Doubly Diastereoconvergent Preparation and Microsolvation-Controlled Properties of (Z)- and (E)-1′-Lithio-1′-(2,6-dimethylphenyl)propenes

Article abstract of DOI:10.1021/jacs.6b11003

A doubly diastereoconvergent reaction can ad libitum generate either one or the other of two diastereomeric products with complete consumption of the diastereomeric precorsors or their mixtures. Thus, the preparation of configurationally pure (Z)-1′-lithio-1′-(2,6-dimethylphenyl)propene [(Z)-1] from any Z,E mixture of the corresponding bromoalkenes with n-butyllithium succeeded by means of a user-friendly (E)-1 → (Z)-1 configurational interconversion. The subsequent treatment of (Z)-1 with a minimum amount of THF afforded exclusively (E)-1 as the other diastereomeric product and was mediated by a beneficial (Z)-1 → (E)-1 interconversion. This behavior provided microsolvation-controlled choices of highly diastereoselective derivatizations of 1. Low-temperature ¹³C NMR spectra established that (Z)-1 was dissolved as a trisolvated monomer in THF but as a disolvated dimer in monodentate, ethereal, non-THF solvents, whereas (E)-1 was always monomeric. Backed by such knowledge, kinetic experiments indicated that the electrophiles 1-bromobutane or ClSiMe₃ in Et₂O reacted at 32 °C with the tiny (NMR-invisible) population of monomeric (Z)-1 that was formed in a mobile equilibrium from the inactive, predominantly dimeric (Z)-1. The equilibration of monomeric (Z)-1 and (E)-1 in THF as the solvent was fast (seconds on the ¹H NMR time scale), whereas the corresponding stereoinversion of both solvated and unsolvated (E)-1 → (Z)-1 in non-THF solvents occurred on the laboratory time scale (minutes at ambient temperatures). Dicyclopropyl ketone added rapidly to the monomers (Z)-1&3THF and (E)-1&3THF with a rate ratio of at least 14:1 in THF at ?78 °C. Di-tert-butyl ketone added less rapidly to the less shielded (Z)-1 [but never to (E)-1]; this singly diastereoconvergent process was much more slowly reversible in THF.

Full text of DOI:10.1021/jacs.6b11003

Subscriber access provided by University of Newcastle, Australia
Article
Doubly Diastereoconvergent Preparation and Microsolvation-Controlled
Properties of (Z)- and (E)-1´-Lithio-1´-(2,6-dimethylphenyl)propenes
Rudolf Knorr, Claudia Behringer, Monika Knittl, Ulrich von Roman, and Ernst Lattke
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.6b11003 • Publication Date (Web): 03 Mar 2017
Downloaded from http://pubs.acs.org on March 3, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 38
Journal of the American Chemical Society
1
2
3
4
5
Doubly Diastereoconvergent Preparation and Microsolvation-Controlled
Properties of (Z)- and (E)-1´-Lithio-1´-(2,6-dimethylphenyl)propenes
6
7
8
Rudolf Knorr,* Claudia Behringer, Monika Knittl, Ulrich von Roman, and Ernst Lattke
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Department Chemie, Ludwig-Maximilians-Universität München, Butenandtstrasse 5-13 (Haus
F), 81377 München, Germany
KEYWORDS Aggregation; Z,E configurational lability; ketone addition; broken kinetic order;
microsolvation-controlled stereospecificity.
ABSTRACT: A doubly diastereoconvergent reaction can ad libitum generate either one or the
other of two diastereomeric products with complete consumption of the diastereomeric precorsors
or their mixtures. Thus, the preparation of configurationally pure (Z)-1´-lithio-1´-(2,6-
dimethylphenyl)propene [(Z)-1] from any Z,E mixture of the corresponding bromoalkenes with
n-butyllithium succeeded by means of a user-friendly (E)-1 → (Z)-1 configurational
interconversion. The subsequent treatment of (Z)-1 with a minimum amount of THF afforded
exclusively (E)-1 as the other diastereomeric product and was mediated by a beneficial (Z)-1 →
(E)-1 interconversion. This behavior provided microsolvation-controlled choices of highly
diastereoselective derivatizations of 1. Low-temperature 13C NMR spectra established that (Z)-1
was dissolved as a trisolvated monomer in THF but as a disolvated dimer in monodentate,
ethereal, non-THF solvents, whereas (E)-1 was always monomeric. Backed by such knowledge,
kinetic experiments indicated that the electrophiles 1-bromobutane or ClSiMe3 in Et2O reacted at
32 °C with the tiny (NMR-invisible) population of monomeric (Z)-1 that was formed in a mobile
equilibrium from the inactive, predominantly dimeric (Z)-1. The equilibration of monomeric (Z)-
1 and (E)-1 in THF as the solvent was fast (seconds on the 1H NMR time scale), whereas the
corresponding stereoinversion of both solvated and unsolvated (E)-1 → (Z)-1 in non-THF
solvents occurred on the laboratory time scale (minutes at ambient temperatures). Dicyclopropyl
ketone added rapidly to the monomers (Z)-1&3THF and (E)-1&3THF with a rate ratio of at least
14:1 in THF at –78 °C. Di-tert-butyl ketone added less rapidly to the less shielded (Z)-1 [but
never to (E)-1]; this singly diastereoconvergent process was much more slowly reversible in
THF.
ACS Paragon Plus Environment
1

Journal of the American Chemical Society
Page 2 of 38
1. INTRODUCTION
1
2
Stereoconvergent processes are attractive if they afford the same stereoisomer as the main
product from different stereoisomeric sources or their mixtures. This will happen when one of
the stereoisomers of the products or its precursors is sufficiently favored in a mobile equilibrium.
With organolithium reagents,1,2 for instance, the preparative stereoselectivity may depend on the
configuration3 of the lithium-bearing carbon atom (C-α). If so, configurational lability of C-α
(mobile equilibrium) at sufficiently high temperatures may either deteriorate or improve the
stereoselectivity according to the equilibrium constant. A recently4 published example involved
3-siloxy-substituted sec-alkyl iodides whose syn or anti diastereomers or their mixtures were
subjected to I/Li interchange reactions: The emerging 3-siloxy-substituted sec-alkyllithium
compounds equilibrated very rapidly at –100 °C in Et2O/hexane with formation of always the
same (ca. 9:1) syn/anti mixture;5 extensive derivatization experiments4 afforded syn/anti product
mixtures with diastereomeric ratios between 90:10 and 99:1. Obviously, such a single
diastereoconvergence would not provide the alternative (anti) diastereomers as the predominant
products. It was also shown5 that replacement of the 3-siloxy substituent led to slower
conversion at –90 °C with formation of ca. 55:45 syn/anti product mixtures. This suggests that
suitable manipulations of a mobile equilibrium might provide the possibility of designing a
doubly diastereoconvergent system: This should be able to furnish either the one or the other
diastereomer as a preponderant product with complete consumption of the different
diastereomeric precursors or their mixtures. We will report here on the suitably mobile
equilibrium between the title compounds (Z)-1 and (E)-1 (Scheme 1) and its manipulation by
judicious choices of the solvents and temperatures.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 1. (E)-1, (Z)-1 and comparable, published 1´-alkenyllithiums with no (2) or with two
substituents at C-2´ (= C-β) of 3 and 4; Don = solvating, electron-pair donating ligand
coordinated at Li; d = microsolvation number.
ACS Paragon Plus Environment
2

Page 3 of 38
Journal of the American Chemical Society
3
3
1
2
3
4
5
6
4
4
2
2
CH3
H3C
5
5
1
1
6
6
.
.
.
.
(Don)
d
Li
(Don)d Li
H
CH3
CH3
1´
1´
2´
2´
H
CH3
trans
CH3
cis
trans cis
7
8
(E)-1
(Z)-1
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
3
3
4
4
4
2
2
2
CH3
CH3
CH3
5
5
5
1
1
1
6
6
.
.
.
6
.
.
Li
.
Li
Li
CH3
CH3
CH3
α
α
1´
β
2´
β
H
H
H3C
cis
CH3
(H3C)2
(CH3)2
cis
trans
trans
2
3
4
Although the bare preparation protocols will be explained in Section 2.1 for (Z)-1 and in
Section 2.3 for (E)-1 and in more detail in the Experimental Section, we deemed it appropriate to
offer also thorough analytic characterizations (Section 2.2) and detailed dynamic knowledge
(Sections 2.4 – 2.6) of the protagonists (Z,E)-1, so that the structural assignments and kinetic
relations become transparent. Comparisons of (Z,E)-1 with the related 1´-arylalkenyllithiums 2,6
3,7 and 48 will be used to confirm the similarities of the ground states with respect to
comparable structural traits, (non)aggregation states, charge delocalization, and microsolvation9
by electron-pair donor ligands (“Don” in Scheme 1) that coordinate to lithium. The
(non)influence of alkyl substituents at C-2´ (or C-β) will become evident in the series of 1 – 3
especially with respect to the important quantification of configurational lability which is
essential for stereoconvergence: As a model for the expected thermal interconversion of (Z)-1
and (E)-1, the cis/trans stereoinversion of monomeric 2 is known6 to occur rapidly in
tetrahydrofuran (THF) as the solvent at room temperature (rt) through a THF-catalyzed, ionic
mechanism.10 The corresponding stereoinversion of aggregated 2 happened less rapidly6 through
unconsolidated mechanisms in Et2O or tert-butyl methyl ether (t-BuOMe) as the solvent and even
in the absence of such solvating ligands. With this background, it will become comprehensible
why (Z,E)-1 could provide a suitable paradigm of a double diastereoconvergence.
2. RESULTS AND DISCUSSION
2.1. Diastereoconvergent Preparation of (Z)-1&t-BuOMe and its Derivatives
The Br/Li interchange reaction (Scheme 2) of n-butyllithium (n-BuLi) with the 1-aryl-1-
bromopropenes11 (Z,E)-5 to give the 1´-aryl-1´-lithiopropenes 1 should be expected to occur with
configurational retention, so that (Z)-5 (with the olefinic Me trans to aryl) should generate (E)-1.
1
In t-BuOMe as the solvent, however, almost only (Z)-1 could be detected by H NMR
ACS Paragon Plus Environment
3

Journal of the American Chemical Society
Page 4 of 38
spectroscopy in situ after the complete, fast consumption of any (Z,E)-5 mixture: Derivatization
by solid CO2 furnished the carboxylic acid (E)-8 as the only detectable diastereomer whose
1
2
3
4
5
1
3
configuration followed from the typically smaller cis H,13C NMR coupling constant J(2´-
6
7
8
9
H,CO2H) = 6.9 Hz. Consistently, the alkene (Z)-7 was obtained through protolysis of another
3
such (Z)-1 solution and exhibited the characteristic cis interproton NMR coupling constant J(1´-
H,2´-H) = 11.3 Hz. This diastereoconvergent preparation of (Z)-1 provided a first (not
unprecedented) indication of configurational lability of 1: The primary product (E)-1, as
generated from (Z)-5, appeared to have been converted rapidly and (almost) completely into (Z)-
1. To be sure, we demonstrated that the bromopropene (Z)-5 reacted hardly slower than the
accompanying (E)-5 with a deficient amount (0.5 equiv) of n-BuLi.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 2. Preparation and derivatization of the 1´-aryl-1´-lithiopropenes (Z)-1 and (E)-1.
3
4
2
CH3
H3C
5
7´ 6´
H3C-CH2-CH2-CH2
5´
4´
1
3´
6
Br
CH3
CH3
2´
H3C
H3C
cis
H
H
trans cis
trans
(E)- and (Z)-5
(E)- and (Z)-6
+ n-BuLi
n-BuBr
-
3
3
4
4
2
2
CH3
H3C
5
5
1
1
6
6
.
.
.
(t-BuOMe)
.
Li
(THF)3 Li
H3C
1
CH3
CH3
1´
1´
+ 3 THF
2´
2´
H3C
cis
H
H
-
t-BuOMe
trans cis
trans
dimeric (Z)-1
monomeric (E)-1
3
3
4
4
2
2
CH3
H3C
5
5
R
1
1
6
6
R
R
CH3
CH3
1´
1´
(Z)-7.
(E)-8. CO2H (Z)-8
(E)-9. (Z)-9
(E)-10 . SnMe3 (Z)-10
H
(E)-7
2´
2´
H3C
H3C
cis
H
H
trans cis
trans
I
(E)-7, (Z)-8
-
10
(Z)-7, (E)-8 - 10
The t-BuOMe solutions of (Z)-1 were unsuitable for long-term measurements, because (Z)-1
was slowly butylated by its co-product 1-bromobutane (n-BuBr in Scheme 2) at rt with formation
of LiBr and the hydrocarbons 6 (Z/E = 49:1), accompanied by ca. 10% of the alkenes 7 (Z/E =
9:1). A two-dimensional interproton NOESY experiment established that the olefinic 2´-H and
the CH2-4´ group of (Z)-6 were on the same side of the C=C double bond and that 2´-H of (E)-6
ACS Paragon Plus Environment
4

Page 5 of 38
Journal of the American Chemical Society
was instead close to the 2,6-dimethyl groups. This butylation proceeded with comparable
velocity (typical first half-reaction time t1/2 = ca. 70 min at 32 °C in Section 2.5) in Et2O as the
solvent but afforded a significantly increased portion of the minor product (E)-6 (Z/E = 86:14).
In the absence of such electron-pair donor solvents, the butylation occurred more slowly in
pentane, benzene, or toluene. In THF, however, it was regrettably fast even at –70 °C, yielding
(Z)-6 and (E)-6 in a 62:38 ratio from the (Z)-1/(E)-1 = 36:64 mixture to be reported further
below; thus, the obviously faster butylation of (Z)-1 afforded the main product (Z)-6. Hence, it
appeared desirable to get rid of the interfering n-BuBr, at least for projected investigations in
ethereal solvents. Gratifyingly, the generally applicable method of distillative removal of n-BuBr
became unnecessary through the precipitation of (Z)-1 from cold t-BuOMe solutions: Suitable
manipulations (described in the Experimental Section) of the precipitate enabled us to grow
colorless platelets that were sufficiently heavy to settle quickly down from the solution. The
supernatant containing all contaminations (n-BuBr, LiBr, etc.) could then be withdrawn under
inert gas cover, and the remnant platelets were washed with dry pentane. After dissolution in dry
Et2O or THF, the purified platelets appeared to have carried one t-BuOMe ligand per carbanion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
unit (found 1.1 ± 0.1 equiv, now at
δH = 3.10 ppm since released from coordination to Li). This
substance, (Z)-1&t-BuOMe, decomposed through proton transfer from the following two solvents
at rt with rather long t1/2 values in the absence of n-BuBr, moisture, and oxygen: In t-BuOMe,
the formation of the alkenes 7 (Z/E = 81:19) and isobutene (from t-BuOMe) with t1/2 ≈ 8 days
pointed to an NMR-invisible amount of (E)-1 that gave (E)-7 and was more reactive than the
preponderant diastereomer (Z)-1 that furnished (Z)-7. In THF, the alkenes 7 (Z/E = ca. 1:1, along
with ethylene from THF) emerged with t1/2 = ca. 3 days and catalyzed7 a slow vinyl-to-allyl
anion isomerization of 1. The LiBr-free crystals of (Z)-1&t-BuOMe were used in most of the
following investigations, which implied that optimized purity may have been preferred to
optimized yields (with respect to the initial bromopropenes 5) of the crude products. For
instance, purified (Z)-1 in t-BuOMe and elemental iodine afforded only 43% of the iodoalkenes 9
in an E/Z = 94:6 ratio, which pointed again to an enhanced reactivity of an invisible trace of (E)-1
that formed (Z)-9 (6%).
Up to this point, we achieved a singly diastereoconvergent preparation of (Z)-1 as the highly
preponderant equilibrium component. This means that a planned diastereoconvergent formation
of (E)-1 rather than (Z)-1 from the same precursors must employ a method that disturbs the Z,E
equilibrium in the opposite direction, as will be shown in the later Section 2.3.
2.2. NMR Analyses and Properties of (Z)-1 and (E)-1 in Solution
ACS Paragon Plus Environment
5

Journal of the American Chemical Society
Page 6 of 38
The aggregational states of isotope-labeled [6Li]1 became evident through 13C NMR
1
2
3
spectroscopy at low temperatures, where intermolecular scrambling of the Li cations was so slow
4
5
6
7
8
9
6
that nuclear magnetic coupling of a carbanionic center 13C-1´ (= C-α) with n Li nuclei became
6
resolved. For the Li isotope with its small nuclear quadrupole moment and its spin quantum
number I = 1, this coupling can split the 13C-1´ resonance line into 2nI + 1 components,
7
producing simpler multiplicities and often better resolution than the Li (I = 3/2) isotope does.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Thus, the CLi2 moiety of a dimeric aggregate (n = 2) should be recognized through a quintet
splitting of 13C-1´ with a 1:2:3:2:1 intensity ratio. We discovered these quintets (Figure 1a–c) of
crystalline (Z)-1&t-BuOMe in Et2O, t-BuOMe, and [D8]toluene as the solvents: The scalar, one-
1
bond NMR coupling constants J(13C,6Li) = 7.6 or 7.5 Hz remained visible up to (at least) –40, –
25, and –10 °C, respectively, at practically identical resonance positions of
δ(C-1´) = 189.9(2)
ppm; therefore, (Z)-1&t-BuOMe is a dimer rather than a cyclooligomer12 (also with n = 2), just
as had been observed with the corresponding α-arylvinyllithiums13 2&t-BuOMe and 2&Et2O.
[The full 13C NMR spectra of clean dimeric (Z)-1&t-BuOMe as the only species of 1 in either t-
1
BuOMe or toluene at both 25 and –70 °C are shown in Figure S4.]14 A similar quintet with JCLi
= ca. 7 Hz was detected for the THF-solvated dimer of 1 in toluene solution; but its resonance
position at
δ
(C-1´) = 192.8(2) ppm in Table S8a14 seemed hardly compatible with
C values resembled those of dimeric (Z)-1
δ(C-1´) of
the dimers in Figure 1a–c, although all of its other
δ
with the non-THF donor ligands (Tables S5a–S7a).14 Despite such doubt, the Z configuration of
this THF-solvated dimer was confirmed by the characteristic magnitude of its three-bond NMR
3
trans coupling constant J(2´-H,C-1) = 13.5 Hz. Furthermore, the correlation peaks in a two-
dimensional HOESY (6Li,1H) NMR experiment revealed that 6Li was cis to either 2´-H in this
dimer of (Z)-1&THF or to 2´-CH3 in the accompanying monomeric (E)-1&3THF. (Of course,
6Li correlated also with THF and the 2-/6-CH3 protons.) We ascribed the increased
δ(C-1´)
value to a superior donor quality of coordinating THF as compared with the weaker non-THF
donors.
ACS Paragon Plus Environment
6

Page 7 of 38
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
6
Figure 1. 13C NMR splitting (1JCLi) of C-1´ (= C-α) by Li: (a) Dimeric (Z)-1 in Et2O at –90 °C
1
1
with JCLi = 7.6 Hz. (b) Dimeric (Z)-1 in t-BuOMe at –70 °C with JCLi = 7.6 Hz. (c) Dimeric
1
(Z)-1&t-BuOMe in [D8]toluene at –70 °C with JCLi = 7.5 Hz. (d) Monomeric (E)-1 and (Z)-1 in
1
THF at –88 °C, both with JCLi = 11.4 Hz. Scales are in ppm at 100.6 MHz.
6
Microsolvation numbers d of electron-pair donating ligands, coordinated at the Li cation,
may be read from the magnitudes of JCLi by means of the empirical equation 1,9 where a is
1
the number of carbanionic centers (C-1´ or C-α) that are in contact with a certain Li nucleus
under consideration. Using the sensitivity factor L = 46 (±1) Hz, as published6 for 2, we
1
calculated d = 1 per Li from the practically equal JCLi values (entries 1–5 in Table 1) of the
dimers (n = a = 2). Hence, (Z)-1 formed disolvated dimers in analogy with those encountered13
in the X-ray structures of 2; this applied also to the above-mentioned THF-solvated dimer (d = 1)
of (Z)-1, as indicated in footnote a of Table S8a.14 These results from the primary NMR criterion
1
of JCLi (eq 1) enabled us to recognize disolvated dimers (Z)-1 also by the secondary NMR
criteria of lithiation shifts ꢀ
(RLi) from special effects that are present also in the “parent” olefin RH, so that the effects of
lithiation may be seen more clearly. Thus, most of the ꢀ values of (Z)-1 are closely similar in
δ = δ(RLi) – δ(RH), which serve to purge the chemical shift values
δ
δ
the three different solvents of entries 1–3 of Table 1, whereas the substantially greater ꢀ
δ
ACS Paragon Plus Environment
7

Journal of the American Chemical Society
Page 8 of 38
magnitudes in entries 6 and 7 exclude a disolvated, dimeric structure for both (Z)-1 and (E)-1 in
THF as the solvent. Actually, the 13C-1´ resonances of both (Z)-1 and (E)-1 in THF were split by
1
2
3
4
5
6
7
8
9
6
the Li (I = 1) nucleus into 2nI + 1 = 3 equally strong components (Figure 1d), which requires
that n = 1 Li nucleus only is in contact with the 13C-1´ center. The temperature-dependent
line-shapes of these triplets are shown14 in Figure S5: triplet coalescences (loss of the 13C,6Li
6
6
spin coherence) through intermolecular Li scrambling is seen to be faster for the sterically less
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
shielded (Z)-1 than for (E)-1. Since both of these C–Li1 species have neither bidentate ligands
nor side-chain donors available for coordination to a second Li-bearing particle, both (Z)-1 and
(E)-1 must be monomeric in THF, so that a = 1 in eq 1. With L = 46 Hz as above, the frequency
1
intervals of JCLi = 11.4 Hz of the 13C-1´ triplet resonances were translated into the
microsolvation number d = 3. Hence, trisolvation by THF applied not only to monomeric α-
arylvinyllithium 2 (entry 9)6 and monomeric 4 with two β-tert-alkyl groups8 (entry 10), but also
to the (Z)-1 and (E)-1 monomers and to monomeric 1´-lithio-2´-methylpropene7 3 (entry 8, and
Table S1314). In contrast, most of the ꢀ
δ values in entries 6–9 depend significantly on the
changing β- or 2´-substituents, with two exceptions: The para-positioned nuclei C-4 and 4-H are
sufficiently remote from both C-1´ and C-2´ to be rather insensitive to proximity effects of both
Li and the 2´-substituents. Instead, ꢀδ(C-4) and ꢀδ
(4-H) in Table 1 were thought6,9,15,16 to be
partially due to the delocalization of negative electric charge from the Li–C(1´) bond into the π-
electron system on the 1´-aryl plane that is held perpendicular to the C=C double-bond plane by
means of a substantial energetic barrier against rotation about the C(1´)–C(1) single bond.15
Compared with entries 6–10, this lihiation-caused π-charge accumulation at C-4 is weaker for the
dimeric species (entries 1–5), because the second Li cation at C-1´ counteracts the π-electron
delocalization from C-1´. Although interpretations of the other ꢀ
δ data are less obvious, their
values characterize the various families of monomers and aggregates and their appertaining
microsolvation numbers d within the validity limits imposed by 13C-1´ NMR splitting.
Complete ꢀ
δ
sets of the various species may be inspected in Figure S1.14
d = L × (n × 1JCLi)–1 – a
(1)
The results of this section established that the ground states of (Z,E)-1 and 2 satisfy the same
NMR-spectroscopic criteria and that all three of them are exclusively monomeric in THF. Like
2, (Z)-1&t-BuOMe generated disolvated dimers in Et2O, t-BuOMe, and toluene. On the other
hand, (E)-1 did not form a persistent dimer and not even a Z,E-mixed dimer, presumably due to
its trans-2´-CH3 substituent that shields the Li–C bond.
ACS Paragon Plus Environment
8

Page 9 of 38
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Table 1. Microsolvation Numbers d and NMR Data of 1´-Lithio-1´-(2,6-dimethylphenyl)propenes (1) compared with the
Related α-Arylalkenyllithiums 2, 3, and 4. a
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
agg b 1J(13C,6Li) d
.
lithiation NMR shifts ꢀ
δ
[ppm] c
entry cpd. solvent
no.
donor
(equiv)
[Hz]
C-1´ C-2´ C-1
C-4 4-H
°C d
–86
–90
ref
e
1
(Z)-1 t-BuOMe t-BuOMe
D
D
qi, 7.6
qi, 7.6
qi, 7.5
qi, 7.5
qi, 7.5
t, 11.4
t, 11.4
t, 11.5(1)
t, 11.6
t, 10.7
1
1
1
1
1
3
3
3
3
3
0
0
+62.2 –7.9 +17.5 –7.4 –0.45
+62.6 –7.4 +17.2 –7.3 –0.45
+62.3 –7.4 +17.5 –7.0 –0.07 f –70
+68.2 –3.6 +19.6 –6.5 –0.02 f –84
2
(Z)-1 Et2O
Et2O
e
3
(Z)-1 toluene
t-BuOMe (1.9) D
t-BuOMe (1.3) D
e
4
toluene
6
e
2
5
Et2O
Et2O
THF
THF
THF
THF
THF
none
none
D
+57.5 –1.1 +19.6 –7.0 –0.48
+69.7 –14.7 +21.6 –11.8 –0.80
+68.5 –15.8 +23.5 –10.7 –0.77
+69.6 –13.6 +25.5 –10.9 –0.79
+76.4 –10.9 +24.2 –11.0 –0.76
–71
–88
–88
3
6
(Z)-1 THF
(E)-1 THF
M
M
M
M
M
e
7
e
8
THF
THF
THF
–117 e
–89
3
2
4
9
6
10
11
12
+64.0 –17.4 +20.8 –12.6 –0.84 Z –85 8
(Z)-1 C5H10
> D none
> D none
+52.0 +2.1 +13.8 –3.9 –0.21
+57.4 +6.3 +15.5 –3.5 –0.19
–57
–35
e
C5H10
6
2
a 1´ = α, 2´ = β, qi = quintet (1:2:3:2:1), t = triplet (1:1:1), C5H10 = cyclopentane. b “D” = dimer, “M” = monomer. ꢀ
δ =
c
e
f
δ(RLi) –
δ
(RH). d Temperature of ꢀ
δ
determination. This work. Effect of the aromatic solvent.
ACS Paragon Plus Environment
9

Journal of the American Chemical Society
Page 10 of 38
1
2
3
4
5
6
7
8
9
2.3. Z,E Equilibrations leading to (E)-1&3THF or Unsolvated (Z)-1 and their Derivatives
The disolvated dimer of (Z)-1 remained unchanged in the solvents t-BuOMe, Et2O, and
toluene, as shown by the insignificant temperature-dependencies of almost all of their chemical
shifts δH and δC between –90 and +25 °C (Figure S4 and Tables S5–S7).14 In contrast, the
purified crystals of (Z)-1&t-BuOMe changed in THF solution immdiately to the equilibrium
mixture of monomeric (Z)-1&3THF and (E)-1&3THF in concentration-independent ratios that
varied from 52:48 (at +25 °C) to 32:68 (at –88 °C) as shown14 in Figures 1d and S5 and Table
S10. This trend established that the fast Z → E transformation of the monomers is exothermic in
THF. In fact, the quantitative analysis (Table S2)14 furnished a stereoinversion enthalpy of
ꢀH0(Z→E) = –0.84 (±0.05) kcal mol–1 and an entropic penalty of only ꢀS0 = –3.0 (±0.3) cal mol–
1 K–1, the latter as expected17 since unchanged microsolvation numbers (as encountered here) do
not contribute to ꢀS0. Complete formation of monomeric (E)-1 was achieved through
perturbation of that equilibrium by crystallization in the following manner. Purified (Z)-1&t-
BuOMe was completely dissolved in the minimum amount of anhydrous THF (ca. 4 equiv) and
then cooled to –70 °C. The finely felted needles of (E)-1&3THF precipitated after 15 min and
were kept undisturbed for a short while, whereupon (still and always at –70 °C) they were
washed with dry pentane and dissolved for the immediate trapping with an electrophile, so to
avoid the E → Z conversion that will be quantified in Section 2.4. For example, carboxylation of
such a solution of (E)-1&3THF in dry toluene furnished the expected carboxylic acid (Z)-8
(Scheme 2) exclusively, whose configuration followed from the NMR trans coupling constant
3J(2´-H,CO2H) = 13.8 Hz. Similarly, titration at –70 °C with a toluene solution of elemental
iodine afforded the iodoalkene (Z)-9 with not more than a trace of (E)-9; the Z configuration of
this liquid diastereomer (Z)-9 was established through the NOESY correlation peak of 2´-H with
the 2,6-CH3 protons. Trapping of (E)-1&3THF in dry toluene at –70 °C with ClSnMe3
appeared to proceed less rapidly, since formation of the diastereomeric stannanes (Z)-10 and (E)-
10 in a 92:8 ratio may have occurred in parallel with some (E)-1 to (Z)-1 transformation. The
stannane configurations were recognized through the trans and cis NMR coupling constants
3J(2´-H,119Sn) = 141.8 Hz for (Z)-10 and 76 Hz for (E)-10, respectively. Stannane (Z)-10 was
hoped to be a suitable precursor of unsolvated (E)-1. However, the Sn/Li interchange reaction of
(Z)-10 in pentane as the solvent with either n-BuLi or tert-butyllithium (t-BuLi) turned out to be
limited to the extremely slow formation of methyllithium (MeLi) through Sn–Me bond cleavage.
Admixtures of t-BuOMe (up to 20 equiv) induced the slow production of (Z)-1 in parallel with
the appearance of MeLi and isobutene (from t-BuOMe with t-BuLi); but significant
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
10

Page 11 of 38
Journal of the American Chemical Society
improvements were not achieved even in t-BuOMe as the solvent. On the other hand, the desired
Sn–C(1´) bond cleavage of (Z)-10 in pentane became the main route with t-BuLi (4 equiv) in
presence of 1,2-bis(dimethylamino)ethane (TMEDA, at least 2 equiv) and was almost complete
(87%) already within 15 min at rt. The Z/E = 3:1 ratio of the emerging 1 was confirmed through
carboxylation; it revealed that the expected primary product (E)-1 [from (Z)-10] had partially
isomerized into (Z)-1 at or below rt in TMEDA/pentane.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Donor-free (Z)-1 was finally obtained through the instantaneous I/Li interchange reaction
(Scheme 3) of n-BuLi (1.1 equiv) with the liquid iodoalkene (Z)-9 in cyclopentane at rt. The co-
product 1-iodobutane (n-BuI) did not react with (Z)-1 in the donor-free solvent during many days
at rt, whereafter carboxylation furnished the acid (E)-8 exclusively. Again, the primary product
(E)-1, expected from (Z)-9, was NMR-invisible because it had readily (< 5 min at ≤ rt) and
entirely been converted into unsolvated (Z)-1. (The mechanism of this donor-free stereoinversion
1
is unknown.18) The H and 13C NMR signals of donor-free (Z)-1 were strangely broad at rt but
became resolved on cooling (Table S11).14 The resultant ꢀ
substantially from those of the solvated dimers and monomers (entries 1–10) in the direction of
donor-free 2 (entry 12). For instance, the diminished magnitudes of ꢀ (C-4) and ꢀ (4-H)
δ data (entry 11 of Table 1) deviated
δ
δ
indicated that π-charge delocalization from the Li–C(1´) bond was much less efficient than in
entries 1–10, possibly due to n > 2 donor-free Li cations bound to C(1´). Regrettably, unsolvated
6
(Z)-1 and 2 did not show resolved Li,13C(1´) splitting that would have established the CLin
connectivities; but the diminution of ꢀδ(C-1´) = +62 ppm for the disolvated dimers in entries 1–
3 to +52 ppm (entry 11) resembled the decrease by 11 ppm from disolvated, dimeric 2 (entry 4 of
Table 1) to +57 ppm for unsolvated 2 (entry 12). Only 4 ppm of these diminutions should be
ascribed to the loss of the last donor ligand per Li, as suggested by a published series with the
same connectivities n = a = 2 and d = 1 versus d = 0.19 The remainder of roughly 6 ppm may
be due to a higher than dimeric aggregation (n = a > 2) of unsolvated (Z)-1 and 2.
Scheme 3. Generation of donor-free (E)-1 which rearranged
readily to unsolvated (Z)-1 (aryl = 2,6-dimethylphenyl).
1´ aryl
I
1´ aryl
1´ Li
Li
aryl
+ n-BuLi
n-BuI
o
-
2´
H3C
H
2´
2´
H3C
H
H3C
H
(Z)-9
(in cyclopentane)
(Z)-1
(> dimeric)
(E)-1
(NMR-invisible)
The preparative results of this section commend that derivatizations of the sterically less
impeded (Z)-1&t-BuOMe should be performed in Et2O, t-BuOMe, or hydrocarbons as the
ACS Paragon Plus Environment
11

Journal of the American Chemical Society
Page 12 of 38
solvents at or below rt, whereas (E)-1&3THF should be derivatized in toluene solution at or
below –78 °C very soon after its preparation. The following two sections provide more
instructive details of Z,E interconversion, dimerization of (Z)-1, a broken-order reaction
mechanism of (Z)-1 derivations, and the significantly different reactivities of (Z)-1 and (E)-1.
1
2
3
4
5
6
7
8
9
2.4. Dynamics of the Z,E Stereoinversion of 1
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The temperature-dependent time scales of Z,E stereoinversion collected in this Section are
intended to provide approximate rate standards for use in the later Sections 2.5 and 2.6. Thus, the
preceding Sections 2.1 and 2.3 presented already some qualitative evidence about the
configurational lability of the diastereomeric alkenyllithiums (Z,E)-1: Unsolvated (donor-free)
(E)-1, obtained from (Z)-9 in Scheme 3, was NMR-invisible because it isomerized into donor-
free (Z)-1 within less than a few minutes (laboratory time scale). A similar behavior of 1 in t-
BuOMe had led to the diastereoconvergent preparation of (Z)-1&t-BuOMe in Section 2.1. Other
than in THF (to be described further below), the Z,E interconversion rates in these non-THF
solvents remained far below the rate regime of possible NMR coalescences, as shown by the
well-resolved NMR signals of the equilibrium component (E)-1 at rt in Et2O (Table S6b)14 or
toluene (with 4 or 15 equiv of THF in Tables S8 and S9). The crystalline, trisolvated monomer
(E)-1&3THF (Section 2.3) maintained its configuration for some time if held at –78 °C in
toluene as the solvent. We report now that such a toluene solution admitted stereoinversion with
the first-order rate constant k(E→Z) = 0.00040 s–1 at –25 °C, as recognized through its
concentration-independent half-reaction time of 29 min that corresponds to a barrier of ꢀG‡(–25
°C) = 18.3 kcal mol–1. This run (Figure S2, Table S1)14 terminated with the Z/E = 83:17
equilibrium ratio. Due to the shortage of THF ligands (total only 3.4 equiv, as expected), this
equilibrated mixture contained mainly disolvated dimeric (Z)-1 along with trisolvated monomeric
(E)-1. We notice that monomeric (Z)-1 was NMR-visible only in cases of solvation by THF
(Section 2.1 and Tables S8–S10).
In THF as the solvent, however, the (E)-1&3THF → (Z)-1&3THF stereoinversion was ≥185-
fold faster than above in toluene at –25 °C, as estimated in the following manner. The two
1
olefinic H NMR quartets (2´-H) of monomeric (Z)-1 and (E)-1 in THF at 32 °C were unresolved
because already in coalescence at 80 MHz, which means that their interconversion was a little
faster than the corresponding diastereotopomerization20 of LiBr-free 2&3THF in THF.
Regrettably, proper rate measurements by total line-shape analyses appeared not feasible with 1
in view of the menacing vinyl-to-allyl anion isomerization7 mentioned in Section 2.1. It
appeared reasonable, however, to anticipate that a single 2´-CH3 substituent in 1 had little
ACS Paragon Plus Environment
12

Page 13 of 38
Journal of the American Chemical Society
influence on the Z,E (or cis,trans) interconversion rate of the monomers [as also on the
dimerization equilibrium of (Z)-1 to be detailed in Section 2.5], so that the known6
1
2
3
4
5
6
7
8
9
stereoinversion rates of 2 might serve as lower limits for the rates of 1 also at other temperatures.
Therefore, we used the extrapolated value ꢀG‡(–25 °C) = 15.7 kcal mol–1 of 2 in THF6 for
calculating kψ(–25 °C) = ca. 0.074 s–1 as a lower limit for 1, which led to the above factor of
kψ(–25 °C)/k(E→Z) ≥ 0.074/0.00040 = ca. 185. This acceleration is partially due to THF
catalysis in the ionic stereoinversion mechanism.10 For a further kinetic orientation, we list here
the extrapolated half-reaction times [t1/2 = (ln2)/kψ] of 2 as upper limits for (Z)-1 in THF: 0.007 s
at 95 °C, 0.16 s at 32 °C, 0.24 s at 25 °C, 9.4 s at –25 °C, 37 s at –40 °C, and 54 min at –78 °C.
On the other hand, the second 2´-CH3 group in 3 (Scheme 1) was shown as follows to decelerate
the E,Z stereoinversion process considerably. In view of the known7 vinyl-to-allyl anion
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
isomerization at about rt, a THF solution of 3 was heated quickly up to 95 °C, where the two H
NMR (60 MHz) 2´-CH3 singlets began to exhibit weak line broadening. The usual total line-
shape analysis led to an estimated maximum of the first-order rate constant of kψ(+95 °C) ≤ 10
(±3) s–1 and a corresponding barrier of ꢀG‡(+95 °C) ≥ 20.0(3) kcal mol–1 for the cis/trans 2´-
methyl interconversion (methyl diastereotopomerization) of 3, to be compared with an
extrapolated20 value of ꢀG‡(+95 °C) = 18.4 kcal mol–1 for monomeric 2 in THF which pointed
to a substantial (ca. 10-fold) retardation of monomeric 3 as compared with monmeric 2.
The data collected in this section established the conditions which provide for a sufficiently
high mobility of the Z,E equilibration of 1.
2.5. Dimerization and the 1.5th-Order Rate Equations of (Z)-1 with RHal
The much faster monomer/dimer (M/D) interconversion of 1 could not be measured in THF
(because of no dimers) and in t-BuOMe (no monomers) or in Et2O [no monomeric (Z)-1], but it
was detected in toluene as the solvent with some THF (4 or 15 equiv, Tables S8 and S9)14 to
occur on the 1H and 13C NMR time scales (that is, very fast), as observed through line broadening
and decoalescences between –32 and –68 °C. Recalling from above (Section 2.2 and footnote a
in Table 8a) the determinations of dD = 1 for disolvated, dimeric (Z)-1&THF (“D”) and dM = 3
for trisolvated, monomeric (Z)-1&3THF (“M”) in toluene, we were able to measure a few of the
temperature-dependent equilibrium constants KMD (not reported here), as defined further below
(eq 2). The small number of these measured KMD values did not suffice for a reliable
thermodynamic analysis; but they were found to be numerically similar to the KMD values of the
endothermic dimerization21 of 2 under similar conditions. This agreement between (Z)-1 and 2
ACS Paragon Plus Environment
13

Journal of the American Chemical Society
Page 14 of 38
established that the single 2´-CH3 cis-substituent in (Z)-1 had little effect on the dimerization
equilibrium.
1
2
3
4
5
6
7
8
9
In Et2O as the solvent, the small amount (≤ 6%) of (E)-1 in Et2O did not provide 13C NMR
information, so that its aggregational state and microsolvation number remained undetermined.
Dimeric, disolvated (Z)-1 (namely, dD = 1 in D&2Et2O) was the main component (Table S6),14
whereas monomeric (Z)-1 (M&dMEt2O) remained NMR-invisible down to –90 °C, so that its
equilibria with dimeric (Z)-1 and with the minor component (E)-1 (known to be mobile on the
laboratory time scale) could not be measured. In spite of such a tiny population of monomeric
(Z)-1, the following two kinetic experiments suggested that (Z)-1 was the most reactive species in
Et2O solution. The reaction of 1 with ClSiMe3 produced both diastereomers of 11 (Scheme 4),
whose configurations were recognized through their interproton NOESY correlations: (E)-11 had
2´-H near to SiMe3 but 2´-CH3 near to 2,6-CH3, whereas (Z)-11 showed 2´-CH3 to be near to
SiMe3 but 2´-H near to 2,6-CH3. If (E)-11 and also the butyl derivative (Z)-6 would be formed
through a direct electrophilic attack of RHal (ClSiMe3 or n-BuBr in the upper line of Scheme 4)
on the dimeric portion D of (Z)-1, the reaction rate would be proportional to the concentrations of
both RHal and D, that is, the rate would be of second order kinetically. We applied the
simplifying condition of a sufficient excess (≥ 8 equiv) of RHal, so that its concentration [RHal]
was practically constant; the rate would then be of pseudo-first order with respect to the
concentration [D(t)], and the corresponding half-reaction time t1/2 should be independent of the
decreasing [D(t)]. Experimentally, however, t1/2 was found to decrease, which excluded D from
being the reactive species and left the NMR-invisible monomer M of (Z)-1 for playing that role
(Scheme 4). The rate should then be of first order in [M(t)] in eq 3, where it is seen to be defined
as the time dependence of the increasing product concentration, which is twice as high as the time
dependence of the decreasing [D(t)]. Together with eq 4 (from eq 2), this leads to the 0.5th-order
(namely, square root) dependence on [D(t)] in eq 5, which defines also the 1.5th-order rate
constant k3/2 = kψ/[RHal]. Both k3/2 (in L0.5 mol–0.5 s–1) and the concentration-dependent rate
constant kψ (in mol0.5 L–0.5 s–1) contain the numerically unknown equilibrium constant KMD (eq
2) and the concentration [free Et2O] with the difference (dM – dD) of the microsolvation numbers
(eq 4). Integration of eq 5 afforded eq 6 in the convenient form of counting the dimer in terms of
its two carbanionic units (as also measured through NMR integration). As derived before,22 eq 6
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ψ
implies a concentration-dependent pseudo-half-reaction time t1/2 in eq 7. With the effective
[RHal] as the average of the excessive initial and final values,, the experimental data of both
electrophiles RHal (Tables S3 and S4)14 obeyed the linear time dependence of the square roots
(2[D(t)])0.5 in eq 6 (Figures 2 and S3); the final product ratios were (E)-11/(Z)-11 = 85:15 with
ACS Paragon Plus Environment
14

Page 15 of 38
Journal of the American Chemical Society
ClSiMe3 and (Z)-6/(E)-6 = 78:13 with n-BuBr. The k3/2 values (Table 2) for either electrophile
were numerically very similar (by factors of 1.5 and 0.6, respectively)22 to those of 2, suggesting
that the 2´-CH3 cis-substituent in (Z)-1 had again little influence also on these rates. All four of
these reactions of 1 and 2 ran on a convenient time scale at 32 °C because the predominating
dimers of 2 or (Z)-1 are inactive and must rapidly (albeit reversibly) deaggregate before reacting
with RHal (Scheme 4). For comparison with n-BuBr, the methylation of a monomeric
alkyllithium compound with CH3I was very fast in THF/Me2O at –132 °C and was shown23 to
proceed via a solvent-separated ion pair as the most reactive equilibrium component.
Trimethylsilylation of (Z)-1 occurred faster than butylation by a factor of 17 (k3/2 values in Table
2); this factor mirrored the relative kM values of the product-forming steps in Scheme 4, since all
other parameters (eqs 3–5) and the temperatures were equal in the two runs; but the absolute
values of these kM of the two monomers remained unknown, of course. Since the
interconversion of monomeric (Z)-1 and (E)-1 (“slow” in Scheme 4) was mobile on the
laboratory time scale (Section 2.4), comparison of the equilibrium portion (6% in Table S6b)14 of
(E)-1 in Et2O with the increased (15 or 13%, respectively) minor products (Z)-11 or (E)-6
pointed to a certain partitioning of monomeric (Z)-1 between the “faster” derivatization and the
“slow” replenishment of (E)-1.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 4. Solvated, dimeric (Z)-1&dDEt2O (= D&2Et2O) reacts with electrophiles
RHal (n-BuBr or ClSiMe3) in Et2O at 32 °C via monomeric (Z)-1&dMEt2O (“M”);
“slow” = mobile on the laboratory time scale; aryl = 2,6-dimethylphenyl.
0.5 {(Z)-1&dDEt2O}2
(= 0.5 D&2Et2O)
+ RHal
+
(dM - dD) Et2O
slow
(Z)-1&dMEt2O (M)
monomeric (Z)-1
(E)-1&dMEt2O
monomeric (E)-1
faster
fast
kM
+ RHal
-
LiHal
aryl
R
R
aryl
H3C
H
H3C
H
R = n-Bu: (Z)-6
R = SiMe3: (E)-11
R = n-Bu: (E)-6
R = SiMe3: (Z)-11
ACS Paragon Plus Environment
15

Journal of the American Chemical Society
Page 16 of 38
1
2
3
M – 2d
KMD = [M]–2 [D] [free Et2O]2d
(2)
D
4
5
6
7
rate = d[product]/dt = –2d[D(t)]/dt
= kM [RHal] [M(t)]
(3)
(4)
8
9
M – dD
[M(t)] = (2KMD)
–0.5 (2[D(t)])0.5 [free Et2O]d
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
rate = –d(2[D(t)])/dt = k3/2 [RHal] (2[D(t)])0.5
= kψ (2[D(t)])0.5
(5)
Integrated: (2[D(t)])0.5 = –0.5 kψ t + (2[D(t=0)])0.5 (6)
ψt1/2 = 2(1 – 2–0.5) kψ–1 (2[D(t=0)])0.5
= 0.586 kψ–1 (2[D(t=0)])0.5
(7)
Figure 2. Time dependence of the square root (pseudo-0.5th order) of the doubled
concentration 2[D(t)] of dimeric (Z)-1 reacting with an excess of ClSiMe3 in Et2O at 32 °C.
ACS Paragon Plus Environment
16

Page 17 of 38
Journal of the American Chemical Society
Table 2. Rate measurements of dimeric (Z)-1 [“D(t)”] with the
electrophiles RHal in Et2O at +32 (±1) °C: Pseudo-0.5th-order rate
1
2
3
4
ψ
constants kψ (in mol0.5 L–0.5 s–1), first pseudo-half-reaction times t1/2,
5
6
7
and 1.5th-order rate constants k3/2 (in L0.5 mol–0.5 s–1); M = mol/L.
8
9
R–Hal:
average [R–Hal]:
[D(t=0)]:
n-BuBr
2.46 M
ClSiMe3
1.60 M
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0.150 M
0.548 M0.5
0.107 M
0.463 M0.5
(2[D(t=0)])0.5:
product fraction: 78% (Z)-6 a 85% (E)-11 b
105kψ:
ψt1/2:
105k3/2:
7.8
70 min
3.2
86
5.2 min
54
a Accompaied by (E)-6 (13%) and the alkenes 7 (9%). b With (Z)-11 (15%).
The data in this section confirmed that the resemblance of (Z)-1 and 2 extended to the
dimerization equilibria which are highly mobile on the NMR time scale already at about –50 °C.
It extended also to the reactivities against ClSiMe3 or n-BuBr in Et2O at 32 °C with their
conveniently long first pseudo-half-reaction times of 70 or 5.2 min, respectively, which are due to
the inactivity of dimeric (Z)-1 that must deaggregate reversibly to maintain a tiny concentration
of the reactive, monomeric (Z)-1.
2.6. Selectivities of (Z)-1 and (E)-1 toward fast and slow Ketones
Recent studies24 revealed that addition reactions of monomeric aryllithium species to aryl
ketones in THF/Me2O can be extremely fast even at –140 °C, while the remnant aggregated
species were observed to react independently but much more slowly. This suggested that
monomeric (Z)-1 and (E)-1 should react with most ketones in THF not only rapidly but also
independently of each other (“non-Curtin-Hammett condition”)22 under the important proviso
that their Z,E equilibration (Section 2.4) would be significantly slower. If so, the adducts (E,Z)-
12 (Scheme 5) would be generated in proportion (eq 8) to the relative reactivities kZ/kE and
populations of (Z,E)-1. However, this benefit of a high velocity of ketone addition may become a
disadvantage if one or both competitors react extensively already within the boundary zone of
incomplete microscopic mixing,22,25,26 which would lead to the feared depressions of observed
competition constants
κobs (1 ≤ κobs ≤ kZ/kE or kE/kZ) if calculated by eq 8. Convinced that we
would anyhow be able to obtain only minimum (depressed) selectivities κobs > 1, we did not try
to find out22 how close our data came to the “true” quotients kZ/kE. Instead, we looked in the
ACS Paragon Plus Environment
17

Journal of the American Chemical Society
Page 18 of 38
following manner for qualitative information. Even without mixing problems (that is, at
sufficiently low temperatures), eq 8 can be valid only as long as the known initial [(E)-1][(Z)-1]–
1
2
3
4
5
1
ratio does not change substantially, namely, either with a large excess of 1 or during the initial
6
7
8
9
reaction phase only. Combining the latter choice with analyses of interrupted runs, we added
deficient amounts of a ketone to solutions of (Z,E)-1 and trapped remnant (Z,E)-1 with solid CO2.
The carboxylic acids, (E)-8 from (Z)-1 and (Z)-8 from (E)-1 (Scheme 5 and Table 3), reported the
final (Z)-1/(E)-1 ratios and the rate of ketone addition to 1 (total yields of the adducts 12). As an
important confirmation of slow Z,E equilibration, the E/Z ratio of 8 may reveal a certain depletion
of the faster diastereomer of 1 that furnished the higher portion of the two products 12. Thus, a
deficiency of the weakly (yet quickly) soluble dicyclopropyl ketone in a THF solution of
monomeric (Z,E)-1 at –78 °C (entry 1 of Table 3) afforded the adducts (E,Z)-12a within 30 s in a
selectivity of κobs ≥ 14. As expected for a continuing conversion (increasing yield), the
corresponding depletion of (Z)-1 was stronger after 65 s (entry 2), as reported by the
progressively lowered ratios of the acids 8 with respect to the initial Z/E ratio of 1. (For
comparison, the Z,E equilibration in THF at –78 °C had been estimated in Section 2.4 to occur
with t1/2 ≤ ca. 54 min). The actual rates of ketone addition cannot be read from entries 1 and 2
because of the slow introduction of dicyclopropyl ketone; however, an excess of this ketone was
found to consume monomeric (E)-1&3THF completely within less than 4 min at –78 °C in THF,
Et2O, or toluene as the solvents. These rapid addition reactions (always faster than the Z,E
interconversion) in any of these three solvents indicated that direct rate measurements with
dicyclopropyl ketone would require either more advanced equipment or much lower
temperatures.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
[(E)-12][(Z)-12]–1[(E)-1][(Z)-1]–1 = kZ/kE = κZ/E
(8)
Table 3. Relative reactivities κobs of (Z,E)-1 in addition reactions to ketones: Adducts (E,Z)-12,
olefins (Z,E)-7, and acids (E,Z)-8, the latter from remnant (Z,E)-1.
en-
try
1
ketone (equiv)
sol- temp.
vent °C
time
acids adducts olefins yield a
Z/E 8 E/Z 12 E/Z 7 Z/E of 12
1
κobs
(eq 8)
cpr2C=O b (< 1.0) THF –78
cpr2C=O b (< 1.0) THF –78
30 s
65 s
20 s
34:66 24:76
34:66 15:85
88:12
85:15
–
–
53%
68%
31%
14 c
11 c
2
d
3
THF
rt
52:48 55:45 E only
E only
33 min 96:4 85:15 E only 88:12 < 40%
t-Bu2C=O (< 1.0) Et2O –40 120 min 96:4 91:9 E only none 53%
trace
t-Bu2C=O (≥ 1.0)
∞
d
4
THF
rt
27 min 52:48 none
trace 100%
t-Bu2C=O (≥ 1.0)
∞
d
5
t-Bu2C=O (< 1.0) Et2O –40
∞
d
6
∞
ACS Paragon Plus Environment
18

Page 19 of 38
Journal of the American Chemical Society
a Fraction of 12 in the sum of 7, 8, and 12. b Dicyclopropyl ketone. c Apparent (minimum)
selectivity. d Stereospecific.
1
2
3
4
5
6
7
8
9
Scheme 5. (Z)-1 and (E)-1 compete for a ketone R2C=O, whereupon
carboxylation checks for the depletion of remnant (Z)-1 or (E)-1.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(E)-8
(Z)-8
CO2
CO2
CH3
H3C
.
.
.
.
Li
Li
CH3
CH3
H3C
H3C
H
H
(E)-1
(Z)-1
+
R2C=O
3
3
4
2
4
2
CH3
1
H3C
R
R
5
5
1
4´
4´
6
6
OH
HO
CH3
H3C
3´
3´
R
R
2´
2´
H3C
cis
H
H3C
trans
H
trans
cis
12
a
R
(E)-12a-c
(Z)-12a
cyclo-
propyl
b
c
C(CH3)3
CH(CH3)2
3
cis
4
2
5´
CO2H
H3C
(H3C)2CH
4´
5
1
(H3C)2C
6
H3C
O
(H3C)2CH
3´
C
syn
H
1´
2´
H
trans
(H3C)2CH
H
anti
13
14
As a counterexample, the more soluble di-tert-butyl ketone (t-Bu2C=O)27 in excess reacted
with monomeric 1 in THF at rt more slowly than the Z,E stereoinversion of 1. This became
evident through the analysis (entry 3 of Table 3) of a portion (about one third of the run) that was
quenched with solid CO2 after 20 s: Although only 31% of (E)-12b [but no (Z)-12b] had been
produced from 1, the (E)-8/(Z)-8 = 55:45 ratio indicated that the initial ratio of (Z)-1/(E)-1 =
52:48 was maintained by means of a fast E → Z conversion in THFat rt (t1/2 ≤ ca. 0.24 s in
Section 2.4). With the remaining two thirds of this run, a rough estimate of the first half-reaction
ACS Paragon Plus Environment
19

Journal of the American Chemical Society
Page 20 of 38
1
time t1/2 = ca. 50 s for (E)-12b formation was possible since the amount of (Z,E)-1 had not yet
been exhausted, as shown by the increasing yields of 12 in entries 3 and 4. In Et2O (wherein 1 is
mainly dimeric) at –40 °C, this reaction appeared to be significantly slower: Comparison of the
1
2
3
4
5
6
7
1
first and the final yields of (E)-12b in entries 5 and 6, respectively, suggested a first t1/2 = ca. 15
8
9
min, while a comparably slow Z,E interconversion had apparently replenished only part of the
consumed (Z)-1 after 33 min (acids 8 in entry 5) and approached the initial (Z)-1/(E)-1 = 96:4
ratio rather slowly (increased ratio of 8 in entry 6). In consideration of the different temperatures,
15 min at –40 °C appeared again to be compatible with the time scale of 0.1 min at 22 °C as
estimated previously22 for t-Bu2C=O with 2 in Et2O. In accord with this time scale, only a trace
of (E)-12b but a lot of the almost pure acid (E)-8 were formed from (Z)-1 at –78 °C in Et2O in
another run that was terminated through carboxylation after 145 min. In toluene as the solvent,
monomeric (E)-1&3THF was completely consumed by t-Bu2C=O within 35 min at rt, but
produced again only (E)-12b (the inverted product) and some olefin (E)-7; clearly, once more
only (Z)-1 had reacted after its formation from the inactive (E)-1. This “slow” (E)-1 → (Z)-1
conversion was so easily recognized because t-Bu2C=O consumed only (Z)-1 but did not react
directly with the source (E)-1. The lithium alkoxide of (E)-12b precipitated occasionally as a
colorless powder from runs in Et2O or THF after at least 30 min. Through simple washing with
dry pentane and subsequent acidification, this powder afforded pure (E)-12b, whose
configuration was established by a NOESY analysis. In analogy with the previously22 reported
retro-addition reaction that regenerated t-Bu2C=O and 2, the reversibility of (E)-12b formation
was demonstrated as follows. In the presence of dicyclopropyl ketone (4 equiv) and the free
alcohol (E)-12b (1 equiv) as a proton source, the lithium alkoxide of (E)-12b regenerated t-
Bu2C=O and (Z)-1 in THF very slowly: After 22 hours at rt, the product mixture contained still
some (E)-12b (12%), the adduct (E)-12a (12%) of dicyclopropyl ketone (the trap for 1), and the
olefins (Z)-7 (71%) and (E)-7 (5%) which arose through proton transfer from (E)-12b to 1. The
ratio of (Z)-7/(E)-12a = 71:12 indicated that the rate constant of that proton transfer to (Z)-1 had
been 24-times (namely, 4 × 71:12) higher than the rate constant of (Z)-1 addition to dicyclopropyl
ketone. The corresponding potassium alkoxide of (E)-12b was generated in THF by the slow,
heterogeneous deprotonation reaction with an excess of KH in the presence of dicyclopropyl
ketone (8 equiv) and decayed within 20 hours at rt to give (E)-12a and a trace of (Z)-12a.
Diisopropyl ketone reacted with monomeric (Z)-1 in THF to give (E)-12c but did not (or not
irreversibly) add to monomeric (E)-1. The NOESY cross-peaks of 2´-H with 5´-H and of 2´-CH3
with the 2,6-dimethyl protons established the E configuration. The olefinic side-products (Z)-7
and (E)-7 (17:14 at rt) arose partially through proton transfer to (Z)-1 and (E)-1, respectively,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
20

Page 21 of 38
Journal of the American Chemical Society
from diisopropyl ketone whose enolate was recognized through final trapping with solid CO2:
The pure 2,2,4-trimethyl-3-oxopentanoic acid28 (13) was isolated in an amount that could be
expected from the olefinic product portion. In toluene as the solvent, dimeric (Z)-1&t-BuOMe
was consumed completely by diisopropyl ketone within 150 s at +2 °C, affording again only (E)-
12c (84%) and the olefin (Z)-7 (16%). In most of these runs, (E)-12c was generated so
selectively that an analytically almost pure sample could be obtained through simple evaporation
of the accompanying volatiles in vacuo at rt. This was fortunate because modest heating
(attempted distillation) of (E)-12c led to spontaneous H2O elimination, which furnished the 1,3-
diene 14. A similar decomposition problem prevented us from purifying and presenting the
adducts of acetone to 1.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
This section demonstrated that monomeric (Z)-1 was at least 14-times more reactive than
monomeric (E)-1 in THF against dicyclopropyl ketone; this addition reaction was faster than the
Z,E equilibration of 1. Compared with that Z,E equilibration, however, the sterically hindered t-
Bu2C=O reacted more slowly with (Z)-1 in THF but more rapidly in Et2O, in accord with the
selectivity reversal observed22 for 2 with t-Bu2C=O in THF versus Et2O. Presumably due to
steric shielding by its 2´-CH3 substituent, (E)-1 did not add to either t-Bu2C=O or diisopropyl
ketone, so that these two ketones consumed both (Z)-1 and (E)-1 with exclusive formation of the
product from (Z)-1 (an unexpected single diastereoconvergence).
3. CONCLUSION
(i) Both (Z)-1 and (E)-1 are trisolvated monomers in THF as the solvent wherein (E)-1 is
thermodynamically more stable than (Z)-1 by only 0.84 kcal mol–1. In contrast, (Z)-1 is (almost)
exclusively either a disolvated dimer in monodentate, non-THF ethereal solvents or an unsolvated
higher aggregate in donor-free hydrocarbon solutions. The most reactive monomeric (Z)-1 was
NMR-visible only with microsolvation by THF. Armed with such knowledge, we became able to
determine the following dynamic properties of 1.
(ii) The endothermic dimerization of (Z)-1&3THF led to a THF-dependent equilibrium in
1
toluene that was highly mobile on the H and 13C NMR time scales at about –50 °C already.
Steric shielding of the Li–C(1´) bond of (E)-1 by the neighboring 2´-CH3 substituent may
explain why (E)-1 did not form a persistent dimer and did not furnish the expected adducts with t-
Bu2C=O or diisoproyl ketone. Even a mixed dimer of 1, bearing one cis and one trans 2´-CH3
group, was never observed because E,Z interconversion would provide for its removal. This
mode of accomodation is not possible for 3 with its two 2´-methyl groups, so that 3 was found to
be dimeric in Et2O (Figure S6 and Table S12) and monomeric in THF (Table S13).
ACS Paragon Plus Environment
21

Journal of the American Chemical Society
Page 22 of 38
(iii) The formation of (Z)-1 from monomeric (E)-1&3THF in toluene occurred with a
conversion-independent half-reaction time of 29 min at –25 °C. A similar time scale (in min at
rt) was estimated in Et2O and in donor-free hydrocarbon solvents, confirming earlier observations
with similar α-arylalkenyllithiums by, for examples, Curtin29 et al. (in benzene), Panek30 et al.
(in hexane with donor additives), van Koten31 et al. (in hexane), or Knorr32 et al. (donor-free [β-
D1]2 in hexane).
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(iv) Disolvated, dimeric (Z)-1 is unreactive and must deaggregate in order to consume n-BuBr
or ClSiMe3 in Et2O. This followed from its 1.5th-order kinetics with these two electrophiles and
can be ascribed to the intermediacy of the monomeric (NMR-invisible) species of (Z)-1 as the
more reactive equilibrium component.
(v) The close similarity of (Z)-1 and the 2´-unsubstituted α-(2,6-dimethylphenyl)vinyllithium
(2) included the above kinetics, the stereoinversion rates, the dimerization equilibrium and its
1
mobility, the reaction rate with t-Bu2C=O, the NMR coupling constants J(13C,6Li) with derived
microsolvation numbers at Li, and the degree of π-electron delocalization. Thus, a single 2´-CH3
substituent in 1 has little influence if cis to the 1´-aryl group. Therefore, it appeared justified to
use the published6 cis/trans stereoinversion parameters of 2 as a model for 1: The E/Z
interconversion of THF-solvated 1 was estimated to be faster in THF at –25 °C (t1/2 ≤ ca. 9.4 s)
than in toluene as the solvent by a factor of at least ca. 185 that appears compatible with catalysis
by THF as expected for the ionic mechanism6 via a THF-separated ion-pair intermediate.
(vi) These observations explain why we encountered the higly selective, doubly
diastereoconvergent formation of either dimeric (Z)-1&t-BuOMe or monomeric (E)-1&3THF by
Br/Li interchange reactions from any mixture of the (Z,E)-bromopropenes 5. Of course, the
commended preparative protocols apply verbatim only to the present paradigm example: (Z)-
1&t-BuOMe may be used in Et2O, t-BuOMe, or hydrocarbons as the solvents at or below rt,
whereas (E)-1&3THF should be employed in toluene solution at or below –78 °C very soon after
its preparation. THF was rarely suitable as a solvent for diastereospecific derivatizations of 1.
(vii) In the rapid addition reactions of dicyclopropyl ketone (always faster than Z,E
interconversion), monomeric (Z)-1 was more successful than monomeric (E)-1 by a factor of at
least 14 in THF. The “slow” di-tert-butyl ketone reacted with (Z)-1 only; the estimated time
scales of this addition were ca. 50 s in THF at rt and ca. 15 min in Et2O at –40 °C.
4. EXPERIMENTAL
General Remarks. For compatibility with our previously published alkenyllithiums such as 2
and 4, we designate here the positions within the 2´-aryl groups by the numbers 1 through 6,
ACS Paragon Plus Environment
22

Page 23 of 38
Journal of the American Chemical Society
1
2
3
whereas C-1´ and C-2´ of the C=C double bond were alternatively also called C-α and C-β,
1
respectively. Previously9 desribed techniques and spectrometers were used. All H and 13C
4
5
6
7
8
9
NMR shifts were referenced against internal TMS, and some of the spectra in THF could be
measured down to –117 °C owing to the presence of saturated hydrocarbons as cosolvents.
NMR abbreviations were d = doublet, m = multiplet, q = quartet, qi = quintet, quat = quaternary,
sept = septet, t = triplet. Concentrations were determined through NMR integrations and
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
comparison either with one of the H NMR 13C satellites of a nondeuteriated solvent or with a
calibrated capillary in situ. Concentrations (M = molar) and molar ratios were counted in terms
of the monomer formulas (such as the carbanionic units in 1) as read from NMR integrals. With
this choice, the numerical sum of reactants and products can remain constant during a
transformation, so that the amounts of all participants can be compared on the same footing. In
contrast, interpretations in terms of dimeric and monomeric particles would imply that the
summed molecular concentrations of reactants and products are not necessarily constant.
(Z)-1´-Lithio-1´-(2,6-dimethylphenyl)propene [Dimeric and Monomeric (Z)-1]. A dry
NMR tube (5 mm) was charged with a Z,E mixture of 1´-bromo-1´-(2,6-
dimethylphenyl)propene11 (5, 120 mg, 0.53 mmol) and anhydrous t-BuOMe (0.4 mL). After
addition of n-BuLi (0.64 mmol) in hexane (0.27 mL) or cyclopentane at –30 °C under dry argon
gas cover, the tube was tightly stoppered, cautiously moved for a partial mixing, and recooled
quickly to –30 °C for mitigating the exothermic Br/Li interchange reaction that was complete
within a few minutes. Dimeric (Z)-1&t-BuOMe began to crystallize in colorless platelets either
soon at rt or with cooling to –20 °C. Too small crystals (< 1 mm) were dissolved through
warming at 50 °C, then reprecipitated through slow cooling and keeping at –20 °C for
(sometimes) up to two days. The crystals were purified at –70 °C under a stream of dry argon
gas by means of a syringe with a long cannula as follows. The supernatant was withdrawn and
replaced by dry pentane (ca. 0.3 mL). Short sucking and re-ejecting of a little pentane served to
whirl up the solid components. The heavier crystals were allowed to settle down, whereas the
powdery side-product LiBr was removed together with the pentane. This washing procedure was
repeated up to four times and terminated by volatilization of remnant pentane in a short (3 s)
stream of dry argon gas as introduced through a long pipette. The purified crystals of (Z)-1&t-
BuOMe (yield up to 85%) were dissolved in the desired anhydrous solvent with cooling under
1
argon gas cover. H NMR of dimeric (Z)-1 (t-BuOMe, 25 °C, 400 MHz)
δ
1.37 (d, 3J = 5.6 Hz,
3H, 2´-CH3 cis to aryl), 1.99 (s, 6H, 2-/6-CH3), 5.85 (q, 3J = 5.5 Hz, 1H, 2´-H), 6.52 (t, 3J = 7.3
Hz, 1H, 4-H), 6.75 (d, 3J = 7.3 Hz, 2H, 3-/5-H) ppm, no (E)-1, assigned through comparison with
1
(Z)-1 in Et2O; H NMR of dimeric (Z)-1 in a Z/E = 94:6 mixture (Et2O, 0.35 M, 25 °C, 400
ACS Paragon Plus Environment
23

Journal of the American Chemical Society
Page 24 of 38
1
2
3
MHz)
δ
1.37 (d, 3J = 5.5 Hz, 3H, 2´-CH3 cis to aryl), 2.00 (s, 6H, 2-/6-CH3), 5.86 (q, 3J = 5.5
1
Hz, 1H, 2´-H), 6.52 (t, 3J = 7.3 Hz, 1H, 4-H), 6.75 (d, 3J = 7.3 Hz, 2H, 3-/5-H) ppm; H NMR of
4
5
dimeric (Z)-1&t-BuOMe ([D8]toluene, 0.34 M, –70 °C, 400 MHz)
δ
1.68 (d, 3J = 5.4 Hz, 3H,
6
7
8
9
2´-CH3 cis to aryl), 2.15 (s, 6H, 2-/6-CH3), 6.12 (q, 3J = 5.4 Hz, 1H, 2´-H), 6.94 (t, 3J = 7.4 Hz,
1H, 4-H), 7.12 (d, 3J = 7.4 Hz, 2H, 3-/5-H), 0.89 and 2.73 (2 × s, coordinated t-BuOMe) ppm, no
1
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(E)-1; H NMR of THF-solvated (4 equiv), dimeric (Z)-1 ([D8]toluene, –85 °C, 400 MHz)
δ
1.74 (d, 3J = 5.5 Hz, 3H, 2´-CH3 cis to aryl), 2.20 (s, 6H, 2-/6-CH3), 5.98 (q, 3J = 5.5 Hz, 1H, 2´-
H), 6.90 (t, 3J = 7.4 Hz, 1H, 4-H), 7.10 (d, 3J = 7.4 Hz, 2H, 3-/5-H) ppm, assigned through
1
comparison with (Z)-1 in Et2O; H NMR of THF-solvated (4 equiv), monomeric (Z)-1
([D8]toluene, –85 °C, 400 MHz)
δ
1.93 (d, 3J = 5.5 Hz, 3H, 2´-CH3 cis to aryl), 2.38 (s, 6H, 2-
/6-CH3), 5.93 (q, 3J = 5.5 Hz, 1H, 2´-H), 6.85 (t, 3J = 7.4 Hz, 1H, 4-H), 7.18 (d, 3J = 7.4 Hz, 2H,
1
3-/5-H) ppm; H NMR of monomeric (Z)-1 in a Z/E = 47:53 mixture (THF, 0.68 M, –10 °C, 400
MHz)
7.4 Hz, 1H, 4-H), 6.57 (d, 3J = 7.4 Hz, 2H, 3-/5-H) ppm, see below for (E)-1; 13C NMR of
dimeric (Z)-1 (t-BuOMe, 25 °C, 100.6 MHz)
17.0 (qd, 1J = 124 Hz, 2J = 8.5 Hz, 2´-CH3 cis to
δ
1.92 (s, 6H, 2-/6-CH3), (2´-CH3 obscured), 5.46 (q, 3J = 5.5 Hz, 1H, 2´-H), 6.21 (t, 3J =
δ
aryl), 21.7 (qd, 1J = 125 Hz, 3J = ca. 5 Hz, 2-/6-CH3), 120.0 (sharp d, 1J = 156 Hz, C-4), 122.0
(dq, 1J = 147 Hz, C-2´), 127.2 (dm, 1J = 153 Hz, C-3/-5), 127.6 (m, apparent J = ca. 6 Hz, C-2/-
6), 154.2 (unresolved m, C-1), 190.5 (dm. 2J = 16 Hz, C-1´); assigned through the γ-aryl33 effect
and comparison with (Z)-1 in Et2O; 13C NMR of dimeric (Z)-1 in a Z/E = 94:6 mixture (Et2O,
0.35 M, 25 °C, 100.6 MHz)
δ
17.2 (qd, 1J = 124 Hz, 2J = 8.5 Hz, 2´-CH3 cis to aryl), 21.5 (qd,
1J = 125 Hz, 3J = 4.9 Hz, 2-/6-CH3), 120.2 (sharp d, 1J = 155 Hz, C-4), 122.7 (dq, 1J = 144 Hz, 3J
= ca. 5 Hu, C-2´), 127.4 (dm, 1J = 155 Hz, C-3/-5), 127.9 (m, apparent J = ca. 6 Hz, C-2/-6),
154.1 (unresolved m, C-1), 191.0 (dm. 2J = 16 Hz, C-1´), the quintet (1:2:3:2:1) splitting of C-1´
by 1J(13C,6Li) was detected at –50 (7.6 Hz), –85 (7.6 Hz), and –90 °C (7.5 Hz); 13C NMR of
dimeric (Z)-1&t-BuOMe ([D8]toluene, 0.34 M, –70 °C, 100.6 MHz)
δ 17.4 (2´-CH3 cis to aryl),
22.2 (2-/6-CH3), 119.9 (C-4), 121.2 (C-2´), 127.2 (C-3/-5), 127.4 (C-2/-6), 154.1 (C-1), 189.8
(quintet 1:2:3:2:1, 1JCLi = 7.5 Hz, C-1´), 26.2, 49.0, and 74.2 (coordinated t-BuOMe) ppm,
assigned through comparison with (Z)-1 in Et2O, the C-1´ quintet (7.5 Hz) was also detected at –
45 and –18 °C; 13C NMR of THF-solvated (4 equiv), dimeric (Z)-1 ([D8]toluene, –85 °C, 100.6
MHz)
δ 18.2 (2´-CH3 cis to aryl), 21.7 (2-/6-CH3), 119.0 (C-4), 122.0 (C-2´), 127.0 (C-3/-5),
127.2 (C-2/-6), 154.8 (C-1), 192.6 (quintet 1:2:3:2:1, 1JCLi = ca. 6.5 Hz, C-1´) ppm, assigned
through comparison with (Z)-1 in Et2O and confirmed at 25 °C through gated decoupling and
selective {1H} decoupling as follows: {2´-CH3} → C-1´ as a d with 2J = 16 Hz, 2´-CH3 as a d
ACS Paragon Plus Environment
24

Page 25 of 38
Journal of the American Chemical Society
with 2J = 8.5 Hz; {2-/6-CH3} → C-1 as a dt with 3J = 13.5 Hz and 5.5 Hz, C-1´as a d, C-2/-6 as a
d with 3J = 6.8 Hz, and 2-/6-CH3 as two narrow d with 2J = 4.9 Hz for (Z)-1 and (E)-1; {2´-H}
→ 2´-CH3 as a sharp q. 13C NMR of THF-solvated (4 equiv), monomeric (Z)-1 ([D8]toluene, –
1
2
3
4
5
6
7
85 °C, 100.6 MHz)
δ 18.5 (2´-CH3 cis to aryl), 22.3 (2-/6-CH3), 114.9 (C-4), 116.0 (C-2´), (C-
8
9
2/-6 obscured), 126.5 (C-3/-5), 158.5 (C-1), 197.4 (C-1´) ppm, assigned through comparison with
monomeric (Z)-1 in THF; 13C NMR of monomeric (Z)-1 in a Z/E = 32:68 mixture (THF, 0.24 M,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
–88 °C, 100.6 MHz)
δ 18.0 (2´-CH3 cis to aryl), 22.0 (2-/6-CH3), 114.2 (C-2´), 114.9 (C-4),
125.0 (C-2/-6), 126.0 (C-3/-5), 158.4 (C-1), 197.6 (triplet 1:1:1, 1JCLi = 11.4 Hz, C-1´) ppm,
assigned through the γ-aryl effect,33 temperature dependent Z/E ratios and comparison with (Z)-1
in Et2O and t-BuOMe.
(E)-1´-Lithio-1´-(2,6-dimethylphenyl)propene [Monomeric (E)-1&3THF]. 1´-Bromo-1´-
(2,6-dimethylphenyl)propene11 [(Z,E)-5, 500 mg, 2.22 mmol)] was transformed as above into
purified, dimeric (Z)-1&t-BuOMe. This product in the NMR tube (5 mm) was treated at rt with
successive small amounts of anhydrous THF (total 0.85 mL) until the last few grains of (Z)-1&t-
BuOMe were dissolved. (More THF will prevent the subsequent crystallization.) Fine white
needles of (E)-1&3THF precipitated on cooling at –78 °C and were after 30 min purified through
washing [see (Z)-1] with dry pentane (3× 0.5 mL, no cyclopentane!) under argon gas cover.
1
Without warming, the needles were dissolved for NMR spectroscopy or derivatization. H NMR
of monomeric (E)-1 in a Z/E = 94:6 mixture (Et2O, 0.35 M, 25 °C, 400 MHz)
Hz, 3H, 2´-CH3 trans to aryl), 2.07 (s, 6H, 2-/6-CH3), 6.14 (q, 3J = 6.0 Hz, 1H, 2´-H), 6.46 (t, 3J
= 7.3 Hz, 1H, 4-H), 6.72 (d, 3J = 7.3 Hz, 2H, 3-/5-H) ppm; H NMR of THF-solvated (4 equiv),
δ
1.91 (d, 3J = 6.0
1
monomeric (E)-1 ([D8]toluene, –85 °C, 400 MHz)
6.44 (q, 3J = 6 Hz, 1H, 2´-H), 6.83 (t, 3J = 7.4 Hz, 1H, 4-H), 7.14 (d, 3J = 7.4 Hz, 2H, 3-/5-H)
δ 2.38 (s, 6H, 2-/6-CH3), (2´-CH3 obscured),
1
ppm, assigned through comparison with (E)-1 in THF; H NMR of monomeric (E)-1 in a Z/E =
47:53 mixture (THF, 0.68 M, –10 °C, 400 MHz)
δ 1.94 (s, 6H, 2-/6-CH3), (2´-CH3 obscured),
5.72 (q, 3J = 5.8 Hz, 1H, 2´-H), 6.20 (t, 3J = 7.4 Hz, 1H, 4-H), 6.57 (d, 3J = 7.4 Hz, 2H, 3-/5-H)
ppm; 13C NMR of THF-solvated (4 equiv), monomeric (E)-1 ([D8]toluene, –85 °C, 100.6 MHz)
δ
23.0 (2-/6-CH3), 25.2 (2´-CH3 trans to aryl), 115.6 (C-4), 116.6 (C-2´), (C-2/-6 obscured),
52
53
54
55
56
57
58
59
60
127.2 (C-3/-5), 161.2 (C-1), 197.8 (C-1´) ppm, assigned through comparison with (E)-1 in THF;
13C NMR of monomeric (E)-1 in a Z/E = 32:68 mixture (THF, 0.24 M, –88 °C, 100.6 MHz)
δ
22.7 (2-/6-CH3), 24.8 (2´-CH3 trans to aryl), 115.4 (C-2´), 116.0 (C-4), 126.1 (C-3/-5), 126.6 (C-
2/-6), 161.2 (C-1), 198.1 (triplet 1:1:1, 1JCLi = 11.4 Hz, C-1´) ppm, assigned through the γ-aryl
effect,33 the temperature dependent Z/E ratios, and comparison with (Z)-1 in Et2O and t-BuOMe.
ACS Paragon Plus Environment
25

Journal of the American Chemical Society
Page 26 of 38
1
2
3
4
5
6
7
8
9
Donor-free (Z)-1. A dry NMR tube (5 mm) was loaded with 1´-iodo-1´-(2,6-
dimethylphenyl)propene [(Z)-9, 40 mg, 0.15 mmol], dry cyclopentane (0.4 mL),
[D12]cyclohexane (0.07 mL), and a trace of TMS, then cooled at –30 °C under argon gas cover.
n-Bu6Li (0.16 mmol) in cyclopentane (0.11 mL) was added to the tube, which was warmed up to
1
rt after 15 min and showed that the I/Li interchange reaction was complete. H NMR (saturated
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
hydrocarbon solvents, –45 °C, 400 MHz)
3J = 5.6 Hz, 2´-H), 6.71 (t, 3J = 7.3 Hz, 1H, 4-H), 6.78 (d, 3J = 7.3 Hz, 2H, 3-/5-H) ppm; 13C
NMR (saturated hydrocarbon solvents, –45 °C, 100.6 MHz) 17.3 (2´-CH3 cis to aryl), 21.1
δ 1.73 (s, 6H, 2-/6-CH3), (2´-CH3 obscured), 5.74 (q,
δ
(qd, 2-/6-CH3), 123.0 (d, C-4), 128.8 (narrow d, C-2/-6), 129.0 (dd, C-3/-5), 131.4 (d, C-2´),
150.3 (narrow m, C-1), 179.5 (unresolved narrowed m, C-1´) ppm, assigned through the γ-aryl
effect33 and selective {1H} decouplimg. After the long-term measurements, the clear solution
was poured onto dry CO2, and the warmed-up material was dissolved in aqueous NaOH (3 M)
and Et2O. Acidification of the alkaline aqueous layer afforded the diastereomerically pure
carboxylic acid (E)-8 (12 mg, 42%).
1´-Lithio-2´-methyl-1´-phenylpropene (3).7 Unsolvated 3 precipitated slowly from a solution
of 1-bromo-2-methyl-1-phenylpropene7 (2.10 g, 9.95 mmol) and n-Bu6Li (11 mmol) in pentane
(2.0 mL). After 4 hours at rt, the precipitate was washed with dry pentane under argon gas cover
and dried in vacuo (ca. 30 min), then vented by slow admission of dry argon and weighed (yield
1
950 mg, 69%). H NMR (hexane, 60 MHz)
δ
6.69 (d, 3J = ca. 7 Hz, 2H, 2-/6-H), 6.73 (t, 3J = ca.
1
6 Hz, 1H, 4-H), 7.05 (overlaid t, 3-/5-H) ppm; H NMR (Et2O, 32 °C, 80 MHz)
δ 1.60 (s, 3H,
2´-CH3 cis to aryl), 1.85 (s, 3H, 2´-CH3 trans), 6.56 (d, 3J = 8.0 Hz, 2H, 2-/6-H), 6.60 (t, 3J = 7.0
1
Hz, 1H, 4-H), 7.03 (t, 3J = 7.2 Hz, 2H, 3-/5-H) ppm; H NMR (THF, –37 °C): as in ref 7; 13C
1
1
NMR (Et2O, –51 °C, 25.1 MHz)
δ 20.2 (qq, J = 122 Hz, 2´-CH3 cis to aryl), 29.9 (qq, J =
3
1
3
122.5 Hz, J = ca. 4 Hz, 2´-CH3 trans), 119.6 (dt, J = 159 Hz, J = 7.5 Hz, C-4), 124.6 (dt, 1J =
3
1
3
154 Hz, J = 6.5 Hz, C-2/-6), 128.2 (dd, J = 156 Hz, J = 7.5 Hz, C-3/-5), 134.4 (quasi-septet,
1
apparent J = ca. 5 Hz, C-2´), 158.0 (t, 3J = 6.5 Hz, C-1), 183.0 (qi 1:2:3:2:1, J(13C,6Li) = 7.5
Hz, C-1´) ppm (Table S12),14 assigned through the γ-aryl effect33 and comparison with 1 in
Et2O, the qi 1J(13C,6Li) = 7.5 Hz of C-1´ was observed also at –71 (Figure S6),14 –79, –91
6
(prepared through Hg/Li interchange), –95 (with LiBr), and –106 °C; 13C NMR (THF, –81 °C,
25.1 MHz)
δ 20.3 (q, 2´-CH3 cis), 30.6 (q, 2´-CH3 trans), 115.0 (d, C-4), 121.0 (quat, C-2´),
122.2 (d, C-2/-6), 127.2 (d, C-3/-5), 164.0 (quat, C-1), 194.0 (unresolved, C-1´) ppm, assigned
through {1H} off-resonance decoupling, the γ-aryl effect,33 and comparison with 1 in THF; the
ACS Paragon Plus Environment
26

Page 27 of 38
Journal of the American Chemical Society
1
triplet (1:1:1) with J(13C,6Li) = 11.5(1) Hz of C-1´ was observed at –106, –117, –124, and –126
°C (Table S13).14
1
2
3
4
5
6
7
8
3´-(2,6-Dimethylphenyl)hept-2´-ene [(Z)- and (E)-6]. A solution of unpurified, dimeric (Z)-
1 in t-BuOMe was prepared as above from the bromoalkenes11 (Z,E)-5 (116 mg, 0.51 mmol) and
kept at rt for 19 hours. The crude, yellowish oil (60 mg), isolated after aqueous workup,
contained (Z)-6, (E)-6, and the olefin (Z)-7 in the molar ratio 90:2:8. The crude material (132
mg) from several runs was purified through chromatography on silica gel (3.5 g): Elution with
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
low-boilimg petroleum ether (7 mL) furnished a clean sample of (Z,E)-6. H NMR of (Z)-6
(CDCl3, 400 MHz)
δ
0.89 (t, 3J = 7.2 Hz, 3H, CH3-7´), 1.30 and 1.42 (2 m, CH2-5´/-6´), 1.31 (dt,
3J = 6.7 Hz, 5J = 1.5 Hz, 3H, CH3-1´ cis to aryl), 2.14 (tm, 3J = 8 Hz, CH2-4´), 2.16 (s, 6H, 2-/6-
CH3), 5.57 (qt, 3J = 6.7 Hz, 4J = 1.5 Hz, 1H, 2´-H), 7.02 and 7.04 (A2B system, 3J = ca. 7 Hz,
1
2+1H, 3-/5-H and 4-H) ppm, assigned through the NOESY correlation 2´-H ↔ CH2-4´; H NMR
of (E)-6 (CDCl3, 400 MHz)
δ
0.86 (t, 3J = 7.2 Hz, 3H, CH3-7´), 1.77 (d, 3J = 7.0 Hz, 3H, CH3-1´
trans to aryl), 2.21 (s, 6H, 2-/6-CH3), 5.24 (qt, 3J = 6.9 Hz, 1H, 2´-H) ppm, others not identified,
assigned through the NOESY correlation 2´-H ↔ 2-/6-CH3; 13C NMR of (Z)-6 (CDCl3, 100.6
MHz)
δ
14.06 [qqi, 1J = 124 Hz, (2J + 3J)/2 = 3.9 Hz, CH3-7´], 14.11 (qd, 1J = 125.2 Hz, 2J = 4.5
Hz, CH3-1´ cis to aryl), 19.61 (qm, 1J = 126 Hz, 2-/6-CH3), 22.90 (tm, 1J = 126 Hz, CH2-6´),
29.85 (tm, 1J = 126 Hz, CH2-5´), 37.25 (tm, 1J = 124.5 Hz, CH2-4´), 120.17 (dm, 1J = 151.5 Hz,
apparent 3J and 2J = 6 Hz, C-2´), 126.11 (sharp d, 1J = 158.0 Hz, C-4), 127.19 (dm, 1J = 156 Hz,
C-3/-5), 135.33 (quasi-qi, apparent J = 5.3 Hz, 2-/6-C), 140.20 (unresolved, C-1), 140.86
3
(unresolved, C-3´) ppm, assigned through the γ-aryl33 effect; C NMR of (E)-6 (CDCl3, 100.6
MHz)
δ 13.47 (q, CH3-7´), 15.28 (q, CH3-1´ trans to aryl), 20.3 (q, 2-/6-CH3), 23.35 (t, CH2-6´),
30.30 (t, CH2-5´), 31.87 (t, CH2-4´), 123.36 (d, C-2´), 125.93 (d, C-4), 127.24 (d, C-3/-5), 135.78
(C-2/-6), 139.78 (C-1), 144.82 (unresolved, C-3´) ppm, assigned as above. Anal. calcd for
C15H22 (202.3): C, 89.04; H, 10.96. Found: C, 88.65; H, 10.88.
1´-(2,6-Dimethylphenyl)propene [(Z)- and (E)-7]. Z,E mixtures of this “parent” olefin 7
were obtained through protolysis and aqueous workup of 1. A sample (183 mg) distilled at 70–
88 °C (bath temp.)/13 Torr (ref 34: 75 °C/4 Torr) to give a colorless mixture (103 mg) of (Z)-7
1
and (E)-7 (63:37). H NMR of (Z)-7 (CDCl3, 400 MHz)
δ
1.45 (dd, 3J = 6.8 Hz, 4J = 1.7 Hz, 3H,
2´-CH3 cis to aryl), 2.19 (s, 6H, 2-/6-CH3), 5.83 (dq, 3J = 11.3 Hz, 3J = 6.8 Hz, 1H, 2´-H), 6.30
(dm, 3J = 11.3 Hz, 1H, 1´-H), 7.02 and 7.06 (A2B system, 3J = 7.4 Hz, 2+1H, 3-/5-H and 4-H)
1
ppm (the literature35 values are too large by ca. 0.18 ppm); H NMR of (E)-7 (CDCl3, 400 MHz)
ACS Paragon Plus Environment
27

Journal of the American Chemical Society
Page 28 of 38
1
2
3
4
δ
1.90 (dd, 3J = 6.5 Hz, 4J = 1.8 Hz, 3H, 2´-CH3 trans to aryl), 2.28 (s, 6H, 2-/6-CH3), 5.68 (dq,
3J = 16.1 Hz, 3J = 6.5 Hz, 1H, 2´-H), 6.33 (overlaid dm, 3J = ca. 16 Hz, 1´-H) ppm (3-/4-/5-H
obscured); 13C NMR of (Z)-7 (CDCl3, 100.6 MHz)
14.2 (2´-CH3 cis to aryl), 20.2 (2-/6-CH3),
δ
5
6
7
126.4 (C-4), 127.0 (C-3/-5), 127.3 (C-1´), 128.4 (C-2´), 136.1 (C-2/-6), 136.6 (C-1) ppm,
assigned through the Z/E ratios and the γ-aryl33 effect; 13C NMR of (E)-7 (CDCl3, 100.6 MHz)
δ
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
18.8 (2´-CH3 trans to aryl), 20.9 (2-/6-CH3), 126.1 (C-4), 127.6 (C-3/-5), 128.5 (C-1´), 130.3 (C-
2´), 136.0 (C-2/-6), 137.7 (C-1) ppm, assigned as above (the published35 1H and 13C NMR
spectra were not assigned and hence unsuitable for calculating ꢀ
δ values); IR (film, Z/E =
63:37) ν 3065, 3009, 2917, 2855, 1465, 768 cm–1. Anal. calcd for C11H14 (146.23, Z/E = 63:37):
C, 90.35; H, 9.65. Found: C, 89.57; H, 9.71.
(E)-2´-(2,6-Dimethylphenyl)but-2-enoic Acid [(E)-8]. Purified dimeric (Z)-1&t-BuOMe was
prepared from the bromoalkenes11 (Z,E)-5 (200 mg, 0.89 mmol), then dissolved in anhydrous t-
BuOMe (2.0 mL) and immediately poured onto solid CO2, warmed up, and diluted with Et2O (15
mL). The mixture was extracted with aqueous NaOH (2 M. 4 × 5 mL), and the Et2O layer was
discarded. The combined aqueous layers were acidified with conc. HCl, then shaken with Et2O
(3 × 15 mL), and thereafter discarded. The combined Et2O phases were washed with distilled
water until neutral, dried over MgSO4, filtered, and concentrated. The dried product (E)-8 (mp
134–139 °C, 73 mg, 43%) contained no (Z)-8 and was recrystallized from cyclohexane (2×),
1
yielding pure (E)-8 with mp 144–146.5 °C. H NMR (CDCl3, 400 MHz)
δ
1.58 (d, 3J = 7.1 Hz,
3H, 3´-CH3 cis to aryl), 2.12 (s, 6H, 2-/6-CH3), 7.06 and 7.13 (A2B system, 3J = 7.4 Hz, 2+1H,
3-/5-H and 4-H), 7.33 (q, 3J = 7.1 Hz, 1H, 3´-H) ppm; 13C NMR (CDCl3, 100.6 MHz)
15.1
δ
(qd, 1J = 127 Hz, 2J = 3.8 Hz, 3´-CH3 cis to aryl), 19.9 (qd, 1J = 126.7 Hz, 3J = 5.0 Hz, 2-/6-
CH3), 127.3 (ddq, 1J = 158 Hz, 3J = 7.3 Hz, 3J = 5.0 Hz, C-3/-5), 127.7 (sharp d, 1J = 159 Hz, C-
4), 132.7 (q, 3J = 6.1 Hz, 2J = 0 Hz, C-2´), 133.7 (unresolved m, C-1), 136.5 (dq, 3J = 7.5 Hz, 2J =
6 Hz, C-2/-6), 143.1 (dq, 1J = 157.7 Hz, 2J = 7.0 Hz, C-3´), 172.0 (d, 3J = 6.9 Hz, CO2H) ppm,
3
assigned through selective {2,6-CH3} decoupling → C-1 as a pseudo-q J = 7.3 Hz (to 3-/5-H
3
and 3´-H), C-2/-6 as a distorted d J = 7.5 Hz, and C-3/-5 as a sharp dd; IR (KBr) ν 3600–2300
(O–H), 2658, 2532, 1687, 1633, 1417, 1286, 936, 767, 714 cm–1. Anal. calcd for C12H14O2
(190.24): C, 75.76; H, 7.42. Found: C, 75.91; H, 7.63.
(Z)-2´-(2,6-Dimethylphenyl)but-2-enoic Acid [(Z)-8]. The purified dimeric (Z)-1&t-BuOMe
was obtained from the bromoalkenes11 (Z,E)-5 (250 mg, 1.11 mmol), then converted into
crystalline (E)-1&3THF that was dissolved at –70 °C in dry toluene (1.0 mL) and immediately
ACS Paragon Plus Environment
28

Page 29 of 38
Journal of the American Chemical Society
poured onto solid CO2. After warm-up and dilution with Et2O (15 mL), the mixture was
extracted with aqueous NaOH (2 M. 4 × 5 mL) and the Et2O layer discarded. The combined
aqueous phases were acidified, shaken with Et2O (3 × 15 mL), and thereafter discarded. The
combined Et2O extracts were washed with distilled water until neutral, dried over MgSO4,
filtered, and concentrated to yield the crude acid (Z)-8 (mp 111–118 °C, 31 mg, 15%) without
1
2
3
4
5
6
7
8
9
1
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
any trace of (E)-8. Pure (Z)-8 had mp 120–122 °C (cyclohexane). H NMR (CDCl3, 400 MHz)
δ
2.20 (s, 6H, 2-/6-CH3), 2.25 (d, 3J = 7.3 Hz, 3H, 3´-CH3 trans to aryl), 6.22 (q, 3J = 7.3 Hz, 1H,
3´-H), 7.04 and 7.11 (A2B system, 3J = 7.5 Hz, 2+1H, 3-/5-H and 4-H) ppm; 13C NMR (CDCl3,
100.6 MHz)
δ
16.2 (qd, 1J = 128 Hz, 2J = 3 Hz, 3´-CH3 trans to aryl), 20.4 (qd, 1J = 126 Hz, 3J =
4.2 Hz, 2-/6-CH3), 127.3 (dm, 1J = 157 Hz, C-3/-5), 127.5 (sharp d, 1J = 159.4 Hz, C-4), 131.4
(m, C-2´), 136.7 (dq, 3J = 7 Hz, 2J = 6 Hz, C-2/-6), 138.3 (unresolved, C-1), 145.2 (dq, 1J = 153
Hz, 2J = 7 Hz, C-3´), 171.1 (d, 3J = 13.8 Hz, CO2H) ppm, assigned through selective {3´-CH3
3
3
and 2,6-CH3} decoupling → 2-/6-CH3 as a d J = 4.2 Hz, C-2/-6 as a d J = 7.0 Hz, C-2´as a s
1
3
(2J = 0), C-3´ as a sharp d J = 152 Hz, C-3/-5 as a d J = 7.5 Hz; IR (KBr) ν 3600–2400 (O–
H), 1689, 1622, 1260, 1244, 1229, 776 cm–1. Anal. calcd for C12H14O2 (190.24): C, 75.76; H,
7.42. Found: C, 75.90; H, 7.49.
(E)-1´-Iodo-1´-(2,6-dimethylphenyl)propene [(E)-9]. The bromoalkenes11 (Z,E)-5 (200 mg,
0.89 mmol) were converted into the purified dimer (Z)-1&t-BuOMe, then dissolved in anhydrous
t-BuOMe (1.0 mL) and stirred at 2 °C under argon gas cover during titration with a solution of
elemental iodine (225 mg, 0.89 mmol) in t-BuOMe (1.5 mL) for a persisting coloring. After
addition of two further droplets of the iodine solution, the mixture was kept at rt for 15 min and
then diluted with water (12 mL) and Et2O. The aqueous layer was shaken with Et2O (3 × 4 mL);
the four Et2O phases were combined, shaken with aqueous NaHSO3 (37%, 2 × 4 mL), washed
with distilled water until neutral, dried over MgSO4, filtered, and concentrated. The crude
material (mp 57–62 °C, 104 mg, 43%) contained (E)-9 and (Z)-9 (94:6) and was recrystallized
1
from methanol (ca. 1 mL) to give analytically pure, colorless (E)-9 (53 mg, mp 63–65 °C). H
NMR (CDCl3, 400 MHz)
6.56 (q, 3J = 7.0 Hz, 1H, 2´-H), 7.01 and 7.09 (A2B system, 3J = 7.4 Hz, 2+1H, 3-/5-H and 4-H)
ppm, assigned through comparison with (Z)-9 and (E)-5; 13C NMR (CDCl3, 100.6 MHz)
17.0
δ
1.36 (d, 3J = 7.0 Hz, 3H, 2´-CH3 cis to aryl), 2.23 (s, 6H, 2-/6-CH3),
δ
(qd, 1J = 127.6 Hz, 2J = 3.4 Hz, 2´-CH3 cis to aryl), 19.5 (qd, 1J = 127 Hz, 3J = ca. 5 Hz, 2-/6-
CH3), 94.7 (dq, 2J = 6.0 Hz, 3J = 8.0 Hz, C-1´), 127.6 (ddq, 1J = 158.3 Hz, 3J = 7.3 Hz, 3J = 5.0
Hz, C-3/-5), 128.0 (sharp d, 1J = 159.1 Hz, C-4), 135.4 (dq, 3J = 7.5 Hz, 2J = 6 Hz, C-2/-6), 137.9
(dq, 3J = 16.0 Hz, 2J = 6.7 Hz, C-2´), 139.5 (unresolved m, C-1) ppm, assigned through
ACS Paragon Plus Environment
29