Planar chiral PHANOLS as organocatalysts for the Diels-Alder reaction via double hydrogen-bonding to a carbonyl group

Article abstract of DOI:10.1055/s-2003-39890

Planar chiral PHANOLs have been shown to catalyze Diels-Alder reactions of α,β-unsaturated aldehydes and ketones with various dienes. Rate accelerations of up to ca. 30-fold were obtained using the electron deficient 4,12-dihydroxy-7,15-dinitro[2.2]paracyclophane as a catalyst. It is proposed that the carbonyl group of the dienophile is activated via a double hydrogen-bonding mode. Although the PHANOLs are inherently chiral, little or no asymmetric induction was observed when using enantiopure (R)-PHANOL.

Full text of DOI:10.1055/s-2003-39890

LETTER
1121
Planar Chiral PHANOLs as Organocatalysts for the Diels–Alder Reaction via
Double Hydrogen-Bonding to a Carbonyl Group
P
H
A
NOLs as
Org
.
a
nocatalysts
C
via
D
ouble Hydr
h
ogen-Bondin
r
g
istopher Braddock,*a Iain D. MacGilp,b Benjamin G. Perrya
a
Department of Chemistry, Imperial College London, South Kensington, London SW7 2AZ, UK
b
GlaxoSmithKline Ltd, Old Powder Mills, Near Leigh, Tonbridge, Kent, TN11 9AN, UK
Fax +44(207)5945805; E-mail: c.braddock@imperial.ac.uk
Received 4 April 2003
activation via double hydrogen-bonding using a chiral
catalyst. In this letter we report our results using 4,12-di-
hydroxy[2.2]paracyclophane (PHANOL) (1), and the
7,15-dinitroderivative 5, as planar chiral organocatalysts
Abstract: Planar chiral PHANOLs have been shown to catalyze
Diels–Alder reactions of a,b-unsaturated aldehydes and ketones
with various dienes. Rate accelerations of up to ca. 30-fold were ob-
tained using the electron deficient 4,12-dihydroxy-7,15-dini-
tro[2.2]paracyclophane as a catalyst. It is proposed that the carbonyl that activate a,b-unsaturated aldehydes and ketones for
group of the dienophile is activated via a double hydrogen-bonding
Diels–Alder reactions. Double hydrogen-bonding to the
mode. Although the PHANOLs are inherently chiral, little or no
carbonyl group of the dienophiles is proposed to be re-
asymmetric induction was observed when using enantiopure (R)-
sponsible for this effect where the hydroxy groups in
PHANOL.
PHANOLs 1 and 5 are correctly positioned to support two
hydrogen bonds to (and so activate) the dienophiles.10
Key words: organocatalysis, planar chiral bisphenols, double hy-
drogen bonding, Diels–Alder reactions, carbonyl activation
R
The use of Lewis acids to bind and activate carbonyl
O
groups via polarization is ubiquitous in synthetic organic
H
O
chemistry.1 Recently, it has been recognized that hydro-
gen bonding can also be used to activate selected carbon-
yl-containing substrates in lieu of traditional Lewis acids.
For instance, Schreiner has employed a thiourea (25
mol% loading) to activate oxazolidinones for Diels–Alder
reactions.2 Philp has recently shown that a diamidopyri-
dine (100 mol%) activates maleimide for 1,3-dipolar cy-
cloadditions.3 Göbel reported that amidinium ions (25–
100 mol%) were capable of activating 1,2-diketones for
Diels–Alder reactions.4 A sulfonamide has been used to
accelerate the addition of pyrrolidine to 2(5H)-furanone.5
In all these examples multiple hydrogen-bonds are being
formed between multiple hydrogen-bond donors and ac-
ceptors and it is their combined effect which results in the
observed activation. A more general solution would be to
activate carbonyl groups by forming multiple hydrogen
bonds to the oxygen of the carbonyl (of e.g., an aldehyde
or ketone) only. To date there is only one report of such
multiple hydrogen-bonding in order to promote cataly-
sis:6,7 Kelly showed that a derivative of biphenylene diol
(40 mol%) could activate a,b-unsaturated aldehydes and
ketones for Diels–Alder reactions presumably via com-
plex of the type I (Figure 1).8 In this complex the two hy-
drogen bonds are made to the two sp2 lone pairs of the
carbonyl group. This configuration is supported by X-ray
crystallographic evidence of a 1:1 adduct of Kelly’s bi-
phenylene diol derivative8 and in closely related struc-
tures.9 However, there are no examples of carbonyl
H
O
NO2
NO2
Figure 1 Kelly biphenylenediol-dienophile complex I
( )-PHANOL (1), and the single enantiomers are avail-
able via our previously published procedure.11 4,12-Dihy-
droxy-7,15-dinitro[2.2]paracyclophane (5) was also
targeted as a potential catalyst since it was expected to be
a more effective hydrogen-bond donor on account of
the electron withdrawing nitro groups. Its synthesis
commenced by treatment of dibromide 212 with nitronium
tetrafluoroborate in sulfolane to give dinitro[2.2]paracy-
clophane (3) (Scheme 1). Copper(I) catalyzed Ullman
coupling with sodium methoxide in DMF at room temper-
ature provided the dimethoxyether 4.
Deprotection to the desired dinitro-PHANOL (5)13 was
achieved by refluxing in concentrated aqueous HBr. 4,16-
Dihydroxy[2.2]paracyclophane (7)12 was prepared by
treatment of commercially available [2.2]paracyclophane
(6) with lead tetraacetate in TFA14 followed by basic hy-
drolysis. The isomeric compound 7 represents the ‘con-
trol’ PHANOL where double hydrogen-bonding to the
same carbonyl group is not possible by consideration of
the positioning of the hydroxy groups on opposite sides of
the [2.2]paracyclophane.
Solubility studies revealed that PHANOL (1) was only
sparingly soluble in solvents such as dichloromethane,
chloroform and hexane. Dinitro-PHANOL (5) was essen-
Synlett 2003, No. 8, Print: 24 06 2003.
Art Id.1437-2096,E;2003,0,08,1121,1124,ftx,en;D08403ST.pdf.
© Georg Thieme Verlag Stuttgart · New York

1122
D. C. Braddock et al.
LETTER
tially insoluble in these solvents.15 Accordingly, we chose
to examine their catalytic potential in neat dienophile/di-
ene mixtures.
OH
Br
Br
NO2BF4
77%
O2N
O2N
OH
Br
Br
The reaction between neat cyclopentadiene and crotonal-
dehyde (Table 1, entry 1) was chosen as the Diels–Alder
cycloaddition to study initially, since the background rate
between the diene and dienophile is slow at room temper-
ature. In the control experiment with no additive, <1%
background reaction to the cycloadduct occurred over 15
hours (Table 1, entry 1). Addition of 10 mol% PHANOL
(1) or dinitro-PHANOL (5) resulted in a 17% and 37%
conversion respectively to the desired cycloadduct over
the same time period. Using 4,16-dihydroxy[2.2]paracy-
clophane (7) (10 mol%) as an additive gave a small in-
crease in conversion relative to the control reaction (ca
3%). Interestingly, we also note that the use of 10 mol%
BINOL16 as a catalyst gave a 9% conversion to the cy-
cloadduct. The relative difference in reactivity between 7
1
2
3
5
4 mol%
Cu(CH3CN)4BF4
OMe
OMe
OH
O2N
O2N
O2N
O2N
HBr
80%
NaOMe, DMF
66%
OH
4
HO
1. Pb(OAc)4, TFA
2. NaOH
17%
OH
6
7
Scheme 1 Synthesis of PHANOLS 5 and 7
Table 1 Catalytic Diels–Alder Reaction Using PHANOLs 1 and 5
Entrya
1c
Diene
Dienophile
Adduct
Time
15 h
Controlb
1b
5b
O
H
<1 (1.9:1)
17 (3.8:1)
37 (1.9:1)
CHO
2c
3c
4d
10 min
15 h
35 (4.3:1)
79 (5.6:1)
74 (6.2:1)
40
O
H
CHO
CHO
12
3e
36
8e
O
H
15 h
13e
O
H
H
CHO
5c
6c
15 h
15 h
34 (1:4.7)
2 (1.1:1)
74 (1:4.5)
5 (1.8:1)
77 (1:6.5)
7 (1.5:1)
O
H
CHO
CHO
O
O
H
Ph
Ph
O
7f
8
10 min
16 h
26 (6.6:1)
70g
72 (14:1)
73g
–
–
O
O
OEt
OEt
a All reactions carried out with 10 mol% catalyst and a 1:1 ratio of diene:dienophile at r.t.
b Percentage conversion as determined by integration of the relevant signals in the 1H NMR spectrum (CDCl3 solvent). The figures in
parentheses give the endo:exo ratio respectively.
c For characterization of this cycloadduct see ref.17
d For characterization of this cycloadduct see ref.18
e Endo only.
f For characterization of this cycloadduct see ref.19
g Endo:exo ratio not determined.
Synlett 2003, No. 8, 1121–1124 ISSN 1234-567-89 © Thieme Stuttgart · New York

LETTER
PHANOLs as Organocatalysts via Double Hydrogen-Bonding
1123
and 1 and 5 is consistent with a double hydrogen-bonding
mode to the carbonyl group possible in 1 but not in 7 and
the increased acidity (and hence hydrogen bonding abili-
ty) of dintro-PHANOL (5) compared to PHANOL (1).
Further, in line with our proposal that carbonyl activation
occurs via double hydrogen-bonding to the carbonyl
group, thus lowering the LUMO of the dienophile as per
classical Lewis acids, the endo:exo ratio of the cycload-
duct increased from 1.9:1 in the control experiment to
3.8:1 when employing catalytic quantities of 1. However,
despite the significant rate acceleration with 5 as the cata-
lyst, the endo:exo ratio remained essentially unchanged.
Acknowledgment
We thank GlaxoSmithKline plc for a CASE award (to B. G. P.) and
the EPSRC for a CNAA award (to D. C. B.).
References
(1) (a) Shambayati, S.; Crowe, W. E.; Schreiber, S. L. Angew.
Chem., Int. Ed. Engl. 1990, 29, 256. (b) Shambayati, S.;
Schreiber, S. L. In Comprehensive Organic Chemistry, Vol.
1; Trost, B. M.; Fleming, I., Eds.; Pergamon Press: Oxford,
1991, 283.
(2) Schreiner, P. R.; Wittkopp, A. Org. Lett. 2002, 4, 217.
(3) Rowe, H. L.; Spencer, N.; Philp, D. Tetrahedron Lett. 2000,
41, 4475.
(4) (a) Schuster, T.; Kurz, M.; Göbel, M. W. J. Org. Chem.
2000, 65, 1697. (b) For the use of an axially chiral
amidinium ion see: Schuster, T.; Bauch, M.; Dürner, G.;
Göbel, M. W. Org. Lett. 2000, 2, 179.
(5) Raposo, C.; Almaraz, M.; Crego, M.; Mussons, M. L.; Pérez,
N.; Caballero, M. C.; Morán, J. R. Tetrahedron Lett. 1994,
35, 7065.
(6) (a) For a urea doubly hydrogen bonding to the ether oxygen
of 6-methoxy allyl vinyl ether to promote a Claisen
rearrangement see: Curran, D. P.; Kuo, L. H. Tetrahedron
Lett. 1995, 36, 6647. (b) For the activation of nitrones with
thioureas see: Okino, T.; Hoashi, Y.; Takemoto, Y.
Tetrahedron Lett. 2003, 44, 2817.
(7) For double hydrogen bonding to an epoxide and concomitant
ring-opening with diethylamine see: (a) Hine, J.; Linden, S.-
M.; Kanagasabapathy, V. M. J. Am. Chem. Soc. 1985, 107,
1082. (b) Hine, J.; Linden, S.-M.; Kanagasabapathy, V. M.
J. Org. Chem. 1985, 50, 5096.
(8) Kelly, T. R.; Meghani, P.; Ekkundi, V. S. Tetrahedron Lett.
1990, 31, 3381.
(9) (a) Saied, O.; Simard, M.; Wuest, J. D. J. Org. Chem. 1998,
63, 3756; and references cited therein. (b) Hine, J.; Ahn, K.;
Gallucci, J. C.; Linden, S.-M. J. Am. Chem. Soc. 1984, 106,
7980; and references cited therein.
Further Diels–Alder reactions are shown in Table 1. Both
a,b-unsaturated aldehydes (entries 2–6) and ketones (en-
try 7) were subject to rate accelerations in the presence of
catalytic quantities of PHANOLs 1 or 5 but esters (entry
8) were not. Typically a doubling or tripling of the conver-
sions was seen, and in general, the dinitrodiol 5 gave
slightly increased conversions compared to PHANOL (1),
as expected on the basis of hydrogen-bond strength (even
though it was observed to be only moderately soluble
even in neat solutions).
Finally, we examined the use of enantiopure PHANOL
(R)-111 as a potential asymmetric catalyst. The Diels–
Alder reaction with the greatest previously observed rate
acceleration was selected so as to minimize uncatalyzed
background reaction i.e., the reaction between cyclopen-
tadiene and crotonaldehyde. However, after analysis of
the resulting cycloadducts by diastereomeric imine for-
mation using a-methylbenzylamine, the diastereomeric
excesses, and hence the enantioselection in the Diels–
Alder reactions were found to be minimal (<5%).
In conclusion, we have shown that the PHANOLs 1 and 5
are capable of catalyzing Diels–Alder reactions between
a,b-unsaturated aldehydes and ketones and dienes with
modest (ca. 2-4 fold) to significant (ca. 30 fold) rate accel-
erations. A reasonable model to explain this reactivity in-
vokes double hydrogen bonding of the PHANOLs to the
two sp2 lone pairs of the carbonyl group (Figure 2).20 This
is the first example of an inherently chiral catalyst acting
in this manner and the potential for asymmetric catalysis
is clear. However, presumably because of the lack of ex-
pression of the planar chirality associated with the paracy-
clophane backbone around the olefin of the dienophile,
the enantioselectivities were found to be minimal. Further
applications of the PHANOL family for catalysis via dou-
ble hydrogen bonding will be reported in due course.
(10) Inspection of X-ray crystal structures of biphenylenediol
shows a phenolic O-O separation of 4.0 Å. Molecular
modelling of PHANOL 1 reveals an expected phenolic O-O
separation of ca. 4.1 Å.
(11) Braddock, D. C.; MacGilp, I. D.; Perry, B. G. J. Org. Chem.
2002, 67, 8679.
(12) Reich, H. J.; Cram, D. J. J. Am. Chem. Soc. 1969, 91, 3527.
(13) Procedures and Data for Compounds 3–5: 4,12-
Dibromo-7,15-dinitro[2.2]paracyclophane (3). A
heterogeneous mixture of nitronium tetrafluoroborate (2.7 g,
20 mmol) in sulfolane (30 mL) in a sealed flask was
immersed in an ultrasound bath until the mixture had
homogenized. The resultant solution was added dropwise to
a solution of 4,12-dibromo[2.2]paracyclophane 2 (2.5 g, 6.7
mmol) in CH2Cl2 (30 mL) at –78 °C under nitrogen. After 15
min the reaction mixture was allowed to warm to r.t., and
was heated to 50 °C for 1 h. The reaction was quenched with
H2O (20 mL), the organic layer was separated from the
aqueous layer, and the volatiles removed under reduced
pressure. The residue was added to the aqueous layer,
forming a precipitate, which was isolated by filtration. The
aqueous layer was extracted with Et2O (3 × 30 mL), and the
combined organic layers added to the precipitate. The
resultant solution was washed with H2O (2 × 50 mL) and
brine (1 × 50 mL), dried over MgSO4, and chromatographed
(1:1 CH2Cl2:petroleum ether) to yield 3 (2.36 g, 77%) as a
yellow solid: Mp 185–190 °C. Rf = 0.35 (1:1
O
H
X
O
X = H, NO2
R = H, Me
R
X
H
O
Figure 2 Proposed double hydrogen-bonding with PHANOLs 1
and 5
CH2Cl2:petroleum ether). IR (DRIFTS): 3100, 2950, 2850,
Synlett 2003, No. 8, 1121–1124 ISSN 1234-567-89 © Thieme Stuttgart · New York

1124
D. C. Braddock et al.
LETTER
1570, 1500, 1470, 1330 cm–1. 1H NMR (300 MHz, CDCl3):
d = 3.04–3.24 (4 H, m), 3.45–3.52 (2 H, m), 3.83–3.90 (2 H,
m), 7.37 (2 H, s), 7.38 (2 H, s). 13C NMR (75 MHz, CDCl3):
d = 32.0, 34.5, 127.9, 132.5, 136.5, 136.7, 141.6, 148.5. MS
(EI): m/z = 458 (M+), 456 (M+), 454 (M+), 229, 227. HRMS
(EI) calcd for C16H12O4N281,81Br2: 457.9123,
CD3SOCD3): d = 2.67 (4 H, m), 3.24 (2 H, m), 3.72 (2 H, m),
6.31 (2 H, s), 7.24 (2 H, s), 10.83 (2 H, s, broad). 13C NMR
(68 MHz, CD3SOCD3): d = 29.9, 33.7, 121.2, 126.7, 130.3,
140.5, 141.4, 162.6. MS (EI): m/z = 330, 314, 300, 282, 165,
149, 136, 120, 106, 79, 65. HRMS (EI): calcd for
C16H14O6N2: 330.0852. Found: 330.0845. Anal. Calcd for
C16H14N2O6: H, 4.27; C, 55.18; N, 8.48. Found: H, 3.99; C,
54.97; N, 8.36.
C16H12O4N279,81Br2: 455.9143, C16H12O4N279,79Br2:
453.9164. Found: 457.9132, 455.9147, 453.9176. 4,12-
Dimethoxy-7,15-dinitro[2.2]paracyclophane (4). DMF
(50 mL) was added to a flask charged with 4,12-dibromo-
7,15-dinitro[2.2]paracyclophane (3) (1.11 g, 2.4 mmol),
Cu(CH3CN)4BF4 (80 mg, 0.24 mmol) and sodium
methoxide (7.9 g, 146 mmol) under nitrogen, and the
heterogeneous mixture was stirred for 30 min. The mixture
was poured into an aqueous solution of HOAc (2 M, 100
mL), and extracted with EtOAc (2 × 50 mL). The combined
organic layers were washed with water (4 × 100 mL) and
brine (50 mL), dried over anhyd MgSO4 and
(14) Sato, T.; Torizuka, K.; Shimizu, M.; Kurihara, Y.; Yoda, N.
Bull. Chem. Soc. Jpn. 1979, 52, 2420.
(15) Both PHANOLs were fully soluble in EtOAc, DMSO and
DMF, where presumably they are able to form strong
hydrogen bonds to the solvent. Their lack of solubility in
non-hydrogen bonding solvents precluded the direct
examination (without solvent interference) of dienophile-
PHANOL interactions by 1H NMR.
(16) 1,1¢-Binaphthyl-2,2¢-diol.
(17) Ishihara, K.; Kurihara, H.; Matsumoto, M.; Yamamoto, H. J.
Am. Chem. Soc. 1998, 120, 6920.
chromatographed (2:1 CH2Cl2:petroleum ether) to yield 4
(576 mg, 66%) as a yellow solid: Mp 178–180 °C. Rf = 0.30
(2:1 CH2Cl2:petroleum ether). IR (DRIFTS): 3067, 2966,
2942, 2857, 1594, 1561, 1505, 1455, 1438 cm–1. 1H NMR
(300 MHz, CDCl3): d = 2.90 (4 H, m), 3.31 (2 H, m), 3.80 (6
H, s), 3.95 (2 H, m), 6.18 (2 H, s), 7.39 (2 H, s). 13C NMR
(75 MHz, CDCl3): d = 29.7, 34.2, 56.0, 116.0, 128.5, 129.5,
140.8, 142.9, 162.5. MS (EI): m/z = 358 (M+), 328, 179.
Anal. Calcd for C18H18N2O6: H, 5.06; C, 60.33; N, 7.82.
Found: H, 4.96; C, 60.25; N, 7.79. 4,12-Dihydroxy-7,15-
dinitro[2.2]paracyclophane (5). 4,12-Dimethoxy-7,15-
dinitro[2.2]paracyclophane (4) (500 mg, 1.4 mmol) was
added to an aqueous solution of hydrobromic acid (50 mL,
48% w/w) and HOAc (5 mL). The solution was heated to
135 °C for 16 h, allowed to cool, diluted with H2O (150 mL),
neutralized to pH = 7 with sat. aq NaHCO3 solution (ca. 200
mL) and extracted with EtOAc (3 × 100 mL). The combined
organic layers were washed with brine, dried over MgSO4
and evaporated to give 5 (371 mg, 80%) as a dark solid: Mp
280 °C (dec.). Rf = 0.22 (2:98 MeOH:CH2Cl2). IR
(18) Burnell, D. J.; Goodbrand, H. B.; Kaiser, S. M.; Valenta, Z.
Can. J. Chem. 1987, 65, 154.
(19) Caselli, A. S.; Collins, D. J.; Stone, G. M. Aust. J. Chem.
1982, 35, 799.
(20) The PHANOL system is clearly related to the Kelly
biphenylene diol8 (Figure 1) in so much as it employs two
phenolic groups held with the correct orientation to doubly
hydrogen bond to the two sp2 lone pairs of a carbonyl group.
It would be instructive to compare their relative reactivity. In
the Kelly system, using 40–50 mol% of catalyst the reactions
are typically run in CD2Cl2, with a 10-fold excess of diene at
55 °C, whereas for PHANOL (10 mol%) a 1:1 stoichiometry
of diene:dienophile was employed with no solvent at
ambient temperature: the conditions are not directly
comparable. However, for the two entries that were run at
ambient temperature in Kelly’s work (1. 10 equiv CpH, 1
equiv MVK, 40 mol% catalyst, 10 min, 90% conversion; 2.
10 equiv CpH, 1 equiv acrolein, 40 mol% catalyst, 30 min,
76%), comparable conversions were obtained using just 10
mol% PHANOL (Table 1, entries 2, 7).
(DRIFTS): 3600–3000 cm–1. 1H NMR (300 MHz,
Synlett 2003, No. 8, 1121–1124 ISSN 1234-567-89 © Thieme Stuttgart · New York