Coring and Core Analysis (Complementary Core Information)

Complementary Core Information

Complementary Core Analysis Information

In addition to the basic data concerning porosity, permeability, and residual saturations, we can gain important information from the visual study of a core. These observable characteristics may be subsequently computerized and analyzed statistically, as in fracture study results, or they may be permanently recorded by the use of white light color photography or ultraviolet light photography. Additional information may be recorded and used in specific applications. A discussion of the more routinely recorded data follows.

Lithology, Texture, and Fluorescence

When conventional cores are taken, a detailed core description should be made. The description should list the thickness of the pay zone; the depth and thickness of shale, anhydrate, and salt zones; the presence of cross-bedding, slumping, or other features; the presence of slickensides and fractures; estimated grain size and fluorescence. In laminated zones, the measured thickness of pay and shale zones forms the basis for net pay determinations. Depositional environment can often be inferred from the occurrence of ripple marks and cross-bedding, and a coarsening or fining of grain size with depth.

Lithologic Description

The lithology of the rock is important because changes in lithology indicate variations in chemical and physical properties as well as in depositional environment. Changes in lithology give rise to changes in grain density, natural radio-activity, acoustic properties, and resistivity — all of which are important in formation evaluation and, in particular, in understanding down hole log response.

The lithologic description of core samples should follow a basic format similar to that used in the description of cuttings. Swanson (1981) has prepared a manual for the American Association of Petroleum Geologists that gives detailed instructions on sample description.

Core descriptions typically include all or a portion of the following information in the order in which it is listed.

Core Description Format:

rock type;

color;

grain or crystal size;

major characteristics;

minor characteristics;

hardness;

oil show and fluorescence.

It is beyond the scope of this module to discuss all lithological traits in detail, although some general comments, on at least two items, are in order. Hydrocarbon reservoir rock types usually break down into either clastics, composed predominantly of siliceous grains (SiO2), or carbonates, composed of limestone (CaCO₃) and dolomite (CaMgCO₃).

The color of the rock varies from all shades of gray through reds and yellows. The color may be uniform or it may be mottled, spotted, or banded. In areas where most productive rocks exhibit essentially the same color, this property may be omitted from core analysis reports.

Texture

Textural characteristics of the rock include grain size, grain size distribution, and the degree of shaliness. These characteristics influence porosity, permeability, and the interstitial water saturation. A commonly used scale for assessing grain size in clastics is the Wentworth Scale (1922). This scale, shown in Figure 1 (Wentworth scale of grain size) links medium, fine, or other reported grain sizes to specific grain diameter. Major characteristics might include the fact that the rock is laminated, or shaly, or contains fossils, as well as the minerals present (quartz, feldspar, clays, etc.).

Minor characteristics might indicate a minor percentage of accessory minerals such as mica, pyrite, or glauconite. Other important sedimentary structures, such as the presence of stylolites, should be recorded when present. Although a hardness scale exists for pure minerals, it is not used for the core description. Instead, the degree of consolidation or unconsolidation, and whether it is friable or well cemented, are traits that are noted in describing a core.

 

Oil Fluorescence

The presence and color of oil stain and oil fluorescence are also routinely recorded. Core samples that contain oil will glow when placed under an ultraviolet light. It is often possible to prove the presence of hydrocarbons by observing this fluorescence, and to predict expected production from a zone. For example, gas condensate typically appears as a uniform bright blue/white, whereas oil productive zones exhibit a uniform bright gold fluorescence. The absence of fluorescence indicates either that oil is absent, or, if oil is present, aromatic compounds have been leached from the oil in the core.

Some minerals exhibit fluorescence under ultraviolet light, but mineral fluorescence is of no interest in the core description. Where it is not possible to differentiate between oil fluorescence and mineral fluorescence, a small portion of the core is chipped off and placed in a clean, dry dish. A solvent, such as chlorothene, is poured on the chip. This dissolves the oil that is present, and fluorescence may be observed in the solvent moving away from the core sample. This is referred to as an oil cut.

 

Typical Core Description

The lithologic, textural, and fluorescent properties of a core are combined to form the description, which might read as follows: ""Sandstone, buff, fine grained, shaly and laminated, friable, blue/white uniform fluorescence."

Grain Density

The grain density of a rock is defined as the weight of the rock (exclusive of the weight of fluids contained in the pore space) divided by the volume of the solid rock material (exclusive of pore space). The density varies with the mineral composition of the rock and the state of hydration of the minerals. In complex lithologies containing inter-mixed limestone, dolomite, sandstones, and heavy minerals grain density will
vary vertically and horizontally. Even in formations described as homogeneous, measured densities often vary considerably from published values for pure components (Table 1., below). Minor amounts of secondary cement, such as calcite or siderite, will cause grain densities to exceed values shown in the table. An example of the variance in interwell and intrawell differences that was recorded in one study is shown in Figure 1 (Grain density variation).

Component	Approximate grain density (g/cm3)
Sandstone	2.65
Limestone	2.71
Dolomite	2.85-2.87
Anhydrite	2.98
Gypsum	2.3
Pyrite	5.0
Siderite	3.9
Clays	2.2-2.9

Table 1.
Grain densities of pure rock or mineral components

Grain density is important in core analysis for two reasons. First, it can be used as a quality control check of the core analysis measurements themselves. Second, a value for grain density is required in the equation that is used to calculate the porosity from the bulk density value that has been sensed by the down hole logging tool. The equation takes the form:

Where:

GD = Grain Density, g/cm³ (assumed or measured)

BD = Bulk Density, g/cm³ (from down hole log)

FD = Fluid Density, g/cm³ (assumed)

= Porosity, fractional

Grain density values that exceed estimated values for pure components occur when carbonate cements are present, or when impurities such as pyrite or siderite are in the pore space. Values can exceed 3.0 g/cm³. In these cases, assumptions of erroneously low grain densities will yield calculated values of porosity that are too low. This potential error is of extreme importance in evaluating poor quality, low porosity rock. When the erroneously low porosity value is substituted into water saturation equations that utilize resistivity data, the calculated water saturations are too high. A combination of underestimated porosity and overestimated water saturations can cause calculated oil-in-place to be underestimated by as much as 25%

An exception to the general rule that impurities in the rock increase its grain density is the case of core samples that contain large quantities of hydratable clays. When such a sample is dried in a humidity oven, the two molecular layers of water are maintained on the clay surfaces and the calculated grain density has a lower value than the rock itself. Removal of this water increases the grain density.

When grain densities are reported from core analysis, it is important to know how the samples were prepared. In some cases, the equations used in the evaluation require that the grain density be measured on samples with all bound water removed from clay surfaces; in other circumstances, it is desirable to measure grain densities and porosities with this water still in place.

After the core has been cleaned and dried to specifications, the measured grain density is accurate to ± 0.01 g/cm³.

Core Photography (Natural and Ultraviolet Light)

Core photography has advanced from black and white photographs to color photography and to pictures taken under ultraviolet light. Photographs are often taken of the core prior to the time it is slabbed, with its residual hydrocarbons still in place. The color and ultraviolet light photographs supplement one another, the former recording what the eye sees and the latter revealing oil fluorescence and, hence, the location of hydrocarbons. In reservoirs that are fractured, enhancement of the fracture pattern is sometimes observed by the ultraviolet photographs.

Photographs are often taken at 1 ft (30 cm) intervals along the core. In other cases, the cores are grouped into lengths of 15 or 20 ft (4.5 or 6.1 m) and a single photograph is made of the sequence. Some detail of the core may be lost in the multiple footage shots, but changes in lithology as well as core structures are sometimes observed better this way. While laminations and other core structures are visible on nonslabbed core, they are sometimes easier to study on a core that has been cleaned, slabbed, and then photographed.

Multiple footage pictures, such as those shown in Figure 1 (Core photograph illustrating structure and depositional environment), are helpful for seeing changes in core bedding and structures that indicate differences in permeability, porosity, and depositional environment. These photographs form a permanent record of the core's appearance and can be easily shipped to interested persons. A photographic record is invaluable in reservoir studies when the rock is no longer available. Written lithologic descriptions are important, but they are complemented by the photographic record. Geologic interpretations may change as additional wells are drilled and new information becomes available. In such cases a permanent record of the core's appearance is a very helpful tool. The photograph graphically reveals zone transitions, stained sections, minerals, fractures, dips, stylolites, shale partings, and other critical formation guideposts.

Core-Gamma Log

The natural gamma radiation of the core may be sensed in the laboratory with instrumentation similar to that used in the downhole log. This radiation is generally attributed to the presence of uranium minerals, or the potassium isotope K-40, or both. Shales typically have high gamma activity, carbonate rock activity is low, and sandstone activity falls between the two. The gamma radiation tool is used for discerning lithology and for correlation purposes; the data it generates are referred to as the Core-Gamma log.

When a core sample reaches the laboratory, it is placed on a motor-driven conveyer belt. The sample travels through a lead tunnel containing a scintillation gamma-ray detector. Impulses from the detector are electronically integrated and the output is recorded on a log to the same scale used for downhole gamma ray logs.

Measurements in the laboratory are restricted to continuous cores; they are not available for sidewall core samples. Data are recorded in chart form for comparison with visual assessment of core lithology, and for depth correlation with the downhole log. Historically, laboratory responses have not been calibrated in API units, but standards are now available so that this can be accomplished.

By aligning the peaks and valleys of the laboratory-measured gamma radiation profile with those of the down hole gamma ray log, the depth recorded on the core analysis can be calibrated to the depth on the downhole log. Adjustments are normally made to the core analysis depth profile to correct it to the downhole log values. Core depths are computed by counting the pipe joints recovered from the hole. In some cases, a joint (30 ft, or 9 m) or stand (90 ft, or 27 m) may be missed in the tally. It is common to find depth discrepancies of 3 to 7 ft (1 to 2 m) between log depths and core depths. When no correlation is found between the two, it is advisable to shift the depth in intervals of 30 ft (9 m) to see if a correlation can be found. The largest depth discrepancy between the two devices noted to date by the author has been 180 ft (55 m).

The natural gamma radiation tool possesses a feature that allows output to be sent to a computer, where it is digitized and stored for timely retrieval. The data can then be replotted to various scales for easier correlation and interpretation, or presented in a tabular printed report. Usually the surface measurements of gamma radiation can be completed within an hour and are ready for immediate comparison with physical characteristics of the core and for correlation with downhole logs from the same or adjacent wells. These data are usually reproduced on a core graph along with other basic laboratory information.

Cores that were not originally tested for gamma radiation and that have been stored for long periods of time are still suitable for this measurement. Samples that were previously analyzed, and cleaned and dried in the process, are also suitable candidates. Even samples that have been preserved with CoreSealTM can be tested without removal of the preservative material. It is recommended that cores not be tested within closed wooden or cardboard boxes because this inhibits proper fitting of the core prior to sensing its gamma activity.

When rubber sleeves are within the unopened boxes, for example, zones where the sleeve has collapsed because of sand washout will not be identified.

A recent advance in surface measurements of gamma radiation emitted from the cores has been the introduction of the spectral gamma analysis. This test presents not only the total gamma activity recorded previously, but also furnishes the portion of the total recording contributed by uranium, potassium, and thorium.

In some areas of the world, the presence of mica and other clays within the pore structure or rock fabric has caused hydrocarbon productive zones to be highly radioactive and to appear to be shales. The use of this spectral log is expected to assist in identifying such zones, as well as others with anomalous radioactivity. Figure 1
(Core-Gamma surface log and Core Spectral log) illustrates a Core-Gamma surface log (traditional laboratory gamma radiation test) and the associated Core Spectral log (special analysis of elements). Figure 2 (Correlation of downhole gamma log with Core-Gamma surface log) shows the correlation between a downhole and Core-Gamma surface log.

Directional Permeability

Permeability is often a directional quantity. In both sandstone and carbonate reservoirs, it is common to find low permeability zones that may be only millimeters thick, yet serve as effective barriers to vertical flow. In such a case, the vertical permeability of a formation is some fraction of the horizontal permeability. A lack of information with respect to this phenomenon can result in improper completion practices, or installation of an improved recovery scheme that does not work, or both. An example of an inappropriate recovery scheme occurred when a crestal injection of liquified petroleum gas was used to improve displacement efficiency from pinnacle reef reservoirs. In some instances, vertical flow barriers extended over large horizontal distances, causing a holdup of the downward advancing liquified hydrocarbon fluids and an unanticipated high residual oil saturation in noncontacted zones.

In thinner formations, where differences in vertical and horizontal permeability may not be significant, there may still exist a preferred permeability direction in the horizontal plane. Many rocks are made up of grains that were transported by and deposited in water currents. If the grains were not spherical, they acted like weather vanes. This means that they came to rest with their long axis parallel to the current, and their greatest cross-sectional areas parallel to the earth's surface. This preferred grain orientation gives rise to directional permeability.

Tests to define directional permeability can be made on plugs cut around the core periphery or by conducting studies on full diameter cores. In either case, these data are most helpful, especially when they are obtained for oriented cores and reported at 30º to 45° increments around the circumference of the rock.

Directional permeability tests are a natural adjunct to oriented core analysis. They can be complemented by studying thin sections cut in specific orientations to explore and explain directional permeability differences.

Cation Exchange Capacity

The cation exchange capacity (CEC) is a measure of the ability of clays, such as montmorillonite, chlorite, illite, or kaolinite, to exchange ions between the clay surfaces and surrounding water. It is normally expressed as milliequivalents per 100 grams of dry rock. Its magnitude is a function of the clay type and amount, and may be determined using wet chemistry techniques that utilize an ammonium or a barium exchange. This technique originated within the soil science discipline. A newer technique furnishes the CEC through a correlation with the adsorbed water content of test samples. Bush and Jenkins (1977) have presented data and test information on the procedure.

Cation exchange capacity is a mechanism that has been related to stripping of ions from injection fluids, trace element accumulation, and over-pressuring of the formation. These data have also been used to identify changes in environment resulting from changes in climate, sediment source, and erosion rate. A primary use of these data in the petroleum industry is in the Waxman and Smits, and Waxman and Thomas equations (1968 and 1974) where they are used for calculation of water saturation in clay-bearing formations. The CEC is converted to an exchange capacity per unit pore volume prior to use.

Clay	CEC (MEQ/100 grams)*
Montmorillonite	80-150
Illite	10-40
Chlorite	10-40**
Kaolinite	3-15

*Units are milliequivalents/100 grams of dry rock.
**In some studies reported the CEC has been as low as 0.0 to 1.0

Table 1. Cation exchange capacities (CEC) of various clays

Application of this equation and its influence on calculated water saturations was demonstrated in a paper by Keel an and McGinley (1979). Ignoring the cation exchange capacity will result in calculated water saturations that are erroneously high. This may cause hydrocarbon productive zones to be overlooked, or at best have their oil-in-place underestimated. Table 1. presents cation exchange capacities for various clays.

Grain Size Distribution

Grain size distribution relates to both porosity and permeability and is an important indicator of reservoir quality. Even though, generally, the more uniform the grain size the higher the porosity and permeability, porosity is actually independent of grain size. Studies have shown that both silt-sized and coarse-grained sands will exhibit essentially the same porosity — but the coarse-grained rock will have much higher permeability.

The grain size distribution of poorly consolidated to unconsolidated sands is often arrived at by grain size analysis. Hayes (1977) emphasized the practical aspects of these measurements. Grain-size data are used for sizing the gravel packs that are employed to prevent sand flow into the wellbore; the data is also one indicator of depositional environment. The new particle size analyzer used in sidewall core analysis provides a rapid measure of grain size distribution and, with it, a better estimate of permeabilities on these small samples.

Two basic analytical techniques are commonly employed to estimate grain size distribution. One involves sieving a disaggregated core sample through a series of screens with increasingly finer mesh. Difficulties occur in this technique because the electrical attraction between disaggregated clay particles causes them to clump and not pass properly through dry screens. An improved procedure allows the sieve analysis to be done using a wet technique. With this procedure the finer clay particles can be washed through a 230 mesh sieve (0.0625 mm) to remove the clay. The remaining portion of the sample is then passed through a nest of various sized screens, and the portion remaining on each screen is weighed. It is common to present the grain size distribution as a function of cumulative weight percent of the total sample, as shown in Figure 1 (Grain size distribution using sieve analysis). Examination of this figure shows the S-shaped distribution curve normally encountered. The further to the left on the horizontal axis that the distribution curve falls, the coarser the material. The more uniform the sample the more vertical the S, and a completely uniform sample would result in a vertical line.

The second analytical technique for measuring grain size distribution utilizes the principles of Stokes' law on the settling time of free-falling grains in a liquid column. The sample is carefully disaggregated and weighed, and then placed in a holder where ultrasonic agitation completely disperses the particles and detaches clay from any sand-sized grains. The dispersed sample is then spread evenly and simultaneously over the surface of the solution in the settling tube. Particles falling to the bottom of the tube collect on a balance as a function of time. Weights are monitored electronically and transmitted to a computer. The results, when analyzed, furnish both tabular and graphical data similar to that illustrated in Figure 1.

A sample may be characterized mathematically by (1) its standard deviation, which indicates the degree of sorting, and (2) the median grain diameter. Other information is also presented. A grain size distribution not symmetrical about the mean grain size is said to be skewed. The coefficients of skewness are calculated, as well as the relationship of the total height of the distribution to its base width. This is referred to as the peakedness (kurtosis) of the distribution. The Trask (1932) sorting coefficient is an indicator of the degree of sorting, with the value of 1.0 representing a well-sorted sample. The coefficient is defined as the square root of the quotient resulting when the grain diameter observed at 25% of the cumulative sample weight is divided by the grain diameter found at 75% cumulative sample weight.

Firmly cemented samples pose problems in measuring grain size distribution. There are two common types of cement: carbonate and silica. Carbonate cement can be removed by washing the core with dilute hydrochloric acid. If the sample is cemented by silica, it may be impossible to disaggregate it without crushing the grains, and the only available technique for grain size distribution determination would be the use of thin sections. Sneider (1981) has reported that even in cemented sandstones the grain size distribution is a helpful indicator of rock quality and has been used successfully in studies of both cuttings and cores. While grain size distribution is an indicator of depositional environment, it normally must be supplemented with other information.

 

Fracture Studies

The purpose of fracture studies is to describe the nature and degree of natural fractures in recovered cores. These data are subsequently used:

to aid in locating exploration and development wells from whose fracture permeability we may benefit;

to assist in the planning of well completion and stimulation by identifying fracture intervals;

to evaluate wireline fracture-finding logs and correlate them with actual rock fracture conditions;

to assist in the structural analysis of hydrocarbon accumulations as they relate to fault orientation and location;

to assist in the calculation of theoretical block height and cross-sectional area of matrix fracture blocks for mathematical modeling.

In order to obtain good fracture data, good core handling procedures must be employed at the wellsite. Care should be taken not to break the rock excessively and to fit it properly together prior to marking. The core must be spatially oriented to be of maximum benefit to the study. This can be accomplished by coring with a special orientation tool, as discussed by Rowley, Burk, and Manual (1967), or by aligning visible core bedding dip with formation dip as identified either by down hole dipmeter or by regional geologic information.

After the core is properly orientated, both the dip and strike of observed fractures are recorded. An example of such a recording is given in Figure 1 (Distribution of measured fracture strike in a core), which shows a well with a major fracture trend that runs in a northwest to southeast direction, with a secondary trend at essentially right angles to it. Figure 2 (Histogram of fracture dip) is a histogram that indicates that the preponderance of fractures found in this well occur at dip angles of 50º or greater.

It is common to record all observed fracture data on a core fracture log, as discussed by Snyder and Craft (1977). This log displays fracture depth, length, frequency, density per foot, and dip, together with lithology and rock hardness, on a foot-by-foot basis. In addition, the log records core porosity and matrix oil saturation. The latter data are of particular importance in thick carbonates that display wide variations in rock properties and whose zones of oil saturation are sporadic, appearing and disappearing at different depths.

 

Calcimetry Measurements

Core samples are sometimes treated with acid to determine their total acid solubility, while in other cases it is desirable to know the percentage of limestone, dolomite, and insolubles present within the core. These data aid in typing rocks, and are often helpful in correlating various information about rock properties. Calcimetry measurements are sometimes made using a wet chemistry technique that is referred to as a versenate analysis. Swanson (1981) furnishes an operational method for making this determination.

There is a rapid, practical, and economical method for determining the ratios of limestone and dolomite and the total amount of acid-soluble material in small samples, using a calcimetry instrument. A one-half gram sample is crushed and placed in a chamber to which acid is added. The reaction of the acid with the rock releases carbon dioxide, which builds up pressure in the chamber. This rise in pressure is related to the rate of reaction and, when monitored, can be used to differentiate between limestones and dolomites, as well as to measure the total solubility of the rock.

 

Pyrochromatography (Thermal Extraction Chromatography)

The term pyrochromatography is derived from the process in which heat is applied in a controlled manner to small hydrocarbon-saturated rock samples, which causes the fluids to leave the sample and be swept into a chromatograph for "fingerprinting." Maness and Price (1977) have shown that these data are helpful in providing a distinction between gas productive, oil productive, and nonproductive zones. The most important application of this procedure, currently, is in evaluating reef formations. Its application requires the use of production as well as other conventional well evaluation data to establish working correlations.

Figure 1 (Chromatographic fingerprints for productive and nonproductive formations) presents chromatographic fingerprints for a nonproductive and an oil productive formation. The oil productive formation is characterized by high percentages of C-15 through C-19 and recognizable percentages of C-20 and above. Note that essentially all of the hydrocarbons less than C-8 are missing. This indicates the rock has permeability and that these components vaporized and escaped on the trip to the surface. Another important indicator is the total hydrocarbon quantity in the sample (the area under the curve), which is sufficient in this case to indicate oil production. The nonproductive figure shows less total hydrocarbon in the core and a high percentage of C-8 and lower. This indicates that the rock contained little residual hydrocarbons and was of such poor quality that the lighter hydrocarbons could not escape as the core moved to the surface.

This technique has been extended to the analysis of cuttings, and it embodies certain of the characteristics of the analysis used to evaluate source rock quality. Research is in process for extending the technique into the evaluation of other rock types.

Maness (1983) has reported on advances in the technology of fluorescence and their recent application to pyrochromatographic measurements. In the procedure he describes, portions of the rock are placed in a chemical solvent to leach out residual hydrocarbons. Fluorescence measurements of the leached solution are taken at predetermined time intervals, using a laboratory instrument that yields relative levels of fluorescence. This provides an indication of how rapidly available hydrocarbons are extracted from the rock, which is an indication of the permeability of the samples. Fluorescence data can then be plotted by depth to identify wellbore intervals that do and do not exhibit fluorescence. Figure 2 (Comparison of hydrocarbon profile by relative fluorescence and chromatographic analysis) is an example of a comparison between fluorescence and chromatographic analysis and includes a prediction of the gas, oil, and water zones.

Salinity of Pore Water

It is sometimes desirable to know the salinity of the water present within a reservoir rock. When prior knowledge exists about the salinity of both the formation brine and the coring fluid, measurement of salinity of water recovered in the core will give an indication of the extent of filtrate flushing. This technique can be applied to cores cut with water-base mud. In samples that have been taken with oil-base mud, where no extraneous water has been added to the core, the salinity of water in the sample should represent that in the reservoir.

In formations where the water level has not been previously penetrated by drilling and no water production has occurred, the salinity of the formation brine may be unknown. This salinity can sometimes be calculated from the spontaneous potential log if a sufficient salinity difference exists between the coring fluid and the in-place formation brine. Even in this case further verification of the calculated salinity is often desired.

A laboratory technique for estimating water salinity that is relatively rapid and often employed uses a sample of core taken adjacent to those used for the Dean-Stark or summation-of-fluids sample. If it is assumed that the porosity and water saturation present in the adjacent sample are the same as that in the summation or Dean-Stark sample previously analyzed, the quantity of water in the salinity sample can be estimated. The sample can then be crushed and washed with a known volume of distilled water. This leaches the salt from the sample. This solution can then be titrated to furnish a measure of water salinity. This technique has been documented in API RP 40.

Salinity measurements on cores taken with oil-base mud have also been used to estimate the vertical sweep efficiency. In those reservoirs where the salinity of the injection water varies sufficiently from that of the formation water, determination of the core salinity on a foot-by-foot basis will yield that portion of the formation that has been contacted by the injected water and, thus, has been swept.

There are practical limitations to making these calculations on low porosity formations that contain moderate water saturations. In such cases the quantity of water present in the test plug may be quite small, and a small error made in estimating its volume will yield a large error in the calculated salinity.

Oil API Gravity

Knowledge of the oil API gravity is helpful in the interpretation of core analysis data, and estimates of its value may be useful prior to the time the well is completed. Estimates of the in-place oil gravity can be made using several techniques. A common method requires the collection of oil samples from the area of interest. The API gravity is determined on these samples, which are then retorted, using procedures similar to that to which the core will be subjected. The API gravity is then determined on the oil recovered in the retorting process, and correlations are made between the retorted oil gravity and the initial, nonretorted oil gravity. Retort oil from the core analysis can be used with the correlations in subsequent wells. Data included on the core analysis report represent a non retorted oil gravity.

In cases involving analysis of heavy oil and tar sands, where no produced oil is available or likely to be available, oil is actually recovered from the core. Sufficient oil can sometimes be recovered by centrifuging the core; in other cases, solvents such as methylene chloride have successfully removed the oil from the core. The solvent is subsequently vaporized and measurements are made on the remaining oil. Application of heat and agitation is necessary to remove the solvent from the oil, and some uncertainty exists as to how representative the gravity of the remaining fluid actually is. Nevertheless, this measurement has proved helpful in those cases where no other source of oil gravity was available.

Three basic techniques exist for the determination of oil gravity. The first technique involves filling and weighing a capillary of small volume with water, and then filling and weighing the capillary with the sample of oil. Comparison of the two weights yields the specific oil gravity from which API gravity is calculated.

The second technique involves the suspension of an oil droplet in a water-alcohol mixture. The mixture is adjusted by water or alcohol addition to cause the oil droplet to float. When this adjustment has been completed, a hydrometer is inserted into the mixture and the specific gravity of both the oil and the mixture is read.

A third technique and one commonly used in sidewall core analysis is to measure the refractive index of the retorted oil and, through the use of correlations, to estimate the API gravity of the non retorted oil. This is a relatively rapid measurement technique. It requires previously developed correlations between non retorted oil API gravity and retorted oil refractive index. The following equations express the relationship between specific gravity and API gravity:

SG = 141.5/(131.5 + API°)