Incorporation of steroidal biomarkers into petroleum model compounds

Article abstract of DOI:10.1002/poc.2931

A series of fully-substituted 5,6-benzoquinolines covalently fused to the chlolestane framework has been synthesized to serve as model compounds for the poorly characterized components found in the heaviest fractions of petroleum, namely the asphaltenes.

Full text of DOI:10.1002/poc.2931

Research Article
Received: 16 December 2011,
Revised: 31 January 2012,
Accepted: 1 February 2012,
Published online in Wiley Online Library: 2012
(wileyonlinelibrary.com) DOI: 10.1002/poc.2931
Incorporation of steroidal biomarkers into
petroleum model compounds
Alexander Scherera, Frank Hampela, Murray R. Grayb, Jeffrey M. Strykerc
and Rik R. Tykwinskia*
A series of fully-substituted 5,6-benzoquinolines covalently fused to the chlolestane framework has been synthesized
to serve as model compounds for the poorly characterized components found in the heaviest fractions of petroleum,
namely the asphaltenes. Acid-catalyzed cyclocondensation of an aromatic imine with 5-a-cholestan-3-one gives
5,6-benzoquinoline cycloadducts incorporating the steroidal biomarker fused to the benzoquinoline system at the
3,4-positions, with various pendent substituents in the 2-position. X-ray crystallography of three such derivatives
shows that the solid-state geometries of these molecules are quite similar, and packing in the solid state appears
to be dominated to a great extent by the nature of the pendent substituent. X-ray crystallography and solution phase
circular dichroism spectroscopy together suggest that the conjugated benzoquinoline moiety adopts a helical
conformation. Copyright © 2012 John Wiley & Sons, Ltd.
Supporting information may be found in the online version of this paper.
Keywords: asphaltenes; benzoquinolines ; chirality ; petroleum biomarkers ; solid state structure ; steroid derivatives
Given the incredible diversity of structures present in a natural
sample, however, global analyses can be difficult to generalize,
INTRODUCTION
because the analytical results may not correlate with the consti-
tution of the bulk sample. We[12–19] and others[20–23] have opted
for an alternative approach to study asphaltene behavior,
particularly with respect to aggregation phenomena. Rather than
the top–down approach to analysis, we consider instead a
‘bottom–up’ approach, which is accomplished through the syn-
thesis and study of complex archipelago model compounds of
rigorously defined structures.
Steroids are prominent among the biomarker components of
petroleum.[24–27] A biomarker is defined as a complex organic
molecule that retains a distinct structural echo of its progenitor
organism.[28,29] Model biomarker compounds have been synthe-
sized, but only relatively simple systems and substitution
patterns have as yet been investigated.[30–39] Larger biomarker
molecules consisting of up to seven fused rings, including pyri-
dyl and quinolinyl systems, have recently been detected in crude
oils; Oldenburg et al.[40] suggest that such large molecules are a
natural component of petroleum.
The increasing global dependence on fossil fuels combined with
decreasing supply is currently one of the most challenging
dilemmas facing future generations. To meet future energy
needs, the use of both renewable energy sources and lower-
grade petroleum sources will be required. Heavy crude oils and
bitumen deposits have a higher content of large, high boiling
components and heptane-insoluble asphaltenes. Strong molecu-
lar aggregation is a serious issue that complicates the processing
of these large petroleum-derived molecules.[1–3] Many of the
problems arising from bitumen aggregation can be addressed
through an improved understanding of the specific molecular
structure of heavy petroleum components. One of the chal-
lenges surrounding the study of crude oil at the molecular level
is, however, the extraordinary diversity of the structures that are
present.[4] For example, the asphaltenes comprise a range of
functional groups, including alkyl rings and chains, aromatic
and heteroaromatic rings, carboxylic acids, basic nitrogen
groups, and porphyrins.[5–7] The exact assembly of these various
fragments into an idealized asphaltene molecule, however,
remains a topic of considerable debate. For almost a century,
the gross structure of an asphaltene molecule was thought to
consist of a large, unsaturated polycyclic aromatic hydrocarbon,
decorated with saturated hydrocarbon groups appended to the
central core. This large ‘island’ motif is often referred to as the
continental model for asphaltene components.[8] Alternatively,
the archipelago model posits that asphaltene molecules are
made up of smaller islands of fused ring systems, both unsatu-
rated and partially saturated, linked by variable-length alkyl
and alkanethiol bridges.[3,7]
* Correspondence to: Rik R. Tykwinski, Department of Chemistry and Pharmacy
and Interdisciplinary Center of Molecular Materials (ICMM), University of
Erlangen-Nuremberg, Henkestrasse 42, 91054 Erlangen, Germany .
E-mail: rik.tykwinski@chemie.uni-erlangen.de
a A. Scherer, F. Hampel, R. R. Tykwinski
Department of Chemistry and Pharmacy and Interdisciplinary Center of
Molecular Materials (ICMM), University of Erlangen-Nuremberg, Henkestrasse
42, 91054 Erlangen, Germany
b M. R. Gray
Department of Chemical and Materials Engineering, University of Alberta,
Edmonton, AB, T6G 2V4 Canada
The continental/archipelago debate continues, in part, be-
cause a ‘top–down’ approach has been traditionally used for
asphaltene characterization, in which naturally occurring asphal-
tene samples are analyzed, degraded, and/or separated.[9–11]
c J. M. Stryker
Department of Chemistry, University of Alberta, Edmonton, AB, T6G 2G2 Canada
J. Phys. Org. Chem. (2012)
Copyright © 2012 John Wiley & Sons, Ltd.

A. SCHERER ET AL.
cholestan-3-one (3), using the mild, iodine-catalyzed conditions
recently described by Wang et al.[42] Unfortunately, the crude
product from this multicomponent reaction was quite complex;
none of the desired product 5a could be identified in the mix-
ture. Thus, this route was abandoned.
Alternatively, Kozlov et al. have extensively explored an
analogous, but stepwise, procedure for benzoquinoline forma-
tion.[43,44] This route relies on the initial isolation of a Schiff base
4 (Scheme 2). In the present case, the requisite Schiff bases 4a–f
were easily obtained by condensation of 2-naphthalenamine 1
with commercially available aromatic aldehydes 2a–f in toluene
at reflux, using a Dean–Stark trap to ensure complete conversion.
The crude products were crystallized from EtOH to afford imines
4a–f in moderate to good yields (Table 1).
H
H
H
H
H
Ph
H
H
I , THF
2X
O
+
Ph
2a
3
O
NH
N
2
5a
1
Whereas aldehydes 2a–f are commercially available, the two
more elaborate archipelago precursors 2g–h required synthesis.
Aldehydes 2g–h were obtained in reasonable yield from a three-
step reaction sequence starting with 4-bromobenzaldehyde
(Scheme 3). Wittig olefination using the ylides derived from 6a
and 6b gave the corresponding vinyl compounds, which were
reduced to give 7a and 7b, respectively. Lithium–halogen
exchange using n-BuLi, followed by the addition of dimethylform-
amide then gave aldehydes 2g and 2h. Although 2g was readily
purified by column chromatography, aldehyde 2h was consis-
tently produced in ~70% yield and isolated in impure form
(~70%). This material could not be further purified by flash column
chromatography and was used as is. Fortunately, however, the
impurities do not affect the conversion to the imine 4h, which
could then be purified easily. Thus, 2g and 2h were converted to
4g and 4h, respectively, using the same protocol as described for
4a–f (Scheme 2).
With imines 4a–h in hand, subsequent reaction with cholesta-
none 3 and a catalytic amount of HCl in boiling EtOH gave
cycloadducts 5a–h (Scheme 2, Table 1); monitoring the reaction
progress by thin layer chromatography indicated consumption
of the imine was complete after ~16 h. Following workup, all of
these reactions afforded crude products 5a–h as impure red–
brown oils clearly containing a range of byproducts. This distinct
lack of selectivity is likely due to the harshly acidic conditions
Scheme 1. Unsuccessful synthesis of compound 5a using a one pot,
multicomponent reaction, as reported by Wang et al.[42]
Petroleum biomarkers play an important role in petroleum
chemistry because they are generally stable to biodegradation,
distillation, thermal degradation, and catalytic reactions under
processing. In a recent study, for example, the thermal behavior
of biomarker model compounds was detailed, demonstrating
clearly that the investigation of petroleum chemistry at the
molecular level is indeed quite valuable.[41] Thus, in the present
study, we report the synthesis of a series of chimeric asphaltene
model compounds constructed from the fusion of an aromatic
benzoquinoline to a cholestane-derived steroidal biomarker.
The incorporation of side chains through this reaction sequence
ultimately provides archipelago-like structures. Preliminary char-
acterization of the chemical properties of these complex chiral
molecules as a function of molecular structure is also presented.
RESULTS AND DISCUSSION
The most direct route to covalent fusion products, such as 5a
(Scheme 1), appeared be to be a one-pot three-component
reaction of 2-naphthalenamine (1), benzaldehyde (2a), and 5-a-
NH2
N
R'
O
R'
a)
+
H
R
4a–h
2a–h
1
H
H
R
b)
H
H
H
H
3
R =
O
R'
N
5a–h
Scheme 2. Syntheses of Schiff bases 4a–h and benzoquinolines 5a–h. Reagents and conditions: (a) Dean–Stark trap, PhMe, reflux, 16 h. (b) HCl, EtOH,
reflux, 16 h
wileyonlinelibrary.com/journal/poc
Copyright © 2012 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. (2012)

INCORPORATION OF STEROIDAL BIOMARKERS INTO PETROLEUM MODEL COMPOUNDS
R
R
CHO
R
a), b)
c)
Cl– Ph3+P
+
Br
OHC
Br
6a R = 1-naphthyl
6b R = Ph
7a R = 1-naphthyl (67%)
7b R = Ph (80%)
2g R = 1-naphthyl (48%)
2h R = Ph (ca. 50%)
Scheme 3. Synthesis of 2g and 2h. Reagents and conditions: (a) LiOH•H2O, i-PrOH, reflux, 16 h; (b) H2 (1200 psi), Pd/C, PhMe, r.t., 48 h. (c) n-BuLi, DMF,
THF, –78 ꢀC to r.t., 16 h
5a–h directly via column chromatography was not possible.
Table 1. Isolated yields of 4a–h and 5a–h
The product-containing oil was thus extracted with hexanes,
filtered, and the hexanes solution saturated with gaseous
Entry
R’
4 yield (%) 5 yield (%)
HCl. This affords the corresponding hydrochloride salts as
an alkane-insoluble precipitate, whereas most of the impuri-
ties remain in solution. The resulting salts were washed
extensively with hexanes to remove the remaining cholesta-
none 3, dissolved in CH2Cl2, basified with NaOH, and the
crude material so obtained was then further purified by col-
umn chromatography, giving biomarker models 5a–h as
spectroscopically pure solids in poor to moderate yields
(10–39%).
a
69
27
b
47
22
nBu
c
64
38
35
27
S
1
Characterization of benzoquinolines 5a–c, f–h by H and 13 C
d
NMR spectroscopy is straightforward; in each case, only one
product isomer is observed. Thus, the enolization of cholesta-
none 3 proceeds in a highly regioselective manner, providing
the ‘parallel’ tautomer 8 exclusively (Figure 1). This is a conse-
quence of the differential torsional strain experienced by the
ring system because it distorts to the lowest energy half-chain
enol conformation.[47] This result is consistent with the observa-
tions of Inhoffen et al.[48] who reported that the conversion of
cholestanone 3 to the corresponding enol ether gives only the
isomer analogous to 8. The 1H NMR spectra for compounds
5a–h each show a characteristic set of doublets in the range of
2–3 ppm, which can be tentatively assigned to Ha and Ha′ of
e
f
67
63
61
73
39
22
10
11
1
isomer 5 (Figure 1). For example, the H NMR spectrum of 5b
features these doublets at 2.95 and 2.35 ppm (2JHH = 17.2 Hz). A
strong correlation is observed between the two doublets in the
correlation spectroscopy (COSY) spectrum, but no other correla-
tions are found for either signal. An heteronuclear single quan-
tum coherence (HSQC) experiment reveals a correlation for each
of the protons to the same secondary carbon atom at 43.4 ppm.
Finally, a three-bond correlation in the heteronuclear multiple
bond correlation (HMBC) spectrum is observed between the
doublet at 2.95 ppm and the resonance of the most downfield-
shifted carbon atom at 159.7 ppm, which is assigned to the
deshielded 2-position of the benzoquinoline system. Thus, these
data are fully consistent with the diastereotopic protons Ha and
Ha′ of compound 5.[49]
g
h
.
and prolonged reaction times, presumably leading to imine
hydrolysis and exchange.[45,46] Purification of cycloadducts
R
R
R
R
R
H
N
H
H
H
H
H
H
H
H
H
Ha
Ha'
R'
H
H
H
H
Hb'
H
R'
Hb
N
O
OH
OH
3
5'
8
5
8'
Figure 1. Possible regioisomeric products 5 and 5’, with respect to enolization of 5-a-cholestan-3-one 3 to give either 8 or 8’
J. Phys. Org. Chem. (2012)
Copyright © 2012 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc

A. SCHERER ET AL.
The nuclear magnetic resonance spectra of pyrenyl and
naphthyl derivatives 5d and 5e, respectively, are also generally
consistent with the description above, but demand additional
comment. The 1H NMR spectrum of 5d exhibits two pairs of
doublets (2.77 and 2.61 ppm, 2.24 and 2.18 ppm), with each pair
of signals integrating to a total of one proton (Figure 2). Likewise,
5e also shows two pairs of doublets (2.82 and 2.56 ppm, 2.30 and
2.15 ppm), each pair again integrating to a total of one proton. In
both cases, the ratio between members of each pair is 3 : 2,
presumably reflecting the presence of two atropisomers (diaster-
eomers) that result from restricted rotation about the bond linking
the pendent pyrenyl or naphthyl group to the benzoquinoline
system. This is comparable to the axial chirality commonly encoun-
tered in biphenyl and binaphthyl derivatives.[50–52] Consistent with
this rationale, variable temperature 1H NMR experiments for
compounds 5d and 5e show that each pair of doublets coalesces
to a broad singlet upon heating to ~95 ꢀC in toluene-d8 (see
supporting Information for detailed spectroscopic data). On the
basis of the coalescence temperature, rotational barriers in the
range of 18.4 ꢁ 0.3 kcal/mol and 18.0 ꢁ 0.3 kcal/mol can be
estimated for 5d and 5e, respectively.[53] These rotational barriers
are comparable to the published values for 1,1-binaphthyl
(~24 kcal/mol),[54] taking into account the decreased rigidity of the
partially saturated tetrahydrobinaphthyl segment of 5d and 5e.
X-ray quality single crystals of 5a (R′ = Ph), 5c (R′ = thienyl), and
5e (R′ = naphthyl) were grown from toluene/MeOH, CH2Cl2/
MeOH, and CHCl3/EtOH, respectively, and crystallographic analy-
sis confirmed the spectroscopic assignment of product isomers
obtained from the condensation reaction (Figure 3). The struc-
ture solution of compound 5c shows two crystallographically
independent molecules in the unit cell (only one is shown in
Figure 3(c)). Several points of comparison for structures 5a, 5c,
and 5e are noteworthy:
appears to also be maintained in the solution state, as mea-
sured by circular dichroism (CD) spectroscopy (vide infra).
(3) Neither the phenyl substituent in 5a nor the naphthyl group
in 5e are coplanar with the conjugated framework of the
benzoquinoline moiety, presumably a consequence steric
repulsion (Figures 3(b) and (d)). In the case of 5e, this results
in axial chirality with respect to the orientation of the
a)
b)
c)
d)
(1) In each case the overall conformation of the benzoquinoline,
steroid, and alkyl side chain are remarkably similar, as
evidenced by the overlaid structures in Figure 3(a), for 5a,
5c (both molecules), and 5e.
(2) In each molecule, the quinoline fragment combines with the
A-ring of the cholesteryl unit to form a tetrahydro[4]helicene
with the (P)-configuration. The chirality of the helicene also
e)
Figure 3. (a) Overlay of the solid-state structures of 5a, 5c (both crystal-
lographically unique molecules), and 5e. (b–d) ORTEP plots of from X-ray
crystallographic analysis of 5a, 5c, and 5e, respectively (hydrogen atoms
not shown). (e) p-Stacking interactions found in the solid state for 5c
(in red). Thermal ellipsoids shown at 50% probability; cocrystallized
chloroform molecule not shown for 5e. Color scheme: carbon atoms in
gray, nitrogen in blue, and sulfur in yellow
Figure 2. 1H NMR spectra of 5d and 5e, showing signals for Ha and Ha′
as a result of the mixture of atropisomers in CDCl3 solution at room
temperature
wileyonlinelibrary.com/journal/poc
Copyright © 2012 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. (2012)

INCORPORATION OF STEROIDAL BIOMARKERS INTO PETROLEUM MODEL COMPOUNDS
naphthyl group. Whereas 5e in solution gives a 3 : 2 mixture
of diastereomers, only one diastereomer is found in the
solid-state structure of 5e, that with the (Sa)-configuration.
To date, single crystals of the other possible diastereomer,
with the (Ra)-configuration, have not been obtained.
finds itself surrounded only by alkyl groups of the choles-
teryl unit.[49] The arrangement of 5a and 5e is consistent
with the previously reported self-assembly of cholesteryl
structures appended to polycyclic aromatics such as anthra-
cene.[55] This observation also supports the suggestion by
Gray et al. [3] that a variety of associative intermolecular forces
are important in the aggregation of large petroleum mole-
cules, not merely aromatic pi-stacking.
(4) The thienyl group in 5c approaches a coplanar orientation
with respect to the pyridyl ring of the benzoquinoline,
showing a tosional angle of 19–20ꢀ (derived from planes
composed of the nonhydrogen atoms of each ring, for each
of the two unique molecules in the unit cell). This coplanarity
is manifested in the red-shifted lmax observed in the UV–Vis
spectrum of 5c relative to the other derivatives (vide infra).
(5) Although the solid-state conformations determined for
each of the four structures (5a, both molecules of 5c,
and 5e) are similar, the nature of intermolecular interac-
tions in the solid-state varies quite substantially. In particu-
lar, aromatic p-stacking is apparent only in the solid-state
structure of thienyl derivative 5c, as highlighted in red in
Figure 3(e), showing close contacts of ~3.5–3.8 Å between
the two crystallographically unique molecules. In the other
two cases (5a and 5e), the p-system of the benzoquinoline
The chirality embedded with the cholestane unit ensures that
cycloadducts 5a–h are also nonracemic.[56,57] Similar to cholesta-
20
none 3 (½aꢂD = +39, c = 0.14, CH2Cl2),[58] most of the compounds
– 5a–c,f,g – also show positive optical rotation values.[59]
Compounds with additional axial chirality based on the pendent
pyrenyl (5d) and naphthyl (5e) groups, on the other hand, show
20
negative optical rotations of ½aꢂD = ꢃ57 (c = 0.26, CH2Cl2) and
20
½aꢂD = ꢃ3.0 (c = 0.28, CH2Cl2), respectively.
The electronic absorption characteristics of substituted 5,6-
benzoquinolines have been extensively studied by Kozlov and
Basalaeva. Substitution at the 2-position typically induces only
small shifts in the electronic spectra.[43] Selected UV–Vis data
from the substituent series 5a–h are used to highlight trends,
a)
a)
b)
b)
c)
Figure 4. (a) UV–Vis and (b) CD spectra for compounds 5a, 5c, and 5e as
measured in CH2Cl2
Figure 5. Arrows illustrate (P)-helicity of benzoquinoline unit in the
solid-state structures of (a) 5a, (b) 5c, and (c) 5e
J. Phys. Org. Chem. (2012)
Copyright © 2012 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc

A. SCHERER ET AL.
as shown in Figure 4(a) for structurally characterized derivatives
behavior of 5a–h in contexts pertinent to modeling asphaltene
5a, 5c, and 5e. The absorption spectra for compounds 5a and
composition and aggregation is currently in progress.
5e show
lmax = 353 nm (e = 4160) and 351 nm (e = 5500),
respectively, and two slightly weaker absorptions at higher
energy in the range of 340–320 nm. The electronic spectrum
of the thienyl derivative 5c deviates from this pattern, showing
a distinctly lower energy HOMO–LUMO gap: the two lower en-
ergy absorptions for 5c are observed at l = 373 nm (e = 8770)
and 357 nm (e = 9860). Thus, the smaller ring size and hetero-
aromatic structure of the pendent thiophene group facilitate
coplanarity with the rest of the framework (consistent with
the crystallographic analysis), allowing the p-system to remain
conjugated with the thienyl group, lowering the HOMO–LUMO
gap for this material.
The CD spectra for 5a, 5c, and 5e are shown in Figure 4(b), as
determined in CH2Cl2 solution. Roughly speaking, two series of
signals appear for each of the three derivatives, one in the high
energy UV absorption region, ~250–310 nm, and one in the
lower energy region, ~310–400 nm. In all three cases, the
observed Cotton effects correspond reasonably well to the lmax
values obtained from the UV–Vis absorption spectra. This is
particularly true for compound 5c, where zero-crossing points
in the CD spectrum at 315, 344, and 383 nm show excellent
correlation to the UV absorptions at 317, 357, and 373 nm.
The strong Cotton effects at lower energy are intriguing,
because these signals presumably arise from the thienyl-
substituted benzoquinoline chromophore. Thus, the CD analysis
suggests that the helical (P)-conformation of the benzoquino-
line unit observed in the solid state (Figure 5) is also preserved
in solution.
EXPERIMENTAL SECTION
General
All experiments were carried out under ambient conditions, unless other-
wise noted. Melting points were recorded on an Electrothermal melting
point apparatus. Infrared (IR) spectra were recorded on a Varian IR-660
spectrometer (Varian / Agilent Technologies, Santa Clara (CA), USA).
UV–Vis measurements were carried out on a Varian Cary 5000 UV–Vis–
NIR spectrophotometer (Varian / Agilent Technologies, Santa Clara (CA),
USA) using 1 cm QS quartz cuvettes from Hellma (Hellma Analytics, Müll-
heim, Germany). CD measurements were performed on a Jasco J-815
spectrophotometer (Jasco Germany, Gross-Umstadt, Germany) using
1 mm QS quartz cuvettes from Hellma (Hellma Analytics, Müllheim, Ger-
many). Optical rotations were measured using a Perkin-Elmer model
431 polarimeter (Perkin-Elmer, Waltham (MA), USA). 1H and 13 C NMR
spectra were recorded on Bruker 300 and 400 MHz spectrometers (Bruker
BioSpin, Rheinstetten, Germany). Chemical shifts were referenced to the
residual solvent signal (1H: CHCl3, 7.24 ppm and toluene-d8, 2.03 ppm;
13 C: CDCl3, 77.0 ppm). Mass spectra were obtained using a Micromass
Zabspec instrument (Micromass / Waters, Milford (MA), USA), a Shimadzu
AXIMA Confidence mass spectrometer (Shimadzu, Duisburg, Germany),
and a Bruker 9.4 T Apex-Qe FTIC (Bruker Daltonics, Bremen, Germany).
Thin layer chromatography was carried out with ALUGRAM SIL G/UV254
plates (Macherey–Nagel, Düren, Germany). Flash column chromatogra-
phy was conducted using silica gel 60 M (Macherey–Nagel, Düren,
Germany).
Benzene, EtOH, i-PrOH, EtOAc, toluene, Et2O, MeOH, CH2Cl2, CHCl3,
and hexanes solvents were commercial grade and used as received. Tet-
rahydrofuran (THF) was distilled from sodium/benzophenone prior to
use. Benzaldehyde, 4-n-butylbenzaldehyde, thiophene-2-carbaldehyde, 1-
naphthaldehyde, 2-naphthaldehyde, LiOH•H2O, Pd/C, 4-bromobenzalde-
hyde, and 5a-cholestan-3-one were purchased from commercial sources
and used as received. 2-Naphthalenamine,[60] pyrene-1-carbaldehyde,[61]
(1-naphthylmethyl)triphenylphosphonium chloride, and benzyltriphenyl-
phosphonium chloride[62] were synthesized as previously reported.
As a final comment regarding chirality, compounds 5 d (R =
pyrenyl) and 5e (R = naphthyl) provide a rather unique pair of
chiral molecules that combine stereochemical contributions
from all three common sources: (i) stereogenic centers, from
the cholesteryl moiety; (ii) helical chirality, from the embedded
tetrahydrohelicene; and (iii) axial chirality, arising from the pen-
dent aromatic polycycle. Further studies are necessary, however,
to isolate and define the relative contributions of each factor to
the overall molecular chirality, particularly with respect to the
CD signatures.
General procedure A: N-(Arylmethylene)-2-naphthalenamines 4a–h
This procedure follows the basic protocol described by Kozlov and
Basalaeva.[43] 2-Naphthalenamine (10.0 mmol) was added to a solution
of the appropriate aldehyde (10.0 mmol) in PhMe (50 mL) and the
solution stirred at reflux in a Dean–Stark apparatus for 16 h. The reaction
was cooled to room temperature (r.t.) and the solvent removed in vacuo.
The resulting residue was crystallized from EtOH (100 mL) and the solid
collected by filtration and dried in vacuo to give the reported products.
CONCLUSIONS
A series of eight cholestane-benzoquinoline conjugates 5a–h
has been synthesized in modest yields (10–39%) by the acid-
catalyzed cyclocondensation of 5a-cholestan-3-one with a series
of aromatic Schiff bases (4a–h). The large polycyclic arrays, partly
aromatic and partly aliphatic, present in compounds 5a–h are
designed to serve as model asphaltene biomarkers, featuring a
5-a-cholestan-3-yl moiety covalently fused to a benzoquinoline,
a typical tricyclic aromatic heterocycle found in bitumen. The
structures of 5a–h were assigned through a combination of spec-
troscopic and single-crystal X-ray analyses. These studies reveal
that only one regioisomeric cycloadduct is obtained, based on
the preferential enolization of the 5-a-cholestan-3-one precursor.
1-Pyrenyl- and 1-naphthylbenzoquinolines 5d and 5e show axial
chirality as a result of the size of these aromatic substituents, and
both are present as diastereomeric mixtures in solution at room
temperature. X-ray crystallography and CD spectroscopy suggest
that the substituted benzoquinoline moiety adopts a (P)-helicene
conformation. Investigation of the chemical and physical
General procedure B: Benzoquinolines 5a–h
5a-Cholestan-3-one (0.50 mmol) was added to a solution of conc. HCl
(0.1 mL) in EtOH (5 mL). The appropriate N-(arylmethylene)-2-naphthale-
namine (4a–h) (0.50 mmol) was added and the mixture stirred at reflux
for 16 h. The reaction was cooled to r.t., the solvent removed in vacuo,
and Et2O (50 mL) and NH3 (25% in water, 25 mL) were added. The organic
phase was separated, washed with NH3 (25% in water, 2 ꢄ 25 mL), brine
(2 ꢄ 25 mL), 2 M NaOH (2 ꢄ 25 mL), dried (Na2CO3), filtered, and the sol-
vent removed in vacuo. The residue was dissolved in hexanes (100 mL)
and filtered through a plug of celite. The solution was saturated with
HCl gas (generated from NaCl and conc. H2SO4). The resulting precipitate
was collected by filtration, washed with hexanes (2 ꢄ 25 mL), dissolved in
CH2Cl2 (50 mL), and filtered through a plug of celite. The solution was
washed with 1 M NaOH (2 ꢄ 25 mL), and the organic phase separated,
dried (MgSO4), and filtered. The solvent was removed in vacuo. Purifica-
tion by column chromatography (silica gel, CH2Cl2/hexanes 1:4–19:1)
gave the desired product.
wileyonlinelibrary.com/journal/poc
Copyright © 2012 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. (2012)

INCORPORATION OF STEROIDAL BIOMARKERS INTO PETROLEUM MODEL COMPOUNDS
153.1, 148.9, 142.8, 134.0, 132.3, 132.0, 130.4, 128.9, 127.9, 127.8, 127.7,
Aldehyde 2g
126.4, 125.3, 121.0, 118.1. EI HRMS calcd. for C25H11NS (M+) 237.0612,
found 237.0604.
Aryl bromide 7a (2.281 g, 7.362 mmol) was added to dry THF (50 mL)
under a nitrogen atmosphere. The mixture was cooled to ꢃ78 ꢀC with stir-
ring and n-BuLi (2.5M in hexanes, 3.24 mL, 8.09 mmol) was slowly added.
After 1 h, dry dimethylformamide (DMF) (0.56mL, 0.59 g, 8.1mmol) was
added, the cooling bath was removed, and stirring was continued for
16h. To the reaction mixture was then added brine (50mL), conc. HCl
(25 mL), and Et2O (100mL). The organic layer was separated, washed with
brine (3 ꢄ 50mL), dried (Na2SO4), and filtered through a plug of silica gel.
The solvent was removed in vacuo to give 2g as a pale yellow solid
(0.919 g, 48%). Mp = 156–157 ꢀC. IR (cast film) 3044, 2822, 1691, 1599,
N-(1-Pyrenylmethylene)-2-naphthalenamine (4d)
The reaction of 2-naphthalenamine (0.716 g, 5.00 mmol) with pyrene-1-
carbaldehyde (1.151 g, 5.004 mmol) according to general procedure A
gives imine 4d as a yellow solid (0.675 g, 38%). Mp = 127–129 ꢀC. IR (cast
film) 3044, 2964, 1579, 1501, 1383 cm–1. 1H NMR (300 MHz, CDCl3) d 9.58
(s, 1H), 9.04 (d, J = 9.3 Hz, 1H), 8.78 (d, J = 8.1 Hz, 1H), 8.24–.8.03 (m, 7H),
7.94–7.87 (m, 3H), 7.75 (s, 1H), 7.63–7.60 (m, 1H), 7.54–7.44 (m, 2H). 13 C
NMR (75 MHz, CDCl3) d 159.0, 150.5, 134.3, 133.5, 132.1, 131.3, 130.64,
130.59, 129.12, 129.07, 129.05, 128.5, 128.1, 127.9, 127.5, 126.8, 126.5,
126.3, 126.2, 126.0, 125.5, 125.1, 124.9, 124.6, 122.4, 124.5, 117.9. EI HRMS
calcd. for C27H16N ([M – H]+) 354.1283, found 354.1291.
1165 cm–1 1H NMR (300 MHz, CDCl3) d 9.95 (s, 1H), 8.04 (d, J = 6.2Hz,
.
1H), 7.86 (d, J = 5.9 Hz, 1H), 7.84–7.70 (m, 3H), 7.52–7.47 (m, 2H),
7.37–7.30 (m, 3H), 7.20 (d, J= 6.2Hz, 1H), 3.40–3.34 (m, 2H), 3.13–3.08
(m, 2H). 13 C NMR (75MHz, CDCl3) d 192.0, 149.2, 136.8, 134.6, 133.9, 131.6,
129.9, 129.1, 128.9, 127.0, 126.1, 126.0, 125.53, 125.45, 123.4, 37.1, 34.5.
EI HRMS calcd. for C19H16NO (M+) 260.1201, found 260.1197.
N-(1-Naphthalenylmethylene)-2-naphthalenamine (4e)
Aldehyde 2h
The reaction of 2-naphthalenamine (0.917 g, 6.40 mmol) with 1-naphthal-
dehyde (0.96 mL, 1.0 g, 6.4 mmol) according to general procedure A pro-
vides imine 4e as a pale brown solid (1.206 g, 67%). Mp = 103–104 ꢀC. IR
(cast film) 3050, 2869, 1580, 1503, 1438 cm–1. 1H NMR (300 MHz, CDCl3) d
9.22 (s, 1H), 9.11 (d, J = 8.7 Hz, 1H), 8.15 (d, J = 7.1 Hz, 1H), 8.00–7.85
(m, 5H), 7.67–7.42 (m, 7H). 13 C NMR (75 MHz, CDCl3) d 160.2, 150.3, 134.1,
133.9, 132.00, 131.96, 131.50, 131.45, 130.0, 129.0, 128.8, 127.9, 127.7,
127.5, 126.4, 126.2, 125.3, 124.3, 121.3, 117.5 (one signal coincidental or
not observed). EI HRMS calcd. for C21H15N (M+) 281.1204, found 281.1213.
Bromide 7b (2.841 g, 10.91 mmol) was added to dry THF (25 mL) under a
N2 atmosphere. The mixture was cooled to ꢃ78 ꢀC with stirring and
n-BuLi (2.5 M in hexanes, 5.4 mL, 13.5 mmol) was slowly added. After
1 h, DMF (0.84 mL, 0.79 g, 14 mmol) was added, the cooling bath was re-
moved, and stirring was continued for 16 h. To the reaction mixture was
then added brine (50 mL), conc. HCl (25 mL), Et2O (100 mL). The organic
layer was separated, washed with brine (3 ꢄ 50 mL), dried (Na2SO4), and
filtered through a plug of silica gel. The solvent was removed in vacuo
to give 2h as a pale yellow wax (1.76 g, ~70%), which was ~70% pure by
1H NMR spectroscopy. This product was exceedingly difficult to purify
further; instead, it was carried on without further purification. 1H NMR
(300 MHz, CDCl3) d 9.94 (s, 1H), 7.77 (d, J = 8.6 Hz, 2H), 7.30–7.14 (m, 7H),
2.98–2.90 (m, 4H). 13 C NMR (75 MHz, CDCl3) d 192.0, 149.0, 140.8, 134.5,
129.9, 129.1, 128.4, 128.3, 126.1, 38.0, 37.3.
N-(2-Naphthalenylmethylene)-2-naphthalenamine (4f)
The reaction of 2-naphthalenamine (1.189 g, 8.309 mmol) with 2-
naphthaldehyde (1.297 g, 8.309 mmol) according to general procedure
A gives imine 4f as a pale brown solid (1.471 g, 63%). Mp = 99–100 ꢀC.
IR (cast film) 3052, 2862, 1588, 1501, 1436 cm–1. 1H NMR (300 MHz, CDCl3)
d 8.74 (s, 1H), 8.23–8.21 (m, 2H), 7.95–7.84 (m, 6H), 7.66 (d, J = 1.4 Hz, 1H),
7.59–7.42 (m, 5H). 13 C NMR (75 MHz, CDCl3) d 160.5, 149.6, 135.0, 134.1,
134.0, 133.1, 132.0, 131.4, 129.0, 128.8, 128.7, 127.9, 127.7, 127.6, 126.6,
126.4, 125.4, 123.9, 121.1, 117.9 (one signal coincidental or not observed).
EI HRMS calcd. for C25H21N (M+) 281.1204, found 281.1195.
N-(Phenylmethylene)-2-naphthalenamine (4a)
2-Naphthalenamine (1.430 g, 10.00 mmol) was treated with benzalde-
hyde (1.02 mL, 1.06 g, 10.0 mmol) according to general procedure A,
giving imine 4a as a pale brown solid (1.595 g, 69%). Mp = 98–99 ꢀC. IR
(cast film) 3053, 2892, 1576, 1501, 1447 cm–1 1H NMR (300 MHz, CDCl3)
.
d 8.48 (s, 1H), 7.88–7.84 (m, 2H), 7.78–7.74 (m, 3H), 7.51 (d, J = 1.8 Hz,
1H), 7.42–7.32 (m, 6H). 13 C NMR (75 MHz, CDCl3) d 160.5, 149.6, 136.2,
134.0, 131.9, 131.4, 128.9, 128.82, 128.75, 127.9, 127.7, 126.4, 125.3,
121.1, 117.7. EI HRMS calcd. for C17H13N (M+) 231.1048, found 231.1039.
N-(4-(2-(1-Naphthalenyl)ethyl)phenylmethylene)-2-naphthalena-
mine (4g)
The reaction of 2-naphthalenamine (0.843 g, 5.89 mmol) with 2 g
(1.531 g, 5.889 mmol) according to general procedure A gives imine 4g
as an off-white solid (1.385 g, 61%). Mp = 123–124 ꢀC. IR (cast film)
N-(4-n-Butylphenylmethylene)-2-naphthalenamine (4b)
3053, 2862, 1591, 1504, 1391 cm–1 1H NMR (300 MHz, CDCl3) d 8.57 (s,
.
1H), 8.11 (d, J = 8.2 Hz, 1H), 7.89–7.83 (m, 6H), 7.74 (d, J = 8.2 Hz, 1H),
7.60–7.25 (m, 10H), 3.44–3.39 (m, 2H), 3.16–3.10 (m, 2H). 13 C NMR
(75 MHz, CDCl3) d 160.4, 149.8, 145.8, 137.3, 134.3, 134.1, 133.9, 131.9,
131.7, 129.00, 128.97, 128.9, 127.9, 127.7, 126.9, 126.4, 126.2, 126.0,
125.5, 125.3, 123.5, 121.2, 117.7, 37.1, 34.8 (two signals coincidental or
not observed). EI HRMS calcd. for C29H23N (M+) 385.1830, found 385.1831.
The reaction of 2-naphthalenamine (1.928 g, 13.48 mmol) and 4-n-butyl-
benzaldehyde (2.26 mL, 2.18 g, 13.5 mmol) according to general proce-
dure A provides imine 4b as an orange solid (1.821 g, 47%). Mp =
90–91 ꢀC. IR (cast film) 3053, 2924, 1586, 1504, 1458 cm–1 1H NMR
.
(400 MHz, CDCl3) d 8.55 (s, 1H), 7.88–7.83 (m, 5H), 7.59 (d, J = 1.7 Hz,
1H), 7.48–7.43 (m, 3H), 7.31 (d, J = 8.0 Hz, 2H), 2.69 (t, J = 7.7 Hz, 2H),
1.69–1.61 (m, 2H), 1.44–1.34 (m, 2H), 0.96 (t, J = 7.4 Hz, 3H). 13 C NMR
(100 MHz, CDCl3) d 160.5, 149.9, 146.9, 134.1, 131.9, 128.88, 128.86,
127.8, 127.7, 126.3, 125.2, 121.2, 117.6, 35.7, 33.4, 22.3, 13.9 (two signals
N-(4-(Phenylethyl)phenylmethylene)-2-naphthalenamine (4h)
coincidental or not observed). EI HRMS calcd. for
287.1674, found 287.1669.
C21H21N
(M+)
The reaction of 2-naphthalenamine (1.616 g, 11.28 mmol) with the crude
aldehyde 2h (2.611 g) according to general procedure A, except using
benzene (50 mL) as solvent, provides imine 4h as a pale brown solid
(2.75 g, 73%). Mp = 111–112 ꢀC. IR (cast film) 3022, 2916, 2851, 1590,
N-(2-Thienylmethylene)-2-naphthalenamine (4c)
1567, 1493, 1452 cm–1 1H NMR (300 MHz, CDCl3) d 8.53 (s, 1H),
.
The reaction of 2-naphthalenamine (1.431 g, 10.00 mmol) with thio-
phene-2-carbaldehyde (0.93 mL, 1.1 g, 10 mmol) according to general
procedure A gives imine 4c as a yellow solid (1.517 g, 64%). Mp =
7.85–7.82 (m, 4H), 7.57 (d, J = 2.0 Hz, 1H), 7.49–7.39 (m, 3H), 7.30–7.15
(m, 8H), 3.02–2.90 (m, 4H). 13 C NMR (75 MHz, CDCl3) d 160.4, 149.8, 145.6,
141.3, 134.14, 134.08, 131.9, 129.0, 128.91, 128.88, 128.5, 128.4, 127.9,
127.7, 126.3, 126.0, 125.2, 121.2, 117.6, 37.9, 37.6. EI MS m/z 335 (M+, 35),
91 ([C7H7]+, 100). EI HRMS calcd. for C25H21N (M+) 335.1674, found
335.1677.
105–106 ꢀC. IR (cast film) 3052, 2882, 1581, 1502, 1423 cm–1 1H NMR
.
(300 MHz, CDCl3) d 8.69 (s, 1H), 7.86–7.82 (m, 3H), 7.61 (d, J = 1.8 Hz,
1H), 7.53–7.40 (m, 5H), 7.15–7.12 (m, 1H). 13 C NMR (75 MHz, CDCl3) d
J. Phys. Org. Chem. (2012)
Copyright © 2012 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc

A. SCHERER ET AL.
Compound 5a
X-ray crystallographic data for 5c
5-a-Cholestan-3-one (0.339 g, 0.876 mmol) and imine 4a (0.202 g,
0.876 mmol) were treated according to general procedure B to give 5a
as a pale yellow solid (0.141 g, 27%). Mp = 216–217 ꢀC. Rf = 0.29 (CH2Cl2/
Single crystals of 5c were grown by slow evaporation of a CH2Cl2/MeOH
solution at rt. Crystal data for 5c (C42H53NS), Fw = 603.91; monoclinic crystal
system; space group P21, a = 12.1984(3) Å, b = 9.5200(3) Å, c = 29.7503(6) Å;
a =90ꢀ, b = 93.040(2)ꢀ, g =90ꢀ; V= 3450.00(15) Å3; Z=4; r(calc) = 1.163 g cm–3;
2θmax = 50.10ꢀ; l = 0.071 nm; m = 0.124 mm–1; T= 173(2) K; total data col-
lected = 11763; R1 = 0.0604 [7965 observed reflections with F20 ≥2s(F02)];
20
hexanes 19:1). ½aꢂD = 54 (c = 0.27, CH2Cl2). UV–Vis (CH2Cl2) lmax (e) 353
(4,160), 337 (3,790), 322 (2,470), 297 (13,100, sh), 271 (35,200, sh), 256
(43,800) nm. IR (cast film) 3052, 2916, 2866, 1552, 1441, 1376 cm–1 1H
.
oR2 = 0.1438 for 793 variables with F20 ≥ 3s(F02); residual electron density =
ꢃ
NMR (400 MHz, CDCl3) d 8.91 (d, J = 8.3 Hz, 1H), 7.99 (d, J = 8.9 Hz, 1H),
7.92–7.85 (m, 2H), 7.66–7.57 (m, 2H), 7.52–7.39 (m, 5H), 3.46–3.29 (m,
2H), 2.91 (d, J = 17.2 Hz, 1H), 2.36 (d, J = 17.2 Hz, 1H), 1.94 (d, J = 12.6 Hz,
1H), 1.82–0.86 (m, 35H), 0.72–0.64 (m, 4H). 13 C NMR (100 MHz, CDCl3) d
159.7, 146.7, 143.6, 141.5, 133.5, 130.6, 130.2, 129.5, 129.1, 128.8, 128.43,
128.38, 128.3 127.9, 126.1, 125.8, 124.2, 56.32, 56.27, 53.9, 43.3, 42.4,
41.7, 39.8, 39.4, 37.6, 36.1, 35.7, 35.4, 33.7, 31.6, 28.9, 28.1, 27.9, 24.1,
0.493 and ꢃ0.301 eÅ–3. CCDC 857439.
Compound 5d
5a-Cholestan-3-one (0.280 g, 0.726 mmol) and imine 4d (0.257 g,
0.726 mmol) according to general procedure B to give 5d as an orange
solid (0.141 g, 27%). In solution, 5d exists as a 3 : 2 mixture of diastereo-
mers and was characterized spectroscopically as this mixture. Mp =
23.8, 22.7, 22.5, 20.8, 18.6, 11.9, 11.6. HR MALDI MS calcd. for C44H56
N
([M + H]+) 598.4407, found 598.4398.
198–199 ꢀC. Rf = 0.20 (CH2Cl2/hexanes 19:1). ½aꢂ20 = ꢃ57 (c = 0.26, CH2Cl2).
D
IR (cast film) 2926, 2855, 1557, 1442, 1310 cm–1. 1H NMR (400 MHz, CDCl3)
d 9.03 and 9.01 (2x d, J = 6.8 Hz and 7.4 Hz, 1H), 8.33–7.91 (m, 10H),
7.76–7.63 (m, 3H), 7.54 (d, J = 9.2 Hz, 1H), 3.55–3.41 (m, 2H), 2.77 and
2.61 (2x d, J = 17.5 Hz and 17.6 Hz, 1H), 2.24 and 2.18 (2x d, J = 17.2 Hz
and 17.3 Hz, 1H), 1.71–0.47 (m, 40H). 13 C NMR (100 MHz, CDCl3) d 159.5,
159.4, 147.1, 146.9, 143.8, 143.4, 136.3, 136.2, 133.6, 131.4, 131.2, 131.1,
131.0, 130.71, 130.65, 130.4, 130.3, 130.0, 129.9, 129.6, 129.5, 129.19,
129.16, 129.0, 128.6, 128.5, 127.94, 127.8, 127.6, 127.5, 126.7, 126.3,
126.24, 126.15, 126.0, 125.92, 125.87, 125.3, 125.2, 125.1, 125.0, 124.94,
124.91, 124.86, 124.82, 124.53, 124.49, 124.45, 56.08, 56.0, 55.9, 53.54,
53.47, 42.9, 42.5, 42.2, 42.1, 41.73, 41.69, 39.44, 39.36, 39.3, 37.7, 37.6,
36.00, 35.98, 35.6, 35.3, 35.2, 33.53, 33.49, 31.42, 31.38, 29.6, 28.9, 28.8,
28.0, 27.9, 24.0, 23.74, 23.66, 22.7, 22.6, 22.4, 20.4, 18.4, 18.3, 11.8, 11.7,
11.6, 11.3 (108 resonances expected for both isomers, 91 observed). HR
MALDI MS calcd. for C54H60N ([M + H]+) 722.4720, found 722.4723.
X-ray crystallographic data for 5a
Single crystals of 5a were grown by slow evaporation of a PhMe/MeOH
solution at r.t. Crystal data for 5a (C44H55N), Fw = 597.89; orthorhombic
crystal system; space group P212121, a = 10.5822(3) Å, b = 11.1289(3) Å,
c = 29.5000(7) Å; a = 90.00ꢀ, b = 90.00ꢀ, g = 90.00ꢀ; V = 3474.17(15) Å3;
Z = 4; r(calc) = 1.143 g cm–3; 2θmax = 146.66ꢀ; l = 0.154 nm; m = 0.479 mm–
1; T = 173.0 K; total data collected = 10789; R1 = 0.0339 [6633 observed
reflections with F20 ≥ 2s(F02)]; oR2 = 0.0837 for 412 variables with F20 ≥
ꢃ3s(F
2
); residual electron density = 0.135 and ꢃ0.145 eÅ–3. CCDC
0
857441.
Compound 5b
5a-Cholestan-3-one (0.193 g, 0.501 mmol) and imine 4b (0.144 g,
0.501 mmol) were treated according to general procedure B to give 5b
as an orange solid (0.072 g, 22%). Mp = 105–106 ꢀC. Rf = 0.35 (CH2Cl2/hex-
Compound 5e
20
anes 19:1). ½aꢂD = 60 (c = 0.26, CH2Cl2). IR (cast film) 3050, 2924, 2860,
1550, 1439, 1365 cm–1 1H NMR (400 MHz, CDCl3) d 8.89 (d, J = 8.5 Hz,
.
5a-Cholestan-3-one (0.222 g, 0.575 mmol) and imine 4e (0.162 g,
0.575 mmol) were treated according to general procedure B to give 5e
as a pale yellow solid (0.145 g, 39%). In solution, 5e exists as a 3 : 2 mix-
ture of diastereomers and was characterized spectroscopically as this
1H), 7.99 (d, J = 8.8 Hz, 1H), 7.89–7.83 (m, 2H), 7.64–7.54 (m, 2H), 7.43
(d, J = 7.8 Hz, 2H), 7.28 (d, J = 7.9 Hz, 2H), 3.43–3.26 (m, 2H), 2.95 (d,
J = 17.2 Hz, 1H), 2.68 (t, J = 7.4 Hz, 2H), 2.35 (d, J = 17.2 Hz, 1H), 1.95
(d, J = 12.6 Hz, 1H), 1.84–0.88 (m, 43H), 0.65 (s, 3H). 13 C NMR (100 MHz,
CDCl3) d 159.7, 146.6, 143.3, 142.4, 138.7, 133.4, 130.6, 129.9, 129.5,
129.0, 128.7, 128.34, 128.31, 128.26, 125.9, 125.6, 124.0, 56.32, 56.29,
53.9, 43.4, 42.4, 41.7, 39.8, 39.5, 37.7, 36.2, 35.8, 35.5, 35.4, 33.7, 33.5,
31.6, 28.9, 28.2, 28.0, 24.2, 23.9, 22.8, 22.6, 22.4, 20.9, 18.6, 14.0, 11.9,
11.7. 1H–1H COSY (400 MHz, CDCl3-, selected data only) d 2.95 , 2.35.
13 C–1H HSQC (400 MHZ, CDCl3) d 43.4 , 2.95, 2.35. 13 C–1H HMBC (400
MHZ, CDCl3) d 159.7 , 2.95; 142.4 , 2.95, 2.35; 128.26 , 2.95, 2.35;
42.4 , 2.95; 35.4 , 2.95, 2.35; 11.7 , 2.95, 2.35. HR MALDI MS calcd.
for C48H64N ([M + H]+) 654.5033, found 654.5026.
20
mixture. Mp = 108–109 ꢀC. Rf = 0.28 (CH2Cl2/hexanes 19:1). ½aꢂD = ꢃ3.0
(c = 0.28, CH2Cl2). UV–Vis (CH2Cl2) lmax (e) UV–Vis (CH2Cl2) lmax (e): 351
(5,500), 335 (5,200), 319 (4,100), 294 (26,100, sh), 284 (32,100), 273
(33,000). IR (cast film) 3048, 2917, 2866, 1553, 1441, 1381 cm–1 1H NMR
.
(400 MHz, CDCl3) d 9.03 and 9.01 (2x d, J = 7.0 Hz and 7.1 Hz, 1H),
8.06–7.92 (m, 5H), 7.74–7.33 (m, 7H), 3.55–3.41 (m, 2H), 2.82 and 2.56 (2x d,
J= 17.4Hz and 17.8Hz, 1H), 2.30 and 2.15 (2x d, J= 17.2Hz and 17.3Hz,
1H), 1.83–1.76 (m, 4H), 1.53–0.60 (m, 36H). 13 C NMR (100 MHz, CDCl3) d
159.0, 158.9, 146.9, 146.7, 143.6, 143.2, 138.8, 133.9, 133.6, 133.5, 131.6,
131.3, 130.60, 130.55, 130.2, 130.1, 129.8, 129.7, 129.5, 129.4, 129.10,
129.07, 128.42, 128.36, 128.3, 128.2, 126.4, 126.24, 126.20, 126.1, 125.81,
125.80, 125.77, 125.7, 125.6, 125.4, 125.0, 124.41, 124.35, 56.3, 56.23,
56.20, 53.7, 53.5, 42.6, 42.3, 41.93, 41.86, 41.8, 39.7, 39.6, 39.5, 37.8, 37.7,
36.1, 35.7, 35.4, 33.6, 33.5, 31.6, 31.5, 29.0, 28.2, 28.0, 26.9, 24.2, 23.8,
22.8, 22.5, 20.6, 20.5, 19.4, 18.6, 11.90, 11.86, 11.8, 11.3 (96 resonances
expected for both isomers, 77 observed). HR MALDI MS calcd. for
C48H58N ([M + H]+) 648.4564, found 648.4558.
Compound 5c
5a-Cholestan-3-one (0.245 g, 0.636 mmol) and imine 4c (0.202 g,
0.852 mmol) were treated according to general procedure B to give 5c
as a yellow solid (0.134 g, 35%). Mp = 147–148 ꢀC. Rf = 0.22 (CH2Cl2/hex-
20
anes 19:1). ½aꢂD = 0.73 (c = 0.29, CH2Cl2). UV–Vis (CH2Cl2) lmax (e) 373
(8,770), 357 (9,860), 317 (16,900), 296 (32,600, sh), 284 (40,700), 269
(39,700) nm. IR (cast film) 2922, 2860, 1546, 1436, 1364 cm–1 1H NMR
.
Xray crystallographic data for 5e
(400 MHz, CDCl3) d 8.80 (d, J = 8.4 Hz, 1H), 7.93 (d, J = 8.9 Hz, 1H),
7.83–7.79 (m, 2H), 7.58–7.51 (m, 3H), 7.46–7.44 (m, 1H), 7.16–7.14 (m, 1H),
3.24–3.18 (m, 3H), 2.55 (d, J = 16.6Hz, 1H), 2.04 (d, J = 12.2 Hz, 1H),
1.56–0.84 (m, 36H), 0.66 (s, 3H). 13 C NMR (100 MHz, CDCl3) d 151.5, 146.7,
145.3, 143.9, 133.5, 130.5, 130.3, 129.3, 129.1, 128.3, 128.0, 127.8, 127.4,
126.0, 125.8, 123.9, 56.3, 56.2, 54.2, 43.7, 42.4, 41.1, 39.9, 39.5, 37.5, 36.1,
35.7, 35.2, 33.5, 31.4, 28.7, 28.1, 27.9, 24.1, 23.9, 22.8, 22.2, 21.2, 18.6, 11.8,
11.7 (one signal coincidental or not observed). HR MALDI MS calcd.
for C42H54NS ([M+ H]+) 604.3971, found 604.3963.
Single crystals of 5e were grown by slow evaporation of a CHCl3/EtOH so-
lution at r.t. Crystal data for 5e•CHCl3 (C29H58Cl3N), Fw = 767.31; Mono-
clinic crystal system; space group P21, a = 10.6416(2) Å, b = 11.0931(2) Å,
c = 17.7810(3) Å; a = 90ꢀ, b = 100.376(2)ꢀ, g = 90ꢀ; V = 2064.69(6) Å3; Z = 2;
r
(calc) = 1.234 g cm–3; 2θmax = 148.30ꢀ; l = 0.154 nm; m = 2.260 mm–1
;
T = 173(2) K; total data collected = 7936; R1 = 0.0553 [7409 observed
reflections with F20 ≥ 2s(F02)]; oR2 = 0.1450 for 478 variables with F20 ≥
ꢃ3s(F
2
); residual electron density = 0.469 and–0.424 eÅ–3. CCDC 857439.
0
wileyonlinelibrary.com/journal/poc
Copyright © 2012 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. (2012)

INCORPORATION OF STEROIDAL BIOMARKERS INTO PETROLEUM MODEL COMPOUNDS
filtration, and dried in vacuo to give 7a[63] as a pale yellow solid
Compound 5f
(1.219 g, 67%). Mp = 128–129 ꢀC. IR (cast film) 3059, 2929, 1484, 1393,
799. 1H NMR (300 MHz, CDCl3) d 7.93 (d, J = 8.0 Hz, 1H), 7.76–7.73
(m, 1H), 7.60 (d, J= 8.2Hz, 1H), 7.43–7.33 (m, 2H), 7.29–7.21 (m, 3H), 7.10
(d, J= 6.9 Hz, 1H), 6.94 (d, J = 8.3Hz, 2H), 3.23–3.18 (m, 2H), 2.90–2.84 (m,
2H). 13 C NMR (75 MHz, CDCl3) d 140.8, 137.2, 133.9, 131.6, 131.4, 130.2,
5a-Cholestan-3-one (0.187 g, 0.486 mmol) and imine 4f (0.136 g,
0.486 mmol) were treated according to general procedure B to give 5f
as a pale yellow solid (0.069 g, 22%). Mp = 167–168 ꢀC. Rf = 0.26 (CH2Cl2/
20
hexanes 19:1). ½aꢂD = 55 (c = 0.26, CH2Cl2). IR (cast film) 2926, 2866,
1561, 1442, 1365 cm–1 1H NMR (400 MHz, CDCl3) d 8.94 (d, J = 8.1 Hz,
.
128.9, 128.4, 126.9, 126.1, 125.9, 125.5, 123.5, 119.7, 36.3, 34.8. EI HRMS
Br (M+) 312.0337, found 312.0334.
81
18 15
1H), 8.03–7.88 (m, 7H), 7.69–7.60 (m, 3H), 7.53–7.50 (m, 2H), 3.52–3.34
(m, 2H), 2.98 (d, J = 17.1 Hz, 1H), 2.44 (d, J = 17.0 Hz, 1H), 1.87–0.62 (m,
40H). 13 C NMR (100 MHz, CDCl3) d 159.7, 146.8, 143.7, 139.1, 133.6,
133.5, 133.0, 130.7, 130.3, 129.5, 129.2, 128.52, 128.50, 128.46, 128.1,
127.9, 127.8, 126.8, 126.22, 126.18, 126.16, 125.8, 124.3, 56.3, 56.2, 53.9,
43.4, 42.4, 41.8, 39.8, 39.4, 37.7, 36.1, 35.7, 35.4, 33.8, 31.6, 29.0, 28.1,
27.9, 24.1, 23.8, 22.7, 22.5, 20.8, 18.5, 11.9, 11.7. HR MALDI MS calcd. for
m/z calcd. for C
H
Compound 7b
Benzyltriphenylphosphonium chloride 6b (2.549 g, 6.559 mmol) was
added to a solution of LiOH•H2O (0.368 g, 8.74 mmol) in i-PrOH (50 mL)
and stirred for 20 min. 4-Bromobenzaldehyde (1.112 g, 6.213 mmol) was
added and the mixture was stirred at reflux for 16 h. The reaction was
cooled to r.t., EtOAc (75 mL) and brine (75 mL) were added, and the or-
ganic phase was separated. The solution was dried (MgSO4), filtered,
and the solvent removed in vacuo. The residue was recrystallized from
EtOH (100 mL), the precipitate collected by filtration, and dried in vacuo.
The crude product ((E/Z)-1-bromo-4-styrylbenzene), PhMe (50 mL), and
10% Pd/C (250 mg) were added to a glass hydrogenation tube. The tube
was placed in an autoclave, the autoclave purged 3ꢄ with hydrogen, and
the hydrogen pressure regulated at 1200 psi for 2 d. The inhomogeneous
mixture was filtered to remove the catalyst, and the solvent was then
removed in vacuo. The resulting solid was recrystallized from EtOH
(100 mL), the precipitate collected by filtration, and dried in vacuo to give
7b[63] as a pale yellow solid (1.262 g, 80%). IR (cast film) 3051, 1589, 1482,
C
48H58N ([M + H]+) 648.4564, found 648.4555.
Compound 5g
5a-Cholestan-3-one (0.237 g, 0.614 mmol) and imine 4g (0.236 g,
0.614 mmol) were treated according to general procedure B to give 5g
as a pale yellow solid (0.046 g, 10%). Mp = 112–113 ꢀC. Rf = 0.33 (CH2Cl2/
20
hexanes 19:1). ½aꢂD = 57 (c = 0.26, CH2Cl2). IR (cast film) 3049, 2924,
2863, 1548, 1440, 1364 cm–1 1H NMR (400 MHz, CDCl3) d 8.92 (d,
.
J = 8.7 Hz, 1H), 8.16 (d, J = 8.7 Hz, 1H), 8.01 (d, J = 8.0 Hz, 1H), 7.94–7.87
(m, 3H), 7.74 (d, J = 8.1 Hz, 1H), 7.67–7.47 (m, 6H), 7.41–7.32 (m, 4H),
3.46–3.30 (m, 4H), 3.16–3.12 (m, 2H), 2.99 (d, J = 17.1 Hz, 1H), 2.41 (d,
J = 17.0 Hz, 1H), 2.00 (d, J = 12.2 Hz, 1H), 1.79–0.67 (m, 39H). 13 C NMR
(100 MHz, CDCl3) d 159.6, 146.7, 143.6, 141.7, 139.2, 137.8, 134.0, 133.5,
131.8, 130.6, 130.1, 129.5, 129.1, 129.0, 128.9, 128.5, 128.4, 128.3, 126.8,
126.2, 126.1, 125.9, 125.8, 125.6, 125.5, 124.1, 123.7, 56.32, 56.29, 53.9,
43.4, 42.4, 41.7, 39.9, 39.4, 37.6, 36.8, 36.1, 35.7, 35.4, 35.0, 33.7, 31.6,
28.9, 28.2, 27.9, 24.1, 23.8, 22.7, 22.5, 20.8, 18.6, 11.9, 11.6. HR MALDI MS
calcd. for C56H66N ([M + H]+) 752.5190, found 752.5184.
1436 cm–1 1H NMR (300 MHz, CDCl3) d 7.37 (d, J = 7.0 Hz, 2H), 7.29–7.12
.
(m, 5H), 7.01 (d, J = 7.0 Hz, 2H), 2.86 (s, 4H). 13 C NMR (75 MHz, CDCl3) d
141.2, 140.6, 131.3, 130.2, 128.4, 128.3, 126.0, 119.6, 37.7, 37.2. EI MS m/z
260 (M+, 9), 91 ([C7H7]+, 100).
Acknowledgements
Compound 5h
This work was generously supported by the University of
Erlangen–Nürnberg, the Centre for Oilsands Innovation at the
University of Alberta, and Syncrude Canada, Ltd. We acknowl-
edge the use of the archipelago image from en.wikipedia.org/
wiki/File:SchärenTurku.jpg (accessed 02.12.2011), adapted for
use in the graphical abstract under terms of the GNU Free Docu-
mentation License. We thank Ms. Annie Tykwinski for creation of
the graphical abstract image.
5-a-Cholestan-3-one (0.196 g, 0.509 mmol) and imine 4h (0.171 g,
0.509 mmol) were treated according to general procedure B to give 5h
as a pale yellow solid (0.039 g, 11%). Mp = 109–110 ꢀC. Rf = 0.31 (CH2Cl2/
1
hexanes 19:1). IR (cast film) 2920, 1604, 1561, 1548, 1441, 1381 cm–1. H
NMR (400 MHz, CDCl3) d 8.91 (d, J = 8.3 Hz, 1H), 7.99–7.86 (m, 3H),
7.67–7.58 (m, 2H), 7.43 (d, J = 8.1 Hz, 2H), 7.30–7.17 (m, 7H), 3.49–3.33
(m, 2H), 2.98 (s, 4H), 2.95 (d, J = 17.8 Hz, 1H), 2.39 (d, J = 17.1 Hz, 1H), 1.97
(d, J= 11.8 Hz, 1H), 1.82–0.66 (m, 39H). 13 C NMR (100MHz, CDCl3) d 159.6,
146.6, 141.7, 141.3, 139.0, 133.4, 130.5, 130.0, 129.4, 129.0, 128.8, 128.5,
128.41, 128.35, 128.3, 126.0, 125.9, 125.7, 124.1, 56.24 56.19, 53.8, 43.3,
42.3, 41.7, 39.8, 39.3, 37.7, 37.6, 37.5, 36.0, 35.6, 35.3, 33.7, 31.5, 28.8,
28.1, 27.8, 24.1, 23.7, 22.6, 22.4, 20.7, 18.5, 11.8, 11.6 (two signals coinci-
dental or not observed). HR MALDI MS calcd. for C52H64N ([M + H]+)
702.5033, found 702.5023.
REFERENCES
[1] R. Tanaka, E. Sato, J. E. Hunt, R. E. Winans, S. Sato, T. Takanohashi,
Energy Fuel 2004, 18, 1118–1125.
[2] K. L. Gawrys, G. A. Blankenship, P. K. Kilpatrick, Energy Fuel 2006, 20,
705–714.
[3] M. R. Gray, R. R. Tykwinski, J. M. Stryker, X. Tan, Energy Fuel 2011, 25,
Compound 7a
3125–3134.
[4] C. A. Hughey, R. P. Rodgers, A. G. Marshall, Anal. Chem. 2002, 74,
4145–4149.
[5] S. K. Panda, J. T. Andersson, W. Schrader, Angew. Chem. Int. Ed. 2009,
(1-Naphthylmethyl)triphenylphosphonium
chloride
6a
(2.812 g,
6.407 mmol) was added to a solution of LiOH•H2O (0.359 g, 8.54 mmol)
in i-PrOH (250 mL) and stirred for 20 min. 4-Bromobenzaldehyde
(1.087 g, 5.873 mmol) was added and the mixture was stirred at reflux
for 16 h. The reaction was cooled to r.t., EtOAc (300 mL) and brine
(150 mL) were added, and the organic phase was separated. The solution
was dried (MgSO4), filtered, and the solvent removed in vacuo. The resi-
due was recrystallized from EtOH (300 mL), the precipitate collected by
filtration, and dried in vacuo. The crude product ((E/Z)-1-(4-bromostyryl)
naphthalene, PhMe (100 mL), and 10% Pd/C (0.250 g) were added to a
glass hydrogenation tube. The tube was placed in an autoclave, the au-
toclave purged 3ꢄ with hydrogen, and the hydrogen pressure regulated
at 1200 psi for 2 d. The inhomogeneous mixture was filtered to remove
the catalyst, and the solvent was then removed in vacuo. The resulting
solid was recrystallized from EtOH (100 mL), the precipitate collected by
48, 1788–1791.
[6] S. B. Jaffe; H. Freund, W. N. Olmstead, Ind. Eng. Chem. Res. 2005, 44,
9840–9852.
[7] O. P. Strausz; E. M. Lown, The Chemistry of Alberta Oil Sands,
Bitumens, and Heavy Oils. Alberta Energy Research Institute (AERI),
Calgary, Alberta, Canada, 2003.
[8] O. C. Mullins, Annu. Rev. Anal. Chem. 2011, 4, 393–418.
[9] E. Durand, M. Clemancey, J. M. Lancelin, J. Verstraete, D. Espinat,
A.-A. Quoineaud, Energy Fuel 2010, 24, 1051–1062.
[10] For an excellent recent review of petroleum analysis, see: R. P.
Rodgers, A. M. McKenna, Anal. Chem. 2011, 83, 4665–4687.
[11] I. Merdrignac, D. Espinat, Oil Gas Sci. Technol. 2007, 62, 7–32.
[12] A. H. Alshareef, A. Scherer, X. Tan, K. Azyat, J. M. Stryker, R. R.
Tykwinski, M. R. Gray, Energy Fuel 2011, 25, 2130–2136.
J. Phys. Org. Chem. (2012)
Copyright © 2012 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc

A. SCHERER ET AL.
[13] B. C. Owen, J. Gao, D. J. Borton, L. M. Amundson, E. F. Archibold, X.
Tan, K. Azyat, R. R. Tykwinski, M. R. Gray, H. I. Kenttamaa, Rapid
Comm. Mass Spectr. 2011, 25, 1924–1928.
[14] D. Borton, D. S. Pinkston, M. R. Hurt, X. Tan, K. Azyat, A. Scherer, R.
Tykwinski, M. Gray, K. N. Qian, H. I. Kenttamaa, Energy Fuel 2010,
24, 5548–5559.
V. Kumar, K. B. Lawrence, M. E. Logan, P. M. Scensny, J. A. Seelye,
B. E. Tomczuk, J. M. Yanni, J. Med. Chem. 1992, 35, 374–378.
[40] T. B. P. Oldenburg, H. Huang, P. Donohoe, H. Willsch, S. R. Larter, Org.
Geochem. 2004, 35, 665–678.
[41] The thermal behavior of compound 5 h was explored in this study,
see reference 12.
[15] A. H. Alshareef, K. Azyat, R. R. Tykwinski, M. R. Gray, Energy Fuel 2010,
[42] X. S. Wang, Q. Li, C. S. Yao, S. J. Tu, Eur. J. Org. Chem. 2008, 3513–3518.
[43] N. G. Kozlov, L. I. Basalaeva, Russ. J. Gen. Chem. 2001, 71, 250–256.
[44] N. G. Kozlov, K. N. Gusak, A. P. Kadutskii, Chem. Heterocycl. Compd.
2010, 46, 505–528.
[45] For discussion of Schiff base stability see: F. C. McIntire, J. Am. Chem.
Soc. 1947, 69, 1377–1380.
[46] For discussion of Schiff base stability see: P. Di Bernardo, P. L. Zanonato,
S. Tamburini, P. Tomasin, P. A. Vigato, Dalton Trans. 2006,
4711–4721.
[47] L. Velluz, J. Valls, G. Nominé, Angew. Chem. Int. Ed Engl. 1965, 4, 181–200.
[48] H. H. Inhoffen, W. Becker, G. Kölling, Liebigs Ann. Chem. 1950, 568,
181–184.
[49] See supporting information for additional details.
[50] For structural assignments of binaphthyl see: K. A. Kerr, J. M. Robertson,
J. Chem. Soc. B 1969, 1146–1149.
[51] For structural assignments of binaphthyl see: P. Wormell, G. L. D.
Ritchie, J. Mol. Struct. 1990, 240, 331–337.
[52] For structural assignments of binaphthyl see: R. B. Kress, E. N.
Duesler, M. C. Etter, I. C. Paul, D. Y. Curtin, J. Am. Chem. Soc. 1980,
102, 7709–7714.
24, 3998–4004.
[16] H. Sabbah, A. L. Morrow, A. E. Pomerantz, O. C. Mullins, X. Tan, M. R.
Gray, K. Azyal, R. R. Tykwinski, R. N. Zare, Energy Fuel 2010, 24,
3589–3594.
[17] D. S. Pinkston, P. Duan, V. A. Gallardo, S. C. Habicht, X. Tan, K. N.
Qian, M. Gray, K. Mullen, H. I. Kenttamaa, Energy Fuel 2009, 23,
5564–5570.
[18] F. Rakotondradany, H. Fenniri, P. Rahimi, K. L. Gawrys, P. K. Kilpatrick,
M. R. Gray, Energy Fuel 2006, 20, 2439–2447.
[19] K. Akbarzadeh, D. C. Bressler, J. N. Wang, K. L. Gawrys, M. R. Gray, P.
K. Kilpatrick, H. W. Yarranton, Energy Fuel 2005, 19, 1268–1271.
[20] A. B. Andrews, A. McClelland, O. Korkeila, A. Demidov, A. Krummel,
O. C. Mullins, Z. Chen, Langmuir 2011, 27, 6049–6058.
[21] M. A. Francisco, R. Garcia, B. Chawla, C. Yung, K. Qian, K. E. Edwards,
L. A. Green, Energy Fuel 2011, 25, 4600–4605.
[22] E. L. Nordgard, G. Sorland, J. Sjoblom, Langmuir 2010, 26, 2352–2360.
[23] E. L. Nordgard, E. Landsem, J. Sjoblom, Langmuir 2008, 24, 8742–8751.
[24] W. Steinkopf, H. Winternitz, W. Roederer, A. Wolynski, J. Prakt. Chem.
1920, 100, 65–85.
[25] W. Steinkopf, Chem. Ber. 1927, 60, 2609–2610.
[53] Estimation based on analysis as described in: Dynamic NMR
spectroscopy (Ed.: J. Sandström). Academic Press, New York, 1982;
chapters 6–7.
[54] L. Meca, D. Reha, Z. Havlas, J. Org. Chem. 2003, 68, 5677–5680.
[55] D. J. Abdallah, R. G. Weiss, Adv. Mater. 2000, 12, 1237–1247.
[56] For related, chiral steroidal heterocycles, see: S. Gladiali, G. Chelucci,
M. S. Mudadu, M. A. Gastaut, R. P. Thummel, J. Org. Chem. 2001, 66,
400–405.
[57] For related, chiral steroidal heterocycles, see: G. Chelucci, G. A. Pinna,
A. Saba, G. Sanna, J. Mol. Cat. A: Chem. 2000, 159, 423–427.
[58] As measured in the present study, see also: D. H. R. Barton, A. da S.
Campos-Neves, R. C. Cookson, J. Chem. Soc. 1956, 3500–3506.
[59] The optical rotation for 5 h was not measured due to a lack of
sample.
[26] N. D. Zelinsky, Chem. Ber. 1927, 60, 1793–1800.
[27] N. D. Zelinsky, Chem. Ber. 1928, 66, 1291–1294.
[28] K. E. Peters, G. L. Scheuerman, C. Y. Lee, J. M. Moldowan, R. N. Reynolds,
M. M. Pena, Energy Fuel 1992, 6, 560–577.
[29] R. P. Philp, Mass Spec. Rev. 1985, 4, 1–54.
[30] H. P. Nytoft, Org. Geochem. 2011, 42, 520–539.
[31] J. Riolo, B. Ludwig, P. Albrecht, Tetrahedron Lett. 1985, 26, 2697–2700.
[32] A. C. Greiner, C. Spyckerelle, P. Albrecht, Tetrahedron 1976, 32, 257–260.
[33] For a review of steroidal heterocycles, see: P. de Ruggieri, C. Gandolfi,
D. Chiaramonti, Gazz. Chim. Ital. 1962, 92, 768–798.
[34] For a review of steroidal heterocycles, see: G. I. Zhungietu,
G. N. Dorofeenko, Russ. Chem. Rev. 1967, 36, 24–37.
[35] For examples of steroidal heterocycles, see: H. Antaki, V. Petrow,
J. Chem. Soc. 1951, 901–904.
[60] H. R. Snyder, C. T. Elston, D. B. Kellom, J. Am. Chem. Soc. 1953, 75,
[36] For examples of steroidal heterocycles, see: T. C. Miller, J. Hetercyclic
Chem. 1966, 3, 338–344.
[37] For examples of steroidal heterocycles, see: A. Hassner, M. J. Haddadin,
J. Org. Chem. 1962, 27, 1911–1914.
2014–2015.
[61] Y. Yang, S. Ji, F. Zhou, J. Zhao, Biosens. Bioelectron. 2009, 24,
3442–3447.
[62] B. Lygo, J. Crosby, T. Lowdon, T. G. Wainwright, Tetrahedron 1999,
[38] For examples of steroidal heterocycles, see: J.-Z. Yan, J. Li, G.-W. Rao,
Acta Crystallogr., Sect. E: Struct. Rep. Online 2006, 62, O2773–O2774.
[39] For examples of steroidal heterocycles, see: B. R. Venepalli, L. D.
Aimone, K. C. Appell, M. R. Bell, J. A. Dority, R. Goswami, P. L. Hall,
55, 2795–2810.
[63] Wittig–reaction according to the general protocol reported by R.
Antonioletti, F. Bonadies, A. Ciammaichella, A. Viglianti, Tetrahedron
2008, 64, 4644–4648.
wileyonlinelibrary.com/journal/poc
Copyright © 2012 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. (2012)