Ferrocenium Cations as Catalysts for the Etherification of Cyclopropyl-Substituted Propargylic Alcohols: Ene-yne Formation and Mechanistic Insights

Article abstract of DOI:10.1002/ejoc.201901330

Commercial ferrocenium hexafluorophosphate ([FeCp₂]PF₆) and ferrocenium boronic acid hexafluoroantimonate ([FcB(OH)₂]SbF₆) were found to be efficient catalysts for the etherification of terminal, tertiary, cyclo

Full text of DOI:10.1002/ejoc.201901330

A Journal of
Accepted Article
Title: Ferrocenium Cations as Catalysts for the Etherification of
Cyclopropyl-Substituted Propargylic Alcohols: Ene-yne Formation
and Mechanistic Insights
Authors: Deva Saroja Talasila, Matthew J. Queensen, Michael Barnes-
Flaspoler, Kellsie Jurkowski, Evan Stephenson, Jordan M.
Rabus, and Eike B. Bauer
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Eur. J. Org. Chem. 10.1002/ejoc.201901330
Link to VoR: http://dx.doi.org/10.1002/ejoc.201901330
Supported by

10.1002/ejoc.201901330
European Journal of Organic Chemistry
Ferrocenium Cations as Catalysts for the Etherification of
Cyclopropyl-Substituted Propargylic Alcohols: Ene-yne
Formation and Mechanistic Insights
[a]
[a]
[a]
Deva Saroja Talasila, Matthew J. Queensen, Michael Barnes-Flaspoler, Kellsie
[a]
[a]
[a]
[a]
Jurkowski, Evan Stephenson, Jordan M. Rabus, Eike B. Bauer*
Submitted to the European Journal of Organic Chemistry
a
University of Missouri - St. Louis, Department of Chemistry and Biochemistry, One
University Boulevard, St. Louis, MO 63121, USA. Tel.: +1-314-516-5311; Fax: +1-314-516-
5342. bauere@umsl.edu. www.eike-bauer.net
Abstract
Commercial ferrocenium hexafluorophosphate ([FeCp ]PF ) and ferrocenium boronic acid
2
6
hexafluoroantimonate ([FcB(OH) ]SbF ) were found to be efficient catalysts for the
2
6
etherification of terminal, tertiary, cyclopropyl-substituted propargylic alcohols through
nucleophilic substitution with primary and secondary alcohols. The alcohol nucleophiles and the
propargylic alcohols were employed in a nearly equimolar amount and no further additives were
required. After 2 h reaction time at 40 °C in CH Cl and 3 to 5 mol% catalyst load, aromatic,
2
2
cyclopropyl-substituted propargylic alcohols gave rearranged, conjugated ene-yne products as
single isomers in 35 to 73% isolated yields. Cyclopropyl-substituted propargylic alcohols bearing
1
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901330
European Journal of Organic Chemistry
a thiophenyl substituent gave the corresponding cyclopropyl-substituted propargylic ethers in 27
to 56% isolated yields (45 °C, 2 h reaction time), where the cyclopropyl unit did not rearrange.
Cyclobutyl-substituted propargylic alcohols gave the corresponding propargyl ether substituted
products in 40 to 55% isolated yields (40 to 45 °C, around 16 h reaction time), and no
rearrangement of the cyclobutyl unit was observed. Only minor amounts of side products were
observed in the reaction mixtures. Experimental evidence points toward an ionic mechanism,
since the more electron-rich thiophenyl-substituted substrates reacted faster.
Keywords: Homogeneous catalysis, iron, isomerization.
Introduction
[1]
Propargylic alcohols (1 in Scheme 1) are valuable starting materials in organic synthesis.
They feature an alkyne and a hydroxyl unit in close proximity, and this bifunctionality allows for
[2,3]
a variety of useful organic transformations.
The triple bond can be engaged in cycloaddition
[4]
reactions and if terminal, it can be deprotonated to act as nucleophile or undergo cross
[5]
coupling reactions. Intramolecular cyclization reactions involving propargylic alcohols have
[6]
been investigated as well. The hydroxyl group can be replaced in nucleophilic substitution
[7]
reactions by a variety of nucleophiles (Scheme 1a). Tandem reactions constituting of a
propargylic nucleophilic substitution reaction followed by a cyclization can afford a variety of
[8]
cyclic and heterocyclic ring systems. As a consequence, propargylic alcohols are employed as
[9]
[10]
starting materials in the synthesis of a variety of natural products, pharmaceuticals
or other
[11]
materials with valuable optical or mechanical properties.
2
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901330
European Journal of Organic Chemistry
The bifunctionality of propargylic alcohols also enables rearrangement and isomerization
reactions. The most common ones are the Meyer-Schuster rearrangement (Scheme 1b), the Rupe
[12]
rearrangement and redox isomerization.
These reactions afford synthetically valuable α,β-
unsaturated carbonyl compounds (such as 3 in Scheme 1b) and can be of value in their own
[13]
right.
However, they can also cause yield-diminishing side reactions. Many catalyst systems,
especially based on Lewis acids such as FeCl , can catalyze intramolecular rearrangement
3
[14]
reactions of propargylic alcohols
as well as intermolecular reactions with other substrates
3
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901330
European Journal of Organic Chemistry
[15]
(e.g. in substitution reactions).
Thus, the search for catalyst systems that catalyze specifically
one reaction over others is still ongoing.
As part of our long-range research activities in the area of catalytic activation of propargylic
alcohols, we were especially interested in the substitution reaction of the –OH functionality of
[16]
propargylic alcohols by other nucleophiles.
A number of catalyst systems are known for that
[17]
[18]
reaction, and they are mainly based on transition metals such as ruthenium,
rhodium,
[19]
[20]
[21]
[22]
gold,
iridium,
copper
or iron.
However, main group compounds such as BiCl can
3
[23]
[24]
also catalyze propargylic substitution reactions.
Even Brønsted acids
can do so for certain
substrates. Still, challenges remain. Some catalyst systems require high reaction temperatures,
which may be problematic for the reaction of more highly sophisticated substrates. High reaction
temperatures can also promote the catalyzed elimination of water from a propargylic starting
[25]
[19]
material.
Furthermore, some catalyst systems are restricted to internal substrates
or show
[16a]
other substrate dependencies, like the restriction to propargylic acetates.
The idea, that a
proton can catalyze the substitution reaction for certain substrates appears appealing. However, a
strong Brønsted acid such as HOTf may not be compatible with more highly sophisticated targets
featuring other functional groups, giving rise to unwanted side reactions. Suppression of
[16c]
[26]
rearrangement side reactions such as the Meyer-Schuster rearrangement
or others
can
also be a challenge. As such, the search for catalyst systems that perform the reaction at
reasonably low temperatures and with a high level of selectivity is still ongoing, as demonstrated
by the number of publications and review articles on the topic that have appeared in the last few
[7]
years.
4
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901330
European Journal of Organic Chemistry
Iron catalysis is a vibrant area of research, because iron-based catalyst systems are affordable,
[27,28]
environmentally benign and tend to be non-toxic.
Ferrocenium cations have been
[29]
employed as stoichiometric oxidants in organic reactions previously,
but their catalytic
[30,31]
activity is not very widely researched.
Recently, we disclosed that ferrocenium
hexafluorophosphate, [FeCp ]PF (subsequently abbreviated as FcPF ), catalyzes the
2
6
6
substitution of the –OH group of aromatic propargylic alcohols by oxygen-centered nucleophiles
–
to give the corresponding propargylic ethers (according to Scheme 1a, where Nu would
– [16c]
formally be RO ).
Besides our own finding based on FcPF , we are only aware of two other
6
[15]
iron based catalysts for the catalytic substitution of propargylic alcohols, FeCl
and
3
+ [32]
[Fe(Cp)(CO) ] .
Interestingly, the substitution reactions catalyzed by FcPF proceeded at 40
2
6
[16]
[33]
°C, which is lower than many other catalyst systems published by us
or others.
This led to
speculation what the cause of the reduced reaction temperature for the FcPF catalyst could be.
6
[29]
Ferrocenium hexafluorophosphate is known to be a one electron oxidant,
which might enable
radical reactions, and we were speculating if propargylic substitution reactions may proceed
through a radical mechanism. To test this hypothesis, we decided to employ radical clock
substrates in the reaction. Radical clocks rearrange when they form radicals throughout the
course of a reaction, and the observation of rearranged products provides evidence for a radical
[34]
mechanism.
Accordingly, we decided to synthesize cyclopropyl-substituted propargylic alcohol substrates
(4 in Scheme 1c) to investigate whether the reaction proceeds through a radical mechanism.
Cyclopropyl-substituted radicals 6 (Scheme 1c) may ring-open to form alkenes, but carbocation
5 may also have this tendency (vide infra). As exemplified in Scheme 1c, ene-ynes 7 can form
5
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901330
European Journal of Organic Chemistry
[35]
through rearrangement if a cyclopropyl-substituted propargylic alcohol 4 is employed.
The
employment of cyclopropyl-substituted propargylic alcohols in reactions with alcohols to give
[36a]
conjugated, achiral enynes has been reported previously only four times, utilizing Yb(OTf) ,
3
[36b]
[36c]
[36d]
triflic acid (TfOH),
HAuCl
and ruthenium complexes
as catalysts. In these reports,
4
[36b,c]
[36a,d]
the nucleophile was either the solvent
or employed in large excess.
Herein, we report
ferrocenium-catalyzed substitution reactions with these substrates and isolated both ene-yne
products and cyclopropyl-substituted products, depending on the substituent R in 4.
Experimental evidence points toward an ionic mechanism through carbocation 5 (Scheme
[37]
1c).
Results
First, we needed to access cyclopropyl-substituted propargylic alcohols, which can be
synthesized by the addition of an ethynyl anion to the carbonyl unit of the corresponding
cycloalkyl ketone substrate precursors. Accordingly, the known cyclopropyl substituted
[36d]
[36a]
propargylic alcohols 8
and 9
were synthesized following literature procedures as well as
[38]
the known cyclobutyl-substituted propargylic alcohol 10 (Figure 1).
From our previous work, we knew that FcPF catalyzed the substitution of tertiary, aromatic
6
[16c]
propargylic alcohols by alcohol nucleophiles.
Accordingly, we did not perform extensive
6
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901330
European Journal of Organic Chemistry
optimization efforts for the FcPF -catalyzed substitution reactions of the cyclopropyl-substituted
6
propargylic alcohols. We assumed that our previously established reaction conditions (CH Cl
2
2
solvent, 40 °C reaction temperature, almost equimolar ratio of propargylic alcohol to alcohol
nucleophile) would work for the cyclopropyl substrates as well, which turned out to be the case.
For the test reaction between cyclopropyl alcohol 9 and n-butanol, the corresponding ene-yne
product 11 (resulting from opening of the cyclopropyl ring) was obtained after 2 h at 40 °C in
CH Cl with a virtual complete conversion rate (Table 1, entry 1). No significant amounts of
2
2
side products, e.g. from rearrangements as shown in Scheme 1b, were observed by GC. As
investigated before, CH Cl seems to be the solvent of choice for the reaction. The reaction also
2
2
runs well in ClCH CH Cl (entry 2). However, as can be seen in Table 1, solvents such as THF
2
2
were less efficient (entry 3) and in toluene, the reaction did not proceed at all (presumably
because the FcPF catalyst is not sufficiently soluble in that solvent, entry 4). Some catalyst
6
systems were reported where the alcohol nucleophile also serves as the solvent or an excess of
[17,20]
the nucleophile over the propargylic alcohol was employed.
However, in our case, when n-
butanol was used as the solvent, the conversion rate dropped to 25% (entry 5). It appears the
alcohol solvent inhibits catalytic activity.
7
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901330
European Journal of Organic Chemistry
Table 1. Screening reactions.
OH
O
FcPF6 (5 mol%)
Ph
11
CH2Cl2
40-45 °C, 2 h
+
10
Ph
OH
Entry a
Conversion (%)
Solvent
CH2Cl2
1
2
97
95
ClCH2CH2Cl
THF
3
4
5
44
toluene
none
25%
n-butanol
6
CH2Cl2 / cat.[CoCp2]PF6
none
7
8
CH2Cl2 / TEMPO
CH2Cl2 / BHT
34 b
98 c
a Typical conditions: propargylic alcohol (around 0.6 mmol)
and alcohol (around 0.6 mmol, equimolar) in the solvent (1
mL) catalyzed by FcPF6 (5 mol%), at 40-45 °C for 2 hours.
Conversions were determined by GC.
b In the presence of 20 mol% TEMPO for 16 h. In the
presence of 100 mol% TEMPO, the conversion was 11%.
c In the presence of 20 or 100 mol% BHT for 16 h.
Interestingly, cobaltocenium hexafluorophosphate [CoCp ]PF does not catalyze the reaction
2
6
(Table 1, entry 6). Cobaltocenium hexafluorophosphate is cationic but an 18 valence electron
complex and not a one electron oxidant, whereas FcPF is a 17 valence electron complex.
6
8
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901330
European Journal of Organic Chemistry
To further investigate the possibility of a radical mechanism, we performed the reaction in the
presence of the radical scavengers (2,2,6,6-tetramethylpiperidin-1-yl)oxyl (TEMPO) and 2,6-di-
[39]
tert-butyl-4-methylphenol (BHT, entries 7 and 8).
As can be seen, BHT did not inhibit the
reaction at all, and TEMPO slowed it down but did not completely stop it. This provides
evidence that the reaction does not involve a free radical chain mechanism.
Taking the data in Table 1 together, preliminary results are somewhat inconclusive. The
[40]
reaction may proceed through a radical mechanism:
the cyclopropyl substrate ring-opens to
the ene-yne product. Cobaltocenium hexafluorophosphate, which does not act as a one-electron
oxidant but possibly as a Lewis acid, is catalytically inactive. On the other hand, radical
scavengers do not completely inhibit the reaction, and additional evidence collected throughout
the course of our studies point toward an ionic mechanism (vide infra).
Next, we investigated the substrate scope of the reaction of the cyclopropyl propargylic
alcohol 9 for the substitution reaction with a variety of alcohol nucleophiles (Table 2). As can be
seen in Table 2, a variety of primary alcohols can be employed in the reaction to give the ene-
yne products in isolated yields ranging from 35 to 73% after a 2 h reaction time. Secondary
alcohols gave on average somewhat lower isolated yields of 35 to 45% (products 18 to 20),
which we tentatively attributed to the lower nucleophilicity of these alcohols. In all cases, the
rearranged ene-yne products were isolated, as judged by NMR spectroscopy. Cyclopropyl rings
13
1
give diagnostic peaks at around 2 and 3 ppm in their C{ H} NMR spectra and peaks below 0.8
1
[36a,b]
ppm in their H NMR spectra,
which were absent in all products in Table 2. Furthermore,
1
the olefinic methine =CH protons gave distinct triplets around 6.3 ppm in the H NMR spectra,
[36a,b]
giving further evidence for the ene-yne products.
The terminal alkyne unit was still present
9
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901330
European Journal of Organic Chemistry
1
in all products, as evidenced by resonances for the alkyne protons around 2.3 ppm in the H
13
1
–1
NMR and around 83 and 80 ppm in the C{ H} NMR spectra. IR stretches around 3300 cm
were also observed for the terminal alkynes. Also, no –OH stretch for alcohols was observed in
the IR spectra, demonstrating that the isolated products did not contain alcohols. In principle, all
products in Table 2 can exist as E or Z isomers. However, only one product isomer was observed
both by GC and NMR spectroscopy. We did not perform NOE NMR experiments for all
products, but we did for 14. We observed an NOE correlation between the olefinic methine =CH
proton and a phenyl proton, pointing toward an E isomer. Product 13 matched literature values,
[36b]
which was also assigned by the authors to the E product.
We tentatively assigned the
configurations of all other products in Table 2 also to be the E isomers. The products in Table 2
turned out to be not very stable. They decompose in solution or in the solid state within days, and
we could not obtain correct elemental analyses, they tended to be low on carbon by about 1%.
Oxidative decomposition may take place, possibly catalyzed by trace amounts of iron present in
the final products. GC-MS data taken of freshly prepared samples match the structures of the
products and the NMR spectra exhibited baseline purity.
10
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901330
European Journal of Organic Chemistry
Table 2. Isolated yields under optimized conditions.[a]
OR
OH
FcPF6 (5 mol%)
+ ROH
CH2Cl2
45 °C, 2 h
Ph
Ph
9
O
O
O
Ph
Ph
13, 73%
Ph
12, 50%
11, 37%
O
O
14, 60%
15, 35%
Ph
Ph
O
Ph
O
3
3
16, 50%
Ph
17, 47%
Ph
O
O
O
Ph
Ph
Ph
18, 41%
19, 35%
20, 45%[b]
[a] Typical conditions: propargylic alcohol (around 0.6 mmol) and
alcohol (around 0.6 mmol, equimolar) in CH2Cl2 (1 mL) catalyzed
by FcPF6 (5 mol%), at 40 - 45 °C for 2 hours (in some cases, the
reactions were run overnight, but were generally complete after 2
h). The products were isolated by column chromatography.
[b] With HBF4 as catalyst.
The reaction times of 1 to 2 h to obtain the products in Table 2 are shorter than for our
previously published FcPF catalyzed reactions employing tertiary propargylic alcohols without
6
[16d]
a cyclopropyl unit (where the reaction times were between 16 h and 3 d).
Consequently, it
appears that the cyclopropyl unit accelerates the reaction significantly. The ring-opened ene-yne
11
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901330
European Journal of Organic Chemistry
products in Table 2 may provide evidence for a radical reaction. However, we observed by NMR
and GC that first the substitution product with a ring-closed cyclopropyl unit formed, which
subsequently ring-opened.
We monitored the reaction between cyclopropyl-substituted propargylic alcohol 9 and n-
butanol over time. After about 20 minutes, the starting material was more or less consumed and
the substitution product 21 with the ring-closed cyclopropyl unit had formed (as determined by
GC and NMR, Scheme 2 and Figure 2). Over the course of another 20 minutes, the cyclopropyl
ring opened to give the eneyne product. It was possible to intercept the intermediate 21 and
purify it by column chromatography to identify its structure by NMR (albeit in only 23% isolated
yield and 95% spectroscopic purity); the diagnostic peaks for the cyclopropyl ring around 2 and
3 ppm were still present in 21. As such, the ring-closed product 21 forms first, which ring-opens
in a second step to the eneyne 11. The course of the reaction is illustrated in Figure 2, where for
the reaction in Scheme 2 the relative ratios of starting material intermediate and ene-yne product
are plotted versus the reaction time. As can be seen, the starting material first converts to the
cyclopropyl-substituted intermediate, which then rearranges to the ene-yne product.
12
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901330
European Journal of Organic Chemistry
100
90
80
70
60
50
%
40
30
20
10
0
0
10
20
30
40
50
60
time / min
Figure 2. Formation of the ene-yne product over time, monitored by GC (average of two
runs). Propargyl alcohol starting material 9: solid line. Cyclopropyl intermediate 21: dashed line.
Ene-yne product 11: dotted line.
This is an important observation. First, the initial formation of the ring-closed product would
indicate an ionic mechanism, and the cyclopropyl substituent efficiently stabilizes that positive
[41]
charge.
Second, it appears possible that the cyclopropyl-substituted propargylic ether 21
rearranges in the presence of a Lewis acid. To support this assumption, we added a small amount
of the Brønsted acid HBF to an NMR sample of the cyclopropyl-substituted propargylic ether
4
21 (Scheme 2). Indeed, ring opening to the ene-yne product 11 (in addition to some side
products) was observed by NMR after an hour. Finally, a tertiary radical intermediate like 6 in
Scheme 1c may not have a large tendency to ring-open to a primary radical. These findings
13
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901330
European Journal of Organic Chemistry
support an ionic mechanism, and that the ring-opening is possible in the presence of Brønsted
acids.
More evidence that Lewis acids can ring-open the cyclopropyl ring in 9 provides product 20
in Table 2. Ene-yne 20 was obtained with HBF as the catalyst. Under these reaction conditions,
4
it appears unlikely that radicals form, further excluding a radical mechanism for the ring-
opening.
We next turned our attention to the cyclobutyl-substituted propargylic alcohol 10. We were
speculating as to whether cyclic substituents as part of the propargylic alcohol can, in general,
accelerate the substitution reactions, whereas we were not expecting the cyclobutyl substituent to
ring-open. Indeed, the substitution reactions also worked for this substrate with five different
alcohols (Table 3), and the cyclobutyl ring system stayed intact. However, the reaction times for
the cyclobutyl-substituted propargylic alcohol 10 were much longer; it took at least 8 hours for
the reaction to go to completion and reactions were typically run overnight. As such, the
cyclobutyl ring does not accelerate the reaction, giving the cyclopropyl substituent a special role
in the substitution reactions described herein.
14
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901330
European Journal of Organic Chemistry
Table 3. Isolated yields under optimized conditions.[a]
OH
OR
3-5 mol% Fc+
+ ROH
CH2Cl2
45 °C, 12 to 20 h
Ph
O
Ph
10
O
23, 40% [b]
O Ph
Ph
Ph
22, 43%
O
Ph
Ph
Ph
24, 43%
25, 55%
O
3
3
26, 42%
[a] Typical conditions: propargylic alcohol (around 0.600
mmol) and alcohol (around 0.600 mmol) in CH2Cl2 (1 mL)
catalyzed by [FcB(OH)2]SbF6 (3 mol%), at 45 °C for 12 to 20
hours. The products were isolated by column
chromatography.
[b] FcPF6 catalyst (5 mol%).
Finally, we became interested in what impact the aromatic ring substituent on the propargylic
alcohol has on the course of the reaction. We employed the thiophenyl-substituted, aromatic
propargylic alcohol 8 in substitution reactions (Table 4), and the results were surprising.
15
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901330
European Journal of Organic Chemistry
No ring opening of the cyclopropyl unit took place no matter the length of the reaction time.
The corresponding cyclopropyl-substituted propargylic ethers were isolated in 27 to 52%
isolated yields. All thiophenyl-substituted products still featured peaks around 2 and 3 ppm in the
13
1
C{ H} NMR spectra, indicating that the cyclopropyl substituent remained intact. Another
feature of the thiophenyl-substituted propargylic alcohol 8 was a dramatically increased reaction
rate with FcPF as the catalyst. As established by GC, the formation of 29 (Table 4) was 80%
6
complete after 10 minutes and it was 100% complete after 30 minutes, which is even somewhat
16
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901330
European Journal of Organic Chemistry
faster compared to the phenyl-substituted propargylic alcohol 9. As such, the thiophenyl
substituent in 8 has an additional accelerating effect on the reaction compared to the phenyl-
substituted substrate 9. In addition, an extended reaction time of 2 h and longer resulted in
conversion of the product 29 back to the ketone 31 (Scheme 3), as established by GC. Loss of the
alkynyl group may give the cationic intermediate 30, which can be hydrolyzed to the ketone 7
with the water generated during the formation of the ether 29.
OH
S
FcPF6
40 °C
30 min
O
O
FcPF6
S
8
40 °C
2 h
29
31
S
+
H2O
OH
O
HO
H2O
S
30
Scheme 3. Speculated reactivity of thiophene-substituted propargylic
alcohol 8.
Overall, it was determined that the course of the reaction of cyclopropyl-substituted
propargylic alcohols depended on the other substituent present on the propargylic substrate.
Phenyl substituents afforded the corresponding ene-yne products after one to two hours, whereas
thiophenyl-substituted substrates gave only cyclopropyl-substituted products after 10 to 30
minutes. The isolated yields were moderate to good. Although we typically observed complete
conversion by GC with only small amounts of side products, we also observed that we lost
product during chromatographic workup. The products appeared to not be especially stable, and
as exemplified in Scheme 3, may decompose over time.
17
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901330
European Journal of Organic Chemistry
Some of the reactions in Tables 3 and 4 were performed with the known ferrocenium boronic
[30c]
acid hexafluoroantimonate ([FcB(OH) ]SbF ) as the catalyst.
This catalyst appears to be a
2
6
little less reactive, as demonstrated with product 29 (Table 4), where the reaction needed to be
run overnight. Ferrocenium boronic acid hexafluoroantimonate gave inseparable ring-opened and
ring-closed product mixtures for the reactions in Table 2. However, for some substrates in Tables
3 and 4, they gave, as opposed to FcPF , analytically pure samples. The influence of the
6
ferrocenium architecture and the counterion is currently under investigation.
Discussion
The above findings shed some light on the mechanism of propargylic substitution reactions
and the ring-opening behavior of cyclopropyl rings. The above data clearly give evidence for an
ionic, S 1-type mechanism (Scheme 4, pathway to the right). As such, FcPF assists in the
N
6
formation of the carbocation 32, which subsequently reacts with the nucleophilic alcohol to give
+
the propargylic ether. The ether may rearrange, assisted by the Fc ion, another Lewis acid, or a
proton, to the ene-yne product, possibly through an intramolecular process. Other authors also
suggested an ionic mechanism for the formation of ene-ynes from cyclopropyl-substituted
[36b]
propargylic alcohols.
18
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901330
European Journal of Organic Chemistry
OH
Fc+
OH
ROH
Ar
32
not
Ar
observed
+ ROH
H+
Fc
OR
OR
OR
Fc+
Ar
Ar
Ar
Scheme 4. Suggested mechanistic sequence.
Several data points give evidence for an ionic mechanism. First, the initial substitution
products for both cyclopropyl-substituted propargylic alcohols are the ring-closed cyclopropyl
propargylic ethers. The proparglic ethers bearing a phenyl-substituent cause an opening of the
cyclopropyl ring over time, whereas the ethers with a thiophenyl substituent do not. As
mentioned above, the thiophenyl-substituted propargylic alcohol 8 reacts much faster than the
phenyl-substituted alcohol 9. The thiophenyl substituent is more electron-rich, facilitating the
formation of the carbocation intermediate 32 in Scheme 4. The cyclobutyl-substituted
propargylic alcohol 10 is the least reactive among those tested in this study, requiring a longer
reaction time of at least 8 hours. The cyclobutyl substituent in 10 seems to behave like a
“regular” substituent in a tertiary propargylic alcohol. We tested tertiary propargylic alcohols
with a methyl instead of a cyclopropyl substituent in previous work, and they also required
[16c]
reaction times above 8 hours.
As such, the cyclopropyl ring alone activates the propargylic
alcohols for nucleophilic substitution reactions compared to an alkyl substituent. This is in line
with the increased aptitude of the cyclopropyl ring to stabilize a positive charge (as in 32)
[41]
compared to a methyl substituent,
further supporting an ionic mechanism.
19
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901330
European Journal of Organic Chemistry
As such, a radical mechanism appears unlikely. Some radical ring opening mechanisms were
reported to proceed under oxidative conditions with an oxidant present in stoichiometric
[40]
amounts.
We employed FcPF (potentially an oxidant) in catalytic amounts. Finally, the ring-
6
opening could also take place by direct attack of the alcohol on the cyclopropyl ring (Scheme 4,
left pathway). However, such a direct attack is primarily observed with electrophilic
[42]
cyclopropanes,
and the cyclopropyl substituents in our starting material may just not be
electrophilic enough.
An open question, however, is why the phenyl-substituted cyclopropyl ether 21 rearranges to
the ene-yne product 11 under ionic conditions (Scheme 2), and the thiophenyl-substituted ethers
do not. As shown in the experiment in Scheme 2 and mechanistically depicted in Scheme 4, a
Lewis acid may catalyze the rearrangement from a cyclopropyl-substituted propargylic alcohol
to the corresponding ene-yne. One explanation may be that the sulfur atom in the thiophenyl ring
system disturbs the ring opening process.
Some of the yields of the reactions in Tables 2 to 4 were only moderate. However, with the
exception of product 13 in Table 2, all catalysis products are new. Especially the quaternary,
cyclobutyl- and thiophenyl-substituted propargyl ethers constitute a new class of compounds, to
which the ferrocenium catalysts provide a new synthetic pathway. This is of importance, because
[43]
[44]
some ene-ynes
activity.
and propargylic ethers
were demonstrated to exhibit pharmaceutical
Conclusion
We demonstrated that ferrocenium cations serve as catalysts in the substitution of –OH groups
in propargylic alcohols by alcohol nucleophiles. Cyclopropyl-substituted propargylic alcohols
20
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901330
European Journal of Organic Chemistry
afforded ene-ynes through ring opening of the cyclopropyl ring, which took place after the
substitution reaction. Thiophenyl- and cyclobutyl-substituted propargylic alcohols afforded the
corresponding propargylic ethers, and no ring-opening took place. These quaternary propargylic
alcohols with a thiophenyl- and cyclopropyl ring are a new compound class, to which
ferrocenium catalysts provide access. We think carbocation intermediates formed during the
reaction, because electron-rich substituents accelerated the reaction and the substitution itself
took place without cyclopropyl ring-opening, which may be more prevalent when a radical
intermediate forms. Also, radical scavengers did at best slow down the reaction but did not
completely inhibit it.
Acknowledgment
We would like to thank the University of Missouri – St. Louis (Research Award) for financial
support. Further funding from the National Science Foundation for the purchase of the NMR
spectrometer (CHE-9974801) is acknowledged. We thank Professor Benjamin J. Bythell (Ohio
University) for providing HRMS data of some of the catalysis products.
Supplementary data
1
13
1
H and C{ H} NMR spectra of all catalysis products in Tables 2 to 4 and for the reaction in
Scheme 2. Supplementary data associated with this article can be found, in the online version,
at xxx.
Experimental
General. All chemicals, ferrocenium hexafluorophosphate (FcPF ) and cobaltocenium
6
hexafluorophosphate ([CoCp ]PF ) were used as supplied from Sigma-Aldrich. CH Cl was
2
6
2
2
21
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901330
European Journal of Organic Chemistry
[36a]
freshly distilled from CaH . 1-Cyclopropyl-1-phenylprop-2-yn-1-ol (9),
1-cyclopropyl-1-
2
[36d]
[38]
(thiophen-2-yl)prop-2-yn-1-ol (8),
1-cyclobutyl-1-phenylprop-2-yn-1-ol (10),
and
ferrocenium boronic acid hexafluoroantimonate ([FcB(OH) ]SbF ) were synthesized following
2
6
[30c]
the literature.
All NMR spectra for characterization were collected at room temperature on a
Bruker Avance 300 MHz instrument; all chemical shifts () are reported in ppm and are
referenced to a residual solvent signal. All assignments are tentative. IR spectra were collected
on a Thermo Nicolet 360 FT-IR spectrometer. HRMS measurements were performed on a
MaXis plus quadrupole time-of-flight mass spectrometer (Bruker) with atmospheric pressure
photoionization (APPI). EI masses were recorded on a HP 5988A GC-MS instrument. Elemental
analyses were performed by Atlantic Microlab Inc., Norcross, GA, USA.
(6-Methoxyhex-3-en-1-yn-3-yl)benzene (12, Table 2): In a screw-cap pressure vial, 1-
phenyl-1-cyclopropyl-2-yne-1-ol (0.100 g, 0.581 mmol) was dissolved in CH Cl (2 mL).
2
2
Methanol (0.0185 g, 0.593 mmol) and FcPF (0.010 g, 0.030 mmol) were added and the sealed
6
vial was heated at 45 °C for 2.5 hours. The reaction mixture was filtered through a short pad of
silica gel using CH Cl (2–4 mL). The product was obtained by column chromatography on
2
2
alumina (2.5 × 30 cm, 9:1 v/v hexanes : EtOAc) as an orange colored oil (0.054 g, 0.290 mmol,
1
50%). H NMR (300 MHz, CDCl ): δ =7.69–7.65 (m, 2H, aromatic), 7.42–7.31 (m, 3H,
3
aromatic), 6.61 (t, J = 7 Hz, C=CH, 1H), 3.60 (t, J = 7 Hz, 2H, OCH ), 3.44 (s, 3H, CH ),
HH
2
3
13
1
3.41 (s, ≡CH, 1H), 2.86 (q, J = 7 Hz, 2H, OCH ) ppm; C{ H} (75 MHz, CDCl ): δ =137.5
HH
2
3
(s, C=CH), 136.2 (s, aromatic), 128.4 (s, aromatic), 127.8 (s, aromatic), 126.0 (s, aromatic),
124.1 (s, C=CH), 83.5 (s, HC≡), 80.7 (s, HC≡C), 71.5 (s, =CCH ), 58.7 (s, CH O), 31.7 (s,
2
2
–1
CH ); IR (ATR, neat): ṽ = 3283 (m), 2922 (m), 2870 (m), 2825 (m) cm ; MS (70 eV): m/z (%):
3
22
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901330
European Journal of Organic Chemistry
+
+
+
+
185 (30) [M–H] , 171 (10) [M–CH ] , 156 (50) [M+H–OCH ] , 141 (100) [M–C H O] , 128
3
3
2 5
+
+
+
(60) [M+H–C H O] , 115 (100) [M+H–C H O] , 45 (90) [C H O] .
3 7
4 8
2 5
[36b]
6-Ethoxy-3-phenyl-3-en-1-yne (13, Table 2):
In a screw cap pressure vial, 1-phenyl-1-
cyclopropyl-2-yne-1-ol (0.116 g, 0.623 mmol) was dissolved in CH Cl (2 mL). Ethanol (0.031
2
2
g, 0.674 mmol) and FcPF (0.010 g, 0.030 mmol) were added and the sealed vial was heated at
6
45 °C for 75 minutes. The reaction mixture filtered through a short pad of silica gel and the
filtrate was chromatographed (2 × 30 cm alumina, 1:1 v/v EtOAc/hexanes 1:9) to obtain the
1
product 6-ethoxy-3-phenyl-3-en-1-yne (0.091 g, 0.455 mmol, 73%) as an orange oil. H NMR
(300 MHz, CDCl ): δ =7.66 (d, J =7.0 Hz, 2H, aromatic), 7.46–7.31 (m, 3H, aromatic), 6.61
3
HH
(t, J =7.4 Hz, C=CH, 1H), 3.65–3.54 (m, 4H, CH OCH ), 3.28 (s, ≡CH, 1H), 2.72 (q, J =7.0
HH
2
2
HH
13
1
Hz, 2H, CH CH ), 1.15 (t, J =7.0 Hz, 3H, CH ) ppm; C{ H} (75 MHz, CDCl ): δ = 137.5
2
3
HH
3
3
(s, C=CH), 136.4 (s, aromatic), 128.4 (s, aromatic), 127.7 (s, aromatic), 126.0 (s, aromatic),
124.0 (s, C=CH), 83.4 (s, C≡CH), 80.7 (s, C≡CH), 69.3 (s, OCH ), 66.2 (s, C=CHCH ), 31.9 (s,
2
2
CHCH O), 15.5 (s, CH ) ppm.
2
3
(6-Butoxyhex-3-en-1-yn-3-yl) benzene (11, Table 2): In a screw cap pressure vial, 1-phenyl-
1-cyclopropyl-2-yne-1-ol (0.105 g, 0.610 mmol) was dissolved in CH Cl (2 mL). n-Butanol
2
2
(0.045 g, 0.608 mmol) and FcPF (0.010 g, 0.030 mmol) were added and the vial was sealed and
6
heated at 45 °C for 2 hours. The reaction mixture was filtered through silica gel, using CH Cl
2
2
(2–4 mL). The product was obtained by column chromatography on alumina (2.5 × 30 cm, 9:1
1
v/v hexanes : EtOAc) as a yellow colored oil (0.052 g, 0.228 mmol, 37%). H NMR (300 MHz,
23
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901330
European Journal of Organic Chemistry
2
CDCl ): δ =7.53–7.51 (m, 2H, aromatic), 7.28–7.16 (m, 3H, aromatic), 6.48 (t, 1H, J =7 Hz,
3
HH
2
2
=CH), 3.49 (t, J =7 Hz, 2H, OCH ), 3.38 (t, 2H, J =7 Hz, OCH ), 3.27 (s, 1H, ≡CH), 2.72
HH
2
HH
2
2
(q, J =7 Hz, 2H, CH ), 1.54–1.34 (m, 2H, CH ), 1.36–1.26 (m, 2H, CH ), 0.85 (t, 3H,
HH
2
2
2
2
13
1
J
=7 Hz, CH ) ppm; C{ H} (75 MHz, CDCl ): δ = 137.6 (s, C=CH), 136.5 (s, aromatic),
3 3
HH
128.4 (s, aromatic), 127.7 (s, aromatic), 126.0 (s, aromatic), 123.9 (s, C=CH), 83.4 (s, C≡CH),
80.8 (s, C≡CH), 70.8 (s, OCH ), 69.5 (s, OCH ), 31.90 (s, CH ), 31.86 (s, CH ), 19.4 (s, CH ),
2
2
2
2
2
−1
14.0 (s, CH ) ppm; IR (ATR, neat): ṽ = 3288 (m), 3055 (w), 2973 (m), 2926 (w) cm ; MS
3
+
+
+
(70 eV): m/z (%): 185 (20) [M–C H ] , 155 (15) [M–C H O] , 141 (50) [M–C H O] , 128
3 7
4 9
5 11
+
+
+
(40) [M–C H O] , 115 (50) [M–C H O] , 57 (100) [C H ] .
6 12
7 13
4 9
(6-(Hexyloxy)hex-3-en-1-yn-3-yl)benzene (14, Table 2): 1-Cyclopropyl-1-phenylprop-2-yn-
1-ol (0.103 g, 0.598 mmol) was added to a screw cap pressure vial and dissolved in CH Cl (2
2
2
mL). 1-Hexanol (0.075 g, 0.669 mmol) was added followed by FcPF (0.012 g, 0.037 mmol).
6
The sealed vial was then heated to 45 °C for 2 hours. The mixture was filtered through a short
pad of silica using CH Cl (2–4 mL). The product was obtained by chromatography on a neutral
2
2
alumina oxide (Aluminar®) column (2.5 × 30 cm, 9:1 v/v hexanes : EtOAc) to give the product
1
as a yellow colored oil (0.093 g, 0.361 mmol, 60%). H NMR (300 MHz, CDCl ): δ =7.54–7.51
3
2
(m, 2H, aromatic), 7.28–7.19 (m, 3H, aromatic), 6.48 (t, J = 7 Hz, 1H, alkene), 4.07 (s,
HH
2
2
0.16H, ferrocene), 3.48 (t, J = 7 Hz, 2H), 3.36 (t, J = 7 Hz, 2H), 3.26 (s, 1H), 2.75–2.68
HH
HH
2
2
2
(q, J = 7 Hz, 2H), 1.50 (q, J = 7 Hz, 2H), 1.31–1.22 (m, 6H), 0.80 (t, J = 7 Hz, 3H)
HH
HH
HH
24
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901330
European Journal of Organic Chemistry
13
1
ppm; C{ H} (75 MHz, CDCl ): δ = 137.6 (s, C=CH), 136.5 (s, aromatic), 128.4 (s, aromatic),
3
127.8 (s, aromatic), 126.0 (s, aromatic), 124.0 (s, C=CH), 83.5 (s, C≡CH), 80.8 (s, C≡CH), 71.1
(s, CH O), 69.5 (s, CH O), 68.0 (s, trace amount ferrocene), 31.9 (s, CH ), 31.8 (s, CH ), 29.8
2
2
2
2
(s, CH ), 26.0 (s, CH ), 22.7 (s, CH ), 14.2 (s, CH ) ppm; IR (ATR, neat): ṽ = 3288 (m), 3060
2
2
2
3
–1
+
(w), 3024 (w), 2927 (s), 2854 (s), 2790 (m) cm ; MS (70 eV): m/z (%): 255 (5) [M–1] , 155
+
+
+
+
(30) [M–C H O] , 141 (100) [M–C H O] , 128 (70) [M–C H O] , 115 (95) [M–C H O] ;
6 13
7 15
8 16
9 17
+
HRMS (APPI): m/z calcd for C H O: 257.1905 [M+H] ; found: 257.1900.
18 25
6-Alloxy-3-phenyl-3-en-1-yne (15, Table 2): In a screw cap pressure vial, 1-phenyl-1-
cyclopropyl-2-yne-1-ol (0.094 g, 0.547 mmol) was dissolved in CH Cl (2 mL). Allyl alcohol
2
2
(0.031 g, 0.534 mmol) and FcPF (0.010 g, 0.030 mmol) was added and the vial was sealed and
6
heated at 45 °C overnight. The reaction mixture was filtered through a short pad of silica gel with
CH Cl (2–4 mL). The product was obtained by column chromatography on alumina (2.5 × 30
2
2
1
cm, 9:1 v/v hexanes : EtOAc) as an orange colored oil (0.041 g, 0.193 mmol, 35%). H NMR
(300 MHz, CDCl ): δ =7.54–7.49 (m, 2H, aromatic), 7.28–7.19 (m, 3H, aromatic), 6.48 (t, J
=
HH
3
7 Hz, C=CH, 1H), 5.90–5.80 (m, CH=CH , 2H), 4.16 (t, J = 5.48 Hz, OCH , 2H), 3.93 (t,
2
HH
2
J
J
= 7 Hz, OCH , 2H), 3.52 (t, J = 7.4 Hz, OCH CH , 2H), 3.27 (s, C≡CH, 1H), 2.73 (q,
HH
HH
2
HH
2
2
13
1
= 7.3 Hz, CHCH , 2H) ppm; C{ H} (75 MHz, CDCl ): δ =137.5 (s, C=), 136.4 (s,
2
3
aromatic), 136.2 (s, C=), 128.4 (s, aromatic), 127.8 (s, aromatic), 126.0 (s, aromatic), 124.1 (s,
C=), 117.1 (C=CH ), 83.5 (s, C≡CH), 80.7 (s, C≡CH), 71.9 (s, OCH ), 69.1 (s, OCH ), 68.0 (s,
2
2
2
trace amount ferrocene), 31.9 (s, =CHCH ) ppm; IR (ATR, neat): ṽ = 3267 (m), 3023 (w), 2853
2
25
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901330
European Journal of Organic Chemistry
–1
+
+
(m) cm ; MS (70 eV): m/z (%): 184 (10) [M–C H ] , 155 (20) [M–C H O] , 141 (50) [M–
2 4
3 5
+
+
+
+
C H O] , 128 (40) [M–C H O] , 115 (50) [M–C H O] , 41 ([C H ] , 100%).
4 7
5 8
6 9
3 5
6-(trans-5-Decoxy)-3-phenyl-3-en-1-yne (16, Table 2): In a screw cap pressure vial, 1-
phenyl-1-cyclopropyl-2-yne-1-ol (0.110 g, 0.639 mmol) was dissolved in CH Cl (2 mL) and
2
2
trans-5-decanol (0.100 g, 0.641 mmol) and FcPF (0.010 mg, 0.030 mmol) were added. The
6
sealed vial was heated at 45 °C for 2 hours. The reaction mixture was filtered through a short pad
of silica gel by using CH Cl (2–4 mL). The product was obtained by column chromatography
2
2
on alumina (2.5 × 30 cm, 9:1 v/v hexanes : EtOAc) as an orange colored oil (0.099 g, 0.319
1
mmol, 50%). H NMR (300 MHz, CDCl ): δ =7.54–7.52 (m, 2H, aromatic), 7.29–7.18 (m, 3H,
3
aromatic), 6.49 (t, J =7 Hz, 1H, C=CH), 5.32 (q, J =3 Hz, 2H, HC=CH), 3.50 (t, J = 7
HH
HH
HH
Hz, 2H, OCH ), 3.36 (t, J = 7 Hz, 2H, OCH ), 3.27 (s, 1H, ≡CH), 2.72 (q, J = 7 Hz, 2H,
2
HH
2
HH
CH ), 1.96–1.80 (m, 4H, 2CH ), 1.55–1.47 (m, 2H, CH ), 1.39–1.31 (m, 2H, CH ), 1.29–1.22
2
2
2
2
13
1
(m, 4H, 2CH ), 0.82 (t, J = 7 Hz, 3H, CH ) ppm; C{ H} (75 MHz, CDCl ): δ = 137.6 (s,
2
HH
3
3
C=), 136.5 (s, aromatic), 130.8 (s, C=), 130.0 (s, C=), 128.4 (s, aromatic), 127.7 (s, aromatic),
126.0 (s, aromatic), 123.9 (s, C=CH), 83.4 (s, C≡CH), 80.8 (s, C≡CH), 70.9 (s, OCH ), 69.5 (s,
2
OCH ), 32.4 (s, CH ), 32.4 (s, CH ), 31.89 (s, CH ), 31.85 (s, CH ), 29.2 (s, CH ), 26.2 (s,
2
2
2
2
2
2
+
CH ), 22.3 (s, CH ), 14.1 (s, CH ) ppm; MS (70 eV): m/z (%): 309 (5) [M–1] , 155 (30) [M–
2
2
3
+
+
+
+
C H O] , 141 (60) [M–C H O] , 128 (50) [M–C H O] , 115 (100) [M–C H O] ; IR
10 19
11 21
12 22
13 23
–1
(ATR, neat): ṽ = 3289 (m), 3021 (w), 2923 (s), 2853 (s), 2790 (w) cm .
26
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901330
European Journal of Organic Chemistry
6-Benzoxy-3-phenyl-3-en-1-yne (17, Table 2): In a screw cap pressure vial, 1-phenyl-1-
cyclopropyl-2-yne-1-ol (0.108 g, 0.627 mmol) was dissolved in CH Cl (2 mL) and benzyl
2
2
alcohol (0.068 g, 0.630 mmol) and FcPF (0.010 g, 0.030 mmol) were added. Then the vial was
6
sealed and heated at 45 °C for 75 minutes. The reaction mixture was filtered through a short pad
of silica gel using CH Cl (2-4 mL). The product 6-benzoxy-3-phenyl-3-en-1-yne was obtained
2
2
by column chromatography on alumina (2.5 × 30 cm, 9:1 v/v hexanes : EtOAc) as an orange
1
colored oil (0.077 g, 0.294 mmol, 47%). H NMR (300 MHz, CDCl ): δ =7.63–7.59 (m, 2H,
3
2
aromatic), 7.37–7.26 (m, 8H, aromatic), 6.57 (t, J = 7 Hz, C=CH, 1H), 4.55 (s, 2H,
HH
2
OCH Ph), 4.16 (0.5 H, ferrocene), 3.36 (t, J = 7 Hz, 2H, OCH ), 3.35 (s, 1H, C≡CH), 2.85
2
HH
2
2
13
1
(q, J = 7 Hz, 2H, CHCH ) ppm; C{ H} (75 MHz, CDCl ): δ = 138.4 (s, C=), 137.6 (s,
HH
2
3
aromatic), 136.42 (s, aromatic), 136.38 (s, aromatic), 128.53 (s, aromatic), 128.49 (s, aromatic),
127.8 (s, aromatic), 127.7 (s, aromatic), 126.0 (s, aromatic), 124.2 (s, =C), 83.7 (s, C≡CH), 80.8
(s, C≡CH), 73.0 (s, OCH ), 69.1 (s, OCH ), 31.9 (s, CHCH ) ppm; MS (70 eV): m/z (%): 261
2
2
2
+
+
+
+
(5) [M–1] , 141 (20) [M–C H O] , 114 (20) [M–C H O] , 91 (100) [C H ] ; IR (ATR, neat):
8 9
10 12
7 7
–1
ṽ = 3282 (w), 3026 (m), 2853 (m) cm .
((4-Phenylhex-3-en-5-yn-1-yl)oxy)cyclooctane (18, Table 2): 1-phenyl-1-cyclopropyl-2-
yne-1-ol (0.071 g, 0.413 mmol) was added to a screw cap vial and dissolved in CH Cl (1 mL).
2
2
Cyclooctanol (0.076 g, 0.593 mmol) was added followed by the addition of FcPF (0.012 g,
6
0.037 mmol). The sealed vial was heated to 45 °C for 100 minutes. The solvent was removed,
and the residue was chromatographed on a neutral alumina (Aluminar®) column (2.5 × 30 cm,
9:1 v/v hexanes : EtOAc) to give the product as an orange colored oil (0.060 g, 0.213 mmol,
52%). For further purification, the product was chromatographed on a silica column (2.5 × 10
27
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901330
European Journal of Organic Chemistry
cm, 9:1 v/v hexanes : EtOAc) to give the product as a yellow colored oil (0.048 g, 0.170 mmol,
1
41%). H NMR (300 MHz, CDCl ): δ =7.54–7.50 (m, 2H, aromatic), 7.28–7.16 (m, 3H,
3
2
2
aromatic), 6.49 (t, J = 7Hz, 1H, =CH), 4.08 (s, 0.16H, ferrocene), 3.49 (t, J = 7 Hz, 2H,
HH
HH
2
CH ), 3.39–3.33 (m, 1H, HCOH), 3.26 (s, 1H, C≡CH), 2.69 (q, J = 7 Hz, 2H), 1.80–1.30 (m,
2
HH
13
1
14H, 7CH ) ppm; C{ H} (75 MHz, CDCl ): δ =137.6 (s, =C), 136.8 (s, aromatic), 128.4 (s,
2
3
aromatic), 127.8 (s, aromatic), 126.0 (s, aromatic), 123.8 (s, =C), 83.4 (s, C≡CH), 80.8 (s,
C≡CH), 79.9 (s, CH O), 68.0 (s, trace amount ferrocene), 66.9 (s, CHO), 32.3 (s, CH ), 31.6 (s,
2
2
CH ), 27.4 (s, CH ), 25.5 (s, CH ), 23.2 (s, CH ) ppm; IR (ATR, neat): ṽ = 3289 (m), 3026 (w),
2
2
2
2
–1
+
+
2915 (s), 2849 (s) cm ; MS (70 eV): m/z (%): 281 (5) [M–1] , 141(80) [M–C H O] , 128 (75)
9 17
+
+
[M–C H O] , 114 (100) [M–C H O] ; HRMS (APPI): m/z calcd for C H O: 283.2061
10 18
11 20
20 27
+
[M+H] ; found: 283.2056.
1,7,7-Trimethyl-2-((4-phenylhex-3-en-5-yn-1-yl)oxy)bicyclo[2.2.1]heptane (19, Table 2):
1-phenyl-1-cyclopropyl-2-yne-1-ol (0.100 g, 0.581 mmol) was added to a screw cap vial and
dissolved in CH Cl (1 mL). l-Borneol (1,7,7-trimethylbicyclo[2.2.1]heptan-2-ol, 0.089 g, 0.576
2
2
mmol) was added followed by the addition of FcPF (0.010 g, 0.037 mmol). The vial was heated
6
to 45 °C for 120 minutes. The solvent was removed, and the residue was chromatographed on a
neutral alumina oxide (Aluminar®) column (2.5 × 30 cm, 9:1 v/v hexanes: EtOAc) to give the
1
product as an orange colored oil (0.063 g, 0.204 mmol, 35%). H NMR (300 MHz, CDCl ): δ
3
=7.53–7.50 (m, 2H, aromatic), 7.28–7.16 (m, 3H, aromatic), 6.50 (t, J = 7 Hz, 1H, =CH), 4.07
HH
2
(s, 0.4H, ferrocene), 3.55–3.42 (m, 3H), 3.27 (s, 1H, ≡CH), 2.68 (q, J = 7 Hz, 2H), 2.05–1.89
HH
28
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901330
European Journal of Organic Chemistry
(m, 2H), 1.62–1.53 (m, 2H), 1.19–1.11 (m, 2H), 0.94 (dd, J = 13 Hz, J = 3 Hz, 1H), 0.80
HH
HH
13
1
(s, 3H), 0.76 (s, 6H) ppm; C{ H} (75 MHz, CDCl ): δ =137.7 (s, =C), 137.2 (s, aromatic),
3
128.4 (s, aromatic), 127.7 (s, aromatic), 125.9 (s, aromatic), 123.6 (s, =C), 84.7 (s, C≡CH), 83.3
(s, OCH), 80.9 (s, C≡CH), 68.9 (s, CH O), 68.0 (trace amount ferrocene), 49.3 (s, CH), 47.8 (s,
2
CH), 45.0 (s, CH), 36.4 (s, CH), 32.3 (s, CH), 28.4 (s, CH), 26.7 (s, CH), 19.9 (s, CH ), 18.9 (s,
3
CH ), 14.1 (s, CH ) ppm; IR (ATR, neat): ṽ = 3303 (m), 3056 (w), 3024 (w), 2945 (s), 2869 (s)
3
3
–1
+
cm ; HRMS (APPI): m/z calcd for C H O: 309.2218 [M+H] ; found: 309.2210.
22 29
(6-(Cyclopentyloxy)hex-3-en-1-yn-3-yl)benzene (20, Table 2): 1-Cyclopropyl-1-
phenylprop-2-yn-1-ol (0.103 g, 0.598 mmol) was added to a 5-mL screw cap vial and dissolved
in CH Cl (1 mL). Cyclopentanol (0.050 g, 0.581 mmol) was added followed by the addition of
2
2
HBF (ca 5.0 μL of a diethyl ether complex). The vial was then heated at 40 °C for 1 hour and
4
the solvent was removed. The residue was chromatographed on a neutral alumina oxide
(Aluminar®) column (2.5 × 30 cm, 9:1 v/v hexanes : EtOAc) to give the product as an orange oil
1
(0.065 g, 0.270 mmol, 45 %). H NMR (300 MHz, CDCl ): δ =7.69–7.65 (m, 2H, aromatic),
3
2
7.43–7.31 (m, 3H, aromatic), 6.62 (t, 1H, J =7 Hz, C=CH), 4.02–3.97 (m, 1H, OCH), 3.60 (t,
HH
2
2
2H, J = 7 Hz, OCH ), 3.42 (s, 1H, C≡CH), 2.84 (q, 2H, J = 7 Hz, OCH CH ), 1.84–1.50
HH
2
HH
2
2
13
1
(m, 8H, 4CH ) ppm; C{ H} (75 MHz, CDCl ): δ = 137.6 (s, =C), 136.7 (s, aromatic), 128.4 (s,
2
3
aromatic), 127.7 (s, aromatic), 125.9 (s, aromatic), 123.9 (s, =C), 83.4 (s, HC≡C), 81.4 (s, OCH),
80.8 (s, HC≡C), 67.5 (s, CH O), 32.4 (s, CH ), 32.2 (s, CH ), 23.6 (s, CH ) ppm. MS (70 eV):
2
2
2
2
29
This article is protected by copyright. All rights reserved.