Concave reagents. 20. Sterically shielded m-terphenyls as selective agents in general protonations

Article abstract of DOI:10.1021/jo961094r

New m-terphenyls with acidic substituants in the 2′-position have been used in general protonations leading to reagent-controlled selectivity enhancements: up to 96:4 for the γ/α-protonation of unsymmetrically substituted allyl anions, up to 97:3 for the protonation of cyclohexyl anions generating preferentially the thermodynamically less stable cis-products. In order to allow a general, reagent-controlled protonation the acidity of the protonating agent should be as low as possible.

Full text of DOI:10.1021/jo961094r

7922
J . Org. Chem. 1996, 61, 7922-7926
Con ca ve Rea gen ts. 20. Ster ica lly Sh ield ed m -Ter p h en yls a s
Selective Agen ts in Gen er a l P r oton a tion s1
U. Lu¨ning,* H. Baumgartner, C. Manthey, and B. Meynhardt
Institut fu¨r Organische Chemie, Olshausenstr. 40, D-24098 Kiel, Germany
Received J une 11, 1996X
New m-terphenyls with acidic substituents in the 2′-position have been used in general protonations
leading to reagent-controlled selectivity enhancements: up to 96:4 for the γ/R-protonation of
unsymmetrically substituted allyl anions, up to 97:3 for the protonation of cyclohexyl anions
generating preferentially the thermodynamically less stable cis-products. In order to allow a general,
reagent-controlled protonation the acidity of the protonating agent should be as low as possible.
Most asymmetrically substituted carbon atoms have
In this work, we introduce 2′-substituted m-terphenyls
as selective proton donors for diastereoselective and
regioselective protonations.
a hydrogen atom as one of the four different substituents.
If this hydrogen atom could be attached selectively to a
prochiral carbon atom, for instance by a reagent-
controlled protonation, many enantio- and stereoselec-
tivity problems would be solved. But there are always
two mechanistic pathways possible for a kinetically
controlled protonation: the general and the specific
protonation (see Figure 1 and ref 2 ). The proton can
either be transferred directly from the acid to the
substrate (reagent control, general protonation) or the
acid is only the source for the protons that are transferred
to the substrate by the solvent or a cosolvent (shuttle
mechanism, specific protonation). Therefore, reagents
and reaction conditions have to be worked out to allow
general protonation exclusively.
Our concept of concave reagents5,9,10 has taken the
geometry of enzymes and has transferred it to standard
reagents of organic chemistry like acids and bases.11 In
these reagents a macrocycle is spanned by a bridge
carrying a functional group in such a way that a reaction
only can occur on the inside of the bimacrocycle. The
nature of this bridge is very important because confor-
mations that allow the functional group to undergo
reactions on the convex outside must be avoided. Bis-
ortho-substituted phenyl derivatives ensure this, and
very useful building blocks are 2′-substituted m-terphe-
nyls such as 1-10, especially if the outer aryl rings are
also ortho-substituted.
Selective protonations have been investigated thor-
oughly. Some progress has been made toward enanti-
oselectivity3 but even the problem of diastereoselective
or regioselective protonation is not solved yet.4 To
increase the selectivities of protonations, concave acids
and other sterically extremely shielded acids have been
developed.5 Concave acids (and the conjugate acids of
concave bases) have been used to govern the diastereo-
selectivity6 and the regioselectivity of the protonation of
nitronate ions (C- vs O-protonation, “soft Nef-reaction”7,8).
* To whom correspondence should be addressed. Fax: +49-431-
880-1558. E-mail: noc03@rz.uni-kiel.d400.de.
Hart12 developed a method to synthesize m-terphenyls
that may be substituted in all ortho-positions of the outer
aryl rings and developed ways to incorporate these into
polymacrocycles. The key intermediate of Hart’s method
is the 2′-Grignard compound 1, which can be quenched
by iodine to give the 2′-iodide 2.
Starting from the iodide 2, we have synthesized various
2′-substituted m-terphenyls via the lithium compound
3: open chain tetra-o-methyl derivatives 4-10,13 bimac-
X Abstract published in Advance ACS Abstracts, October 1, 1996.
(1) Concave Reagents. 19. Ross, H.; Lu¨ning, U. Tetrahedron 1996,
52, 10879-10882.
(2) For general and specific acid catalysis see: Maskill, H. The
Physical Basis of Organic Chemistry; Oxford University Press: New
York, 1993. For general and specific protonation see ref 20.
(3) Review: (a) Waldmann, H. Nachr. Chem. Tech. Lab. 1991, 39,
413-418. Recent examples: (b) Fehr, C.; Galindo, J . Angew. Chem.
1994, 106, 1967-1969; Angew. Chem., Int. Ed. Engl. 1994, 33, 1888-
1889. (c) Ishihara, K.; Kaneda, M.; Yamamoto, H. J . Am. Chem. Soc.
1994, 116, 11179-11180. (d) Fuji, K.; Kawabata, T.; Kuroda, A.; Taga,
T. J . Org. Chem. 1995, 60, 1914. (e) See the references in refs 3a-d,
4c.
(4) Examples for diastereoselective protonations of cyclic carban-
ions: (a) Hu¨nig, S.; Keita, Y.; Peters, K.; v. Schnering, H. G. Chem.
Ber. 1994, 127, 1495-1500. (b) Gerlach, U.; Haubenreich, T.; Hu¨nig,
S.; Keita, Y. Chem. Ber. 1993, 126, 1205-1215. (c) Krause, N. Angew.
Chem. 1994, 106, 1845-1847; Angew. Chem., Int. Ed. Engl. 1994, 33,
1764-1765 and references cited therein. For diastereoselective pro-
tonations of nitronate ions: (d) Zimmerman, H. E. ; Mariano, P. S. J .
Am. Chem. Soc. 1968, 90, 6091-6096. (e) Zimmerman, H. E. Acc.
Chem. Res. 1987, 20, 263-268. (f) See ref 6. Examples for regioselective
protonations: (g) Tanaka, J .; Nojima, M.; Kusabayashi, S.; Nagase,
S. J . Chem. Soc., Perkin Trans. 2 1987, 673-678. (h) Ikeda, Y.; Ukai,
J .; Ikeda, N.; Yamamoto, H. Tetrahedron 1987, 43, 743-753.
(5) Lu¨ning, U. Top. Curr. Chem. 1995, 175, 57-99.
(7) Lu¨ning, U.; Baumstark, R.; Mu¨ller, M.; Wangnick, C.; Schillinger,
F. Chem. Ber. 1990, 123, 221-223.
(8) Review for the Nef-reaction: Pinnick, H. W. Org. React. 1990,
38, 655-792.
(9) Lu¨ning, U. Liebigs Ann. Chem. 1987, 949-955.
(10) Our concept of concave reagents has been adopted, and “reaction
bowls” have been synthesized: Goto, K.; Tokitoh, N.; Okazaki, R.
Angew. Chem. 1995, 107, 1202-1203; Angew. Chem. Int. Ed. Engl.
1995, 34, 1124-1126.
(11) Recent reviews for enzyme mimics: (a) Kirby, A. J . Angew.
Chem. 1996, 108, 769-790; Angew. Chem., Int. Ed. Engl. 1996, 35,
705-724. (b) Murakami, Y.; Kikuchi, J .; Hisaeda, Y.; Hayashida, O.
Chem. Rev. 1996, 96, 721-758.
(12) Hart, H.; Vinod, T. K. Top. Curr. Chem. 1994, 172, 119-178.
(13) Lu¨ning, U.; Baumgartner, H. Synlett 1993, 571-572.
(6) Lu¨ning, U.; Mu¨ller, M. Angew. Chem. 1992, 104, 99-102; Angew.
Chem., Int. Ed. Engl. 1992, 31, 80-82.
S0022-3263(96)01094-8 CCC: $12.00 © 1996 American Chemical Society

Sterically Shielded m-Terphenyls
J . Org. Chem., Vol. 61, No. 22, 1996 7923
rocyclic m-terphenyl carboxylic acids,14 a bimacrocyclic
sulfinic acid,15 and a bimacrocyclic thiol acetate.13 Other
nonmacrocyclic m-terphenyl-2′-carboxylic acids have been
investigated by Siegel16 and others.12,17
In the bimacrocyclic 2′-substituted m-terphenyls as
well as in the open-chain precursors the functional groups
are sterically well shielded. Although in bimacrocylic
m-terphenyls the shielding of a functional group in the
2′-position is much larger than in the open-chain precur-
sors (see methyl carboxylate cleavages14b), a tetra-o-
substitution of a 2′-substituted m-terphenyl has still a
strong steric effect (for reaction retardations of a sulfonyl
chloride and a carbaldehyde oxime see ref 13, for cleft
molecules see ref 18 ).
We have therefore investigated tetra-o-methyl-m-ter-
phenyls carrying an acidic function in the 2′-position in
protonation reactions. In addition to the 2′-carboxylic
and sulfinic acids 4 and 5 for which the acidities have
been determined,15,16 also the 2′-thiol 6 and the 2′-carbinol
7 can also be used as proton sources if the pKa of the
anion to be protonated is large enough.
In two model reactions (protonation of the anions 11
and 16) we have investigated m-terphenyl derivatives as
acids and compared them to other acids.
Ta ble 1. Regioselectivities of th e P r oton a tion of th e
Allyl An ion s 11a a n d 11b by Va r iou s Acid s22
acid
12a :13a 12b:13b
ref
14b (R ) Me)
6 (X ) SH)
3:97
4:96
7:93
9:91
10:90
20:80
20:80
a
a
14a (R ) H)
21:79
20/a
a
a
a
a
a
25
a
4 (X ) COOH)
7 (X ) CH2OH)
thiophenol
5 (X ) SOOH)
2,2,5,5-tetramethylcyclopentanol23 25:75
2′-hydroxy-5′-nitro-m-terphenyl24
2,6-dimethylphenol
tert-butyl alcohol
benzoic acid
32:68
34:66
36:64
37:63
39:61
53:47
90:10
a
benzyl alcohol
1514a
a
diethyl malonate
80:20
20/a
a
This work.
A reversed regioselectivity was observed when mal-
onates were used as protonating species. The R-proto-
nated product 12a was formed in excesses of up to 9:1.20
An explanation for this reversed selectivity may be the
formation of complexes between the malonates and the
lithium ions in the contact ion pairs prior to the proton
transfer from the malonate to the anion 11a .
The nature of lithium-11a aggregates plays an im-
portant role for R- and γ-selectivity of the protonation.
The reaction is very sensitive to solvent or stoichiometry
changes and to the addition of co-solvents. With diethyl
ether as solvent and a 1:2:2.5 stoichiometry of 12:HMPT:
n-BuLi, reagent-controlled protonations were possible.20
This stoichiometry has been used in this work, as well
(see Table 1).
Variation of the substituents at the silicon atom (11c)
or in the γ-position of the allyl anion (11d ) led to
substrate-controlled R/γ-selectivities. Change of the
proton sources from 2,6-di-tert-butylphenol (14a ) to di-
ethyl malonate had no influence on the R/γ-selectivities:
<10:>90 for 11c, >90:<10 for 11d .21
Regioselective P r oton a tion of Allyl An ion s
If an unsymmetrical allyl anion is protonated, two
products are possible because the proton can attack the
R- or the γ-position. Using the (triphenylsilyl)allyl anion
11a 19 as a model compound, specific reagents and reac-
tion conditions for a selective protonation have been
found to produce the allyl or the vinyl compounds 12a
or 13a selectively in large excesses.20 While most acids
led to a mixture of 12a and 13a , 2,6-di-tert-butylphenol
(14a ) formed the vinyl compound 13a with a 9:1 selectiv-
ity presumably caused by steric control.
We have now applied m-terphenyls carrying acidic
functionalities in the 2′-position in the protonation of 11a
and a second allyl anion 11b. The regioselectivities are
compared in Table 1.
Table 1 clearly shows that the preference of the
γ-protonation cannot only be achieved by changing the
methyl groups of 2,6-dimethylphenol to tert-butyl groups
(1420). The 13/12-selectivity of other acids is also strongly
increased when the acidic function (thiol, carboxylic acid,
or methanol) is located in the 2′-position of a tetra-o-
(14) (a) Lu¨ning, U.; Wangnick, C.; Peters, K.; v. Schnering, H. G.
Chem. Ber. 1991, 124, 397-402. (b) Lu¨ning, U.; Wangnick, C. Liebigs
Ann. Chem. 1992, 481-484.
(15) Lu¨ning, U.; Baumgartner, H.; Wangnick, C. Tetrahedron 1996,
52, 599-604.
(16) (a) Chen, C.-T.; Siegel, J . S. J . Am. Chem. Soc. 1994, 116, 5959-
5960. (b) Chen, C.-T.; Chadha, R.; Siegel, J . S. Tetrahedron Lett. 1995,
36, 8403-8406.
(21) Meynhardt, B. Diploma Thesis, Universita¨t Kiel, 1995.
(22) 13a can be formed as the cis- or trans-isomer, but only trans-
13a has been detected.
(23) Synthesis: Beckwith, A. L. J .; Lawrence, T. J . Chem. Soc.
Perkin Trans. 2 1979, 1535-1539. We thank Dr. M. Gelbert for a
sample.
(24) Synthesis: J ohnson, B. P.; Gabrielsen, B.; Matulenko, M.;
Dorsey, J . G.; Reichardt, C. Anal. Lett. 1986, 19, 939-962.
(25) Wangnick, C. Ph.D. Thesis, Universita¨t Freiburg, 1991.
(17) J ones, D. H.; Ragg, W. R. J . Chem. Soc. C 1968, 2154-2155.
(18) Zimmerman, S. C. Top. Curr. Chem. 1993, 165, 71-102.
(19) (a) Corriu, R.; Masse, J . J . Organomet. Chem. 1973, 57, C5-
C8. (b) Corriu, R. J . P.; Masse, J .; Samate, D. J . Organomet. Chem.
1975, 93, 71-80.
(20) Lu¨ning, U.; Wangnick, C.; Ku¨mmerlin, M. Chem. Ber. 1994,
127, 2431-2433.

7924 J . Org. Chem., Vol. 61, No. 22, 1996
Lu¨ning et al.
Ta ble 2. cis/tr a n s-Ra tio of th e Ester 17 a fter
P r oton a tion of th e Lith iu m En ola te 16 by Va r iou s Acid s
a t RT in THF a n d THF /DMF (4:1)
methyl-m-terphenyl (6, 4, 7). 2′-Hydroxy-5′-nitro-m-
terphenyl and the 3,3′′,5,5′′-substituted acid 15 only
formed 13a and 12a in a 2:1 and 1:1 ratio, respectively.
The tetra-ortho-substitution is necessary to obtain 13/
12-selectivities of >90:<10. Thiol 6 was the most selec-
tive m-terphenyl, reflecting the slightly higher selectivity
of thiophenol in comparison to benzoic acid and benzyl
alcohol.
The selectivity increase can be rationalized by the
geometry of the tetra-ortho-substituted m-terphenyls 4,
6, and 7. The tetra-ortho-substitution leads to an almost
perpendicular twist between the aryl rings of the
m-terphenyl.16b
It is remarkable that the 13/12-selectivities obtained
with the m-terphenyls 4, 6 and 7 do not vary much
although the acidity of the functional group in the 2′-
position varies from alcohol (7) to carboxylic acid (4). Only
the use of the most acidic proton source, the sulfinic acid
5, leads to a smaller selectivity. Whether this is a result
of its acidity or of its hygroscopic behavior cannot be
answered. But the almost nonexisting acidity depen-
dence of the selectivities is in contrast to the results
observed for the protonation of the cyclohexyl anion 16
where a strong dependence of the selectivities on the
acidity of the proton source was observed.
cis-17/trans-17
acid
THF
THF/DMF (4:1)
7 (X ) CH2OH)
triphenylmethanol
6 (X ) SH)
tert-butyl alcohol
4 (X ) COOH)
14b (R ) Me)
benzyl alcohol
94:6
81:19
64:36
58:42
97:3
81:19
54:46
50:50
51:49
51:49
Also, the di-tert-butylphenol 14b was unselective, even
at low temperature (-70 °C, 14b: cis-17/trans-17 53:47).
But the use of alcohols that are less acidic proton sources
was successful in the diastereoselective protonation of 16.
The best result was obtained with the m-terphenyl
methanol 7 that formed the cis-product in an excess of
g94:<6.
Dia ster eoselective P r oton a tion of Cycloh exyl
An ion s
If a carbanion is part of a ring system that carries
appropriate substituents protonation can lead to cis- and
trans-isomers.4 The cis/trans-ratio should be controllable
if a general protonation was possible. By using buffers
of substituted 1,10-phenanthrolines, a cis-selective gen-
eral protonation of cyclic nitronate ions is possible.6 Due
to the concave wrapping of the proton transferred from
the protonated 1,10-phenanthroline to the nitronate ion,
the thermodynamical less stable cis-products were formed
in large excess (>12:1). This contrathermodynamic pro-
tonation is favored because the smallest substituent in
the product, the proton, is a very large pseudosubstituent
in the transition state when it is still partly bound to
the shielded 1,10-phenanthroline.
As Table 2 shows, good selectivities have only been
found for acids that are sterically shielded and have a
relatively low acidity. Steric shielding alone is not
sufficient for a cis-selective protonation of 16. In the thiol
6 the proton is more shielded than in the methanol 7
because the distance between the aryl ring and the proton
is one atom longer in the alcohol 7. But 7 gives the larger
cis/trans-ratio. When the alcohol 7, the thiol 6, and the
acid 4 are compared with one another there is a clear
trend: the more acidic the m-terphenyl the more unse-
lective is the protonation.
Table 2 also compares different alcohols. The cis-
selectivity increases from the nonshielded benzyl alcohol
via tert-butyl alcohol and triphenylmethanol to the m-
terphenyl alcohol 7. The contrathermodynamic cis-
selectivity of the protonation by 7 can be understood if
the transition states leading to trans- and cis-17 are
looked at. As for the cis-selective protonation of a cyclic
nitronate ion,6 the smallest substituent in the product,
the proton, is a very large pseudosubstituent in the
transition state when it is still partly bound to the
alcohol. In the pseudoequatorial attack leading to cis-
17, the interaction with adjacent hydrogen atoms of the
cyclohexane ring is smaller. Therefore, this pathway
dominates.
We have tried to extend this selective protonation to
ester stabilized cyclic carbanions for which diastereose-
lective protonations have already been investigated by
Hu¨nig et al.4b (see Table 2). In the case of the methyl
ester they found remarkable “uniform” cis/trans-selectivi-
ties (max 73:27).
The protonations of the tert-butyl-substituted ester
enolate 1626 with the m-terphenyl acid 4 or even the
m-terphenyl thiol 6 hardly showed any selectivities.
(26) To deprotonate 17 LDA was used, although problems in
selective deuterations have been encountered when LDA has been
used: (a) Seebach, D. Angew. Chem. 1988, 100, 1685-1715; Angew
Chem., Int. Ed. Engl. 1988, 27, 1624-1654. (b) Mohrig, J . R.; Lee, P.
K.; Stein, K. A.; Mitton, M. J .; Rosenberg, R. E. J . Org. Chem. 1995,
60, 3529-3532.
For the protonation of other anions like the allyl anion
11 or the nitronate anions by sterically shielded acids,
no acidity dependence has been observed. A possible

Sterically Shielded m-Terphenyls
J . Org. Chem., Vol. 61, No. 22, 1996 7925
2,2′′,6,6′′-Tetr am eth yl-m-ter ph en yl-2′-car baldeh yde (10).
Under nitrogen, 8.7 mL (20.0 mmol) of a 2.3 M solution of
n-butyllithium in n-hexane was added to a solution of 8.28 g
(20.0 mmol) of m-terphenyl iodide 2 in 50 mL of dry cyclohex-
ane. A colorless precipitate formed. After the mixture was
stirred at rt for 15 h, 7.0 mL (6.7 g, 100 mmol) of dry DMF
was added to the suspension at 0 °C within 5 min under
nitrogen. During stirring at rt for an additional 3 h, the
mixture became more viscous. After hydrolysis with 50 mL
of water, the mixture was extracted with diethyl ether (100
mL total). The organic layer was washed with water (50 mL
twice). After evaporation of the solvent, the remaining slightly
yellow material was recrystallized from n-hexane yielding 3.82
g of 10 as thin colorless needles, mp 145 -146 °C (n-hexane,
dec). IR (KBr): ν 1695 (CdO), 1570 (arom). 1H-NMR (400
MHz, CDCl3): δ 1.98 (s, 12 H), 7.10 (d, J ) 8.0 Hz, 4 H), 7.15-
7.30 (m, 4 H), 7.66 (t, J ) 8.0 Hz, 1 H), 9.61 (s, 1 H). MS (EI,
70 eV): m/z 314 (5), 299 (12), 296 (100), 281 (46), 266 (14),
165 (28). Anal. Calcd for C23H22O (314.44): C, 87.86; H, 7.05.
Found: C, 87.86; H, 7.04.
2,2′′,6,6′′-Tetr a m eth yl-m -ter p h en yl-2′-m eth a n ol (7). (a )
By Rea ction of 3 w ith Ga seou s F or m a ld eh yd e. Under
nitrogen, 3.0 mL (7.5 mmol) of a 2.5 M solution of n-
butyllithium in n-hexane was added to a solution of 2.80 g (7.00
mmol) of 2 in 20 mL of dry cyclohexane. A yellow color
appeared quickly, and a colorless solid precipitated. After 15
h of stirring at rt, the solvents were evaporated under nitrogen
with reduced pressure and the residue was dried in vacuo. The
lithium salt 3 was dissolved at -50 °C in 20 mL of dry diethyl
ether and stirred for 10 min. In a second flask ca. 5 g (ca.
0.17 mol) of dry paraformaldehyde (previously dried for several
days in a vacuum desiccator over P4O10) was heated to 180
°C, and the developing formaldehyde was transported by a
slight nitrogen flow into the solution of 3, which was vigorously
stirred at -50 °C. After complete depolymerization of the
paraformaldehyde, the turbid reaction mixture was hydrolyzed
by addition of 50 mL of ice-water and stirred at rt for 30 min.
The mixture was filtered and extracted with diethyl ether (100
mL total), and the organic layer was washed with water (twice
50 mL) and dried with CaCl2. After evaporation of the
solvents, a slightly yellow oil remained from which colorless
needles crystallized at -18 °C. Recrystallization from n-
hexane gave 1.30 g (59%) of fine colorless crystals, mp 161-
162 °C (n-hexane).
(b) By Red u ction of 10 w ith LiAlH4. A 1.50 g (4.78
mmol) portion of 10 dissolved in 30 mL of dry diethyl ether
was added within 1.5 h to a refluxing suspension of 181 mg
(4.77 mmol) of lithium aluminum hydride in 25 mL of dry
diethyl ether. After being refluxed for 2 h, the mixture was
carefully hydrolyzed. Aluminum hydroxide was dissolved by
addition of 30 mL of 10% H2SO4 solution. The organic layer
was washed with 100 mL of water, and the water layer was
extracted twice with diethyl ether (100 mL total). The
combined organic layer was dried with CaCl2. After evapora-
tion of the solvent 1.5 g of crude product yielded 1.35 g (89%)
of 7as colorless crystals by recrystallization from n-hexane,
mp 159-160 °C (n-hexane). IR (KBr): ν 3522 (OH), 1579
(arom). 1H-NMR (400 MHz, DMSO-d6): δ 1.95 (s, 12 H), 3.77
(s, 2 H), 3.98 (s, 1 H), 6.98 (d, J ) 8 Hz, 2 H), 7.06-7.18 (m,
6 H), 7.45 (t, J ) 8 Hz, 1 H). 1H-NMR (250 MHz, CDCl3): δ
2.06 (s, 12 H), 4.04 (s, 2 H), 7.10 (d, J ) 8 Hz, 2 H), 7.20-7.30
(m, 6 H), 7.46 (t, J ) 8 Hz, 1 H). MS (EI, 70 eV): m/z 316
(M+, 5), 298 (100), 283 (59), 268 (22), 253 (21). Anal. Calcd
for C23H24O (316.46): C, 87.30; H, 7.64. Found: C, 87.19; H,
7.63.
F igu r e 1. General and specific protonation of an anion A-.
Direct proton transfer from the acid of the buffer (XH) to the
anion A- leads to reagent-controlled general protonation, while
dissociation of the acid and proton transfer via the protonated
solvent So‚H+ gives a specific protonation that is not control-
lable by the nature of the reagent.
explanation is given in Figure 1. In all protonations the
two pathways, the general and the specific protonation,2
have to be looked at. Only a general protonation will lead
to a reagent-controlled protonation.
In the solution of the proton source, the protons may
be bound to the protonating agent (XH) or to the solvent
(So‚H+). The equilibrium depends on the acidity of the
buffer acid and on the basicity of the solvent. If other
bases, e.g., cosolvents (So′), are present in solution they
also may act as proton shuttles, and one more species
capable of protonating the anion A- is present (So′‚H+).
Furthermore, the rate of reaction between the anion and
the protonated solvent will depend on the basicity of the
anion. These relations may explain the experimental
findings.
To estimate the reaction rates for the general and the
specific protonation the rate constants as well as the
concentrations of XH and So‚H+ have to be known. If
the protonating acid becomes more acidic the equilibrium
will be shifted from XH to So‚H+, increasing the rate for
the specific protonation (see Table 2: 7, 6, 4). The XH/
So‚H+ equilibrium will also shift toward So‚H+ (and
So′‚H+) if the basicity of the solvent/cosolvent increases
[diethyl ether/HMPT (Table 1) vs THF/diisopropyl amine
(Table 2)]. Therefore, the allyl anions 11 can be selec-
tively protonated by m-terphenyls with varying acidity,
while the selective protonation of 16 is restricted to the
alcohol 7.
In order to understand the results of the nitronate ion
protonation6 the basicity of the anions have to be
compared. Nitro compounds are much more acidic than
allyls and esters. Therefore, a larger concentration of
So‚H+ would be necessary to protonate a nitronate ion
with the same rate as the ions 11 and 16.
As a result of these investigations the following re-
quirements for selective reagent-controlled protonations
have to be met: (i) the solvent and all cosolvents should
be as slightly basic as possible, (ii) the acidity of XH
should be as low as possible, but it must be guaranteed
that the conjugate base of the buffer will not be able to
deprotonate the product AH.27 Otherwise, a thermody-
namically controlled protonation instead of a kinetically
and reagent-controlled protonation would occur.
m-Terphenyls seem to do this task because their acidity
can be varied widely while keeping the shielding com-
parable. Substitution with tert-butyl groups14,28 increases
the solubility of m-terphenyls even in very unpolar
solvents.
(c) By Red u ction of 4 w ith LiAlH4. Under nitrogen 40
mL of dry THF was added to 277 mg (0.839 mmol) of 4 and
159 mg (4.19 mmol) of LiAlH4. The mixture was refluxed.
After being refluxed for 5 d the mixture was hydrolyzed with
50 g of ice, extracted with 100 mL of dichloromethane, and
dried with MgSO4. Evaporation gave a solid that was recrys-
tallized from n-hexane to give 204 mg (77%) of colorless
crystals.
Exp er im en ta l Section
For general remarks see ref 29.
2,2′′,6,6′′-Tet r a m et h yl-m -t er p h en yl-2′-su lfon yl Ch lo-
r id e (9). Under nitrogen 4.4 mL (10.0 mmol) of a 2.2 M
solution of n-butyllithium in n-hexane was added to a solution
(27) Reference 3d and references cited therein.
(28) Ku¨hl, C. Diploma Thesis, Universita¨t Kiel, 1995.
(29) Lu¨ning, U.; Mu¨ller, M. Liebigs Ann. Chem. 1989, 367-374.

7926 J . Org. Chem., Vol. 61, No. 22, 1996
Lu¨ning et al.
of 4.14 g (10.0 mmol) of 2 in 30 mL of dry cyclohexane. The
mixture turned yellow, and a colorless solid precipitated. After
the mixture was stirred for 15 h at room temperature, the
solvents were evaporated under nitrogen with reduced pres-
sure and the residue was dried in vacuo. At -50 °C the
lithium salt was dissolved in 20 mL of dry diethyl ether and
stirred for 10 min. This solution was thoroughly stirred, and
0.93 mL (1.55 g, 11.5 mmol) of freshly distilled sulfuryl chloride
(bp 68-69 °C) was added through a septum within 1 min. The
solution turned yellow to red. After being stirred for 1 h at
-50 °C, the mixture was slowly warmed to rt, stirred for 2 h,
poured onto 50 mL of saturated NaHSO4 solution, and
extracted with diethyl ether (70 mL total). The organic layer
was washed with 50 mL of water and dried with CaCl2.
Evaporation of the solvent gave 3.7 g of a slightly red product
that was recrystallized from a small amount diethyl ether.
Yield: 2.86 g (75%) (55% yield in a 0.1 mol batch). Mp: 215-
216 °C (diethyl ether), 231 °C [after chromatography, silica
gel/cyclohexane/dichloromethane (1:1)]. IR (KBr): ν 1567
(arom), 1172 (S)O). 1H-NMR (250 MHz, CDCl3): δ 2.08 (s,
12 H), 7.11-7.30 (m, 8 H), 7.77 (t, J ) 7.7 Hz, 1 H). MS (EI,
70 eV): m/z 386, 384 (M+, 4, 12), 350 (1), 319 (6), 40 (100).
Anal. Calcd for C22H21ClO2S (384.94): C, 68.65; H, 5.50.
Found: C, 68.56; H, 5.44.
2,2′′,6,6′′-Tetr a m eth yl-m -ter p h en yl-2′-su lfin ic Acid (5).
(a ) See r ef 13. (b) By Red u ction of 9 w ith LiAlH4. A 322
mg (8.48 mmol) portion of LiAlH4 suspended in 20 mL of
diethyl ether was added to a refluxing solution of 1.35 g (3.50
mmol) of 9 in 80 mL of dry diethyl ether. Because TLC after
3 h showed that the reaction was not complete, an additional
188 mg (4.95 mmol) of LiAlH4 was added, and the mixture
was refluxed for an additional 12 h. After hydrolysis with 50
mL of water, the mixture was acidified with ca. 100 mL of 2
N HCl. The organic layer was washed twice with 100 mL of
water and dried with CaCl2. Concentration to dryness gave a
colorless solid that was recrystallized from cyclohexane/ethyl
acetate (1:1), yielding 625 mg of 5 as fine colorless needles,
mp 173 °C (cyclohexane/ethyl acetate, dec).
2,2′′,6,6′′-Tetr a m eth yl-m -ter p h en yl-2′-th iol Aceta te (8).
A 3.0 g (7.8 mmol) portion of 9 was added to a boiling mixture
of 20 mL of acetic acid p.a., 3 mL of acetic anhydride, 600 mg
(19.4 mmol) of red phosphorus, and 40 mg (0.32 mmol) of
iodine. The red mixture was refluxed for 15 h, and then the
remaining red phosphorus was filtered off. The solution was
mixed with 100 mL of dichloromethane and 100 mL of 10%
sodium dithionite solution. The organic layer was washed
twice with 100 mL of Na2CO3 solution and 100 mL of 10%
NaCl solution. After the organic layer was dried with MgSO4,
the solvent was evaporated and the remaining brown-green
oil was dissolved in n-hexane and filtered. Storage at -18 °C
gave 2.24 g (80%) of slightly red crystals, mp 130 °C (n-
hexane). IR (KBr): ν 1705 (CdO). 1H-NMR (250 MHz,
CDCl3): δ 1.81 (s, 3 H), 2.02 (s, 12 H), 7.06 (d, J ) 7.3 Hz, 4
H), 7.09-7.21 (m, 4 H), 7.55 (t, J ) 7.7 Hz, 1 H). MS (EI, 70
eV): m/z 360 (M+, 26), 318 (82), 303 (100), 288 (22). Anal.
Calcd for C24H24OS (360.53): C, 79.96 H, 6.71. Found: C,
79.93; H, 6.70.
amount of dry diethyl ether, yielding 6.52 g (80%) of 6, mp
130 °C (diethyl ether).
(b) By Red u ction of 5 w ith LiAlH4
. A 4.0 g (105 mmol)
portion of LiAlH4 was added to a solution of 7.10 g (20.2 mmol)
of 5 in 100 mL of dry diethyl ether within 1 h. After being
refluxed for 9 h and stirred at rt for 14 h, the mixture was
poured carefully onto ice and extracted with 700 mL of diethyl
ether. The organic layer was washed with 200 mL of water
and dried with CaCl2. Evaporation of the solvents gave a
slightly yellow residue that was dissolved in a small amount
of dichloromethane/cyclohexane (1:1) and filtered through
silica gel (6 × 4 cm). The solvent was evaporated, and the
colorless residue was recrystallized from little diethyl ether,
giving 2.5 g (39%) of 6.
(c) By Red u ction of 9 w ith LiAlH4. A 1.5 g (40 mmol)
portion of LiAlH4, suspended in 20 mL of diethyl ether, was
added to a refluxing solution of 1.35 g (3.50 mmol) of 9 in 80
mL of dry diethyl ether within 1 h. After being refluxed for 2
d, the mixture was carefully hydrolyzed with 50 mL of water
and acidified with 2 N HCl. The organic layer was washed
twice with 100 mL of water and dried with CaCl2. Evaporation
of the solvent gave a colorless solid that was dissolved in a
small amount of dichloromethane/cyclohexane (1:1) and fil-
tered through silica gel (6 × 4 cm). After evaporation to
dryness the solid was recrystallized from little diethyl ether,
yielding 550 mg (49%) of 6 as colorless needles, mp 140 °C
(diethyl ether). IR (KBr): ν 2557 (SH), 1575 (arom). 1H-NMR
(250 MHz, CDCl3): δ 2.05 (s, 12 H), 2.97 (s, 1 H), 7.04 (d, J )
7.7 Hz, 2 H), 7.08-7.28 (m, 7 H). MS (EI, 70 eV): m/z 318
(M+, 6), 303 (100), 288 (24). Anal. Calcd for C22H22
(318.49): C, 82.97; H, 6.96. Found: C, 82.89; H, 6.97.
S
Dep r oton a tion of 17 a n d Rep r oton a tion . Under nitro-
gen, a 2.5 M solution of n-butyllithium in n-hexane (560 µL,
1.40 mmol) was added dropwise to a solution of diisopropyl-
amine (180 µL, 1.40 mmol) in THF (4 mL) at 0-5 °C. The
solution was stirred for 60 min. After addition of ester 1730
(100 mg, 472 µmol) in THF (1 mL), stirring was maintained
for 60 min. The proton source (71 µmol) was dissolved in the
solvents (200 µL) listed in Table 2. At rt the ester enolate
solution (300 µL, 24 µmol) was added slowly with stirring, and
the solution was stirred for another 14 h. Then a 0.1 M HCl
solution (200 µL) was added, and the aqueous layer was
extracted two times with 2 mL of diethyl ether. The solvents
of the combined organic layers were evaporated, and the
residue was dissolved in diethyl ether (200 µL) and analyzed
by GC (SE 30, OV 101, or OV 17 capillary columns, no special
conditions needed).
Unused ester enolate solution was treated with ethyl iodide
(100 µL, 1.2 mmol) in THF to check for completion of depro-
tonation by GC analysis (less than 4% of ester 17 was
detected).
P r oton a tion of 11. 11 was protonated as previously
described,20 but the experiments were carried out on a slightly
larger scale (200 µL of the anion solution). At least a 40-fold
excess of the acid was used. The degree of deprotonation of
12/13 (generally >95%) was checked by alkylation of the
resulting 11 with ethyl iodide. The GC analyses were carried
out on a 25 m OV17 column connected to 25 m of OV 1701.
Synthesis: 11a ,31 11b,32 11c,33 11d .34
2,2′′,6,6′′-Tetr a m eth yl-m -ter p h en yl-2′-th iol (6). (a ) By
Ba sic Hyd r olysis of 8. A 9.23 g (25.6 mmol) portion of 8
was mixed with 140 mL of water/ethanol (1:1) and 1.43 g (25.5
mmol) of KOH and refluxed for 7 h. After extraction with 250
mL of dichloromethane, the organic layer was washed with
50 mL of water and dried with MgSO4. Evaporation of the
solvents gave a solid that was recrystallized from a small
J O961094R
(31) Eisch, J . J .; Gupta, G. J . Organomet. Chem. 1979, 168, 139-
157.
(32) Hagen, G.; Mayr, H.; J . Am. Chem. Soc. 1991, 113, 4954-4961.
(33) Muchowski, J . M.; Naef, R.; Maddox, M. L. Tetrahedron Lett.
1985, 26, 5375-5378.
(34) Matarasso-Tchiroukhine, E.; Cadiot, P. J . Organomet. Chem.
1976, 121, 155-168.
(30) Karger, B. L.; Stern, R. L.; Zanucci, J . F. Anal. Chem. 1968,
40, 727-735.