. ..

Formation Evaluation by Analysis of

Hydrocarbon Ratios
B. O. pix]er, SPE-AIME, Bamid Div. National Lead Co.

Introduction
Mud logging was first offered commercially in Aug., A long-accepted premise is that as formations are
1939. This logging method quickly gained favcr drilled, the drilling mud filtrate partially flushes the
among many operators because the type of fluid in the formation fluid ahead of the bit, It was generally
formation could be determined within minutes after thought that the formations were flushed to an irre-
the formation was drilled, The presence and magni- ducible minimum — generally considered to be about
tude of the methane show was and is the most im- 30 percent of in-place fluid. Experience in mud log-
portant factor in mud log interpretation. However ging, however, has shown that this rarely happens.
this magnitude in some instances was improperly This partial flushing does not prevent mud lo@ng
understood, and as a consequence some operators from successfully determining productive or non-
still do not use mud logging, even though the early productive formations. Experienced logging engi-
technique frequently made the difference between a neers, in possession of quantitative gas analyses, make
successful well and an abandoned hole. Both the “hot interpretations that take into account the flushing that
wire” log of gas combustibles in the sample and the results in rocks of various permeabilities, the effect
percent-of-gas log obtained with the conventional gas of overbalanced mud weight and the effect of initial
trap and the gas chromatographyindicate only that the filtrate 10ss,
reservoir in question contains hydrocarbons, These
methods do not necessarily indicate the quantitative Method ●

amounts of the various hydrocarbons in the mud. Ordinarily, when formation cuttings are drilled they
The addition of a new Steam Still-Reflux gas sam- retain much of the formation pore fluid. This fluid is
pling system to gas chromatography enables accurate released to the mud column as the cuttings travel up
determination of the composition of the mud gas the annulus. Most of the formation fluid in the cut-
sample. A knowledge of gas composition makes it tings will be “produced” into the drilling mud during
possible to establish the relationship of methane to the top 500 ft of hole travel. Conventionally, a mud
the heavier hydrocarbon shows. An awareness of this sample is diverted to a mechanically operated gas trap
relationship led to a new, additional mud log inter- to obtain a sample of the gas in the mud. The effi-
pretative technique that permits relating the quanti- ciency of this trap is from 15 to 70 percent, depending
tative amounts of methane (C,), ethane (C,), pro- upon the gel strength of the mud, the amount of mud
pane (CJ, butane (C,), and pentane (CJ to in-place flowing through the trap and the rotation speed of
reservoir fluid content. the trap impeller. The magnitude of the conventional

The ratio of methane to the heavier hydrocarbon components — ethane, propane,

butane, and pentane — is indicative of gas, oil and water productive potential,
The Steam Still-Re@x Unit, used in conjunction with mud logs and gas chromatography
yields a quantitative analysis from which this ratio can be plotted.
1

1969 665
 .“

gas show is, therefore, quantitative ordy to the air-gas at the Reflux-Condensing Unit, withdrawn with a
sample obtained, The sample is accurately analyzed syringe, diluted to the standard chromatography sam-
by the gas chromatography; but, because the sample ple size and injected into the chromatography for
furnished by the conventional gas trap represents only analysis. The Reflux-Condensing Unit removes only
a fraction of the gas present in the mud, and because the lighter paraffin series hydrocarbons from the mud
that fraction is not representative of the total gases in sample tested. For example, if the mud contains diesel
the mud, the results are still only qualitative. oil, the more complex hydrocarbons — COand above
When the Steam Still-Reflux Unit is used to obtain — cordense and drop back into the mud-steam mix-
the gas sample, the gas sample will represent almost ing chamber. Therefore, regardless of whether the
100 percent of the hydrocarbon fractions C, through fluid phase of the mud is oil or water, the gas sample
C, that were in the mud sample. This enables the analyzed contains only the light fractions through Cb,
chromat~graph analysis to be related quantitatively and the analysis is representative of the formation gas.
to the mud, and the readings to be reported as parts The full importance of determining formation gas
per million of each hydrocarbon vapor (C, through composition has not always been apparent. At first
C,) to mud volume. it was observed that if the magnitude of butane in
Because the cuttings from a particular formation the mud was greater than the magnitude of either
“produce” the gas they contain into the drilling mud, propane or ethane, the zone in question would pro-
it was reasonable to assume that this same formation, duce water and hydrocarbons, Later, the ratios of
if completed, would produce gases of a similar com- methane to each of the heavier hydrocarbon com-
position. This assumption led to a comparison of ponents were plotted on semilog paper. Hydrocarbon
ppm Logs of hydrocarbon vapors with similar data ratio plots obtained from ppm Logs and available
from producing wells. Plots were made of the ratio data from wellhead gas sample analyses were com-
of methane to each of the heavier hydrocarbons from pared, The comparison of the plots from ppm Logs
many analyses of wellhead samples. These plots were and wellhead gas analysis data showed a striking cor-
compared with plots, made from ppm Logs, of gas relation. The correlation demonstrated that ppm Logs
in mud. Both groups of plots showed definite patterns made with Steam Still-Reflux samples could be in-
between (1) the magnitude of the ratios of methane terpreted in terms of in-place formation content.
to each of the heavier hydrocarbons, and (2) the The magnitude of the methane-to-ethane ratio de-
slope of the lines of the plotted ratios, These, in termines if the reservoir contains gas or oil or if it is
turn, indicate productive potential and reservoir nonproductive. The slope of the line of the ratio plot
permeability. of C1/C2,C1/CS,CJC4, and CJC6 indicates whether
The Steam Still-Reflux Unit consists of a small the reservoir will produce hydrocarbons or hydro-
steam boiler, mud-injection port, mud-steam mixing carbons and water. Positive line slopes indicate pro-
chamber, Reflux-Condensing Unit and a gas-extrac- duction; negative slopes indicate water-bearing for-
tion port, Five ml of mud ~veinjected into the purged mations. An undersaturated reservoir may show a
mud-steam mixing chamber. The mud is rolled with negative slope, but such occurrences are rare. The
2,000 to 4,000 volumes of steam. The hydrocarbons ratio plots may not be definitive for low permeability
(C, through CJ extracted from the mud are collected zones, but unusually steep plots indicate tight zones.

400
300
200-
/

@ MISSISSIPPIAN
OIL RESERVOIR
40 @ Gas-condensate,
FrioSand,
30 l=%, ‘j @ ;&&Ar\Er~ATE Hidalgo CountY,Texas, analysis
from wellhead sample.
RESERVOIR @ )!ol#5;:::A~eil;il;cur, analysis

@ State tract 49 No. 1 well, Nueces

County, Texas, 12690 section, analysis
from gas show on p~m Log.

4-
3 H7
2Lc&_++l#
I I
I
J_
q ~ c, ~
HYDROCARBON RATIOS
Fig. l—Hydrocarbon ratio plots obtainad from wellhead Fig. 2—hydrocarbon ratio plots, productive reservoirs,
sample analyses data, limestone reservoirs, South Texas.”
Rocky Mountain area.

666 JOURNAL OF PETROLEUM TECHNOLOGY

..-

Aratioof C,/C, between2 and 15 indicates oil. A Experience shows that if the C,/Cz ratio is above
ratio of Cl/C? between 15 and 65 indicates gas. The 65 the zone is too tight for commercial production. ‘
!ower the Cl/C, ratio, the richer the gas or the lower Fig. 3 shows the ratio plots obtained from ppm Logs
the 011gravity, If the ratio of CJCz is below about 2 on Texas Gulf Coast wells that were nonproductive
or above about 65 the zone is nonproductive. in the zones of interest. Plot 1 is from the ppm Log
of the R. A. Tally No. 1 Well, Victoria County, Tex.
Field Examples The C,/C, ratio was 470. The zone was tested ex-
Fig, 1 shows average hydrocarbon ratio plots from tensively but it was a low permeability reservoir that
limestone reservoirs in the Rocky Mountain area. could not be commercially completed. Plots 2 and 3
Plot 1 is derived from analyses of gases from Mis- are from the State Tract 49 No. 1 Well, Nueces
sissippian oil-producing reservoirs, The CJC2 ratio County, Tex. Plot 2 was from a sand encountered at
is 3.5. The slope of the line i: positive and not steep. about 8,060 ft. The relatively high ratios of CJC2,
Plot 2 was obtained from analyses of gases from wells C,/C,, C,/C?, and C,/C, indicated that the zone was
producing gas-condensate from the Silurian. The nonproductive because of the low permeability. This
C,/C, ratio is 12; the line slope is again positive and was subsequently verified by testing. Plot 3 was ob-
not steep. Plot 3 is from gas-condensate wells pro- tained from a sand at 9,130 ft. The negative slope of
ducing from the Ordovician. The C,/C, ratio is 15 the ratio plot, C,/C, to CJC3, indicated that the
and, again, the slope of the line is not steep; all three zone was water-bearing, Subsequent formation tests
plo~s ~how slopes-favorable for producti&. Plot 4 “ showed water and non:commer~al amounts of gas.
;hows ratios obtained from an analysis of gas from Plot 4 was obtained from the ppm Log”ofthe Kovar
the Lower Ordovician, which produced gas and No. 1 Well, Victoria County, Tex. The sand encoun-
water, The plot shows a negative slope of the section tered from which the plot was made is at 10,120 ft.
from the C,/C, ratio to the CJC, ratio. Many tests The gas show appeared to be good, but a negative
have verified the fact that if a ratio plot shows a slope of the CJCS ratio to the CJC4 ratio was posi-
negative slope, the zone in question is water-bearing. tive identification of a water-bearing .ormation.
Fig. 2 shows plotted hydrocarbon ratios for pro-
ductive reservoirs in South Texas. Plot 1 was made Evaluation Technique
from an analysis of a wellhead sample of gas- It is apparent that with this evaluation system, poten-
condensate produced from a Frio sand, Hidalgo tial production can be accurately predicted, The only
County. The production is rich in liquid hydrocar- significant time lapse between penetration of the
bons as indicated by the low C,/C, ratio. Plot 2 is formation and evaluation of its productive possilMi-
from an analysis of a wellhead gas sample from an ties is the time required to pump the mud from the
11,000-ft oil reservoir, North Lindsey field. The bottom of the hole to the surface and analyze it by
pentane was not reported, but the low C,/C, ratio the Steam Still-Reflux and Chromatography method.
indicates oil production. Plot 3 was obtained from a Fig, 4 shows the evaluation technique, which maybe
gas show at 12,690 ft on the ppm Log of the State described as follows.
Tract 49 No. 1 Well, Nueces County, Tex. Formation First, record the net increase of each gas com-
tests resulted in gas production. ponent over the background gas; next, plot the ratios

~ooo:_ : ,; ,_ :-+. -:.:

6000”
4000
3000
2000
:

/’-
: : “-”L–.
,.—
‘- .. @ R.;.T/&V$u;;E;;XAS

STATE TRACT 49 No. 1 WELL,

NUECES COUNTY, TEXAS
@ Use only gas increase over
background.
@ Plot ratios on lines indicated.
mm:. DEPTH 8060”
@Evaluate section forprobable
600 STATE TRACT 49 No. 1 WELL, production as indicated by the plotted
NUECES COUNTY, TEXAS curve within the following limits:
400 DEPTH 9130
300 (a) Productive dry gas zones may show
200 A. KOVAR No. 1 WELL, only Cl but abnormally high C, only
VICTORIA COLINTY,TEXAS shows are usually indicative of
DEPTH 10120 salt water.
(b) If the C,/C, ratio is low in the oil
secLon and the Cl/C, ratio Is high in
the gas section the zone is probably

“ ““’:2=
10-, :Z,,
8 F. -~:
---,: -.-=u,;–Z-;T.-_=
::- ~-
L,&-—&-,q
nonproductive.
(c) If any ratio (C,/C, excepted if oil is
used in the mud) is lower than a
preceding ratio the zone Is probably
nonproductive. For example, if
6=-:-: _- :: —.. . . .._- . ‘— . . . . -
C,/C, is less than C,/C, the zone is
4----;- -.:--- --–= ----; probably wet.
~.–.. - . . . . . —. -—- —-
~2
2, ~. .& . . ..&.. d (d) The ratios may not be definitive for
& HYDR$AR80; RATI;S
low Wrmeability zones, however
c, C c. c, ste~p ratio plots may indicate
HYDROCARBON RATIOS tight zones.

Fig. 3—Hydrocarbon ratio lots, nonproductive reservoirs, Fig. 4-ppm Log and repoti form for evaluation of show.
South Texas—analyses, f rom gas shows on ppm Logs.

JUNE 1969 667

 .

C,/C,, C,/C,,, C,/C,, C,/C, on the ratio lines as what appeared cm the ppm Log to be a good sand
indicated. Then evaluate, within the following limits, body, but the ppm Log showed only a nominal in-
the section in question for probable production as crease in gas. After the sand was penetrated and the
indicated by the plotted curve: well deepened, hole trouble was encountered, No
1. Productive dry gas zones may show only C,, other information of interest was available on the
but abnormally high shows of C, only are usually sand. The cost of the side-tracking to investigate the
indicative of salt water. sand was sizable. Tight hole conditions and the low
2. If .he C,/C, ratio is low in the oil section and magnitude of the gas show indicated that the sand
the C,/C . .Jio is high in the gas section the zone is had good permeability and” that possibly formation
probably nonproductive. hydrocarbons had been flushed ahead of the bit, A
3, If any ratio (CJC, excepted if oil is used in the plot of the hydrocarbon ratios indicated oil produc-
mud ) is lower than a preceding ratio, the zone is tion, Therefore, at considerable expense, the sand
probably nonproductive. For example, if C,/C., is was investigated and a new oil field was found,
less than C,/C,, the zone is probably water-bearing. An interesting well recently drilled in St. Martin
4. The ratios may not be definitive for low perme- Parish, La., was the No, 1 St. Martin Bank and Trust
ability zones; however, steep ratio plots may indicate located on the southeast flank of the Arise La Butte
tight zones. Dome. A good sand encountered at about 8,000 ft
showed oil, but the negative slope of the ratio plot
Application indicated that the sand was water-bearing, The well
The ppm Log is only one of many tools that are was deepened to approximately 9,600 ft. One of the
ordinarily used for formation evaluation. But in many partners, a successful independent with a talent for
instances, the ppm Log has furnished the vital infor- finding oil by “feel” and by prudent use of the latest
mation necessary to make the final decision on a well. technology, decided that the formations in which the
One well drilled in inland waters of Louisiana had well was being drilled were tilted to almost vertical.

ppm LOG ~. —.r.

r I I I I 1
I DR I II c, I C21C31C41C5 I
400,
300
200

100
80
“60
-.!s
I
A
1#- 1

-GAS ,’~ “ 1
40 1 Z.T 1 ,8,, 4
30
20
?&=H
10
8
6
4
&
3 -OIL
I
1 1

-t+
2

TTT71
NONPRODUCTIVE

Fig. 5—ppm Log and hydrocarbon ratio plots, No. 1 St. Martin Bank and Trust Well, St., Martin Parish, La.

668 JOURNAL OF PETROLEUM TECHNOLOGY

On his recomrnendaion the well was plugged back TABLE l—MUD GAS COMPONENTS, PERCENT OF
to about 7,000 ft and sidetracked. The sand that was TOTAL GAS
drilled at 8,000 ft in the first hole was encountered
in the directional hole at approximately 7,400 ft and —%;h . c, c?
—. c, c, c’
the entire sand was hydrocarbon saturated. 7,460 93.1 3.6 1,5 _ii( -iii
The ppm Log and the ratio plots from the sand in 7,475 82.4 9.6 5.9 1.4 0.7
the sidetrack hole are shown in Fig, 5, Table 1 shows 7,485 74$4 13.3 10.0 1.6 0.7
7,490 78.0 11.4 8,6 1.4 0.6
the mud gas components related to percent of total
7,500 77.0 14.3 7.2 1.1 0.4
gas. In actual practice the ppm gas shows obtained 7,515 761 17.1 5.1 1,4 0.3
from the ppm Log are not converted to percent of
total gas; but note the general decrease in percent
methane in the lower section of the sand compared the bottom of the sand,
with that in the upper section, The magnitude of the Another example of the application of the ppm
gas show in the straight hole and in the sidetrack Log is No, 1 State Tract 198 Well, Aransas County,
hole was significant, An accurate determination, how- Tex. Many sands were encountered showing the
ever, of the composition of the gas in both cases led presence of hydrocarbons. The logging crew sub-
to correct conclusions on the potential productivity mitted more than 60 ratio plots to the operator during
of tile sand at the different depths in each hole. Note the drilling of the well. In almost all instances subse-
that the ratio Plot 1 at the top of the sand indicates quent information verified the logging engineers’ pre-
a gas cap. As shown in Table 1, the gas was 93.1 dictions of probable productivity based on the ratio
percent methane. Subsequent plots indicated that plots, Fig. 6 shows a section of the ppm Log and the
production would be oil. In each of these cases the ratio plots for this well. The gas composition relating
C,/C, ratio was less than 9. The lowest ratio, 4.5, is the percent of each gas component to the total gas is
shown in Plot 6, which was made from the show at shown in Table 2. Gas condensate production is indi-

ppm LOG
4oo~NONpR()~ucflvE+l
---
SW
200
FF

3oJzf=Etl

201++-t-H ! ! !.
10 ;
8. 1 I
6 -OIL
4
3
2“
NONPRODUCTIVE

-..,,.,
1’
HYDROCARBON RATIOS

PLOT

M
11300 ......’! 1 M 71 84 230
● R.: 2 20 61 92 102
. 5“ 3 19.5 78 233 310
4 16 93 232 a
5 19.3 58 115 185
 TABLE 2—MUD GAS COMPOSITION, PERCENT OF bons in mud is indicated, and other supplemental
TOTAL GAS information may be necessary to evaluate the forma-
tion in terms of potential productivity. However, if
‘i%h CJ_ c, c1 c, c, the composition of the gas sample obtained from the
10,110 92.0 i> G = G mud is representative of the in-place formation gas,
10,115 92.0 4.6 1.5 1.0 0.9 then the gas analysis is accurate. The use of the Steam
10,450 93.3 4.8 1.2 0.4 0.3 Still-Reflux Unit makes possible a report of forma-
10,520 92.8 5.8 1.0 0.4 0.0 tion gas composition on the ppm Log. Meaningful
11,305 92.3 4.8 1.6 0.8 0.5
ratio plots of gas composition can then be made. Even
though many factors affect the amount of reservoir
cated by the ppm Log and ratio plots as shown, The fluid released to the drilling mud, reservoir “potential
zones are tight marine deposits — especially the productive capabilities can be determined by a study
10,520-ft zone. Plot 4 has the steepest line slope; of the ratio of methane to each of the heavier hydro-
pentane was not present. TIIe slope of Plot 3 is steep, carbon components. The hydrocarbon ratio plot is a
Plots 2 and 5 show more favorable (less steep) line unique technique and provides the operator with new
slopes. The electric log and subsequent formation information for evaluating productive possibilities of
tests made of each zone indicated probable produc- exploratory wells.
tion. The well was completed as a gas condensate Computer programs involving percent gas in mud
producer in the 11,300-ft section, which is the section (ppm Log) and gas composition are being used in
plotted as No. 5. special cases to determine reservoir potential produc-
tion. The use of computers in mud log interpretation,
Conclusions although new, will contribute significantly towards a
Only qualitative shows of hydrocarbons in the mud better application of the data shown on the ppm Log.
can be derived from conventional mud logs, If iJPT
chromatography is used, only a general indication of
in-place gas composition is obtainable. Such hydro-
carbon shows may be reported as units of gas or Orlglnal manuscript received in Society of Potro!eum Engineers
office Aug. 5. 1968. Ravlsad manuscript received -March 5, 1969.
percent hydrocarbons or parts per million of gas in Paper (SPE 2254) was presented at SPE 43rd Annual Fall Meeting
the air-gas mixture tested. Only the presence in rela- halcl In Houston, Tax., Sept. 29.Ott. 2, 1968. C3 Copyright 1969
American institute of Mining, Metallurgical, and Petroleum Errgl.
tive amounts, not the actual quantity, of hydrocar- neers, Inc.

670 JOURNAL OF PETROLEUM TEHINOM30Y