could be run through casing~ and which are
based on measurements reflecting the abundance of elements other
than chlorine. These include logs designed to reflect the activa-
tion of oxygen nuclei as well as logs designed to measure the
, ~ ~ "
1~5~6S~
inelastic scattering of nuclei by carbon and oxygen. Oxygen
abundance can also be gained Erom background correction
applied to some versions of pulsecl-neutron-capture logs. But
the recorded signals from each of the logging devices required
to carry out -the above-mentioned methods only indicate the
presence of formation fluids such as water, rather than its -
abundance, mainly due to difficulty in separating responses
due to oil from those associat~d with the formation water.
In summary, the existing prior art methods of which
I am aware are not quantitative enough to provide accurate
determination of the fractional content of water in a reservoir
rock containing oil, water and gas, especially if the formation
brine is nonsaline and/or has been varied by previous water-
floods.
OBJECT OF THE INVENTION .~;
It is the object of the present invention in one
aspect thereof to provide a novel method for accurately
measuring water saturation in a formation penetrated by a well
bore using logs that produce signals that not only vary
linearly with fluid saturations but in a manner associated
with other properties of the well bore and rock matrix, such
that general calibration (and recalibration) of the logging
tool that provide the logs is unnecessary. ;~
SUMMARY OF THE INVENTION
~ In accordance with one aspect of this invention
there is provided a method of measuring the concentration of
water in an earth formation constituting a given depth
~ .
interval of a region surrounding a well bore, comprising:
(a) running, over said given depth interval, a log whose
response to the oxygen content of said region is substantially
linear throughout the range of expectable oxygen contents,
~ _ 4
'''.' ~' .
11 ~55~
and running, over another depth interval known to contain
substantially no water, the same said log, (b) injecting .
into said region adjacent to said given depth interval a
solvent capable of replacing substantially all of previous
fluids in the part of said adjacent region to which said
log is responsive, (c) running said log over said given
. depth interval a second time to obtain signals changed
from those of the first running, the changes reflecting
the effects of Step (b), (d) calculating the water concen-
tration from the measured difference in the signals from
Steps (a) and (c).
In accordance with another aspect of this invention
there is provided a method of measuring the concentration of
oil in an earth formation constituting a given depth interval
- of a region surrounding a well bore, comprising:
(a) running, over said given depth interval, a
log whose response to the oxygen content of said region is
substantially linear throughout the range of expectable
oxygen contents, and running, over another depth interval
known to contain substantially no oil, the same said log,
(b) injecting into said region adjacent to said
given depth interval a solvent capable of replacing sub-
stantiall~ all of the previous liquids in the part of said
adjacent region to which said log is responsive,
(c) running said log over said given depth
interval a second time to obtain signals changed from those
of the first running, the changes reflecting the effects of
Step (b),
(d) calculating the water concentration from the
measured difference in the signals from Steps (a) and (c).
By way of added explanation, in accordance with
the present invention in one embodiment thereof, signals
indicative of the oxygen nucleus concentrations in a poreus
';
~ -4a-
. , : . ~ . . ,~
,
~L05965~
reservoir rock containing oil, water and gas are obtained
using a radioactive logging tool in a manner that allows
the omission of general calibration (and recalibration)
of the tool that provides the logs, even in applications
where the reservoir is nonsaline or where the formation
brine has been significantly varied by previous water-
floods.
The first step is to record the response of a
log sensitive to the oxygen nuclei in the rock, for
example, an ; -
`':
.
~ ~ .
'':
"-. .
'~' '. ~ .
'
I .
. -4b- ;`
~5~6S~L
:
oxygen log, a carbon/oxygen-ratio log, or a pulsed neutron-
capture log, in a manner that assures that the fluid saturations
around the well bore represents those in the bulk of the
reservoir. The next step consists of making a purposeful change
in the region surrounding the well bore by injecting a sufficient
quantity of a chemical solvent or combination of solvents to
displace substantially all of the formation fluids far enough
away from the well bore such that the displaced water formation
cannot be detected by the logging tool. The log is then run
again. The chemical solvent is then displaced itself by brine of
known saline concentration. Then the log is run a third time.
Thereafter, the water content of the reservoir is accurately
determined from the log responses combined with known properties
of the fluids. During the logging steps, the tool does not
undergo general calibration (or recalibration), since the present
invention involves the utillzation of the differences between,
rather than the absolute magnitudes of, the log signals.
Further objects, features and advantages of the present
) invention will become more apparent to those skilled in the art
from the detailed reading o the following description of prefer-
red embodiments thereof, when taken in consideration with the
accompanying drawings.
DESCRIPTION OF THE DRAWINGS
.
FIG. 1 is a side elevation of a borehole penetrating
a reservoir partially cut away to illustrate a logging sonde
connected uphole to a logging truck for carrying out radioactive
logging of the reservoir for the determination of water satura~
tion in the presence, say, of oil and/or gas;
FIGS. 2 and 3 are characteristic amplitude-vs.~time
decay waveforms provided by pulsed-neutron well logging means
which could be used within the logging sonde of FIG. l; and
FIG~ 4 is a schematic block diagram illustrating the
-- 5 --
- . . .
~QS9~Sl
steps used in carrying out the method of the present invention.
PREF13RRED EMBODIMENTS OF THE INVENTION
Referring now to the drawings, particularly to FIG. 1,
there is illustrated a radioactive well logging sonde 10 for
carrying out a survey operation within well bore 11 penetrating
earth formation 12. The well bore 11 is cased with casing 13,
sealed to the formation 12 by cement 14. The casing 13 is
perforated to provide openings 15 for production of gas and oil
from the formation 12, or for injection of flllids from source ';
tanks 16 or 17 at the earth's sur~ace. The passage of the
injecting fluids is via valves 1~, 19 and pump 20 at the earth's
surface and thence downhole through the interior of tubing 21
into the formation 12. ~-
In the arrangement shown, the tubing 21 is packed off ~.
from its upper and lower extremities by packers 22. The packers
22-allow the ~luids to pass relative to a selected (and isolated) .
region of the formation 12, say into sandstone stratum 23 between
.:
- ca~bonaceous and cap rock layers 24 and 25, respectively. At
the earth's surface, the tublng 21 terminates in a connector 26 ~ .
having an opening.27 through which a logging cable 28 passes.
The cable 28 is reeled via sheaves 29 and 30 attached to supports ~ . -
31 of a logging truck 32 for translation of the logging sonde 10
... . .
. through the well bore 11. . . -
.
Constructional characteristics of the cable 28 are : ~.
conventional: it includes a plurality of weight-bearing members :
surrounding a series of conductors.. A portion of the conductors
is set aside for control purposes, which allows signals to pass
from a control and processing unit 33 within the truck 32 down-
hole to a radioactive logging system 34 within the sonde 10.
Another portion of the conductors is set aside for detection
purposes which conduct detected signals representing various
downhole conditions uphole from the logging system 34 to the
l(~S9651
control and processing unit 33. The detected signals form a
log-in digital form--which represents various formation character~
istics by which later mathematical processes of the present
invention can be carried out.
Peripherals of the control and processing unit 33 can
include a magnetic tape deck 36 where the log signals can be
sto~ed; a~ter a multiplicity of signals have been received and
- stored, say from a plurality of logging xuns, processing via
processor 37 occurs. Of course, the signals of each logging run
are identified on a selected track of the tape ("log") by log-
ging run as well as by depth interval. Each data bit of each
"log" is thus identifiable so as to be combinable with other
information in the manner set forth.
Processor 37 is preferably a minicomputer (a relative-
ly low-cost, small, short-word-length (12-16 bits), limited-core-
storageable, microprogrammable device). Control of the processor
37 is provided by appropriate software in cooperation with key-
board terminal 38. Display of the results can occur at the key-
) board terminal 38 or via printer 39.
Logging system 34 is controlled in concert with
selected injection of fluids from source tanks 16 and 17 to
- provide meaningful logging measurements as the sonde 10 is trans-
lated through the well bore 11. Characteristics of the radio-
active system 34 vary with the types of application and data to -
be indicated, but in general include a detector, a radioactive
source and various shields. Most are operated in a manner so as
to gate the detector means and provide one or more radioactive
logs as a function of depth at uphole control and processing unit
33 within the truck 32. ~mong logging systems 34, available for
use within the logging sonde 10 which have special relevance to
the method of the present invention are the following: pulsed-
neutron-capture and oxygen-activation logging systems. Such
; .
. .
., ., . . :. .
1()5g65~ ,
systems are in~ividually available on a commercial basis and
have in common a data ou~put storable on a selected medium such
as magnetic tape processable by associated equipment in a manner
suitable for carrying out the method of the present invention.
Since the logs themselves are indicative of the
operations of the elements which comprise each corresponding -
logging system, a brief discussion of each log is in order and
is presented belo~.
Pulsed-Neutron-Capture Log
Pulsed-neutron-capture logs are designed to accurately
measure the rate of decay of thermal neutrons at a detector fol-
lowing an intense pulse from a high-energy neutron source. The
detector is synchronized with the source to operate while the
latter is off. The radiation detected can be either slow
neutrons or gamma rays resulting fro~ neutron capture; in either
"
case, the signal response is related to the population of ther-
mal neutrons in the formation surrounding the borehole. From
the time the source is turned off, the slow neutrons are gradual-
- ! ly captured and the amount of radiation detected per unit of `!
time decreases until the source is again turned on. By
measuring the number of thermal neutrons (or gamma rays produced
by the~thermal neutrons) present at any particular time, the
rate of decay of the thermal neutron population may be measured.
The rate of decay is dependent upon the nuclei of the material
present in the formations and varies from formation to formation.
This rate of decay is related to the time required for those
nuclei to capture the thermal neutrons, therefore, the measure-
ment is related to this time and hence to the lifetime of the
neutrons in the formation.
FIG. 2 illustrates the decline of the thermal neutron
population as a function of time as measured by the number of
thermal neutrons (or gamma rays produced by the thermal neutrons)
'' '
,,
1059~5~ `
that are present in the vicinity oE the well bore. Thus, the
rate of decay of curve 45 of FIG. 2 is dependent upon the nuclei
of the material present in the adjacent ~ormationO
Of particular importance in water saturation deter-
minations are the oxygen-abundance signatures provided by
monitoring the background indication of the above pulsed~neutron-
capture logs. Returning to FIG. 2, background radiation appears
in region 46 of the curve 45. Note in this regard that curve
45 results from utilization of a pair of counting-rate curves
plus a curve indicative of the rate of decline of the neutron
) population. The early gate, Nl, portion of curve 45 is normally
derived during the interval of 400 to 600 microseconds after time
zero ~the time of termination of the neutron pulse from the
neutron source), and is a measure of the radiation intensity
detected during that interval. A later gate, N2, portion of the
curve 45 is a similar measurement derived during the interval
from 700 to 900 microseconds after time zero. By further modi-
fication in the manner taught in U. S. Patents 3,705,304, issued
) December 5, lg72, E. C. ~Iopkinson et al and 3,706,884, issued
December 19, 1972, A. H. Youmans, extraction of the oxygen-
abundance level of the surveyed formation fluids is possible,
say ~rom the relative intensity of background region 46.
FIG. 3 illustrates to another method of gaining an
- indication of oxygen abundance of a surveyed reservoir by
measuring only a component of signals associated with the back-
- ground radiation of the pulsed-neutron-capture log.
Assume in FIG. 3 that the formation has been irradiated
with a pulse o~ neutrons over a time ~ in a region 48 which is
varied to equal the measured exponential decay time for the dis-
appearance o~ thermal neutrons. Note that time r is maintainedby continuous corrections to provide a logic rule that curve 49
detected by the logging tool in the interval 2~ - 3~ is equal to
twice that in the interval 3~ - 5 ~ . Amplitudes of curve 49 are
_ g _
~lQS~6~
also determined, in part, from the signal measured during the
time interva~ 6~ - 9~ , called the "Gate 3" signal. In this later
time interval, the signal from thermal neutrons has decreased to
less than 0.0025 times the original thermal neutron signal; so
the intensity of the detected signal in this interval represents
natural gamma radiation and radiation from the decay of unstable
N 6 formed by neutron activation of ol6 nuclei. Thus the "Gate
3" response ~rom the above-described logging tool from either the
near or the far detectors thereof provides a signal that varies
with the oxygen content of the fluids under survey. Time of
! neutron activation and signal recordation during one cycle of
sliding-gate pulsed-neutron logs is, of course, directly propor- '!~.
tional to time ~ . ~Iowever, the number of pulses is inversely
proportional to ~. During any period containing a number of
pulses, then the "Gate 3" response is independent of ~ . The
sliding gate pulsed neutron log is commercially available and is
described in detail in Transactions of the Society of Profession-
al Well Log ~lalysts~ "Thermal Neutron Decay Time Logging Using
) Dual Detector", J. T. Dewan, C. W. Johnstone, L. A. Jacobson,
W. B. Wall, and R. P. Alger~ 1973.
Recorded background response of the log of FIG. 2 as
well as the "Gate 3" response of the slidiny-gate pulsed neutron
log of FIG. 3 can be characterized by Equation 1:
SO = Ctl + a~2f(oxygen) '(1) ~.
wherein
SO is the measured response for a certain logging speed
at a certain depth;
~1 is a constant reflecting gamma radiation whose origin
is not activated oxygen radiation, which includes both
natural gamma radiation as well as radiation from
activated nuclei other than oxygen;
- .~ .
-- 10 -- ~ ~
:
. .
105~S~
~2 is a constant reflecting source strength and detector
sensitivity to gamma radiat`ion emitted by the unstable
N16 nucleus that results from activation of ol6; and
f (oxygen) is some function of the oxygen content in the
well, the casing cement (if the well is cased), the
reservoir rock, and the fluids contained in the forma-
tion.
Referring specifically to Equation (1), it should be
noted that the constant ~1 and ~2 are sensitive to source
strength, detector sensitivity and logying speed, and thus are
) different (in an absolute sense) for the different logs depicted
in FIGS. 2 and 3. It is necessary to take steps to insure that
these parameters remain constant for successive runs~ such as by
checking and adjusting logging parameters while the tool is at
depths where saturations are not to be changed during the
measurement process. The function f (oxygen) will probably not
be simply proportional to the over-all oxygen content of the
material in the well, the casing, and the formation, because
) each logging tool has different sensitivity to matter at differ-
ent distances. The pulsed-neutron-capture log is designed to
minimize the signal from matter near the tool, however; field
experience has moreover shown that well bore fluids and casing
cement can influence measurements by the tool; so the signal ;
thexefrom cannot be ignored. For the measurement process of the
- present invention, however, it is necessary only that changes in
log response be proportional to changes in oxygen content in the ;
fluids in the formation if all other factors are held constant.
Use of background radiation data from pulsed-neutron-capture log -
responses in the manner of FIGS. 2 and 3 have several additional
benefits: (1) existing tools have been developed small enough to
be run in tubing, (2) the primary pulsed-neutron-capture log ;~
response is also available for supplementary measurement such as
,:
~59~5~L
a porosity measurement, and (3) depth of investigation is great-
er.
Oxygen Logs Specifically Designed to Record Oxygen
Neutron-activation logging to measure the oxygen con-
tent of the formation is also available, and one system is pro- ;
posed in U. S. Patent 3,46S,151 by A. H. Youmans, issued
September 2, 1969. In that system, th~ neutron-activated radia-
tion is sensed downhole in a manner that such radiation can be
related to the oxyyen content of the region around the well bore.
The essential difference between such a log and the above-
) discussed background signal from pulsed-neutron-capture logs of
FIG. 3 is that in a log specifically designed for oxygen measure-
ment, more effort must be applied to exclude signals unrelated
to oxygen content. In other words, radiation not associated
with the decay of the unstable N16 nucleus which is produced by
interaction of high-energy neutrons with ol6 must be excluded.
This is achieved by noting that such radiation in such logs dies
away with a time constant of 7.3 seconds; thus it can be distin-
guished from other radiation on the basis of its persistence.
Since the uses of the present invention involve the
difference between, rather than the absolute magnitudes of, log
signals, the exclusion of unrelated signals is also of less
importance. Therefore, logs specifically designed to measure
oxygen activ~tion are suitable or carrying out the method of
the present invention. If source strength, detector sensitivityt -
and logging speed are constant, the signal from the oxygen log
can be described by Equation (1), where SO represents the
oxygen-log response. The parameters ~ 2~ and f (oxygen)
have the same meanings discussed above but are related to the
characteristics of an oxygen log.
FIG. 4 illustrates the steps of the present invention
in detail.
- 12 -
~05~6~
While logg;ng techniques are commercially available to
measure parameters of interest related to formation and formation
fluids about the well bore, it is necessary that steps be taken
to insure that the fluid content of the region of interest
reflects as nearly as possible the fluid content throughout the
reservoir. Effects associated with rapid production can be
minimized by producing the well slowly or by shutting-in the well
to be logged for some time before the logs are recorded. After
the water saturation in the vicinity of the well bore has been
made equal to that in the reservoir, the formation can be logged
by one of the several methods previously described. For ease of
discussion, however, FIG. 4 reflects the operation of the present
invention with the particular logging method: the pulsed-
neutron-capture logging system.
In accordance with the present invention, the first
step consists of logging the formation interval of interest to
provide log signals SOl. Thereafterj a purposeful change is made
in the region by injecting into it fluid or flulds from source
I tank 17, FIG. 1, capable of removing all of the formation fluids.
One process by which the formation fluids can be removed consists
of injecting solvent followed by alcohol; a second consists of
injecting a micellar solution preceded, if necessary, by
sufficient fresh water to prevent emulsification of the solution. ;
These techniques have been described for use with other well
1oggin~ methods, for example, by R. P. Murphy, W. W. Owens, and
D. L. Dauben in U. S. Patent 3,757,575, issued September 11, 1973.
For the measurement process herein described, it has been ;
mentioned that the displacing fluid must have known properties.
Specifically, it is necessary that the volume fraction of oxygen
in the displacing fluid be known. If fluids of unknown composi-
tion are used, this fraction can be accurately measurçd with
currently available laboratory techniques.
~ .
- 13 -
1059t;5~
The next step consists of running the log ~ second time
to provide the signal SO2 of FIG. 4~ while the displacing fluid
is in the porous reservoir rock. Because part of the signal
recorded by logging tools is determined by the fluid within the
well bore, it is desirabl~ to insure that this fluid is nearly
the same for all logging runs.
The fourth step of the process consists of injecting
brine from source tank 16 of FIG. 1 into the formation to insure
that the solvent that displaced the formation fluids is itself
displaced into the formation beyond the region of tool response.
,~ The preferred brine for this step is produced brine of known
characteristics.
The fifth step consists of recording the log response a
third time when the formation is filled with brine to provide the
log signal SO3 of FIG. 4.
Calculations can then occur at the well site using the
processor 37 in truck 32 of FIG. 1 to indicate water saturation.
The basis of the calculations are set forth below. ~ -
) The magnitude of a pulsed neutron log signal reflect- -;
ing oxygen abundance can be represented by a more detailed form
of E~uation 1:
Sl - dl + d2f(~ C~ ~r~ p wl) (2)
wherein -
is the mass of oxygen contained in fluids in the welli -
~L C is the mass of oxygen contained in the cement casing;
J~ r is the mass of oxygen contained in the reservoir rock;
and
is the mass of oxygen contaiDed in the water in the
porous part of the formation after the well is
produced.
The magnitude of the second oxygen-related signal, SO~,
- 14 -
~s9~s~
1 can be represented by ~guation 3:
S2 - al + a2~Qf'QC' r' p~D
2 where Qp D ls the mass of oxygen contai~ed in the displacing
3 agent ~hich ~i~ls th~ poraus rock at the time of the second log-
4 ging run.
~he mag~itude of the 02ygen-related cig~al from the
6 third log, S03, can b~ represented by Eguation 4:
3 ~1 2g~Qf'Qc'Qr'Qp,w2~ (4)
. .
`)7 where Qp is the mass of oxygen coDtained in the ~ater which
8 no~ co~pletel~ fills the Eorous part o~ the ~ormati~n~ .
9 . The o~ygen content of the water in t~e porous reser~oir
at the time o~ $he first set of logs, np wl~ is described by . .
11 Equatio~ 5: :
p,wl ~Pw wFo,wV (5) ~ ,~
,.~: :
12 wherein . ~:
~ = the po}asitl of the formation;
14 Pw = the densit~ ~f the formation brine;
Fo w = the ratio ~f the mass cf oxygen to the over-all ~ass
16: o:~ the formation brine; j~
17 V = the ef fe~t-1~e volume sensed by the log; and
18 S~ = the ~at~r saturatio~ represe~tative of the reservoir. :
.:
1g ~he o~ygeD coDtent of the displacing agent filling the :
porous reser~oir ~en the s~cond log is run, , can be repre-
21 :sented by E~uatioD 6: .
n p D = ~PDFo~D (6)
,',".
22 whe rei2
23 p = deIlsity of tbe displacing agent; . .
~ ~ D = ratio of the mass of oxygen to the over-all mass of
. .
- 15 -
1059GS~
1 the displacing agent.
2 The oxygen content of the porous reservoir fill~d with
3 brine ~hen the third log is run,Qp w2 , can be represented by
4 Equatlon 7:
Qp,w2 = wFo,w
S ~ater saturations (Sw~ can ~e comput~d at each depth
6 ~rom the difference~ in the l~g responses described above. The
7 ratio between dif~erences of recorded signals can be represented
8 ~y Equations 8, 9 and 1~:
n 2 = ~ p F V (8)
f ( nf ~ nC ~ Qr ~ np ,W2~ - f (nf ~ nC ~ Qr ~ np ,wl) (g) ..
SO3-502 ~(nf~nc~nr~slp~w2)- f (nf~S?c'S2r'n D)
.af ~ -
3n (Slp ,w2 np ,Wl)
anp Inp,w2~ np,D)
(Pw o,w SwPwFo,w) .. :
w o, w D o, D
Sw) Pw o ,w . .,
:Sv = 1 - [1 - ~ ~oL ~10)
9 It is also ev~aent that if the formation ~oes not con~
tain gas ~ithin thc zoDe of interest then the oil saturatio~ [So~
11 can be inferred fr~m Eguatio~ 10, vis: ;
50 = 1 - Sw = ~ - ~ 503~5~
12 I~ can be seen from tbe equations above tbat ~alidity of the
13 method claimed does not depend OD the guantitati~e accuracy of ' ~`
14 the method for logging measurement of oxygen_ The method is
15~ valid so long as cha~qes in log responses are linearl~
-- 1 6 --
~059~Sl
proportional to changes in oxygen content throughout the ranye
corresponding to changes in fluid content in the pore space.
Illustrative Example
The following is an example of the use of pulsed-
neutron-capture log data to measure water saturation around a
well. A waterflood in the field replaces an unknown fraction
of the original formation brine by water with a different
salinity. Economic evaluation of a proposed tertiary recovery ~
program requires accurate determination of porosity, and , ;
inter alia, water saturation through a 20' interval. Determina-
) tion is to be made from signals available from a sliding-gate
pulsed-neutron-capture log.
Step number one consists of running the log in the
well after production has terminated. The logging tool is
run through the depth interval of interest a number of times in ~
order to reduce statistical errors of the measured values. The ~` ;
average of the gate three (P3) signals at a particular depth is '
103.6 count. The average capture cross section is lS.7 c.u.
) The latter value is not used in any of the calculations that
follow, however, because the salinity of the brine in any
particular part of the formation is unknown.
The second step is accomplished by the injection of a ~
miscible solvent such as alcohol or methanol followed by enough t..~,.,''
~,
isopropanol to 1nsure complete displacement of the solvent
several feet away from the well bore. This injection is ~.
accomplished by setting a plug below and a packer above the
interval logged. After the injection step is completed, the `
packer is unseated and enough produced brine is injected to
replace the alcohol in the annulus between tubing and casingO
This insures that the same fluids are present within the well
bore at each logging step so that differences in logged values
reflect only changes of the fluids in the formation.
- 17 -
~L059~
The third step consists of a second running of -the
pulsed-neutron-capture log. Data are recoxded through the
inte~val of interest as well as an adjacent interval in order
to confirm that efforts were successful in keeping logging
panel settings and logging speeds the same as they were when -
the original sets of logs were rec~orded. From the multiple
logging passes through the interval of interest, it is deter-
mined that at the particu1ar depth the cap-ture cross-section,
~ 2~ is now 14.8 c.u., and the average gate three signal,
(F3)2, is 82.0 counts.
Step number four consists of injection of approxi
mately 100 bbls of brine of known capture cross-section into
the formation. The salinity of the brine is chosen to be
50,000 ppm NaCl.
The final step consists of repeating the pulsed-
neutron-capture log. The average capture cross-section measured
at the particular depth of interest is now ~ 3 = 19.6 c.u., and
the average gate three signal is now (F3) 3 = 137~2 counts.
In order to calculate the water saturations it is
necessary to use the following constants:
density of isopropanol at reservoir temperature, Pd=
0.757 g/cm3 ;~
fractlonal mass of oxygen in isopropanol, Fo d =
0.266
Thermal-neutron-capture cross-section of isopropanol,
d = 20-0 c.u.
density of brine at reservoir temperature, Pw =
1.027 g~cm3
fractional mass of oxygen in brine, Fo = 0~855 `-
Thermal-neutron-capture cross-section of 50,000 ppm
NaCl brine,~ w = 40 0 c.u.
- Equation 10, supra, can be evaluated:
- 18 -
., :: , . .
; -. . . ' , . ,:
. ,. ~: .-; .'. -' , , ' ': .
1~i965~L
_ ~ 7 2-103 ~ 1 00757*0.266
~w - 1 - ~1 ~ 1 027*0.855
Porosity can also be determined from the ~ross-sec-tions ,~
measured by the pulse-neutron-capture log:
40 0-20 0 = 0.24
Modification
In some applications, it is possible to omit the
last injection of brine surrounding the well bore and the
subsequent logging of the fluids by the logging system depict~
ed in FIG. 1. For the modification to be effective, it must
be recognized from the data that a given depth interval in the
well has properties substantially equivalent to the tested
interval of interest and the former is known to contain no
water. Water saturation is then calculated using the signal
recor~ed in this region in place of the signals which would
have ~een recorded after the final injection stage. Con~erse-
ly, if oil saturation calculations in accordance with Equation
I1 ars to be carried out over the given depth interval supra,
the region used for normalization purposes must be known not
to contain oil. Oil saturation is then calculated using the
signal recorded in that region in place of the normal SO
signals.
Also, replacement of the displacing fluid within ~;
... . .
the cased hole can often be accomplished by unseating the
packer used to seal off the interval of interest and inject-
ing produced brine. Since this brine is heavier than the
displacing fluid, the lower part of the hole can be filled
with the brine, without raising the-pressure enough to cause
its injection into the formation. It is often useful to put
enough brine into the hole so that it fills the hole through
not only the formation of interest but also another region
in the hole. Logs recorded through this latter region can be
-- 19 -- ,~
lU~96~ ~
used to check the log responses.
While specific embodiments o~ the invention have :
been described in detail, it should be understood that the ; :
invention is not limited thereto as many variations will be
readily apparent to those skilled in -the art~ and thus the
invention is to be given the broadest possible interpreta-
tion within the terms of the following claims.
.. .
,
~,
'. ~'
;`:
~-
.
- 20 -
.
