Determination of the pore space parameters in microporous rocksby means of thermal methods

Małgorzata Labus a,n, Krzysztof Labus a, Petr Bujok b

a Institute for Applied Geology, Silesian University of Technology, Akademicka 2, 44-100 Gliwice, Polandb Institute of Geological Engineering, VŠB-Technical University of Ostrava, 17. listopadu 15, 708 33 Ostrava, Czech Republic

a r t i c l e i n f o

Article history:Received 17 September 2014Accepted 4 February 2015Available online 14 February 2015

Keywords:microporosityporosityshalesthermal analysisretort extraction

a b s t r a c t

The aim of the study was to examine the possibility of determining the pore space in microporous rocksusing a set of methods: differential thermal analysis, retort distillation, and Dean-Stark. These methodsare used for low-permeability rock containing hydrocarbons, such as shale gas, shale oil and tight gas. Byusing these methods it is possible to specify the amount of water contained in the sample and,consequently, determine the different levels of saturation and porosity.

The study showed shale porosity values ranging from 0% to more than 5%. It was found that thesamples with the highest porosities in the combustion process evolve relatively large amounts of carbondioxide, which indicate that the porosity is related to the presence of organic matter. There was nocomplete correlation between the various methods' test results. It seems that for determining theporosity of tight shale rocks, the most effective would be the use of several complementary methods.

& 2015 Elsevier B.V. All rights reserved.

1. Introduction

In recent years, the high importance in fuel industry is attributed tothe so-called “unconventional gas”. From an economic point of view,unconventional shale reservoirs are more difficult and less profitableto operate than traditional (conventional) deposits. They are charac-terized by extremely low petrophysical parameters, such as porosityand permeability, representing a tremendous challenge for the oilindustry. The specific nature of these reservoir rocks is determinedby the analytical difficulties and technical challenges of producingnatural gas.

Effective porosity, saturation of rocks with water, oil and gas, aswell as the permeability of the basic reservoir parameters are nece-ssary to evaluate the quantity of hydrocarbons and reservoir efficiency.In the case of gas accumulated in the clay-slate rocks (so-called “gasshale”) the evaluation of the above parameters is a huge technicalchallenge. In the classical approach, shale was treated as a sealingrock, due to its very low effective porosity and permeability (fromhundreds to thousands of times smaller than in conventional reservoirrocks). The consequence of low porosity is correspondingly lowcontent of reservoir fluids. Currently, the approach to shale haschanged in a fundamental way; the shale rock is treated as a source,sealing and reservoir rock. The porosity of shales also has a differentcharacter (Milner et al., 2010; Such et al., 2011). It is related to

(1) pores in the organic substance, (2) intragranular pores, and(3) intergranular pores.

n case of the first mentioned pores, in the organic matter, theyrefer to the isolated intrakerogen pores. As they are very small andnot effectively connected to natural and hydraulic fractures, theyresult in huge challenges in gas production. Hence, the primaryquestion is the separation of organic and inorganic porosity inshales. Organic porosity is critical in terms of accumulation andproduction of hydrocarbons, while in the pores of the inorganicgenesis predominantly accumulates water (Milner et al., 2010).This separation and potential hydraulic discontinuity betweenbrines and hydrocarbons calls into question standard concepts ofpore saturation and the relative mobility of pore fluids in the rock(Handwerger et al., 2011). Another challenge in the case of shalerock is to distinguish “free” pore water from “bound” or “matrix”water. To understand the nature of unconventional reservoir rockproperties there is a need not only to reconsider the meaning ofpreviously defined concepts of the pore space, but also to developnew methods for measuring the pore space. Petrophysical analysisof fine-grained shale rocks should focus on the evaluation of actualreservoir capacity of rocks and assess the fluids transport.

In order to evaluate the shale rock hydrocarbons capacity it isessential to determine (Such and Leśniak, 2011): (1) total porosity(including isolated and interconnected pores), (2) effective porosity(sum of interconnected pore space), (3) % pore volume occupied bywater, (4) % pore volume occupied by liquid hydrocarbons, (5) % porevolume occupied by dispersed organic matter.

Contents lists available at ScienceDirect

journal homepage: www.elsevier.com/locate/petrol

Journal of Petroleum Science and Engineering

http://dx.doi.org/10.1016/j.petrol.2015.02.0140920-4105/& 2015 Elsevier B.V. All rights reserved.

n Corresponding author.E-mail addresses: malgorzata.labus@polsl.pl, mlabus@o2.pl (M. Labus).

Journal of Petroleum Science and Engineering 127 (2015) 482–489

To test unconventional shale reservoirs, there are used traditionalmethods of analysis for conventional reservoir rock, modified how-ever, for these particular needs (e.g., the Dean-Stark fluid extractionmethod or mercury porosimetry), as well as other methods, e.g., gasadsorption, thermal analysis or FTIR spectroscopy, used to assess thecomposition of fluids present in the sample. The most commonly usedtechniques for assessing petrophysical properties of shale are theabove-mentioned methods of Dean-Stark extraction and retort extrac-tion. These methods do not provide the most consistent results of thetotal porosity, and therefore there are often carried out experimentsaimed to compare and link the results obtained from differentmethods (Handwerger, 2012).

The measurements are usually performed on core or crushedsamples, and the legitimacy of the use of one selected technique is thesubject of a number of comparative studies and remains controversial(Sondergeld et al., 2010; Clarkson et al., 2011). Nanopores, andvirtually no diffusive flow means that in the case of tight shales themeasurement of pore space is incomplete, and the results obtainedare unique. It is possible, however, to obtain the parameters of porespace in the shale rocks by mercury porosimetry, according to themethodology developed at the Institute of Oil and Gas (Darłak et al.,2011). Such and Leśniak (2011) suggested to perform all analyses,except for bulk density, on the crushed material (even porosimetricmeasurements (Such, 2012)), which would give a better reproduci-bility of the results.

2. Scope and methods of the analysis

2.1. Samples of shale rocks

Seven, fine grained clastic sedimentary rock samples, were selectedfor the analysis. They were collected from outcrops or drill cores inPoland, Czech Republic and Estonia. The characteristic of thesesamples is presented in Table 1. The mineral composition presentedin the table was determined basing on microscopic observations. Inthis table the total organic carbon (TOC) content is also included. TheTOC value (wt%) is a result of Rock Eval analysis (pyrolysis technique).Gas generation potential of gas shales is mostly generated by the totalorganic carbon content, type and maturity of the organic matter.TOCrepresents the concentration of organic matter in rocks by the weightpercent of organic carbon. Samples 3 and 4 have low organic carboncontent (o1%), samples 1, 2 and 7 represent rocks of TOC consideredthe minimum for shale gas reservoirs (1–2%), while Estonian rocks'(sample 5 and 6) TOC is very high (412%). In case of Kukersite oilshale (sample 6) most of the organic matter is derived from fossilgreen alga (Elenrum et al., 2008).

To complete the information on the mineral composition of theexamined rocks and to determine the presence of different forms ofwater, the thermal analysis (TA) was performed. The analysis in aDean-Stark apparatus was performed to determine the water contentin the samples, including the capillary bound water in clays. The retortmethod allowed the extraction of liquid and gaseous hydrocarbonsand pore water, hygroscopic and crystallization water.

2.2. Analysis of mineral composition

Thermal analysis (TA) enables determination of mineral com-position of the rock samples containing clay minerals, as well aswater content in the sample (hygroscopic and chemically com-bined). For the purpose of this research, the TA analysis wasperformed in the Institute of Geonics in Ostrava, with use ofSETSYS 24 Evolution by Setaram. The samples of about 50 g wereheated to 1200 1C with heating rate 10 1C/min.

Thermal analysis (TA) is a general term, which covers a variety oftechniques, recording the physical and chemical changes in a sub-stance as a function of temperature (Lever et al., 2014). Two of theranges of TA techniques were applied here: thermogravimetry (TG orTGA) and differential thermal analysis (DTA). TG measures the changein mass of a material as a function of time over a temperature range,using a predetermined heating rate. DTA is a technique where thetemperature difference between a sample and a reference material ismeasured as a function of time or temperature. A DTA curve providesdata on the changes in the sample, either exothermic or endoth-ermic (e.g., crystallization, melting, dehydration, organic matter comb-ustion, etc.).

The interpretation of TG and DTA curves in case of rocks is not aseasy as in case of pure minerals. In composite material, the decom-position temperatures could be shifted and some thermal effectsoverlap. However, based on examples of the use of thermal methodsin geology (Földvári, 2011), particularly with regard to the interpreta-tion of mineral composition of clay rocks, the obtained thermogramswere interpreted.

2.3. Analysis in Dean-Stark apparatus

The Dean-Stark solvent extraction is applied commercially,according to the methodology outlined by the Gas ResearchInstitute (GRI method) (Handwerger, 2012). The method enableswater extraction from the sample, which helps in total porositydetermination. It is possible to separate from the rock samplecapillary and interlayered water associated to the clays. The GRImethod is a multiple-weeks process of extraction and drying ofrock core samples. The results obtained by this method are oftenaffected by error, consisting in the extraction of more water thanwould be justified by taking into account free and capillary water(Handwerger, 2012). This could be explained by the contaminationwith atmospheric water or release of the hydroxyl water from thecrystal structure of clay minerals. In the studies described in thispaper, the rock material, before being placed in a Dean-Starkapparatus, was grinded. This approach is in line to the methodol-ogy proposed by Handwerger et al. (2011) as well as Such andLeśniak (2011). The fragmentation of the sample prior to extractionfacilitates getting solvent (toluene) into the pores in the rock.

Upon heating, the water comes out of the sample, togetherwith a solvent (toluene). The azeotropic mixture evaporating inthe form of vapors passes into the flask and the reflux condenser,where it is condensed. The condensed mixture is collected in avertical tube with a scale where it is partitioned into two layers bycooling to room temperature. Toluene as a liquid of lower densityaccumulates in the upper layer. The solvent is recycled to thereaction system automatically by the diagonal tube, whereas watercan be drained down periodically into the vessel (Fig. 1). For thepurpose of this study, the crushed shale samples (to a particle sizefrom 0.5 to 2 mm) were weighed and, sequentially, placed in theDean-Stark apparatus. Each sample was flooded with toluene andthen heated to the boiling point of the solvent (in case of toluene itis 115 1C). The experiment was completed after the stabilization ofthe condensed water volume.

The pore volume in the rock, occupied by water, was calculatedaccording to the following formula (API, 1998):

Ww ¼ Vwρw

Ws100%

where Ww is the pore volume in the rock, occupied by water [%],Vw is the volume of separated water [cm3], ρw is the density ofdistilled water [g/cm3] (0.9997 g/cm3 in room temperature), andWs is the initial weight of sample [g].

M. Labus et al. / Journal of Petroleum Science and Engineering 127 (2015) 482–489 483

Table 1Characteristic of the shale samples selected for the analysis.

Sampleno.

Name Photograph Age Origin Description Mineral content determined bymicroscope observations

TOC(wt%)

1 Gas shale Silurian Lublinprovince(Poland)

Silt-grain fraction. Darkgrey color is aconsequence of organicmatter content

Illite, kaolinite, quartz, plagioclase,pyrite

6.2

2 Gas shale Silurian Lublinprovince(Poland)

Silt-grain fraction, darkgrey

Illite, chlorite, quartz, plagioclase,pyrite

0.1

3 Gas shale Silurian East-EuropeanCraton (BalticBasin)

Silt-grain fraction, darkgrey

Illite, chlorite, quartz, plagioclaseand organic matter

1.5

4 Graptolite slate Silurian Holy CrossMountains(Poland)

Dark gray slaty shalerock

Clay minerals, quartz and organicmatter

0.7

5 Dictyonema oilshale

LowerOrdovician(PakerortStage)

Mäekülatunnels,Tallin(Estonia)

Argillaceous rock of abrown color

Clay minerals, sapropelite organicmatter (10–20%) and pyrite (9%)

10.3

6 Kukersite OilShale

UpperOrdovician,(KukruseStage)

VanakülaQuarry,(North-EastEstonia)

Argillaceous rock ofbrown color with alarge amount of whiteorganic detritus

Clay minerals, calcite and syderite.The content of organic matter comesto over 40% (basic organiccomponent is telalginit)

37.6

7 Metamudstone“roof slate”

Carboniferous(Culm)

HořejšiKunčice,Nizki Jesenik(CzechRepublic)

Gray slaty rock Muscovite, biotite, small grains ofquartz and plagioclase

1.1

M. Labus et al. / Journal of Petroleum Science and Engineering 127 (2015) 482–489484

2.4. Retort extraction

Retort method allows the extraction of liquid and gaseoushydrocarbons and pore water, hygroscopic and crystallization. Italso enables the measurement of the volume of separated compo-nents, which can then be subjected to further analysis. Thisanalysis includes determining the gas composition and the che-mical composition of pore waters.

In the basic retort method, there is a used apparatus composedof an oven capable of holding multiple retorts (Fig. 2). Such retortsare designed to hold 100–175 g of crushed core material. Eachretort is equipped with a threaded cap, with a gasket to preventleakage of condensable gases. A long, stainless steel condensingtube is connected to the opposite end of the retort (API, 1998).

On the basis of the volume of separated constituents, the volumeof pore space is determined. It is possible to broaden the furtherinterpretation of rock parameters, such as specific surface area. In thecase of shale rock samples, they are crushed and weighed prior toextraction. The samples were then placed in the retort and heatedcontinuously to achieve the characteristic temperature thresholds,diagnostic for the individual components of the reservoir fluids.During the process of heating, water (or oil and gas) in both freeand bound states is vaporized, collected, condensed and separated(Handwerger et al., 2011). In this work the retort analyses wereperformed in a retort furnace of VINCI Technologies (France), in thelaboratory of the Institute of Clean Technologies for Mining andUtilization of Raw Materials for Energy Use in Ostrava (Fig. 3). Theretort furnace is not under vacuum.

Concentration of the captured gases mixture was examined bygas chromatography, with the use of thermal conductivity detector(TCD) and a flame ionization detector (FID).

3. Results and discussion

3.1. Mineral composition of the shale rocks

Test results obtained by thermal analysis are presented below.The example of the sample thermogram (TG and DTA curves) isgiven in Fig. 4. The reactions, which were determined in thesamples, are given in Table 2.

It should be noted here that the presence of certain compo-nents of shale rocks cannot be precisely identified by thermalanalysis, due to the overlapping of the thermal effects on TG andDTA curves, when the reactions occur in the same temperaturerange. At lower temperatures, the oxidation of organic substancesoccurs, which is modified by the peak derived from quartz phasetransformation (α/β transition) at 573 1C. In the range of 50–700 1C, the dehydroxylation of clay minerals takes place, and atabout 850 1C the decomposition of carbonates occurs.

In case of goethite FeO(OH), which has been found in all shalesamples (except for kukersite – sample 6) the only recordedreaction is its decomposition to hematite and water, what isvisible as an endothermic peak on the DTA curve. The reactionbegins at a temperature of about 200 1C, and ends at 700 1C. In

Fig. 1. Dean-Stark extraction apparatus for volumetric determination of water(API, 1998).

Fig. 2. Diagram of retort oven (API, 1998). Fig. 3. Retort furnace of VINCI Technologies.

M. Labus et al. / Journal of Petroleum Science and Engineering 127 (2015) 482–489 485

case of samples composed of great amount of goethite, themaximum of the reaction falls between 300 and 400 1C (Meher,2014). However, in the case of shale samples, where goethite isstrongly fine-grained and dispersed, the maximum of decomposi-tion peak falls at lower temperatures, typically about 300 1C (Goss,1987) (Fig. 4).

In two samples: graptolite slate (sample 4), and kukersite shale(sample 6), a significant amount of calcite (CaCO3) was stated. Theevidence of the presence of calcite in the sample is an endother-mic peak on the DTA curve, extending generally in the range of600–950 1C. It results from the thermal decomposition reaction ofcalcite, accompanied by CO2 release. The temperature of max-imum peak on DTA curve, as well as the mass loss effect on TGcurve, usually decreases in proportion to the amount of calcite inthe sample (Földvári, 2011). In case of carbonate-rich kukersite oilshale, the decomposition of calcium carbonate takes place attemperature over 790 1C, which was proven also by the thermalFig. 4. Thermogram (TG and DTA) for sample 1. TG – thermogravimetrical curve,

DTA – differential thermal analysis curve.

Table 2TG/DTA analysis results for the shale rocks.

Sample Initial sample mass(mg)

Mineral composition Reactions Temperature range of the reaction(Beginning–(maximum)–ending)

1 49.34 Illite, kaolinite, quartz, goethite and organic matter 1. Clay mineral dehydration2. Goethite dehydration3. Organic matter combustion4. Kaolinite dehydroxylation5. Spinel phases crystallization

40–(190)–220 1C220–(300)–330 1CMax. 450 1C480–(550)–620 1CMax. 880 1C

2 50.90 Clay minerals, quartz, goethite and organic matter 1. Clay mineral dehydration2. Goethite dehydration3. Organic matter combustion4. Polymorphic quartz

transformation

130–(190)–230 1C230–(290)–330 1CMax. 340 1C575 1C

3 50.10 Illite, quartz, goethite and smaller amounts of organic matter 1. Clay minerals dehydration2. Goethite dehydration3. Organic matter combustion4. Clay minerals

dehydroxylation

120–(180)–220 1C220–(290)–340 1CMax. 480 1C480–(550)–610 1C

4 48.70 Illite, calcite, quartz, goethite, organic matter 1. Clay minerals dehydration2. Goethite dehydration3. Organic matter combustion4. Calcite thermal

decomposition

120–(190)–250 1C220–(290)–340 1CMax. 480 1C710–(800)–850 1C

5 49.40 Illite, goethite, organic matter 1. Clay minerals dehydration2. Goethite dehydration3. Clay minerals

dehydroxylation4. Organic matter combustion

130–(190)–240 1C250–(300)–340 1C500–(510)–560 1CMax. 500 1C (overlapping at reaction3)

6 50.24 Carbonates (calcite and siderite), small amounts of quartz andorganic matter

1. Clay minerals dehydration2. Thermal dissociation of

siderite3. Organic matter

combustion clay4. Calcite thermal

decomposition

150–(180)–220 1C325–(420)–510 1CMax. over 500 1C (overlapping atreaction 2)700–(785)–810 1C

7 49.98 Clay minerals, goethite 1. Clay minerals dehydration2. Goethite dehydration3. Clay minerals

dehydroxylation

130–(190)–210 1C250–(300)–350 1C490–(520)–610 1C

M. Labus et al. / Journal of Petroleum Science and Engineering 127 (2015) 482–489486

processing of kukersite in the technological furnace (Elenrumet al., 2008).

The combustion of organic matter is shown on the DTA curvesas an exothermic effect, what is connected to the mass loss visibleon TG curves. The maximum temperature of this effect is differentfor each shale rock sample (Table 2). The lowest (340 1C) is forsample 2, at a medium level (480 1C) in case of samples 1, 3 and 4,and the highest (over 500 1C) for samples 5 and 6. The tempera-ture of organic matter combustion (or pyrolisis in case of non-oxidizing atmosphere) is indicative to the maturity of shale rock(Jaco and Grebowicz, 2013). In our case, the higher the tempera-ture of the exothermic effect, the higher is the TOC value (Table 1),which confirms the mentioned relationship.

The broad peak of organic matter combustion may sometimesmask other reactions, e.g., exothermic peak characteristic for pyrite,located between 400 and 550 1C. The abundance of pyrite was provedby microscope observations (Table 1) in samples 1, 2 and 5, but itcould not be stated by means of TG/DTA method.

3.2. Pore volume analysis

The analysis of thermal curves (TG and DTA) allowed the determi-nation of percentage by weight of bound water, released as a result ofdehydration of clay minerals (illite and kaolinite), goethite dehydra-tion and, in some cases, dehydroxylation (releasing of OHþ groupsfrom clay minerals). This proportionwas converted (knowing the bulkdensity of the samples) to the volume percent (Table 3). For samples2 and 3, due to the disturbed TG curves, the quantification was notpossible.

As a result of the analysis in a Dean-Stark apparatus, the watercondensation was obtained only for samples 1, 4 and 5 (Table 3).The percentage of volume occupied by water (Ww) was 2.45%,1.17% and 2.47%, correspondingly. Measurement time was variedfrom 3 to 16 h. The measurement was terminated after a markedstabilization of the amount of water in the trap. For some samples(2, 3, 6 and 7) the presence of water was not found even after 24 hfrom the start of the experiment.

The measurement in the retort distillation apparatus was carriedout (for each of the samples) for 230 min. The oven temperature wasinitially maintained at 177 1C (Hensel, 1982) to remove pore water,absorbed water, interlayer clay water (e.g., smectite), and waters ofhydration (e.g., CaSO4), but not hydroxyl clay water. The watervolumes were recorded and the oven temperature was raised to atemperature between 538 and 649 1C. In case of the considered shalesamples, gypsum or montmorillonite were not identified; therefore, itis assumed that here their breakdown at high temperature was noinfluence on water saturation (and porosity). The final retortingtemperature of 538–649 1C might yield incorrect oil volumes, because

of decomposition at these temperatures of some solid hydrocarbons,e.g., kerogen, gilsonite, that might be contained in the sample material.

According to numerous works describing the results of ther-mooxidation of oil shales (e.g. Levy and Stuart, 1984; Kaljuveeet al., 2004, 2011), we should be aware of the phenomena, whichmight also play a role in the retort analysis of certain shale rocks,when the retort furnace is not under vacuum conditions. Thethermooxidative decomposition of shale proceeds in the followingstages (depending on the heating rate): (I) thermooxidation ofvolatile organic compounds (between 200 and 485 1C), (II) ther-mooxidation of heavier part of kerogen and fixed carbon (up to465–700 1C), and (III) decomposition of carbonates (up to 705–870 1C).

The last stage was not the case performed for this researchanalysis, as the samples were heated to a temperature of only650 1C.

Usually, a release of adsorbed water takes place at tempera-tures up to 250 1C (Kaljuvee et al., 2011). Kaljuvee et al. (2004)observed also the emission of a small amount of HCl, identified at240–330 1C and the beginning of emission of SO2 at 280–320 1C,with the maximum absorbance at 450–500 1C, some of whichcould be attributed also to the thermooxidation of pyrite, resultingalso in the production of Fe2O3 (Levy and Stuart, 1984).

In case of analyzed samples in the temperature range from 150to 200 1C, the first water condensation drops appeared (Fig. 5) andcapturing (in receiving tube) of the released gas started. In thistemperature range, the released water includes free water andwater from the dehydration of clay minerals (Table 4). Anotherthreshold temperature at which the increased release of fluidappeared, in some of the samples, is at about 500 1C. Thisphenomenon is related to the process of dehydroxylation. Forsamples 1, 4 and 5 at a temperature of about 450 1C the dark

Table 3Pore volume occupied by water in analysed samples [%].

Sample Bulkdensity(g/cm3)

Results from TG/DTA analysis Resultsfrom D-SapparatusDehydration Dehydroxylation Total

volume ofboundwater

Byweight

Byvolume

Byweight

Byvolume

Byvolume

1 2.80 0.98 0.27 2.16 0.60 0.87 2.452 2.87 n.o. n.o. 0 0 0 03 2.72 n.o. n.o. n.o. n.o. n.o. 04 2.73 0.70 1.91 0 0 1.91 1.175 2.50 2.80 0.70 8.04 2.00 2.70 2.406 1.54 1.36 0.21 0 0 0.21 07 2.72 0.71 0.19 1.60 0.44 0.63 0

n.o. – Determination not possible.

Fig. 5. Water content in the shale rocks [vol%], condensed in retort analysis.

Table 4Retort analysis results.

Sample Released water [%] in theprocess of

Liquidhydrocarbons

Pore spacesaturated withfluids in total (%)

Dehydration Dehydroxylation(%) (%)

1 0 4.54 0.7% 5.242 0 0 0 03 0.82 1.38 Paraffin 2.204 1.32 1.05 0 2.345 3.44 0.14 Traces 3.586 0.42 0.22 Oil 0.647 0.17 0.18 0 0.35

M. Labus et al. / Journal of Petroleum Science and Engineering 127 (2015) 482–489 487

substance condensed, which is associated with the transformationof kerogen into oil. The graph presented in Fig. 5 shows a suddenincrease in the volume of the condensed water in the analysis at650 1C. This is structural water, or in the case of samples contain-ing kreogen–water from its decomposition.

Sample 1, representing the gas shale from Lublin province, hasthe highest value of the pore space. In contrast, sample 2, from thesame area, does not reveal any liquid condensates in any of themethods used. The relatively high water content was determinedfor Dictyonema oil shale (sample 5), which confirms the result ofthe analysis carried out in a Dean-Stark apparatus and the result ofthermal analysis.

Our results can be compared, in the relevant temperature range, tothe outcomes of experiments on clay standards, (illite, chlorite,kaolinite and smectite) and mudstones, performed by Handwergeret al. (2011). In this respect, at temperatures up to 250 1C, pore waterwas released from all of the samples (except for sample 2).

Cumulative water yield patterns confirmed that in the case ofsamples 1, 3, 6 and 7, water bound by capillary, double-layer andsurface forces is released in the interval between 250 1C and600 1C (Fig. 5). The curve for sample 1 presents a steep rise ofexpelled water, possibly connected with illite/smectite, while theremaining samples' behaviour could be attributed to the presenceof chlorite. The further temperature increase – over 600 1C –

produced the structural water from the samples 1, 3, 4 and 5.The late effect for the samples 4 and 5 might be indicative to thepresence of kaolinite. High abundance of smectite typical forDictyonema oil shale (reported by Kilk et al., 2010) was notdetected in sample 5. According to Handwerger et al. (2011) forsmectite a constant water release would be observed, with noplateau between 250 1C and 650 1C.

Gas chromatography test results of captured gas mixture are givenin Table 5 (the total volume of gas was considered 100%, the result isthe percentage of gas in the mixture).

In the above table (Table 5) the attention is drawn by the relativelyhigh carbon dioxide content, in particular in samples 1, 5 and 6. Thisgas probably comes from the combustion of organic matter in theretort analysis. The high organic matter content in these samples isconfirmed by thermal analysis (Table 2), as well as the TOC value(Table 1). The increased amount of released carbon dioxide furthercorrelates with high values of porosity for samples 1 and 5, whichmay indicate that the porosity is related to the presence of organicmatter. In samples 3 and 4, pore space saturated with fluids (Table 4)is rather low, but higher than 2%. As the TOC value for these rocks is

low, the pore space is probably the intercrystalline porosity, notrelated to organic matter.

4. Conclusions

The tested samples of shale rocks demonstrate variable poros-ity value from 0% to greater than 5%. There was no completecorrelation of the obtained results for various test methods, but allthe methods show that the rock of the highest porosity is gas shalefrom Lublin province (sample 1) and Dictyonema oil shale (sample5). In both cases, the presence of high carbon dioxide contentcaptured in the analysis indicates that the porosity is related to thepores in the organic substance.

In conclusion, it should be noted that the determination of theporosity of tight shale rock is a difficult challenge, and demandingthe use of several complementary methods.

The advantage of the applied thermogravimetric methods (TG,DTA) is the requirement of a small amount of crushed sample.Furthermore, these methods allow the distinction between freeand bound water in the clay minerals, in contrast to the Dean-Stark extraction method. The most complete results were obtainedwith use of the retort distillation method. This analysis enables theconsistent separation of all phases of fluids in the rock by acontrolled temperature rise.

The temperature of organic matter combustion determined bymeans of thermogravimetric analysis is correlated to the totalorganic carbon content (TOC), and is indicative to the maturity ofshale rock.

Acknowledgment

The research was conducted thanks to the financial support bythe Institute of Applied Geology, Silesian University of Technology(BK-256/RG-7/2013).

References

API, 1998. Recommended Practices for Core Analysis, Recommended Practice 40.⟨http://w3.energistics.org/RP40/rp40.pdf⟩.

Clarkson, C.R., Jensen, J.L., Blastingame, T.A., 2011. Reservoir engineering forunconventional gas reservoirs: what do we have to consider? In: SPE Paper145080 Presented at the SPE North American Unconventional Gas Conferenceand Exhibition, 14–16 June, The Woodlands, TX, USA.

Table 5The composition of gas [vol%] captured during retort distillation.

Gas Sample

1 2 3 4 5 6 7

Acetylene o0.001 o0.001 o0.001 o0.001 o0.001 o0.001 o0.001Ethylene 0.043 0.002 0.002 0.018 0.075 0.035 o0.001Ethane 0.35 0.001 0.001 0.051 0.096 0.095 o0.001Propylene 0.023 o0.001 o0.001 0.009 0.032 0.070 o0.001Propane 0.19 o0.001 o0.001 0.061 0.070 0.076 o0.001i-Butane 0.025 o0.001 o0.001 0.020 0.012 0.019 o0.001n-Butane 0.072 o0.001 o0.001 0.022 0.030 0.023 o0.001i-Pentane o0.001 o0.001 o0.001 o0.001 o0.001 0,003 o0.001n-Pentane o0.001 o0.001 o0.001 o0.001 o0.001 0,011 o0.001n-Hexane o0.001 o0.001 o0.001 o0.001 o0.001 o 0.001 o0.001Methane 1.2 0.059 0.040 0.15 0.28 0.11 0,050CO2 6.2 0.90 0.90 1.5 18 8.3 0,50CO 0.40 o0.05 o0.05 0.070 1.7 1.0 o0,05Hydrogen 10 o3.0 o3.0 o3.0 o3.0 o3.0 o3,0Oxygen 17 22 22 21 13 15 22Nitrogen 63 78 78 78 64 75 75

M. Labus et al. / Journal of Petroleum Science and Engineering 127 (2015) 482–489488

Darłak, B., Kowalska-Włodarczyk, M., Such, P., 2011. Methodological aspects ofporosity and pore space measurements in shale rocks. Nafta-Gaz 5, 326–330.

Elenrum, A., Oja, V., Tali, E., Tearo, E., Yanchilin, A., 2008. Thermal processing ofDictyonema Argillite and Kukersite oil shale: transformation and distribution ofsulfur compounds in pilot-scale galoter process. Oil Shale 28 (3), 325–334.

Földvári, M., 2011. Handbook of thermogravimetric system of minerals and its usein geological practice. Occasional Papers of the Geological Institute of Hungary,p. 213.

Goss, C.J., 1987. The kinetics and reaction mechanisms of the goethite to hematitetransformation. Min. Mag. 51, 437–451.

Handwerger, D., 2012. Retort and Dean-Stark saturations in unconventional shalereservoirs. In: Materials Shale Science Conference “Developing the mentalpicture of reservoir quality and completion quality for tight shales”, Warsaw.

Handwerger, D.A., Suarez-Rivera, R., Vaughn, K.I., Keller, J.F., 2011. Improvedpetrophysical core measurements on tight shale reservoirs using retort andcrushed samples. In: SPE Paper 147456 Presented at the SPE Annual TechnicalConference and Exhibition, 30 October–2 November, Denver, CO, USA.

Hensel Jr., W.M., 1982. An improved summation-of-fluids porosity technique. SPE J.22, 193–201.

Jaco. Z., Grebowicz. J., 2013. Combined methods of thermal analysis for character-ization of rocks containing hydrocarbon. In: Proceedings of the 41st AnnualConference of the North American Thermal Analysis Society, 4–7 August,Bowling Green, KY, USA.

Kilk, K., Savest, N., Hruljova, J., Tearo, E., Kamenev, S., OJA, V., 2010. Solvent swellingof dictyonema oil shale. Oil Shale 27 (1), 26–36.

Kaljuvee, T., Pelt, J., Radin, M., 2004. TG-FTIR study of gaseous compounds evolvedat thermooxidation of oil shale. J. Therm. Anal. Calorim. 78, 399–414.

Kaljuvee, T., Keelmann, M., Trikkel, A., Kuusik, R., 2011. Thermooxidative decom-position of oil shales. J. Therm. Anal. Calorim. 105, 395–403.

Lever, T., Haines, P., Rouquerol, J., Charsley, E.L., van eckeren, P., Burlett, D.J., 2014.ICTAC nomenclature of thermal analysis (IUPAC Recommendations 2014). PureAppl. Chem. 86 (4), 545–553.

Levy, J.H., Stuart, W.I., 1984. Thermal properties of Australian oil shales: character-ization by thermal analysis and infrared spectrophotometry. In: Proceedings ofthe Symposium on Characterization and Chemistry of Oil Shales PetroleumChemistry. American Chemical Society St. Louis Meeting, April 8–13.

Milner, M., Mclin, R., Petriello, J., 2010. Imaging texture and porosity in mudstonesand shales: comparison of secondary and ion-milled backscatter SEM methods.In: SPE Paper 138975 Presented at the Canadian Unconventional Resources &International Petroleum Conference, 19–21 October, Calgary, AB, Canada.

Meher, S.N., 2014. Thermal analysis of nalco red mud. Int. J. Chem. Stud. 1 (5), 1–9.Sondergeld, C.H., Ambrose, R.J., Rai, C.S., 2010. Micro-structural studies of gas shale.

In: Proceedings of the Unconventional Gas Conference. SPE 131771, 23–25February, Pittsburgh, PA, USA.

Such, P., Leśniak, G., 2011. Badania petrofizyczne dla złóż niekonwencjonalnych.Wirtualny Nowy Przemysł. ⟨http://gazownictwo.wnp.pl⟩.

Such, P., 2012. Przestrzeń porowa skał łupkowych. Nafta-Gaz 9, 561–565.Such, P., Leśniak, G., Buniak, A., Ziemianin, K., Słota-Valim, M., 2011. Factors affecting

tight gas reservoir properties in deep laying rotliegend sandstones. In:Proceedings of the 73rd EAGE Conference and Exhibition 2011: UnconventionalResources and the Role of Technology. Incorporating SPE EUROPEC 2011, vol. 6,pp. 4465–4469.

M. Labus et al. / Journal of Petroleum Science and Engineering 127 (2015) 482–489 489