Using Oil Geochemistry to Predict Vertical and Lateral Variations in Oil Quality

In Section I of this article, we discuss how oil geochemistry can be used to derive oil viscosity and gravity data from analysis of sidewall cores or conventional cores. In Section II, we discuss how oil geochemistry can be used to develop predictive models by characterizing the filling history and the PVT conditions at which the fluids were emplaced. In Section III, we discuss how oil geochemistry can be used to better define gas/oil contacts and oil/water contacts. In Section IV, we discuss how oil geochemistry can be used to determine fluid properties for oil samples contaminated with oil-based mud or other drilling additives.

I – Predicting Oil Gravity and Viscosity from Core Extracts

Four methods are available for acquiring oil gravity and viscosity data for an oil accumulation:

1. Estimation of oil properties based on analogy to nearby accumulations. Obviously, this approach is limited by the availability of analogous oils, and by the geological validity of the analogy.
2. Direct measurement of oil properties using oil derived from RFT, MDT, or DST testing of an interval. RFT, MDT, and DST samples are typically not available due to the great expense of sample acquisition. Even when RFT testing is done, the primary objective is typically to acquire pressure data; as a result, fluid samples are often not collected or not saved for subsequent analysis.
3. Direct measurement of oil properties using oil derived from centrifuging core material. This approach has limited applicability because:
   • Under most circumstances, insufficient oil can be centrifuged from either conventional core or sidewall cores to allow physical properties to be directly measured.
   • Evaporative loss of low-molecular-weight hydrocarbons from core during storage and handling will raise the apparent viscosity and lower the apparent gravity of oil centrifuged from the core.
   • Using solvent (in place of centrifugation) to extract oil from core will cause the extract to be of limited use for gravity and viscosity measurements, since it is not possible to remove all the solvent from the extract without also losing much of the light ends of the oil.
4. Estimation of oil properties by using a transform to convert a geochemical analysis of a core extract into a calculated gravity and/or viscosity value.

This last approach is the most useful of the four, and is discussed below.

Differences in oil gravity between fields or between reservoirs within one field are a consequence of:

• Differences in oil source-rock characteristics, such as source-rock type and/or thermal maturity.
• Differences in post-generation processes, such as (i) oil biodegradation, (ii) evaporative fractionation during gas migration, (iii) in-reservoir deasphalting of oil, and/or (iv) water-washing at the oil/water contact

Oil geochemistry can reveal precisely which of these processes control gravity and viscosity variations in a given field. Once the cause of the variation is determined, then a geochemical parameter sensitive to that process can be measured on a set of produced oils covering the gravity or viscosity range. A cross-plot of those geochemical data against measured viscosity or gravity then yields a transform that can be used to convert geochemical data from core extracts into values of viscosity or gravity.

Baskin and Jones (1993) demonstrated the applicability of this approach to predicting the gravity of Miocene Monterey Formation oil (California, USA) from geochemical analyses of sidewall core and ditch cuttings extracts. The oil gravity variations in their study were due to differences in oil maturity. Therefore, they constructed their geochemical/gravity transforms using geochemical parameters sensitive to maturity. That study saved millions of dollars by avoiding costly DST tests in unproductive intervals during Chevron’s development of the off-shore Monterey (D. Baskin, personal communication to M. McCaffrey).

Smalley et al. (1996) and McCaffrey et al. (1996) demonstrated how oil viscosity in biodegraded heavy oil accumulations can be predicted from core and cuttings extracts prior to well testing by constructing geochemical/viscosity transforms using oil geochemistry parameters sensitive to biodegradation. Specifically, Smalley et al. (1996) noted that lateral and vertical variations in oil quality (viscosity, gravity) in a North Slope field were controlled by two phenomena:

1. Lateral and vertical variations in the magnitude of oil biodegradation (a process that increases oil viscosity and decreases oil gravity), and
2. Lateral and vertical variations in the relative abundance of a secondary, condensate-like charge, a material that reduces the oil viscosity and raises the oil gravity.

Smalley et al. (1996) then identified oil geochemistry parameters that are sensitive to the degree of oil biodegradation and to the quantity of the secondary charge. The authors then developed transforms that related those oil geochemistry parameters to oil quality (viscosity or gravity). They then used those transforms to predict oil quality from geochemical analysis of sidewall cores. McCaffrey et al. (1996) used a similar approach to predict oil viscosity in a biodegraded heavy oil accumulation in the San Joaquin Valley, California.

In cases where oil property variations are due to varying contributions of two oil types from two discrete sources, biomarker parameters sensitive to oil source can be used to construct the geochemical/oil-property transforms.

II – Understanding the Impact Of Fluid Filling History on Oil Quality

The discussion above has focused on how oil geochemistry can be used to determine the quality of oil (viscosity, API gravity) in a sidewall core or core sample. However, oil geochemistry can also be used to develop predictive models of how fluid properties are likely to vary laterally and vertically in a field.

When oils are biodegraded, models can be constructed which seek to predict lateral and vertical variations in oil biodegradation (e.g., Larter et al., 2006). Such vertical and lateral variations in the extent of biodegradation typical fall into two categories:

1. Variations due to distance from the oil-water contact. Because biodegradation occurs at or near the oil-water contact (e.g., Head et al., 2003), biodegraded oil columns commonly are compositionally graded, with the most biodegraded oil occurring near the oil-water contact.
2.  Variations due to the “pulsed” nature of reservoir filling. Since the time scale of biodegradation is often similar to the time scale of reservoir charging (e.g., Larter et al., 2003), a biodegraded oil column may consist of a mix of oil that arrived first in the reservoir (here called the “primary charge”) and subsequent pulses of oil that arrived  later (here called the “secondary charges” to the accumulation).  The primary charge may be more biodegraded than the secondary charge, since the primary charge has been subjected to in-reservoir biodegradation for a longer period of time. Therefore, depending on the migration pathways into the reservoir, vertical variations in the relative abundance of “primary” vs. “secondary” charges may cause vertical variations in the oil fluid properties (e.g., API gravity and viscosity).

In NONDEGRADED oil accumulations, the role of geochemistry in modeling fluid property variations is quite different.  For example,  gas chromatography (GC) data can be used to perform a "Slope Factor Analysis" in which the paraffin distribution in an oil is compared with an existing PVT dataset to characterize the PVT conditions under which the oils were originally emplaced into the reservoir (e.g., Thompson 2002, 2003, 2006).   In this approach, plots of:

log (n-alkane concentration)  vs. Paraffin Carbon number

are constructed, and the resulting slopes in the different regions of the plot are compared with data from a large PVT dataset.  This approach reveals the PVT conditions under which the petroleum was originally emplaced, and also whether or not there has been a multi-part charge history.

Development geologists can construct predictive models of how fluid properties are likely to vary laterally and vertically in a field by integrating an understanding of the field geology with an understanding of the PVT conditions of fluid emplacement and the number of discrete charge events (derived from Slope Factor analysis).

III – Identifying Fluid Contacts (Gas/Oil/Water) from Core Extracts

In addition to helping predict oil viscosity and gravity, oil geochemistry (oil fingerprinting) can be used to better identify fluid contacts in cases where wireline log data are not definitive. Wireline log experts can typically assess the type of reservoir fluid (oil/gas/water) in sand-shale sequences by using a combination of (1) a neutron-density tool that detects low hydrogen and low electron densities typical of gas zones, and (2) a repeat formation tester (RFT), which uses both the pressure gradient and sample acquisition techniques to evaluate reservoir fluid. However, in some areas, sands exhibit a poor neutron-density response to gas, and RFT testing is not performed due to poor hole conditions and fear of tool loss. In such cases, geochemical fingerprinting of residual hydrocarbons chemically extracted from sidewall core samples can provide an independent means of identifying reservoir fluid type. Baskin et al. (1995) discuss this approach in detail, using a Niger Delta case study to illustrate how geochemistry can be used to either identify or corroborate fluid contacts from core or sidewall core analyses.

IV- Predicting Oil Occurrence and Quality in Samples Contaminated with Oil or Synthetic - Based Drilling Muds

Presently, many oil and gas wells are being drilled with non-water-based drilling muds. Nonaqueous mud bases include diesel oil, petroleum mineral oil, and synthetic fluids. The non-water-based drilling muds were developed to:

• Provide more consistent drilling performance
• Facilitate drilling of cleaner, more stable holes with less sloughing
• Facilitate drilling of difficult horizontal or extended reach wells
• Minimize the incidence of stuck tools
• Create a lower volume of drill cuttings
• Enable recycling of the mud (i.e., multiple uses on multiple wells)

Oil-based muds (OBM) incorporate diesel or petroleum mineral oil as their base. These muds are typically harmful to the environment and may require complicated disposal procedures, whether the drilling is on land or offshore. Synthetic fluid based muds (SBM) use α-olefins, esters, or ethers as their base. SBM formulations are expensive but have:

• Lower toxicity
• Faster biodegradability
• Lower bioaccumulation potential
• Greater potential for recycling

Oil-based and synthetic-based drilling filtrates rapidly invade the well bore region during drilling, and the core (conventional) during coring. OBM and SBM fluids mix with or displace the in situ fluids (Lugol et al., 2000 and Wenger et al., 2003). If the formation fluids are not completely displaced, then they are often highly contaminated with the oil or synthetic base fluids. Consequently, short-term tests such as RFT’s and MDT’s typically collect highly contaminated fluids in their chambers, since the testing program fails to remove all of the drilling fluid from the well bore region prior to filling the chamber (Hashem et al., 1999).

Any geochemical fluid characterization (such as assessment of the presence of hydrocarbons in the reservoir or assessment of the quality of the oil) will be impacted by the oil or synthetic base used in the OBM or SBM. If the formation fluid has been completely flushed from the test region of the well bore, then analysis of the RFT fluids or fluids extracted from conventional core or sidewall core will not yield accurate prediction of formation fluid properties.

If the OBM or SBM filtrate has not completely displaced the formation fluid, then geochemical analysis of the RFT fluid or core extract may yield results adequate to determine formation fluid type and a qualitative prediction of the oil properties (e.g., Schafer, 1992). Success depends on the type of formation fluid and the hydrocarbon type in the tested formation (Table 1). Prediction of oil properties is more successful when:

• The drilling mud contamination is low, and/or
• The OBM or SBM base is comprised of only a few components, and/ or
• The formation contains non-biodegraded, moderate gravity oil.

The potential for success is unknown when wells are drilled with mineral oil-based mud unless the chemical makeup of mineral oil is known and contains a different range of hydrocarbons than the formation oil (i.e., mineral oil may range from complex mixtures such as crude oil to a limited distillation or boiling point range).

Hydrocarbon Type	OBM Diesel-base	OBM Mineral oil-base	SBM α-olefins, esters, or ether base
Gas	Poor	Unknown	Good
Condensate	Poor-fair	Unknown	Good
Non-biodegraded oil	Fair	Unknown	Good
Biodegraded or immature oil	*Poor-fair	Unknown	*Poor-good

* Success depends on level of biodegradation; heavy-severely biodegraded oils have poor success potential.

The same geochemical techniques are used with OBM and SBM contaminated oils as described earlier in this section. Analytical methods may include high resolution gas chromatography (oil fingerprinting approach) and biomarker analysis. Sample RequirementsAccurate oil property predictions require high quality samples. The list below shows the sample type and sizes required for analysis. The sample types are in order of best to worst.

1. RFT sample — 1 quart of whole fluid from the test interval.
2. Conventional core — several 2" cubes from the test interval. The cubes should be taken from the center of core preferably away from the invasion halo. Following collection, the samples should be immediately wrapped in foil and frozen to minimize loss of light ends (see Sampling Techniques).
3. Sidewall core — several whole cores from the tested interval. SWC’s are the poorest choice of sample as they are usually highly contaminated. Following collection, the samples should be immediately wrapped in foil and frozen to minimize loss of light ends (see Sampling Techniques).

In addition to the samples above, a one quart sample of the whole mud used at the time of testing is required. OilTracers LLC requires this, since oil- and synthetic-base muds are recycled and may be contaminated with crude oil prior to testing the formation.

For more information on the geochemical techniques described here, or to discuss a specific project, e-mail us a info@oiltracers.com, or call us at (214) 584-9169.

References

Baskin D. K. and Jones R. W. (1993). Prediction of oil gravity prior to Drill-stem testing in Monterey Formation reservoirs, offshore California: AAPG Bulletin  v. 77 (9), 1479-1487.

Baskin D. K., Hwang R. J. and Purdy R. K. (1995). Predicting gas, oil, and water intervals in Niger delta resevoirs using gas chromatography: AAPG Bulletin v. 79 (3), 337-350.

Hashem, N. M., E. C. Thomas, R. I. McNeil, and O. Mullins, 1999, Determination of Producible Hydrocarbon Type and Oil Quality in Wells Drilled with Synthetic Oil-Based Muds: Society of Petroleum Engineers, April 1999

Head, I. M., D. M. Jones, and S. R. Larter, 2003, Biological activity in the deep subsurface and the origin of heavy oil: Nature, v. 426, p. 344-352.

Larter, S., H. Huang, J. Adams, B. Bennett, O. Jokanola, T. Oldenburg, M. Jones, I. Head, C. Riediger, and M. Fowler, 2006, The controls on the composition of biodegraded oils in the deep subsurface: Part II - Geological controls on subsurface biodegradation fluxes and constraints on reservoir-fluid property prediction: AAPG Bulletin, v. 90, p. 921-938.

Larter, S., A. Wilhelms, I. Head, M. Koopmans, A. Aplin, R. Di Primio, C.Zwach, M. Erdmann, and N. Telnaes, 2003, The controls on the composition of biodegraded oils in the deep subsurface-Part I: biodegradation rates in petroleum reservoirs: Organic Geochemistry, v. 34, p. 601-613.

Lugol R. L., C.A. Dodman and J. Crowe, 2000, On the Quest for Ester-Base Mud System Impact on Data Acquisition, SPE Annual Technical Conference and Exhibition held in Dallas, Texas, 1-4 October 2000, Society of Petroleum Engineers Inc.

McCaffrey M. A., Legarre H. A. and Johnson S. J. (1996). Using biomarkers to improve heavy oil reservoir management: An example from the Cymric field, Kern County, California: AAPG Bulletin v. 80 (6), 904-919.

Schafer, D., 1992, Rock and Oil Properties Used to Confirm and Differentiate Proven Pay and Low Resistivity Pay in Wells Drilled with Oil-Base Mud, Offshore Gulf of Mexico: AAPG National Convention 1992, Calgary, Oral presentation (abstracts book p.114.)

Smalley, P. C., N. S. Goodwin, J. F. Dillon, C. R. Bidinger, and R. J. Drozd, 1996, New Tools Target Oil Quality Sweetspots in Viscous Oil Accumulations: SPE Paper No. 36652, p. 911-917.

Thompson, K. F. M., 2006, Mechanisms controlling gas and light end composition in pyrolysates and petroleum: applications in the interpretation of reservoir fluid analyses: Organic Geochemistry, v. 37, p. 798-817.

Thompson, K. F. M., 2003, Interpretation of charging phenomena based on reservoir fluid (pvt) analyses, in J. Cubbitt, W. England, S. Larter, and G. Macleod, eds., Conference Abstracts: Geochemistry of Reservoirs II:  Linking Reservoir Engineering and Geochemical  Models (Geological Society of London, February 3-4, 2003), Geological Society of London.

Thompson, K. F. M., 2002, Compositional regularities common to petroleum reservoir fluids and pyrolysates of asphaltenes and kerogens: Organic Geochemistry, v. 33, p. 829-841.

Wenger, L. M., J. R. Gormly, J. M. Evensen, and L. C. Davis, 2003, Geochemical Evaluation in Deepwater Wells: Impact of Modern Drilling Mud Systems and Fluid Testing Procedures: Abstracts 21st International Meeting on Organic Geochemistry, September 8-12, 2003, Krakow, Poland, p.OXXIV/1.