Fuel 89 (2010) 1863–1871

Contents lists available at ScienceDirect

Fuel

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

Geochemical characterization of solid residues, bitumen and expelled oil basedon steam pyrolysis experiments from Irati oil shale, Brazil: A preliminary study

Noelia Franco a,*, Wolfgang Kalkreuth b, Maria do Carmo Ruaro Peralba c

a Geology Department, UFRJ, Cidade Universitária, Ilha do Fundão, CEP 21949-900, Rio de Janeiro, RJ, Brazilb Geology Department, UFRGS, Av. Bento Gonçalves 9500, CEP 91501-970, Porto Alegre, RS, Brazilc Chemistry Institute, UFRGS, Av. Bento Gonçalves 9500, CEP 91501-970, Porto Alegre, RS, Brazil

a r t i c l e i n f o a b s t r a c t

Article history:Received 5 June 2009Received in revised form 7 November 2009Accepted 13 November 2009Available online 3 December 2009

Keywords:Steam pyrolysisIrati oil shaleParaná BasinBitumenExpelled oil

0016-2361/$ - see front matter � 2009 Elsevier Ltd. Adoi:10.1016/j.fuel.2009.11.018

* Corresponding author. Fax: +55 21 2598 9465.E-mail address: noeliaf@yahoo.com (N. Franco).

Steam pyrolysis experiments were performed on immature samples from the Irati oil shale, Paraná Basin,Brazil, using a maximum temperature of 350 �C with up to 98 h exposure time at that temperature. Theobjectives were to study geochemical and petrographical changes in the source material during stepwiseincrease in maturity, in steam conditions, comparing the properties of expelled oil with the bitumenretained in the solid residue after experimentation.

Petrographical and geochemical parameters such as vitrinite reflectance and Tmax, indicated an increasein maturity related to the exposure time of the organic matter to the maximum temperature. However,biomarker ratios such as 22S/(22S + 22R) C31 and C32 homohopanes, 20S/(20S + 20R) and abb/(abb + aaa) C29 sterane, which are considered to be indicators of organic matter maturity levels, didnot reach their equilibrium values. Some biomarkers frequently used as indicators of specific sourcesand/or paleoenvironments of deposition such as hopane/sterane ratio, and the concentrations of C27

and C29 steranes showed significant variations related to the stage of maturity. Based on the evaluationof Rock-Eval parameters, the transformation ratios in steam pyrolysis conditions reached levels higherthan 80% in samples having 9 and more hours of exposure time to maximum temperature. Bitumenwas found to be enriched in components of heavier molecular weight (resins and asphaltenes), whereasthe expelled oils contained higher quantities of aliphatic and aromatic components. At relatively lowmaturity levels the n-alkane distribution of expelled oils indicate a somewhat higher maturity level whencompared to the n-alkane distribution of the bitumen retained in the source rock, whereas at highermaturity levels the n-alkane distribution for the expelled oil and for the bitumen is very similar.

� 2009 Elsevier Ltd. All rights reserved.

1. Introduction temperatures changes the physical properties of these types of

Pyrolysis is a technique that can be defined as the thermaldecomposition of organic matter through the application of heatin the absence of oxygen. Pyrolysis experiments can be carriedout with or without the presence of liquid water as well as in thepresence of water steam. In the petroleum industry, pyrolysis isemployed to simulate oil generation during the subsidence of a ba-sin, carrying simulations in the absence (anhydrous pyrolysis) andin the presence (hydrous pyrolysis) of liquid water during the heat-ing process. The products of these two types of pyrolysis provide abetter understanding of the mechanisms taking place in naturalsystems during the process of oil generation and have been widelystudied by a number of authors [1–6]. On the other hand, pyrolysisexperiments in the presence of water steam, known as steam pyro-lysis, are frequently used in the petroleum industry aiming therecovery of heavy oils due to the fact that water steam at high

ll rights reserved.

oil, mainly by decreasing their viscosities [7,8]. Another use forsteam pyrolysis is the extraction of solid fuels, being reported a sig-nificant yield increase and variation in liquid products compositionwhen compared with other extraction techniques [9,10]. However,no study has been carried out on possible variations in geochemi-cal and petrographic parameters in rocks submitted to steam pyro-lysis as well as biomarkers ratios in the liquid products generated.

The present work aimed to register the influence of water steamon the maturation parameters Tmax and vitrinite reflectance deter-mined in the Irati oil shale, as well as the principal biomarker ratiosdetermined in the generated liquid hydrocarbons.

2. Sampling and methods

2.1. Sampling and sample preparation

The sample used in this study was from the Permiam IratiFormation, Paraná Basin, Brazil, containing organic-rich rocks oftype I/II kerogen, considered to have the greatest oil generation

1864 N. Franco et al. / Fuel 89 (2010) 1863–1871

potential of the basin [11–15]. The oil shale was collected from anopen-pit mine in São Mateus do Sul, Paraná State. The sample wassubsequently crushed, homogenized, and the 0.1–0.2 cm size frac-tion was retained for steam pyrolysis experiments.

2.2. Steam pyrolysis experiments

Steam pyrolysis experiments were performed in a stainlesssteel reactor (Parr Instrument Company, model 4651, 500 ml) con-nected to a Parr temperature controller Model 4842. Fifty grams ofcrushed sample and 50 ml of distilled water were put in the Stain-less Steel reactor, which was purged with nitrogen prior to thepyrolysis experiment in order to remove the air. To determinethe compositional variations in the bitumen and expelled oil, 6experiments were carried out at the same pressure and tempera-ture condition, 2100 psi and 350 �C, respectively. The exposuretime to the maximum temperature varied between 0 (immediateheating shutoff) and 98 h. The heating rate was 3.7 �C/min andthe temperature was continuously monitored and kept within±1 �C of the setting point.

2.3. Extraction and liquid chromatography

Liquid products were separated after cooling. Expelled oil wasseparated from water in a separation funnel and the autoclavewas rinsed with dichloromethane in order to remove traces ofexpelled oil. The bitumen generated under steam pyrolysis condi-tions was extracted from the pyrolyzate sample by Soxhlet extrac-tion with dichloromethane for 48 h. After extraction, the bitumenwas transferred to a rotary vacuum evaporator for solvent removal.Saturate, aromatic and NSO (resins and asphaltenes) fractions fromboth bitumen and expelled oil were separated in a Medium Pres-sure Liquid Chromatograph (MPLC), equipped with a HPLC pumpand SiO2 column. The solvents used for separation of the fractionswere n-hexane for the aliphatic fraction, n-hexane/toluene (60/40)for the aromatic fraction, and toluene/methanol (60/40) for theheterocomponents (NSO).

2.4. TOC and Rock-Eval pyrolysis

Total organic carbon (TOC) and Rock-Eval pyrolysis were carriedout on the original sample (unheated) and on the solid residuesobtained from the steam pyrolysis experiments. TOC analyses wereperformed in a carbon and sulfur non-dispersive analyzer SC-444,Leco� equipped with an infrared detector and Rock-Eval pyrolysisanalyses were carried out in a pyrolyser Rock-Eval 6, Vinci Tech-nologies equipped with a flame ionization detector (FID).

2.5. Vitrinite reflectance determinations (%Rrandom)

Unheated sample and solid residues from steam pyrolysis wereprepared for optical characterization. Mean random vitrinitereflectance (%Rrandom) measurements were carried out using aLeitz MPV3-SP microscope; 50�/0.85 oil immersion objective.Standards with known reflectance used were Sapphire 0.579%,Yttrium–Aluminium–Garnet 0.897% and Garnet–Gadolinium–Gal-lium 1.697%.

2.6. GC and GC–MS analyses

Aliphatic fraction of bitumen and whole expelled oil were runon a Hewlett–Packard 5890A gas chromatograph with a flame ion-ization detector (FID). Samples were injected (1 ll) in the splitlessmode with hydrogen as carrier gas. The oven was programmed foran initial temperature of 55 �C for 2 min., followed by heating ratesof 30 �C/min (55–190 �C), 1.5 �C/min (190–250 �C) and 2.0 �C/min

to a final temperature of 300 �C. The chromatography columnwas a high-resolution capillary column model J&W Scientific DB-5 (30 m � 0.25 mm i.d., film thickness 0.25 lm).

GC–MS analyses were carried out on aliphatic fractions frombitumen and expelled oils. The fractions were run on a gas chroma-tography Hewlett–Packard model 5890/II with a high-resolutioncapillary column model J&W Scientific DB-5 (30 m � 0.25 mmi.d., film thickness 0.25 lm), connected to a HP-5870MSD massspectrometer operating at 70 eV and selected-ion-monitoring(SIM). The SIM saturate analyses focused primarily on m/z 217(steranes) and m/z 191 (terpanes).

3. Results and discussion

3.1. Total organic carbon (TOC), Rock-Eval pyrolysis and vitrinitereflectance (%Rrandom)

The trends of TOC, Rock-Eval pyrolysis parameters (S2 peak andTmax) and vitrinite reflectance (%Ro) measurements obtained forthe sample from the Irati Formation before and after the steampyrolysis experiments are shown in Fig. 1. It is apparent that, asthe time of interaction between the source rock and the vaporwater at the maximum temperature (350 �C) increases, the TOCcontent decreases significantly (Fig. 1a), from 17.30 wt.% in the ori-ginal sample to 5.16 wt.% after 18 h at maximum temperature.However, a small increase in TOC content after this experimentaltime is also observed. A similar decreasing trend is observed forthe S2 peak, determined by Rock-Eval pyrolysis (Fig. 1b), whereasother parameters, such as Tmax and vitrinite reflectance used toevaluate the thermal evolution of the organic matter increase(Fig. 1c and d).

The trend of reduction in TOC content (Fig. 1a) is explained bythe fact that more and more organic matter is converted intopetroleum products as exposure to maximum temperature in-creases. At the same time the S2 peak decreases, since there is lessorganic matter available for the generation of hydrocarbons. Thisdecreasing trend is observed until experimental time for 18 h. Atlonger experimental times the S2 peak remains constant (Fig. 1b)suggesting that the kerogen in the sample has reached the highestpossible decomposition level for generation of hydrocarbons.

The Rock-Eval temperature of maximum hydrocarbon genera-tion (Tmax �C) employed as a maturity parameter for source rocks,increased with the time of exposure to maximum temperature(350 �C), as shown in Fig. 1c. This pattern was expected and repre-sents an increase in the thermal evolution of the organic matterpresent in the samples. The original sample has a Tmax value of433 �C, indicative of the transition from the immature to maturezone for petroleum generation [16]. As the time of steam pyrolysisexperiments was increased, Tmax values increased to more than450 �C in the experiments with 36, 72 and 98 h, which correspondsto the final mature zone in the oil window. This increase in matu-rity, as indicated by Tmax, was also confirmed by the vitrinite reflec-tance value (%Rrandom) (Fig. 1d), ranging from 0.26%Rrandom inthe original sample to 1.18%Rrandom in the experiments with 72and 98 h.

The hydrogen index (HI) is commonly used as a parameter todetermine the type of kerogen in immature samples and to esti-mate the organic matter thermal evolution [17,18]. In our studythe hydrogen index (HI) value of 812 mg HC/g TOC determinedin the original sample prior to pyrolysis indicates type I kerogen,and it can be clearly seen that this parameter decreases with theincrease of the experimental time (Fig. 2). This trend is alsoobserved for changes of the n-alkanes profile, as showed in Fig. 2.

Determination of the transformation ratio (TR) based on param-eters from Rock-Eval analysis and TOC measurements (Table 1),

Fig. 1. Changes in: (a) TOC content, (b) Tmax, (c) S2 peak and (d) vitrinite reflectance (%Ro) related to the increasing of the experimental time. Open diamonds represent theoriginal sample (unheated).

N. Franco et al. / Fuel 89 (2010) 1863–1871 1865

show that TR increased regularly with increasing exposure time tomaximum temperature, with a significant increase in experimentscarried out between 0 and 18 h (TR = 17% and 74%) and a less sig-nificant increase with values ranging from 74% to 86%, for experi-ments lasting longer than 18 h. Since crushed rock samples (0.1–0.2 cm size) were used in our experiments, the TR values may actu-ally be higher than those expected for larger fragments or blockswhich would be more representative of natural conditions ofpetroleum generation in a basin environment.

The observed TR values were high for samples having vitrinitereflectances lower than 0.6%Rrandom. The discrepancy may beexplained by the fact that vitrinite reflectances are suppressed bythe presence of alginite compounds [20,21]. Interlaboratory exer-cise on the Irati shale by the ICCP (International Committee of Coaland Organic Petrology) has shown that other maturity parameterssuch as lambda max and Q from spectral fluorescence measure-ments, Tmax from Rock-Eval analysis and the C29 steranes 20S/(20S + 20R) ratio all confirm a higher maturity level as indicatedby vitrinite reflectance measurement [22].

The studied sample of the Irati Formation contained a quite la-bile type I kerogen that probably generated petroleum in a rathernarrow range of kinetic activation energies [15]. Being relativelyfast under the steam pyrolysis conditions, the increase in petro-leum generation and transformation ratio of the kerogen in theIrati Formation is not steadily accompanied by vitrinite reflectance,which undergoes an increase under much slower kinetic parame-ters. This apparent discrepancy is more evident under experimen-tal conditions requiring higher temperatures to compensate for theshort time periods.

3.2. Gas chromatography

The gas chromatograms of bitumens show an increase in therelative abundance of low molecular weight n-alkanes with the in-crease of the experimental time at 350 �C in steam pyrolysis condi-tions, as observed in Fig. 2. The original sample shows a lowabundance of n-alkanes, regardless of their molecular weight.The gas chromatogram of the original sample (unheated) also

shows a greater abundance of the isoprenoids pristane (Pr) andphytane (Ph), when compared to the n-alkanes C17 and C18, as wellas an abundance of biomarker compounds. Both features beingcharacteristic of a low maturity level [23,24].

The relative abundances of n-alkanes with low to mediummolecular weight (n-C14–n-C20), as well as the heavier n-alkanes(n-C21–n-C42), increase with the time of the experiments (Fig. 2).This increase is caused by thermal cracking of the heavier mole-cules, such as the substituted aromatic compounds, resins, asphalt-enes and cyclic compounds. With the increase in exposure time tomaximum temperature (350 �C), significant changes also occur inthe relative abundance of the isoprenoids pristane and phytaneand C17 and C18 n-alkanes. Relative abundance of pristane and phy-tane is higher than C17 and C18 n-alkanes in experiments up to 18 h,whereas C17 and C18 n-alkanes predominate in experiments thatlasted more than 36 h.

Comparing the chromatograms of the expelled oil and bitumens(Fig. 3), it is apparent that at relatively low maturation level(experiment for 0 h), the expelled oil is richer in low molecularweight n-alkanes than the bitumen, suggesting a somewhat highermaturity level for the expelled oil as compared to its bitumencounterpart in the source rock, as expected. However, at highermaturity levels (experiment for 72 h) the chromatograms of bitu-men and expelled oil are very similar. Based on the variations ob-served in the chromatograms from Fig. 3, it can be suggested thatoils expelled by the source rocks at a low maturity levels tend toshow a higher thermal evolution than the corresponding bitumen.However, for samples with a higher thermal evolution, such differ-ences were not noticed.

Affected by the thermal evolution are also the pristane/phytane,pristane/n-C17 and phytane/n-C18 ratios and the Carbon Preferen-tial Index (CPI) in the bitumen and expelled oil (Table 1). Byincreasing the experimental time the pristane/phytane ratioincreases in bitumen and expelled oil, while pristane/n-C17,phytane/n-C18 and the CPI values decrease as expected.

The increase in the pristane/phytane ratios with increasingof the experimental time is due to greater stability of pristaneas compared to phytane, which results in the relatively high

Fig. 2. Trends of the hydrogen index (IH), oxygen index (OI) and alkanes profile with the increase of the steam pyrolysis time at 350 �C.

Table 1Data of solid residues and aliphatic fraction for generated bitumen and expelled oils after steam pyrolysis at 350 �C and 2100 psi in Irati oil shale.

Experimental time(h)

S1/(S1 + S2)(%)

TRa

(%)Aliphatics(%)

Aromatics(%)

NSO(%)

Pr/Ph Pr/n-C17 Ph/n-C18 CPIb

B O B O B O B O B O B O B O

Unheated 3 – 44 – 20 – 36 – 1.53 – 2.00 – 3.15 – 1.72 –0 6 17 27 29 21 41 52 30 1.56 1.66 2.20 2.38 2.20 2.58 1.81 1.359 8 54 24 19 26 37 50 44 1.87 1.92 1.92 1.75 1.59 1.26 1.11 1.0518 16 74 19 32 28 29 53 39 2.00 2.23 1.37 1.36 1.15 0.87 1.11 1.1936 11 77 23 36 10 40 67 24 2.00 2.88 0.82 0.88 0.64 0.67 1.15 1.0872 24 84 25 32 15 45 60 23 2.12 2.76 0.75 1.02 0.52 0.72 1.05 1.0698 22 86 15 30 31 38 54 32 1.93 3.22 0.92 1.02 0.70 0.48 1.04 1.08

NSO (nitrogen, sulfur, oxygen); Pr (pristane); Ph (phytane); B (bitumen); O (expelled oil).a %TR (transformation ratio) = {[(HIo � (HIr � TOCr))/TOCo]/HIo} � 100. The subscript ‘‘o” and ‘‘r” represent the values of the original and residual samples, respectively. The

TR formula, as proposed by Espitalié and Bordenave [19], considers the amount of generated petroleum as the difference between the original and the residual hydrogenindices.

b CPI (Carbon Preferential Index) = ½[(RC25 � C33/RC24 � C32) + (RC25 � C33/RC26 � C34)].

1866 N. Franco et al. / Fuel 89 (2010) 1863–1871

preservation of pristane during thermal evolution. The decreasedvalues found for the pristane/n-C17 and phytane/n-C18 ratios at

higher maturity levels are caused, according to Peters et al. [25],by thermal cracking of high molecular weight (>C18) n-alkanes. A

Fig. 3. Gas chromatograms of the aliphatic fraction showing the n-alkanes distribution of the bitumen (left column) and expelled oil (right column) obtained after steampyrolysis at 350 �C for 0 and 72 h.

N. Franco et al. / Fuel 89 (2010) 1863–1871 1867

relative increase of C17 and C18 n-alkanes in respect to pristane andphytane isoprenoids occurs, resulting in smaller values for theseratios as maturity increases, as showed in Fig. 2 under steam pyro-lysis conditions. The Carbon Preferential Index (CPI) decreases inboth bitumen and expelled oil as a function of the experimentaltime, due to the preferential thermal cracking of molecules withan odd number of carbon atoms over those with an even number[26].

The trends found for the n-alkanes profile and for the ratios cal-culated from isoprenoids and n-alkanes in bitumens as well as ex-pelled oils in the presence of water steam were similar to thosereported for natural and hydrous pyrolysis conditions [5,27–32].

3.3. Biomarker maturity ratios

Some biomarker ratios, which are considered to be indicators oforganic matter maturity levels, such as the 22S/(22S + 22R) ratio ofC31 and C32 homohopanes showed a slight trend to increase withincreasing experimental time (Fig. 4a), whereas the 20S/(20S + 20R) and abb/(abb + aaa) C29 sterane ratios show a muchstronger trend (Fig. 4b). However, these ratios did not reach theirequilibrium values. On the other hand, it can be observed inFig. 4c that the Ts/(Ts + Tm) ratio tends to decrease although theopposite was expected [25].

The increase of the 22S/(22S + 22R), 20S/(20S + 20R) e abb/(ab-b + aaa) ratios is frequently interpreted to be related to the mech-

anism of isomerization of the less stable biological configuration(22R, 20R, aaa) to the more stable biological configuration (22S,20S, abb) during the geological diagenesis process [33–36]. Thissame isomerization mechanism is proposed for the isomers Tmand Ts (Ts is more stable under geological conditions). However,a qualitative analysis of the chromatograms for the m/z 191 ofthe original samples and after experiments for 98 h, shown inFig. 5, suggest that the observed trend for the Ts/(Ts + Tm) ratiois result of the concentration increase of the isomer Tm liberatedfrom the kerogen, with the increase of the experimental time,without isomerization to the more stable Ts compound.

Studies by Farrimond et al. [37] and Dzou et al. [38] showedthat the compound liberation rates from kerogen and the preferen-tial degradation of one isomer over the other may influence theaforementioned trends. Taking into account that the purpose ofthe present work was to observe the behavior of the biomarkers ra-tio under water steam conditions and not the quantification of thecompounds, it was not possible to establish the influence of thoseliberation rates and degradation with the data set created in thepresent study.

3.4. Effect of maturity on hopane and sterane distribution

Observing the trends in the ratios tricyclic/pentacyclic terpanes(Fig. 4d) and hopanes/steranes (Fig. 4e) it can be noticed that theincrease of the steam pyrolysis time promoted the cracking of

Fig. 4. Trend of the ratios used to establish maturity level (22S/(22S + 22R) of C31 and C32 homohopanes (a), 20S/(20S + 20R) of C29 sterane (b), Ts/(Ts + Tm) (c) and tricyclic/pentacyclic ratios (d)) and organic matter source (hopane/sterane ration (e) and percentage of C27, C28 and C29 sterane (f)) with the increase of the experimental time at350 �C.

1868 N. Franco et al. / Fuel 89 (2010) 1863–1871

the higher molar weight molecules. This effect was more pro-nounced for steranes than for hopanes (Fig. 5), and confirms earlierstudies by Eglinton and Douglas [2]. On the other hand, the inver-sion in the relative abundance of the C27 and C29 steranes in longerexperiments (36 and 72 h) was also observed. The C27 and C29 ster-

anes show a distinct pattern with an increase in C27 and a decreasein C29 steranes with increasing experimental time (Fig. 4f),whereas the C28 steranes remained unchanged.

The increase observed in the relative content of the C27 steranescan be attributed to preferential thermal cracking of the C29

Fig. 5. m/z 191 and m/z 217 Mass chromatograms of the bitumen from the unheated sample and solid residues from steam pyrolysis at 350 �C for 18 and 98 h of oil shale fromIrati. TT = tricyclic terpane, g = gammacerane, empty and filled dots indicate (22S) and (22R)-17a, 21b(H) isomers, respectively.

N. Franco et al. / Fuel 89 (2010) 1863–1871 1869

steranes, which occurs during the thermal evolution of the organicmatter. The variations in the abundance of C27 and C29 steranesrelated to the level of maturity have implications on the interpre-tation of the organic matter source affecting the information thatcan be obtained from these, e.g. in immature organic matter thegreater abundance of the C27 sterane suggests a planktonic or mar-ine source, whereas the predominance of the C29 sterane is consid-ered to indicate contribution from terrestrial plants [25,26]. Sincethese parameters are affected by temperature care should be takenwhen the steranes abundance is used in the interpretation of depo-sitional environments.

3.5. Composition of bitumen and expelled oil

The chemical composition of bitumen and expelled oils (Table 1)showed that: (i) the amounts of aliphatic, aromatic and NSO com-pounds (resins and asphaltenes) in the bitumen extracted fromthe solid residues vary significantly, when compared to the bitu-men extracted from the original sample; (ii) the increase of theexperimental time do not affected significantly the chemical com-position of both expelled oil and bitumen as expected (an increaseof the aliphatic and aromatic fractions relative to the decrease ofpolar compounds with the increase of maturity level as showed

1870 N. Franco et al. / Fuel 89 (2010) 1863–1871

by Tissot & Welte [16, p. 422]); and (iii) when the composition ofexpelled oil and bitumen are compared, the percentages of aliphat-ics and aromatics, in the expelled oils, are always greater than thosein the bitumen, whereas the heaviest fraction (resins and asphalt-enes or NSO) is less abundant. This difference in the chemicalcomposition of expelled oil and bitumen has also been reportedfor hydrous pyrolysis conditions [4,39] and natural sequences[16,26].

Table 1 shows that bitumen extracted from the unheated sam-ple presents aliphatic and polar (NSO) fractions in almost identicalamounts, while bitumens generated during experiments presentan aliphatic fraction much smaller than the polar one. This differ-ence can be attributed to the fact that increasing in temperaturegenerates, initially, bitumens with chemical composition enrichedin heavier compounds (NSO) [4,16,26]. On the other hand, the dif-ference in chemical composition between bitumens and oils is dueto the preferential retention in the rock of polar compounds (NSO)when compared to aliphatic and aromatic compounds during theprimary migration process. During the primary migration thechemical composition of expelled oils is relatively enriched in ali-phatic and aromatic hydrocarbons which are less retained by thesource rocks, resulting a bitumen with a smaller fraction of thesecompounds and enriched in the heavier and polar fraction (resinsand asphaltenes or NSO). The preferential retention of resins andasphaltenes by the source rock occurs because these compoundsare macromolecules with high molar masses and with heteroatoms(N, S, O) as part of their structures conferring a significant polarityand a strong interaction with the matrix.

4. Conclusions

– Pyrolysis experiments at 350 �C in the presence of water steamshowed that increasing of the experimental time results inincrease of the maturation level of organic matter in the Iratioil shale. The Tmax (�C) values and vitrinite reflectances (%Rran-dom) showed that during the longest experimental time used(98 h), the maturity level of the rock reached the end of the oilgeneration window. However, the biomarker ratios fromhopanes and steranes (22S/(22S + 22R), 20S/(20S + 20R) e abb/(abb + aaa), determined in bitumens and expelled oils gener-ated, did not reach the equilibrium values, suggesting that theexperimental conditions had a significant effect on geochemicaland petrographical parameters of the solid residues when com-pared with the biomarkers ratios of the liquid products.

– Maturity levels in the expelled oil based on n-paraffin distribu-tions (gas chromatograms) may indicate an elevated maturitylevel, which does not correspond to the maturity level of thebitumen in the source rock, where the expelled oil was gener-ated. These differences decrease with increasing maturity.

– Biomarker ratios used to interpret depositional environments,like the pristane/phytane ratio, the hopanes/steranes ratio andthe distribution of C27, C28 and C29 steranes, were also affectedby the experimental conditions.

Acknowledgements

The authors thank the Geochemistry Section of PetrobrasResearch and Development Center (CENPES) for providing the sam-ple and bulk geochemical analyses, and the National PetroleumAgency (ANP) and CTPetro-Finep for financial support (Grant65.00.0026.00). W. Kalkreuth acknowledges the support by CNPqin form of a personal research grant (304991/2003-1). The authorsappreciate the helpful reviews provided by anonymous reviewer.

References

[1] Comet PA, McEvoy J, Giger W, Douglas AG. Hydrous and anhydrous pyrolysis ofDSDP Leg 75 kerogens – a comparative study using a biological markerapproach. Org Geochem 1986;9:171–82.

[2] Eglinton TI, Douglas AG. Quantitative study of biomarker hydrocarbonsreleased from kerogens during hydrous pyrolysis. Energy Fuels 1988;2:81–8.

[3] Koopmans MP, Carson FC, Sinninghe Damsté JS, Lewan MD. Biomarkergeneration from Type II-S kerogens in claystone and limestone duringhydrous and anhydrous pyrolysis. Org Geochem 1998;29:1395–402.

[4] Lewan MD. Laboratory simulation of petroleum formation: hydrous pyrolysis.In: Engel MKM, editor. Organic geochemistry. New York: Plenum Press; 1993.p. 419–42.

[5] Lewan MD, Bjorøy M, Dolcater DL. Effects of thermal maturation on steroidhydrocarbons as determined by hydrous pyrolysis of phosphoria retort shale.Geochim Cosmochim Acta 1986;50:1977–87.

[6] Lewan MD, Kotarba MJ, Curtis JB, Wieclaw D, Kosakowski P. Oil-generationkinetics for organic facies with Type-II and -IIS kerogen in the meniliteshales of the Polish Carpathians. Geochim Cosmochim Acta 2006;70:3351–68.

[7] Hongfu F, Yongjian L, Liying Z, Xiaofei Z. The study on composition changes ofheavy oils during steam stimulation processes. Fuel 2002;81:1733–8.

[8] Liu Y, Fan H. The effect of hydrogen donor additive on the viscosity of heavy oilduring steam stimulation. Energy Fuel 2002;16:842–6.

[9] Minkova V, Razvigorova M, Goranova M, Ljutzkanov L, Angelova G. Effect ofwater vapour on the pyrolysis of solid fuels: 1. Effect of water vapour duringthe pyrolysis of solid fuels on the yield and composition of the liquid products.Fuel 1991;70:713–9.

[10] Razvigorova M, Goranova M, Minkova V, Russyanova N. Comparative analysisof extractables and steam pyrolysis products from high-volatile bituminouscoal. Fuel 1995;74:1333–42.

[11] Padula VT. Oil shale of Permian Irati Formation, Brazil. AAPG Bull1969;53:591–602.

[12] Neto CC, Furtado EG, Concha FJM, Cardoso JN, Quadros LP. Anomalies in thestratigraphic distribution of hydrocarbons in the Irati oil-shale. Chem Geol1978;23:181–92.

[13] Da Silva ZCC, Cornford C. The kerogen type, depositional environment andmaturity, of the Irati shale, upper Permian of Paraná Basin, Southern Brazil.Org Geochem 1985;8:399–411.

[14] Mendonça Filho JG. Estudo petrografico e organo-geoquímico de amostras defolhelhos da Formação Irati, Permiano Superior da Bacia do Paraná. Dissertaçãode Mestrado, Universidade Federal do Rio Grande do Sul; 1994.

[15] Araujo LM. Análise da expressão estratigráfica dos parâmetros de geoquímicaorgânica e inorgânica nas seqüências deposicionais de Irati. Tese deDoutorado, Universidade Federal do Rio Grande do Sul; 2001.

[16] Tissot BP, Welte DH. Petroleum formation and occurrence. 2nded. Berlin: Springer-Verlag; 1984.

[17] Espitalié J, Laporte JL, Madec M, Marquis F, Leplat P, Paulet J, et al. Méthoderapide de caractérisation des roches mères, de leur potentiel pétrolier et deleur degré d’evolution. Reveu de L’Institute Français du Pétrole1977;32:23–42.

[18] Peters KE, Moldowan JM, Schoell M, Hempkins WB. Petroleum isotopic andbiomarker composition related to source rock organic matter and depositionalenvironment. Org Geochem 1986;10:17–27.

[19] Espitalié J, Bordenave ML. Rock-Eval pyrolysis. In: Bordenave ML, editor.Applied petroleum geochemistry. Paris: Éditions Technip; 1993. p. 237–61.

[20] Hutton AC, Cook AC. Influence of alginite on the reflectance of vitrinite fromJoadja, NSW, and some other coals and oil shales containing alginate. Fuel1980;59:711–4.

[21] Kalkreuth W, Macauley G. Organic petrology and geochemical (Rock-Eval)studies on oil shales and coals from the pictout and antigonish areas, NovaScotia, Canada. Bull Can Petrol Geol 1987;35:263–95.

[22] Borrego AG, Araujo CV, Balke A, Cardott B, Cook AC, David P, et al. Influence ofparticle and surface quality on the vitrinite reflectance of dispersed organicmatter: comparative exercise using data from the qualifying system forreflectance analysis working group of ICCP. Int J Coal Geol 2006;68:151–70.

[23] Mackenzie AS. Application of biological markers in petroleum geochemistry.Adv Org Geochem 1984;1:115–21.

[24] Thomas BH, Moller-Pederesen P, Whitaker MF, Shaw ND. Organic facies andhydrocarbon distributions in the Norwegian North Sea. In: Thomas BM, editor.Petroleum geochemistry in exploration of the Norwegianshelf. London: Graham and Trotman; 1985. p. 3–26.

[25] Peters KE, Walters CC, Moldowan JM. The biomarker guide: biomarkers andisotope in petroleum exploration and earth history. 2nd ed. CambridgeUniversity Press; 2005.

[26] Hunt JM. Petroleum geochemistry and geology. 2nd ed. New York: W.H.Freeman and Company; 1995.

[27] Alsaab D, Suarez-Ruiz I, Elie M, Izart A, Martinez L. Comparison of generativecapacities for bitumen and gas between carboniferous coals from Donets Basin(Ukraine) and a cretaceous coal from Sabinas-Piedras Negras Basin (Mexico)during artificial maturation in confined pyrolysis system. Int J Coal Geol2007;71:85–102.

[28] Koopmans MP, Rijpstra WIC, de Leeuw JW, Lewan MD, Damsté JSS. Artificialmaturation of an immature sulfur- and organic matter-rich limestone from theGhareb Formation, Jordan. Org Geochem 1998;28:503–21.

N. Franco et al. / Fuel 89 (2010) 1863–1871 1871

[29] Monthioux M, Landais P, Durand B. Comparison between extracts from naturaland artificial maturation series of Mahakam delta coals. Org Geochem1986;10:299–311.

[30] Monthioux M, Landais P, Monin J-C. Comparison between natural and artificialmaturation series of humic coals from the Mahakam delta, Indonesia. OrgGeochem 1985;8:275–92.

[31] Soldan AL, Cerqueira JR. Effects of thermal maturation on geochemicalparameters obtained by simulated generation of hydrocarbons. OrgGeochem 1986;10:339–45.

[32] Elie M, Mazurek M. Biomarker transformations as constraints for thedepositional environment and for maximum temperatures during burial ofOpalinus clay and Posidonia shale in northern Switzerland. Appl Geochem2008;23:3337–54.

[33] Mackenzie AS, Lewis CA, Maxwell JR. Molecular parameters of maturation inthe Toarcian shales, Paris Basin, France–IV. Laboratory thermal alterationstudies. Geochim Cosmochim Acta 1981;45:2369–76.

[34] Mackenzie AS, McKenzie D. Isomerization and aromatization of hydrocarbonsin sedimentry basins formed by extension. Geol Magaz 1983;120:417–70.

[35] Seifert WK. Steranes and terpanes in kerogen pyrolysis for correlation of oilsand source rocks. Geochim Cosmochim Acta 1978;42:473–84.

[36] Seifert WK, Michael Moldowan J. Applications of steranes, terpanes andmonoaromatics to the maturation, migration and source of crude oils.Geochim Cosmochim Acta 1978;42:77–95.

[37] Farrimond P, Taylor A, TelnÆs N. Biomarker maturity parameters: the role ofgeneration and thermal degradation. Org Geochem 1998;29:1181–97.

[38] Dzou LIP, Noble RA, Senftle JT. Maturation effects on absolute biomarkerconcentration in a suite of coals and associated vitrinite concentrates. OrgGeochem 1995;23:681–97.

[39] Peters KE, Moldowan JM, Sundararaman P. Effects of hydrous pyrolysis onbiomarker thermal maturity parameters: Monterey phosphatic and siliceousmembers. Org Geochem 1990;15:249–65.