Organic Geochemistry 42 (2011) 228236

Contents lists available at ScienceDirect

Organic Geochemistry
journal homepage: www.elsevier.com/locate/orggeochem

Structural insights from boron tribromide ether cleavage into lignites and low
maturity coals from the New Zealand Coal Band
Clemens Glombitza , Kai Mangelsdorf, Brian Horseld
Helmholtz Centre Potsdam GFZ German Research Centre for Geosciences, 4.3 Organic Geochemistry, Telegrafenberg, 14473 Potsdam, Germany

a r t i c l e

i n f o

Article history:
Received 5 July 2010
Received in revised form 1 November 2010
Accepted 20 December 2010
Available online 24 December 2010

a b s t r a c t
Ether functionalities form an important cross-linking structure within the macromolecular organic
matrix of lignites and coals. To obtain a deeper insight into the complex internal structure of such macromolecules and the maturation related changes of the ether compounds within the network structure,
boron tribromide (BBr3) ether cleavage was applied to a series of lignite and coal samples of different
maturity (R0 0.270.80%) obtained from coal mines and natural outcrops from the North and South Island
of New Zealand. Terminal ether-bound alcohols rapidly decrease during diagenesis and occur only in low
amounts during the catagenetic stage. Comparison between ester- and ether-bound terminal alcohols
indicates a parallel decreasing trend during the diagenetic stage, suggesting that the stability differences
between both linkages are not large enough to be observed in maturation processes over geological time
scales. Polyether compounds were detected with chain length up to ve carbon atoms. After a small
decrease during the diagenetic phase these compounds occur in relatively high concentrations, even in
the main catagenetic stage. This suggests that these linkage structures represent important cross-linking
substructures within the macromolecular matrix of lignites and coals being sterically protected within
the macromolecular network during the maturation process. Additional cross-linking substructures were
(poly)ether aromatics, esters and ketones.
2010 Elsevier Ltd. All rights reserved.

1. Introduction
The alteration of sedimentary organic matter during the geological maturation process is characterised by the degradation of
reactive functional groups within the organic matrix. The most
abundant hetero atom in deposited organic matter is oxygen, forming carboxyl (esters and acids), carbonyl (ketones and aldehydes),
ether and hydroxyl functional groups (Petersen et al., 2008). Especially during diagenesis, the maturation related compositional
transformation of terrigenous organic matter (Type III kerogen) is
characterised by a loss of oxygen being indicated by a strong decrease of the atomic O/C ratio of the organic matter as visualised
in a van Krevelen diagram (van Krevelen, 1961).
During this geological alteration of buried organic matter structural reorganisation is observed, e.g. by the decrease of C@O signals
in IR spectroscopy (Robin, 1975; Lis et al., 2005; Petersen et al.,
2008) indicating the loss of carbonyl and carboxyl functional
groups. A parallel ongoing aromatisation can be observed from
NMR and IR spectroscopy by an increase of aromatic CH signals
(Robin, 1975; Hatcher et al., 1982, 1989; Suggate and Dickinson,
2004; Lis et al., 2005; Werner-Zwanziger et al., 2005). Additionally,
Corresponding author. Present address: University of Potsdam, Institute for
Earth and Environmental Sciences, Geomicrobiology Group, Karl-Liebknecht-Str.
24, Haus 27, 14476 Potsdam, Germany.
E-mail address: clemens.glombitza@geo.uni-potsdam.de (C. Glombitza).
0146-6380/$ - see front matter 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.orggeochem.2010.12.005

increasing aromatisation and the loss of heteroatom-bound aliphatic chains leading to higher material density and increasing p
electron mobility is indicated by an increase in vitrinite reectance
(Burnham and Sweeney, 1989; Larter, 1989; Carr, 1999).
To investigate changes of the structural composition within the
macromolecular organic matter during maturation selective chemical degradation procedures are appropriate tools used in organic
geochemistry. A broad suite of chemical reactions is well established for a selective cleavage of different bond types (Rullktter
and Michaelis, 1990; Richnow et al., 1992; Schaeffer et al., 1995;
Schaeffer-Reiss et al., 1998). Oxygen containing compound linkages
within the kerogen matrix are ester or ether bonds. In a stepwise
procedure ester cleavage has to be performed prior to ether cleavage due to the fact that ether cleavage methods will also attack ester functions.
Ester cleavage of a series of New Zealand lignites and subbituminous coals of different maturity (R0 0.270.80%) from ve sedimentary basins in New Zealand (Northland, Eastern Southland,
Waikato, West Coast and Taranaki basins) was already the focus of
two previous studies (Glombitza et al., 2009a,b). In these studies it
was shown that the New Zealand coals and lignites release signicant amounts of low molecular weight organic acids (formate, acetate and oxalate) as well as ester-bound high molecular weight
fatty acids and alcohols during maturation. The main phase of
degradation of these compounds was found to be associated with
diagenesis but also occurs during the early catagenesis. The amounts

C. Glombitza et al. / Organic Geochemistry 42 (2011) 228236

of released compounds decrease signicantly with ongoing maturation of the samples pointing to compositional alteration within the
macromolecular structure. Additionally, kinetic investigations on
aqueous hydrolysis being a favourable natural process to cleave ester-bound compounds within low maturity sediments showed that
the release of acids by hydrolysis is slowed with increasing maturation and that the required activation energies for the cleavage reaction increase with ongoing maturation (Glombitza, 2010)
conrming the structural alteration within the organic
macromolecule.
In the current study, ether cleavage was applied to the same sample set to investigate the changes of ether-bound compounds with
increasing maturation and to obtain a deeper insight into structural
properties of the investigated lignite and coal samples. For the cleavage of ether bonds several methods are available, such as the use of
HI (Hld et al., 1998; Putschew et al., 1998; Schaeffer-Reiss et al.,
1998), boron trichloride (BCl3) (Chappe et al., 1982; Richnow et al.,
1992), boron tribromide (BBr3) (Schwarzbauer et al., 2003; Wollenweber et al., 2006), boron triuoride (BF3) (Narayanan and Iyer,
1965; Ambls et al., 1996), triuoroacetic acid (Bhatt and Kulkarni,
1983) or trimethylsilyliodide (TMSI) (Michaelis et al., 1989). For this
study, we used the boron tribromide (BBr3) procedure because the
produced bromides are more stable than e.g. the iodides and can
be analysed by gas chromatographymass spectrometry (GCMS)
without a further reduction step.
2. Materials and methods
2.1. Sample material
Sample material for the current study came from a set of coal
samples gathered in 2002 from different coal mines and natural
outcrops from the North and South Island of New Zealand. The
sample set represented an almost continuous maturity series from
lignites to sub-bituminous coals covering a maturity range from
early diagenesis to the middle of the main catagenesis. The New
Zealand Coal Band represents one of the best coal series in the
world to investigate maturity related processes in terrigenous kerogen (Sykes and Snowdon, 2002; Sykes and Johansen, 2007). The
selected sample set covers a time period from Late Cretaceous to
Pleistocene age. Several investigations of this coal sample series
were previously reported. The kerogen composition determined
from open system pyrolysis and RockEval data were presented
in Vu et al. (2008), low molecular weight organic acids from water
extraction of these samples were reported in Vieth et al. (2008), the
molecular composition of the aliphatic bitumen fraction obtained
by solvent extraction is investigated in Vu et al. (2009), results of
alkaline ester cleavage in pre-extracted samples were reported in
Glombitza et al. (2009b) for the low molecular weight organic
acids (LMWOAs) and in Glombitza et al. (2009a) for the high
molecular weight organic acids (HMWOAs).
For this study, 12 lignite and coal samples of different maturity
ranging between 0.27% and 0.80% vitrinite reectance (R0) were selected from ve different basins. Two samples from the Northland
(G001986 and G001988) and two samples from the Eastern Southland Basin (G001976 and G001978) are in the lignite stage. Three
samples from the Waikato Basin (G001980, G001982 and
G001983), three samples from the West Coast Basin (G001993,
G01995 and G001996) and two samples from the Taranaki Basin
(G001991 and G001994) are sub-bituminous coals (Table 1).
2.2. Methods
2.2.1. Sample preparation
The dried and nely ground coal samples were extracted with
water for 1 day using Soxhlet extraction to remove the water solu-

229

ble compounds (Vieth et al., 2008). This procedure was followed by

a 3 day Soxhlet extraction using an azeotropic solvent mixture of
acetone (38%), methanol (30%) and chloroform (32%) and an additional extraction with pure MeOH for 1 day (removal of bitumen).
Subsequently, alkaline ester cleavage reaction was performed
(Glombitza et al., 2009a,b). The residue of this ester cleavage procedure was used for BBr3 ether cleavage that is described below.
2.2.2. Ether cleavage procedure
The ether cleavage reaction was applied as reported by Wollenweber et al. (2006) using the Lewis acid BBr3 as reagent. This
ether cleavage reaction leads to mono- and poly-brominated alkyl
and aryl compounds. The bromine position represents the previous
ether linkage to the macromolecular matrix (for details on the
mechanism see Benton and Dillon, 1942). Approximately 150 mg
of the coal sample was weighed into a 10 ml GC vial, equipped with
a magnetic stirrer. Five ml of a BBr3 solution (1 M in dichloromethane, Alfa Aescar GmbH & Co KG, Karlsruhe, Germany) were added
and the vial was sealed. The suspension was then placed in an ultrasonic bath for 2 h, stirred for 24 h at room temperature and nally
ultrasonicated for another 2 h. Finally, the vial was opened and
approximately 3 ml of diethyl ether were carefully added dropwise
while stirring until the excess of BBr3 was deactivated. The residue
was ltered using Whatman glass bre lters (pore size 0.7 lm)
and rinsed three times with 5 ml of diethyl ether. The combined organic phases were washed three times with water and dried with
Na2SO4. Finally, the solvent was reduced at ambient pressure to
approximately 2 ml.
2.2.3. Gas chromatographymass spectrometry (GCMS)
Prior to GCMS analysis 5 lg of 5a(Y)-androstane solved in
dichloromethane were added to each sample as injection standard
for quantication. The peak areas of all compounds identied were
quantied relatively to the peak area and known amount of the
injection standard. Response factors of the standard to the different
compound groups investigated were not determined. Finally, the
compound amount was related to the total organic carbon (TOC)
content of the respective sample. To determine the GC retention
times of the monobromo-alkanes a standard mixture of 1-bromo-dodecane, 1-bromo-tetradecane, 1-bromo-hexadecane and 1bromo-octadecane was analysed. GCMS analysis was performed
using a Thermo Trace GC Ultra (Thermo Fisher Scientic Inc., Waltham, MA, USA) coupled to a Thermo DSQ mass spectrometer.
Oven temperature was set to 50 C (1 min isothermal), heated at
3 C/min to 310 C and held isothermal for 30 min at 310 C. Samples were injected with a PTV splitless injector heated from 50 C
to 300 C at 10 C/s. The column used was a BPX5, length 50 m,
diameter 0.22 mm and 0.25 lm lm thickness. Helium was used
as the carrier gas. Mass spectra were obtained in the electron impact mode at 70 eV. The applied scan interval was 50650 amu
at 2.5 scans/s. The temperature of the ion source was set to
230 C. To identify mono-brominated alkanes the mass trace
m/z = 135 was used, being the mass of the characteristic pentacyclic fragment C4H8Br+.
3. Results and discussion
3.1. Qualitative evaluation of the BBr3 cleavage products from the New
Zealand coal series
The investigated extracts obtained from BBr3 ether cleavage
contained various amounts of previously ether-bound compounds
and some ester-bound compounds that appeared to have survived
the previously applied ester cleavage procedure (Table 1). The
ether cleavage products comprise several groups of compounds

230

C. Glombitza et al. / Organic Geochemistry 42 (2011) 228236

Table 1
Formations, coalelds, ages and sedimentary basins of the selected sample set and concentrations of monobromo-, dibromo-, tribromo- and tetrabromo-alkanes, brominated
esters, ketones, aromatics and phenols as well as ethyl esters after BBr3 ether cleavage experiments from NZ coal samples ranging from diagenesis to the main catagenesis
(R0 0.270.80%) stage. Data are presented in lg/g total organic carbon (TOC), R0: vitrinite reectance, NB: Northland Basin, ESB: Eastern Southland Basin, WB: Waikato Basin,
WCB: West Coast Basin, TB: Taranaki Basin, : not detected; small letters (ag) indicate the peaks labelled in the chromatogram of sample G001978 (Fig. 1).

R0 (%)
Basin
Formation

Coaleld
Age

G001986

G001988

0.27
NB
Unknown

0.27
NB
Unknown

G001976

G001982

G001983

G001980

G001996 G001995 G001994

G001993 G001991

0.29
ESB
Gore
lignite
measures
Kapuka

0.40
WB
Waikato
coal
measures
Rotowaro

0.41
WB
Waikato
coal
measures
Maramarura

0.44
WB
Waikato
coal
measures
Rotowaro

0.52
WCB
Brunner
coal
mesures
Reefton

0.76
WCB
Brunner
coal
mesures
Garvey
Creek
Eocene

0.80
TB
Mangahewa

Pleistocene Pleistocene Oligocene Oligocene Eocene

Eocene
Miocene
Miocene
Oligocene Oligocene

0.52
WCB
Brunner
coal
mesures
Reefton

0.61
TB
Rakopi

Eocene
Eocene
Oligocene

Eocene

Late
Cretaceous

5.09
13.15

Monobromo-alkanes (lg/gTOC)
C14
58.80
C16
10.45
C18
48.56
C20
32.73
C22
35.96
C24
37.95
C26
81.57
C28
131.37
C30
58.29

48.99
15.00
49.87
25.94
26.73
37.88
52.52
106.52
33.89

17.32
9.14
32.25
13.52
12.63
35.67
71.67
70.26
48.83

15.10
9.25
29.30
13.01
16.23
47.57
40.00
43.73
42.46

Total

397.34

311.29

256.65

22.79

10.84
31.08
25.87

20.18
14.20

183.49
166.08
5.43

132.80
116.94

67.79

389.38

177.88

249.74

495.68

Di-, tri- and tetra-bromoalkanes (lg/gTOC)

C3H5Br3

C3H4Br4

C4H8Br2 (g)

C5H9Br3a(c)
393.64
236.72
C5H9Br3a (c)
327.91
188.38
C5H8Br4

Total

721.55

425.10

22.79

92.57
85.31

28.70

11.00

Unknown

6.48
10.88

5.77
7.13

Unknown
Eocene

7.93

38.80

11.00

18.24

17.36

12.90

7.93

70.70
61.37
6.62

167.02
143.68

20.31
21.56

193.19
210.22

224.58
205.39

193.99
238.60

138.69

310.70

41.87

422.41

429.97

432.59

Aromates (lg/gTOC)
C6H(CH3)3Br2a (d)
C6H(CH3)3Br2a (d)

12.27
10.11

12.34
7.80

5.41
5.85

6.76
5.72

3.37
5.11

8.48
9.19

9.43
8.73

Total

22.38

20.14

11.26

12.48

8.48

17.66

18.16

15.87
8.20

Phenols (lg/gTOC)
C6H2Br3OH (e)

C6H2C(CH3)3Br2OH
(f)

C6H3Br2OH
Total

Total

Ethyl esters (lg/gTOC)

C24
C26
C28
C30

24.07

6.61

6.61

3.97
15.88
21.67

41.52

13.05

29.14
26.01
53.74
12.13

37.62
42.08
57.24
13.24

50.96
84.04
100.76
63.18

38.03
61.86
65.33
46.42

121.02

150.18

298.94

211.64

21.17
36.98

Brominated esters and ketones (lg/gTOC)

CO(CH2Br)2 (b)

CH2BrCO2C2H5 (a)
843.25
761.92
CH3CO2CH(CH3)
198.01
260.05
C2H4Br a
268.53
480.05
CH3CO2CH(CH3)
C2H4Br a
Total
1309.79
1502.02
a

G001978

0.28
ESB
Gore
lignite
measures
Sweetwater Sweetwater Waimumu

58.15

30.34
15.32

45.66

15.41

15.41

13.05

Regioisomers or diastereoisomers.

such as monobromo-alkanes, polybromo-alkanes, ethyl esters, brominated esters and ketones. In addition, some brominated aromatic and phenolic compounds could be detected (Table 1).
Fig. 1 shows an example chromatogram (Fig. 1B) of the products
obtained from sample G001978 (Eastern Southland Basin). The
mass trace m/z 135 was used to identify the mono-brominated

alkanes with aliphatic chains containing more than three carbon

atoms (Fig. 1C) and the mass trace m/z 88 was used to identify
the ethyl esters (Fig. 1D). For the identication of the aliphatic
monobromo-alkanes the chromatogram of a standard mixture
(containing n-C12, n-C14, n-C16 and n-C18 1-bromo-alkanes) is
shown (Fig. 1A).

C. Glombitza et al. / Organic Geochemistry 42 (2011) 228236

231

Fig. 1. (A) Total ion current (TIC) chromatogram of a monobromo-alkane standard mixture (n-C12, n-C14, n-C16, n-C18) and (B) TIC-chromatogram of BBr3 cleavage products
from sample G001978 (Eastern Southland Basin, R0 0.28%), (C) all mono-brominated aliphatic compounds with more than three carbon atoms (mass trace m/z 135) and (D)
ethyl esters (mass trace m/z 88). Numbers indicate the carbon atoms of the aliphatic chain; small letters indicate the following compounds: (a) CH2BrCOOC2H5, (b)
CO(CH2Br)2, (c) C5H9Br3, (d) C6H(CH3)3Br2, (e) C6H2Br3OH, (f) C6H2Br2C(CH3)3OH and (g) C4H8Br2.

3.1.1. Mono- and poly-brominated alkanes

Monobromo-alkanes were found to show quite identical mass
spectra. The signals of the different molecular ions were not
detectable. Therefore, to identify the monobromo-alkanes, a standard mixture of commercially available monobromo-alkanes with
12, 14, 16 and 18 carbon atoms (SigmaAldrich Chemie GmbH,
Buchs, Switzerland) was measured for comparison. Thus, the
monobromo-alkanes obtained from the lignite and coal samples
were identied according to their GC retention times. The carbon
numbers of the longer bromo-alkanes were assessed by their elution time within the homologue series.
The released monobromo-alkanes are the ether cleavage products of formerly ether bound n-alcohols. They range from n-C14
to n-C30 with a strong predominance of even carbon numbers (Table 1). This coincides with the ether cleavage product distributions
found in other studies using different ether cleavage procedures
(Richnow et al., 1992; Hld et al., 1998; Schaeffer-Reiss et al.,
1998). In other studies, applying the BBr3 procedure, the brominated alkanes were often described as bulk parameters due to their
indistinguishable mass spectra (e.g. Wollenweber et al., 2006). In
the Northland and Eastern Southland Basin samples, maxima in
the distribution of the monobromo-alkanes occur at n-C18 and
n-C28. The sub-bituminous coal samples from the Waikato, West
Coast and Taranaki basins contained lower numbers of monobromo-alkanes. Only the n-C18 bromo-alkane was found in almost
all of these samples (Table 1). Additionally, n-C16 bromo-alkane occurs in some samples of the West Coast and Taranaki basins.
The monobromo-alkanes were linked to the organic matter via
one ether bridge and, therefore, they represent terminal-bound

alcohols. These structural units are proposed to derive from land

plant waxes such as cutin and were also described by Wollenweber
et al. (2006) after BBr3 treatment of organic matter in Paleozoic
coals of different locations in Europe, Asia, South America and
Antarctica.
Additionally, several poly-brominated alkanes were found in almost all samples investigated. In contrast to the mono-brominated
alkanes these compounds have characteristic mass spectra. The
highest number of compounds was detected in the Eastern Southland lignite samples containing tribromo-propane (C3H5Br3, sample
G001976 only), tetrabromo-propane (C3H4Br4, sample G001976
only), dibromo-butane (C4H8Br2 (g), sample G001978 only), two
regioisomers of tribromo-pentane (C5H9Br3 (c)) and tetrabromopentane (C5H8Br4, sample G001976 only) (Table 1). The two isomers of tribromo-pentane showed the highest concentrations of
all poly-brominated alkanes. Only these two components were
found to be present in all samples. Wollenweber et al. (2006) also
described the nding of huge amounts of poly-brominated alkanes.
Additionally, they identied dibromo-alkanes with chain length of
1020 carbon atoms in some samples, which were not detected
in the New Zealand coals investigated in this study. In previous
studies of ligneous material, a,x-dicarboxylic acids were found in
products of ruthenium tetroxide oxidation of macromolecular organic material (Allard and Derenne, 2007; Valkova et al., 2009).
These acids might also derive from the oxidation of ether bridges
or the corresponding dialcohols by ruthenium tetroxide (Schouten
et al., 1998).
The poly-brominated compounds were identied by interpretation of their mass spectra and comparison with MS library data

232

C. Glombitza et al. / Organic Geochemistry 42 (2011) 228236

(National Institute of Standards and Technology, NIST). In addition,

the mass spectra of tetrabromo-propane and tetrabromo-pentane
were also compared with the spectra published in Wollenweber
et al. (2006). With respect to their mass spectra, several isomers
of the detected poly-brominated alkanes are possible due to the
fact that different regioisomers show similar fragmentation patterns. Comparison with mass spectra from NIST library suggests
for tribromo-propane a 1,1,2- or 1,2,2-conguration. The dibromo-butane is more likely a vicinale than a geminale dibromobutane due to the observed mass peak intensities of the NIST
library reference spectra. However, it has to be kept in mind that

for all these isomers only the mass spectra of some compounds
can be found in the MS library. The mass spectra of the polybrominated alkanes with exception of the two spectra of tetra-brominated alkanes, already published in Wollenweber et al. (2006),
are shown in Fig. 2.
The poly-brominated alkanes are suggested to form crosslinking structures within the macromolecular organic matter by
the formation of several ether bridges. Thus, they appear to represent structural units connecting aromatic substructures and forming the complex network structure of the organic matrix. However,
it has also be kept in mind that during the BBr3 treatment a possi-

Fig. 2. Mass spectra of (A) tribromo-propane, (B) dibromo-butane and (C) tribromo-pentane showing the characteristic bromine isotopic doublets (caused by 79Br and 81Br)
and triplets (caused by 79/79Br2, 79/81Br2 and 81/81Br2) for the mono-and di-brominated fragments. The molecular ion quadruplet for the tribromo-propane and -pentane was
not detected. m/z values are rounded.

C. Glombitza et al. / Organic Geochemistry 42 (2011) 228236

ble transformation of hydroxy groups to bromine substituents has

to be considered and, therefore, some of the bromine substituents
might be former hydroxy groups (Wollenweber et al., 2006). In
addition to these polyether bridges, it was shown in previous studies that in the ester fraction of the New Zealand coals (obtained
after saponication) dicarboxylic low (oxalic acid) and high molecular weight organic acids (n-C16 to n-C28) provide additional crosslinking structures forming important bridges in the macromolecular kerogen network (Glombitza et al., 2009b; Glombitza, 2010).
Various short chain cross-linking units such as dicarboxylic acids
(e.g. glycolic, malonic and succinic acid) and alcohols with two or
more hydroxy groups (e.g. 2-methyl-propane-1,2-diol, glycol)
were also detected by Valkova et al. (2009) after oxidative cleavage
reaction in humic acids and lignites. Their ndings support the
importance, high complexity and variability of short chain crosslinking subunits in terrestrial macromolecular organic matter.

3.1.2. Ethyl esters and (poly-)brominated aromatics, phenols, esters

and ketones
Two regioisomers of dibromo-trimethylbenzene (C6H(CH3)3Br2
(d)) were found in the Eastern Southland Basin lignites samples
and in one of the Waikato Basin coals (G001980) as well as in
the three West Coast Basin coals (G001993, G001995, G001996)
and in one of the Taranaki Basin samples (G001994) but they were
absent in all other samples investigated. The phenols 2,4,6-tribromo-phenol (C6H2Br3OH (e)) and 2,6-dibromo-4-tert-butyl-phenol (C6H2C(CH3)3Br2OH (f)) were only detected in one of the
Eastern Southland Basin lignites (G001978). 2,6-dibromo-phenol
(C6H3Br2OH) was only found in sample G001980 from the Waikato
Basin. All compounds were identied by interpretation of the mass
spectra and comparison with spectra from NIST MS library.
It might be speculated that the detected aromatic compounds
are most likely related to altered substructures of lignin although
they are not exactly matching the structural properties of lignin
dominantly consisting of phenylpropyl units. Wollenweber et al.
(2006) also reported the ndings of several aromatics and phenols
and correlated these compounds to the maceral composition. The
treatment of inertinite rich samples did not release these compounds or only in minor amounts. Some of these compounds were
released from vitrinite rich samples but the highest amounts were
obtained from liptinite rich samples. Because liptinite is not

233

predominately derived from woody tissue they concluded that a

primary lignin binding for these compounds cannot be assumed.
Nevertheless, these structural units appear to be fragments of
the macromolecular network. The former linkage positions are
indicated by the position of the bromine substituents. In Fig. 3A selected aromatic substructures detected in the samples are shown.
Ethyl esters with predominantly even carbon numbers in the
range of n-C24 to n-C30 were found in the lignites from the Northland and Eastern Southland basins as well as in sample G001980
from the Waikato Basin (with the exception of n-C30). They are related to terrigenous plant derived fatty acids, either linked to the
kerogen matrix via an ester bond or belonging to the bitumen fraction being trapped within the macromolecular structure. Although
all samples were previously treated by alkaline ester cleavage reaction as described in Glombitza et al. (2009a), these fatty acids obviously had remained within the macromolecular matrix. It can be
speculated that the cleavage of ether-bound compounds opened
smaller substructures within the kerogen matrix exposing prior
protected ester functionalities for the cleavage reaction or liberating trapped bitumen fatty acids. Valkova et al. (2009) also showed
that a notable proportion of free fatty acids can be trapped within
the macromolecular matrix and are liberated by the application of
chemical degradation reactions decomposing the organic macromolecular matrix. The BBr3 treatment of esters or free acids led
to highly reactive acid bromides. The following addition of diethyl
ether resulted nally in the formation of ethyl esters. This could
explain why no other fatty acid esters than the ethyl esters were
detected. Surprisingly, only long chain ethyl esters were detected.
A reason why there are no short chain ethyl esters (e.g. n-C16,
n-C18) is puzzling, especially, because the short chain fatty acids
are the dominating fatty acids in the fatty acid distribution in the
kerogen-bound ester fraction (Glombitza et al., 2009a). In another
study it was shown that in the bitumen fraction of these samples
the short chain fatty acids are less dominant (Glombitza, 2010).
However, if the ethyl esters represent bitumen fatty acids being released from some inner inclusions in the kerogen matrix after
ether cleavage, at least a small proportion of short chain fatty acids
can be expected. Thus, it is not clear which proportion of the organic matrix is represented by the ethyl esters. However, these results
might point to some differences between the outer and inner kerogen matrix becoming accessible during the stepwise degradation
procedures when the macromolecular structure is degraded to a

Fig. 3. Selected structures of detected brominated aromatics, phenols, ethyl esters and ketones from the New Zealand coal samples forming substructures within the
macromolecular organic matter. The position of bromine reveals the former ether linkage position, the ethyl esters are suggested to derive from former ester linkage. The
small letters (ag) indicate the peaks labelled in the chromatogram of sample G001978 (Fig. 1).

234

C. Glombitza et al. / Organic Geochemistry 42 (2011) 228236

certain degree. The fatty acid signature of the inner kerogen matrix
appears to resemble a more pristine plant derived signal. In contrast, the outer signature might be mixed with organic matter from
other sources (such as microbial biomass) containing a high proportion of short chain (mainly n-C16 and n-C18) fatty acids
(Glombitza et al., 2009a).
The ketone 1,3-dibromo-propanone (CO(CH2Br)2 (b)) was only
detected in one of the Eastern Southland lignite samples
(G001978). The bromo-acetic acid ethyl ester (CH2BrCO2C2H5 (a))
was found in all samples of the Northland, Eastern Southland
and Waikato Basins except in sample G001982 (Waikato Basin).
This compound most likely derives from glycolic acid linked via
an ester bond to the kerogen matrix (or being trapped within the
matrix), resulting in an ethyl ester during the BBr3 cleavage procedure (see above), while the former hydroxy group previously may
have been linked via an ether bond to the macromolecular matrix.
Furthermore, two stereoisomers of acetic acid 2-bromo-1-methylpropyl ester (CH3CO2CH(CH3)C2H4Br) were found in the lignites
from the Northland Basin and in low amount in one of the Eastern
Southland Basin samples (G001976). These compounds (except the
acetic acid 2-bromo-1-methylpropyl ester) are also most likely
cross-linking structures within the kerogen matrix. The structures
are shown in Fig. 3B. The variability in the occurrence of the before
mentioned compounds might either be related to facies differences
or to the increasing maturity of the samples.
The ndings of poly-brominated alkanes, aromatics, phenols as
well as the brominated esters and ketones reveal a deep insight
into the network structure of the organic macromolecule. Linkages
of structural subunits in the kerogen by oxygen functionalities are
characterised by various ether bridges of short chain length bearing sometimes further oxygen functionalities such as ketones.
Thus, together with some low and high molecular weight organic
dicarboxylic acids, especially oxalic acid (Glombitza et al.,
2009b), ether lipids form an important cross-linking bridge within
the kerogen network of the New Zealand coals.

3.2. Quantitative evaluation of the BBr3 cleavage products from the

New Zealand coal series
3.2.1. Mono- and poly-brominated alkanes
The highest concentrations of monobromo-alkanes were found
in the immature Northland and Eastern Southland Basin samples. A
strong decrease of these compounds is observed during early diagenesis. At the end of the diagenesis and in the early as well as
main catagenesis stage samples only small concentrations are
detectable (Fig. 4A). Although facies differences have to be considered for the immature samples, the data indicate a rapid degradation of terminal ether-bound alcohols during diagenesis.
The maturation related decrease of the ether-bound alcohols
(Fig. 4A) is comparable to the observed decrease of the ester-bound
alcohols (Glombitza et al., 2009a). This result is surprising because
esters are thought to be less stable to thermal decomposition than
ethers. The thermal decomposition of ethers proceeds via either a
four centre transition state or a free radical chain decomposition
which should require more energy than the decomposition of esters by a quasi six-membered ring transition state (Blake et al.,
1961). Despite that, the observed differences in decomposition
temperatures for various esters and ethers are small. Blake et al.
(1961) measured ranges of approximately 270350 C for esters
and 200450 C for ethers depending on different substituents.
Therefore, it may be speculated that the difference in stability between ester and ether bonds, at least for the terminal bound compounds, is not large enough to be observed in a maturation series
representing alterations within geological time spans.
Interestingly, the poly-brominated alkanes (mainly tribromopentanes (c)) do not show the same maturity related decrease as
observed for the monobromo-alkanes (Fig. 4B). Highest amounts
were also found in the Northland and the Eastern Southland Basin
samples with decreasing concentrations during early diagenesis.
With exception of two samples (G001978 and G001995), showing
extremely low values, a small minimum at still comparably high

Fig. 4. Concentrations of BBr3 ether cleavage products of New Zealand lignite and coal samples of different maturity, grouped into mono-brominated alkanes (A), polybrominated alkanes (B), ethyl esters (C) as well as brominated esters and ketones (D). Note different y-axis scales.

C. Glombitza et al. / Organic Geochemistry 42 (2011) 228236

concentrations in the distribution of poly-brominated alkanes is

observed during the late diagenesis to early catagenesis samples
from the Waikato Basin. In contrast to the monobromo-alkanes,
the poly-brominated alkanes occur in relatively high concentrations in the catagenesis samples from the West Coast and Taranaki
basins. Their distributions remain relatively constant or might
even slightly increase. This indicates that after the early diagenesis,
where facies related differences play a more important role
(Glombitza et al., 2009a), the concentrations of these compounds
are not decreasing with increasing maturation level (at least in
the range investigated here). Due to their suggested function as
linkage structures within the macromolecular network it can be
speculated that they are mainly located within the network concomitantly being sterically protected by the organic matrix and,
therefore, are less effected by alteration processes than the terminal ether-bound aliphatic moieties.
As mentioned above, long chain ether bridges were not detected
in the New Zealand samples. In contrast, in a former study ester
bridges formed by long chain dicarboxylic acids could be found,
however, only in the early diagenetic stage samples of the Northland and the Eastern Southland basins and in comparably smaller
amounts (Glombitza et al., 2009a; Glombitza, 2010) than the short
chain ether bridges described above. A short chain dicarboxylic
acid (oxalic acid) was also detected and decreases rapidly during
diagenesis and early catagenesis but was still present in signicant
amounts at higher maturities (Glombitza et al., 2009b). This suggests that within the organic macromolecular matrix of the New
Zealand coals bridges formed by short chain ether and ester bonds
are considerably more important and stable than those formed by
long chain compounds, which are either absent or seem to decrease rapidly during early diagenesis.
3.2.2. Ethyl esters and (poly)brominated aromatics, phenols, esters and
ketones
Poly-brominated aromatics and phenols identied from the
samples do not show any relation to maturation (Table 1). They
are probably internal aromatic substructures of the macromolecular organic matter and, therefore, they might be more facies dependent and less effected by maturation processes.
The ethyl esters, formed by long chain fatty acids, only show
signicant amounts in the diagenetic stage samples (Fig. 4C) and
they are absent in the catagenesis samples. Therefore, they show
a different pattern with ongoing maturation than the much slower
decreasing kerogen bound fatty acids obtained after saponication
(Glombitza et al., 2009a). Together with the fact that no short chain
ethyl fatty acids were detected (as discussed above) the origin of
these ethyl esters remains unclear. They might represent an inner
bitumen fatty acid fraction trapped within the kerogen structure
being strongly dominated by a plant signature.
The highest amounts of bromo-acetic acid ethyl ester
(CH2BrCO2C2H5 (a)) were found in the lignites from the Northland
Basin with approximately 20 times higher concentration than in
the Eastern Southland Basin samples. The Waikato Basin samples
released approximately half of the concentrations obtained from
the Eastern Southland Basin samples (Table 1). The huge difference
between the Northland and Eastern Southland samples is most
likely facies related rather than the result of thermal maturation
because the difference in thermal maturation between the Northland (R0 0.27%) and the Eastern Southland Basin samples
(R0 0.280.29%) is only small. On the other hand, the fact that this
compound was absent in the West Coast and Taranaki Basin samples indicates that its degradation appears also during diagenesis
to early catagenesis (Fig. 4D). These ndings combined with the
fact that e.g. acetic acid 2-bromo-1-methyl-propyl ester
(CH3CO2CH(CH3)C2H4Br) were only found in the Northland Basin
samples in signicant concentrations and 1,3-dibromo-propanone

235

(CO(CH2Br)2 (b)) only in one Eastern Southland Basin sample suggest that the linkages structures highly depend on the chemical
composition of the contributing source material and that these
compounds rapidly degrade during diagenesis.
4. Conclusion
The BBr3 cleavage of kerogen ether-bound compounds in lignite
and sub-bituminous coal samples from the Northland, Eastern
Southland, Waikato, West Coast and Taranaki basins of New Zealand
resulted in a complex set of ether cleavage products such as monoand poly-brominated alkanes, brominated aromatic compounds,
ethyl esters and brominated esters and ketones.
The lignite samples of the Northland and Eastern Southland
basins showed signicantly higher contents of mono-brominated
alkanes than the sub-bituminous coal samples from the Waikato,
West Coast and Taranaki basins. The data show a rapid decrease
of the mono-brominated alkanes during the diagenetic stage, indicating the major loss of terminal ether-bound alcohols during this
phase. Although ether bonds are known to be more stable than ester bonds, the main decomposition of ether- and ester-bound alcohols can be observed in the same range during diagenesis. Thus, it
is speculated that the stability differences between both terminal
linkages are not large enough to be observed in maturation processes over geological times.
In contrast, short chain poly-brominated alkanes (mainly n-C5)
were also found in signicant amounts during the catagenetic
stage, suggesting that these compounds represent important
cross-linking bridges within the kerogen matrix. Thus, together
with short chain ester moieties such as oxalic acid (Glombitza
et al., 2009a) these short chain ether link-bridges appear to be of
high relevance for the linkage of substructures within the macromolecular network.
Additionally, poly-brominated aromatic compounds, brominated esters and ketones also seem to form substructures or
bridges in the macromolecular network. However, the maturity related distribution of brominated esters and ketones suggests a
higher dependence to the contributing plant facies and a higher
sensitivity to maturation related degradation compared to the
poly-brominated alkanes and aromatic compounds.
Acknowledgements
We thank Jan Schwarzbauer (RWTH Aachen, Germany) for sharing knowledge about BBr3 ether cleavage procedures. Hans Peter
Nytoft and an anonymous reviewer are thanked for their constructive and helpful suggestions. Cornelia Karger (GFZ Potsdam,
Germany) is thanked for GC-MS measurement and Richard Sykes
(GNS Science, New Zealand) for providing background information
and parameters on the sample material. Finally, we gratefully
acknowledge the Helmholtz Association for the nancial support
of this study.

Associate EditorGeoff Abbott

References
Allard, B., Derenne, S., 2007. Oxidation of humic acids from an agricultural soil and a
lignite deposit: analysis of lipophilic and hydrophilic products. Organic
Geochemistry 38, 20362057.
Ambls, A., Grasset, L., Dupas, G., Jacquesy, J.C., 1996. Ester- and ether bond cleavage
in immature kerogens. In: 17th International Meeting on Organic
Geochemistry. Pergamon-Elsevier Science Ltd., Donostia-San Sebastian, Spain,
pp. 681690.
Benton, F.L., Dillon, T.E., 1942. The cleavage of ethers with boron tribromide. I. some
common ethers. Journal of the America Chemistry Society 64, 11281129.
Bhatt, M.V., Kulkarni, S.U., 1983. Cleavage of ethers. Synthesis 249, 282.

236

C. Glombitza et al. / Organic Geochemistry 42 (2011) 228236

Blake, E.S., Hammann, W.C., Edwards, J.W., Reichard, T.E., Ort, M.R., 1961. Thermal
stability as a function of chemical structure. Journal of Chemical & Engineering
Data 6, 8798.
Burnham, A.K., Sweeney, J.J., 1989. A chemical kinetic model of vitrinite maturation
and reectance. Geochimica et Cosmochimica Acta 53, 26492657.
Carr, A.D., 1999. A vitrinite reectance kinetic model incorporating overpressure
retardation. Marine and Petroleum Geology 16, 355377.
Chappe, B., Albrecht, P., Michaelis, W., 1982. Polar lipids of archaebacteria in
sediments and petroleums. Science 217, 6566.
Glombitza, C., 2010. New Zealand Coals A Potential Feedstock for Deep Microbial
Life. PhD thesis, Technische Universitt, Berlin.
Glombitza, C., Mangelsdorf, K., Horseld, B., 2009a. Maturation related changes in
the distribution of ester bound fatty acids and alcohols in a coal series from the
New Zealand Coal Band covering diagenetic to catagenetic coalication levels.
Organic Geochemistry 40, 10631073.
Glombitza, C., Mangelsdorf, K., Horseld, B., 2009b. A novel procedure to detect low
molecular weight compounds released by alkaline ester cleavage from low
maturity coals to assess its feedstock potential for deep microbial life. Organic
Geochemistry 40, 175183.
Hatcher, P.G., Breger, I.A., Szeverenyi, N., Maciel, G.E., 1982. Nuclear magnetic
resonance studies of ancient buried wood II. Observations on the origin of coal
from lignite to bituminous coal. Organic Geochemistry 4, 918.
Hatcher, P.G., Lerch III, H.E., Bates, A.L., Verheyen, T.V., 1989. Solid-state 13C nuclear
magnetic resonance studies of coalied gymnosperm xylem tissue from
Australian brown coals. Organic Geochemistry 14, 145155.
Hld, I.M., Brussee, N.J., Schouten, S., Sinninghe Damst, J.S., 1998. Changes in the
molecular structure of a Type II-S kerogen (Monterey Formation USA) during
sequential chemical degradation. Organic Geochemistry 29, 14031417.
Larter, S., 1989. Chemical models of vitrinite reectance evolution. Geologische
Rundschau 78, 349359.
Lis, G.P., Mastalerz, M., Schimmelmann, A., Lewan, M.D., Stankiewicz, B.A., 2005.
FTIR absorption indices for thermal maturity in comparison with vitrinite
reectance R0 in type-II kerogens from Devonian black shales. Organic
Geochemistry 36, 15331552.
Michaelis, W., Richnow, H.H., Jenisch, A., 1989. Structural studies of marine and
riverine humic matter by chemical degradation. The Science of the Total
Environment 81 (82), 4150.
Narayanan, C.R., Iyer, K.N., 1965. Regeneration of steroid alcohols from their methyl
ethers. Journal of Organic Chemistry 30, 17341736.
Petersen, H.I., Rosenberg, P., Nytoft, H.P., 2008. Oxygen groups in coals and alginiterich kerogen revisited. International Journal of Coal Geology 74, 93113.
Putschew, A., Schaeffer-Reiss, C., Schaeffer, P., Koopmans, M.P., de Leeuw, J.W.,
Lewan, M.D., Sinninghe Damst, J.S., Maxwell, J.R., 1998. Release of sulfur- and
oxygen-bound components from a sulfur-rich kerogen during simulated
maturation by hydrous pyrolysis. Organic Geochemistry 29, 18751890.
Richnow, H.H., Jenisch, A., Michaelis, W., 1992. Structural investigations of sulphurrich macromolecular oil fractions and a kerogen by sequential chemical
degradation. Organic Geochemistry 19, 351370.

Robin, P.L., 1975. Carractrisation des krognes et de leur volution par

spectroscopie infrarouge. PhD thesis, Universit de Louvain, Louvain.
Rullktter, J., Michaelis, W., 1990. The structure of kerogen and related materials. A
review of recent progress and future trends. Organic Geochemistry 16, 829
852.
Schaeffer, P., Harrison, W.N., Keely, B.J., Maxwell, J.R., 1995. Product distributions
from chemical degradation of kerogens from a marl from a Miocene evaporitic
sequence (Vena del Gesso, N. Italy). Organic Geochemistry 23, 541554.
Schaeffer-Reiss, C., Schaeffer, P., Putschew, A., Maxwell, J.R., 1998. Stepwise
chemical degradation of immature S-rich kerogens from Vena del Gesso
(Italy). Organic Geochemistry 29, 18571873.
Schouten, S., Moerkerken, P., Gelin, F., Baas, M., de Leeuw, J.W., Sinninghe Damst,
J.S., 1998. Structural characterization of aliphatic, non-hydrolyzable
biopolymers in freshwater algae and a leaf cuticle using ruthenium tetroxide
degradation. Phytochemistry 49, 987993.
Schwarzbauer, J., Ricking, M., Littke, R., 2003. Quantitation of nonextractable
anthropogenic quantitation of nonextractable anthropogenic sediments after
chemical degradation. Acta Hydrochimica et Hydrobiologica 31, 469481.
Suggate, R.P., Dickinson, W.W., 2004. Carbon NMR of coals: the effects of coal type
and rank. International Journal of Coal Geology 57, 122.
Sykes, R., Johansen, P.E., 2007. Maturation characteristics of the New Zealand Coal
Band: part 1 evolution of oil and gas products. In: Poster: The 23rd
International Meeting on Organic Geochemistry, vol. 9 no. 14, p. 571.
Sykes, R., Snowdon, L.R., 2002. Guidelines for assessing the petroleum potential of
coaly source rocks using Rock-Eval pyrolysis. Organic Geochemistry 33, 1441
1455.
Valkova, D., Grasset, L., Ambls, A., 2009. Molecular compounds generated by
ruthenium tetroxide oxidation and preparative off line thermochemolysis of
lignite humic acids from South Moravia: implications for molecular structure.
Fuel 88, 21132121.
van Krevelen, D.W., 1961. Coal: Typology Chemistry Physics Constitution.
Elsevier, The Netherlands.
Vieth, A., Mangelsdorf, K., Sykes, R., Horseld, B., 2008. Water extraction of coals
potential for estimating low molecular weight organic acids as carbon feedstock
for the deep terrestrial biosphere. Organic Geochemistry 39, 985991.
Vu, T.A.T., Horseld, B., Sykes, R., 2008. Inuence of in-situ bitumen on the
generation of gas and oil in New Zealand coals. Organic Geochemistry 39, 1606
1619.
Vu, T.T.A., Zink, K.-G., Mangelsdorf, K., Sykes, R., Wilkes, H., Horseld, B., 2009.
Changes in bulk properties and molecular compositions within New Zealand
Coal Band solvent extracts from early diagenetic to catagenetic maturity levels.
Organic Geochemistry 40, 963977.
Werner-Zwanziger, U., Lis, G., Mastalerz, M., Schimmelmann, A., 2005. Thermal
maturity of type II kerogen from the New Albany Shale assessed by 13C CP/MAS
NMR. Solid State Nuclear Magnetic Resonance 27, 140148.
Wollenweber, J., Schwarzbauer, J., Littke, R., Wilkes, H., Armstroff, A., Horseld, B., 2006.
Characterisation of non-extractable macromolecular organic matter in Palaeozoic
coals. Palaeogeography, Palaeoclimatology, Palaeoecology 240, 275304.