Nucleophilic Substitutions of Alcohols in High Levels of Catalytic Efficiency

Article abstract of DOI:10.1021/acs.orglett.8b01023

A practical method for the nucleophilic substitution (S_N) of alcohols furnishing alkyl chlorides, bromides, and iodides under stereochemical inversion in high catalytic efficacy is introduced. The fusion of diethylcyclopropenone as a simple Lewis base organocatalyst and benzoyl chloride as a reagent allows notable turnover numbers up to 100. Moreover, the use of plain acetyl chloride as a stoichiometric promotor in an invertive S_N-type transformation is demonstrated for the first time. The operationally straightforward protocol exhibits high levels of stereoselectivity and scalability and tolerates a variety of functional groups.

Full text of DOI:10.1021/acs.orglett.8b01023

Letter
pubs.acs.org/OrgLett
Cite This: Org. Lett. XXXX, XXX, XXX−XXX
Nucleophilic Substitutions of Alcohols in High Levels of Catalytic
Efficiency
Tanja Stach, Julia Drager, and Peter H. Huy*
̈
Institute of Organic Chemistry, Saarland University, P.O. Box 151150, D-66041 Saarbruecken, Germany
S
* Supporting Information
ABSTRACT: A practical method for the nucleophilic
substitution (SN) of alcohols furnishing alkyl chlorides,
bromides, and iodides under stereochemical inversion in
high catalytic efficacy is introduced. The fusion of
diethylcyclopropenone as a simple Lewis base organocatalyst
and benzoyl chloride as a reagent allows notable turnover
numbers up to 100. Moreover, the use of plain acetyl chloride
as a stoichiometric promotor in an invertive SN-type
transformation is demonstrated for the first time. The
operationally straightforward protocol exhibits high levels of stereoselectivity and scalability and tolerates a variety of functional
groups.
ucleophilic substitutions (SN) at sp3-hybridized carbon
centers, in particular those employed on alcohol starting
Scheme 1. Comparison of This Work with Previous Protocols
N
materials, are some of the most fundamental and widely used
chemical transformations.1 Especially, bimolecular nucleophilic
substitutions (SN2), which proceed under stereochemical
Walden inversion, provide a straightforward access to a variety
of important functional groups, such as alkyl halides, amines, and
ethers. To address drawbacks such as (1) low levels of functional
group compatibility and (2) stereoselectivity, (3) poor
sustainability2 and (4) low cost-efficiency, there is a high
demand for the development of more effective protocols for SN-
type conversions that proceed under chemical inversion.3
Despite the significance of invertive SN-reactions in general,
catalytic procedures have only been introduced very re-
cently.4−11 The first example of a catalytic approach, which
engages a Lewis base catalyst with a carbonyl group as the key
structural motif, has been reported by Lambert and co-workers
(Scheme 1A).4a Therein, a cyclopropenone derivative promoted
the transformation of alcohols 1 into alkyl chlorides 2.
Subsequently, Nguyen’s group introduced tropone as a potent
catalyst (Scheme 1B).6a Afterward, our group established an
efficient catalytic method for conversions of type 1 → 2, which is
based on formamide catalysts like N-formylpyrrolidine (FPyr)
and utilizes benzoyl chloride (BzCl) as reagent (Scheme 1C).8
The application of weakly electrophilic BzCl allowed for a
significant improvement of the functional group tolerance, cost
efficiency, and sustainability because weakly acidic benzoic acid is
formed as byproduct instead of strongly acidic HCl. However, a
drawback of all approaches so far is a relatively low catalytic
efficiency. This is evidenced by (1) high catalyst loadings
(typically 10−20 mol %) and hence low turnover numbers
(TON ≤ 10),12 (2) high reaction temperatures of up to 80 °C,
and (3) the fact that acetyl chloride (AcCl) could not be
employed as a reagent so far.8,13 Indeed, the low molecular
weight of AcCl would facilitate a significant enhancement of the
atom efficiency in comparison to conventional reagents such as
BzCl, oxalyl and thionyl chloride (SOCl2), and phosgene
(COCl2). Against this background, we envisioned that these
severe disadvantages could be overcome when carboxylic acid
chloride reagents like BzCl and AcCl are merged with
cyclopropenones as strong Lewis base catalysts.14,15
Herein, we present a diethylcyclopropenone catalyzed
protocol for the conversion of alcohols 1 into chloro, bromo
Received: March 30, 2018
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.8b01023
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
7). The higher electrophilicity of AcCl in comparison to BzCl
could explain why less active formamide catalysts like FPyr are
unable to promote dehydroxychlorination with AcCl (7% yield
for 21, entry 8). The increased reactivity also called for higher
catalyst loadings (10 mol %) to minimize formation of acetate
31b. Interestingly, triphenylphosphine oxide (TPPO) and
tropone, which have been reported as potent catalysts for the
conversion 1 → 2 mediated by oxalyl chloride,6,7 failed to deliver
21 when either AcCl or BzCl were applied as reagents (entries 9
and 10 and ref 8). As illustrated in Scheme 2, the combination of
Table 1. Method Development Benzylic Substrates
a
Scheme 2. Substrate Scope Utilizing Benzoyl Chloride
a
1
Ratio 21/31 determined from the H NMR of the crude material.
b
c
Yields refer to isolated material. Yields were determined by internal
d
e
NMR standard. Reaction performed at 40 °C. Reaction conducted
in dioxane.
and iodo alkanes of type 2, 6 and 7, respectively, with BzCl and
AcCl, respectively (Scheme 1D). The method development was
commenced with the simple model substrate 11, BzCl, and
commercially available diphenylcyclopropenone as catalyst
(DPC, Table 1). A primary solvent screen revealed environ-
mentally friendly MTBE2c,d as optimal (see Table S1 in the
Supporting Information, (SI)), and only 2.5 mol % of DPC were
necessary for the formation of benzylic chloride 21 in 88% yield
(entry 1). In addition, an excellent level of chemoselectivity was
attained, which is reflected in the ratio of the chloro alkane 21 to
the undesired ester 31a of 95:5. In fact, reaction of alcohol 11 with
BzCl and AcCl, respectively, without an appropriate catalyst,
exclusively provides esters of type 31a and 31b, (21/31 ≤ 2:98,
entries 6 and 11).
Next, we discovered that DPC could be replaced by simplified
diethylcyclopropenone (DEC), which resulted in an enhance-
ment of the chemoselectivity under otherwise identical
conditions (21/31a 95:5 → ≥98:2, entry 2). Fortunately, the
synthesis of DEC according to Gundersen’s group,16a which
furnished this catalyst reproducibly in multigram quantities and
yields >80% (see SI, section 4.3.1), is significantly higher yielding
than that of DPC (44%).16b Remarkably, the catalyst loading
could be minimized to 1 mol %, still affording benzylic chloride
21 in 90% isolated yield (entry 3).16c A comparison reaction with
1 mol % FPyr,8 which gave 21 as minor side product in 7% yield,
revealed the strongly improved catalytic efficiency (entry 4).
Even at levels as low as 0.5 mol % DEC, an excellent selectivity
21/31a of 97:3 was preserved, although full conversion of 11 was
not achieved within 24 h of reaction time (entry 5). Remarkably,
with a yield of 87%, this results in a TON of 174.13 To our
delight, DEC was also able to shift chemoselectivity in favor of
substitution product 21 when AcCl was utilized as reactant (entry
a
All yields refer to isolated material if not otherwise specified. (a) With
30 mol % DEC and 1.3 equiv NiBu3. (b) Prepared from the
enantiomeric alcohols (99% ee). ee determined by chiral GC. (c) Yield
determined with internal NMR standard. (d) Reaction conducted in
MeCN. (e) With 20 mol % DPC in dioxane and 2.3 equiv BzCl for
218. (f) Reaction conducted at 80 °C with 20 mol % DEC in dioxane.
DEC and BzCl allowed the effective transformation of a broad
range of primary, secondary, and tertiary alcohols 1 into alkyl
chlorides of type 2. A variety of functional groups was tolerated
under the optimized conditions, which includes acid sensitive
silyl ethers, a tert-butyl ester, and acetals (Scheme 2A). Only in
the case of MOM- and aliphatic TBDMS-ether 24 and 26, which
are highly susceptible toward cleavage through HCl, the reaction
had to be conducted in the presence of a base (N-iBu3).
B
DOI: 10.1021/acs.orglett.8b01023
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
a
In these cases, to outcompete the condensation of 1 with BzCl
(furnishing benzoates of type 3a), which is accelerated by bases,
higher catalyst loadings of 30 mol % had to be applied. Notably,
our previous formamide catalyzed protocol8 did not allow access
to these products in an efficient manner. In the case of the
enantioenriched alcohols 17−19 (99% ee), dehydroxychlorina-
tion furnished the alkyl halides 27−29 in 93−99% ee under
stereochemical inversion (Scheme 2B), which strongly suggests
that these reactions proceeded via a SN2 mechanism.17 Although
a small decrease in enantiopurity was observed in the
transformation of benzylic alcohol S-17, the ee of 93% for the
product R-27 accounts for the highest reported values to date.4a,8
As an important feature, linear allylic alcohols like geraniol E-111
could be converted to the corresponding linear chlorides with
minimal formation of the undesired branched regioisomer
(Scheme 1C). Because chlorinations of E-111 with simple
reagents such as thionyl chloride or phosgene proceed in poor
regioselectivities,8 isomerization sensitive allylic alcohols have
not been included in the substrate scopes of previous catalytic
protocols.4,6,7 Additionally, the production of E-211 on a 27 g
scale with just 4 mol % of DEC highlights the excellent scalability
and operational simplicity of our method, which uses standard
laboratory equipment (volume ≤0.5 L) and does not require
rigorous exclusion of water. Furthermore, BzCl can be
conveniently added via dropping funnel, whereas literature
protocols accomplish addition of the reagent as solution via
syringe pump.4−7 Chlorination of the secondary allylic alcohol b-
Scheme 3. Substrate Scope Using Acetyl Chloride
a
All yields refer to isolated material if not otherwise specified. (a) Yield
determined with internal NMR-standard. (b) Prepared from the
enantiomeric alcohols in (99% ee). ee determined by chiral GC.
amounts. Nevertheless, the superior catalytic efficiency of
cyclopropenones was demonstrated by a comparison reaction
with FPyr,8 which did not furnish the product 213 at all. In
alignment to Lambert’s4 and our previous work,8 a mechanism is
proposed in Scheme 4A.19 Initially, cyclopropenone DEC is
1
12 furnished the branched allylic chloride b-212 predominantly.
Scheme 4. (A) Tentative Mechanism and (B) Proof of
Intermediates II
a
Thus SN2-type attack of chloride is favored over SN2′ in moderate
regioselectivities.
While literature protocols demand high reaction temperatures
of up to 80 °C for the transformation of aliphatic alcohols 1,4−6,8
cyclopropenone catalysis enables milder conditions ≤40 °C
(Scheme 1D). For example, dodecyl chloride 213 was prepared at
room temperature using just 5 mol % of DPC in good yield,
whereas the same reaction with 20 mol % FPyr did not deliver the
desired substitution product. In the instance of aliphatic
substrates of type 1, DPC emerged to be more efficient than
DEC. This is likely the result of faster decomposition of DEC
compared to DPC under the applied reaction conditions. In the
case of steroid derived diol 117, the sterically less encumbered
primary hydroxyl group could be selectively addressed to furnish
product 217, while an excess of BzCl gave rise to the dichloride
218. Moreover, the challenging phthaloyl protected β-amino
chlorides 219 and 220 could be produced in good yields. In
contrast, amino alcohols carrying other protecting groups and an
indole moiety bearing substrate, which could be converted to the
corresponding chloro alkanes with our previous method,8 were
not feasible (see SI, section 3).18 Both substrate types have not
been included in previous catalytic approaches.4−7 It is
noteworthy that alkyl bromides and iodides 6 and 7, respectively,
are also accessible under Finkelstein-type reaction conditions
using sodium halides in acetone (Scheme 1E).
To our delight, DEC (10 mol %) also enabled the utilization of
AcCl as reagent for the synthesis of a range of benzylic and allylic
chlorides, some of which contain acid-labile groups (Scheme
3A). The preparation of the chiral chlorides R-27 and S-28 in 94−
99% ee was accomplished under inversion of the stereochemistry
from the respective alcohols S-17 and R-18 (99% ee), which
implies a SN2 mechanism again (Scheme 3B). Aliphatic alcohols,
on the other hand, could not be accessed in synthetically useful
yields. For instance, dodecyl chloride 213 was obtained in a
moderate yield of 46% alongside with dodecyl acetate in similar
a
(a) Corresponds to the ratio of II, starting material 1 and benzoate 3a
1
according to H NMR. (b) Chemical shift difference in relation to 1.
expected to be acylated with either BzCl or AcCl to furnish the
charged intermediate I. Subsequently, the hydroxyl group of
alcohol 1 would be activated to an enhanced leaving group by
means of intermediate I, which delivers cyclopropenium ion II.
Finally, SN displacement of the catalyst moiety through the
chloride counterion allows for the formation of product 2 and the
1
catalyst. While Lambert already proved intermediate II by H
NMR, we were able to stabilize and fully characterize this
transient species by means of NMR spectroscopy after anion
exchange with non-nucleophilic hexafluorophosphate (Scheme
1
4B). Apparent from the H NMR spectrum of II is the large
difference of the chemical shift of the multiplets of the CH2 group
bound to oxygen Δδ of approximately 1 ppm relative to the
starting material 1. The conjunction of the cyclopropenylium
fragment and the alkyl group of the starting material could be
C
DOI: 10.1021/acs.orglett.8b01023
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
unambiguously proven by 3JC,H couplings in the 2D-NMR. The
suggested intermediate I could not be verified by this approach.
In summary, the amalgamation of cyclopropenone type
catalysts and carboxylic acid chloride reagents facilitated the
straightforward transformation of alcohols 1 into chloro, bromo,
and iodo alkanes of type 2, 6, and 7 in high catalytic efficacy.
Remarkably, the present protocol allows (1) catalyst loadings as
low as 1 mol % and (2) application of AcCl as reagent for the first
time in an SN process under inversion of the stereochemistry.
Additionally, the novel procedure is distinguished by several
other important advantages such as high levels of (3)
compatibility with acid sensitive functional groups, (4) stereo-
and regioselectivity, and (5) scalability. In view of these
significant enhancements and the operational simplicity, our
novel protocol is expected to have a significant impact on
synthetic chemistry laboratories in academia and industry.
Lambert, T. H. Org. Lett. 2011, 13, 740−743. (d) Li, L.; Ni, C.; Wang, F.;
Hu, J. Nat. Commun. 2016, 7, 13320.
(6) For the application of tropone in SN-protocols, see: (a) Nguyen, T.
V.; Bekensir, A. Org. Lett. 2014, 16, 1720−1723. (b) Nguyen, T. V.;
Lyons, D. J. M. Chem. Commun. 2015, 51, 3131−3134.
(7) For triphenylphosphine catalyzed nucleophilic substitution
approaches, see: (a) Denton, R. M.; An, J.; Adeniran, B. Chem. Commun.
2010, 46, 3025−3027. (b) Denton, R. M.; An, J.; Adeniran, B.; Blake, A.
J.; Lewis, W.; Poulton, A. M. J. Org. Chem. 2011, 76, 6749−6767.
(8) (a) Huy, P. H.; Motsch, S.; Kappler, S. M. Angew. Chem., Int. Ed.
2016, 55, 10145−10149 Angew. Chem. 2016, 128, 10300−10304.
(b) Huy, P. H.; Filbrich, I. Chem.−Eur. J. 2018, DOI: 10.1002/
chem.201800588.
(9) For variants of the Mitsunobu reaction that are promoted by Lewis
base catalysts (5−10 mol % typically), see: (a) But, T. Y. S. B.; Toy, P. H.
J. Am. Chem. Soc. 2006, 128, 9636−9637. (b) O’Brien, C. J. PCT Int.
Appl. WO2010/118042A2, 2010. (c) Hirose, D.; Taniguchi, T.;
Ishibashi, H. Angew. Chem., Int. Ed. 2013, 52, 4613−4617 Angew.
Chem. 2013, 125, 4711−4715. (d) Hirose, D.; Gazvoda, M.; Kosmrlj, J.;
Taniguchi, T. Chem. Sci. 2016, 7, 5148−5159. (e) Buonomo, J. A.;
Aldrich, C. C. Angew. Chem., Int. Ed. 2015, 54, 13041−13044. Angew.
Chem. 2015, 127, 13233−13236 (f) Hirose, D.; Gazvoda, M.; Kosmrlj,
J.; Taniguchi, T. Org. Lett. 2016, 18, 4036−4039.
(10) For selected reviews on unimolecular nucleophilic substitution
(SN1) protocols, see: (a) Emer, E.; Sinisi, R.; Capdevila, M. G.;
Petruzziello, D.; De Vincentiis, F.; Cozzi, P. G. Eur. J. Org. Chem. 2011,
2011, 647−666. (b) Naredla, R. R.; Klumpp, D. A. Chem. Rev. 2013, 113,
6905−6948. (c) Kumar, R.; Van der Eycken, E. V. Chem. Soc. Rev. 2013,
́
42, 1121−1146. (d) Baeza, A.; Najera, C. Synthesis 2014, 46, 25−34.
(e) Dryzhakov, M.; Richmond, E.; Moran, J. Synthesis 2016, 48, 935−
959.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.orglett.8b01023.
Optimization data, experimental procedures, character-
ization of new compounds, and spectral data (PDF)
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: peter.huy@uni-saarland.de
ORCID
(11) For a striking Brønsted acid catalyzed cyclodehydration protocol
under SN2 inversion, see Bunrit, A.; Dahlstrand, C.; Olsson, S. K.; Srifa,
Peter H. Huy: 0000-0002-5099-5988
Notes
P.; Huang, G.; Orthaber, A.; Sjoberg, P. J. R.; Biswas, S.; Himo, F.;
̈
Samec, J. S. M. J. Am. Chem. Soc. 2015, 137, 4646−4649.
(12) The lowest catalyst loading has been reported by Denton in ref 7b.
2-Chlorooctane could be prepared in 82% yield (determined via internal
standard) utilizing 3 mol % of triphenylphosphineoxide as catalyst,
which corresponds to a TON of 27.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
(13) Only certain tertiary benzylic and allylic alcohols can be efficiently
converted with carboxylic acid chlorides (e.g., AcCl) to alkyl chlorides:
(a) Bachmann, W. E.; Hauser, C. R.; Hudson, B. E., Jr. Org. Synth. 1943,
23, 100−104. (b) Kishali, N.; Polat, M. F.; Altundas, R.; Kara, Y. Helv.
Chim. Acta 2008, 91, 67−71.
(14) For the initial discovery of cyclopropenones, see: (a) Breslow, R.;
Haynie, R. R.; Mirra, J. J. Am. Chem. Soc. 1959, 81, 247−248.
(b) Kursanov, D. N.; Volpin, M. E.; Koreshkov, Y. D. Izv. Akad. Nauk
SSSR, Otd. Khim. Nauk 1959, 560.
(15) For reviews on cyclopropenium ions and cyclopropenones, see:
(a) Krebs, A. W. Angew. Chem., Int. Ed. Engl. 1965, 4, 10−22 Angew.
Chem. 1965, 77, 10−22. (b) Potts, K. T.; Baum, J. S. Chem. Rev. 1974,
74, 189−213. (c) Komatsu, K.; Kitagawa, T. Chem. Rev. 2003, 103,
1371−1428.
We thank the German Research Foundation (DFG) and the
Fonds of the Chemical Industry (Liebig fellowship for P.H., and
Ph.D. fellowship for T.S.) for generous support.
REFERENCES
■
(1) (a) Larock, R. C. Comprehensive Organic Transformations; Wiley-
VCH: New York, 1999; pp 689−702. (b) Margaretha, P. Science of
Synthesis; Schaumann, E., Ed.; Georg Thieme: Stuttgart, 2007; Vol. 35,
pp 1−188. (c) Smith, M. B. March’s Advanced Organic Chemistry, 7th ed.;
Wiley: Hoboken, NJ, 2013; pp 500−503. (d) An, J.; Denton, R. M.;
Lambert, T. H.; Nacsa, E. D. Org. Biomol. Chem. 2014, 12, 2993−3003.
(e) Huy, P. H.; Hauch, T.; Filbrich, I. Synlett 2016, 27, 2631−2636.
(2) Selected reviews about sustainability in chemistry: (a) Sheldon, R.
A. Green Chem. 2007, 9, 1273−1283. (b) Li, C. J.; Trost, B. M. Proc. Natl.
Acad. Sci. U. S. A. 2008, 105, 13197−13202. (c) Newhouse, T.; Baran, P.
S.; Hoffmann, R. W. Chem. Soc. Rev. 2009, 38, 3010−3021. (d) Anastas,
P.; Eghbali, N. Chem. Soc. Rev. 2010, 39, 301−312. (e) Sheldon, R. A.
Chem. Soc. Rev. 2012, 41, 1437−1451. (f) Dunn, P. J. Chem. Soc. Rev.
2012, 41, 1452−1461.
(16) (a) Netland, K. A.; Gundersen, L.-L.; Rise, F. Synth. Commun.
2000, 30, 1767−1777 For selected examples for the production of
diarylcyclopropenones, see. (b) Breslow, R.; Posner, J. Org. Synth. 1967,
47, 62−63. (c) West, R.; Zecher, D. C.; Goyert, W. J. Am. Chem. Soc.
1970, 92, 149−154.
(17) The substitutions in the case of products 22 and 216 are more likely
to follow an SN1 pathway.
(3) Constable, D. J. C.; Dunn, P. J.; Hayler, J. D.; Humphrey, G. R.;
Leazer, J. L., Jr.; Linderman, R. J.; Lorenz, K.; Manley, J.; Pearlman, B. A.;
Wells, A.; Zaks, A.; Zhang, T. Y. Green Chem. 2007, 9, 411−420.
(4) For cyclopropenone catalyzed SN2-type protocols, see: (a) Vanos,
C. M.; Lambert, T. H. Angew. Chem., Int. Ed. 2011, 50, 12222−12226
Angew. Chem. 2011, 123, 12430−12434. (b) Nacsa, E. D.; Lambert, T.
H. Org. Lett. 2013, 15, 38−41.
(5) For cyclopropenone and 1,1-dihalogenocyclopropene mediated
nucleophilic substitutions, see: (a) Kelly, B. D.; Lambert, T. H. J. Am.
Chem. Soc. 2009, 131, 13930−13931. (b) Hardee, D. J.; Kovalchuke, L.;
Lambert, T. H. J. Am. Chem. Soc. 2010, 132, 5002−5003. (c) Kelly, B. D.;
(18) A screening of various additives in the model reaction 11 → 21
according to the robustness screen of Glorius uncovered that certain H-
bond-donor function can hamper successful chlorination; see SI, section
3 and (a) Collins, K. D.; Glorius, F. Nat. Chem. 2013, 5, 597−601.
(b) Collins, K. D.; Glorius, F. Tetrahedron 2013, 69, 7817−7825.
(19) For a recent theoretical study on cyclopropenone catalyzed
chlorinations according to ref 4a, see: Tataroglu, M. M.; Sungur, F. A. J.
Mol. Graphics Modell. 2017, 77, 106−114.
D
DOI: 10.1021/acs.orglett.8b01023
Org. Lett. XXXX, XXX, XXX−XXX