Studies directed toward the exploitation of vicinal diols in the synthesis of (+)-nebivolol intermediates

Article abstract of DOI:10.3762/bjoc.13.56

While the exploitation of the Sharpless asymmetric dihydroxylation as the source of chirality in the synthesis of acyclic molecules and saturated heterocycles has been tremendous, its synthetic utility toward chiral benzo-annulated heterocycles is relatively limited. Thus, in the search for wider applications of Sharpless asymmetric dihydroxylation-derived diols for the synthesis of benzo-annulated heterocycles, we report herein our studies in the asymmetric synthesis of (R)-1-((R)-6-fluorochroman-2-yl)ethane-1,2-diol, (R)-1-((S)-6-fluorochroman-2-yl)ethane-1,2-diol and (S)-6-fluoro-2-((R)-oxiran-2-yl)chroman, which have been used as late-stage intermediates for the asymmetric synthesis of the antihypertensive drug (S,R,R,R)-nebivolol. Noteworthy is that a large number of racemic and asymmetric syntheses of nebivolol and their intermediates have been described in the literature, however, the Sharpless asymmetric dihydroxylation has never been employed as the sole source of chirality for this purpose.

Full text of DOI:10.3762/bjoc.13.56

Studies directed toward the exploitation of vicinal diols in the
synthesis of (+)-nebivolol intermediates
Runjun Devi and Sajal Kumar Das*
Letter
Open Access
Address:
Beilstein J. Org. Chem. 2017, 13, 571–578.
Department of Chemical Sciences, Tezpur University, Napaam,
Tezpur, Assam-784028, India
doi:10.3762/bjoc.13.56
Received: 04 January 2017
Accepted: 27 February 2017
Published: 21 March 2017
Email:
Sajal Kumar Das* - sajalkdas@gmail.com
* Corresponding author
Associate Editor: T. P. Yoon
Keywords:
© 2017 Devi and Das; licensee Beilstein-Institut.
License and terms: see end of document.
dihydroxylation; epoxide-ring opening; heterocycles; nebivolol;
syn-epoxide
Abstract
While the exploitation of the Sharpless asymmetric dihydroxylation as the source of chirality in the synthesis of acyclic molecules
and saturated heterocycles has been tremendous, its synthetic utility toward chiral benzo-annulated heterocycles is relatively
limited. Thus, in the search for wider applications of Sharpless asymmetric dihydroxylation-derived diols for the synthesis of
benzo-annulated heterocycles, we report herein our studies in the asymmetric synthesis of (R)-1-((R)-6-fluorochroman-2-yl)ethane-
1,2-diol, (R)-1-((S)-6-fluorochroman-2-yl)ethane-1,2-diol and (S)-6-fluoro-2-((R)-oxiran-2-yl)chroman, which have been used as
late-stage intermediates for the asymmetric synthesis of the antihypertensive drug (S,R,R,R)-nebivolol. Noteworthy is that a large
number of racemic and asymmetric syntheses of nebivolol and their intermediates have been described in the literature, however,
the Sharpless asymmetric dihydroxylation has never been employed as the sole source of chirality for this purpose.
Findings
Chiral chromans are prevalent components of a large number of be a potent β1-adrenergic receptor blocker [2,3]. On the other
natural products, natural product-like molecules and pharma- hand, the corresponding enantiomeric form, (R,S,S,S)-nebivolol
ceutical drugs, possessing diverse biological activities [1]. In or (−)-nebivolol (1b, Figure 1) was found to be devoid of
view of their wide spectrum of biological profiles, chiral chro- β1-antagonist activity, but it showed a significant synergistic
mans have become attractive synthetic targets in academia and effect on the antihypertensive proficiency of the (S,R,R,R)
pharmaceutical industry [1]. Nebivolol (1, Figure 1) is a isomer [6-9].
chroman-based antihypertensive drug that was first reported in
the racemic form [2,3]. Chiral HPLC was subsequently em- A number of syntheses of nebivolol and its intermediates have
ployed to access various stereoisomers of 1 in enantiomerically been described in the literature [10-19]. Most of these have
pure form [4,5]. Out of the ten possible stereoisomers of 1, demonstrated that the synthesis of 1a could be achieved using
(S,R,R,R)-nebivolol or (+)-nebivolol (1a, Figure 1) was found to the 2-substituted chroman derivatives (R)-1-((R)-6-fluo-
571

Beilstein J. Org. Chem. 2017, 13, 571–578.
Figure 1: The chroman-based antihypertensive drug nebivolol, its biologically active stereoisomers and late-stage intermediates for its synthesis.
rochroman-2-yl)ethane-1,2-diol (2) and (R)-1-((S)-6-fluo- construct 3 (Scheme 1, method 1). For this purpose, the neces-
rochroman-2-yl)ethane-1,2-diol (3) or the corresponding sary epoxide 6 could be obtained from the parent E-allylic
chroman epoxides 4 and 5 as late-stage intermediates. Al- alcohol through Sharpless asymmetric epoxidation (SAE) [11].
though the consensus synthetic strategy for 1a involves the However, the corresponding parent Z-allylic alcohol appears to
convergent assembly of chroman-based key subunits, the ques- be not suitable to provide 7 under SAE conditions [20]. This has
tion of how best to access them remains open. The intramolecu- eliminated the possibility of obtaining 2 via intramolecular
lar ring-opening of enantiomerically pure epoxides by the epoxide ring-opening of 7 (Scheme 1, method 2). Consequently,
phenolic hydroxy group is one of the most popular methods to an alternative pathway involving the Mitsunobu inversion of 9
Scheme 1: Synthetic strategies toward late-stage intermediates of 1a.
572

Beilstein J. Org. Chem. 2017, 13, 571–578.
(obtained by intramolecular epoxide ring-opening of 8 which is For the synthesis of chroman derivative 2, first a base-mediated
the enantiomer of 6) has been followed to obtain 2 (Scheme 1, intramolecular SNAr reaction was envisioned for the aryl C–O
method 3) whereupon the overall yield of the reaction sequence bond formation under transition-metal-free conditions [28-30].
diminished [11].
The additional benefit of this strategy would be the non-require-
ment of any protecting group to construct the chroman ring. To
On the other hand, the Sharpless asymmetric dihydroxylation test this seemingly straightforward approach, we initially
(SAD) has been a workhorse as a synthetic tool for accessing attempted to synthesize (±)-2 utilizing the cyclization of (±)-
enantiopure vicinal diols [21]. The extensive work in this field triol 14 (Scheme 2) as the key step.
has resulted in the discovery of a number of cinchona alkaloid-
derived ligands which allow the dihydroxylation of alkenes of Thus, commercially available 2,5-difluorobenzaldehyde (10)
almost all substitution patterns with high enantioselectivity. was treated with the Wittig reagent Ph3P=CHCO2Et and the re-
Noteworthy is that the SAD is not limited to only E-allylic alco- sulting unsaturated ester was then hydrogenated with Pd–C and
hols in its choice of substrates as is the SAE process. Moreover, H2 at room temperature to obtain compound 11 in 91% yield
the SAD is much more superior in terms of operational over two steps (Scheme 2). DIBAL-H (1.1 equiv, –78 °C)
simplicity as SAE, it can be run at 0 °C in water as a co-solvent reduction of the ester group of 11 followed by Wittig olefina-
and under an atmosphere open to air.
tion of the resulting crude aldehyde with Ph3P=CHCO2Et pro-
vided (E)-α,β-unsaturated ester 12 (80% over two steps). A
The application of SAD-derived vicinal diols in the synthesis of further DIBAL-H (2.5 equiv, 0 °C) reduction of 12 delivered
acyclic molecules and saturated heterocycles has been aston- (E)-allylic alcohol 13 (90%) which was then dihydroxylated
ishing. However, their utilities in the synthesis of chiral benzo- under Upjohn conditions to obtain triol (±)-14 (92%). With
annulated heterocycles are relatively limited [22-26]. In this access to (±)-14 we were in a position to investigate the key
paper, we describe our efforts toward the synthesis of 2, 3 and 5 cyclization involving an intramolecular SNAr to deliver 2.
using different cyclization strategies. To the best of our know- Unfortunately, all attempts of cyclizing (±)-14 to obtain
ledge, nebivolol or its intermediates have never been synthe- chroman derivative (±)-2 under various SNAr reaction condi-
sized using Sharpless asymmetric dihydroxylation as the sole tions were not successful. Treatment of (±)-14 with KOt-Bu/
source of chirality [27].
THF (65 °C), NaH/DMF (80 °C), NaH/DMSO (100 °C) and
Scheme 2: Attempted synthesis of (±)-2 via intramolecular SNAr reaction.
573

Beilstein J. Org. Chem. 2017, 13, 571–578.
KOt-Bu/toluene (110 °C) did not lead to any conversion. How- generated cyclic orthoesters [33]. We speculated that a similar
ever, more forcing conditions such as NaH/NMP (130 °C) reaction on an appropriately positioned diol with a tethered
resulted in a to partial decomposition of the starting material. o-hydroxyphenyl group might produce 2-substituted chroman
These results indicated that an intramolecular SNAr reaction of derivatives via a 6-exo-tet cyclization (Scheme 3).
triol (±)-14 to form (±)-2 is not feasible. Thus, the presence of
an activating substituent (e.g., a nitro group) at the C-5 position To substantiate this hitherto unexplored approach in the context
of the benzene ring might be helpful in synthesizing molecules of synthesizing (+)-nebivolol intermediates, we first needed to
similar to 2 [31,32].
synthesize the syn-dihydroxy esters 19 and 20 (Scheme 4).
Toward that objective, 2-allyl-4-fluorophenol (15) was benzyl-
The failure to achieve an intramolecular cyclization of the diol ated with BnCl and anhydrous K2CO3 in the presence of KI in
via an SNAr reaction caused us to investigate other cyclization acetone under reflux conditions to obtain benzyl ether 16
approaches towards these chroman derivatives. In 2005, Borhan (Scheme 4). The subsequent hydroboration of the allyl group in
and co-workers described the construction of tetrahydrofuran 16 with 9-BBN and the oxidation of the resulting organoborane
and tetrahydropyran structures from 1,2,n-triols via an elegant with NaOH and H2O2 furnished alcohol 17 in 96% yield. A
cyclization involving Lewis acid-mediated cyclization of in situ one-pot PCC oxidation–Wittig olefination (with
Scheme 3: Speculation on the synthesis of a 2-substituted chroman derivative based on Borhan’s approach.
Scheme 4: Synthesis of syn-2,3-dihydroxy esters 19 and 20.
574

Beilstein J. Org. Chem. 2017, 13, 571–578.
Ph3P=CHCO2Et) of 17 provided (E)-α,β-unsaturated ester 18 in obtain the β-hydroxy-α-tosyloxy esters 24 and 25, respectively
85% yield over 2 steps. Compound 18 was then subjected to a (Scheme 6).
Sharpless asymmetric dihydroxylation with AD-mix-α in
t-BuOH/H2O (1:1) at 0 °C for 24 h furnishing syn-2,3-di-
hydroxy ester 19 in a high yield of 92%. For the synthesis of
syn-2,3-dihydroxy ester 20, AD-mix-β was employed.
Debenzylation of diol 20 with Pd–C and H2 at room tempera-
ture produced compound 21 having a tethered o-hydroxyphenyl
group (Scheme 5). Next, compound 21 was exposed to
Borhan’s reaction conditions with the hope to obtain 22. To our
dismay, however, the reaction furnished the hydrolyzed prod-
uct 23 instead of cyclized product 22. It is important to mention
that strict anhydrous conditions were maintained for this trans-
Scheme 6: Synthesis of β-hydroxy-α-tosyloxy esters 24 and 25.
formation in order to prevent the nucleophilic attack of H2O
leading to the formation of 23. However, the formation of 23
instead of 22 clearly indicates that the in situ generated cyclic Panda and co-worker applied a three-step reaction sequence in-
orthoester, after getting activated by Lewis acid, did not experi- volving epoxidation/debenzylation/epoxide ring-opening to
ence a nucleophilic attack of the phenolic hydroxy group; convert the β-hydroxy-α-tosyloxy ester into the corresponding
instead it reacted with water during the work-up process. This 2-substituted chroman derivative. However, it has been
different outcome of this reaction compared to Borhan’s results reviewed that not only benzylic epoxides but also non-benzylic
might be attributed to the lower nucleophilicity of phenols com- epoxides are sensitive to the standard hydrogenation/debenzyla-
pared to alcohols.
tion conditions [35]. Whereas benzylic epoxides are highly
sensitive to hydrogenation conditions, non-benzylic epoxides,
Previously, Panda and co-worker converted succesfully a syn- depending on the reaction conditions, may produce traces to
2,3-dihydroxy ester into a 2-substituted chroman derivative significant amounts of side-products via hydrogenolysis. Thus,
[23]. The above-described unfortunate failures eventually we decided to modify Panda’s synthetic route to significantly
forced us to turn our attention to utilize this methodology for increase the overall yield. We hypothesized that this problem
the synthesis of 2-substituted chroman derivatives. Thus, diols might be circumvented by performing the debenzylation reac-
19 and 20 were subjected to a monotosylation reaction [34] to tion prior to the epoxide-ring formation. Further we speculated
Scheme 5: Attempted cyclization according to Borhan’s method.
575

Beilstein J. Org. Chem. 2017, 13, 571–578.
that compound 26, in the presence of a base, might undergo a cific rotations of 2 and 29 matched with those reported in the
simultaneous epoxidation–intramolecular epoxide-ring opening literature [11,14,15]. For the synthesis of compound 3, stereo-
to produce 27 (Scheme 7) as the corresponding benzoxepin ring isomer 29 was subjected to classical two-step Mitsunobu inver-
formation via intramolecular displacement of –OTs group by sion protocol which was successful but poor yielding
ArO− is unresponsive [23].
(Scheme 9).
Not surprised by the poor yield of this transformation, we
focused on the conversion of 28 into 5 which has also been used
as a late-stage nebivolol intermediate (5 is a more advanced
intermediate compared to 3). Thus, the tosylation of compound
28 followed by LiBH4 reduction of the resulting tosylate and
subsequent epoxidation of the obtained tosyloxy alcohol with
anhydrous K2CO3 in absolute ethanol (Scheme 10) provided
compound 5.
Scheme 7: Speculation of simultaneous epoxidation/epoxide-ring
opening.
To test this hypothesis, first compound 24 was subjected to the
debenzylation reaction with 10% Pd–C in EtOH under an H2 at-
mosphere (Scheme 8) at room temperature. After completion of
the debenzylation process, K2CO3 was added to the reaction
mixture in the same reaction vessel. The reaction mixture, after
being run for additional 6 h, provided compound 27 in 70%
yield which is significantly higher than the literature yield
(53%) for the similar transformation [23]. The similar strategy
was applied in converting 25 into chroman derivative 28. Next,
Scheme 10: Synthesis of chroman epoxide 5.
LiAlH4 reduction of 27 and 28 provided 2 and 29, respectively, The NMR spectra and specific rotations of 5 also matched those
in 93% yield. It is to be mention that the NMR spectra and spe- reported in the literature [17-19,36].
Scheme 8: Synthesis of chroman diols 2 and 29, respectively.
Scheme 9: Conversion of 32 into 3 via Mitsunobu inversion.
576

Beilstein J. Org. Chem. 2017, 13, 571–578.
7. Van de Water, A.; Xhonneux, R. S.; Reneman, R. S.; Janssen, P. A. J.
Eur. J. Pharmacol. 1988, 156, 95–103.
Conclusion
In summary, in the context of exploiting Sharpless asymmetric
dihydroxylation-derived vicinal diols in the synthesis of (R)-1-
((R)-6-fluorochroman-2-yl)ethane-1,2-diol, (R)-1-((S)-6-fluo-
rochroman-2-yl)ethane-1,2-diol and (S)-6-fluoro-2-((R)-oxiran-
2-yl)chroman, which have previously utilized as late-stage
intermediates for the synthesis of (S,R,R,R)-nebivolol, we have
extensively studied different cyclization strategies. The
construction of 2-substituted chroman derivatives using
phenolic hydroxy-mediated intramolecular ring opening of syn-
2,3-diol ester-derived cyclic orthoester or intramolecular SNAr
reaction of a triol containing a tethered 2,5-difluorophenyl sub-
stituent were not successful. However, the exposure of
β-hydroxy-α-tosyloxy esters to a one-pot, three-step process
(debenzylation–epoxidation–intramolecular epoxide ring
opening) enabled us to acheive the target molecules. To the best
of our knowledge, this is the first use of the Sharpless asym-
metric dihydroxylation as the sole source of chirality for the
synthesis of nebivolol intermediates.
doi:10.1016/0014-2999(88)90151-3
8. Xhonneux, R.; Wouters, L.; Reneman, R. S.; Janssen, P. A. J.
Eur. J. Pharmacol. 1990, 181, 261–265.
doi:10.1016/0014-2999(90)90087-M
9. Bowden, C. R.; Marchione, C. S. J. Pharmacol. Exp. Ther. 1989, 251,
599–605.
10.Johannes, C. W.; Visser, M. S.; Weatherhead, G. S.; Hoveyda, A. H.
J. Am. Chem. Soc. 1998, 120, 8340–8347. doi:10.1021/ja981378o
11.Chandrasekhar, S.; Reddy, M. V. Tetrahedron 2000, 56, 6339–6344.
doi:10.1016/S0040-4020(00)00572-X
12.Yang, Y. X.; Wang, N. X.; Xing, Y. L.; Wang, W. W.; Zhao, J.;
Wang, G. X.; Tang, S. Chin. Chem. Lett. 2005, 16, 1577–1580.
13.Yu, A.-G.; Wang, N.-X.; Xing, Y.-L.; Zhang, J.-P.; Yang, Y.-X.;
Wang, W.-W.; Sheng, R.-l. Synlett 2005, 1465–1467.
doi:10.1055/s-2005-868508
14.Bai, Y.; Chen, X. J. Chem. Res. 2006, 807–808.
doi:10.3184/030823406780199802
15.Wang, N.-X.; Yu, A.-G.; Wang, G.-X.; Zhang, X.-H.; Li, Q.-S.; Li, Z.
Synthesis 2007, 1154–1158. doi:10.1055/s-2007-965993
16.Carreño, M. C.; Hernández-Torres, G.; Urbano, A.; Colobert, F.
Eur. J. Org. Chem. 2008, 2035–2038. doi:10.1002/ejoc.200800201
17.Bartoli, S.; Cipollone, A.; Fattori, D. Process for the preparation of
nebivolol. PCT Patent Application WO2011/098474 A1, Aug 18, 2011.
18.Mauro, S.; Fattori, D.; D’Andrea, P.; Cipollone, A. Process for the
preparation of nebivolol. PCT Patent Application WO2012/095707 A1,
July 19, 2012.
Supporting Information
Supporting Information File 1
Experimental procedures, characterization data and copies
of 1H and 13C NMR spectra for final compounds are
available.
19.Jas, G.; Freifeld, I.; Kesseler, K. Method for producing nebivolol. PCT
Patent Application WO2011/091968 A1, Aug 4, 2011.
20.Khandavalli, P. C.; Spiess, O.; Boehm, O. M.; Freifeld, I.; Kesseler, K.;
Jas, G.; Schinzer, D. J. Org. Chem. 2015, 80, 3965–3973.
doi:10.1021/acs.joc.5b00263
[http://www.beilstein-journals.org/bjoc/content/
supplementary/1860-5397-13-56-S1.pdf]
21.Kolb, H. C.; VanNieuwenhze, M. S.; Sharpless, K. B. Chem. Rev.
1994, 94, 2483–2547. doi:10.1021/cr00032a009
22.Manabe, Y.; Kanematsu, M.; Osaka, M.; Yoshida, M.; Shishido, K.
Heterocycles 2014, 88, 441–452. doi:10.3987/COM-13-S(S)50
23.Das, S. K.; Panda, G. Tetrahedron 2008, 64, 4162–4173.
doi:10.1016/j.tet.2008.03.001
Acknowledgements
We acknowledge the financial support provided by the Depart-
ment of Science and Technology [(DST), SB/FT/CS-073/2013],
and the University Grant Commission [F30-33/2014(BSR)],
New Delhi, India.
24.Das, S. K.; Dinda, S. K.; Panda, G. Eur. J. Org. Chem. 2009, 204–207.
doi:10.1002/ejoc.200800661
25.Khadem, S.; Joseph, R.; Rastegar, M.; Leek, D. M.; Oudatchin, K. A.;
Arya, P. J. Comb. Chem. 2004, 6, 724–734. doi:10.1021/cc049941o
26.Kishuku, H.; Yoshimura, T.; Kakehashi, T.; Shindo, M.; Shishido, K.
Heterocycles 2003, 61, 125–131. doi:10.3987/COM-03-S4
27.Previously, Sharpless asymmetric dihydroxylation was used to create
one of the two chiral centers of each chroman unit of nebivolol. See
reference [16] above.
References
1. Shen, H. C. Tetrahedron 2009, 65, 3931–3952.
doi:10.1016/j.tet.2009.02.002
2. De Crée, J.; Geukens, H.; Leempoels, J.; Verhaegen, H.
Drug Dev. Res. 1986, 8, 109–117. doi:10.1002/ddr.430080114
3. Van Lommen, G. R. E.; De Bruyn, M. F. L.; Schroven, M. F. J.
Derivatives of 2,2'-iminobisethanol. U.S. Patent US4,654,362, March
31, 1987.
28.Devi, R.; Gogoi, D.; Bora, P.; Das, S. K. Tetrahedron 2016, 72,
4878–4888. doi:10.1016/j.tet.2016.06.059
See for the construction of the aryl C–O bond of benzo-fused
oxo-heterocycles via intramolecular SNAr reaction between a hydroxy
group and unactivated aryl fluoride; see also references [29,30] below.
29.Singha, R.; Ahmed, A.; Nuree, Y.; Ghosh, M.; Ray, J. K. RSC Adv.
2015, 5, 50174–50177. doi:10.1039/C5RA07532G
4. Pauwels, P. J.; Gommeren, W.; van Lommen, G.; Janssen, P. A. J.;
Leysen, J. E. Mol. Pharmacol. 1988, 34, 843–851.
5. Van Lommen, G.; De Bruyn, M.; Schroven, M. J. Pharm. Belg. 1990,
45, 355–360.
6. Midol-Monet, M.; Davy, M.; Heimburger, M.; Beslot, F.; Cohen, Y.
J. Pharm. Pharmacol. 1991, 43, 504–509.
30.Shi, G. Q.; Dropinski, J. F.; Zhang, Y.; Santini, C.; Sahoo, S. P.;
Berger, J. P.; MacNaul, K. L.; Zhou, G.; Agrawal, A.; Alvaro, R.;
Cai, T.-q.; Hernandez, M.; Wright, S. D.; Moller, D. E.; Heck, J. V.;
Meinke, P. T. J. Med. Chem. 2005, 48, 5589–5599.
doi:10.1111/j.2042-7158.1991.tb03523.x
doi:10.1021/jm050373g
577

Beilstein J. Org. Chem. 2017, 13, 571–578.
31.Zezula, J.; Singer, L.; Przybyl, A. K.; Hashimoto, A.; Dersch, C. M.;
Rothman, R. B.; Deschamps, J.; Lee, Y. S.; Jacobson, A. E.;
Rice, K. C. Org. Biomol. Chem. 2008, 6, 2868–2883.
doi:10.1039/b803433h
See for the construction of the aryl C–O bond of benzo-fused
oxo-heterocycles via intramolecular SNAr reaction between a hydroxy
group and activated aryl fluoride.
32.Kurimura, M.; Liu, H.; Sulima, A.; Hashimoto, A.; Przybyl, A. K.;
Ohshima, E.; Kodato, S.; Deschamps, J. R.; Dersch, C. M.;
Rothman, R. B.; Lee, Y. S.; Jacobson, A. E.; Rice, K. C. J. Med. Chem.
2008, 51, 7866–7881. doi:10.1021/jm800913d
33.Zheng, T.; Narayan, R. S.; Schomaker, J. M.; Borhan, B.
J. Am. Chem. Soc. 2005, 127, 6946–6947. doi:10.1021/ja043002i
34.Fleming, P. R.; Sharpless, K. B. J. Org. Chem. 1991, 56, 2869–2875.
doi:10.1021/jo00008a051
35.Thiery, E.; Le Bras, J.; Muzart, J. Eur. J. Org. Chem. 2009, 961–985.
doi:10.1002/ejoc.200800739
36.Horiguchi, A.; Kuge, Y.; Mochida, K.-i. J. Mol. Catal. B: Enzym. 1997, 3,
285–292. doi:10.1016/S1381-1177(97)00013-1
License and Terms
This is an Open Access article under the terms of the
Creative Commons Attribution License
(http://creativecommons.org/licenses/by/4.0), which
permits unrestricted use, distribution, and reproduction in
any medium, provided the original work is properly cited.
The license is subject to the Beilstein Journal of Organic
Chemistry terms and conditions:
(http://www.beilstein-journals.org/bjoc)
The definitive version of this article is the electronic one
which can be found at:
doi:10.3762/bjoc.13.56
578