REVIEW OF LITERATUR.....

SPE Society of Petroleum Engineer'S

SPE 21004

Characterization of Tar From a Carbonate Reservoir in Saudi Arabia: Part I-Chemical Aspect A.S. Harouaka* and H.K. Asar, * KFUPM/RI; AA AI-Artaj and A.H. AI-Husaini, KFUPM; and W.A. Notal, KFUPM/RI 'SPE Members

This paper was prepared for presentation at the SPE International Symposium on Oilfield Chemistry held in Anaheim, California, February 20-22, 1991.

This paper was selected for presentation by an SPE Program Committee following review of information contained in an abstract submitted by the author(s). Contents of the paper, as presented, have not been reviewed by the Soci~ty of Petroleum Engineers and are sUbject to correction by the author(s). The material, as presented, does not necessarily reflect any position of the Society of Petroleum En~meer~, ItS officers, or members. Papers presented at SPE meetings are subject to publication review by Editorial Committees of the Society of Petroleum Engineers. Permission to copy IS restricted to an abstract of not more than 300 words. Illustrations may not be copied. The abstract should contain conspicuous acknowledgment of where and by whom the paper is presented. Write Publications Manager, SPE, P.O. Box 833836, Richardson, TX 75083-3836 U.S.A. Telex, 730989 SPEDAL.

ABSTRACT

Tar. mats are extra heavy oil zones sandwiched between aquifers and adjoining oil columns. They isolate either partially or completely an oil reservoir from its aquifer. This results in a rapid pressure drop, a premature high gas-oil ratio and a low primary oil recovery; all of which point to some form of pressure maintenance early in a field's life.

Eventhough tar mats represent considerable hydrocarbon reserves as they are rather common in the Middle East and Africa, it is their impact on oil recovery from adjoining oil columns which is of interest for the time being. Tar must be characterized to evaluate its mobility and ways of establishing contact between an oil column and its aquifer as well as to design an optimum water injection scheme.

The present paper discusses a detailed chemical characterization of tar from a carbonate reservoir in Saudi Arabia. Thermal stability variation was evaluated by thermal gravimetry' (TG), and differential thermal analysis (DTA). Elemental analysis of preserved and non preserved samples were carried out with a Carlo Erba 1106 elemental analyser. The sulfur content was also determined by two different ASTM methods. The presence of Nickel, Vanadium and Iron, the major metals usually found in hydrocarbons, was investigated by X-ray fluorescence (XRF) spectroscopy. The

89

tar major hydrocarbon group components were separated and quantified by high performance liquid chromatography (HPLC).

Experimental results showed variation, in tar properties, with depth and area within the same field. The carbon to hydrogen ratio increased systematically with a decrease in API gravity. The sulfur contents obtained with the Carlo Erba elemental analyser were in good agreement with those obtained by the general bomb method (ASTM D-129-64). The content of Hexane insolubles was relatively high at about 38 % by weight. The polar compounds ranged between 5 and 9 % by weight.

INTRODUCTION Experimental work has been conducted at The

Research Institute of KFUPM to characterize tar from the tar mat of a carbonate reservoir in Saudi Arabia. This reservoir will be referred to as Field A throughout the remainder of this paper.

As indicated earlier, tar occurence is generally between an oil column and its aquifer. This situation creates several problems from a production standpoint, to say the least, keeping in mind that a tar zone or mat can be as thick as the oil column above it.

Tar mats are rather common in the Middle East and their importance as a hydrocarbon resource has been discussed elsewhere (1). However, very little is known about tar properties and their variation with depth and area.

2 CHARACIERIZATION OF TAR FROM A CARBONATE RESERVOIR IN SAUDI ARABIA - PART I: CHEMICAL ASPECT

SPE21004

Tar characerization is an important part of an ongoing investigation to evaluate tar mobility and methods of improving injectivity in tar mats. As such, this work has been directed towards a characterization mainly from a reservoir engineering standpoint.

Experimental results showed that several tar properties do vary with depth and area within the same field.

This paper presents results of a laboratory evaluation of several tar chemical properties. The samples analyzed were either extracted from non preserved cores or retrieved as bottomhole fluid samples. Both extracted and bottomhole samples were from the tar mat of Field A. Part I of this paper presents the chemical aspect of this characterization while part 2 is devoted to the physical aspect.

SAMPLING

It is well known that representative tar samples are extremely difficult to obtain. There are basically two methods for retrieving a fluid sample from a tar mat:

I) by extraction from cores or as a

2) bottomhole fluid sample.

Experimental work was carried out on several extracted and bottomhole samples from the tar mat present in Field A.

EXTRACTED TAR SAMPLES

The extraction process was conducted in two steps: Dean Stark extractors were used to determine the water content of core plugs while tar extraction was conducted in Soxhlet extractors with Toluene as a solvent. Few samples were extracted with Carbon disulfide and the tar-solvent separation was carried out at room temperature.

A good material balance along with a simulated distillation of the solvent used were the main criteria for evaluating the outcome of an extraction process. Another criteria has been the comparison of thermogravimetric analysis (TGA) and simulated distillation for a tar sample extracted with toluene to similar tests conducted on a twin sample

90

extracted with a low boiling point solvent such as Carbon disulfide.

Unfortunately, the cores available for this investigation were not preserved. Extracted tar samples were considerably altered by aging and exposure. Under these conditions measured chemical properties cannot be considered representative of the reservoir conditions in Field A. However, analyses of these samples were useful for comparative purposes especially, for assessing tar properties variation with depth and area.

BOTTOMHOLE SAMPLES.

Bottomhole samples were repeat formation tester (RFT) samples. Two RFT tars, from two different wells in Field A, were made available for this investigation. One of these two samples was found severly contaminated with mercury. The cores available for extraction and the RFT fluid samples were from different wells.

RFT samples are preserved, they were retrieved, under pressure, from the test chamber of an RFT sampler. These samples are believed to be representative of the tar mat in Field A.

EXPERIMENTAL PROCEDURE AND CONDITIONS

The following is a description of the procedures and conditions under which the chemical characterization of tar samples from the tar mat in field A was achieved. Table 1 shows a brief description of some of the analyzed samples. Among the samples listed in Table I, FI is believed to be the most representative of the tar present around the tar/water contact in well WFI. Samples F5A, F6A and F7A along with F5B, F6B and F7B were extracted from a nonpreserved core. The latter has been left to the open atmosphere for few months before extractions took place.

ELEMENTAL ANALYSIS

Amounts of Carbon, Hydrogen, Nitrogen, Oxygen and Sulfur present in a tar sample were obtained with a Carlo Erba 1106 elemental analyser. Two additional methods (ASTM D-1551- 68 and ASTM D-129-64) were used to determine

SPE21004 A. S. HAROUAKA, H. K.ASAR, A. A. AL-ARFAJ, A. R. AL-HUSAINI, W. A. NOFAL 3

the sulfur content of the tar samples under consideration.

Measurements were carried out in duplicate for reproducibility purposes. The level of accuracy of the elemental analysis has been evaluated as follows: Seven parts from the same tar, taken at random, were analysed for Carbon, Hydrogen and Nitrogen content.

HYDROCARBON GROUP DETERMINATION

High performance liquid chromatography (HPLC) is probably the best technique available for separation and quantification of hydrocarbon groups from tar and heavy crude samples.

Several preserved and nonpreserved tar samples from the tar mat in Field A were analyzed with a High Performance Liquid Chromatograph (HPLC). The column applied was a specialised 3.9 mm * 30 cm energy analysis (NH2) liquid chromatographic column designed specifically for hydrocarbon separation.

The calibration procedure has been proposed by the manufacturer. Briefly, this calibration was achieved by collecting the fraction (aromatics and polars) peaks, evaporating the solvent and weighing the fractions. The response factor or the fraction concentration (weight %) divided by the fraction peak area was calculated for each group. Sample F6A was utilized to conduct the calibration.

With three fractions known by weight (Hexane insolubles, Aromatics and Polar compounds), the Saturates were calculated by the difference. Response of the differential in refractive index between Hexane and sample components was extremely smalL A determination of the Saturates fraction based on the refractometer response would lead to large errors. Also, determination by difference gave a better reproducibility.

X-RAY FLUORESCENCE

X-ray fluorescence (XRF) analyses of nine tar bearing rock samples were conducted to determine the presence, if any, of metals such as Nickel, Vanadium and Iron. In energy dispersive X-ray fluorescence spectroscopy, only those elements

91

having an atomic number of 11 or more can be detected. Also, the sensitivity to low atomic number elements is not good because Mylar foil was utilised and the specimen chamber of the X-ray fluorescence spectrometer was flushed with helium. The smallest aperture collimator was used. The main components of tar must be carbon, hydrogen and eventually oxygen, none of which can be detected by X-ray fluorescence.

Clays and silts with a two microns diameter or less can mix with tar during the extraction process. An extracted tar sample taken at random was dissolved in Hexane (two gram of tar in 80 ml of Hexane). After filtration through a 0.45 micron Millipore filter, the dry part was analysed by XRF along with tar bearing rock samples.

THERMAL ANALYSIS

The thermal analysis included thermal gravimetry (TG) and differential thermal analysis (DTA). The simultaneous thermal analyser used in this investigation performs TG and DTA simultaneously. The temperature of the sample is measured with thermocouples of Platinum and Platinum plus 10 % Rhodium.

The parameters recorded were temperature , change in weight, derivative of change in weight and the difference in temperature between the tar sample and a reference.

The instrument was calibrated using Ammonium Nitrate. Helium was employed to generate a nonoxidizing atmosphere while Oxygen provided an oxidizing atmosphere.

RESULTS AND DISCUSSIONS

The chemical characterization of several tar samples from the tar mat in Field A included sulfur and overall elemental analysis, hydrocarbon group separation and quantification, TGA and DTA. Whenever feasible data from the literature were added for comparative purposes.

A comparison of the properties of samples F5A, F6A and F7 A gives a good indication of these properties variation with depth. Similar indications are evident from a comparison of samples F5B, F6B and F7B properties. Areal properties variation

4 CHARACfERIZATION OF TAR FROM A CARBONATE RESERVOIR IN SAUDi AKAHIA - PAKT 1: CHEMICAL ASPECT

may be illustrated by comparing the properties of sample F2 and either F7A or F7B.

ELEMENTAL ANALYSIS

The results of the elemental analysis are summarized in Table 2. Amounts of Carbon, Hydrogen, Nitrogen, Oxygen and Sulfur present in a tar sample were determined. The results are shown in Table 2 unless the concentration was at or below the detection limit of the instrument.

It can be seen from the data summarised in Table 2 that the total concentrations of the analyzed elements (CHNOS) exceed 94.8 % indicating that other elements soluble in Toluene (and Carbon disulfide for samples F5A, F6A and F7 A) not included for analysis were, as expected, only present in minute quantities. The C/H ratio increased as the API gravity decreased.

Tar samples extracted with Carbon disulfide may show a sulfur content above normal probably due to solvent entrapment within the tar. Carbon disulfide has a low boiling point and the tar-solvent separation is carried out at room temperature. This procedure may and usually does inflate the sulfur content of the sample extracted under these conditons.

The level of accuracy of this elemental analysis has been evaluated, under prevailing experiemntal conditions. The results of this evaluation are shown in Table 3. Reproducibility with respect to carbon and hydrogen was very good. Nitrogen content was at or below the detection limit of the instrument.

SULFUR ANALYSIS

The sulfur content obtained with the Carlo Erba 1106 elemental analyser was compared to the results of the Quartz tube method (ASTM D-1551-68) and the general bomb method (ASTM D-129-64). The results are shown in Table 4. The data obtained with the quartz tube method are somewhat different. This method is recommended for the determination of sulfur in concentrations ranging from 0.1 to 5 % by weight. It is reported to give inaccurate results when applied to samples containing phosphorous, nitrogen and metallo-organic compounnds. Based on the fact that the results of the general bomb

92

method agree with those of the Carlo Erba analyser, the data obtained using these two methods are believed to be more reliable.

X·RAY FLUORESCENCE ANALYSES

Table 5 shows the results of sample F6A XRF analysis. Detected elements were normalised to 100 %. For instance, the sulfur content of sample F6A is about 4.3 % by weight, as indicated in Table 4.

Nickel and Vanadium contents of several tar bearing rock samples, from the same depth as samples F5A, F6A and F7 A, were below the detection limit of the instrument.

HYDROCARBON GROUP DETERMINATION

The Hexane insoluble content of the tar samples analyzed ranged between 13 and 44 % by weight, while the content of aromatics and polar compounds ranged between 38 to 66 % and 4 to 17 % by weight respectively. Saturates were determined by the difference and their content ranged between 2 to 30 % by weight. Saturates determination is considered the least accurate mainly because of the way it has been conducted, that is by difference. The detailed hydrocarbon groups determination results are displayed in Table 6. Other results from samples taken outside Field A were also included in Table 6 for comparative purposes.

Within the scope of the samples analysed there is a noticeable difference in fractions distribution when comparing samples from different wells. However, the differences between fractions distribution of tar samples taken from the same well but at various depths are considered to be inconclusive or even within experimental errors.

Elemental analyses of the tar samples presented in Table 2 showed a good agreement between the variation in N,O,S content of tar samples and the variation in the content of HPLC polar compounds. For instance, the sulfur content of samples F5A, F6A and F7A ranged between 4.2 and 4.7 % by weight. This gradual increase is relatively close when compared to the HPLC data for the same samples where the polar compounds ranged between 4.77 and 6.00 % by weight. A similar comparison

SPE2IOO4 A. S. HAROUAKA, H. K.ASAR, A. A. AL-ARFAJ, A. R. AL-HUSAINI, W. A. NOFAL 5

3)

2)

5)

4)

1) The presence of a tar mat disrupts natural water drive as the reservoir is partially to completely isolated from the aquifer beneath. The knowledge of tar properties becomes essential for an efficient management of the oil column above the tar zone.

Tar properties can vary with depth and area within the same field. This variation becomes more pronounced in the neighborhood of the tar/water contact.

SUMMARY AND CONCLUSIONS

ACKNOWLEDGEMENT

The authors wish to acknowledge the support of Saudi Aramco for this work under KFUPMjRI project No. 21061.

The contribution -to this work by H. Alpustun, B. Mtawea, A. Fuseni and A. lob is also gratefully acknowledged.

The aging factor must be taken into consideration for a proper interpretation of the generated chemical paramet~rs. Since it is v~ry difficult and costly to obtam a representative tar sample, non preserved samples may be considered provided the aging effect is well defined.

Field A tar has a high content of Hexane insolubles, above 36 % by weight. The content of Aromatics and Polars ranged between 38 to 66 and 4 to 7 % by weight respectively.

The HPLC technique is a fast and adequate way for the separation and quantification of hydrocarbon groups from tars and heavy crudes. However, major errors may result from weighing and the peak collection process.

6) The elemental analysis showed a consistent trend, as the API gravity decreased the C/H ratio increased. Three methods were used to determine the sulfur content of various tar samples from Field A. The results obtain~d with the Carlo Erba elemental analyser were In good agreement with those obtained by the general bomb method or ASTM D-I29-64.

THERMAL ANALYSIS

between samples FI and -F2 shows that the (N,O,S) content of samples FI and F2 varied between 5.00 and 5.90 % by weight. This relative variation matched the HPLC results where the content in polar compounds for sample FI was 5.62 % by weight and that of sample F2 was higher at 8.6 % by weight.

TGA tests for samples Fl and F2 along with a refinery residue were conducted at a heating rate of

18 of/min in a Helium atmosphere. The data are shown in Figure 1. For the refinery residue and sample F2 (extracted tar), significant weight losses were not observed until the temperature exceeded

300 OF. However, for sample FI (dead RFT tar), a gradual weight loss started at the beginning o~ ~he test, showing the presence of low volatilIty components. Clearly, aging and exposure have stripped sample F2 from most of its light components. At a higher temperature range, the thermal weight loss characteristics of samples FI and F2 were quite similar a indicated in Figure 1.

Sample F4 is a separator oil which was distilled until it showed the same viscosity as sample FI under reservoir conditions of pressure and temperature. The thermograms for samples Fl and F4 are given in Figure 2.

Thermograms for samples F5A, F6A and F7A indicated that sample F7A, which was taken near the tar/water contact, has more heavier components than samples F5A and F6A. The differential thermal analyses of the latter samples produced a marked difference between the three samples when a nonoxidative atmosphere is compared to an oxidative one. Figures 3,4 and 5 show the DTA for samples F5A, F6A and F7 A respectively. The endothermic peaks are shown downwards and the exothermic peaks upward.

TGA and DTA results clearly indicate that tar lighter compnents content decreased with depth. The same holds for the impact of an oxidative (air) atmosphere which diminished with depth.

93

6 CHARACTERIZATION OF TAR FROM A CARBONATE RESERVOIR IN SAUDI ARABIA - PART I: CHEMICAL ASPECT

SPE 21004

REFERENCES

1) Harouaka A.S. and Asar H.K.:"Tar Mats Evaluation-A Resource and a Nuisance,"First Saudi Symposium on Energy Utilization and Conservation, March 4-7, 1990, Jeddah Saudi Arabia.

2) Chirinos, M.L., Gonzalez, J. and Layrisse I.: "Rheological Properties of Crude Oil From the Orinoco Oil Belt and Their Mixtures With Diluents,"Rev. Tec. Intevep July, 1983.

3) Starr, J., Prats, J.M. and Messulam, S.A.:"Chemical Properties and Reservoir Characteristics of Bitumen and Heavy Oil From Canada nad Venezuela, "First International Conference on the Futur of Heavy Crude and Tar Sands Sponsored by UNITAR, AOSTRA and U.S. DOE, June 4-12, 1979.

4) Hasan, M.U., Ali, S.M. and Bukhari, A.:"Structural Characterization of Saudi Arabian Heavy Crude Oil by NMR Spectroscopy,"Fuel, May, 1983.

94

TABLE 1

* This sample has been distilled until its viscosity matched that of sample FI under reservoir conditions.

Fluid Samples Description

Sample ID

FI

F2

F3

F4*

F5

F5A

F6A

FlA

F5B

F6B

FlB

F8

Depth Interval

Bottom of tar mat

Bottom of tar mat

Top of tar mat

Middle of tar mat

Bottom of tar mat

Top of tar mat

Middle of tar mat

Bottom of tar mat

Bottom of tar mat

Comments

RFT sample from well WF1.

Toluene extracted- core from well WF2.

Separator oil from well WF 1.

Distilled separator oil from well WF1.

Distilled separator oil from well WF3.

CSz extracted - core from well WF4.

CSz extracted - core from well WF4.

CSz extracted - core from well WF4.

Same as F5A but toluene extracted.

Same as F6A but toluene extracted.

Same as FlA but toluene extracted.

RFT sample from well WF3 contaminated with mercury.

SPE 2 1004

TABLE 2

Elemental Analysis

Tar/Oil Sample (.API) Element (wt %)

C H CIH N a S Total

Saskatchewan Heavyffi (14.2) 83.0 11.0 7.55 0.5 <0.1 3.0 -97.6

Venezuela Heavl (11.0) 0.7 5.2

Athabasca Heavy3 (8-10) 83.1 10.6 7.84 0.4 1.1 4.8 100.0

Fl m

(10.7) 79.6 10.2 7.80 0.5 <0.1 4.4 -94.8

F2m (10.2) 83.0 10.5 7.91 0.5 1.2 4.2 99.4

Saudi Heavy 4

(7.7) 84.1 10.7 7.86 0.2 5.0 -100.0

F3 (-28) 85.0 12.1 7.07 0.1 2.8 100.0

F4 (-15) 85.1 11.4 7.46 0.2 3.6 100.3

F5A ( ) 84.8 10.9 7.78 4.2 -99.9

F6A ( ) 84.5 10.7 7.90 4.3 -99.5

FlA ( ) 84.0 10.3 8.16 4.6 -98.9

m: Values measured at KFUPM/RI.

2,3,4: Values from the literature, the supercripts correspond to entries in the reference list.

TABLE 3

Statistical Analysis of the Elemental Measurements

TABLE 4

Sulfur Analysis (% by weight) Run #

2

3

4

5

6

7

Mean

Std. dev.

ReI. std. dev.

Precision at 95% confidence level

%N

0.2

0.33

0.2

%C %H

84.48 11.29

11.36 Tar/Oil Sample· Carlo Erba Bomb Quanz-Tube84.82

Analyzerl MethodZ Method3

85.06 11.34

84.66 11.37 Saskatchewan Heavy 3.0 3.50 2.75

84.77 11.38 California Heavy 6.2 6.10 5.95

84.80 11.39 FI 4.4 4.21 3.84

84.72 11.37 F2 4.2 4.80 3.31

84.76 11.36 F5A 4.2 4.2

0.176 0.0336 F6A 4.3 4.1

0.21 0.3 FlA 4.6 4.7

0.41% 0.60% I: Model 1106 2: ASTM 0-129-64 3: ASTM D-1551-68

95

•

TABLE 5

X-ray Fluorescence Analysis of Sample F6B

Component Conc (%)

S 84.266

Ca 6.725

V 1.771

Fe 2.422

Ni 0.567

Cu 1.759

Zn 2.490

TABLE 6

Hydrocarbon Group Type Characterization

Hydrocarbon groups (wt %)

Hexane Sample Saturates Aromatics Polars Insolubles

F5A 17.20 40.65 5.29 36.86 F6A 17.24 42.00 5.29 34.76 • F7A 13.46 43.51 4.77 38.26 F6B 14.26 38.00 4.64 43.10 F3 29.23 47.31 9.58 13.88 F4 2.43 65.47 16.40 15.70 FI 15.29 39.79 5.62 39.30 F5 6.76 43.60 7.70 41.94 F2 4.93 43.17 8.60 43.30 F8 13.37 26.74 3.87 56.02·· California Heavy 4.32 58.00 12.00 25.68 Saskatchewan Heavy 7.46 62.00 18.00 12.54

Sample F6A was used to calculate the aromatic and polar response factors .

•• Sample F8 was contaminated with mercury.

The most precise value is the aromatics content since they elute without backflushing.

The least precise is the saturates value, it was obtained by difference. However, it showed good reproductivity.

100

90

80

70

g 60e..

E 50.. c..... .r;

40en "iii ~

30

20

10

Chevron Tor

F2

**** Fl

1050 1200 1350 1500 O+-----'f----+---+---+----1---+--+---I-----+---f

o 150 300 450 600 750 SOO Temperature (Deg F)

Figure 1 Thermogravimetric Analysis of Three Tar Samples

96

SPE 21004

100

90

80

70 g Ol 60

.IE: c 'S

50E '"0::

-oJ 40-§, '; :t

30

20

10

O-l---+--~--+---I----1---I--+-':""""'-+--~--l

o 150 300 450 600 750 900 1050 1200 1350 1500 Temperature (Oeg F)

~~~~o~ravimetrie Analysis in Air Medium for Samples F1 and F4

u .~

:5 o '0 C

W

Air Medium

Helium Medium

.. _ .

200,T'""-------------------------......,u

~ "........."-....... ::g

, , x O+-'.,---r-----,r---:'f/'-'--..--."":-....--r----,r---._-.--,-__.,-_w"-'-l

-..:: _----_ ~ .;,,' ; "'. .. .....:------.. :

. ~~---- ......,.. ,:.t'

\\ / ~ : ! : .' ..' ..' ..' . : ~ ; 'V

100

....... !! o > e -100 u ~ ~-200

~ ~

"2 -300 '" ~

:!- -400 D ~ -500 f

~ -600a

-700

200 400 600 800 1000 1200 1400 1600 1800 2000 2200 Temperature (Oe9 F)

-800-1----------------------------' o

Figure 3 Differential Thermal Analysis for F5A Tar Sample

97

100

2oo-r---------------------------TIu-, .~

~ ~

//"" ....// \ ~ :B O+"='""-...,.---r---+...----..--"7.,..---r---r---":T""-,'...,.---r---'w~_j "0 r:...... .. ..' \ I U > '. '------------,' ~ .~ ~ -100 . -"":::::~::-:>",. i ~ '-' -200 ~.~ '\: ~ f 0_. \ :

~ -300 '-~ r\ : GJ ~ : \ :

: ~:: Ai, "0&= It 1; f 'u .,:·u ., I ~ -600 Hc11um Mcd:um ~ :; V

-700

800 1000 1200 1400 1600 1800 2000 2200 Temperature (Deg F)

600400200 -800..J...--------------------------'

o

Figure 4 Differential Thermal Analysis for F6A Tor Sample

'- .......... .. --- :".'..,---------"

.......\ . f[

.~, {'f\o., ,;':.::'"\J.,

\\ Ii \ Ii \ :: ~f,.

Air Medium

100

-600

-700

15 ~ -500 ~

~ o

200.,.----------------------------u,......,

1 ~

..c: ".,-........., ... -........ 15

./ , x 2 O+...:..".......,.----,r--o'..<-/...----..--·.,,·,..-,r---...----.---,..-,............-...----r--'W-'--l "0 -------------.... > e -100 u ~ -- -200 ~ :;)

1! -300 ~

E ~ -400

800 1000 1200 1400 1600 1800 2000 2200 Temperature (Deg F)

600400200 -800.J----------------------------.J

o

Figure 5 Differential Thermal Analysis for F7A Tor Sample

98