Solvent choice and kinetic isotope effects (KIEs) dramatically alter regioselectivity in the directed ortho metalation (DoM) of 1,5-dichloro-2,4-dimethoxybenzene

Article abstract of DOI:10.1002/chem.201405981

The regioselective formation of the 6-lithio derivative of 1,5-dichloro-2,4-dimethoxybenzene (i.e., 12) by directed ortho metalation (DoM) with nBuLi in THF is described. Although literature reports suggest direct deprotonation at C6, a series of time-course and labelling studies has revealed that deprotonation rather occurs exclusively at C3 followed by isomerization of the anion to C6. By contrast, when DoM was performed in Et 2O, deprotonation again occurred selectively at C3, but now no isomerization occurs, and electrophilic capture produces the regioisomer of that produced in THF. In these labeling studies, it has been found that deuterium has an enormous kinetic isotope effect (KIE) that suppresses not only the original DoM reaction at C3 when deuterium is present there, but also suppresses isomerization to C6 when the label is at that site. Remarkably, this "protectinggroup" role of the deuterium is unique to THF; in ether, full deprotonation of the deuterium at C3 was observed.

Full text of DOI:10.1002/chem.201405981

DOI: 10.1002/chem.201405981
Communication
&
Synthetic Methods
Solvent Choice and Kinetic Isotope Effects (KIEs) Dramatically
Alter Regioselectivity in the Directed ortho Metalation (DoM) of
1,5-Dichloro-2,4-dimethoxybenzene
Jennifer L. Farmer,[a] Robert D. J. Froese,[b] Edward Lee-Ruff,[a] and Michael G. Organ*[a]
Abstract: The regioselective formation of the 6-lithio de-
rivative of 1,5-dichloro-2,4-dimethoxybenzene (i.e., 12) by
directed ortho metalation (DoM) with nBuLi in THF is de-
scribed. Although literature reports suggest direct depro-
tonation at C6, a series of time-course and labelling stu-
dies has revealed that deprotonation rather occurs exclu-
sively at C3 followed by isomerization of the anion to C6.
By contrast, when DoM was performed in Et2O, depro-
tonation again occurred selectively at C3, but now no
isomerization occurs, and electrophilic capture produces
the regioisomer of that produced in THF. In these labeling
studies, it has been found that deuterium has an enor-
mous kinetic isotope effect (KIE) that suppresses not only
the original DoM reaction at C3 when deuterium is pres-
ent there, but also suppresses isomerization to C6 when
the label is at that site. Remarkably, this “protecting-
group” role of the deuterium is unique to THF; in ether,
full deprotonation of the deuterium at C3 was observed.
Scheme 1. Natural products obtainable from a central DoM reaction of 1,5-
dichloro-2,4-resorcinol derivatives (5).
Directed ortho metalation (DoM)[1,2] is a powerful tool to pre-
pare aryl products of considerable complexity in an efficient
manner by eliminating the need for a highly functionalized
precursor, such as the corresponding halide, thus, reducing
step count and minimizing waste.[3] Although the primary use
of DoM has been to trap the aryllithium directly with electro-
philes by nucleophilic substitution, transmetalation has also
been shown as an effective pathway into cross-coupling
chemistry.[4] This approach appealed to us when we considered
a general strategy to prepare members of the benzofuran cate-
chol-based family of natural products shown in Scheme 1.
Based on a report by Kraus and Zeng,[5] we felt that 5 could be
deprotonated at C6 to provide the lithium salt 4 selectively
that following transmetalation to the corresponding zinc deriv-
ative would be amenable for selective cross-coupling with the
iodide center in 3. Subsequent stereo- and regioselective
cross-coupling with the requisite terprenylborate would give
the desired final targets using methodology worked out in our
laboratories by using our Pd-PEPPSI (PEPPSI=pyridine-en-
hanced precatalyst preparation stabilization and initiation) pre-
catalysts.[6] In the course of these studies, we have uncovered
a dramatic solvent effect on the selectivity of the DoM process,
which in turn has revealed some interesting insights into the
creation of kinetic and thermodynamic aryl anions.
We began our DoM study by lithiating 6 with nBuLi in THF
at ꢀ788C for 30 minutes and quenching with I2 or CD3OD
(Table 1, entries 1 and 5, respectively). As was observed by
Kraus and Zeng,[5] the only products isolated were those result-
ing from substitution at C6 between the two chlorides (e.g., 7),
which had led them to conclude that this was also the site of
deprotonation. Although ortho lithiation between two chlor-
ides is with precedent,[7] we wondered if it was not more likely
that deprotonation was occurring at C3 between the methoxy
groups kinetically and then the resultant anion (i.e., 11,
Scheme 2) was actually isomerizing to the other site (i.e., 12).
There are two competing effects operating herein. The me-
thoxy groups are better templaters[3] that would direct lithia-
tion between them, but the site between the two chlorides is
more acidic.
[a] J. L. Farmer, Prof. E. Lee-Ruff, Prof. M. G. Organ
Department of Chemistry, York University
4700 Keele Street, Toronto, Ontario, M3J1P3 (Canada)
E-mail: organ@yorku.ca
[b] Dr. R. D. J. Froese
The Dow Chemical Company, Midland, Michigan, 48674 (USA)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201405981.
Chem. Eur. J. 2014, 20, 1 – 7
1
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Communication
Table 1. DoM of 6 and electrophilic capture in THF and Et2O.
Table 2. Regioselective lithiation of 1,5-dichloro-2,4-dimethoxybenzene
(6) with nBuLi: a kinetic study.
Entry
Solvent
T [8C]
E
R1
R2
Product yield [%][a]
Entry
t [s][a]
Product Distribution[b]
7/8
1
2
3
4
5
THF
Et2O
Et2O
Et2O
THF
ꢀ78
ꢀ78
25
25
ꢀ78
I2
I2
I2
H
H
I
D
H
I
7 (90)
H
H
H
D
6 (88)[b]
8 (66)
7
8
6
1
0.5
0.5
30
31
31
59
79
5
6
2
2
64
63
39
19
6.2:1
5.2:1
29.5:1
39.5:1
2[c]
3
CD3OD
CD3OD
9 (75)
10 (80)
4
60
[a] Yields are reported for isolated products following silica-gel chroma-
tography. [b] Represents recovered starting material (6).
[a] Time that solution was stirred following addition of nBuLi and prior to
the addition of I2. [b] Ratios were determined by H NMR spectroscopy on
1
crude reaction mixtures. [c] TMEDA (1.2 equiv) was mixed with nBuLi and
stirred for 5 min prior to being added to solution of 6.
To examine how fast isomerization truly is, we turned back
to the lithium/halogen exchange because it operates at near
diffusion-limited rates (Table 3). In this way, 11 could be gener-
ated immediately and completely, thus allowing isomerization
to be more meaningfully tracked. When the reaction was
quenched, just after the last drop of tBuLi was added, all of 8
was consumed and isomerization to 12, which leads to 10,
had proceeded to approximately 70% (entry 1). Isomerization
then clearly slows as it takes one full hour to complete conver-
sion to 10 (entries 3, 5, 6), which is suggestive that isomeriza-
tion occurs by an intermolecular mechanism (vide infra). Tem-
perature is a significant factor, as the same isomerization study
at ꢀ1008C, rather than ꢀ788C, showed far less isomerization
at identical time intervals (e.g., see entries 2 vs. 1 and 4 vs. 3).
Scheme 2. Determination of the thermodynamically more stable anion of 6
through isomerization studies in THF by Li/I exchange at C6 (7) and C3 (8).
To address this, we prepared compounds
7 and 8
(Scheme 2) and reacted them under metal/halogen exchange
conditions with tBuLi, which is necessary to avoid quenching/
alkylation problems associated with the nBuI by-product of the
exchange with nBuLi. Lithiation of 8, after quenching, revealed
that under nearly identical reaction conditions that the DoM
reactions were carried out, isomerization of 11 to anion isomer
12 was complete. This is consistent with the possibility that in
THF 12 is the thermodynamically more stable anion and de-
protonation between the methoxy groups occurs first. In sup-
port of this, lithiation of 7 resulted solely in electrophilic cap-
ture at the site between the chlorides (i.e., 10). Thus, one can
say that the more stable anion resides at C6 and this is also
supported by computation (vide infra).
Solvent effects
Anion stability is greatly influenced by solvent, so we were en-
couraged to look at the DoM process in a lower dielectric/co-
Table 3. Lithium–halogen exchange of 1,5-dichloro-3-iodo-2,4-dimeth-
oxybenzene (8) with tBuLi: a kinetic study.
To get a feel for the rates involved in the various processes
described above, time-course studies were conducted. Depro-
tonation was followed with 6 (Table 2) and was found to be
not fast at ꢀ788C, because roughly 20% of 6 remained after
one minute (entry 4). However, this study did reveal that de-
protonation does indeed occur between the methoxy groups
(C3), at least to some extent, and that isomerization does in
fact occur and that it is very fast. In entry 1, the quench with I2
was done immediately following the last drop of nBuLi. Assum-
ing for now that deprotonation only occurs at C3, isomeriza-
tion of 11 to 12 is clearly very rapid.
Entry
t[a]
Product distribution[b]
9/10
9
10
8
1
2
3
4
5
6
0.5 s
28
47
20
43
20
0
72
53
78
57
80
0
0
2
0
0
0
1:2.6
1:1.1
1:3.9
1:1.3
1:4
0.5 s[c]
30 s
30 s[c]
30 min
60 min
100
0:1
[a] Time that solution stirred following addition of tBuLi and prior to the
addition of CD3OD. [b] Ratios were determined by 1H NMR spectroscopy
on crude reaction mixtures. [c] Reactions were performed at ꢀ1008C.
&
&
Chem. Eur. J. 2014, 20, 1 – 7
www.chemeurj.org
2
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Communication
ordinating solvent. When 6 was lithiated and quenched at
ꢀ788C in Et2O, no apparent reaction took place (Table 1,
entry 2). However, when lithiation was conducted at room
temperature, complete consumption of starting materials was
observed, but remarkably only the quenched products of
anion 11 were observed (entries 3 and 4). This constitutes
a complete reversal of regioselectivity compared to when the
identical reactions were conducted in THF. Although there are
a number of examples in the literature when such an abrupt
change in DoM regioselectivity was observed, they involve
highly coordinating solvents and additives, such as hexame-
thylphosphoramide (HMPA) or tetramethylethylenediamin
(TMEDA) that significantly alter the organolithiums ability to
template to directing groups.[8,9] When TMEDA was added to
the DoM reaction of 6 in THF, there was no change in regiose-
lectivity from that observed with THF alone (Table 2, entry 2 vs.
1). Further, we also found that when TMEDA was added to the
DoM reaction of 6 in Et2O, again there was no change in regio-
selectivity (i.e., the product obtained was from the anion at
C3). Thus, the regiochemical change observed in the present
study with two similar etheral solvents appears to be without
precedent and the DoM selectivity observed is unique when
compared with other reports of solvent changes (e.g., TMEDA)
in the literature.[8,9]
(Table 4). Compound 9 possessing a deuterium between the
two methoxy groups was reacted with nBuLi in the usual way,
and after stirring for 30 minutes was quenched with methanol
(entry 1). Evaluation of the reaction mixture revealed that only
9 was present, which on first glance might give the impression
that no deprotonation occurred owing to a pronounced kinetic
isotope effect (KIE).
Table 4. Effect of deuterium KIE on DoM of compounds 9 and 10.
Entry
Starting
E+
Product distribution[a]
10 13
6
9
1
2
3
4
9
10
9
CH3OH
CH3OH
CD3OD
CD3OD
0
15
0
100
0
0
0
85
0
0
0
100
85
10
0
0
15
Although this solvent effect is a very important discovery for
synthetic purposes, it also offers an interesting mechanistic
puzzle. For example, does the site of deprotonation change in
ether, is it a solubility issue when the reaction in Et2O becomes
heterogeneous following deprotonation, or does the thermo-
dynamic anion stability change that dramatically in the two
solvents? To address these questions, deprotonation was first
conducted at room temperature in Et2O, which leads exclusive-
ly to anion 11 (Scheme 3). Thirty minutes after the addition of
1
[a] Product distributions and ratios were determined by H NMR spectrosco-
py on the crude reaction mixtures.
When compound 10 was treated under the same reaction
conditions, again, mostly the starting material was returned
following the methanol quench (entry 2). Although the course
of events with 9 was uncertain, there is little reason to suspect
that DoM at C3 in 10 had not taken place, but now
without the normally spontaneous rearrangement
due to another KIE for the deuterium at C6. The sig-
nificant, reaction-path-altering KIE at C3 supports the
possibility that 9 may in fact have been deprotonat-
ed, but at C6 and not at C3.[10] To further probe the
mechanism, we repeated these two reactions, but
Scheme 3. Exchange of THF for Et2O following the DoM of 6.
now quenched with a deuterium source instead of
hydrogen to more effectively track reaction progress.
Indeed, when the reaction with 9 was quenched with
nBuLi, the solvent was carefully removed anaerobically, the
flask cooled to ꢀ788C, and then THF was added. After stirring
for an additional 60 minutes, the reaction was quenched with
CD3OD and the only product obtained (i.e., 10) was derived
from the complete isomerization of 11 to 12. Thus, when 11
was formed completely in Et2O, there was now no doubt that
6 was not involved in the conversion of 11 to 12 in THF and
that THF is unquestionably the trigger for isomerization.
CD3OD, full incorporation of a second deuterium occurred,
leading to 13 confirming that the deuterium “protected” C3
and DoM did occur now selectively at C6 (entry 3). The com-
plementary experiment was also done on 10 (entry 4), which
did confirm, as was expected, that the deuterium at C6 had no
effect on normal DoM at C3, and 13 was produced.
In a very surprising result, we found that despite the clearly
huge KIE mentioned above (and discussed further below) asso-
ciated with the deprotonation of the H/D at C3 on 6/9, it is
dramatically solvent dependent. Although the deuterium
blocked deprotonation of 9 in THF, leading nBuLi instead to
deprotonate at C6 (Table 4, entries 1 and 3), the same reaction
in Et2O led to exclusive deprotonation of the deuterium
(Scheme 4). To the best of our knowledge, there is one exam-
Isotope effects
In an attempt to ascertain details about the mechanism of the
deprotonation/isomerization sequence, deuterium-labeled
starting materials were prepared and used in the DoM reaction
Chem. Eur. J. 2014, 20, 1 – 7
www.chemeurj.org
3
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Communication
4.0 kcalmolꢀ1 (G3MP2B3) indicative of high kinetic selectivity
at C3. These two TSs and some of the critical bond parameters
are shown in Figure 1 and for the DoM site, the TS is mostly
CꢀHꢀC bond breaking/forming with the Li idling by interacting
with the oxygen. If we assume that MeLi can reversibly bind to
Scheme 4. Deprotonation of the C3 deuteron in compound 9 by nBuLi in
ether at RT.
ple in the literature, in which the kinetic DoM site was re-
tained, when the hydrogen was replaced by a deuterium, de-
spite a pronounced KIE.[9] However, in that instance, it was the
unique properties of the unusual ethoxyvinyllithium base, rela-
tive to conventional alkyllithiums that was attributed to that
unanticipated outcome. In the current example, it is almost in-
conceivable that two solvents of such similar structure and di-
electric (e values for THF and Et2O are 7.6 and 4.3, respectively)
could have such a profound effect to overcome the large KIE
at C3 on 9.[11]
Figure 1. Important bond-length parameters in the two TSs defining the
selectivity.
Computational studies
M062X and G3MP2B3 calculations were carried out on the de-
protonation of 6 by LiMe as a model for DoM to reduce com-
putational time required to deal with the conformers of
nBuLi.[12] The data presented in Table 5 for deprotonation were
the oxygen atom in the complex, which is likely true due to
the 11.2 (M062X) or 13.4 kcalmolꢀ1 (G3MP2B3) difference be-
tween the TS and the complex, then the “templating” effect
must be kinetic in nature and due to stabilization of the lithi-
um by the oxygen in the transition state. Figure 1 supports
this argument. From the kinetically preferred pathway, two iso-
mers of 11 were computed, one with a distorted LiꢀO interac-
tion, and one without this interaction with the former being
lower in energy. However, the geometric isomer with the Li re-
siding between the chlorides (12) is 3.6 (M062X) or 2.7 kcal
molꢀ1 (G3MP2B3) more stable. Although no low-energy isom-
erization pathways could be discerned, the thermodynamically
more stable product 12 is opposite of the kinetic product, as
was demonstrated experimentally.
Table 5. Thermodynamic parameters for important structures for the lith-
iation reaction of 6 at C6 to form 12 and C3 to form 11.[a]
M062X/6-311+G** in THF
DH DS DG
G3MP2B3 in THF
DS DG
Compound
DH
0.0
6+LiMe
0.0
ꢀ1.4
7.8
0.0
0.0
3.8
0.0
0.0
3.1
16.5
6 complex
TS (6$11)
TS (6$12)
11 +CH4
ꢀ17.4
ꢀ24.1
ꢀ25.4
ꢀ5.0
ꢀ1.7
10.9
14.4
ꢀ18.4
ꢀ21.0
ꢀ16.0
ꢀ18.6
ꢀ20.6
ꢀ3.3
15.0
17.7
10.1
20.5
ꢀ19.5
ꢀ22.6
ꢀ18.0
ꢀ21.6
ꢀ17.4
ꢀ20.1
12+CH4
ꢀ3.5
ꢀ2.9
[a] Calculations use M062X/6-311+G** and G3MP2B3 with PCM/THF. En-
thalpies and free energies (258C) are in kcalmolꢀ1, whereas entropies are
in calmolꢀ1 K relative to 6+LiMe.
Synthetic utility
The remarkable ability to selectively form one anion in one sol-
vent and another anion in a different one offers great opportu-
nities for synthesis. Deprotonation of 6 in THF followed by ad-
dition of either electron-rich (14) or electron-poor (15) alde-
hydes gave selective attack at C6 (Scheme 5). Conversely, the
same sequence conducted in Et2O gave only the products of
addition to C3 (i.e., 16 and 17).
In another application, lithiation of 6 in THF followed by
transmetalation with ZnCl2 gave 18 that upon addition of Pd–
PEPPSI-IPr and oxidative addition partner 19 led exclusively to
the formation of 20 (Scheme 6i).[15,16] Alternatively, DoM in Et2O
followed by transmetalation locks the organometallic at C3
(i.e., 21), because the zinc derivative does not isomerize once
the THF is added that is necessary for the Negishi coupling to
give isomer 22 (Scheme 6ii). Without this unique solvent fea-
ture, the direct cross-coupling of these two different sites on 6
would not be possible.
generated by using MeLi base,[13] whereas Collum and co-work-
ers used Me2Li in their computational studies.[14] Interestingly,
when the polarization continuum model (PCM, THF) was in-
cluded, MeLi binding to the methoxy groups is enthalpically
only slightly exothermic (M062X, ꢀ1.4 kcalmolꢀ1) compared to
the gas phase (ꢀ17.6 kcalmolꢀ1). No MeLi binding to the chlor-
ides could be found in solution whereas it does bind in the
gas phase. Although the entropies are included as bimolecular
values (6 +MeLi), the true entropies will depend on the
degree of solvation/aggregation, thus, the absolute barriers
are not anticipated to be accurate but the selectivities and
thermodynamics will be.
As shown in Table 5, selectivity favors attack at C3 (organo-
lithium 11) over C6 (organolithium 12) by 2.7 (M062X) and
&
&
Chem. Eur. J. 2014, 20, 1 – 7
www.chemeurj.org
4
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Communication
verted to 11 in Et2O, and it was not until the Et2O was re-
moved and the THF added that isomerization to 12 took place.
The unusually large selectivity observed when C3 is deuter-
ated can be rationalized in terms of a rate-determining depro-
tonation of the complexed lithiated species in the DoM pro-
cess. The unusually large deuterium isotope effect (DIE>99)
observed is consistent with tunneling associated with cyclic (5
or 6 centered) transition states, in which small changes in ge-
ometry (narrow barriers) are involved in hydrogen atom- or
proton-transfer reactions.[17] DIE values of 13000 (at ꢀ1508C)
have been reported for the isomerization of 2,4,6-tri-tert-butyl-
phenyl to 3,5-di-tert-butylneophyl.[18] The DIE decreases with
increasing temperatures (1400 and 80 at ꢀ100 and ꢀ308C, re-
spectively) following nonlinear Arrhenius plots typical for those
processes involving tunneling. Such temperature effects have
also been observed in directed lithium deprotonation in the re-
arrangement of organic phosphates to phosphonates where
a near-normal DIE of 6 was observed at ꢀ508C, which in-
creased to ꢁ100 at ꢀ788C in a-deprotonation.[19] The large
KIEs observed in some of these transformations have led to
the use of deuterium as a “protecting group” in directing re-
gioselectivity, in which it is removed at a later stage.[8,20,21] In
the current case, DoM occurs by rate-determining deprotona-
tion at C3 via a 6-centered transition state (see Figure 1 TS
6$11). The corresponding activated complex for the C3-deu-
terated species lies 0.8 kcalmolꢀ1 higher, which would predict
a normal DIE of 7.9 at ꢀ788C in contrast to the observed selec-
tivity, in which the DIE must be at least 99 to account for it.
Deprotonation at C6 does not involve DoM and the difference
in activation energies between the C6 deuterated and proton-
ated transition states of 0.9 kcalmolꢀ1 leads to a normal pre-
dicted value of 10.3 at ꢀ788C, which is more consistent with
the observed selectivity.
Scheme 5. Complementary regioselective nucleophilic additions to alde-
hydes in ether and THF.
In summary, a remarkable solvent effect has led to the selec-
tive creation of two isomeric aryl lithium species during the
DoM of 1,5-dichloro-2,4-dimethoxybenzene (6). Calculations
have been used to determine that the C6-lithio species (12) is
the thermodynamically more stable anion, and that templated
deprotonation at C3 is the kinetically favored process leading
to formation of 11. Kinetic deprotonation at C3 by using nBuLi
in THF can be deflected to C6 by placing a deuterium at C3,
whereas in Et2O, deprotonation (of D+) still occurs at C3.
These unique observations have been shown to have very
useful applications in synthetic chemistry. Regioisomeric prod-
ucts can now be produced with full and absolute regioselectiv-
ity simply by changing the solvent from THF to Et2O, thus elim-
inating the need for high-coordinating solvents, such as
TMEDA or HMPA that are difficult to work with.[8] In fact, the
addition of TMEDA to either the DoM reaction of 6 in THF or
Et2O did not alter the regioselectivity at all. The ability to readi-
ly reduce or further manipulate the chlorides (e.g., cross-
couple) and methoxy groups will allow the preparation of di-
verse, highly substituted aryl compounds as single isomers,
something that remains a formidable challenge in synthetic
chemistry. This is especially important, because aromatic iso-
mers are typically inseparable.
Scheme 6. Complementary regioselective Negishi cross-coupling reaction of
6 in ether and THF.
Mechanistic considerations
At first glance, it would be tempting to suggest that the start-
ing material (6) is lithiated at C3 by nBuLi that is templated
strongly by the methoxy group(s) to produce 11, which then
deprotonates another molecule of starting material to produce
12 and in doing regenerates 6. This would create a cycle that
would systematically work its way until all of 6 is consumed
and the lithiation mixture contains only 12. The rates from
Table 2 might suggest that this is possible, because DoM itself
is comparatively slow, thus giving isomerization sufficient time
to occur. However, the lithium/halogen exchange results in
Table 3 suggest that such a straightforward mechanism is not
the primary means by which 12 is formed. Here, 11 is formed
instantaneously, and it is clear that complete isomerization
takes more than 30 minutes. Thus, there has to be complete
isomerization of 11 to 12 in THF that does not involve 6,
which does not happen in ether. This was further supported
by the results shown in Scheme 3, in which 6 was fully con-
Chem. Eur. J. 2014, 20, 1 – 7
www.chemeurj.org
5
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Communication
[14] S. T. Chadwick, R. S. Rennels, J. L. Rutherford, D. B. Collum, J. Am. Chem.
Soc. 2000, 122, 8640–8647.
Acknowledgements
[15] a) L. A. McCann, M. G. Organ, Angew. Chem. Int. Ed. 2014, 53, 4386–
4389; Angew. Chem. 2014, 126, 4475–4478; b) M. Pompeo, J. L. Farmer,
R. D. J. Froese, M. G. Organ, Angew. Chem. Int. Ed. 2014, 53, 3223–3226;
Angew. Chem. 2014, 126, 3287–3290; c) C. Valente, M. Pompeo, M.
Sayah, M. G. Organ, Org. Process Res. Dev. 2014, 18, 180–190; d) M.
Sayah, M. G. Organ, Chem. Eur. J. 2013, 19, 16196–16199; e) K. H. Hoi,
J. A. Coggan, M. G. Organ, Chem. Eur. J. 2013, 19, 843–845; f) M. Sayah,
A. Lough, M. G. Organ, Chem. Eur. J. 2013, 19, 2749–2756; g) M.
Pompeo, N. Hadei, R. D. J. Froese, M. G. Organ, Angew. Chem. Int. Ed.
2012, 51, 11354–11357; Angew. Chem. 2012, 124, 11516–11519; h) C.
Valente, S. C¸alimsiz, K. H. Hoi, D. Mallik, M. Sayah, M. G. Organ, Angew.
Chem. Int. Ed. 2012, 51, 3314–3332; Angew. Chem. 2012, 124, 3370–
3388.
[16] For earlier reports detailing the transmetallation of aryllithium com-
pounds to the corresponding zinc for Negishi cross-coupling purposes,
see: a) C. E. Tucker, T. N. Majid, P. Knochel, J. Am. Chem. Soc. 1992, 114,
3983–3985; b) P. Knochel, R. D. Singer, Chem. Rev. 1993, 93, 2117–2188;
c) I. Klement, M. Rottlꢁinder, C. E. Tucker, T. N. Majid, P. Knochel, Tetrahe-
dron 1996, 52, 7201–7220.
[17] D. R. Anderson, N. C. Faibish, P. Beak, J. Am. Chem. Soc. 1999, 121,
7553–7558.
[18] a) G. Brunton, D. Griller, L. R. C. Barclay, K. U. Ingold, J. Am. Chem. Soc.
1976, 98, 6803–6811; b) G. Brunton, J. A. Gray, D. Griller, L. R. C. Barclay,
K. U. Ingold, J. Am. Chem. Soc. 1978, 100, 4197–4200.
[19] F. Hammerschmidt, S. Schmidt, Eur. J. Org. Chem. 2000, 2239–2245.
[20] E. Vedejs, J. Little, J. Am. Chem. Soc. 2002, 124, 748–749.
[21] M. F. Wood, S. Bissiriou, C. Lowe, A. M. Norrish, K. Senechal, K. M. Wind-
eatt, S. J. Coles, M. Hursthouse, Org. Biomol. Chem. 2010, 8, 4653–4665.
This work was supported by NSERC (Canada) and Eli Lilly and
Company through the Lilly Research Award Program (LRAP).
Keywords: isotopes
·
Negishi coupling
·
PEPPSI
·
regioselectivity · solvent effects
[1] H. Gilman, R. Bebb, J. Am. Chem. Soc. 1939, 61, 109–112.
[2] G. Wittig, G. Fuhrman, Chem. Ber. 1940, 73, 1197–1218.
[3] For reviews on DoM, see: a) P. Beak, V. Snieckus, Acc. Chem. Res. 1982,
15, 306–312; b) V. Snieckus, Chem. Rev. 1990, 90, 879–933.
[4] a) E. J.-G. Anctil, V. Sniekus, J. Organomet. Chem. 2002, 653, 150–160;
b) J. E. Milne, S. L. Buchwald, J. Am. Chem. Soc. 2004, 126, 13028–13032.
[5] G. A. Kraus, Y. Zeng, Synth. Commun. 2000, 30, 2133–2141.
[6] J. L. Farmer, H. N. Hunter, M. G. Organ, J. Am. Chem. Soc. 2012, 134,
17470–17473.
[7] a) K. Shiina, T. Brennan, H. Gilman, J. Organomet. Chem. 1968, 11-15,
471–477; b) T. H. Kress, M. R. Leanna, Synthesis 1988, 803–805; c) M.
Iwao, J. Org. Chem. 1990, 55, 3622–3627.
[8] For changes in DoM regioselectivity by the inclusion of TMEDA as an
additive, see: a) D. W. Slocum, G. Book, C. A. Jennings, Tetrahedron Lett.
1970, 11, 3443–3445; b) D. W. Slocum, C. A. Jennings, J. Org. Chem.
1976, 41, 3653–3664.
[9] M. Shimano, A. I. Meyers, J. Am. Chem. Soc. 1994, 116, 10815–10816.
[10] J. Clayden, J. H. Pink, N. Westlund, F. X. Wilson, Tetrahedron Lett. 1998,
39, 8377–8380.
[11] Source of values of tetrahydrofuran and Et2O (diethyl ether), see: http://
depts.washington.edu/eooptic/linkfiles/dielectric_chart[1].pdf.
[12] Please see the Supporting Information for the computational details.
[13] N. J. R. van Eikema Hommes, P. v. R. Schleyer, Tetrahedron 1994, 50,
5903–5916.
Received: November 6, 2014
Published online on && &&, 0000
&
&
Chem. Eur. J. 2014, 20, 1 – 7
www.chemeurj.org
6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Communication
COMMUNICATION
&
Synthetic Methods
J. L. Farmer, R. D. J. Froese, E. Lee-Ruff,
M. G. Organ*
&& – &&
Solvent Choice and Kinetic Isotope
Effects (KIEs) Dramatically Alter
Regioselectivity in the Directed ortho
Metalation (DoM) of 1,5-Dichloro-2,4-
dimethoxybenzene
Simply changing the solvent from THF
to diethyl ether dramatically alters the
outcome of directed ortho metalation
(DoM) of 1,5-dichloro-2,4-dimethoxy-
benzene. Kinetic deprotonation in both
solvents occurs between the two me-
thoxy groups; however, in THF, the 3-
lithio derivative rearranges rapidly to
the 6-lithio isomer, which is the thermo-
dynamically more stable anion (see
scheme).
Chem. Eur. J. 2014, 20, 1 – 7
www.chemeurj.org
7
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ