the occurrence and distribution of sedimentary zeolites in the … · 2017-09-28 · the occurrence...

The Occurrence and Distribution of Sedimentary Zeolites in the Main Pass Area, Gulf of Mexico Basin

Matthew W. Totten, Sr.1, Sheri Simpson2, Elizabeth Powers3, and Iris M. Totten1

1Department of Geology, Kansas State University, Manhattan, Kansas 66506

2Chevron Energy Technology Company, 1600 Smith St., Houston, Texas 77002

3Shell Exploration & Production Company, Two Shell Plaza, Houston, Texas 77022

ABSTRACT The Main Pass area, offshore Louisiana, is a large, salt-dome dominated, Miocene

sandstone reservoir where sporadic production problems have been encountered. The production problems appear to correlate with the presence of authigenic zeolites within the sandstones.

This study was undertaken to understand the distribution of zeolites and to create a predictive model for their distribution in the Main Pass area. Zeolites were concen-trated by density separation from samples obtained from sidewall cores, and examined under a Scanning Electron Microscopy with Energy Dispersive X-ray Spectroscopy. Positive identification of zeolite mineralogy was confirmed by X-ray Diffraction.

The distribution of specific zeolite minerals is dependent upon distance from the salt dome. The sodium-rich zeolite analcime is found from 0-500 feet from the dome, and calcium-rich clinoptilolite is found from 0-2000 feet from the dome. Beyond 2000 feet from the dome unaltered glass shards, and minor amounts of clinoptilolite are observed. The zeolites formed from precursor volcanic glass shards within the reservoir. We pos-tulate that the zeolitization is controlled by the salinity of the formation waters, which is controlled by distance to the salt dome. As the salinity increases, sodium-rich zeolites increase in abundance.

This predictive model of zeolite-mineralization is potentially applicable to other sandstone reservoirs with a volcanic ash component. Overall, recovery can be increased by recognizing specific zeolites and properly treating the wells with appropriate comple-tion fluids to minimize formation damage and the resulting production loss.

INTRODUCTION Zeolites have been encountered in several Gulf of Mexico sandstone reservoirs, and have been reported to

cause production problems offshore Louisiana (Underdown et al., 1990). They often inhibit the permeability of producing sands, particularly near the wellbore as fine-grained zeolites migrate with fluids during production. Authigenic mineralization of zeolites within the pore space of sands further reduces fluid flow.

Zeolites are hydrated aluminosilicates with large open spaces within their crystal structure. They have out-standing qualities of adsorption and ion-exchange, similar to clay minerals (Colella, 1996). There are over 20 zeolite minerals recognized in sedimentary rocks, but laumontite, clinoptilolite, and analcime are the most com-mon zeolites present in hydrocarbon reservoirs (Sheppard and Hay, 2001). The chemical formula of these com-

Totten, M. W., Sr., S. Simpson, E. Powers, and I. M. Totten, 2007, The occurrence and distribution of sedimentary zeolites in the Main Pass area, Gulf of Mexico Basin: Gulf Coast Association of Geological Societies Transactions, v. 57, p. 717-727.

717

Totten et al.

Table 1. Formulas of common zeolites in Gulf of Mexico reservoirs. Clinoptilolite (K0.8Na0.4Ca2.8)Al6.8Si29.2O72 • 26H2O Analcime Na10.2Al10.2Si25.8O72 • 12H2O Laumontite (K0.6Na0.6Ca5.4)Al12.0Si24O72 • 24H2O

mon zeolites is given in Table 1. The differences in zeolite chemistry are very subtle, and are dependent on small differences in the chemistry of formation fluids, temperature, and pressure. The ratio of Si/(Si+Al) is a good determinant of the species of zeolite present, as is the ratio of the major cations K, Na, and Ca (Reed et al., 1993).

Zeolites often form in the diagenetic environment often by alteration of pre-existing material, recrystalliza-tion of volcanic glass, feldspar, amorphous alumino-silicate gel, smectite, kaolinite, feldspathoids, and other pre-cursor zeolites (Mumpton, 1975). Volcanic glass fragments are the most important and predominant precursor (Sheppard and Hay, 2001). Volcanic glass shards are amorphous, and highly reactive. As ash shards within clas-tic sediments are buried, the increased chemical potential promotes their alteration into both zeolites and clay minerals (Boles, 1993).

Previous studies have emphasized the role of increasing temperature during burial diagenesis as the major control on zeolite mineralization (Iijima, 1988; Iijima, 2001; Dunn et al. 1993). A diagenetic sequence of vol-canic glass to clinoptilolite to analcime and laumontite is most often cited in these studies. However, none of these studies report variations in formation water chemistry.

Because of their reactivity, zeolites are very sensitive to the chemistry of completion fluids, drilling muds, oil mobilization fluids, and mixing of formation waters. Analcime has also been known to form from precursor zeolites during well stimulation techniques, reducing the permeability of the formation (Boles, 1993). Stimula-tion using hydrochloric acid on analcime-cemented sandstones formed a silicate gel that heavily damaged the wellbore, while acetic acid treatments dissolved the analcime (Underdown et al., 1990). Clearly, knowledge of the presence of zeolites within sandstones and a model for their distribution will enhance development of these reservoirs.

AREA OF STUDY The area of study involved a field in the Main Pass Protraction Area (Fig. 1) operated by Chevron Corpora-

tion. The Main Pass area surrounds a large salt dome that extends from 2,000 feet to over 25,000 feet below sea level. Production is from sandstones between 4,000 feet and 8,000 feet. Rates of production were below initial estimates, and were found to be due to the presence of zeolites within the reservoir sands. Lower production rates were presumed to involve migration of fine-grained zeolites within the reservoir during production, clogging screens and obstructing pore throats near the well bore. Additional problems were encountered during acid stimulation treatments in some wells.

An internal study was conducted by Chevron to investigate the distribution of zeolites within these reser-voirs. Whole-rock x-ray diffraction (XRD) indicated the presence of clinoptilolite and analcime in several sam-ples. The wells with the highest XRD peaks, presumably due to the highest concentration of zeolite, exhibited the worst production rates. A precursor material was not identified, and the distribution appeared to be random. The samples were not identified petrographically to confirm the presence of zeolites.

OBJECTIVES The objectives of this study are to develop a method to concentrate zeolites, enhancing their detection using

XRD methods. We also wanted to confirm zeolite mineralogy using scanning electron microscopy (SEM). We further wished to identify the precursor material involved in zeolitization, and ultimately, to develop a model to predict their distribution.

718


METHODS Forty-eight side-wall cores from the Main Pass area were obtained from the University of New Orleans

Chevron Earth Science laboratory for analyses. Samples were specifically selected using a previous internal XRD study by Chevron that contained variable amounts of zeolites present, as well as those that did not contain zeolite minerals. Proprietary maps of the field were used to measure the distance of each sample to the salt dome. Temperature data was obtained from well logs, and corrected for circulation time before logging. Chevron also provided proprietary water chemistry from the producing horizons.

Identification of Zeolites Identification of zeolites within clastic sediments is difficult, particularly when they occur in trace amounts.

The accepted method to identify zeolites is by XRD, if they occur in significant quantity. Zeolites are difficult to quantify by XRD, however, because of their variable chemistry, small crystal size (<5 microns), poor structure factors, and overlapping peaks with many other silicates. In addition, XRD cannot directly detect volcanic glass, the most common precursor material. Glass is amorphous, without a regular crystalline structure to diffract the x-ray beam in an identifiable pattern.

Microscopic identification of zeolites is also impractical due to their small size. The use of a scanning elec-tron microscope (SEM) allows magnification high enough to identify zeolites, but is also problematic if the zeo-lite concentration is too small. The efficacy of both XRD and SEM methods is improved if the zeolite fraction can be concentrated from the whole rock. This was accomplished using variable density liquids.

Light Mineral Separation Clinoptilolite has a density of 2.15 g/cc, while analcime has a density of 2.30 g/cc. Almost all other minerals

in a sandstone have a higher density. A light mineral separation was performed to concentrate these zeolites from the rest of the rock using lithium meta tungstate (LMT), a liquid of variable density. The method is similar to that described by Totten et al. (2002) to separate clay minerals from shales. Disaggregated samples were added

Figure 1. Location of Main Pass area in the Gulf of Mexico.

719

to the LMT of the appropriate density. Two separations were performed, the first at 2.4 g/cc to concentrate all of the zeolites, followed by 2.3 g/cc to separate the clinoptilolite from the analcime.

X-Ray Diffraction (XRD) The material recovered in the light-mineral separation was analyzed by XRD using a Siemens D5000 at the

Chevron Algiers Upstream Technology Center in Algiers, Louisiana. The acetate filters that contained the low-density concentrate from the LMT separation were prepared for XRD by using double-sided tape to attach the filter paper onto a clay-smear stage.

Although glass concentration cannot be ascertained by XRD techniques because it is non-crystalline, an amorphous silica halo can occur when a sample contains significant amorphous glass shards. The amorphous silica halo is centered around the quartz peaks (Osawa and Bertan, 2005). Each sample was examined for the presence of this halo to infer the potential for glass shards within the sample.

Scanning Electron Microscopy (SEM) The SEM work was also performed at the Chevron Algiers Upstream Technology Center. This SEM is

equipped with an energy dispersive spectrometer (EDS) that provides a spectrum of the chemical elements pre-sent under the electron beam. This helps determine the specific mineralogy of each sample, based upon gross chemistry.

The acetate filters that were produced from the LMT separation were prepared for the SEM stage by using double-sided tape to attach the filter paper onto the stage. Once the samples were attached to the stage, a mixture of palladium and gold was coated to the sample to allow conductivity of the sample.

The crystal habits of clinoptilolite and analcime are diagnostic. Clinoptilolite is platy, euhedral, and coffin-shaped, while analcime forms trapezohedra. Their EDS spectra is quite similar, therefore the distinguishing char-acteristic between these two zeolites is their crystal habit. The distinguishing feature of glass shards is their con-coidal fracture.

Additional details of the methods used in this study are reported in Powers (2006).

RESULTS

The concentration of the low-density zeolites from the bulk rock greatly improved the ability to detect these

minerals. Figure 2 is an x-ray diffractogram of a sample containing significant analcime, but little clinoptilolite. The background is fairly low, suggesting little amorphous material, hence no glass shards. Figure 3 is an exam-ple of a sample with significant clinoptilolite and analcime present. In addition, the amorphous “hump” suggests the presence of glass shards within the same sample. In this manner, the occurrence of each zeolite and glass shards were recorded throughout the field.

SEM results confirmed the XRD data. An example of a volcanic glass shard is illustrated in Figure 4. These shards are easily identified by their concoidal fracture. This example shows a pristine shard, other samples had shards that were altered. In some cases zeolite material could be seen on the altered shard. Figure 5 is an exam-ple of clinoptilolite. The platy texture and small crystal size (5-10 microns) are apparent. In contrast, analcime shows a relatively large (20 microns) euhedral, analcime trapezohedron (Fig. 6).

The distribution of glass shards, clinoptilolite, and analcime relative to the salt dome is shown in Figure 7. Glass shards were only observed in samples distant from the salt, while analcime was only observed close to the salt dome. Clinoptilolite was observed in most samples, over a much wider spread in distance from the dome.

DISCUSSION It is inferred from the results of this study that volcanic glass shards are the precursor to the zeolites in Main

Pass. This is in agreement with many previous studies suggesting that volcanic glass is the most important pre-

Totten et al.

720

Figure 2. X-ray diffractogram of sample 80. This sample has a strong analcime peak, but no clinopti-lolite. The small increase in background x-rays surrounding the quartz peak suggests that little vol-canic glass is present in this sample.

Figure 3. X-ray diffractogram of sample 117. This sample has a strong analcime peak, and a strong clinoptilolite peak. The large increase in background x-rays surrounding the quartz peak suggests the presence of significant amorphous, volcanic glass. All three phases are present in this sample, which is a rare occurrence in close proximity to the salt dome.


721

Figure 5. An example of clinoptilolite from sample 89 under SEM with typical euhedral, coffin-shaped crystal habit.

Figure 4. An example of a pristine volcanic glass shard as seen under SEM. Pristine shards were re-stricted to samples distant from the salt dome.

Totten et al.

722

Figure 6. An example of a euhedral, analcime trapezohedron from sample 117. The presence of anal-cime was restricted to samples within 500 feet of the salt dome.

Figure 7. Variation of zeolite mineralogy with distance to the salt dome. Unaltered glass shards are restricted to distances greater than 2,000 feet and analcime is restricted to samples within 500 feet of the salt dome.


723

cursor to zeolite formation in hydrocarbon-bearing sandstones (Sheppard and Hay, 2001; Boles, 1993). Volcanic glass shards sourced from western North American eruptive centers have been reported in the Gulf of Mexico (Totten et al., 1998; Totten et al., 2005). It is also important to note that the volcanic glass is not detectible using whole-rock XRD, but is detectible by concentrating the light density fraction using our method. The glass shards are confirmed on SEM of the concentrated fraction. While easier detection is an obvious benefit to concentrating the light fraction, information about the actual in situ distribution of the zeolites within the sandstone porosity is lost.

The control of distance to the salt dome on zeolite mineralogy is apparent in Figure 7. Based upon previous work relating zeolitization with increased temperature during burial diagenesis (Boles, 1993; Iijima, 2001), we first assumed our observed relationship in Figure 7 was due to a temperature gradient, as salt is a good thermal conductor. However, no correlation between zeolite mineralogy and formation temperature is apparent (Fig. 8).

The distance between a sample location and the salt dome exerts a strong influence on formation water chemistry. Figure 9 illustrates this control on samples from Chevron’s Main Pass area. A strong gradient in in-creased sodium concentration of the produced waters exists as the formation nears the salt dome. With increased distance from the salt, sodium levels decrease relative to calcium.

At a distance of 500 feet from the salt dome there is a major change in water chemistry and a corresponding change in zeolite mineralogy (Figs. 7 and 9). A second, less pronounced change is also seen in both water chem-istry and zeolite mineralogy 2,000 feet from the salt. The influence of distance from salt on both variables sug-gests that the major control on zeolitization in this Main Pass field is formation water chemistry. As calcium concentrations increase, precursor volcanic glass is altered into the calcium-rich zeolite clinoptilolite. This altera-tion begins approximately 2000 feet from the salt, and continues until approximately 500 feet from the dome. At this point, a significant increase in sodium occurs, promoting the formation of sodium-rich analcime at the ex-pense of previously formed clinoptilolite.

Figure 10 is a simplified model of our predictive model for this field. If any precursor volcanic glass is pre-sent within a sample, which in the case of this field it was very widespread, changing chemical concentrations of calcium and sodium promote alteration of this glass into zeolites. This alteration begins at 2000 feet from the salt. Because calcium levels are highest relative to sodium at this point, the calcium-rich clinoptilolite is the zeo-lite that forms. At 500 feet from the dome, analcime becomes stable due to increased sodium concentrations, and recrystalizes from earlier formed clinoptilolite.

Future drilling should expect to see specific zeolite minerals depending on the distance of the wellbore from the salt dome when the producing sandstone is penetrated. This could be confirmed using the concentration methods in this study, combine with XRD or SEM. Completion techniques could be tailored to account for the specific zeolite encountered maximizing production rates.

CONCLUSIONS The distribution of zeolites within a producing field in Main Pass, is dependent on consistently identifying

the zeolite mineralogy. This is problematic in bulk rock samples, but becomes reliable by concentrating the frac-tion of the rock that contains zeolites. This was accomplished in this study by a light mineral separation using LMT. An additional benefit of this method is that precursor volcanic glass, if present, is also concentrated. After concentration of this light-density fraction, glass and zeolites are identified using XRD and SEM techniques.

In contrast to previous studies of zeolitization during burial diagenesis, our results indicated no control of temperature on the formation of zeolites. Within this field in Main Pass, formation water chemistry exerts the major control on zeolitization. Calcium-rich waters promote the formation of clinoptilolite, and sodium-rich wa-ters promote analcime. Not surprisingly, the highest sodium concentrations are found closest to the salt dome, therefore analcime is only found close to the dome. Clinoptilolite is first identified 2,000 feet from the dome, and persists in some samples to very near the dome, but is most abundant between 2,000 feet and 500 feet away. Only precursor glass shards are found farther than 2,000 feet from the salt.

The predictive model proposed in this study should aid in determining best completion techniques to avoid the production problems previously encountered due to zeolite minerals. It might also be useful in other zeolite-bearing fields in the Gulf of Mexico.

Totten et al.

724

Figure 8. Variation of zeolite mineralogy with formation temperature. No correlation between tem-perature and zeolitization is observed.

Figure 9. Variation in formation water chemistry with distance from the salt dome. Note the similar distance to salt on changes in water chemistry with changes in zeolite minerals seen in Figure 7. Higher calcium concentrations promote formation of clinoptilolite, higher sodium concentrations favor anal-cime.


725

Figu

re 1

0. S

impl

ified

cro

ss se

ctio

nal m

odel

of z

eolit

e fo

rmat

ion

with

in th

e M

ain

Pass

Fie

ld.

Totten et al.

726

ACKNOWLEDGMENTS The authors would like to thank Chevron Corporation for providing the many resources necessary to com-

plete this study. The generous donation of the Chevron Sample Warehouse to the University of New Orleans allowed access to the samples used in this study. An internship to E. Powers, and SEM and XRD analyses at the Algiers Upstream Technology Center are sincerely appreciated. Access to proprietary field maps and water chemistry were vital to development of our model.

REFERENCES CITED

Boles, J. R., 1993, Zeolite cements in hydrocarbon reservoirs, in D. W. Ming and and F.A. Mumpton, eds., Zeolite ’93: Program and Abstracts of the 4th International Conference on the Occurrence, Properties, and Utilization of Natu-ral Zeolites, Boise, Idaho, p. 51-53.

Colella, C., 1996, Ion exchange equilibria in zeolite minerals: Mineral Deposita, v. 31, p. 554-562. Dunn, T. L., M. D. C. Destefano, and O. O. Decastelli, 1993, Zeolites in petroleum evaluation of volcaniclastic sand-

stones, San Jorge Basin, Argentina, in D. W. Ming and and F.A. Mumpton, eds., Zeolite ’93: Program and Ab-stracts of the 4th International Conference on the Occurrence, Properties, and Utilization of Natural Zeolites, Boise, Idaho, p. 83-85.

Iijima, A., 1988, Application of zeolites to petroleum exploration, in D. W. Ming and and F.A. Mumpton, eds., Zeolite

’93: Program and Abstracts of the 4th International Conference on the Occurrence, Properties, and Utilization of Natural Zeolites, Boise, Idaho, p. 29-37.

Iijima, A., 2001, Zeolites in petroleum and natural gas reservoirs: Reviews in Mineralogy and Geochemistry, v. 45,

p. 347-402. Mumpton, F. A., 1975, Morphology of zeolites in sedimentary rocks by scanning electron microscopy: Clays and Clay

Minerals, v. 24, p. 1-23. Osawa, C., and C. Bertan, 2005, Mullite formation from mixtures of alumina and silica sols: Mechanism and pH effect:

Journal of Brazilian Chemistry Society, v. 16, p. 251-258. Powers, E. A., 2006, Mineralization and prediction of sedimentary zeolites in the Main Pass area, Gulf of Mexico Ba-

sin: M.S. thesis, Kansas State University, Manhattan, 97 p. Reed, J. K, M. Gipson, and D. Vass, 1993, Hydrocarbon potential of sandstone reservoirs in the east Slovakian Basin,

part 2: Zeolites and clay minerals: Journal of Petroleum Geology, v. 16, p. 223-236. Sheppard, R., and R. Hay, 2001, Occurrences of zeolites in sedimentary rocks: Reviews in Mineralogy and Geochemis-

try, v. 45, 217-234. Totten, M. W., M. A. Hanan, B. B. Hanan, and T. Kratochvil, 1998, Improved regional ties to global geochronology

using Pb-isotope signatures of volcanic glass shards from deep water Gulf of Mexico ash beds: Gulf Coast Asso-ciation of Geological Societies Transactions, v. 48, p. 95-106.

Totten, M. W., M. A. Hanan, D. Mack, and J. Borges, 2002, Characteristics of mixed-layer smectite/illite density sepa-

rates during burial diagenesis: American Mineralogist, v. 87, p. 1571-1579. Totten, M.W., M. Jurik, and M. A. Hanan, 2005, The occurrence and seismic expression of volcanic ash beds in the

Gulf of Mexico: Gulf Coast Association of Geological Societies Transactions, v. 55, p. 810-820. Underdown, D. R., J. Hickey, and S. K. Kalra, 1990, Acidization of analcime-cemented sandstone, Gulf of Mexico:

Society of Petroleum Engineers, p. 97-102.


727

NOTES

728

the occurrence and distribution of sedimentary zeolites in the … · 2017-09-28 · the occurrence...

Documents