analysis.

The molecular characterization and quantication of all individual gases generated were per-

formed by GC. C
5

compounds, whic h are onlypartiall y recovered unde r vacuum, were d etectedbut not quantied.For HiP reactors, the

volume of generated gaswas measured by connecting the reactor to an exter-nal stainless steel line of known volume (ca. 14 ml)equipped with

a barometer and set under vacuum.The measured gas pressure (after equilibration fora few minutes) was compared to the initial heliumpressure i

nstalled in the re actor (before heat ing)using the sa me connection line. The dierence ingas pressure was con verted into a gas volume,correspon ding to the

amount of gas generated uponpyrolysis.Th e entire volume o fgaswascollected inaglass bottle at room temperature and pressure andsubsequently

analysed by gas chromatography.

3.3.5. Liquid fraction recovery and characterization

Tube opening was carried out at room temperature under

atmospheric pressure. For characteriza-tion of the liquid fraction, the pyrolysis productswere suc cessively extract ed rst with
n

-pentane,then with dichloromethane (DCM). The gold tubewas pierced in a special device equipped with a nee-dle and full with
n

-pentane. After the gas release, the

n

-pentane solution was transferred to a ask. Thegold tube was recovered and cut into

small piecesthat were transferred into the

n

pentane solution. To assure good extraction eciency, the insolubleresidue

was crushed in an agate mortar full of

n

-pentane prior to extraction. During this procedure, $

5 mg of insoluble residue was systematically lostbecause some crushed organic particles remaine don the mortar wall. Then, the content of the mortarwas

transferred into the ask containing the

n

-pen-tane solution, stirred for 1 h at reux and ltered.After ltration, the

n

-pentane solution was concen-trated to 10 ml and weighed for volume quantica-tion. A rst aliquot of $ 2 ml was weighed and

usedto quantify t he C
6

C
14

compounds. This aliquotwas fractionated by liquid chromatography

into sat-urates and aromatics. The two recovered fractionswere not evaporated, but were directly injected intoa gas chromatograph (GC) after adding an

internalstandard (squalane) to quantify the C

6

to C
14

com-pounds in the two solutions. The C

5

compounds coeluted with the solvent were not quantied. A second aliquot of three ml was evaporated and weighedbefore fra ctionation by liquid

chromatography intosaturates, aro matics, and

n

pentane NSOs w hichcorrespond to resin compounds because

asphaltenesare not soluble in

n

-pentane.After
n

-pentane extraction, the insoluble fractionrecovered on the lter was

extracted with DCM.This extra ct was veri ed to contained less t han1 wt% of saturates and aromatics; therefore, >98%are polar c ompounds, whic

h include the tot alasphaltene fraction and resins that are not solublein
n

-pentane. After solvent evaporation, this extractwas dried

and weighed. Finally, the insoluble resi-due together with the gold pieces were dried andweighed. Because the empty tube was weighed, thetotal

amount of insoluble residue could be determined. However, if small gold particles were lostduring tube cutting, the weight of the insoluble res-idue

could be overestimated.Aft er gas recovery in the HiP reactor, the totalcontent of th e reactor was tra nsferred to a as kfor drying at ambient

atmosphere before crushingto between 60 and 104

l

m. This step was necessarybecause the recovered sample was very wet afterpyrolysi

s and could not be eciently extracted withorganic solvent. During this drying step, a signi-cant part of the low molecular weight hydrocarbonfracti

on may be lost. Nevertheless, the ir residualamount was systematical ly quantied. Th e sameanalytical procedure as that for liquid characterization of pyrolysat

e from gold tube reactor wasappli ed.The complete analytical procedure is summarizedinFig. 2.
3.3.6. Mass balance calculations

For the Type II kerogen experiments, the follow-ing chemical classes were quantied: CO
2

,H
2

S, C
1

C
4

,C
6

C
14

saturates and aromatics, C

14+

saturatesand arom atics,

n

pentane and DC M NSOs andthe solid residue. Water and C

5

compounds weren ot recovered in this analytical procedure. The C

5

compounds were present in very low amounts andtheir contrib ution to the mas

s balance wasne glected. For water, a total yield of 4.8% was previously determi ned on the same sample (Beharet al., 1995). Therefore, at low

kerogen conversion,the generated water was expected to be low andthen, to incre ase to a maximu m value of 4.8%. For the insoluble residue, a loss of

$ 5 mg was estimated, representing 20 mg/g of the total mass bal-ance as the initial amount of kerogen was 250 mg.For experiments on

the Type III lignite, the calculation is more complicated beca use of the freewater i nside the initial lignite. First, the amount

F. Behar et al. / Organic Geochemistry 39 (2008) 122

of residual kerogen was estimated using a carbonmass balance. Carbon, hydrogen and

oxygen content of the insolu ble residue and t he

n

pentaneand DCM NSOs were dete rmined by eleme ntalanalysis. Ato

mic composition was directly dete r-mined on the gas fraction. The saturate and aromatic hydrocarb on fractions wer e estimated toha ve an average va lue of 85% carb

on and 15%hydr ogen and 90% ca rbon and 10% hy drogen,respective ly. Then, the corresponding oxygen bal-ance was done on all products except water whichwas

estimated by dierence with the initial oxygencontent in the lignite. It was checked that the estimated water los s did not exceed

the maximumyi eld determined at $ 15% in a previo us study(Behar et al., 2003).For speci c experiments on the
n

pentane andDCM extracts, the tot al

n

pentane solution wasused only for quanti cation of the C

14+

fractiondue to the low amount of material initially available. Therefore, neither the amou nt of the C
14

fraction, nor the elemental compo sition of theC

14

fraction was determined. For experiments onasphaltenes ge nerated from the

Type II kerogen,t otal mass balance was calculated on the liquid fraction and on the insoluble residue. Gases were quantied only on a subset of experiments at

dierenttemperat ure/time conditions.

3.4. Kinetic modeling

For kinetic modeling of thermal cracking underlaboratory pyrolysis, we ass

ume that thermal lyunstable chemical classes decompose through rstorder processes that do not depend on oil composi-tion. This means that

there is no chemical interaction between ch emical classes d uring thermalcra cking; hence eac h decomposition reactioninvolves one reactant only. Moreover,

reaction ratesare temperature dependent and do not vary withpressure. Ra te constants are derived from the Arrhenius equation:
k

exp
E
a

RT

1 where
k

= rate constant,
A

= frequency facto r,
E
a

= activation ener gy,

R

= 8.314 J mol

,and
T

= temperature. The average rate of decompo-sition of a chemical class can be accounted for by aset of independent, parallel reactions, i.e.

X
i

!
P
i
;

; k
i
;

a
i
;

1
;

X
1

a
i
;

1
;

6
i

X
i


a
i
;

1
;

X
m

!
P
i
;

; k
i

a
i
;

2
;

X
1

a
i
;

2
;

6
i

X
i

a
i
;

2
;

X
m

.........

!
P
i
;

; k
i
;

a
i
;

n
;

a
i
;

n
;

6
i

X
i

a
i
;

n
;

X
m

8>>>> >><>> >>>>:

2


Gases (vacuum line)C
14+

recoverylight HC : GC
C, H, OcontentKerogen weight lossC
14+

Extract

Saturates Aromat icsPentane NSOs

SAR npentaneextractionn -C5residue

DCM NSOs
Pentane NSOs= soluble in n -C
5

DCM NSOs= s oluble in DCM = resins 2 + asphaltenes DCM extraction C

14+

ExtractDCMresidu e= resins 1C
14+

Fig. 2. Analytical procedure for closed system experiments.6

F. Behar et al. / Organic Geochemistry 39 (2008) 122

where
X
i

is a reacting chemical class and

X
j
5

aregenerated classes,
k
i ,

is the rate constant of reaction

j

,
P
i , j

is the weight fraction of

X
i

that reacts throughrea ction

j

,
a
i

, j , h

is the stoichiometric coecient forthe conversion of

X
i

into
X

in reaction
j

,
m

is thenumber of ch emical classes in the model, and nthe

number of reacti ons involved in t he thermaldecom position of

X
i

.In this latter example, mass conservation impliesthat

X
nj

P
i
;

3 and for each reaction

j

X
mh

1
h

6
i

a
i
;

j
;

1 4

Note that if
c
i , j

is the conversion of reaction

j

duringthe decomposition of
X

, then the global rate (

q
i

) of thermal cracking of
X
i

is given by
q

X
nj

P
i
;

d
c

i
;

d
t

5 withd
c
i
;

d
t

k
i
;

1
c
i
;

6 Kinetic (
E
a

and
A

for each reaction in the model)and stoichiometric (

P
i , j

and
a
i , j

, h

) parameters werenumerically calibrated according to pyrolysis massbalances. Optimisation was achieved with

GeoKinComposit ional software, an IFP kinetic simulatorthat allows rate parameters to be adjusted by nding the minimum of an error funct ion

, corresponding to the sum of squared di erencesbetwe en measured and computed amounts:

F

X
si

X
mj

r
2

ij

7 where
s

= the number of experiments,

m

= the num-ber of chemical classes of the model, and

r
ij

= residual, i.e., the di erence between measured andcal

culated amounts of compound

j

in experiment
i

(
X
ij measured

ij calculated

).The minimization of
F

was performed using amodied Levenb ergMarquardt algorithm with ac

ombination of two constraints: mass conservation(the sum of the stoichiometric coecients for eachreaction must be equal to 100% of the

experimentalvalu e) and hydrogen conservation (using the average hydrogen co ntent of each clas s, each stoichiometric reaction must respect hyd rogen balancebet

ween reactants an d products).

4. Results and discussion 4.1. Bulk kinetics in open system

The Type II kerogen and Type III lignite werepyrolysedin

anopensystematd i erentheatingra tesasexplainedin Section3.Pyrolys is datawereusedt oderivethebulkki neticparameters( Table3).Inordert o compare kinetic parameters for

kerogen crackingin open and closed pyrolysis systems, it is necessaryto subtr act the partial act ivation energies corre-sponding to those of late gas

coming from the resid-ual kerogen (Lorant and Behar, 2002). Thus, thepartial
E
i

above 54 kcal/m ol were subtracte

d forboth sample and the resulting bulk kinetics (Table3). For Type II, the resulting transformation ratio(TR) is very similar to that using data

fromTable4becau se the higher part ial

E
i

account for only1 .8%. For the lignite sample, these higher partial

E
i

Table 3Bulk kinetics obtain ed on the Type II kerogen and Type IIIlignite: initial and corrected values of S2
i

(mg/g per kcal/mol)after subtraction of

E
i

above 54 kcal/mol
E
a

(kcal/mol) Type II kerogen Type II li gniteInitial Correc ted Initial Correct ed4 0 0 . 0 0 . 0 0 . 0 0 . 0 4 2 0 . 0 0 . 0 0 .

0 1 0 6 9 7 2 1 5 4 8 4

0 . 0 4 4 . 4 1 . 4 . 3 0 . 3 4 3 . 8 3 . 5 . 1 5 . 4 8 1 9 . 7 0 . 1 1 3 . 0 4 . 5 5 0 6 2 . 6 3 . 8 2 5 . 2 8 . 4 5 2 . 7 8 . 8 3 . 5 3 8 . 6 5

4 5 0 0 4 5 0 1 2 0 0

2 . 0 1 1 . 2 5 6 0 . 0 3 . 0 5 8 0 . 0 0 . 0 . 3 0 . 5 0 0 . 2 1 . 2 6 4 0

2 1 . .

6 . .

. 2 7 7 0 2 0 0 0 0 0 1

0 .

. .

6 . .

0 0 1 4 0 0 0 0 0 0 0 2

. 0 . 0 6 0 . 0 . . 0 . 5 0 . 0 . 7 2 . 0 . 0 7 0 .

0 6 0 0 0 0 0 0 0 0 4 0

. 8 0 0 6 8 . 0 . 0 0 0 . 1 . 0 0 0

. .

7 . .

. .

0 0 . 0 7 6 0 . 0 0 . 0 0 . 0 0 . 0 7 8 0 . 0 0 . 0 0 . 0 0 . 0 8 0 0 . 0 0 . 0 0 . 0 0 . 0 Total 10 0.0 100.0 100 .0 100.0

A

(s

1 . 3 5

10
13

4.11

10
13

Sopt (mg/g) 366.6 129. 9

F. Behar et al. / Organic Geochemistry 39 (2008) 122

represent 10.6% and the resulting TR are shifted tolower values

without exceeding a dierence of 9%.Using these corrected kinetic parameters, it was possible to determin e the kerogen T R whatever thep

yrolysis conditions in a closed system.The chemical composition of the S2 peaks aregiven inTable 5with duplicate q uantication of t

he liquid fractions. In both samples, the NSO com-pounds predominate. The total hydrocarbons repre-sent only 152154 mg/g for the Type II kerogen and62

mg/g for the lignite.

4.2. Kinetic model for the Type II kerogen and associated NSOs in closed system pyrolysis

Mass balances are given inTable 6.

In order tocalculate total recovery, the loss of the insoluble res-idue of 20 mg/g was added in each experiment. Thetotal recover y is between 92

% and 105%. A sexplainedinSect ion3,thispercenta gedecreaseswithi ncreasing maturity because the generated water wasnot quanti ed. The total recovery is

higher than100% for m ost of the experi ments at TR belo w15%. This excess can be explained by a loss of somesmall gold particles when opening

and cutting thegold reactor. At such low maturity, not enough gasis generated to inate the gold reactor. Once heated,the gold tube remains

therefore, at and opening itrequires cutting the tube into smaller pieces.Th e amount of insoluble residue (Fig. 3) starts todecrease at the lowest

temperature with the releaseof nonhydrocarbon gas es. At TR higher than40%, the amount of insoluble residue reaches a rel-

atively constant value $ 4850%, and then increasesat the most severe conditions. This means that thereis anoth er source of soli

d residue that is addedthrough secondary cracking reactions. The sum of CO

2

and H
2

S reaches 8% in the most severe condi-tions. Added to the yield of water at 4.8%, the totalyield of nonhydrocarbon gases shows a maximumvalue of

13%.Fig. 3shows a plot of the generation curves forthe two types of NSOs and the total hydrocarbon(HC ) yield, which is the sum of the C
1

C
4

,C
6

C
14

,C
14+

saturated and C
14+

aromatic compounds.The

DCM NSOs are generated as soon as kerogenstarts to crack, following a trend with a very sharpslope that reduces at kerogen TR of 10%. A

steadystate is observed over a large TR range from 25% to70% and nally, a continu ous decrease occ urs athigher TR. This general tren d means that DC

MNSOs are prim ary products of k erogen crackinga nd they are labile compounds. In contrast, the generation curve of the
n

-pentane NSOs for kerogenTR between 0% and 10% shows a slightly concavetrend tha t is typical of a secondary prod uct. Thus, these compounds are

believed to be mainlyproduced by secondary cracking of the DCM NSOsand are not generated from the kerogen. As these
n

-pentane NSOs are observed at the onset of kerogencracking, the DCM NSOs appear to be undergoingsecon dary cracking as soon as they are

generated.The st eady state of the

n

-pentane NSOs yield isshif ted to a higher TR range 45 90% than that of the DCM NSOs.

Table 4Comparison of trans formation ratio (TR) using bulk kineticsfromTabl e 3(TR 1) with that using parameters fromTable 3(TR2):
D

TR = TR 2 TR 1Sample

C)
t

(h) TR 1 (%) TR 2 (%)

D

TR (%)T y p e I I k e r o gen 250 24 1.1 1.1 0.02 7 5 2 4 5.3 5.4 0.

1 300 24 19.7 20.1 0.4 325 24 54.1 55.1 1.0 350 24 89.7 91.4 1.8 Type III lignit e 250 21 1.6 1.7 0.22 7 5 2 4 8 . 5 9.5 1.0 30 0 24

24.6 27.4 2.8 325 22 49.4 55.2 5.8 350 22 76.6 85.3 8.7 Table 5Chemical composition (mg/g) of the S2 peak for the Type II and Type III samples: HC = C
1

C
4

+ total C
6

C
14

+C
14+

saturates andaromaticsS a m p l e C O
2

C
1

C
2

C
4

C
6

C
14

fraction C
14+

fraction Tot alS a t A r o S a t A r o N S O s S 2 H C Type II ker ogen 38.0 10.

5 20.7 23.0 1 2.4 24.8 60.5 215.7 367.6 1 51.910.5 20. 7 22.1 12.7 25.1 63.2 22 0.2 374.5 15 4.3Type III lignite 10.6 8.2 7.6 6.7 16.5 12.3 4 5.1 107.0 6

1.9113.9 10 .6 8.2 7.5 6 .4 15.9 13. 0 46.7 108. 3 61.68

F. Behar et al. / Organic Geochemistry 39 (2008) 122

Table 6Mass balances (%) obtained on the Type II kerogen articially matured in a closed pyrolysis system ( = under limit detection): HC = C
1

C
4

+ total C
6

C
14

+C
14+

saturates andaromatics
T

C)
t

( h ) T R ( % ) T o t a l g a s C
6

C
14

C
14+

HC NSOs fracti on Residue Tota l recovery T o t a l H C CO

2

H
2

SC
1

Sat Am Sat Ar o Pentane DC M Recov. Lost 2 2 5 2 4 0 . 1 1 . 4 0 . 1 0 . 2 1 . 8 9 4 . 3 2 . 0 9 9 . 8 2 2 5 7 2

0 . 4 0 . 5 0 . 9 5 0 1 2 5 . 4 8 0 0 .

6 . 0

1 . 9 1 . 8 4 2 . 2 . 2

0 9 0 1 . 8 0 . . 1 3 2 . 5

9 3 1 . 1 1 0 . 0 9 3 1 0 1 2 2 . . 7

. 0 2 . 7

4 2 . 0 0 . 9 0 5 0 2 4 1 1 . 7 0 . 1 . 4 2 . 6 . 9 2 . 0 1 . 5 0 . 5 0 7 2 7 2 . 2 0 0 . 1

9 1 1 . .

0 4 0 2 7 3 0 9

0 .

. 6 3 . 4 . 5 2 . 0 3 . 4 0 . 7 5 3 0 1 . 9 0 0 . 1 . 4 1 . 7 0 . 8 2 . 9 7 . 0 0 1 2 7 5 6

1 . 0 0 8 9 7 2 1 0 9

. 9 5 . 1 . 0 0 . 9 . 9 . 5 . 2 0 . 1 5

2 . 0

0 . 0 8 4 . 9 4 2 . 0 7 0 . 2 2 9 2 . 6 0 . 5 0 . . 0 0 . 0 0 . 2 0 . . 5 8 5 .

9 2 . 0 9 7 . 4 0 . 5 2 7 5 2 4 5 . 4 2 . 2 0 . 6 0 . 2 0 . 1 0 . 0 0 . 1 0 . 4 1 . 2 7 . 0 8 6 . 2 2 . 0 1 0 0 . 1 0 . 8 2 7 5 4 8 8 . 8 2 . 6 0 . 7 0 . 2 0 . 1 0 . 1 0 . 1 0 . 7

2 . 1 8 . 7 8 2 . 6 2 . 0 9 9 . 9 1 . 2 2 7 5 7 2 1 1 . 5 3 . 0 0 . 8 0 . 3 0 . 2 0 . 1 0 . 2 1 . 1 2 . 6 9 . 0 7 8 . 9 2 . 0 9 8 . 3 2 . 0 3 0 0 1 2 . 0 2 . 0 0 . 5 0 . 1

 0 . 7 4 . 2 9 2 . 0 2 . 0 1 0 1 . 5 0 . 1 3 0 0 2 3 . 5 2 . 1 0 . 6 0 . 2 0 . 0 0 . 0 0 . 1 0 . 2 0 . 8 5 . 0 8 9 . 9 2 . 0 1 0 0 . 9 0 . 5 3 0 0 3 4 . 8 2 .

3 0 . 1 6 . . 0 . 2 . 9 0 . 1 2 2 . 9 .

0 . 6 0 . 2 1 0 . 1 0 . 0 . 7 1 . 3 0 8 7 . 8 2 1 0 1 . 3 1 3 0 0 6 7 2 . 8 1 . 0 . 3 0 . 2 0 0 . 2 0 . 9 . 0 7 . 8 8 3 2 . 0 9 5 1 . 7 3 0

0 2 . 4 0 . 8 5 2 2 1 . 4 2 . . 5

8 . . 9 0 . 3 9 2 . . 4 0 . 2 0 . 3 7

1 0 . 5 1 . 1 0 . 2 0 . 1 1 . 4 2 . 0 7 8 . 0 9 8 . 7 3 0 0 2 4 1 3 . 6 0 . 7 0 . 2 0 . 5 5 . 2 1 1 0 . 8 2 .

0 9 8 . 3 4 . 0 3 0 0 4 8 3 0 . 8 3 . 7 1 . 2 1 . 0 0 . 6 0 . 3 0 . 8 3 . 2 6 . 2 1 3 . 5 6 6 . 2 2 . 0 9 8 . 7 5 . 9 3 0 0 7 2 3 8 . 6 4 . 2 1 . 5 1 . 3 0 . 8 0 . 4 1 . 0 3 . 8 8 . 2

1 5 . 1 6 1 . 7 2 . 0 9 9 . 8 7 . 1 3 2 5 1 7 . 7 2 . 8 0 . 8 0 . 3 0 . 2 0 . 1 0 . 2 0 . 9 1 . 9 8 . 1 7 9 . 5 2 . 0 9 6 . 6 1 . 6 3 2 5 2 1 2 . 6 3 . 8 1 . 0 0 . 5 0 . 3 0 . 1 0

. 4 1 . 7 3 . 5 1 0 . 7 7 8 . 4 2 . 0 1 0 2 . 5 3 . 0 3 2 5 3 1 6 . 6 3 . 8 1 . 3 0 . 7 0 . 4 0 . 2 0 . 6 2 . 4 4 . 2 1 1 . 4 7 7 . 8 2 . 0 1 0 4 . 7 4 . 3 3 2 5 6 2 6 . 1 3 . 7 1 . 5

0 . 9 0 . 6 0 . 3 0 . 8 3 . 0 6 . 3 1 4 . 7 6 2 . 3 2 . 0 9 6 . 2 5 . 6 3 2 5 9 3 3 . 3 4 . 1 1 . 8 1 . 5 0 . 7 0 . 4 0 . 9 3 . 5 7 . 7 1 5 . 0 5 9 . 3 2 . 0 9 6 . 9 7 . 0 3 2 5 2 4 5

5 . 2 4 . 4 2 . 1 2 . 1 1 . 3 0 . 7 1 . 7 5 . 0 8 . 2 1 4 . 3 5 3 . 6 2 . 0 9 5 . 3 1 0 . 6 3 2 5 4 8 6 2 . 2 4 . 7 2 . 3 2 . 8 1 . 8 0 . 9 2 . 2 5 . 5 8 . 4 1 2 . 9 5 1 . 8 2 . 0 9 5 . 1

1 3 . 2 3 2 5 7 2 8 1 . 9 4 . 6 2 . 3 3 . 1 2 . 1 1 . 0 2 . 6 6 . 0 8 . 2 1 1 . 2 4 8 . 4 2 . 0 9 1 . 5 1 4 . 8 3 5 0 1 2 3 . 9 3 . 8 1 . 4 0 . 3 0 . 2 0 . 4 1 . 9 6 . 3

1 4 . 7 6 4 . 2 2 . 0 9 5 . 5 2 . 8 3 5 0 2 3 6 . 5 3 . 9 1 . 7 0 . 6 0 . 3 0 . 7 2 . 9 8 . 0 1 5 . 0 5 8 . 9 2 . 0 9 4 . 4 4 . 5 3 5 0 3 4 5 . 6 4 . 0 1 . 9 1 . 9 1

. 0 0 . 5 1 . 3 3 . 2 8 . 8 1 6 . 8 5 3 . 6 2 . 0 9 5 . 0 7 . 9 3 5 0 6 6 2 . 7 4 . 3 2 . 1 2 . 2 1 . 5 0 . 8 1 . 9 4 . 4 9 . 0 1 5 . 3 5 0 . 4 2 . 0 9 3 . 8 1 0 . 8 3 5 0 9 7 3 . 0 4 .

6 2 . 2 2 . 7 1 . 8 0 . 9 2 . 3 5 . 5 9 . 1 1 4 . 1 5 0 . 1 2 . 0 9 5 . 1 1 3 . 1 3 5 0 2 4 9 1 . 4 5 . 0 2 . 7 4 . 3 2 . 6 1 . 3 3 . 2 6 . 4 7 . 5 9 . 4 4 9 . 7 2 . 0 9 4 . 1 1 7 . 7 3

5 0 7 2 5 . 1 3 . 8 4 . 0 3 . 1 4 . 9 5 . 8 7 2 . 0 1 9 . 0

9 7 . 5 . 0 5 1 . 9 . 3 3 5 3 . 9 2 . 5

F .B e h ar e t a l . / Or g a n i c G e o c h e mi s t r y3 9 ( 2 0 0 8 ) 1 2 2
9

The generation curve of the total hydrocarbonyield also starts at the very early stages of kerogencracking (TR below 3 4%) with a linear trend upto a

kerogen TR of 80%, after which the generationrate seems to be faster. The maximum yield is 19%at total kerogen conversion. The relative distribu-

tion of the hydrocarbons generated above TR of 5% are reported inTable 7. There are no signicantchange s over the TR range from 5% to 2530%: theC

14+

aromatics predom inate with a relative pr opor-tion of 55 57%, the light and heavy saturates are insimilar propor tions with 9

10% and 11 13%,respectively ; and the hydrocarbon gases represent17 20%. Above kero gen TR at 30%, the relativepropo rtion of the C

14+

aromatics continu ouslydecreases from 5557% to 23%, leading to a relativeincrease of all other hydrocarbon classes. A steadystate is obs

erved for the gen eration yield of t heC

14+

aromatics at 325

C between 48 h and 72 hand at 350

C between 9 h and 24 h followed by aclear decrease at 350

C/72 h. For all other classes,no decrea se is observed in

yield except for theC

14+

saturates at 350

C/72 h. These observationsare i n good agreemen t with kinetic stu dies per-

formed on model compounds pyro lysed in goldtube s, such as

n

-C
25

(Behar and Vande nbroucke,1996) a nd dodecylbenze

ne (DDB) (Behar et al.,2002). A subset of conversions for these two compounds at 300, 325 and 350

C is inTable 8. Resultsshow that

-C
25

starts to decompose at 350

C/24 hwhereas 20% of the DDB is already converted.

Inthe most sever e conditions (35 0

C/72 h), almosth alf of the DDB is cracked.In conclusion, based on the dierences in

thermalreactivity between the kero gen, the two typ es of NSOs, and the total hydroca rbon fractions, it ispossible to p ropose a general kinetic scheme i nwhich:

using equivalent f requency factor, t he minimumactiv ation energy for kerogen cracking is lowerthan that in open system because of the early gen-eration of non-

hydrocarbon compounds (CO

2

,H
2

S, H
2

O), which is not taken into account whencalculating

bulk kinetics wit h the RockEvaltechnique. DCM NSOs are p rimary products fr om kerogencracki ng and have similar kinetic parameters (
E

and
A

) to kerogen because their secondary cracking largely overlaps that of kerogen.

solid residue

406080100

kerogen TR (%)kerogen TR (%)kerogen TR (%)kerogen TR (%)

y i e l d ( % ) y i e l d ( % ) y i e l d ( % ) y i e l d ( % )

DCM NSOs
05101520

n -pentane NSOs
0369120 0 4 0 0 8 0 0 0 0 0 4 0 0 8 0 0 0 0 0 4 0 0 8 0 0 0 0 0 4 0 2 6 1 2 6 1 2 6 1 2 6

0 0

8 0

Total HC
0510152025

Fig. 3. Evolution curves for the insoluble residue, the DCM and
n

-pentane NSOs and total hydrocarbon fraction during

articialmaturation of the Type II kerogen in a closed system with increasing transformation ratio (TR) determined from bulk kinetics in an opensystem.10
F. Behar et al. / Organic Geochemistry 39 (2008) 122