New facts concerning the reaction of K-, K+(15-crown-5)2 with phenyl glycidyl ether: Unexpected formation of potassium cyclopropoxide

Article abstract of DOI:10.1016/S0022-328X(99)00444-1

A new mechanism of the reaction of K^-, K⁺(15-crown-5)₂ with phenyl glycidyl ether is presented. The linear ether bond is attacked only to a small extent, if at all. As the main reaction path the oxirane bond in the β-posit

Full text of DOI:10.1016/S0022-328X(99)00444-1

www.elsevier.nl/locate/jorganchem
Journal of Organometallic Chemistry 590 (1999) 153–157
New facts concerning the reaction of K−, K+(15-crown-5)2 with
phenyl glycidyl ether: unexpected formation of potassium
cyclopropoxide
Zbigniew Grobelny a, Andrzej Stolarzewicz a,*, Adalbert Maercker b,
Wolfgang Demuth b
a Centre of Polymer Chemistry, Polish Academy of Sciences, 41-819 Zabrze, Poland
b Institut fu¨r Organische Chemie, Uni6ersita¨t Siegen, D-57068 Siegen, Germany
Received 5 June 1999; accepted 20 July 1999
Abstract
A new mechanism of the reaction of K−, K+(15-crown-5)2 with phenyl glycidyl ether is presented. The linear ether bond is
attacked only to a small extent, if at all. As the main reaction path the oxirane bond in the b-position is cleaved, followed by the
g-elimination of potassium phenoxide and the formation of potassium cyclopropoxide. Crown ether ring opening also occurs in
reactions with organometallic intermediates. © 1999 Elsevier Science S.A. All rights reserved.
Keywords: Potassium anion; Phenyl glycidyl ether; 15-Crown-5; Potassium cyclopropoxide
1. Introduction
at a molar ratio of reagents equal to 1:2. The disappear-
ance of potassium anions occurred at the same mo-
ment, as it was found in Ref. [1]. The reaction mixture
was then treated with benzyl bromide. That enabled the
chromatographic analysis of liquid benzylated products
formed from non-volatile organometallic compounds.
The signals of two main reaction products were
found in the chromatogram of the GC–MS analysis.
One of them was identified as benzyl cyclopropyl ether,
i.e. benzyl derivative of potassium cyclopropoxide, in
28% yield. The other product was found to be benzyl
phenyl ether, e.g. the benzyl derivative of potassium
phenoxide, in 62% yield.
The chromatogram also exhibited small signals due
to allyl benzyl ether, 2-benzyloxypropyl phenyl ether
and tetraethylene glycol benzyl vinyl ether in total yield
of about 10%. The latter was observed earlier by us as
the benzylated product of a crown ring-opening reac-
tion [2].
In our previous paper [1] preliminary results on the
investigation of the phenyl glycidyl ether reaction with
K−, K+(15-crown-5)2 1 were presented. Anisole and
potassium hydroxide as well as propylene and ethylene
were found in the reaction mixture after treating it with
methyl iodide. It was postulated that in the first step the
metal ion attacks the linear ether bond of phenyl
glycidyl ether, leading to the formation of glycidyl-
potassium 2 and potassium phenoxide 3 (Scheme 1).
In the present work we report new experimental
results concerning the study on this process. A new
reaction mechanism is proposed.
2. Results and discussion
The addition of phenyl glycidyl ether into the blue
K−, K+(15-crown-5)2 solution caused its discoloration
The presence of benzyl cyclopropyl ether in the reac-
tion mixture was also proven by NMR spectroscopic
analysis (1H and 13C). All signals of the authentic
sample were found in the NMR spectra of the reaction
* Corresponding author. Fax: +48-32-2712969.
E-mail address: polymer@uranos.cto.us.edu.pl (A. Stolarzewicz)
0022-328X/99/$ - see front matter © 1999 Elsevier Science S.A. All rights reserved.
PII: S0022-328X(99)00444-1

154
Z. Grobelny et al. / Journal of Organometallic Chemistry 590 (1999) 153–157
Scheme 1.
mixture. The presence of potassium hydroxide, propy-
lene and ethylene in the reaction mixture was confirmed
in separate experiments.
The formation of potassium cyclopropoxide 5 as the
main reaction product shows that cleavage of the linear
ether bond (Scheme 1) is not the main reaction path,
because glycidylpotassium 2 cannot rearrange to 5 for
stereoelectronic reasons. Therefore, the oxirane ring
bond is assumed to cleave in the b-position by electron
transfer from K−, followed by formation of 4 and
g-elimination (Scheme 2).
The intermediate 4 is partly trapped by protonation to
9, finally yielding the 2-benzyloxypropyl phenyl ether 11
upon benzylation and tetraethylene glycol benzyl vinyl
ether 12 as the benzylated crown ether ring-opening
product 10. An example for a similar ring closure to a
cyclopropane derivative is known in the literature [3]
(Scheme 3).
That glycidylpotassium 2 does rearrange into potas-
sium allyloxide 13 was indirectly confirmed by an inde-
pendent synthesis of glycidylsodium yielding exclusively
sodium allyloxide. However, the allyloxide 16 obtained
in our reaction starting with phenyl glycidyl ether need
not be formed via glycidylpotassium 2 because cleavage
of the other oxirane ring bond followed by b-elimination
would yield the same product (Scheme 4).
Finally the formation of propylene and potassium
hydroxide has to be mentioned. Propylene is found only
with excess of K−, K+(15-crown-5)2, i.e. when the phenyl
glycidyl ether solution is dropped into the K−, K+(15-
crown-5)2 solution. Scheme 5 shows the most probable
path to propylene.
Scheme 2.
Scheme 3.
Additional K−, K+(15-crown-5)2 is used to cleave the
ether bond of 4 to form 17, which could eliminate KOH
or K2O to give unstable organopotassium intermediates.
Protonation of these intermediates by crown ether finally
yields propylene and 10, which is known to decompose
under the influence of metal anions into ethylene and
potassium glycoxides [2].
Two further paths that could lead to propylene were
excluded by us as shown in Scheme 6. Compound 18,
which we prepared by treatment of propylene oxide with
K−, K+(15-crown-5)2, does not give propylene. Com-
pound 18 might also be formed by cleavage of 9 with K−,
K+(15-crown-5)2, but we did not find the benzyl deriva-
tive of 18. In another experiment we treated 13 with K−,
K+(15-crown-5)2 and did not get any propylene.
It was assumed that another gaseous product, namely
ethylene, is formed in a side reaction of potassium anion
with 10.
In one experiment we dropped the K−, K+(15-crown-
5)2 solution into an excess of phenyl glycidyl ether
solution (molar ratio 1:10). In this case no propylene and
ethylene, as well as potassium hydroxide, were formed
and only a small amount of tetraethylene glycol benzyl
vinyl ether 12 was observed. The other products of ether
cleavage remained unchanged.

Z. Grobelny et al. / Journal of Organometallic Chemistry 590 (1999) 153–157
155
Scheme 4.
Scheme 6.
3. Conclusions
In the reaction of phenyl glycidyl ether with K−, K+
(15-crown-5)2 predominantly the least-substituted CꢀO
bond of the strained oxirane ring is cleaved, corre-
sponding to alkyl glycidyl ethers. In contrast to the lat-
ter, the so-formed dianion can cleave off the relative
stable potassium phenoxide with closure of a cyclo-
propane ring (g-elimination). This reaction is an inter-
esting new route to the anion of cyclopropanol.
Scheme 5. RH is crown ether or other compounds losing hydrogen.
All the reactions are very fast. It was very important
to use benzyl bromide as derivatising agent to detect
the products 5 and 16. The analysis of the samples
treated with methyl iodide as reported in Ref. [1] did
not give the information we needed for the explanation
of the reaction mechanism.
4. Experimental
Gas chromatography–mass spectrometry (GC–MS)
analyses were conducted on a 30 m-long fused silica
column coated with DB 1701 using a Varian 3300 gas
chromatograph equipped with a Finnigan MAT 800

156
Z. Grobelny et al. / Journal of Organometallic Chemistry 590 (1999) 153–157
AT ion trap detector; the mass spectrum of benzyl
cyclopropyl ether was obtained with a Hewlett–
Packard HP 5988 A quadrupole mass spectrometer.
Diethylene glycol dimethyl ether was used as the
4.3. Tetraethylene glycol monobenzyl ether
Tetraethylene glycol monobenzyl ether (HO[CH2-
CH2O]4CH2Ph): a 80% dispersion of NaH (7.8 g contain-
ing 0.26 mol NaH) in paraffin was washed twice with
tert-butyl methyl ether and decanted; the NaH was
suspended in 100 ml tetrahydrofuran and then a mixture
of tetraethylene glycol (48.5 g, 0.25 mol) and 50
ml tetrahydrofuran was added dropwise. After the evo-
lution of hydrogen had stopped, benzylbromide (25.6 g,
0.15 mol) was added and the reaction mixture stirred for
2 h. Water was added, the organic layer separated and
the aqueous phase extracted with tert-butyl methyl
ether. The combined organic phases were dried and the
solvents removed under reduced pressure. The crude
product (37 g) was used in the next step without further
purification.
1
internal standard for the yield measurement. H- and
13C-NMR spectra of benzyl cyclopropyl ether were
taken at 200 MHz on a Bruker AC 200 spectrometer;
those of the reaction mixture were recorded at 20°C on
1
a Varian VXR-300 spectrometer operating at the H
resonance frequency of 300 MHz and the 13C resonance
frequency of 75 MHz.
Starting materials and the reaction procedure were
described in Ref. [1]. In order to identify non-volatile
organometallic compounds, benzyl bromide was added
to the reaction mixture to form liquid products. Allyl
benzyl ether and benzyl phenyl ether (both Aldrich)
were used as model compounds.
4.1. Benzyl cyclopropyl ether (7)
4.4. Tetraethylene glycol benzyl 6inyl ether (11)
Benzyl cyclopropyl ether (7) was prepared according
to a general method of Furukawa et al. [4]. To a stirred
mixture of benzyl vinyl ether [5] (6.7 g, 50 mmol) and
ZnEt2 (4.0 ml, 40 mmol) in 25 ml of dry diethyl ether was
added CH2I2 (17.5 g, 65 mmol) dropwise during 30 min
at room temperature under N2 atmosphere. The reaction
mixture was poured slowly into ice–diluted HCl under
stirring. The organic layer was washed with water and
diluted NaHCO3 solution, and dried over MgSO4. After
evaporation of Et2O, the residue was distilled through a
short column. Several fractions were collected, b.p.
88–92°C (30 mbar), which consisted mainly of benzyl
cyclopropyl ether contaminated with benzyl alcohol and
some benzyl vinyl ether (total 4.2 g). The last of these
fractions, b.p. 92°C (30 mbar) was stirred over night with
lithium metal and redistilled to give pure (\98%) benzyl
Tetraethylene glycol benzyl vinyl ether (11) was pre-
pared by the transetherification method [6]. A stirred
solution of crude tetraethylene glycol monobenzyl ether
(37 g, 0.13 mol), butyl vinyl ether (60 g, 0.60 mol) and
mercury trifluoroacetate (1.2 g, 2.8 mmol) was heated for
1 h under reflux, then anhydrous potassium carbonate
(2 g, 20 mmol) was added and the excess of butyl vinyl
ether was removed under reduced pressure. A sample
of the residue was distilled in a Kugelrohr apparatus;
the fraction boiling at 145°C (0.05 mbar) consisted of
tetraethylene glycol benzyl vinyl ether. 1H-NMR CDCl3
l: 7.34 (m, 5H, C6H5); 6.50 (dd, J=14.4, 6.8 Hz,
1H, OCHꢁ); 4.56 (s, 2H PhꢀCH2); 4.23 (dd, J=14.4,
2.2 Hz, 1H, CH2ꢁ); 3.97 (dd, J=6.8, 2.2 Hz, 1H, CH2ꢁ);
3.63–3.86 (m, 16H, OCH2). 13C-NMR CDCl3 l: 151.7
(OCHꢁ); 138.2, 128.3, 127.7, 127.5 (C6H5); 86.5 (CH2ꢁ);
73.2 (PhꢀCH2); 67.1–70.7 (OCH2, six signals). Mass
spectrum (m/e): 310 (M+, 2); 223 (1); 177 (5); 133
(15); 105 (17); 91 (100); 73 (19); 45 (62); 43 (30).
1
cyclopropyl ether, b.p. 46°C (2 mbar). H-NMR CDCl3
l: 7.30 (m, 5H, C6H5); 4.52 (s, 2H, PhꢀCH2); 3.32 (m,
1H, OCH); 0.63 (m, 2H, CH2 cis); 0.46 (m, 2H, CH2
1
trans). H-NMR C6D6 l: 7.28 (m, 2H, C6H5); 7.15 (m,
3H, C6H5); 4.38 (s, 2H, PhꢀCH2); 3.10 (m, 1H, OCH);
0.58 (m, 2H, CH2 cis); 0.23 (m, 2H, CH2 trans). 13C-
NMR CDCl3 l: 138.17, 128.35, 127.88, 127.60 (C6H5);
72.79 (PhꢀCH2); 53.01 (OCH); 5.69 (CH2). 13C-NMR
C6D6 l: 139.04, 128.30, 127.84, 127.58 (C6H5); 72.65
(PhꢀCH2); 53.08 (OCH); 5.84 (CH2). Mass spectrum
(m/e): 147 (M−1, 0.3); 130 (0.7); 104 (25); 91 (100); 77
(4); 65 (20); 63 (6); 51 (6); 39 (11).
4.5. 2-Benzyloxypropyl phenyl ether (10)
Potassium hydride (0.10 g, 2.5 mmol) and tetrahydro-
furan (10 cm3) were introduced into the reactor. Then,
1-phenoxy-2-propanol (0.61 g, 2.5 mmol) was added
dropwise while stirring at 25°C. The course of the
reaction was followed by measuring the amount of
hydrogen evolved. After 6 h benzyl bromide (0.43 g, 2.5
mmol) was added to the mixture. The precipitated
potassium bromide was filtered off, and 2-benzyl-
oxypropyl phenyl ether (10) was distilled from the
solution; the fraction boiling at 140°C at 0.1 mbar was
4.2. Glycidylsodium
Glycidylsodium was prepared as an unstable interme-
diate from bromomethyloxirane and sodium in tetrahy-
drofuran solution at room temperature. It could not be
detected itself, but decomposed spontaneously to sodium
allyloxide.
1
collected in 80% yield. H-NMR acetone-d6 l:7.5–6.9
(m, 10H, Ph+OPh); 4.64 (s, 2H, OCH2Ph); 4.10–3.98
(m, 1H, OCH); 3.96–3.83 (m, 2H, CH2OPh); 1.27 (d,
J=6.1 Hz, 3H, CH3). 13C-NMR acetone-d6 l: 159.9

Z. Grobelny et al. / Journal of Organometallic Chemistry 590 (1999) 153–157
157
(COPh ipso); 140.1 (CPh ipso); 130.2 (COPh meta); 128.9
(CPh meta); 128.2 (CPh ortho); 128.0 (CPh para); 121.3
(COPh para); 115.3 (COPh ortho); 74.0 (OCH); 72.2
(OCH2Ph); 71.5 (CH2OPh); 17.5 (CH3). Mass spectrum
(m/e): 242 (M+, 30); 135 (9); 91 (100); 77 (25); 65 (16);
51 (7); 39 (7).
(s, 2H, OCH2). 13C-NMR acetone-d6 l: 158.8 (COPh
ipso); 137.1 (CPh ipso); 129.5 (COPh meta); 128.6 (CPh
meta); 127.9 (CPh para); 127.5 (CPh ortho); 120.9 (COPh
para); 114.9 (COPh ortho); 69.9 (OCH2). Mass spectrum
(m/e): 184 (M+, 92); 91 (100); 77 (10); 65 (78); 51 (19);
39 (36).
4.6. Allyl benzyl ether (13)
1H-NMR acetone-d6 l: 7.36–7.24 (m, 5H, Ph); 5.95
(m, J=17.4, 10.5, 5.3 Hz, 1H, CHꢁ); 5.29 (d of app-
arent q, J=17.4, 2 Hz, 1H, CH2ꢁ); 5.14 (ddt, J=10.5,
2, 1.5 Hz, 1H, CH2ꢁ); 4.5 (s, 2H, OCH2Ph); 4.0 (dt, J=
5.3, 1.5 Hz, 2H, OCH2). 13C-NMR acetone-d6 l: 139.8
(CPh ipso); 136.3 (CHꢁ); 129.0 (CPh meta); 128.3 (CPh
ortho); 128.2 (CPh para); 116.5 (ꢁCH2); 72.5 (OCH2Ph);
71.6 (OCH2). Mass spectrum (m/e): 148 (M+, 2); 118
(2); 107 (16); 91 (100); 79 (21); 51 (11); 39 (16).
References
[1] A. Stolarzewicz, Z. Grobelny, M. Kowalczuk, J. Organomet.
Chem. 492 (1995) 111.
[2] Z. Grobelny, A. Stolarzewicz, in: Proceedings of the Scientific
Conference PTChem, Wroclaw, 1998, p. 118.
[3] W. Moene, M. Schakel, G.J.M. Hoogland, F.J.J. de Kanter,
G.W. Klumpp, A.L. Spek, Tetrahedron Lett. 31 (1990)
2641.
[4] J. Furukawa, N. Kawabata, J. Nishimura, Tetrahedron 24 (1968)
-
53.
[5] E.J. Corey, J.D. Bass, R. LeMahieu, R.B. Mitra, J. Am. Chem.
4.7. Benzyl phenyl ether (8)
Soc. 86 (1964) 5579.
[6] W.H. Watanabe, L.E. Comlon, J. Am. Chem. Soc. 79 (1957)
2828.
1H-NMR acetone-d6 l: 7.5–6.9 (m, 10H+OPh); 5.1
.