Concerning the Role of Supercritical Carbon Dioxide in SN1 Reactions

Article abstract of DOI:10.1002/chem.201604151

A series of S_N1-type reactions has been studied under various conditions to clarify the role of supercritical carbon dioxide (scCO₂) as reaction medium for this kind of transformations. The application of scCO₂ did not result in higher yields in any of the experiments in comparison to those under neat conditions or in the presence of other inert compressed gases. High-pressure UV/Vis spectroscopic measurements were carried out to quantify the degree of carbocation formation of a highly S_N1-active alkyl halide as a function of the applied solvent. No measureable concentration of carbocations could be detected in scCO₂, just like in other low polarity solvents. Taken together, these results do not support the previously claimed activating effect via enhanced S_N1 ionization due to the quadrupolar moment of the supercritical fluid.

Full text of DOI:10.1002/chem.201604151

DOI: 10.1002/chem.201604151
Full Paper
&
Supercritical Fluids
Concerning the Role of Supercritical Carbon Dioxide in SN1
Reactions
Yun X. Qiao,[a] Nils Theyssen,[a] Tobias Eifert,[b] Marcel A. Liauw,[b] Giancarlo Franciꢀ,[b]
Karolin Schenk,[b] Walter Leitner,*[a, b] and Manfred T. Reetz*[a, c]
Abstract: A series of SN1-type reactions has been studied
under various conditions to clarify the role of supercritical
carbon dioxide (scCO2) as reaction medium for this kind of
transformations. The application of scCO2 did not result in
higher yields in any of the experiments in comparison to
those under neat conditions or in the presence of other
inert compressed gases. High-pressure UV/Vis spectroscopic
measurements were carried out to quantify the degree of
carbocation formation of a highly SN1-active alkyl halide as
a function of the applied solvent. No measureable concen-
tration of carbocations could be detected in scCO2, just like
in other low polarity solvents. Taken together, these results
do not support the previously claimed activating effect via
enhanced SN1 ionization due to the quadrupolar moment of
the supercritical fluid.
Introduction
Recently, Gonzꢁlez-NfflÇez and co-workers reported the use of
supercritical carbon dioxide (scCO2)[1] as solvent for SN1 reac-
tions of appropriate alkyl halides such as tert-butyl and 1-ada-
mantyl halides (Cl and Br), as well as other potentially SN1-
active secondary halides such as 1-chloro-1-phenylethane (1).[2]
Based on their experimental observations, the authors conclud-
ed that scCO2 had an activating effect also on electrophilic aro-
matic substitution reactions. In a typical model reaction, a Frie-
del–Crafts alkylation of 1,3-dimethoxybenzene (2) was per-
formed at 608C and 250 bar by pressurizing the reaction com-
ponents with CO2 in a stainless steel reactor for 5 h, which was
reported to provide the alkylation products 3o,p and 3o,o in
82% yield (Scheme 1). In a most recent study, the same group
investigated the influence of the presence of 1-phenylethanol
and ethanol as protic additives, which led to product inhibition
in the electrophilic aromatic substitution reaction of 2 with 1.[3]
The authors proposed that the strong quadrupolar moment
(ꢀ14.3”10ꢀ40 Cm2) of scCO2 facilitates the SN1 ionization of
the alkyl halides in the absence of Lewis or Brønsted acids,
polar protic solvents, catalysts, or additives. The results of race-
Scheme 1. Lewis acid free Friedel–Crafts alkylation of 1,3-dimethoxybenzene
as reported in ref. [2]. Double-substituted isomers are not shown.
mization experiments using (+)-1-choro-1-phenylethane
seemed to support the hypothesis of the scCO2-assisted ionic
dissociation. The apparent ability of scCO2 to ionize carbon–
halogen bonds with subsequent dissociation of the ion pairs
was related to a clustering effect previously discussed in other
systems.[1,4]
In the present study, we report the results from a detailed
assessment of scCO2 as reaction medium in this and related
SN1-type reactions. In particular, the results from the test reac-
tions are corroborated with UV/Vis spectroscopic investigations
that quantify the degree of carbocation formation from SN1-
active alkyl halides as a function of the applied solvent.
Results and Discussion
[a] Dr. Y. X. Qiao, Dr. N. Theyssen, Prof. Dr. W. Leitner, Prof. Dr. M. T. Reetz
Max-Planck-Institut fꢀr Kohlenforschung
Inspired by the intriguing literature reports, we attempted to
perform other types of CꢀC bond forming reactions using alkyl
halides that normally undergo SN1 processes only in the pres-
ence of Lewis acids, to probe their reactivity as alkylating
agents in the supposedly activating solvent scCO2. Specifically,
we targeted the Lewis acid mediated tert-alkylation of enolsi-
lanes with formation of a-alkylated ketones.[5] If successful, the
use of scCO2 was hoped to eliminate the use of stoichiometric
amounts of such strong Lewis acids such as TiCl4 and allow for
a “greener” version of such synthetically important transforma-
Kaiser-Wilhelm-Platz 1, 45470 Mꢀlheim (Germany)
[b] T. Eifert, Prof. Dr. M. A. Liauw, Dr. G. Franciꢁ, K. Schenk, Prof. Dr. W. Leitner
Institut fꢀr Technische und Makromolekulare Chemie
RWTH Aachen University, Worringerweg 1, 52074 Aachen (Germany)
E-mail: leitner@itmc.rwth-aachen.de
[c] Prof. Dr. M. T. Reetz
Department of Chemistry, Philipps-Universitꢂt Marburg
Hans-Meerwein-Straße, 35032 Marburg (Germany)
E-mail: reetz@mpi-muelheim.mpg.de
Supporting information for this article can be found under:
http://dx.doi.org/10.1002/chem.201604151.
Chem. Eur. J. 2017, 23, 1 – 6
1
ꢂ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
tions. Using 1 as well as tert-butyl chloride as the alkylating re-
agents, a series of experiments was performed under various
conditions in scCO2 using the enolsilane 5 derived from cyclo-
hexanone as the nucleophile in the absence of any Lewis acid
or protic solvent (see the Supporting Information for details).
However, no alkylation products were detected in any of these
experiments. This argues against the formation of significant
amounts of the respective carbocations that would be expect-
ed to be intercepted immediately by the nucleophile.
peratures of 60–708C at CO2 pressures between 213–240 bar.
Significantly better results were obtained either under argon at
a pressure level of 205 bar (20% yield) or under CO2 at a lower
pressure of 100 bar (35% yield), while the temperature was
held at 608C.
To investigate the influence of the CO2 pressure and reaction
temperature—and hence density—in more detail, we analyzed
a series of experiments in which the temperature was varied
between 50–708C and the amount of CO2 between 5–25 g
using a design of experiment (DoE) approach (Table 1, En-
tries 1–11).
We, therefore, turned to the more SN1-active benzhydryl
chloride (4)[6] as alkylating agent. Indeed, it reacted with the
enolsilane 5 to give the alkylation product 6 and the byprod-
uct trimethyl silyl chloride 7 in 86–90% conversion (Scheme 2),
Table 1. Results of the Friedel–Crafts alkylation of 1,3-dimethoxybenzene
(2) with tert-butyl bromide (8) leading to 9a/9b.[a]
Entry Temperature[b] m (CO2) Pressure Conversion 8 Yield 9a+9b
[8C]
[g]
[bar]
[%]
[%]
1
2
3
4
5
6
7
8
60.0
60.0
67.1
60.0
52.9
60.0
52.9
60.0
67.1
50.0
70.0
60.0
70.0
70.0
15.0
5.0
22.1
15.0
7.9
15.0
22.1
25.0
7.9
15.0
15.0
0
108
52
192
109
68
107
135
183
76
90
–
0
17
0
48.0
49.4
18.8
42.5
33.4
26.1
22.9
15.0
42.4
26.8
45.0
65.0
76.4
91.2
47.5
48.9
18.6
42.1
33.1
25.8
22.7
14.9
42.0
26.5
44.6
64.3
75.6
90.3
Scheme 2. Lewis acid free alkylation of enolsilane 5.
9
when kept at a total pressure of 190 bar and 708C for 5 h in
a stainless steel reactor. To check whether the use of scCO2
plays a decisive role in this transformation, we performed the
same reaction under identical conditions by replacing CO2 with
compressed nitrogen or argon. At pressures of 190 bar at
708C, this corresponds to fluid densities of 0.27 gmLꢀ1 and
0.17 gmLꢀ1 for Ar and N2, respectively, which is far below their
critical densities (Ar: 0.54 gmLꢀ1; N2: 0.31 gmLꢀ1).[7] Thus—and
in contrast to the corresponding experiment with CO2—these
inert gases cannot act as solvents or reaction media for the
substrates or products under these conditions. However, in
both cases, high conversions were observed (93–97% and 93–
99%, respectively). In all cases, product selectivities between
95–99% were measured. These results cast significant doubt
on a specific “activating effect” of scCO2 in these reactions.
These observations prompted us to reinvestigate the effect
of scCO2 on the SN1-type Friedel–Crafts reaction using the
alkylation of 1,3-dimethoxybenzene (2) with tert-butyl bromide
(8) as chemical probe (Scheme 3). We chose this particular re-
action because it was reported to proceed quite smoothly to
give quantitative conversion under the conditions of Gonzꢁlez-
NfflÇez and co-workers.[2] In a series of experiments, employing
the same substrate loading in two different high pressure reac-
tors, we obtained, however, product yields (9a+9b) ranging
consistently only between 1% and 8% after 5 h applying tem-
10
11
12
13[c]
14[c]
[d]
2.0
0
[a] Reactions were carried out in a high pressure reactor (V=31 mL) man-
ufactured from Alloy C-276. A reaction time of 5 h was chosen for all ex-
periments. The reaction mixture was stirred magnetically with 1000 rpm.
n(8)=0.9 mmol, n(2)=3.6 mmol. Product selectivities were around 99%
in all cases. [b] Given is the individual set temperature of the autoclave
heating unit (hysteresis: ꢁ0.18C). Up to 15 min were needed to reach
this value in its equilibrium state. [c] Reaction time: 16 h. [d] Not mea-
sured.
Statistical analysis revealed that the conversion in the
probed experimental space can be described by a significant
model (p-value=0.0172), whereby higher reaction tempera-
tures seem to be beneficial (but not significantly, p-value=
0.2286) and reduced amounts of carbon dioxide undoubtedly
increase the yields of 9a and 9b (p-value=0.0078). It should
be noted that p-values (Prob>F) less than 0.0500 indicate sig-
nificant model terms. This outcome prompted us to study the
reaction without any addition of CO2. Indeed, an experiment
performed at 608C, 5 h reaction time, and in the total absence
of CO2, resulted in 65% yield (Table 1, Entry 12). The same be-
havior was found at a slightly higher temperature of 708C and
a prolonged reaction time of 16 h: The presence of 2 g CO2 re-
sulted in 76% yield (Table 1, Entry 13), which could be in-
creased by the total absence of CO2, leading to 91% yield
under neat conditions (Table 1, Entry 14).[8]
Conversion/time profiles at three different CO2 amounts (0,
12, and 24 g weighed into the reactor) confirm the statistical
analysis, which clearly shows a higher conversion at lower CO2
Scheme 3. Lewis acid free Friedel–Crafts alkylation.
&
&
Chem. Eur. J. 2017, 23, 1 – 6
www.chemeurj.org
2
ꢂ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
loadings at any given reaction time (Figure S1 in the Support-
ing Information). In summary, these data demonstrate a retard-
ing, rather than activating effect of CO2 on the reaction under
scrutiny in this set-up and under these conditions.
In this context, it is crucial to note that we discuss the effect
of CO2 in comparison to neat reaction conditions, whereas
Gonzꢁlez-NfflÇez and co-workers used reactions in conventional
solvents (n-hexane, diethyl ether, carbon disulfide, acetonitrile,
and dimethylformamide) as benchmark.[2] Whereas the data of
Gonzꢁlez-NfflÇez may indicate that scCO2 can be a possible sol-
vent with beneficial physicochemical properties in certain
cases, our test reactions lead to the conclusion that it does not
act as activating agent for the ionic reactivity as compared to
neat reaction conditions.
The same qualitative trend was observed upon switching to
tert-butyl chloride as the alkylating agent, from which the
same coupling products 9a and 9b were obtained. Notably,
the presence of carbon dioxide (200 bar) resulted again in a de-
creased conversion. The yields after 16 h at 708C were 28%
with CO2 and 77% without CO2 under otherwise comparable
conditions. In addition, we carried out the reaction of tert-butyl
chloride and 2 under similar reaction conditions to those re-
ported in ref. [2] (608C, 5 h, 4 equiv of 2) in our batch reactor.
Neither in the presence (200 bar) nor in the absence of carbon
dioxide did the amount of products 9a/9b exceed more than
1% yield.
To collect further evidence for the questionable ionizing po-
tential of carbon dioxide, we investigated the degree of ioniza-
tion as a function of the applied reaction medium by high-
pressure UV/Vis spectroscopy. For this purpose, 4,4’,4’’-tri-
methoxytrityl chloride (10), an extraordinarily sensitive indica-
tor molecule for ionization in the visible regime, was used.[9]
The individual degrees of ionization were compared in scCO2,
cyclohexane, toluene, dichloromethane, and acetonitrile
(Figure 1). In all cases, a stainless steel reactor equipped with
two sapphire windows and an external UV/Vis probe was ap-
plied. Complete dissolution of the probe molecule in these dif-
ferent media had been ensured in advance by additional tests.
As can be seen from Figure 1, a broad band at around
490 nm is visible upon dissolving 10 in dichloromethane and
acetonitrile. This is assigned to the carbocation (4-Me-Ph)3C+
that results from dissociation of the CꢀCl bond.[10]
We then focused our efforts on elucidating possible reasons
for the discrepancy between the previous[2] and present re-
sults. Gonzꢁlez-NfflÇez and co-workers have performed a variety
of seemingly well-designed control experiments to verify the
beneficial role of scCO2 as a solvent (see the Supporting Infor-
mation of their paper).[2] While we do not question the actual
data reported by Gonzꢁlez-NfflÇez and co-workers, the primary
difference appears to lie in the individually applied experimen-
tal set-up, procedures, and reference basis.
We considered the possibility that the CO2 used and/or reac-
tor set-up applied by us might contain contaminants that
could inhibit the reaction. This appeared highly unlikely as we
used CO2 with a quality of 99.995% under operating condi-
tions that are checked for compatibility even with highly sensi-
tive organometallic complexes. Nevertheless, we verified our
results by performing the reaction of 8+2 in a different labo-
ratory environment (address given under [b] instead of [a]),
using a completely different infrastructure, two different high-
pressure reactors (one of them built completely new), and two
different CO2 qualities (4.5 and 5.6). Again, reactions performed
under neat conditions were by far more efficient than under
CO2 pressure of 250 bar at 608C (96% vs. 20–30% in reactor
1 and 95% vs. 1–5% in reactor 2, see Table S1 in the Support-
ing Information for further details), thus confirming the results
independently.
The absorbance maxima of the respective carbocation could
be quantified even for trace amounts by means of sophisticat-
ed data treatment (Figure 1). These measurements were per-
formed at room temperature, because 10 is not stable in ace-
tonitrile at elevated temperatures. Spectra of 10 in scCO2, cy-
clohexane, toluene, and DCM at 608C are depicted in the Sup-
porting Information (Figure S2), which confirm the same ab-
sorption behavior at both temperatures.
With this technique, the presence of the 4,4’,4’’-trimethoxy-
trityl cation can be detected in liquid CO2, but in marginal
amounts only. The relative carbocation concentration derived
from peak integration is about 3500 lower than the one ob-
tained in acetonitrile (Table 2). The degree of carbocation for-
mation decreases in the order MeCN>DCM@toluene@cyclo-
hexane>scCO2 following typical solvent polarity scales such
as Reichhardt’s ET(30)[11] in an almost perfect manner. These
findings demonstrate that there is no enhanced dissociation of
compound 10 in scCO2 detectable, which argues against an
“activating effect” on ionization of this alkyl halide in this
medium.
In our experiments, the reactors were loaded with the sub-
strates and filled at room temperature with a weighed amount
of CO2. Thus, the reaction mixture was heated under continu-
ous stirring under a compressed CO2 atmosphere from the be-
ginning. In contrast, the previously reported experimental pro-
cedure involved the addition of carbon dioxide as the last
step, pressurizing an already heated tubular reactor containing
the substrates without any agitation. Taking together the find-
ings of both groups, it is conceivable that the reaction occurs
in the neat state and the supercritical CO2 helps to bring the
reagents into contact in some of the experiments described in
the earlier paper. This could happen, for example, by dissolving
and transporting the alkyl halide out of its container in the
protocols in which both substrates 8 and 2 were placed in the
reactor separately.
Conclusions
In conclusion, our study shows that the presence of CO2 under
supercritical conditions has no activating effect on alkylating
reactions through ionization of potentially SN1-active alkyl ha-
lides, at least for the transformations under scrutiny here. Our
results for the Friedel–Crafts reactions 1+2!3 and 8+2!9,
which are also in line with our observations regarding the
transformation 4+5!6, lead to conclusions significantly dif-
ferent from those previously reported on the basis of reactions
Chem. Eur. J. 2017, 23, 1 – 6
www.chemeurj.org
3
ꢂ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Figure 1. Visible region of the UV/Vis spectra of 4,4’,4’’-trimethoxytrityl chloride (10) in liquid CO2, cyclohexane, toluene, dichloromethane (DCM), and acetoni-
trile at 258C. Individually applied concentrations of 10 are given in the Figure.
the SN1-type transformations.[15] A more extensive study would
Table 2. Quantification of the relative ionization degrees of 10 in differ-
ent media by peak integration of spectra shown in Figure 1. ET(30) values
be needed to clarify the exact mechanistic details of the trans-
formations in each individual case. However, taking all findings
together, we conclude that scCO2 is not endowed with special
characteristics that promote a SN1 ionization of alkyl halides as
activating effect for substitution or coupling reactions with nu-
cleophiles as recently claimed.[2] The physicochemical proper-
ties of scCO2 can be associated with a range of positive effects
on chemical transformations such as reduced melting points,
increased heat and mass transfer, enhanced mixing, etc., and
those effects may also facilitate alkylation reactions that
cannot be conducted neat due to physical limitations. Howev-
er, the results of our present study lead to the conclusion that
the medium does not specifically promote alkyl halide ioniza-
tions in the herein studied typical SN1-type reactions[16] that
normally require protic or polar solvents, Lewis or Brønsted
acids, catalysts, or in special cases photolysis.[17]
are given for comparison.
Solvent
Concentration 10 Normalized ratio of
[mmolL-1]
ET(30)[a]
carbocation formation [kcalmol-1]
scCO2
cyclohexane
toluene
0.2974
0.3122
0.2872
1
2.8
11.3
768
3467
28.5
30.9
33.9
40.7
45.6
dichloromethane 0.0663
acetonitrile 0.0151
[a] Value for scCO2 (measured at 408C and 150 bar) taken from ref. [12],
the values of the organic solvents are taken from ref. [13].
obtained with a different experimental setup and procedure.[2]
They are, however, consistently corroborated with investiga-
tions of the degree of CꢀX dissociation using UV/Vis spectros-
copy for highly sensitive chemical probes. These data and our
interpretation are consistent with the general notion that
scCO2, which has a zero dipole moment and very low polariz-
ability, is generally considered to constitute a medium of very
low polarity.[1,3,11] Beneficial effects on reaction rates are typical-
ly associated with enhanced mass transfer, melting point de-
Experimental Section
See the Supporting Information for full experimental details.
pression, or increased gas solubility.[1] Effects of local clustering Acknowledgements
have also been described but are associated with nonpolar
transition states, or ion-pair formation rather than dissocia-
tion.[3,14]
We are indebted to Prof. Herbert Mayr for his suggestion to
perform UV/Vis measurements to compare the degree of ioni-
zation as a function of the applied reaction medium, in order
to assess the degree of ionization directly. We also gratefully
acknowledge Prof. Sergio Castillon at URWV Tarragona for fruit-
ful discussions and sharing unpublished results on alkylation
reactions in carbohydrate chemistry using scCO2 with us. Fur-
Some of the alkylation reactions studied here appear to be
triggered by the metallic surface of the autoclave when carried
out under high-pressure conditions.[8] Furthermore, the forma-
tion of protons from the electrophilic aromatic substitution re-
actions might also be involved in the mechanisms facilitating
&
&
Chem. Eur. J. 2017, 23, 1 – 6
www.chemeurj.org
4
ꢂ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
[7] Densities were calculated on basis of gas-specific numerical equations
using the web-based tool “Thermophysical Properties of Fluid Systems”
available at http://webbook.nist.gov./chemistry/fluid/.
[8] Additional results on the reaction under neat conditions, which are not
related to the scCO2 effect, are summarized in the Supporting Informa-
tion.
[9] a) H. Mayr, J. Ammer, M. Baidya, B. Maji, T. A. Nigst, A. R. Ofial, T. Singer,
J. Am. Chem. Soc. 2015, 137, 2580–2599; b) E. Price, N. N. Lichtin, Tetra-
hedron Lett. 1960, 1, 10–16.
[10] M. Hojo, H. Hasegawa, H. Tsurui, K. Kawamura, S.-i. Minami, A. Mizobe,
Bull. Chem. Soc. Jpn. 1998, 71, 1619–1627.
[11] C. Reichardt, T. Welton, Solvents and Solvent Effects in Organic Chemistry,
4th ed., Wiley-VCH, Weinheim, 2011.
thermore, we acknowledge the mechanical workshop of the
Max-Planck-Institut für Kohlenforschung for the fabrication of
various reactor types to conduct the control experiments de-
scribed herein. We also thank Ralf Thelen and the mechanical
workshop (ITMC, RWTH Aachen University) and Hanns Eckhardt
(tec5, Oberursel) for their technical contributions concerning
the UV/Vis measurements. Moreover, the technical support of
Niklas Fuhrmann and Lars Winkel (Max-Planck-Institut für Koh-
lenforschung) is highly acknowledged.
[12] R. Eberhardt, S. Lçbbecke, B. Neidhart, C. Reichardt, Liebigs Ann. 1997,
1195–1199.
[13] C. Reichardt, Chem. Rev. 1994, 94, 2319–2358.
Keywords: carbon dioxide · solvent effects · solvolysis ·
supercritical fluids · sustainable chemistry
[14] P. G. Jessop, W. Leitner, Chemical Synthesis using Supercritical Fluids,
Wiley-VCH, Weinheim, 1999.
[1] For recent reviews on the use of scCO2 as reaction medium, see:
a) Handbook of Green Chemistry, Vol. 4: Supercritical Solvents (Eds.: W.
Leitner, P. G. Jessop), Wiley-VCH, Weinheim, 2010; b) W. Leitner, Acc.
Chem. Res. 2002, 35, 746–756; c) G. R. Akien, M. Poliakoff, Green Chem.
2009, 11, 1083–1100.
[2] T. Delgado-Abad, J. Martinez-Ferrer, A. Caballero, A. Olmos, R. Mello,
M. E. Gonzꢁlez-NfflÇez, P. J. PØrez, G. Asensio, Angew. Chem. Int. Ed.
2013, 52, 13298–13301; Angew. Chem. 2013, 125, 13540–13543.
[3] T. Delgado-Abad, J. Martinez-Ferrer, R. Acerete, G. Asensio, R. Mello,
M. E. Gonzꢁlez-NfflÇez, Org. Biomol. Chem. 2016, 14, 6554–6560.
[4] a) J. F. Brennecke, J. E. Chateauneuf, Chem. Rev. 1999, 99, 433–452;
b) M. Saharay, S. Balasubramanian, J. Phys. Chem. B 2007, 111, 387–392;
c) J. F. Kauffman, J. Phys. Chem. A 2001, 105, 3433–3442.
[15] H. Mayr, A. R. Ofial, Acc. Chem. Res. 2016, 49, 952–965.
[16] a) A. Streitwieser, Solvolytic Displacement Reactions, McGraw Hill, New
York, 1962; b) G. A. Olah, P. v. R. Schleyer, Carbonium Ions, Vol. 1–5,
Wiley, New York, 1968–1976; c) C. K. Ingold, Structure and Mechanism
in Organic Chemistry, 2nd ed., Cornell University Press, Ithaca, 1969;
d) P. Vogel, Carbocation Chemistry, Elsevier, Amsterdam, 1985; e) X.
Creary, Advances in Carbocation Chemistry, Vol. 1, JAI, Greenwich, 1989;
f) J. M. Coxon, Advances in Carbocation Chemistry, Vol. 2, JAI, Greenwich,
1995.
[17] J. Ammer, H. Mayr, J. Phys. Org. Chem. 2013, 26, 956–969.
[5] Review of a-alkylation of carbonyl compounds via Lewis acid induced
reaction of enolsilanes with SN1-active alkyl halides: M. T. Reetz, Angew.
Chem. Int. Ed. Engl. 1982, 21, 96–108; Angew. Chem. 1982, 94, 97–109.
[6] a) H. F. Schaller, H. Mayr, Angew. Chem. Int. Ed. 2008, 47, 3958–3961;
Angew. Chem. 2008, 120, 4022–4025; b) H. Mayr, B. Kempf, A. R. Ofial,
Acc. Chem. Res. 2003, 36, 66–77.
Manuscript received: September 1, 2016
Revised: January 23, 2017
Final Article published: && &&, 0000
Chem. Eur. J. 2017, 23, 1 – 6
www.chemeurj.org
5
ꢂ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
FULL PAPER
&
Supercritical Fluids
Y. X. Qiao, N. Theyssen, T. Eifert,
M. A. Liauw, G. Franciꢁ, K. Schenk,
W. Leitner,* M. T. Reetz*
&& – &&
Not super here: Supercritical CO2 as
a solvent does not activate alkyl halides
by SN1-type ionization for alkylation re-
actions as recently claimed.
Concerning the Role of Supercritical
Carbon Dioxide in SN1 Reactions
&
&
Chem. Eur. J. 2017, 23, 1 – 6
www.chemeurj.org
6
ꢂ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!