Kinetic resolution of amines with enantiopure 3-N,N-diacylaminoquinazolin-4(3H)-ones

View Article Online / Journal Homepage / Table of Contents for this issue

Kinetic resolution of amines with enantiopure 3-N,N-

diacylaminoquinazolin-4(3H)-ones

Abdullah G. Al-Sehemi, Robert S. Atkinson* and John Fawcett

1

Department of Chemistry, Leicester University, Leicester, UK LE1 7RH

Received (in Cambridge, UK) 4th July 2001, Accepted 16th November 2001

First published as an Advance Article on the web 18th December 2001

The title compounds (DAQs) are chiral when the two N-acyl groups are diﬀerent because of the absence of rotation

around the N–N bond (a chiral axis). Enantiopure DAQs have been obtained by incorporation of a chiral centre in

enantiopure form either into the substituent at the Q2-position or into one of the N-acyl groups, or into both,

followed by separation of diastereoisomers. This separation is unnecessary in one case because conversion of the

N-monoacylaminoquinazolinone (MAQ) into the DAQ is completely diastereoselective. Neither is separation of

diastereoisomers necessary with 3-[N,N-di-(S)-2-acetoxypropanoylamino]-2-diphenylmethylquinazolin-4(3H)-one

37a: this DAQ 37a has its N–N bond rendered a chiral axis by the bias in its imide moiety wholly in favour of one

exo/endo conformation.

The high chemoselectivity exhibited by N,N-diacetyl- or N,N-dibenzoylaminoquinazolinones in reaction with the

less hindered of two secondary amines (pyrrolidine in the presence of 1 eq. of piperidine) has a stereoselective

counterpart: reaction of the above enantiopure DAQs enantioselectively with racemic amines leading to kinetic

resolution. Using 1 eq. of DAQ and 2 eq. of amine, both the derivatised and unreacted amine enantiomers are

recovered with high enantiomeric excess (ee) (better than 90% ee in some cases). Some of the higher ees are found in

the recovered amides where non-chemoselective attack on both N-acyl groups of the DAQ has occurred: from the

opposite conﬁgurations of the amine component in the two amides and from the low enantiopurity of the recovered

unreacted amine, reaction of each of the N-acyl groups with complementary enantiomers of the amine is occurring

(parallel kinetic resolution).

Although higher ees are, in general, obtained using secondary amines, high ees are obtained in some cases using

1-phenylethylamine and, in particular, amino acid esters (valine and alanine).

The sense of enantioselectivity in the reactions of these DAQs with amines is controlled by the conﬁguration of

the N–N axis: replacing the Q group in an N-(S)-2-acetoxypropanoyl-N-acetyl-bearing DAQ by phthalimide, thus

eliminating the N–N chiral axis, drastically reduces the level of kinetic resolution.

Introduction

3-N,N-Diacylaminoquinazolinones (DAQs) e.g. 1 are highly

chemoselective acylating agents for primary amines in the

presence of secondary amines and for the less sterically

hindered of two secondary amines.¹

We have improved on our previously obtained level of chemo-

selectivity, as measured by the preference for attack on pyrrol-

idine in preference to piperidine, by the use of 3-N,N-di-

acetylaminoquinazolinone 2: reaction of the corresponding

3-N,N-dibenzoylaminoquinazolinone 3 at Ϫ10 ЊC with the

same mixture of amines was even more chemoselective

(Scheme 1).²

Of greater interest was whether this chemoselectivity had a

stereoselective counterpart: would an enantiopure DAQ react

selectively with one enantiomer of a racemic amine leading to

kinetic resolution of that amine?³In DAQs having non-

identical N-acyl groups, the N–N bond is a chiral axis, con-

ﬁgurationally stable at room temperature, with the Q ring and

the imide moiety contained in orthogonal planes. The presence

of an additional chiral centre as in (racemic) DAQ 4, allows the

separation of diastereoisomers and, from the rate of conversion

of one of the diastereoisomers into the other on heating in

toluene, a barrier to N–N bond rotation ∆G^#= 121 kJ mol^Ϫ1

was calculated.⁴

To prepare the enantiopure DAQs required to test the stereo-

Scheme 1

selectivity of their reactions with racemic amines, it was expedi-

ent to incorporate a chiral centre in enantiopure form into the

DAQ and to separate the two enantiopure diastereoisomers

obtained: the chiral centre could be located either in the Q2-

substituent or in one of the N-acyl groups.

DOI: 10.1039/b105917n

J. Chem. Soc., Perkin Trans. 1, 2002, 257–274

This journal is © The Royal Society of Chemistry 2002

257

View Article Online

Previously we had shown that kinetic resolution of 1-phenyl-

ethylamine using the separated enantiopure diastereoisomers

of DAQ 5 was feasible albeit with low enantioselectivity (55%

ee and 40% ee, respectively).⁴In these resolutions, the sense of

enantioselectivity was controlled by the conﬁguration of the N–

N bond since the two diastereoisomers of DAQ 5, diﬀering in

conﬁguration at this bond, reacted with diﬀerent enantiomers

of the amine. Subsequently, we found that acylations using

DAQs were more chemoselective when applied to two second-

ary amines (cf. Scheme 1)¹and so our expectation was that

greater enantioselectivity would be obtained in kinetic resolu-

tions of secondary amines using enantiopure DAQs.

The barrier to interconversion of DAQ¹s 8a and 8b was not

quantiﬁed but was clearly lower than that separating the two

diastereoisomers of DAQ 4 since interconversion between them

took place slowly in chloroform solution at room temperature

and was complete after heating at 60 ЊC for 2 h giving a 1 : 1

ratio of 8a and 8b.

By contrast, acetylation of N-benzoylaminoquinazolinone

MAQ¹7 with acetyl chloride–pyridine was completely diastereo-

selective giving a crystalline sample of DAQ¹9 (76%). The

X-ray crystal structure of DAQ¹9 (Fig. 1) was analogous

In this paper we report the preparation of diastereopure

and highly enantiopure DAQs 8a, 8b, 9, 31a, 31b and 37a and

kinetic resolution experiments using 2- and 3-methylpiperidine,

1-phenylethylamine and amino acid esters with the above

DAQs and with DAQs 24a–d and 35.

Results and discussion

3-(N-Acyl-N-benzoylamino)quinazolinones 8 and 9 (DAQ¹s 8

and 9)

3-Aminoquinazolinone 6 (Q¹NH₂) was prepared as described

previously from ()-valine.⁵DAQ¹s 8 and 9 were obtained by

successive N-acylation of Q¹NH₂6 with benzoyl chloride and

then with 2-methylpropanoyl chloride or acetyl chloride

respectively (Scheme 2).

Fig. 1 Molecular structure of 9 showing the atom label scheme.

Displacement parameters are shown at the 30% level. H atoms bonded

to chiral centres are shown with dashed bonds, all other H atoms are

omitted for clarity.

to that of DAQ¹8a. However the conformation around the

bond linking Q¹to its C-2 chiral centre was diﬀerent in the two

structures.

Not only was DAQ¹9 the only diastereoisomer formed in the

acetylation of MAQ¹7 but it also appeared to be the thermo-

dynamically preferred since heating it at 130 ЊC for ∼1 minute

produced no additional signals from another diastereoisomer

in its NMR spectrum.

The barrier to rotation around the N–N bond in MAQ¹7 is,

as expected, lower than that in DAQ¹8a and does not allow

separation of diastereoisomers at room temperature.⁶However,

the NMR spectrum of MAQ¹7 at room temperature shows the

presence of two doublets at δ 5.08 and 6.03, both J = 7 Hz (ratio

7 : 1) presumably from PrⁱCHOSi in diastereoisomers arising

from N–N bond rotation at a rate which is slow on the NMR

timescale. When a crystalline sample of this MAQ¹7 was

dissolved in CDCl₃at Ϫ50 ЊC and an ¹H NMR spectrum

measured at this temperature,‡ this PrⁱCHOSi signal appeared

as two doublets (δ 4.68 and 4.43 ppm) now ratio 1 : 1.5 and,

signiﬁcantly, with diﬀerent J values (2 and 7 Hz respectively).

The disparate concentrations and J values for the species

present in the spectra run at Ϫ50 ЊC and 27 ЊC and the complex

changes in the spectra run at intermediate temperatures suggest

the presence of a second temperature-dependent process

associated with restricted rotation within the Q¹-2 substituent.

An X-ray crystal structure (Fig. 2) of MAQ 11 (Scheme 3),

prepared from the corresponding 3-aminoquinazolinone 10,

Scheme 2 Reagents: i, PhCOCl, pyr. CH₂Cl₂; ii, PrⁱCOCl, pyr. CH₂Cl₂;

iii, CH₃COCl, pyr. CH₂Cl₂.

Separation of DAQ¹8 into oily 8a and crystalline 8b

diastereoisomers was achieved by kieselgel chromatography. An

X-ray crystal structure previously determined for 8b^3ballowed

assignment of conﬁguration to the N–N axis and also revealed

that the conformation of the imide moiety was the less common

exo/exo† (exo carbonyl cis to Q¹). A striking feature of the

crystal structure is the non-planarity of the imide with torsion

angles of 24.4Њ and 9.7Њ for the N–N–C(᎐O) and N–C᎐O planes

᎐

of 2-methylpropanoyl and benzoyl respectively and with the

imide nitrogen pyramidalised (Σθ = 352Њ for the sum of bond

angles around nitrogen). With the benzoyl group also, the

planes containing the carbonyl group and the benzene ring are

inclined at an angle of 50Њ.

‡ It is likely that signals from the minor N–N bond rotamer are absent

from this spectrum: enrichment of the major N-rotamer (second-order

asymmetric transformation) and its non-interconversion with the minor

N-rotamer at Ϫ50 ЊC has been demonstrated with another MAQ.²⁶

† DAQ¹s 8b and 9 are the only ones with exo/exo conformations for the

imide moieties in their crystal structures compared with 12 structures

having exo/endo imide conformations that we have obtained.

258

J. Chem. Soc., Perkin Trans. 1, 2002, 257–274

View Article Online

Scheme 4 Reagents and conditions: i, 2-Methylpiperidine (2 eq.),

Ϫ20 ЊC, 2 h then 5 ЊC, 12 h; ii, 2-methylpiperidine (2 eq.), Ϫ20 ЊC, 2 h

then 5 ЊC, 30 h; iii, (S)-CH₃CH(OAc)COCl, pyr. CH₂Cl₂.

Reaction of the crystalline DAQ¹8b with 2-methylpiperidine

was considerably slower than for its diastereoisomer above

(Scheme 4). Signiﬁcantly, it was the other amine enantiomer

which reacted preferentially giving (S)-N-benzoylamide 12

(81% ee). Since DAQ¹s 8a and 8b have the same conﬁguration

for their chiral centres but opposite conﬁgurations for their

chiral N–N axes, it is the latter which control the sense of

enantioselectivity.

Fig. 2 Molecular structure of 11. Details as for Fig. 1.

Although the oily DAQ¹8a gives high enantioselectivity in

its reaction with 2-methylpiperidine, its separation from DAQ¹

8b requires careful chromatography. By contrast, DAQ¹

9

is formed completely diastereoselectively in good yield (see

earlier). However, its reactions, unlike those of DAQ¹8a,

are not chemoselective: reaction with 2-methylpiperidine

(Scheme 5) gave a mixture of N-benzoyl- and N-acetyl- amides

Scheme 3

shows the benzoyl group with its carbonyl exo as expected.

Even in MAQ 11 there is a substantial deviation of the amide

group from planarity (QN–CO torsion angle 14Њ): the benzene

ring, however, is now more nearly coplanar with the carbonyl

group (torsion angle, 6.6Њ).

Reactions of DAQ¹s 8a, 8b and 9 with racemic amines. Reac-

tion of oily DAQ¹8a (1 eq.) with 2-methylpiperidine (2 eq.) in

dichloromethane at Ϫ20 ЊC for 2 h and then at 5 ЊC for 12 h

takes place highly enantioselectively. Unreacted 2-methyl-

piperidine was extracted with aqueous hydrochloric acid and

the freed amine converted into N-benzoylamide 12 [α]_D= ϩ30

(c 0.64, CHCl₃). The benzoylamide 12 formed in the reaction

mixture was separated from the only other product MAQ¹13

(Scheme 4) by chromatography and had [α]_D= Ϫ31.4 (c 0.8,

CHCl₃).

The absolute conﬁgurations and ees given in Scheme 4 are

based on the speciﬁc rotation of a sample of (S)-N-benzoyl-2-

methylpiperidine [α]_D= ϩ32.9 (c 0.8, CHCl₃), prepared from

(S)-2-methylpiperidine, [α]_D= 8.9 (c 2, EtOH) itself obtained

from the racemic amine by resolution using (R)-mandelic

acid.⁷The enantiopurity of this (S)-2-methylpiperidine was

independently conﬁrmed by its derivatisation with (S)-2-

acetoxypropanoyl chloride and by comparison of the NMR

spectrum of the product 14a with the mixture of diastereo-

isomers formed from racemic 2-methylpiperidine: at 400 MHz

and 50 ЊC, the OCOCH₃signals from the two diastereoisomers

in this mixture were completely separated.

Scheme 5

12 and 15 together with their complementary MAQ¹s 16 and 7.

From this ratio of MAQ¹s 16 and 7, the ratio of amides 12 : 15

formed in the crude reaction mixture was 2.5 : 1. This ratio was

conﬁrmed from the NMR spectrum of the crude reaction

product at 400 MHz and Ϫ40 ЊC: interpretation of the

spectrum at this low temperature is facilitated by separation of

the broadened signals for the amides 12 and 15 into two sets

of signals from the component N–CO bond rotamers.

The N-benzoylamide 12 (41%) isolated by chromatography

was found to be the (S)-enantiomer of 91% ee by comparison

with an authentic sample. Although the N-acetylamide 15

was not isolated, the unreacted 2-methylpiperidine recovered as

its hydrochloride salt was found to be of low enantiopurity

(11% ee) by comparison of its speciﬁc rotation with that of an

enantiopure sample. It appears that the two imide carbonyl

groups are reacting with complementary enantiomers of the

racemic amine i.e. parallel kinetic resolution is occurring⁸

with the two acyl groups of the imide behaving pseudoenantio-

topically towards the reacting amine.

A quantitative measure of the enantioselectivity in reaction

of DAQ¹8a with 2-methylpiperidine was obtained from its

rate of reaction with each of the separated enantiomers of

this amine, k₁(R) and k₂(S). From the rotation values for the

separated enantiomers, the faster reacting (R) was of 95% ee

and the ratio k₁(R) : k₂(S) was at least 27 : 1.

J. Chem. Soc., Perkin Trans. 1, 2002, 257–274

259

View Article Online

Scheme 6 Reagents and conditions: i, 3-Methylpiperidine (2 eq.), Ϫ20

ЊC, 2 h then 5 ЊC, 30 h; ii, TBAF, THF.

Scheme 8

Reaction of DAQ¹9 with racemic 3-methylpiperidine was

also studied (Scheme 6). As with 2-methylpiperidine, attack on

both imide carbonyl groups occurs. Chromatography separated

MAQ¹7 but amide 17 and MAQ¹16 were co-eluted. After

desilylation of MAQ¹16 in this mixture by treating with

tetrabutylammonium ﬂuoride (TBAF) in tetrahydrofuran,

separation by ﬂash chromatography gave N-benzoyl-3-methyl-

piperidine 17 (32%) whose (S)-conﬁguration and enantiopurity

(85% ee) were based on a sample prepared by benzoylation of

(R)-3-methylpiperidine.

(R)-3-Methylpiperidine was obtained from the racemate by

fractional crystallisation of the tartrate salts.⁹From the ratio

of MAQ¹s 16 and 7 in the crude reaction mixture, the ratio

of N-benzoyl- to (unrecovered) N-acetyl-2-methylpiperidine

17 : 18 was 3 : 1.

9 with 1-phenylethylamine

(Scheme 7) both amides 20 and 21, resulting from attack on

Scheme 9

All of these four diastereoisomeric DAQ¹s react completely

chemoselectively with 1-phenylethylamine at the 2-acetoxy-

propanoyl group except DAQ¹24b (Scheme 10).

In the reaction of DAQ¹

The enantioselectivity arising from attack of the amine on

the 2-acetoxypropanoyl group was quantiﬁable directly by

NMR spectroscopy after chromatography, being equivalent

to the diastereoisomer excess (de) of the amide product and

measurable by comparison with spectra of authentic samples

of these diastereoisomers; control experiments conﬁrmed that

no change in de occurred on chromatography.

The relative rates of reaction of DAQ¹s 24a–d (and other

DAQs in this paper) with amines have been used to probe into

the mechanism and to rationalise the stereochemistry of these

acylations and will be discussed elsewhere. For the present it

is noteworthy that, of these four DAQ¹s, 24b has the largest

torsion angle between N–N–C(᎐O) and N–C᎐O bonds for

᎐

the 2-methylpropanoyl (25.1Њ) and also the largest disparity in

Scheme 7

N–C(᎐O) bond lengths for 2-methylpropanoyl and 2-acetoxy-

᎐

propanoyl groups (1.525 Å vs. 1.424 Å) in the crystal structure.⁵

In all cases in Scheme 10, the sense of enantioselectivity is

controlled by the conﬁguration of the N–N axis. Thus DAQ¹s

24b and 24d which are homochiral at their N–N axes reacted

with the (S)-enantiomer of 1-phenylethylamine and likewise

DAQ¹s 24a and 24c reacted with the (R)-enantiomer of this

amine.

2-Methylpiperidine and 2-propylpiperidine (coniine) reacted

with DAQ¹24c, the fastest reacting diastereoisomer of DAQ¹s

24a–d, highly chemo- and enantioselectively (Scheme 11).

(R)-(Ϫ)-2-Propylpiperidine hydrochloride has [α]_D= Ϫ7.3

(c 0.33, ethanol).¹⁰Since the recovered 2-propylpiperidine

enantiomer in Scheme 11 has [α]_D= Ϫ6.7 (c 0.42, ethanol) (ee

89%), the major reacting enantiomer is (S) and amide 29a has

the (S,S)-conﬁguration i.e. 2-propyl- and 2-methylpiperidine

react with DAQ¹24c in the same enantiosense.

imide benzoyl and acetyl groups respectively, were recovered

and shown to have resulted from reactions of complementary

enantiomers of the amine. As in Scheme 6, the recovered

unreacted amine was of low enantiopurity (16% ee).

In contrast to the non-chemoselectivity in reactions of DAQ¹

9 in Schemes 5–7, racemic valine methyl ester reacts highly

chemoselectively with the benzoyl group and gave amide 22

of high enantiopurity (94% ee) by comparison with the rotation

of an authentic sample (Scheme 8).

From the ratio of MAQ¹s 16 and 7 isolated by chroma-

tography, the chemoselectivity is 21 : 1 favouring attack on the

benzoyl group.

Reaction of DAQ¹s 24a–d with amines. In a previous paper⁵

we described the preparation of DAQ¹s 24a–d by reaction of 3-

aminoquinazolinone 6 (Q¹NH₂) with (S)-2-acetoxypropanoyl

chloride followed by 2-methylpropanoyl chloride (Scheme 9).

Considerable epimerisation at the (S)-2-acetoxypropanoyl

centre occurs in the second N-acylation step, resulting in the

formation of DAQ¹s 24a and 24d, (R)-conﬁgured at this centre

as well as 24b and 24c.

3-[N-2-(2S)-Acetoxypropanoyl-N-acetylamino]-2-diphenyl-

methylquinazolinones 31a and 31b (DAQ²s 31a and 31b)

Further acylation of MAQ²30 with (S)-2-acetoxypropanoyl

chloride gave a 1 : 1 mixture of diastereoisomers 31a and

260

J. Chem. Soc., Perkin Trans. 1, 2002, 257–274

View Article Online

Scheme 10

to Q²) for their imides but, as expected, opposite conﬁgurations

for their N–N chiral axes.

Because there is no additional spectator chiral centre in

DAQ²s 31a and 31b as there is in DAQ¹s 24a–d, the extent of

epimerisation at the 2-acetoxypropanoyl chiral centre occurring

in their preparation, if any (cf. Scheme 9), is not apparent from

the NMR spectra of the products; any such epimerisation will

result in loss of their enantiopurity. In fact, the space group of

DAQ²31a in the crystal structure (Fig. 3a) was determined to

be P21/n which indicates that the molecule in this crystal is

racemic. Although some epimerisation in the formation of

DAQ 31a (and probably DAQ 31b) does occur, therefore, the

extent must be very small and is not revealed in an enantio-

purity assay (see below).

Scheme 11 Reagents and conditions: i, 2-Methyl- or 2-propyl-

piperidine, Ϫ10 ЊC, 2 h then 5 ЊC 24 h; ii, NaOH–H₂O; iii, (S)-

CH₃CH(OAc)COCl, pyr.

By analogy with the conformational equilibria present in

DAQ¹s 24a–d, DAQ²31a would be expected to have the same

single exo/endo conformation for its imide in solution present in

the crystal structure Fig. 3a whereas DAQ 31b would be present

as an equilibrium mixture of exo/endo and endo/exo forms.

Thus, in the alternative endo–exo form of DAQ²31a, 31aЈ, there

would be an unfavourable interaction between the CH(OAc)-

CH₃and Ph₂CH groups but this would be absent in the case of

the corresponding DAQ²31bЈ.

Some estimate of the likely ratio of exo/endo–endo/exo imide

conformations present in 31b can be arrived at from com-

parison with DAQ¹24c since both have the same conﬁgurations

at their chiral axes and CH(OAc)CH₃chiral centres albeit

with diﬀerent Q2-substituents and second N-acyl groups. At

equilibrium, the exo/endo–endo/exo ratio for DAQ¹24c favours

the former by 2.7 : 1 at Ϫ50 ЊC. Since the acetyl methyl group

of DAQ²31b is clearly smaller than the isobutanoyl group of

DAQ¹24c and since the smaller this alkyl group the more likely

it is to be located in the exo-position (i.e. with its carbonyl

endo), the equilibrium position for the imide in DAQ²31b

would be expected to favour the exo/endo conformation by a

ratio considerably greater than 2.7 : 1.

Scheme 12

The NMR spectra of DAQ²s 31a and 31b were broadly in line

with the expectations above: the spectrum for DAQ²31a shows

sharp signals for e.g. Ph₂CH and CH(OAc)CH₃protons and

was unchanged when run at low temperature (Ϫ50 ЊC). For

DAQ²31b, both the corresponding proton signals were slightly

broadened and, when the spectrum was run at 0 ЊC, Ϫ10 ЊC and

31b (Scheme 12) which were easily separable by ﬂash chroma-

tography as a result of their widely diﬀering R_fvalues (∆R_f

ca. 0.2 using light petroleum–ethyl acetate 1 : 1). Both crystal

structures for these DAQ²s (Fig. 3a,b) have the same exo/endo

conformation (exo has 2-acetoxypropanoyl carbonyl oxygen cis

J. Chem. Soc., Perkin Trans. 1, 2002, 257–274

261

View Article Online

Scheme 13

amide 33 was a 1 : 1 mixture of diastereoisomers, the N-acetyl-

amide 34 was of high enantiopurity (97% ee) based on its

speciﬁc rotation (Scheme 14).

Scheme 14

The reaction in Scheme 14 was repeated using excess pure

()-alanine ethyl ester and the amide isolated was shown to be

diastereopure 33a by NMR spectroscopic comparison with an

authentic sample. Thus DAQ²31a is of high enantiopurity and

little epimerisation at the 2-acetoxypropanoyl centre takes place

in its formation from MAQ²30 (cf. above).

Diastereoisomeric DAQ²31b exhibits very diﬀerent chemo-

and stereo-selective behaviour to that of DAQ²31a with 1-

phenylethylamine and alanine ethyl ester (Scheme 15). Thus the

reaction with 1-phenylethylamine is now non-chemoselective

but amides 21 and 25 are formed with 85% ee and 80% de

respectively. Alanine ethyl ester reacted almost completely

chemoselectively and the N-2-acetoxypropanoyl amide 33 was

obtained with 94% de.

As yet, an explanation for these extraordinary diﬀerences in

stereochemistry and chemo- selectivity is not available: follow-

ing the reactions by NMR spectroscopy showed that DAQ²31b

reacted at least 500 times faster with 1-phenylethylamine than

DAQ²31a.

Fig. 3 a and b. The molecular structures of 31a and 31b respectively.

Details as for Fig. 1.

Ϫ40 ЊC, showed further broadening then sharpening as the

temperature was lowered. At the lower temperature, however,

separated signals assignable to the expected minor endo/exo

conformation were not visible. That the major conformation in

solution was the exo/endo 31b was supported by the chemical

shift of the CH(OAc)CH₃proton at δ 6.07, consistent with its

deshielding by the endo acetyl carbonyl oxygen.⁵

3-[N,N-Bis((S)-acetoxypropanoyl)amino]-2-alkylquinazolinones

35, 36 and 37 (DAQs 35, 36 and 37)

The preparation of DAQs 35 and 36 from reaction of the corre-

sponding 3-aminoquinazolinones with excess (S)-2-acetoxy-

propanoyl chloride was also described previously (Scheme 16a).

Although these DAQs had signiﬁcant optical rotations, some

epimerisation at both (S)-2-acetoxypropanoyl chiral centres,

leading to partial racemisation, could not be ruled out particu-

larly since some meso-isomer was isolated from the reaction

mixture in the case of DAQ 35.

Reaction of DAQ²s 31a and 31b with amines. Reaction of

DAQ²31a with 1-phenylethylamine was highly chemo- and

stereo-selective (Scheme 13).

Reaction of DAQ²31a with alanine ethyl ester, however,

showed little chemoselectivity and the two amides were

obtained with very diﬀerent stereoselectivities. Thus, using four

equivalents of the amine, whereas the N-2-acetoxypropanoyl

In DAQ 35 also, there was a conformational bias within

the imide moiety with the expected exo/endo

endo/exo

equilibrium in Scheme 16a wholly on the 35 exo/endo side.

262

J. Chem. Soc., Perkin Trans. 1, 2002, 257–274

View Article Online

Scheme 15 Reagents and conditions: i, 1-Phenylethylamine Ϫ10 ЊC, 30 min, CH₂Cl₂; ii, alanine ethyl ester, Ϫ10 ЊC, 6 h, CH₂Cl₂.

Scheme 16 Reagents: i, (S)-CH₃CH(OAc)COCl, pyr. CH₂Cl₂; ii, rac-

CH₃CH(OAc)COCl, pyr. CH₂Cl₂.

Fig. 4 The molecular structure of 37a. Details as for Fig. 1.

It was proposed that as for DAQ’s a–d this bias arose from a

high conformational preference within the 2-acetoxypropanoyl

groups which led to the steric interaction shown between the

methyl group in the 2-acetoxypropanoyl and that on the Q2-

position thus destabilising the 35 endo/exo conformation. This

conformational bias in DAQ 35 is important because the N–N

bond is now clearly a chiral axis whose presence is required for

high levels of enantioselectivity in reaction with racemic

amines§ (see below).

In the X-ray crystal structure of DAQ²37a (Fig. 4) the imide

is present in the usual exo/endo conformation but the torsion

angle ꢀ between the C᎐O and C–OAc bonds in the exo-oriented

᎐

COCH(OAc)Me group (130.9Њ) is very diﬀerent not only from

the corresponding ꢀ in DAQ 35 (20.8Њ) but also from all values

for

ꢀ in other DAQ crystal structures containing this

COCH(OAc)CH₃group that we have determined.¶

DAQ²37a was prepared in good yield from the 3-amino-

quinazolinone 38 by N,N-diacylation with (S)-2-acetoxy-

propanoyl chloride via the isolable MAQ²32 (Scheme 16b).

Conversion of MAQ²32 into DAQ²37a was unaccountably

faster than the analogous reaction forming DAQ²35 and,

importantly, no meso isomer 37b was present in the crude

reaction mixture. An authentic sample of meso DAQ²37b was

obtained by reacting MAQ²32 with rac-2-acetoxypropanoyl

chloride; the ratio of 37a : meso 37b formed in this second

N-acylation was 10 : 1 i.e. this reaction was highly diastereo-

selective. The NMR spectrum of meso-DAQ²37b showed one

broad signal for the CHOAc proton at δ 5.45 which separated at

Ϫ44 ЊC into two broadened signals at δ 4.9 and 5.8 (ratio 1 : 1):

this meso-isomer 37b would be expected to be present as a 1 : 1

mixture of the exo/endo and endo/exo conformers as is the

analogous meso-DAQ 35.⁵

The ¹H NMR spectrum of DAQ²37a also diﬀered from

those of DAQs 35 and 36 in showing broadening of the two

CHOAc proton signals at δ 4.88 and 5.80 ppm which sharpened

at Ϫ50 ЊC (δ 4.54 and 6.10 ppm); small additional signals

also appeared at δ 4.40, 5.07 and 6.07 ppm (∼20% of major

signals) (the spectra of DAQs 35 and 36 were unchanged when

run at Ϫ50 ЊC). Although the process giving rise to this signal

broadening in DAQ²37a is not known it may arise from the

abnormal ꢀ value in the crystal structure above.|| In any event,

it does not appear that DAQ²37a is undergoing fast exo/endo–

endo/exo interconversion at room temperature.

¶ For the endo-oriented COCH(OAc)CH₃group in DAQ²37a ꢀ is in the

more normal range (37.6Њ).

|| Possibly there is an equilibrium in solution between the conform-

ation having the normal value of φ as in DAQ 37a and that in the

crystal structure.

§ If attack by the amine on both exo carbonyl groups of either exo/endo

and endo/exo or exo/exo conformations of DAQ 35 were to occur, little

enantioselectivity would be expected if the eﬀect of the chiral centres

on the latter was small as is believed to be the case.

J. Chem. Soc., Perkin Trans. 1, 2002, 257–274

263

View Article Online

Scheme 17 Reagents and conditions: i, 35, PhCH(CH₃)NH₂, CH₂Cl₂, Ϫ20 ЊC, 2 h the 0 ЊC, 10 h; ii, 37a PhCH(CH₃)NH₂(2 eq.), CH₂Cl₂, Ϫ20 ЊC,

2 h; iii, 37a, PhCH(CH₃)NH₂, (5 eq.), CH₂Cl₂, Ϫ20 ЊC, 2 h.

Reactions of DAQs 35 and 37a with amines. The reaction

products of DAQs 35 and 37a with 1-phenylethylamine are

given in Scheme 17.

At least part of the inferior diastereopurity of the product of

reaction with DAQ 35 can be ascribed to its enantioimpurity.

Thus its reaction with excess (S)-1-phenylethylamine under the

conditions in Scheme 17 gave an 8 : 1 ratio of diastereoisomers

of amide 25 that must have resulted from partial epimerisation

at both chiral centres in preparation of DAQ 35. By contrast,

reaction of DAQ 37a with excess (R)-1-phenylethylamine

using the conditions in Scheme 17 gave only 25c the (R,S)-

diastereoisomer of amide 25 conﬁrming that DAQ 37a is

enantiopure.

Although the level of enantioselectivity in reaction of pri-

mary amines (1-phenylethylamine) with DAQ²37a is modest

(even with 5 eq. of amine), with secondary amines 2-methyl-

piperidine and 2-propylpiperidine (coniine) the enantioselec-

Scheme 19 Reagents and conditions: i, CH₃COCl, pyr. CH₂Cl₂, RT,

24 h; ii, (S)-CH₃CH(OAc)COCl, pyr. CH₂Cl₂; iii, PhCH(Me)NH₂,

CH₂Cl₂, 6 h, Ϫ10 ЊC.

tivity is excellent (Scheme 18). Assignment of conﬁguration to

the major product 29a was made by NMR spectroscopic com-

parison with the sample prepared previously (Scheme 11).

Repetition of this experiment but using excess pure (S)-

phenylethylamine gave only amide 25d conﬁrming that DAP 41

was enantiopure.

Conclusions

All of the DAQs examined in this work react enantioselectively

with the limited range of racemic amines used. Even under

the conditions of stoichiometry (1 eq. DAQ–2 eq. amine)

enantioselectivities sometimes >90% are obtained in both the

derivatised amine enantiomer (amide) and in the recovered

enantiomer. Where reaction of the amine with the DAQ is non-

chemoselective and amides derived from both the DAQ imide

carbonyl groups are formed (and must be separated), each of

these carbonyl groups reacts preferentially with complementary

enantiomers of the amine leading to enhanced enantiopurity

in the isolated amides (parallel kinetic resolution). By this

means even 3-methylpiperidine was recovered with 85% ee as

the N-benzoyl derivative by reaction with DAQ¹9.

The sense of enantioselectivity in reactions of all these DAQs

with amines is determined by the conﬁguration of the N–N

bond: the very low levels of enantioselectivity obtained from

reaction of the N,N-diacylphthalimide 41 with 1-phenyl-

ethylamine conﬁrms that the conﬁguration of the chiral centre

is not important for the high levels of kinetic resolution

achieved.

Scheme 18 Reagents and conditions: i, 2-Methylpiperidine, CH₂Cl₂,

Ϫ20 ЊC, 3 h then 0 ЊC, 12 h; ii, 2-propylpiperidine (4 eq.), CH₂Cl₂, 5 ЊC,

16 h.

To assess the contribution of the 2-acetoxypropanoyl chiral

centres to the stereoselectivities in reactions of DAQs contain-

ing this group, we prepared N-[(S)-2-acetoxypropanoyl]-N-

acetylaminophthalimide (DAP) 41 and reacted it with 1-phenyl-

ethylamine (Scheme 19). At Ϫ10 ЊC, reaction took place on both

acetyl and 2-acetoxypropanoyl groups (ratio 1.2 : 1 respectively)

but the enantioselectivity/diastereoselectivity in each case was

minimal.

The requirement for separation of the enantiopure diastereo-

isomeric DAQs used in these kinetic resolutions is obviated in

the case of DAQs 9 and 37a because their formation is com-

pletely diastereoselective.

264

J. Chem. Soc., Perkin Trans. 1, 2002, 257–274

View Article Online

1692s and 1609s; δ_H(mixture of N–N bond rotamers), major

Experimental

rotamer; Ϫ0.01 and 0.14 (6H, 2 × s, CH₃SiCH₃), 0.9 and 1.02

(6H, 2 × d, J 6.6, CH₃CHCH₃), 0.96 [9H, s, (CH₃)₃CSi],

2.13 [1H, (7 peaks), CH₃CHCH₃], 4.49 (1H, d, J 7.0, CHOSi),

7.4–7.61 [4H, m, 3 × CH(Ph) and 6-H(Q)], 7.79 [2H, m, 7 and

8-H(Q)], 8.0 [2H, m, 2 × CH(Ph)], 8.2 [1H, d, J 8.2, 5-H(Q)]

and 9.35 (1H, s, NH); δ_Hminor rotamer (observable signals),

2.42 [1H, (7 peaks), CH₃CHCH₃], 4.65 (1H, d, J 7.0, CHOSi)

and 8.85 (1H, s, NH); from comparison of the signal at δ 4.49

and δ 4.65 the ratio of N–N bond rotamers was 7 : 1; δ_C(100.6

MHz at 50 ЊC) Ϫ4.7 and Ϫ4.4 (CH₃SiCH₃), 17.9 [C(CH₃)₃],

18.2 and 19.7 (CH₃CHCH₃), 25.8 [(CH₃)₃C], 33.1 [CH(CH₃)₂],

78.6 [br (CHOSi)], 120.8 [CCO(Q)], 126.9, 127.5, 127.8, 128.1,

129.8, 132.9 and 134.7 [5 × CH(Ph) and 4 × CH(Q)], 132.5

¹H and ¹³C nuclear magnetic resonance (NMR) spectra were

recorded on a Bruker ARX-250 spectrometer at 250 MHz and

63 MHz respectively at room temperature in deuterochloroform

(CDCl₃) unless stated otherwise. ¹H NMR spectra at 400 MHz

and ¹³C at 100.6 MHz were recorded on a Bruker DRX

spectrometer at room temperature in deuterochloroform unless

otherwise indicated. Infra-red (IR) spectra of crystalline

compounds were determined using Nujol mulls and of liquids

either in dichloromethane or chloroform solutions, or neat, on

a Perkin–Elmer 298 spectrophotometer. Melting points (mps)

were determined with a Koﬂer hot stage and are uncorrected.

Mass spectra were determined using a Kratos Concept mass

spectrometer using electron impact (EI), chemical ionisation

(CI) or fast atom bombardment (FAB) and high resolution

masses (accurate masses) were obtained by peak-matching

using perﬂuorokerosene; except for the molecular ion M^ϩor

MH^ϩ, only peaks ≥20% of the base peak are given. Some mass

spectra were also determined using a Micromass Quattro lc

(MQlc) spectrometer with ionisation by Electrospray and

operation via “open Access” software with an autosampler,

Elemental analysis was carried out by CHN Analysis, Wigston,

Leicester. Optical rotations were determined on a Perkin–Elmer

[C(Ph)], 146.6 [C–C᎐N(Q)], 160.7 [CN(Q)], 165.9 [CO(Q)] and

᎐

169.1 (PhCO); m/z (%) 452 (MH^ϩ, 100), 394 (71), 275 (19),

and 187 (13).

Low temperature NMR studies were carried out on a sample

of 7 crystallised from light petroleum and dissolved in CDCl₃at

Ϫ50 ЊC (see text).

3-Benzoylamino-2-[(S)-1-tert-butyldimethylsilyloxyethyl]-

quinazolin-4(3H)-one 11. The general procedure I above was

followed using 3-aminoquinazolinone 10¹¹(3 g, 9.4 mmol),

pyridine (1.1 cm³, 13.9 mmol), dichloromethane (3 cm³) and

benzoyl chloride (1.6 g, 11.3 mmol). After work-up the yellow

oil obtained was triturated with ethyl acetate–light petroleum

and the solid obtained gave the title 3-benzoylaminoquin-

azolinone 11 as colourless crystals (3.3 g, 82%), mp 181–182 ЊC

(from light petroleum–ethyl acetate) (Found: C, 65.0; H, 6.9; N,

9.8. C₂₃H₂₉N₃O₃Si requires C, 65.2; H, 6.9; N, 9.9%) (Found:

MH^ϩ424.2056. C₂₃H₂₉N₃O₃Si, requires MH^ϩ424.2056);

ν_max/cm^Ϫ13250w, br, 1690s and 1610s; δ_H(mixture of N–N bond

rotamers), major rotamer; Ϫ0.01 and 0.1 (6H, 2 × s, CH₃Si-

CH₃), 0.8 (9H, s, (CH₃)₃C), 1.44 (3H, d, J 6.6, CH₃CHOSi), 4.91

(1H, q, J 6.6, CHOSi), 7.3–7.5 [4H, m, 6-H(Q) and 3 ×

CH(Ph)], 7.51–7.7 [2H, structured m, 7- and 8-H(Q)], 7.85 [2H,

structured m, 2 × CH(Ph)], 8.12 [1H, d, J 7.9, 5-H(Q)] and 9.4

(1H, s, NH); minor rotamer (observable signals), Ϫ0.09 and

Ϫ0.01 (6H, 2 × s, CH₃SiCH₃), 0.79 (9H, s, (CH₃)₃C), 1.2 (3H, d,

J 6.6, CH₃CHOSi), 5.0 (1H, q, J 6.6, CHOSi) and 8.9 (1H, s,

NH); from comparison of the intensities of signals at δ 4.91

and 5.0 the ratio of N–N bond rotamers is 2 : 1; δ_CϪ4.8 and

Ϫ4.3 (CH₃SiCH₃), 18.7 [(CH₃)₃C], 22.5 (CH₃CHOSi), 26.2

[(CH₃)₃C], 70.8 (CHOSi), 121.4 [CCO(Q)], 127.5, 128.2, 128.5,

129.2, 131.1, 133.1 and 135.3 [5 × CH(Ph) and 4 × CH(Q)], 31.2

341 polarimeter at 589 nm and are given in 10^Ϫ1deg cm²g^Ϫ1

.

Flash chromatography was carried out using silica gel C60 (35–

70) (supplied by Merck & Co.) and kieselgel chromatography

using kieselgel 60, 230–400 mesh. TLC was conducted on

aluminium plates pre-coated with a 0.2 mm layer of silica,

manufactured by Merck & Co. Puriﬁcation by Chromatotron

(Harrison Research California) was performed using model

7924T with circular and kieselgel 60 (PF₂₅₄) and also kieselgel

60 (GF₂₅₄) silica plates supplied by Merck & Co. Light

petroleum refers to the fraction (bp 60–80 ЊC).

Pyridine, 2-methylpiperidine and 3-methylpiperidine were

puriﬁed by distillation from calcium hydride. Ether refers to

diethyl ether and was sodium dried prior to use. Tetrahydro-

furan (THF) was distilled from sodium and benzophenone

immediately prior to use. Routine drying of organic solutions

was carried out using magnesium sulfate unless otherwise indi-

cated. Solvent removal under reduced pressure means using a

Buchi rotary evaporator and a water pump (∼12 mmHg) unless

otherwise indicated. n-Butyllithium (1.6 M) was used as

received from Aldrich Chemical Co. Lead tetraacetate (LTA)

(damp with acetic acid) was dried prior to use under vacuum

using an oil pump (∼1 mmHg) for 15 min. All reaction products

were dried using an oil pump (∼1 mmHg) prior to spectroscopic

analysis. Yields are isolated ones unless otherwise stated.

[C(Ph)], 147.4 [CN᎐C(Q)], 161.4 [C᎐N(Q)], 166.3 and 167.7

᎐

[CO(Q) and PhCO]; m/z (%) (FAB) 424 (MH^ϩ, 100), 366 (51),

292 (22) and 145 (20).

General procedure I for the mono-N-acylation of 3-amino-

quinazolinones

A crystal of compound 11 suitable for X-ray structure

determination (Fig. 2) was obtained from methanol.

To the 3-aminoquinazolinone dissolved in dry dichloromethane

(3 cm³g^Ϫ1) containing dry pyridine (1.5 mol eq.) was added the

acid chloride (1.5 mol eq.) dropwise with stirring. After stirring

for 12 h at room temperature, more dichloromethane was added

(3 cm³g^Ϫ1) and the solution washed with saturated aqueous

sodium hydrogen carbonate, then water, dried and the solvent

removed under reduced pressure. The product was puriﬁed by

crystallisation or chromatography as indicated.

3-(2-Methylpropanoylamino)-2-[(S)-1-tert-butyldimethylsilyl-

oxy-2-methylpropyl]quinazolin-4(3H)-one 13. The general pro-

cedure I for monoacylation was followed using 3-amino-

quinazolinone 6⁵(2 g, 5.8 mmol), pyridine (0.68 g, 8.6 mmol),

dichloromethane (4 cm³) and 2-methylpropanoyl chloride

(0.73 g, 6.9 mmol). The brown oil obtained on work-up was

triturated with ethyl acetate–light petroleum and the solid

obtained crystallised to give the title 3-(2-methylpropanoyl-

amino)quinazolinone 13 as colourless crystals (1.9 g, 79%),

mp 116–118 ЊC (from light petroleum) (R_f0.38, 3 : 1 light

petroleum–ethyl acetate); [α]_D= ϩ27 (c 2.1 CHCl₃) (Found:

MH^ϩ418.2526. C₂₂H₃₆N₃O₃Si requires MH^ϩ418.2526);

δ_H(mixture of N–N bond rotamers), major rotamer Ϫ0.02 and

0.16 (6H, 2 × s, CH₃SiCH₃), 0.94 and 1.04 (6H, 2 × d, J 6.6,

CH₃CHCH₃), 0.97 [9H, s, (CH₃)₃C], 1.36 and 1.39 [6H, 2 × d,

J 6.9, (CH₃)₂CHCO], 2.08 [1H, m, (7 peaks) CH₃CHCH₃],

2.73 [1H, h (heptet),J 6.9, (CH₃)₂CHCO], 4.45 (1H, d, J 7.0,

CHOSi), 7.53 [1H, ddd, J 8.2, 7.0 and 1.6, 6-H(Q)], 7.75 [1H,

3-Benzoylamino-2-[(S)-1-tert-butyldimethylsilyloxy-2-methyl-

propyl]quinazolin-4(3H)-one 7. The general procedure I was

followed using 3-aminoquinazolinone 6⁵(2 g, 4.4 mmol), pyri-

dine (0.7 cm³, 8.8 mmol), dichloromethane (3 cm³) and benzoyl

chloride (0.97 g, 6.9 mmol). The yellow oil obtained on work-

up was triturated with ethyl acetate–light petroleum and the

solid obtained gave the title 3-benzoylaminoquinazolinone 7

as colourless crystals (2.1 g, 80%), mp 147–149 ЊC (from light

petroleum) (Found: MH^ϩ452.2369. C₂₅H₃₃N₃O₃Si, requires

MH^ϩ452.2369); [α]_D= ϩ26 (c 1, CHCl₃); ν_max/cm^Ϫ13350 w, br,

J. Chem. Soc., Perkin Trans. 1, 2002, 257–274

265

View Article Online

ddd, J 8.2, 7.0 and 1.6, 7-H(Q)], 7.82 [1H, dd, J 8.2 and 1.6, 8-

H(Q)], 8.21 (1H, s, NH) and 8.31 [1H, d, J 8.2, 5-H(Q)]; minor

rotamer (observable signals) 0.04 and 0.12 (6 H, 2 × s,

CH₃SiCH₃), 2.3 [1 H, m, (7 peaks) CH₃CHCH₃] and 4.65 (1 H,

d, J 7.0, CHOSi); from comparison of the signals at δ 4.45 and

δ 4.65, the ratio of N–N bond rotamers was 6 : 1; δ_CϪ4.9 and

Ϫ4.4 (CH₃SiCH₃), 18.6 [C(CH₃)₃], 19.4, 19.7, 33.2 and 34.5

(4 × CH₃), 26.2 [(CH₃)₃C], (CHOSi) missing, 121.4 [4 × CCO-

added the acid chloride (2–3 eq.) dropwise over 10 min and

the mixture stirred and heated under reﬂux for 2–4 days,

monitoring the disappearance of the starting 3-monoacyl-

aminoquinazolinone by TLC. After cooling, additional

dichloromethane was added, and the solution was washed with

aqueous sodium hydrogen carbonate, then water, dried, and the

dichloromethane removed under reduced pressure. The bulk

of the residual pyridine was removed using an oil pump and the

product puriﬁed by ﬂash chromatography.

(Q)], 127.5, 128.8 and 135.1 [4 × CH(Q)], 146.8 [CN᎐C(Q)],

᎐

160.4 [CO(Q)] and 174.1 (CO); m/z (%) (FAB) 418 (MH^ϩ, 100),

360 (77), 275 (23) and 216 (31).

3-Diethanoylamino-2-[(S)-1-tert-butyldimethylsilyloxyethyl]-

quinazolinone-4(3H)-one 2. 3-Aminoquinazolinone 10¹¹(0.5 g,

1.24 mmol) was dissolved in acetic anhydride (1 cm³), pyridine

(1 cm³) was added and the reaction mixture stirred at room

temperature for 48 h. The solution was poured into water

(1 cm³), excess acetic anhydride decomposed by stirring for

∼10 min and the product extracted into dichloromethane

(2 × 20 cm³). The combined organic extracts were washed with

saturated aqueous sodium hydrogen carbonate (3 × 20 cm³),

then water, dried, and the solvents removed under reduced

pressure. Chromatography of the oily product obtained with

light petroleum–ethyl acetate (5 : 1) as eluent gave the title DAQ

2 (R_f0.32) as a colourless oil (0.51, 80%) (Found: M^ϩ403.2005.

C₂₀H₂₉N₃O₄Si requires M^ϩ403.2005); δ_HϪ0.07 and 0.02 (6H,

2 × s, CH₃SiCH₃), 0.79 [9H, s, (CH₃)₃C], 1.39 (3H, d, J 6.6,

CH₃CHO), 2.29 and 2.32 (6H, 2 × s, 2 × CH₃CO), 4.8 (1H,

q, J 6.6, CHOSi), 7.4 [1H, ddd, J 8.2, ∼7 and ∼1.5, 6-H(Q)], 7.65

[1 H, dd, J 8.2 and ∼1.5, 8-H(Q)], 7.7 [1H, ddd, J ∼8, 7.0 and

∼1.5, 7-H(Q)] and 8.15 [1H, dd, J ∼8 and ∼1.5, 5-H(Q)]; m/z (%)

403 (M^ϩ,100), 362 (35), 346 (62), 304 (55), 262 (22), 230 (26)

and 188 (21).

3-Acetylamino-2-diphenylmethylquinazolin-4(3H)-one

30.

The general procedure I was followed using the 3-amino-

quinazolinone 38⁴(2 g, 6.12 mmol), pyridine (0.95 cm³,

12 mmol), dichloromethane (4 cm³) and acetyl chloride (0.58 g,

7.3 mmol) and the reaction mixture stirred for 24 h at room

temperature. After work-up the brown oil obtained was puriﬁed

by column chromatography on silica using light petroleum–

ethyl acetate (1 : 1) as eluent to give the title 3-ethanoyl-

aminoquinazolinone 30 as colourless crystals (1.8 g, 80%) (R_f

0.34; 1 : 1 ethyl acetate–petroleum); mp 228–231 ЊC (from ethyl

acetate) (Found: MH^ϩ370.1556. C₂₃H₁₉N₃O₂, requires MH^ϩ

370.1556); δ_H2.30 (3H, s, CH₃CO), 5.71 (1H, s, PhCHPh),

7.28–7.48 [10H, m, 10 × CH(Ar)], 7.53 [1H, ddd, J 8.0, 6.9 and

1.1, 6-H(Q)], 7.69 [1H, d, J 8.0, 8-H(Q)], 7.80 [1H, ddd, J 8.0,

6.9 and 1.1, 7-H(Q)], 8.12 (1H, s, NH) and 8.21 [1H, dd, J 8.0

and 1.1, 5-H(Q)]; δ_C(d₆-DMSO) 20.8 (CH₃CO), 52.7 (Ph₂CH),

121.0 [CCO(Q)], 126.8, 127.1, 127.4, 127.5, 127.9, 128.4, 128.9,

129.4, 129.6 and 135.4 [10 × CH(Ar) and 4 × CH(Q)], 139.8 and

140.4 [2 × C(Ph)], 146.4 [CN᎐C(Q)], 159.2 and 159.3 [C᎐N(Q)

᎐

and CO(Q)] and 169.9 (CH₃CO).

3-Dibenzoylamino-2-[(S)-1-tert-butyldimethylsilyloxyethyl]-

quinazolin-4(3H)-one 3. The general procedure II for diacyl-

ation was followed using MAQ 11 (3 g, 9.4 mmol), pyridine

(1.5 g, 19 mmol), dichloromethane (5 cm³) and benzoyl chloride

(2.64 g, 19 mmol) and the reaction mixture heated under reﬂux

for 10 h. The yellow oil obtained after work-up was puriﬁed by

column chromatography on silica using light petroleum–ethyl

acetate (3 : 1) as eluent and gave DAQ 3 (R_f0.64) as colourless

crystals (1.5 g, 41%), mp 127–129 ЊC (from light petroleum)

(Found: C, 68.0; H, 6.3; N, 7.9. C₃₀H₃₃N₃O₄Si requires C, 68.2;

H, 6.3; N, 7.9%) (Found: MH^ϩ528.2319. C₃₀H₃₃N₃O₄Si,

requires MH^ϩ528.2319); δ_HϪ0.02 and 0.01 (6H, 2 × br s,

CH₃SiCH₃), 0.9 [9H, s, (CH₃)₃C], 1.84 (3H, br d, J 6.6, CH₃-

CHOSi), 5.27 (1H, br q, J 6.6, CHOSi), 7.1–7.4 [7H, m, 6-H(Q)

and 6 × PhCH], 7.6 [1H, ddd, J 8.2, 7.0 and 1.3, 7-H(Q)], 7.80–

7.87 [5H, m, 8-H(Q) and 4 × CH(Ph)] and 8.4 [1H, d, J 8.2,

5-H(Q)]; δ_CϪ4.8 and Ϫ4.4 (CH₃SiCH₃), 19.0 (CH₃CHO-

Si), 22.7 [(CH₃)₃CSi], 26.3 [(CH₃)₃CSi], 71.6 (CHOSi), 122

[CCO(Q)], 127.8, 128.4, 128.6, 130.1, 130.5, 132.8 and 134.8

[10 × CH(Ph) and 4 × CH(Q)], 135.0 and 135.5 [2 × C(Ph)],

2-[(S)-1-tert-Butyldimethylsilyloxy-2-methylpropyl]-3-

ethanoylaminoquinazolin-4(3H)-one 16. The general procedure

I was followed using 3-aminoquinazolinone 6⁵(1 g, 2.9 mmol),

pyridine (0.34 g, 4.4 mmol), dichloromethane (2 cm³) and acetyl

chloride (0.34 g, 4.4 mmol) and the mixture was stirred for

24 h at room temperature. After work-up the brown oil

obtained was puriﬁed by ﬂash chromatography on silica using

light petroleum–ethyl acetate (5 : 1) as eluent to give the 3-

ethanoylaminoquinazolinone 16 as a colourless oil (0.72 g, 64%)

(R_f0.31) (Found: MH^ϩ390.2213. C₂₀H₃₁N₃O₃Si requires MH^ϩ

390.2213); ν_max/cm^Ϫ13400m 1700s, 1610s, 1470s and 1075s;

δ_H(mixture of N–N bond rotamers) major rotamer, Ϫ0.02 and

0.15 (6H, 2 × s, CH₃SiCH₃), 0.93 and 1.15 (6H, 2 × d, J 6.6,

CH₃CHCH₃), 0.97 [9H, s, (CH₃)₃C)], 2.08 [1H, m, (7 peaks),

CH₃CHCH₃], 2.3 (3H, s, CH₃CO), 4.49 (1H, d, J 6.9, CHOSi),

7.73 [1H, ddd, J 8.2, 7.0 and 1.3, 6-H(Q)], 7.76 [1H, dd, J 8.2

and 7.0 7-H(Q)], 7.81 [1H, dd, J 8.2 and 1.3, 8-H(Q)], 8.20 [1H,

d, J 8.2, 5-H(Q)] and 8.65 (1H, s, NH); δ_Hminor rotamer,

(observable signals), 4.62 (1H, d, J 6.9, CHOSi) and 8.72 (1H,

s, NH); δ_CϪ4.7 and Ϫ3.9 (CH₃SiCH₃), 18.4 [C(CH₃)₃], 147.3

147.3 [CN᎐C(Q)], 157.4 [C᎐N(Q)], 160.5, [CO(Q)], 170.1 and

᎐

170.9 (2 × CO); m/z (%) (FAB), 528 (MH^ϩ, 100) and 470 (70).

Further elution with the same solvent mixture gave unreacted

MAQ³11 as colourless crystals (1.5 g).

[C–C᎐N(Q)], 157.8 [CN(Q)], 161.3 [CO(Q)] and 170.9 (CO).

᎐

From comparison of the intensities of signals at δ 4.49 and

δ 4.62 the ratio of rotamers was 4 : 1; δ_CϪ4.9 and Ϫ4.3

(CH₃SiCH₃), 18.6 [(CH₃)₃C], 18.7, 19.9 and 32.8 (3 × CH₃),

26.3 [(CH₃)₃C], 33.6 [(CH₃)₂CH], 79.0, [br (CHOSi)], 121.2

[CCO(Q)], 127.3, 127.5, 128.5 and 135.3 [4 × CH(Q)], 146.9

3-(N-Benzoyl-N-2-methylpropanoylamino)-2-[(S)-1-tert-

butyldimethylsilyloxy-2-methylpropyl]quinazolin-4(3H)-one 8.

General procedure II for diacylation was followed using 3-

benzoylaminoquinazolinone 7 (1 g, 2.2 mmol), dry dichloro-

methane (2 cm³), dry pyridine (0.36 g, 4.6 mmol, 2 eq.) and 2-

methylpropanoyl chloride (0.47 g, 4.7 mmol) and the mixture

stirred at room temperature for 5 days. The yellow oil obtained

on work-up (1.2 g) was puriﬁed by ﬂash chromatography

over silica gel and elution with light petroleum–ethyl acetate

(5 : 1) to give DAQ¹8 (R_f0.48) as a colourless oil (1 g, 83%) and

as a mixture of diastereoisomers. Re-chromatography using

kieselgel with light petroleum–ethyl acetate (10 : 1) as eluent

gave the faster running DAQ¹diastereoisomer 8a as a viscous

[C–C᎐N(Q)], 157.1 [C᎐N(Q)], 160.7 [CO(Q)] and 169.4 (CO);

᎐

m/z (%) (FAB) 390 (MH^ϩ, 100) 374 (20), 332 (84), 290 (21), 216

(44) and 187 (20).

General procedure II for preparation of 3-diacylaminoquin-

azolinones (DAQs) from 3-monoacylaminoquinazolinones

(MAQs)

To a solution of the 3-acylaminoquinazolinone (1 eq.), pre-

pared as described above, in dry dichloromethane (2 cm³g^Ϫ1

)

containing dry pyridine (1.5 eq.) and dry DMF (4 drops), was

266 J. Chem. Soc., Perkin Trans. 1, 2002, 257–274

View Article Online

colourless oil (0.26 g, 23%) (R_f0.42) (Found: MH^ϩ, 522.2789.

C₂₉H₃₉N₃O₄Si requires MH^ϩ, 522.2789); ν_max/cm^Ϫ11700s and

1605s; δ_HϪ0.13 and 0.01 (6H, 2 × s, CH₃SiCH₃), 0.98 [9H, s,

(CH₃)₃CSi], 1.05 and 1.13, 1.28 and 1.39 (12H, 4 × d, J 6.6, 2 ×

CH₃CHCH₃), 2.2 [1H, m, (7 peaks), CH₃CHCH₃], 2.95 [1H, h,

J 6.6, (CH₃)₂CH], 4.62 (1H, br d, J 5.0, CHOSi), 7.50–7.75 [4H,

m, 3 × CH(Ph) and 6-H(Q)], 7.8–8.10 [4H, m, 7- and 8-H(Q)

and 2 × CH(Ph)] and 8.42 [1H, dd, J 8.0 and 1.3, 5-H(Q)];

δ_CϪ4.9 and Ϫ3.7 (CH₃SiCH₃), 18.8 [(CH₃)₃CSi], 19.7, 20.6,

20.9, 32.3 and 35.3 (4 × CH₃and 2 × CH), 26.3 [(CH₃)₃C],

(CHOSi missing), 121.6 [CCO(Q)], 127.7, 127.8, 128.4, 129.2,

129.3, 133.7 and 135.5 [5 × CH(Ph) and 4 × CH(Q)], 134.1

[C(Ph)], 146.7 [CN᎐C(Q)], 156.6 [C᎐N(Q)], 160.5 [CO(Q)] and

A crystal suitable for X-ray crystallographic structure

determination was obtained from ethanol (Fig. 1).

3-[N-(S)-2-Acetoxypropanoyl-N-ethanoylamino]-2-diphenyl-

methylquinazolin-4(3H)-one 31. General procedure II was

followed using MAQ²30 (1.1 g, 2.7 mmol), dichloromethane

(2 cm³), pyridine (0.4 g, 5.1 mmol) and (S)-2-acetoxypropanoyl

chloride (0.82 g, 5.4 mmol) and the mixture stirred at room

temperature for 24 h. Chromatography of the yellow oil (0.95 g)

obtained using ﬂash silica and light petroleum–ethyl acetate

(2 : 1) as eluent gave the minor diastereoisomer 31a as colour-

less crystals (0.43 g, 33%); mp 163–165 ЊC (from methanol)

(R_f0.43) (Found: MH^ϩ484.1873. C₂₈H₂₅N₃O₅requires MH^ϩ

484.1873); ν_max/cm^Ϫ11745s, 1710s, 1610 and 1602s; δ_H1.08 (3H,

s, CH₃CO), 1.51 (3H, d, J 6.9, CH₃CHOAc), 2.03 (3H, s,

CH₃CO₂), 5.48 (1H, s, PhCHPh), 6.01 (1H, q, J 6.9, CH₃CH-

OAc), 7.03–7.26 [10H, m, 10 × CH(Ar)], 7.31 [1H, ddd, J 8.3,

6.6 and 1.3, 6-H(Q)], 7.50 [1H, dd, J 8.2 and 1.3, 8-H(Q)], 7.59

[1H, ddd, J 8.2, 6.6 and 1.3, 7-H(Q)] and 8.05 [1H, dd, J 8.2 and

᎐

170.2 and 178.8 [2 × CO]; m/z (%), (FAB) 522 (MH^ϩ, 66%), 452

(80), 436 (21), 394 (100), 275 (60), 232 (47) and 187 (33).

Further elution with the same solvent mixture gave the major

more polar DAQ¹diastereoisomer 8b (R_f0.37) as colourless

crystals (0.46 g, 40%); mp 126–128 ЊC (from light petroleum)

(Found: C, 66.8; H, 7.5; N, 8.1. C₂₉H₃₉N₃O₄Si requires C, 66.8;

H, 7.5; N, 8.1%); ν_max/cm^Ϫ11700s and 1605s; δ_HϪ0.02 and 0.07

(6H, 2 × s, CH₃SiCH₃), 0.89 [9H, s, (CH₃)₃CSi], 1.08, 1.14, 1.20

and 1.36 (12H, 4 × d, J 6.6, 2 × CH₃CHCH₃), 2.23 [1H, m,

(7 lines), CH₃CHCH₃], 2.78 [1H, h, J 6.6, (CH₃)₂CHCO], 4.66

(1H, d, J 7.2, CHOSi), 7.53–7.67 [4H, m, 3 × CH(Ph) and

6-H(Q)], 7.8–7.94 [2 H, m, 7- and 8-H (Q)], 7.95–8.05 [2H, m,

2 × CH(Ph)] and 8.42 [1H, dd, J 8.3 and 1.0, 5-H(Q)]; δ_CϪ4.6

and Ϫ3.8 (CH₃SiCH₃), 18.9 [(CH₃)₃CSi], 19.6, 20.3, 21.0, 32.5

and 35.6 (4 × CH₃and 2 × CH), 26.4 [(CH₃)₃C], (CHOSi

missing), 121.7 [CCO(Q)], 127.7, 128.3, 135.4, 129.1, 129.7,

133.5 and 136.4 [5 × CH(Ph) and 4 × CH(Q)], 135.0 [C(Ph)],

1

1.3, 5-H(Q)]; The H NMR spectrum at 400 MHz at Ϫ50 ЊC

remained unchanged; δ_C16.2, 20.9, and 22.0 (3 × CH₃), 52.9

(PhCHPh), 72.2 (CH₃CHOAc), 121.0 [CCO(Q)], 127.2, 127.8,

128.1, 128.6, 128.9, 129.2, 129.8, 129.9, 130.3 and 135.8 [10 ×

CH(Ar) and 4 × CH(Q)], 138.1 and 140.0 [2 × C(Ph)], 147.0

[CN᎐C(Q)], 157.9 [C᎐N(Q)], 160 [CO(Q)] and 171.8, 172.5, and

᎐

172.7 (3 × CO); m/z (%) (FAB), 484 (MH^ϩ, 100), 154 (100), 136

(82) and 115 (40).

An X-ray structure determination was obtained on a crystal

from methanol (Fig. 3a).

Further elution with the same solvent mixture gave the DAQ²

diastereoisomer 31b (R_f0.23) as colourless crystals (0.45 g,

35%) mp 153–155 ЊC (from ethyl acetate) (Found: MH^ϩ

484.1873. C₂₈H₂₅N₃O₅requires MH^ϩ484.1873); ν_max/cm^Ϫ1

1740s, 1705s, 1608 and 1600s; δ_H1.25 (3H, s, CH₃CO), 1.82

(3H, d, J 6.6, CH₃CHOAc), 2.24 (3H, s, CH₃CO), 5.44 (1H, s

br, PhCHPh), 6.07 (1H, q br, J 6.6, CH₃CHOAc), 7.25–7.53

[10H, m, 10 × CH(Ar)], 7.57 [1H, ddd, J 8.2, 7.1 and 1.3,

6-H(Q)], 7.50 [1H, dd, J 8.2 and 1.3, 8-H(Q)], 7.59 [1H, ddd,

J 8.2, 7.1 and 1.3, 7-H(Q)] and 8.05 [1H, dd, J 8.2 and 1.3,

5-H(Q)]; δ_C17.4, 21.4, and 23.1 (3 × CH₃), 53.9 [(Ph)₂CH], 71.0

(CH₃CHOAc), 121.3 [CCO(Q)], 127.5, 127.9, 128.0, 128.6,

128.8, 129.8, 130.0 and 135.7 [10 × CH(Ar) and 4 × CH(Q)],

146.8 [CN᎐C(Q)], 156.3 [C᎐N(Q)], 160.4 [CO(Q)] and 169.9

᎐

and 179.4 (2 × CO). From comparison of signals at δ 2.78 and

2.95 in NMR spectrum of the crude reaction product the

ratio of 8b–8a was 1.6 : 1; m/z (%) (FAB) 522 (MH^ϩ, 66%), 452

(80), 436 (21), 394 (100), 275 (60), 232 (47) and 187 (33). An

X-ray structure determination was carried out on a crystal of

diastereoisomer 8b obtained from ethanol.^3b

DAQ¹8b (25 mg) was dissolved in CDCl₃(0.5 cm³) and

heated at 60 ЊC for 14 h to give a 1 : 1 ratio of 8a–8b by com-

parison of the NMR spectrum of the solution with those of

authentic samples above.

N-Benzoyl-N-ethanoylamino-2-[(S)-1-tert-butyldimethylsilyl-

oxy-2-methylpropyl]quinazolin-4(3H)-one 9. General procedure

II for diacylation was followed using 3-benzoylaminoquin-

azolinone 7 (1 g, 2.22 mmol) dissolved in dry dichloromethane

(2 cm³) containing dry pyridine (0.15 g, 1.9 mmol) with acetyl

chloride (0.29 g, 3.7 mmol) added dropwise over 5 min, and the

mixture stirred for 3 days with heating under reﬂux. The yellow

oil obtained on work-up (1.2 g) was puriﬁed by ﬂash chroma-

tography over silica gel with light petroleum–ethyl acetate (4 : 1)

as eluent to give DAQ¹9 (R_f0.56) as colourless crystals (0.84 g,

76%), mp 136–138 ЊC (from light petroleum), [α]_D= Ϫ222.4

(c 2.0, CHCl₃) (Found: C, 65.0; H, 7.1; N, 8.4; C₂₇H₃₅N₃O₄Si

requires C, 65.1; H, 7.1; N, 8.5%) (Found: MH^ϩ494.2475.

C₂₇H₃₅N₃O₄Si requires MH^ϩ494.2475); ν_max/cm^Ϫ11700s and

1600s; δ_HϪ0.02 and 0.02 (6H, 2 × s, CH₃SiCH₃), 0.82 [9H, s,

(CH₃)₃CSi], 1.09 and 1.24 (6H, 2 × d, J 6.6, CH₃CHCH₃), 2.23

(3H, s, CH₃CO), 2.45 [1H, m, (7 peaks), CH₃CHCH₃], 4.63

(1H, d, J 8.2, CHOSi), 7.59–7.74 [4H, m, 3 × CH(Ph) and 6-

H(Q)], 7.83–8.08 [4H, m, 7- and 8-H(Q) and 2 × CH(Ph)] and

8.43 [1H, d, J 8.2, 5-H(Q)]; δ_CϪ4.6 and Ϫ3.9 (CH₃SiCH₃), 18.9

[(CH₃)₃C], 19.1 and 20.07 (CH₃CHCH₃), 26.1 (CH₃CHCH₃),

26.4 [(CH₃)₃C], 32.5 (CH₃CO), 82.9 (CHOSi), 121.6 [CCO(Q)],

127.8, 127.9, 128.4, 129.1, 129.4, 130.2 and 135.5 [5 × CH(Ph)

137.4 and 139.6 [2 × C(Ph)], 146.7 [CN᎐C(Q)], 157.3 [C᎐N(Q)],

᎐ ᎐

159.9 [CO(Q)] and 170.4, 171.5, and 171.9 (3 × CO); from

comparison of signals δ_H6.07 and 6.01 in the NMR spectrum

of the product obtained after the ﬁrst column chromatography,

the ratio of 31b–31a diastereoisomers present was 1.1 : 1; the

signals at δ 6.07 and 5.44 above showed (400 MHz) the follow-

ing temperature dependence: 0 ЊC ∼6.0, s, br (coalescence) and

5.29 s, br; Ϫ25 ЊC 6.1 s, br and 5.11 s, br; 40 ЊC 6.09, q, sharp

and 5.1 s, sharp, respectively; m/z (%) (FAB), 484 (MH^ϩ, 74),

154 (100), 136 (82) and 115 (40).

An X-ray structure determination was obtained on a crystal

from methanol (Fig 3b).

3-{Bis[(S)-2-acetoxypropanoyl]amino}-2-diphenylmethyl-

quinazolin-4(3H)-one 37a. Using general procedure II, a

mixture of 3-[(S)-2-acetoxypropanoyl]aminoquinazolinone 32⁴

(1.1 g, 2.5 mmol), dichloromethane (3 cm³), pyridine (0.39 g,

5.0 mmol) and distilled (S)-2-acetoxypropanoyl chloride

(0.56 g, 3.74 mmol) was stirred continuously for 24 h at room

temperature. The brown oil obtained after work-up was puri-

ﬁed by column chromatography on silica using light petroleum–

ethyl acetate (2 : 1) as eluent to give a colourless oil (R_f0.43)

which solidiﬁed on standing. Crystallisation aﬀorded DAQ²

37a (1.2 g, 87%) as a colourless solid, mp 169–170 ЊC (from

methanol) (Found: C, 66.9; H, 5.2; N, 7.6. C₃₁H₃₀N₃O₇requires

C, 67.0; H, 5.2; N, 7.6%) (Found: MH^ϩ556.2083. C₃₁H₃₀N₃O₇

requires MH^ϩ556.2084); ν_max/cm^Ϫ11750s, 1689s and 1600s;

δ_H1.51 (6H, br d, J 6.7, CH₃CHOAc), 2.10 and 2.25 (6H, 2 × s,

and 4 × CH(Q)], 134.8 [C(Ph)], 146.9 [CN᎐C(Q)], 155.8

᎐

[CN(Q)], 160.4 [CO(Q)] and 169.3 and 171.8 (CO); m/z (%)

(FAB) 494 (MH^ϩ, 100), 436 (63) and 394 (65). The 1H NMR

spectrum of a sample that had been heated at 137 ЊC for ∼1 min

showed no change.

J. Chem. Soc., Perkin Trans. 1, 2002, 257–274

267

View Article Online

1

2 × CH₃CO₂), 5.08 and 6.03 (2H, 2 × q br, CHOAc), 5.84 (1H,

br s, PhCHPh), 7.31–7.72 [11H, m, 10 × CH(Ar) and 6-H(Q)],

7.86 [1H, dd, J 8.3 and 1.0, 8-H(Q)], 7.95 [1H, ddd, J 8.3, 7.0

and 1.0, 7-H(Q)] and 8.35 [1H, dd, J 8.0 and 1.0, 5-H(Q)];

δ_H(400 MHz, Ϫ50 ЊC) 1.26 and 1.52 (6H, 2 × d, J 6.6, 2 ×

CH₃CHOAc), 2.0 and 2.4 (6H, 2 × s, 2 × CH₃CO₂), 4.54 and

6.11 (2H, 2 × q, J 6.6, 2 × CH₃CHOAc), 5.82 (2H, s, PhCHPh)

additional signals (∼20% of those given previously) were

present at δ 4.40 (br s), 5.07 (s), 6.07 (br s) and 7.07 (br s);

δ_C(25 ЊC) 16.8, 20.7, 20.9 and 21.3 (4 × CH₃), 52.8 (PhCHPh),

69.6 and 70.9 (2 × CHOAc), 120.9 [CCO(Q)], 127.6, 127.8,

128.2, 128.6, 128.8, 128.9, 129.2, 129.8, 129.9 and 136.2 [10 ×

CH(Ph) and 4 × CH(Q)], 139.0 and 139.6 [2 × C(Ph)], 146.6

chloroform (0.5 cm³) was stirred for 6 h at 0 ЊC. An H NMR

spectrum (400 MHz) showed the ratio of N-acetylpyrrolidine

to N-acetylpiperidine was 20 : 1 from comparison of signals at

δ 3.42 and 3.50 with those in the spectra of authentic samples.

The same procedure was carried out using DAQ 3 (0.1 g,

0.19 mmol), pyrrolidine (14 mg, 0.19 mmol) and piperidine

(16 mg, 0.19 mmol) in deuterochloroform (0.5 cm³) for 4 h at

Ϫ10 ЊC. An ¹H NMR spectrum (400 MHz) showed the ratio of

N-benzoylpyrrolidine to N-benzoylpiperidine was ∼30 : 1 from

comparison of signals at δ 3.6 and 3.7 with those in the spectra

of authentic samples.

General procedure III for enantioselective acylation (kinetic

resolution) of racemic amines

[CN᎐C(Q)], 157.3 (C᎐N), 160.7 [CO(Q)] and 169.9, 170.7,

᎐

171.5 and 172.3 (4 × CO); m/z (%) (FAB), 156 (MH^ϩ, 100), 442

(96), 311 (51) and 167 (99).

To the DAQ (1 eq.) dissolved in the dichloromethane (1 cm³

100 mg^Ϫ1) was added racemic amine (2 eq.) and the solution

stirred at Ϫ20 to Ϫ10 ЊC and then at 5 ЊC for the time given,

monitoring the disappearance of the starting material by TLC

A crystal grown from methanol was suitable for X-ray struc-

ture determination (Fig. 4).

at 5 ЊC. To work up, further dichloromethane (2 cm³100 mg^Ϫ1

)

Preparation of meso-bis[(S)-2-acetoxypropanoylamino]-2-

diphenylmethylquinazolin-4(3H)-one 37b. Using general pro-

cedure II, a mixture of 3-[(S)-2-acetoxypropanoylamino]-

quinazolinone 32 (0.7 g, 1.5 mmol), dichloromethane (2 cm³),

pyridine (0.23 g, 2.96 mmol) and racemic-2-acetoxypropanoyl

chloride (0.33 g, 2.2 mmol) was stirred continuously for 24 h at

room temperature. Flash column chromatography of the yellow

oil obtained on work-up over silica with light petroleum–ethyl

acetate (2 : 1) as eluent gave a colourless solid whose TLC

showed two spots at R_fs 0.43 and 0.37.

was added and the solution washed with hydrochloric acid

(2 M) to remove unreacted amine, then with water, dried, and

evaporated under reduced pressure. Separation of product was

carried out using a chromatotron or ﬂash chromatography.

The unreacted amine was recovered from the aqueous acid

extract as the hydrochloride salt by evaporation of the bulk of

the acid-water under reduced pressure ﬁrst using an oil pump

and then drying in a desiccator containing P₂O₅for 24 h.

For NMR spectroscopic examination of the progress of the

reactions, deuterochloroform was substituted for dichloro-

methane in the procedure above. Yields of amide were based on

the (eq. amine used)/2.

Further elution with light petroleum–ethyl acetate (1 : 1)

gave unchanged MAQ²32 as a colourless oil (0.49 g) (R_f0.5,

1 : 1 light petroleum–ethyl acetate).

Reaction of DAQ¹8a with a 2-methylpiperidine. General

procedure III was followed using DAQ¹8a (0.1 g, 0.19 mmol)

and racemic 2-methylpiperidine (38 mg, 0.38 mmol) in dichloro-

methane (1 cm³) and the mixture stirred at Ϫ20 ЊC for 2 h and

then at 5 ЊC for 12 h. Flash chromatography of the product with

light petroleum–ethyl acetate (3 : 1) as eluent gave MAQ¹13 (69

mg, 86%) (R_f0.38, 3 : 1 light petroleum–ethyl acetate) identical

with an authentic sample prepared as described above.

Further elution with the same solvent mixture gave (R)-N-

benzoyl-2-methylpiperidine 12 as a colourless oil (33 mg, 85%);

[α]_D= Ϫ31.4 (c 0.8, CHCl₃), ee 95% by comparison with an

authentic sample prepared as described below.

Re-chromatography of the solid mixture above using a

chromatotron and light petroleum–ethyl acetate (2 : 1) as eluent

gave DAQ²37a as a colourless solid (0.08 g, 10%) (R_f0.43)

identical with that isolated previously.

Further elution with the same solvent mixture gave meso-

DAQ²37b (R_f0.37) as a colourless oil (0.01 g, 1%) (Found:

MH^ϩ556.2085. C₃₁H₃₀N₃O₇requires MH^ϩ556.2084); δ_H1.3

(6H, d, J 6.7, 2 × CH₃CHOAc), 2.06 (6H, s, 2 × CH₃CO₂),

5.45 (2H, br q, J 6.7, 2 × CH₃CHOAc), 5.91 (1H, s, PhCHPh),

7.2–7.9 [13H, m, 10 × CH(Ar) and 6-, 7- and 8-H(Q)] and 8.26

[1H, dd, J 8.0 and 1.0, 5-H(Q)]. In variable-temperature NMR

studies (400 MHz) (27, 0, Ϫ25 and Ϫ44 ЊC) the broadened

signal at δ 5.45 of CH₃CHOAc was split into two broad signals

(δ 5.0 and 5.80) with a coalescence temperature at ∼0 ЊC.

General procedure IV; derivatization of the unreacted amine

(i) With benzoyl chloride. The unreacted 2-methylpiperidine

enantiomer was recovered as the hydrochloride acid salt from

the experiment above as a colourless solid (25 mg, 96%). This

salt (22 mg, 0.16 mmol) was dissolved in pyridine (0.5 cm³),

benzoyl chloride (34.2 mg, 0.24 mmol) added and the reaction

mixture stirred for 2 h at room temperature. After addition

of dichloromethane (1 cm³) the solution was washed with

hydrochloric acid (2 M; 0.5 cm³) followed by water, dried

and evaporated and the pale yellow oil (41 mg) obtained was

puriﬁed by column chromatography over silica with light

petroleum–ethyl acetate (3 : 1) as eluent to give (S)-N-benzoyl-

N-[(S)-2-Acetoxypropanoyl-N-ethanoylamino]phthalimide

41. General procedure II was followed using N-acetylamino-

phthalimide 40¹²(0.6 g, 2.94 mmol), dichloromethane (2 cm³),

pyridine (0.46 g, 5.9 mmol) and (S)-2-acetoxypropanoyl

chloride (0.89 g, 5.9 mmol) with stirring at room temperature

for two days. The pale brown oil (1.1 g) obtained was puriﬁed

by ﬂash column chromatography using light petroleum–ethyl

acetate (1 : 1) as eluent to give N-phthalimido-imide 41 as a

colourless oil (0.7 g, 75%) (R_f0.67) (Found: MH^ϩ319.0930.

C₁₅H₁₄N₂O₆requires MH^ϩ319.0930); ν_max/cm^Ϫ11800s, 1730s,

and 1360s; δ_H1.54 (3H, d, J 6.9, CH₃CHOAc), 2.08 (3H, s,

CH₃CO₂), 2.43 (3H, s, CH₃CO), 5.72 (1H, q, J 6.9, CHOAc)

and 7.83–8.20 [4H, m, CH(Ar)]; δ 16.8, 20.7 and 24.6 (3 ×

CH₃), 70.1 (CH₃CHOAc), 124.8, 124.9, 135.6 and 135.8 [4 ×

CH(Ar)], 130.2, 130.3 [2 × C(Ar)] and 164.8, 170.2, 170.4 and

171.1 (5 × CO); m/z (%) (FAB), 319 (MH^ϩ, 48), 277 (45), 259

(100), 154 (60), 137 (62) and 115 (71).

2-methylpiperidine 12 as a colourless oil (23.1 mg, 70%); [α]_D

=

ϩ30 (c 0.64, CHCl₃), ee 91% by comparison with an authentic

sample [α]_D= ϩ32.8 (c 0.64, CHCl₃) (see below).

(ii) With (S)-2-acetoxypropanoyl chloride. The aqueous acid

extract in another experiment using DAQ¹8a carried out as

described above, containing unreacted 2-methylpiperidine

hydrochloride, was made alkaline with sodium hydroxide (2 M)

and the solution extracted with ether (3 × 1 cm³), the combined

ether layers, dried over powdered potassium hydroxide, pyridine

(24 mg, 0.3 mmol) and (S)-2-acetoxypropanoyl chloride

(46 mg, 0.3 mmol) added and the mixture stirred for 2 h at

Competitive reactions of pyrrolidine and piperidine with DAQs

2 and 3

A solution of DAQ 2 (0.1 g, 0.25 mmol), pyrrolidine (18 mg,

0.25 mmol) and piperidine (21 mg, 0.25 mmol) in deutero-

268

J. Chem. Soc., Perkin Trans. 1, 2002, 257–274

View Article Online

room temperature. Work-up as described above gave amide 14

(24 mg, 72%) whose H NMR (400 MHz, 50 ЊC) showed it to

respectively. Chromatotron chromatography with light

1

petroleum–ethyl acetate (2 : 1) as eluent gave MAQ¹7 as colour-

less crystals (18 mg, 20%), (R_f0.50), identical with that isolated

previously.

contain an 18 : 1 mixture of (S,S)–(R,S) diastereoisomers 14a–

14b (de 89%) from comparison of the intensities of signals in

its NMR spectrum at δ 2.09 and 2.10 respectively, adding

incremental amounts of an authentic sample of the (R,S)

diastereoisomer, prepared as described below, and monitoring

the increase in the signal at 2.10 ppm.

Further elution with the same solvent mixture gave a mixture

of MAQ¹16 and amide 17 which was dissolved in THF (1 cm³),

an excess of TBAF in THF (1 M, 0.3 g) added and the mixture

stirred for 6 h at room temperature. The bulk of the solvent was

removed under reduced pressure, the residual oil dissolved in

dichloromethane (3 cm³) and the solution washed successively

with saturated aqueous sodium hydrogen carbonate, brine, and

water, then dried to give a yellow oil (0.11 g). Flash column

chromatography with light petroleum–ethyl acetate (2 : 1) as

eluent gave N-benzoyl-3-methylpiperidine 17 as a colourless oil

(13 mg, 32%), [α]_D= ϩ36.7 (c 0.3, CDCl₃); ee 85% based on an

enantiopure sample prepared as described below.

Further elution with the same solvent mixture gave 3-

ethanoylamino-2-(1-hydroxy-2-methylpropyl)quinazolinone 19

as a colourless oil (23 mg, 41%) (Found: MH^ϩ276.1348.

C₁₄H₁₇N₃O₃, requires MH^ϩ276.1348); ν_max/cm^Ϫ13380w, 3240w,

1700s and 1612s; δ_H(mixture of N–N bond rotamers) major

rotamer 0.95 and 1.03 (6H, 2 × d, J 6.9, CH₃CHCH₃), 2.40 (1H,

m CH₃CHCH₃), 2.52 (3H, s, CH₃CO), 4.74 (1H, d, J 3.3,

CHOH), 7.59–7.75 [1H, m, 6-H(Q)], 7.8–8.08 [2H, m, 8- and

7-H(Q)] and 8.36 [1H, d, J 8.1, 5-H(Q)]; minor rotamer

(observable signals), 2.49 (3H, s, CH₃CO) and 4.93 (1H, d,

J 3.0, CHOH); δ_C20.6, 20.8 and 21.3 (3 × CH₃), 32.9

[(CH₃)₂CH], 72.9 (CHOH), 120.8 [CCO(Q)], 127.3, 127.5,

Reaction of DAQ¹8b with a 2-methylpiperidine

General procedure III was followed using DAQ¹8b (0.1 g, 0.19

mmol) and 2-methylpiperidine (38 mg, 0.38 mmol) in deutero-

chloroform (0.5 cm³) and the mixture stirred at Ϫ20 ЊC for 2 h

and then at 5 ЊC for 30 h. Flash chromatography of the product

with light petroleum–ethyl acetate (3 : 1) as eluent gave MAQ¹

13 (69 mg, 86%) (R_f0.38, 3 : 1 light petroleum–ethyl acetate)

identical with that isolated previously.

Further elution with the same solvent mixture gave (S)-N-

benzoyl-2-methylpiperidine 12 as a colourless oil (32 mg, 82%);

[α]_D= ϩ26.7 (c 0.75, CHCl₃) ee 81% by comparison with an

authentic sample (see below).

The unreacted 2-methylpiperidine enantiomer was recovered

as its hydrochloride salt from the extraction with hydrochloric

acid (2 M) as a colourless solid (28 mg, 97%). Following general

procedure IV this salt (26 mg, 0.19 mmol), dissolved in pyridine

(0.5 cm³) was treated with benzoyl chloride (54 mg, 0.38 mmol)

and the reaction mixture stirred for 2 h at room temperature.

After work-up, the pale yellow oil (70 mg) obtained was

puriﬁed by column chromatography over silica with light

petroleum–ethyl acetate (3 : 1) as eluent to give (R)-N-benzoyl-

127.8 and 135.7 [4 × CH(Q)], 146.2 [C–C᎐N(Q)], 158.4 [C᎐

᎐

N(Q)], 160.1 [CO(Q)] and 171.0 (CO). From comparison of the

intensities of signals at δ 4.74 and δ 4.93 the ratio of rotamers is

1.5 : 1; m/z (%) (FAB) 276 (MH^ϩ, 100), 216 (46) and 154 (51).

The unreacted 3-methylpiperidine was obtained as its hydro-

chloride salt as a colourless solid (21 mg, 78%).

2-methylpiperidine 12 as a colourless oil (27 mg, 69%); [α]_D

=

Ϫ26.7 (c 0.7, CHCl₃) ee 81% (see above).

Reaction of DAQ¹9 with 2-methylpiperidine

Reaction of DAQ¹9 with 1-phenylethylamine

General procedure III was followed using DAQ¹9 (0.1 g, 0.2

mmol) and racemic 2-methylpiperidine (40 mg, 0.4 mmol) in

dichloromethane (1 cm³) and the mixture stirred at Ϫ20 ЊC for

2 h and then at 5 ЊC for ∼30 h. After work-up, a proton NMR

spectrum of the crude reaction product showed the presence of

a ∼2.5 : 1 ratio of N-benzoyl-2-methylpiperidine to N-ethanoyl-

2-methylpiperidine, inferred from comparison of signals at

δ 8.20 [5-H(Q) for MAQ¹s 16 and 7] and δ 8.0 [2 × CH(Ph) for

MAQ¹7] respectively. Chromatotron chromatography with

light petroleum–ethyl acetate (2 : 1) as eluent gave MAQ¹7 as

colourless crystals (16 mg, 17%), identical with that isolated

previously.

General procedure III was followed using DAQ 9 (0.1 g,

0.2 mmol) and 1-phenylethylamine (49 mg, 0.4 mmol) in

deuterochloroform (0.5 cm³) and the solution reaction stirred

at Ϫ20 ЊC for 2 h and then at 5 ЊC for 12 h. A proton NMR

spectrum of the solution showed the presence of N-benzoyl-

and N-acetyl-1-phenylethylamine 20 and 21 in a ratio of 3 : 1 by

comparison of signals at δ 6.60 and 5.97 respectively with those

of authentic samples at (δ 6.61 and 5.90) and by comparison

of signals at δ 8.20 [5-H(Q) for both MAQs 7 and 16] and 8.0

[2 × CH(Ar) for MAQ¹7]. Chromatotron chromatography

with light petroleum–ethyl acetate (3 : 1) as eluent gave MAQ¹7

(17 mg, 19%) identical with that isolated previously.

Further elution with the same solvent mixture gave MAQ¹16

as a colourless oil (35 mg, 41%), identical with that isolated

previously.

Further elution with the same solvent mixture gave (S)-N-

benzoyl-2-methylpiperidine 12 as a colourless oil (17 mg, 41%),

[α]_D= ϩ30 (c 0.5, CHCl₃), ee 91% by comparison with an

authentic sample (see below).

Further elution with the same solvent mixture gave amide

(S)-20 as a colourless solid (21 mg, 46%); [α]_D= Ϫ16.3 (c 0.4,

CHCl₃), ee 82% by comparison with an authentic sample [α]_D=

Ϫ20 (c 0.4, CHCl₃).

Further elution with the same solvent mixture gave MAQ¹16

(32 mg, 41%) identical with that isolated previously.

The unreacted 2-methylpiperidine was obtained as its hydro-

chloride salt as a colourless solid (23 mg, 83%); [α]_D= ϩ0.5 (c 2,

H₂O), ee 11% by comparison with (R)-2-methylpiperidine

hydrochloride, prepared from the R-enantiomer of the amine;

[α]_D= ϩ4.6 (c 2.2, H₂O).

Further elution with ethyl acetate gave amide (R)-21 as a

colourless solid (8 mg, 23%) [α]_D= ϩ94 (c 0.5, CHCl₃) ee 76%

by comparison with authentic sample [α]_D= ϩ124 (c 0.5,

CHCl₃).

The hydrochloride salt of unreacted 1-phenylethylamine was

obtained as a colourless solid (26 mg, 80%), [α]_D= Ϫ1.1 (c 2.6,

H₂O) ee 16% by comparison with hydrochloride salt prepared

from enantiopure material [α]_D= ϩ6.6 (c 2.7, H₂O).

Reaction of DAQ¹9 with a 3-methylpiperidine

General procedure III was followed using DAQ¹9 (0.1 g,

0.2 mmol) and racemic 3-methylpiperidine (0. 04 g, 0.4 mmol)

in deuterochloroform (1 cm³) and the mixture stirred at Ϫ20 ЊC

Reaction of DAQ¹9 with valine methyl ester hydrochloride

1

for 2 h and then at 5 ЊC for 30 h. An H NMR spectrum of

A solution of racemic valine methyl ester hydrochloride (68 mg,

0.4 mmol) in water (1 cm³) was treated with aqueous sodium

hydrogen carbonate and extracted with dichloromethane

(1 cm³). After drying and following general procedure III,

the dichloromethane solution was added to a cold solution of

the solution showed the presence of 3 : 1 ratio of N-benzoyl-

3-methylpiperidine 17 to N-ethanoyl-3-methylpiperidine 18

respectively, inferred from comparison of signals at δ 8.20

[5-H(Q) of MAQ¹s 16 and 7] and δ 8.0 [2 × CH(Ph) of MAQ¹7]

J. Chem. Soc., Perkin Trans. 1, 2002, 257–274

269

View Article Online

DAQ¹9 (0.1 g, 0.2 mmol) in dichloromethane (0.5 cm³) and the

mixture stirred at Ϫ10 ЊC for 2 h and then at 5 ЊC for 12 h.

Unreacted alanine ethyl ester was extracted with aqueous

hydrochloric acid (2 M, 1 cm³). Chromatotron chromatography

of the product from the organic extract with light petroleum–

ethyl acetate (3 : 1) as eluent gave MAQ¹7 (3 mg, 3%) identical

with that isolated previously.

Reaction of DAQ¹24c with amines

(i) With 1-phenylethylamine. General procedure III was

followed using DAQ¹24c (0.1 g, 0.19 mmol) and 1-phenyl-

ethylamine (45 mg, 0.38 mmol) in dichloromethane (1 cm³) and

the solution stirred at Ϫ10 ЊC for 6 h. After work-up a proton

NMR spectrum of the crude mixture showed the presence of

only MAQ¹13 and amide 25. Chromatotron chromatography

with light petroleum–ethyl acetate (3 : 1) as eluent gave MAQ¹

13 (64 mg, 82%) identical with that isolated previously.

Further elution with the same solvent mixture gave amide

(S)-22 as a colourless solid (32 mg, 67%), mp 110–112 ЊC

(from light petroleum) (lit.¹³mp 110.5–111 ЊC) [α]_D= ϩ43.7

(c 0.65, CHCl₃), 94% ee by comparison with an enantiopure

Further elution with the same solvent mixture gave amide 25

as a colourless oil (24 mg, 77%) whose H NMR showed it to

1

authentic sample [α]_D= ϩ46.2 (c 0.65, CHCl₃), lit.¹³[α]_D

=

contain a 15 : 1 mixture of diastereoisomers 25c–25d (de 88%)

by comparison of the signals at δ 2.12 and 2.11 in the NMR

spectrum with those of authentic samples.

The (S)-hydrochloride salt of unreacted 1-phenylethylamine

was recovered as a colourless solid (26 mg, 90%), [α]_D= Ϫ5.9

(c 1.9, H₂O), 91% ee by comparison with the rotation of the

enantiopure material [α]_D= Ϫ6.4 (c 1.9, H₂O).

ϩ46.0 (c 0.4, CHCl₃); δ_H0.99 and 1.02 (6H, 2 × d, J 6.6,

CH₃CHCH₃), 2.28 (1H, dh, J, 6.6 and 4.8, CH₃CHCH₃),

3.78 (3H, s, OCH₃), 4.79 (1H, dd, J 8.3 and 4.8, CHNH), 6.63

(1H, d, J 8.3, NH), 7.4–7.57 [3H, m, CH(Ar)] and 7.78–7.85

[2H, m, CH(Ar)].

Further elution with the same solvent mixture gave MAQ¹16

(51 mg, 65%) identical with that isolated previously.

The (R)-hydrochloride salt of valine methyl ester was

recovered as a colourless solid (29 mg, 85%), [α]_D= Ϫ13.3 (c 0.9,

H₂O), ee 92% by comparison with enantiopure (S)-valine

methyl ester hydrochloride [α]_D= ϩ14.4 (c 0.9, H₂O) lit.¹⁴[α]_D=

ϩ15.7 (c 2, H₂O).

(ii) With 2-methylpiperidine. General procedure III was

followed using DAQ¹24c (0.1 g, 0.19 mmol) and racemic 2-

methylpiperidine (37 mg, 0.37 mmol) the mixture stirred at Ϫ20

ЊC for 2 h and then at 5 ЊC for 24 h. After work-up, ﬂash

chromatography with light petroleum–ethyl acetate (3 : 1) as

eluent gave MAQ¹13 (61 mg, 77%) (R_f0.38, 3 : 1 light

petroleum–ethyl acetate) identical with an authentic sample.

Further elution with the same solvent mixture gave amide 28

as a colourless oil which crystallised on standing (37 mg, 79%)

Reaction of DAQ¹24a with 1-phenylethylamine

General procedure III was followed using DAQ¹24a (80 mg,

0.15 mmol) and 1-phenylethylamine (36 mg, 0.30 mmol) and

the solution stirred at Ϫ10 ЊC for 2 h and then at 5 ЊC for 35 h.

After work-up, a proton NMR spectrum of the crude product

showed presence of only MAQ¹13 and amide 25. Chromato-

tron chromatography with light petroleum–ethyl acetate (3 : 1)

as eluent gave MAQ¹13 (43 mg, 68%) identical with that

isolated previously.

1

whose H NMR (400 MHz, at 50 ЊC) showed it to contain a

∼45 : 1 mixture of diastereoisomers 28a–28b (de 95%) from

comparison of the intensities of signals at δ 2.11 and 2.12.

The unreacted 2-methylpiperidine enantiomer was recovered

as the colourless, solid hydrochloride (22 mg, 88%) and, follow-

ing general procedure IV, was dissolved in pyridine and deriva-

tised by reaction with (S)-2-acetoxypropanoyl chloride to

give amide 28 (24 mg, 81%) whose ¹H NMR (at 400 MHz and

50 ЊC) showed it to contain a 17 : 1 mixture of diastereoisomers

28b–28a (de 89%) by comparison of the signals at δ 2.10 and

2.09 respectively with those of authentic samples.

Further elution with the same solvent mixture gave amide 25

1

as a colourless oil (22 mg, 63%) whose H NMR spectrum

showed it to contain a 8 : 1 mixture of diastereoisomers 25a–

25b (R,R : S,R) from comparison of signals at δ 2.15 and 2.16

with those of authentic samples.

The (S)-hydrochloride salt of unreacted 1-phenylethylamine

was recovered as a colourless solid (19 mg, 81%), [α]_D= Ϫ4.6

(c 1.9, H₂O) (70% ee) by comparison with the rotation of an

enantiopure sample [α]_D= ϩ6.5 (c 2, H₂O).

(iii) With 2-propylpiperidine. General procedure III was

followed using DAQ¹24c (0.1 g, 0.19 mmol), in dichloro-

methane with racemic 2-propylpiperidine (48 mg, 0.38 mmol)

and the reaction mixture stirred at 5 ЊC for 24 h. After work-up,

a proton NMR spectrum of the crude mixture showed the

presence of only MAQ¹13 and amide 29. Separation of these

compounds was carried out using ﬂash chromatography with

light petroleum–ethyl acetate (3 : 1) as eluent to give MAQ¹13

(65 mg, 82%) identical with that isolated previously.

Further elution with the same solvent mixture gave amide 29

as a colourless oil which crystallised on standing (32 mg, 71%)

whose ¹H NMR spectrum showed it to contain a 50 : 1 mixture

of diastereoisomers 29a–29b (de 96%) from comparison of the

intensities of signals at δ 2.11 and 2.13 in the NMR spectrum at

400 MHz and 50 ЊC. Assignment of absolute conﬁguration to

29a and 29b (S,S and S,R respectively) was based on the R-

conﬁguration of the unreacted 2-propylpiperidine enantiomer,

recovered as the colourless hydrochloride salt (27 mg, 87%); mp

219–221 ЊC (lit.¹⁵218 ЊC); δ_H1.0 (3H, t, J 6.7, CH₃), 1.3–2.2

(10H, m, 5 × CH₂), 2.8–3.1 (2H, m, NCH₂), 3.50 (1H, br d,

J 6.7, CHN) and 9.28 and 9.60 (2H, 2 × s br, NHH); [α]_D= Ϫ6.7

(c 0.42, ethanol), lit.¹⁰[α]_D= Ϫ7.3 (c 0.33, ethanol). The ee of

this salt is therefore ∼90%.

Reaction of DAQ¹24b with 1-phenylethylamine

General procedure III was followed using DAQ¹24b (0.1 g,

0.19 mmol) and 1-phenylethylamine (0.046 g, 0.38 mmol) and

the reaction mixture stirred at Ϫ10 ЊC for 2 h and then at 5 ЊC

for 24 h. After work-up, a proton NMR spectrum of the crude

product showed the presence of amides 25 and 26 in a 2 : 1 ratio

from integration comparison of signals at δ 6.42 and 5.81 in

authentic samples (see below) together with the corresponding

MAQ¹s 13 and 27.⁵Chromatotron chromatography with light

petroleum–ethyl acetate (3 : 1) as eluent gave MAQ¹13 (17 mg,

22%) identical with that isolated previously.

Further elution with the same solvent mixture gave amide 25

1

as a colourless solid (19 mg, 42%) whose H NMR showed

it to contain a ≥12 : 1 mixture of diastereoisomers 25d–25c

(S,S : R,S) (de 85%) by comparison of the intensities of signals

at δ 2.16 and 2.17 with those of authentic samples (see below).

Further elution with the same solvent mixture gave a mixture

of MAQ¹27 and amide 26 (∼0.075 g) from comparison of

signals at δ 5.50 to 5.12.

The hydrochloride salt of unreacted 1-phenylethylamine was

recovered as a colourless solid (24 mg, 80%), [α]_D= Ϫ0.5 (c 2,

H₂O) 7.7% ee by comparison with the rotation of an enantio-

Reaction of DAQ¹24d with 1-phenylethylamine

General procedure III was followed using DAQ¹24d (0.1 g, 0.19

mmol) and 1-phenylethylamine (46 mg, 0.38 mmol) and the

mixture stirred at Ϫ10 ЊC for 8 h. After work-up, an ¹H NMR

spectrum of the crude reaction mixture showed the presence of

pure sample of (R)-1-phenylethylamine hydrochloride, [α]_D

ϩ6.5 (c 2, H₂O).

=

270

J. Chem. Soc., Perkin Trans. 1, 2002, 257–274

View Article Online

only MAQ¹13 and amide 25. Chromatotron chromatography

with light petroleum–ethyl acetate (3 : 1) as eluent gave MAQ¹

13 (63 mg, 80%) identical with an authentic sample.

Further elution with the same solvent mixture gave amide 25

as a colourless oil whose ¹H NMR spectrum showed it to con-

tain an 8 : 1 (78% de) mixture of diastereoisomers 25b –25a by

comparison of signals at δ 2.13 and 2.12 with those of authentic

samples.

Enantiopurity assay of DAQ²31a

The reaction above was repeated under the same conditions

but using pure (S)-alanine ethyl ester hydrochloride (98 mg,

0.64 mmol) and DAQ²31a (62 mg, 0.13 mmol). After

separation using chromatotron chromatography as above, a

proton NMR spectrum of the crude product showed the pres-

ence of only one amide ester diastereoisomer 33b (S,S).

The (R)-hydrochloride salt of unreacted 1-phenylethylamine

was recovered as a colourless solid (26 mg, 87%) [α]_D= ϩ5

(c 2, H₂O), 77% ee by comparison with enantiopure material

[α]_D= ϩ6.5 (c 2, H₂O).

Reactions of DAQ²31b

(i) With 1-phenylethylamine. General procedure III was

followed using DAQ²31b (80 mg, 0.17 mmol) and racemic 1-

phenylethylamine (40 mg, 0.33 mmol). After work-up, a proton

NMR spectrum of the crude mixture showed the ratio of

amides 21 to 25 was ∼1 : 1 respectively by comparison of signals

at δ 6.10 and 6.41 with those of authentic samples and con-

ﬁrmed by product isolation (see below). Chromatotron chroma-

tography with light petroleum–ethyl acetate (2 : 1) as eluent

gave MAQ²30 as a colourless solid (27 mg, 37%) identical with

that isolated previously.

Reactions of DAQ²31a

(i) With 1-phenylethylamine. General procedure III was

followed using DAQ²31a (87 mg, 0.18 mmol) and 1-phenyl-

ethylamine (44 mg, 0.36 mmol) at Ϫ10 ЊC for 10 h. After work-

up, a proton NMR spectrum of the crude reaction mixture

showed the ratio of amides 21 to 25 was >23 : 1 from com-

parison of signals at δ 5.90 and 6.39 with those of authentic

samples (δ 590 and 6.35) respectively (see below). Chromato-

tron chromatography with light petroleum–ethyl acetate (2 : 1)

as eluent gave MAQ²32 as a colourless solid (48 mg, 61%),

identical with that isolated previously.

Further elution with the same solvent mixture gave amide 25

as a colourless oil (13 mg, 33%) whose H NMR spectrum

1

showed it to contain a 9 : 1 mixture of diastereoisomers 25c–

25d (S,R)–(S,S) from comparison of signals at δ 2.13 and 2.12

with those of authentic samples (see below).

Further elution with the same solvent mixture gave MAQ²30

as a colourless solid (3 mg, 4%) identical with that isolated

previously.

Further elution with ethyl acetate gave (R)-N-(1-phenyl-

ethyl)acetamide 21 (19 mg, 63%); [α]_D= ϩ116 (c 0.5, CHCl₃),

ee 93% by comparison with an authentic sample [α]_D= ϩ124

(c 0.5, CHCl₃); lit.¹⁶[α]_D= ϩ129.5 (c 1, CHCl₃).

Unreacted (S)-1-phenylethylamine was recovered as its hydro-

chloride salt as a colourless solid (25 mg, 87%); [α]_D= Ϫ5

(c 2, H₂O), ee 77%, by comparison with an authentic sample

[α]_D= ϩ6.5 (c 1, H₂O).

Further elution with the same solvent mixture gave MAQ²32

as a colourless solid (21 mg, 34%) identical with that isolated

previously.

Further elution with ethyl acetate gave (S)-N-(1-phenyl-

ethyl)acetamide 21 (9 mg, 33%); [α]_D= Ϫ100 (c 0.4, CHCl₃)

ee 85% by comparison with an authentic sample [α]_D= ϩ117

(c 0.4, CHCl₃), prepared as described below.

Unreacted 1-phenylethylamine was recovered as hydro-

chloride salt as a colourless solid (21 mg, 80%); [α]_D= Ϫ0.5

(c 2.1, H₂O), ee 7.7%, by comparison with the rotation of an

enantiopure sample [α]_D= ϩ6.5 (c 2, H₂O).

(ii) With alanine ethyl ester. A solution of racemic alanine

ethyl ester hydrochloride (0.114 g, 0.74 mmol) was treated with

excess sodium hydrogen carbonate and the free amino acid ester

extracted into dichloromethane (2 cm³). After drying, and

following general procedure III, the dichloromethane solution

was added to a cold solution of DAQ²31a (90 mg, 0.19 mmol)

in dichloromethane (0.5 cm³) and the mixture stirred at Ϫ10 ЊC

for ∼10 h. Unreacted alanine ethyl ester was extracted with

aqueous hydrochloric acid (2 M, 1 cm³). The ratio of amides 33

and 34 in the crude product was ∼1 : 1 by comparison of the

signals at δ 6.67 and 6.13 with those of authentic samples

(δ 6.68 and 6.13) respectively. Chromatotron chromatography

of the crude product with light petroleum–ethyl acetate (2 : 1)

as eluent gave MAQ²32 (32 mg, 39%) identical with that

isolated previously.

(ii) With alanine ethyl ester. A solution of racemic alanine

ethyl ester hydrochloride (60 mg, 0.39 mmol) in water (1 cm³)

was converted to free amine by addition of aqueous sodium

hydrogen carbonate and extracted into dichloromethane

(1 cm³). After drying, and following general procedure III, the

dichloromethane solution was added to a cold solution of

DAQ²31b (0.09 g, 0.19 mmol) in dichloromethane (0.5 cm³)

and the mixture stirred at Ϫ10 ЊC for 6 h. The ratio of amides

33 to 34 was ∼20 : 1 based on yields of recovered MAQ²30

and MAQ²32 (see below). Separation of these compounds

was carried out using a chromatotron with light petroleum–

ethyl acetate (2 : 1) as eluent and gave MAQ²32 (3.5 mg,

3%) as colourless crystals (R_f0.32), mp 183–185 ЊC (from

light petroleum) identical with an authentic sample (see

above).

Further elution with the same solvent mixture gave amide 33

as a colourless oil (16 mg, 36%) whose H NMR spectrum

Further elution with the same solvent mixture gave amide 33

as a colourless oil (34 mg, 77%) whose H NMR spectrum

1

showed it to contain a 1 : 1 mixture of diastereoisomers 33a–

33b (R,S–S,S) from comparison of the intensities of signals

inter alia at δ 5.12 and 5.16 with those in the NMR spectra of

an authentic mixture (see below).

showed it to contain a ∼30 : 1 mixture of diastereoisomers 33b–

33a (S,S–R,S) from comparison of the intensities of signals at

δ 5.23 and 5.19 with those at δ 5.16 and 5.12 in the spectra of

authentic samples (Found: MH^ϩ232.1185. C₁₀H₁₇NO₅requires

MH^ϩ232.1185); ν_max/cm^Ϫ13440s, 1740s, 1680s, 1525s and

1455s; δ_H1.30 (3H, t, J 7.1, CH₃CH₂O), 1.42 (3H, d, J 7.1,

CH₃CHN), 1.47 (3H, d, J 6.9, CH₃CHOAc), 2.17 (3H, s,

CH₃CO₂), 4.23 (2H, q, J 7.1, CH₃CH₂O), 4.56 (1H, m,

CH₃CHNH), 5.23 (1H, q, J 6.9, CH₃CHOAc) and 6.68 (1H, d

br, J 6.4, NH); minor diastereoisomer (observable signal) 5.22

(1H, q, J 6.9, CH₃CHOAc).

Further elution with the same solvent mixture gave MAQ²30

as colourless crystals (23 mg, 33%) identical with an authentic

sample.

Further elution with ethyl acetate gave (R)-amide 34 as a

colourless oil (7 mg, 23%); δ_H1.29 (3H, t, J 7.1, CH₃CH₂O),

1.39 (3H, d, J 7.4, CH₃CHN), 2.02 (3H, s, CH₃CO), 4.21 (2H,

q, J 7.1, CH₃CH₂O), 4.58 (1H, dq, J 7.4 and 7.0, CH₃CHNH)

and 6.1 (1H, br, NH), [α]_D= Ϫ15.7 (c 0.7, CDCl₃), ee > 97% by

comparison with an authentic sample of the (S)-enantiomer

[α]_D= ϩ15.7 (c 0.7, CDCl₃) (see below).

Further elution with the same solvent mixture gave MAQ²30

as colourless crystals (49 mg, 71%); (R_f0.34; 1 : 1 ethyl acetate–

petroleum ether) identical with an authentic sample (see above).

The hydrochloride salt of unreacted (R)-alanine ethyl ester

was recovered as a colourless solid (49 mg, 86%), [α]_D= Ϫ7.5

The hydrochloride salt of alanine ethyl ester was recovered as

a colourless solid (51 mg, 79%).

J. Chem. Soc., Perkin Trans. 1, 2002, 257–274

271

View Article Online

(c 2, MeOH) ee 93% by comparison with an authentic sample

[α]_D= Ϫ8 (c 2, MeOH).

salt (18 mg) in water (1 cm³) was basiﬁed with sodium hydrox-

ide (2 M), the 2-methylpiperidine was extracted with ether (2 ×

1.5 cm³), and the combined extracts were dried over powdered

sodium hydroxide and the ether removed carefully under

reduced pressure to give (R)-2-methylpiperidine (9 mg, 69%)

[α]_D= Ϫ7.8 (c 0.9, ethanol) ee 89% by comparison with the

speciﬁc rotation of an enantiopure sample [α]_D= Ϫ8.7 (c 1.5,

ethanol) (see below).

Reaction of DAQ 35 with 1-phenylethylamine

General procedure III was followed using DAQ 35 (sub-

sequently shown to be enantio-impure; see below) (75 mg,

0.19 mmol) and 1-phenylethylamine (45 mg, 0.37 mmol) and

the solution stirred at Ϫ20 ЊC for 2 h and then at 0 ЊC for 10 h.

After work-up, chromatotron chromatography of the product

with light petroleum–ethyl acetate (1 : 1) as eluent gave amide

25 as a colourless oil (33 mg, 75%) as a 2 : 1 ratio of diastereo-

isomers 25c–25d by comparison of the signals at δ 2.11 and

2.10 with those of authentic samples

Further elution with the same solvent mixture gave MAQ 39

(30 mg, 60%) (R_f0.36) identical with authentic material.⁵

The hydrochloride salt of unreacted (S)-amine was obtained

as a colourless solid (17 mg, 76%), [α]_D= Ϫ2.2 (c 1.7, H₂O)

ee 34% by comparison with the speciﬁc rotation of authentic

sample [α]_D= Ϫ6.5 (c 2, H₂O).

(iv) With 2-methylpiperidine (5 eq.). Repetition of the experi-

ment above using 5 eq. of 2-methylpiperidine gave amides 27a

and 27b in a ratio of ∼40 : 1 (de 95%).

(v) With 2-propylpiperidine (4 eq.). As in the previous experi-

ment, a solution of DAQ²37a (0.1 g, 0.18 mmol) and 2-

propylpiperidine (coniine) (0.092 g, 4 eq.) in dichloromethane

(1 cm³) was stirred at 5 ЊC for ∼16 h. After work-up, chroma-

totron chromatography with light petroleum–ethyl acetate

(2 : 1) as eluent gave MAQ²32 (65 mg, 82%) (R_f0.32).

Further elution with the same solvent mixture gave N-((S)-2-

acetoxypropanoyl)-2-propylpiperidine 29 as a colourless oil

(31 mg, 71%) as a ∼45 : 1 ratio of diastereoisomers 29a–29b

(S,S–R,S) (96% de) by comparison of signals at δ 2.094 and

2.107 with those of the mixture isolated previously (using

DAQ¹24c).

Enantiopurity assay of DAQ 35

The reaction was repeated under the same conditions but using

pure (S)-1-phenylethylamine (72 mg, 0.60 mmol) and DAQ 35

(60 mg, 0.15 mmol). An NMR spectrum of the crude product

showed the presence of an 8 : 1 ratio of amide diastereoisomers

25c–25d; DAQ 35 is therefore of 77% ee.

Reaction of N-[(S)-2-acetoxypropanoyl-N-ethanoylamino]-

phthalimide (DAP) 41 with 1-phenylethylamine

Reaction of DAQ²37a with amines

General procedure III was followed using DAP 41 (0.1 g, 0.31

mmol) and racemic 1-phenylethylamine (76 mg, 0.62 mmol)

in dichloromethane and the reaction mixture stirred for 6 h at

(i) With 1-phenylethylamine (2 eq.). General procedure III

was followed using DAQ²37a (76 mg, 0.14 mmol) and racemic

1-phenylethylamine (33 mg, 0.27 mmol) in dichloromethane

(0.5 cm³) and the solution stirred at Ϫ20 ЊC for 2 h. After work-

up, chromatotron chromatography of the product with light

petroleum–ethyl acetate (2 : 1) as eluent gave MAQ²32 (51 mg,

85%) (R_f0.32), identical with that prepared above.

Further elution with the same solvent mixture gave amide 25

as a colourless oil (25 mg, 75%) as a 5 : 1 ratio of diastereo-

isomers 25c–25d (67% de) by NMR spectroscopy and com-

parison with authentic samples as previously.

1

Ϫ10 ЊC (0.5 cm³). After work-up, a H NMR spectrum of the

crude mixture showed the ratio of amides 21 to 25 was 1.2 : 1

respectively by comparison of signals at δ 5.90 and 6.39 with

those of authentic samples. Chromatotron chromatography

with light petroleum–ethyl acetate (1 : 1 containing 2% metha-

nol) as eluent gave a 52 : 48 ratio of amides 25d–25c by com-

parison with authentic samples (see below).

Further elution with the same solvent mixture gave a pure

sample of amide 21 (15 mg, 30%) [α]_D= ϩ8.9 (c 0.4, CHCl₃),

ee 7.8% by comparison with an authentic sample.

The hydrochloride salt of unreacted (S)-amine was obtained

as a colourless solid (18 mg, 84%), [α]_D= Ϫ3.9 (c 1.5, H₂O),

ee 60% comparison with an authentic sample [α]_D= Ϫ6.5 (c 1.5,

H₂O).

Enantiopurity assay on DAP 41

The reaction above was repeated under the same conditions

but using pure (S)-1-phenylethylamine (0.085 g, 0.7 mmol)

and DAP 41 (0.11 g, 0.35 mmol). After work-up, a H NMR

spectrum of the crude product showed the same ratio of

amides 21–25 (1.2 : 1 respectively). After separation using

chromatotron as above, a proton NMR spectrum showed the

presence of only one diastereoisomer (25d) of amide 25: DAP

41 is therefore enantiopure.

1

(ii) With 1-phenylethylamine (5 eq.). Repetition of the

experiment above using 5 eq. of 1-phenylethylamine gave amide

25 as a colourless oil as an 8 : 1 ratio of diastereoisomers

25c–25d (78% de).

Enantiopurity assay on DAQ²37a

The reaction above was repeated under the same conditions but

Preparation of authentic samples

using pure (R)-1-phenylethylamine (22 mg, 0.18 mmol) and

1

DAQ²37a (33 mg, 0.06 mmol). An H NMR spectrum of the

General procedure V for N-acylation of amines. To a stirred

solution of the amine (0.2 g) in dichloromethane was added

pyridine (2 eq.) followed by dropwise addition of the acid

chloride (1.3 eq.). After 1 h, further dichloromethane (10 ml)

was added, the solution was then washed successively with

saturated aqueous sodium hydrogen carbonate, hydrochloric

acid (2 M), water, then dried and the solvent removed under

reduced pressure to give the amide.

crude product showed only one signal for the CH₃CO₂of amide

25c showing DAQ²37a is enantiopure.

(iii) With 2-methylpiperidine (2 eq.). General procedure III

was followed using DAQ²37a (86 mg, 0.16 mmol) and racemic

2-methylpiperidine (32 mg, 0.32 mmol) in dichloromethane

(0.5 cm³) and the solution stirred at Ϫ30 ЊC for 3 h and then at

1

5 ЊC for ∼24 h. After work-up, a H NMR spectrum of the

The following were prepared by the above method.

crude mixture (at 50 ЊC and 400 MHz) showed the presence of

N-2-[(S)-acetoxypropanoyl]-2-methylpiperidine 28a–28b as a

33 : 1 ratio of diastereoisomers (94% de) by comparison of the

signals at δ 2.096 and 2.103 with those of authentic samples

(see below).

(i) N-Acetylpyrrolidine as a pale yellow oil (84%), δ_H(300

MHz) 1.89–2.0 (4H, m, 2 × CH₂), 2.1 (3H, s, CH₃CO) and 3.42

(4H, struct. m, 2 × CH₂N), lit.^{17 1}H NMR (300 MHz, CDCl₃)

δ 1.86–2.0 (m, 4H), 2.06 (s, 3H) and 3.45 (dd, J 7.1, 16.0, 4 H).

(ii) N-Acetylpiperidine as a pale yellow oil (83%), δ_H(300

MHz) 1.45–1.65 (6H, m, 3 × CH₂), 2.05 (3H, s, CH₃CO) and

3.37 and 3.54 (4H, 2 × t, J 5.6, 2 × CH₂N); lit.^{18 1}H NMR

The unreacted 2-methylpiperidine was recovered as the solid,

colourless hydrochloride salt (18 mg, 90%). A solution of this

272

J. Chem. Soc., Perkin Trans. 1, 2002, 257–274

View Article Online

(CDCl₃) δ 1.61 (m, 6H, 3 × CH₂), 2.06 (s, 3H, COCH₃) and 3.50

(m, 4H, 2 × CH₂).

from (S)-2-methylpiperidine (0.1 g scale) [prepared by reso-

lution of the racemic amine with (Ϫ)-mandelic acid according to

the literature method⁷[α]_D= ϩ8.9 (c 2, ethanol), lit.⁷[α]_D= 7.2

(c 6, 95% ethanol)] and (S)-2-acetoxypropanoyl chloride. After

work-up, column chromatography of the crude product (0.12 g)

over silica gave amide 28a as a colourless oil (0.15 g, 75%).

(iii) N-Benzoylpyrrolidine as a light brown oil (76%), δ_H1.72–

1.96 (4H, m, 2 × CH₂), 3.40 and 3.60 (4H, 2 × t, J 6.6,

2 × CH₂N), 7.3–7.4 [3H, m, 3 × CH(Ph)] and 7.45 [2H, m,

2 × CH(Ph)]; lit.¹⁹δ_H1.80–2.00 (4H, m, 2 × CH₂), 3.40, 3.64

(2 × 2H, 2 × t, J 6.4, 2 × CH₂N), 7.35–7.50 (5H, m, ArH).

(iv) N-Benzoylpiperidine as a pale yellow oil (81%), δ_H1.3–

1.9 (6H, m, 3 × CH₂), 3.30 and 3.70 (4H, 2 × s, br, 2 × CH₂N)

and 7.3–7.4 [5H, m, CH(Ph)]; lit.²⁰7.42–7.36 (m, 5H), 3.72 (br s,

2H), 3.34 (br s, 2H), 1.68 (m, 4H) and 1.52 (m, 2H).

(v) (S)-N-Benzoyl-1-phenylethylamine 20 from (S)-1-

phenylethylamine as a pale yellow oil (0.30 g, 81%), δ_H1.55 (3H,

d, J 6.9, CH₃CHNH), 5.3 (1H, q, CH₃CHNH), 6.61 (1H, br d,

J 6.9, NH) and 7.1–7.9 [10H, m, CH(Ph)]; [α]_D= Ϫ20 (c 0.4,

CHCl₃), lit.²¹[α]_D= Ϫ20.07 (1.02, CHCl₃).

(xii)

N-[(S)-2-Acetoxypropanoyl]-(R)-2-methylpiperidine

28b from (R)-2-methylpiperidine (0.1 g scale) [prepared by reso-

lution of the racemic amine with (ϩ)-mandelic acid according

to the literature method⁷[α]_D= Ϫ8.5 (c 0.5, methanol)] and (S)-

2-acetoxypropanoyl chloride. Chromatography of the product

as above gave amide 28b as a colourless oil (0.16 g, 80%). For a

1 : 1 mixture of 28a and 28b (Found: MH^ϩ214.1655. C₁₃H₂₃NO₃

requires MH^ϩ214.1655); ν_max/cm^Ϫ11704s, 1645s, 1510s, 1375s

and 1250s; δ_H(mixture of N–CO rotamers) (2-methyl axial;²²

rotamer A, C᎐O/C cis, rotamer B, C᎐O/C cis), 1.25–1.45 (6H,

᎐

2

6

(vi) (R)-N-Acetyl-1-phenylethylamine 21 from (R)-1-

phenylethylamine as a light yellow oil (0.23 g, 85%), δ_H1.29

(3H, d, J 7.0, CH₃CHNH), 1.76 (3H, s, CH₃CO), 4.9 (1H, m,

CH₃CHNH), 5.90 (1H, s br, NH) and 7.0–7.2 [5H, m, CH(Ph)];

m, CH₃CHOAc and CH₃CHN), 1.46–1.80 (6H, m, 3 × CH₂),

2.1 (3H, s, CH₃CO₂), 2.65 [0.5H, ddd, J ∼12.5, ∼12.5 and ∼2,

H₆-ax. (A or B)], 3.08 [0.5H, ddd, J ∼12.5, 12.5 and ∼2, H₆-ax.

(B or A)], 3.52 (0.5H, br d, J ∼12, H₆-eq. A), 3.95 (0.5H, br m,

H₂-eq. B), 4.37 (0.5H, dd, J ∼12.5 and ∼4, H₆-eq. B), 4.83 (0.5H,

m, H₂-eq. A) and 5.33 [1H, 2 × q, J ∼7, CH₃CH (A and B)];

from the signals at δ 2.65 and 3.08 the ratio of N–CO rotamers

was 1 : 1. Examination of the product by NMR spectroscopy

(400 MHz at ϩ50 ЊC) showed it to be a 1 : 1 mixture of diastereo-

isomers from comparison of signals at δ_H2.11 and 2.12. As

for 25c/d above δ for both signals varied depending on

concentration/spectrum reproducibility but ∆δ was maintained

at 0.01 ppm. δ_C15.8, 17.2 and 21.2 (3 × CH₃), 19.1 (3 × CH₂),

45.0 (CHN), 67.5 (CH₃CHOAc) and 169.1 and 171.0 (2 × CO).

(xiii) (S)-2-Acetoxypropanoyl-2-propylpiperidine 29 from

racemic 2-propylpiperidine and (S)-2-acetoxypropanoyl

chloride. After work-up, column chromatography over silica

with light petroleum–ethyl acetate (1 : 1) as eluent gave amide 29

as a colourless oil (0.164 g, 86%) as a mixture of diastereo-

isomers (Found: MH^ϩ242.1756. C₁₃H₂₃NO₃requires MH^ϩ

242.1756); ν_max/cm^Ϫ11740s, 1643s, 1510s, 1375s and 1250s;

δ_H(mixture of N–CO rotamers) (2-propyl axial; rotamer A,

[α]_D= ϩ117 (c 0.4, CHCl₃); 124 (c 0.5, CHCl₃) lit.¹⁶[α]_D

ϩ129.5 (c 1, CHCl₃).

=

(vii) (2S,1ЈR)-2-Acetoxy-N-(1-phenylethyl)propanamide 25c

using (R)-1-phenylethylamine and (S)-2-acetoxypropanoyl

chloride as colourless crystals (82%); δ_H1.55 and 1.60 (6H,

2 × d, J 6.9, 2 × CH₃), 2.12** (3H, s, CH₃CO₂), 5.11–5.30 (2H,

m, 2 × CH), 6.35 (1H, br, NH) and 7.30–7.48 [5H, m, 5 ×

CH(Ph)].

(viii) (2S,1ЈS)-2-Acetoxy-N-(1-phenylethyl)propanamide 25d

using (S)-1-phenylethylamine and (S)-2-acetoxypropanoyl

chloride as a colourless oil (92%); δ_H1.46 and 1.51 (6H, 2 × d,

J 6.9, 2 × CH₃), 2.11 (3H, s, CH₃CO₂), 5.08–5.20 (2H, m,

2 × CH), 6.40 (1H, br d, J 7.4, NH) and 7.20–7.38 [5H, m,

5 × CH(Ph)].

(ix) (R)-N-(2-Methylpropanoyl)-1-phenylethylamine 26

from (R)-1-phenylethylamine as a colourless solid (from ethyl

acetate) (84%) δ_H1.14 (6H, 2 × d, J 6.6, CH₃CHCH₃), 1.47

(3H, J 6.9, CH₃CH), 2.34 (1H, h, CH₃CHCH₃), 5.12 (1H,

m, CH₃CHOAc), 5.81 (1H, br s, NH) and 7.2–7.38 [5H, m,

5 × CH(Ar)]; [α]_D= ϩ76.7 (c 0.3, CHCl₃).

(x) N-[(S)-2-Acetoxypropanoyl]alanine ethyl ester 33a and

33b from racemic alanine ethyl ester hydrochloride and (S)-2-

acetoxypropanoyl chloride. After work-up the pale yellow oil

(0.41 g) obtained was puriﬁed by column chromatography over

silica with light petroleum–ethyl acetate (2 : 1) as eluent to give

amides 33a and 33b (ratio 1 : 1) as a colourless oil (0.27 g, 90%)

(Found: MH^ϩ232.1185. C₁₀H₁₇NO₅requires MH^ϩ232.1185);

ν_max/cm^Ϫ13440s, 1740s, 1680s, 1525s and 1455s; δ_H1.22 and 1.23

(6H, 2 × t, J 7.1, CH₃CH₂O), 1.35 and 1.37 (6H, 2 × d, J 7.1,

CH₃CHN), 1.40 and 1.41 (6H, 2 × d, J 6.9, CH₃CHOAc), 2.09

(6H, s, CH₃CO₂), 4.14 and 4.15 (4H, 2 × q, J 7.1, CH₃CH₂O),

4.49 (2H, m, J 7.1, CH₃CHNH), 5.12 and 5.16 (2H, 2 × q, J 6.9,

CH₃CHOAc) and 6.68 (2H, d, J ∼6, NH); δ_C14.4, 18.1, 18.7

and 21.4 (4 × CH₃), 48.2 (CH₃CHNH), 62.0 (CH₃CH₂O), 70.8

(CHOAc) and 169.8, 170.3 and 173.1 (3 × CO); m/z (%) (FAB)

232 (MH^ϩ, 100) and 190 (24).

An authentic sample of 33b was prepared as described above

from (S)-alanine ethyl ester hydrochloride as a colourless oil

δ_H1.29 (3H, t, J 7.1, CH₃CH₂O), 1.44 (3H, d, J 7.1, CH₃CHN),

1.48 (3H, d, J 6.9, CH₃CHOAc), 2.16 (3H, s, CH₃CO₂), 4.21

(2H, q, J 7.1, CH₃CH₂O), 4.56 (1H, q, J 7.1, CH₃CHNH), 5.19

(1H, q, J 6.9, CH₃CHOAc) and 6.77 (1H, d, br, J 6.4, NH).

(xi) N-[(S)-2-Acetoxypropanoyl]-(S)-2-methylpiperidine 28a

C᎐O/C cis, rotamer B, C᎐O/C cis) 0.85–1.05, 1.2–1.85 (13H,

᎐

2

6

2 × m, 5 × CH₂, CH₃), 2.12 (3H, s, CH₃CO₂), 2.66 (0.25H,

br dd, J ∼12.5 and 12.5, H₆-ax. B), 3.17 (0.75H, ddd, J 12.5,

12.5 and ∼2, H₆ax.-A), 3.58 (0.75H, br d, J 12, H₆-aq. A), 3.93

(0.25H, m, H₂-eq. B), 4.48 (0.25H, br d, J ∼12, H₆-eq. B), 4.72

(0.75H, m, H₂-eq. A), 5.37 (0.75H, q, J 7, CH₃CH, A) and 5.48

(0.25H, q, J 7, B) (ratio A : B is 3 : 1). Examination of the

product by NMR spectroscopy (400 MHz at ϩ50 ЊC) showed it

to be a 1 : 1 mixture of diastereoisomers from comparison of

signals at δ_H2.11 and 2.13; δ_C14.4, 17.1 and 19.6 (3 × CH₃),

19.2, 25.7, 26.4, 26.6, 28.3 and 28.5 (6 × CH₂), 48.8 (CHN), 68.9

(CH₃CHOAc) and 169.2 and 170.8 (2 × CO); m/z (%) (FAB)

242 (MH^ϩ, 100), 200 (48) and 154 (24).

(xiv) (S)-N-Benzoyl-2-methylpiperidine 12 from (S)-(ϩ)-2-

methylpiperidine [prepared by resolution of the racemic amine

with (Ϫ)-mandelic acid according to the literature method⁷

[α]_D= ϩ8.9 (c 2, ethanol), lit.⁷[α]_D= 7.2 (c 6, 95% ethanol)] as a

colourless oil (80%) which solidiﬁed after setting aside for two

days, mp 43–45 ЊC (from light petroleum) (lit.²²42–43 ЊC) [α]_D=

ϩ32.9 (c 0.8, CHCl₃). δ_H(1 : 1 mixture of N–CO rotamers)

(2-methyl axial; rotamer A, C᎐O/C cis, rotamer B, C᎐O/C

᎐

2

6

cis) 1.14 and 1.25 (3H, d, J 7, CH₃CHN), 1.40–1.83 (6H, m,

3 × CH₂), 2.91 [0.5H, ddd, J ∼12, ∼12 and ∼2, H₆-ax. (A or B)],

3.13 [0.5H, ddd, J ∼12, ∼12 and ∼2, H₆-ax. (B or A)], 3.55 (0.5H,

br d, J ∼12, H₆-eq. A), 4.01 (0.5H, m, H₂-eq. B), 4.60 (0.5H, br

d, J ∼12, H₆-eq. B) and 5.06 (0.5H, m, H₂-eq. A).

** The signals at δ 2.12 in 25c and δ 2.11 in 25d used for diastereoisomer

diﬀerentiation, both varied depending on concentration/spectrum

reproducibility but the ∆δ was 0.010 0.002 ppm with 25d at higher

and 25c at lower chemical shift as shown by addition of incremental

amounts of either one of them and observing the increase in the corres-

ponding resonance. For spectroscopic comparison 25c ≡ 25b and 25d ≡

25a.

(xv) N-Acetyl-2-methylpiperidine 15 as a colourless liquid

(84%), δ_H(CDCl , 400 MHz, Ϫ40 ЊC) (1 : 1 mixture of NC᎐O

᎐

3

rotamers) (2-methyl axial; rotamer A C᎐O/C cis; rotamer B

᎐

2

C᎐O/C cis) 1.15 and 1.25 (3H, 2 × d, J 7.0, CH CHN), 1.3–1.8

᎐

6

3

(6H, m, 3 × CH₂), 2.12 and 2.15 (3H, 2 × s, CH₃CO), 2.67 and

3.19 [1H, 2 × ddd, J 16.0, 14.0 and 3.0, H₆-ax. (A and B)], 3.61

J. Chem. Soc., Perkin Trans. 1, 2002, 257–274

273

View Article Online

Table 1 Crystal data

9

11

31a

31b

37a

Formula

M

System

Space group

a/Å

b/Å

c/Å

C₂₇H₃₅N₃O₄Si

493.67

Monoclinic

P2₁

9.649(2)

9.283(6)

15.152(5)

90

C₂₃H₂₉N₃O₃Si

423.58

Orthorhombic

P2₁2₁2₁

8.889(2)

11.391(2)

23.399(4)

90

C₂₈H₂₅N₃O₅

483.51

Monoclinic

P2₁/n

10.167(2)

25.688(3)

10.311(2)

90

C₂₈H₂₅N₃O₅

483.51

Orthorhombic

P2₁2₁2₁

9.280(2)

14.826(1)

18.246(2)

90

C₃₁H₂₉N₃O₇

555.57

Orthorhombic

P2₁2₁2₁

11.315(3)

14.210(2)

17.368(5)

90

α/Њ

β/Њ

96.69(1)

90

1347.9(10)

293

90

2369.1(8)

200

4

115.86(1)

90

2423.3(7)

200

90

2510.4(6)

200

4

90

γ/Њ

u/Å^Ϫ3

T /K

Z

2792.4(11)

190

4

2

4

µ(Mo-Kα)/mm^Ϫ1

0.123

0.126

0.092

0.089

0.095

Reﬂn. measured

Reﬂn. independent

Rint

3369

2842

0.020

0.045

2759

2687

0.015

0.054

5740

4685

0.049

0.107

3166

3088

0.021

0.059

3891

3692

0.029

0.054

R1 {I > 2σ(I )}

wR2(F ²) all data

0.099

0.139

0.359

0.144

0.146

(0.5H, dd, J 14.0 and 3.0, H₆-eq. A), 4.13 (0.5H, m, H₂-eq. B),

4.49 (0.5H, dd, J 14.0 and 3.0, H₆-eq. B) and 4.90 (1H, m,

H₂-eq. A).

WI, USA, 1997]. All non hydrogen atoms were reﬁned with

anisotropic displacement parameters. All hydrogen atoms were

included in reﬁnement cycles riding on bonded atoms.††

(xvi) (R)-N-Benzoyl-3-methylpiperidine 17 from (R)-3-

methylpiperidine [prepared by resolution of the racemic amine

with (ϩ)-tartaric acid by the literature method⁹[α]_D= Ϫ3.3

(c 25, methanol), lit.⁹[α]_D= Ϫ0.61 (neat)] as a light yellow oil

(71%); δ_H(300 MHz) (1 : 1 mixture of N–CO rotamers) 1.35–

1.92 (4H, br m, 2 × CH₂), 2.44 and 2.61 (1H, 2 × m, CH), 2.81

and 2.94 (1H, 2 × m, CH), 3.62 (1H, m, CH), 4.54 (1H, m, CH)

and 7.35–7.45 [5H, m, CH(Ph)]; [α]_D= Ϫ43 (c 0.3, CHCl₃) or

[α]_D= Ϫ51 (c 1, methanol); lit.²³[α]_D= Ϫ51.9 (c 1, methanol),

lit.²⁴(for (S)-enantiomer) [α]_D= ϩ49.5 (c 1, methanol).

(xvii) (S)-N-Benzoylvaline methyl ester 22 from -valine

methyl ester hydrochloride. Column chromatography of the

residual pale yellow oil over silica with light petroleum–ethyl

acetate gave (S)-N-benzoylvaline methyl ester 22 as a viscous

colourless oil which crystallised on standing (91%), mp 110–112

ЊC (from light petroleum), [α]_D= ϩ46.2 (c 0.65, CHCl₃), lit.¹³

mp 110.5–111 ЊC, [α]_D= ϩ46.0 (c 0.4, CHCl₃); δ_H0.99 and 1.01

(6H, 2 × d, J 6.6, CH₃CHCH₃), 2.28 (1H, dh, J 6.6 and 4.8,

CH₃CHCH₃), 3.78 (3H, s, OCH₃), 4.78 (1H, dd, J 8.3 and 4.8,

CHNH), 6.68 (1H, d, J 8.3, NH), 7.37–7.55 [3H, m, CH(Ar)]

and 7.78–7.83 [2H, m, CH(Ar)].

(xviii) (S)-N-Ethanoylalanine ethyl ester 34 from -alanine

ethyl ester hydrochloride. The pale yellow oil obtained was

puriﬁed by column chromatography over silica with light

petroleum–ethyl acetate to give (S)-N-ethanoylalanine ethyl

ester 34 as a colourless viscous oil which crystallised on

standing (0.19 g, 91%). ν_max/cm^Ϫ13440s, 1740s, 1670s, 1515s

and 1455s; δ_H1.28 (3H, t, J 7.1, CH₃CH₂O), 1.39 (3H, d,

J 7.4, CH₃CHN), 2.02 (3H, s, CH₃CO₂), 4.19 (2H, q, J 7.1,

CH₃CH₂O), 4.49 (1H, dq, J 7.4 and ∼6.0, CH₃CHNH) and 6.13

(1H, d, J ∼6.0, NH); δ_C13.1, 17.1 and 21.9 (3 × CH₃), 47.1

(CH₃CHNH), 60.4 (CH₃CH₂O) and 169.1 and 172.3 (2 × CO);

[α]_D= ϩ15.7 (c 0.7 CHCl₃) lit.²⁵[α]_D= ϩ10.4 (c 1, CHCl₃);

δ_H1.29 (t, J 7, 3H), 1.4 (d, J 7, 3H), 2.02 (s, 3H), 4.2 (q, J 7, 2H),

4.6 (q, J 7, 1H) and 6.2 (m, 1H).

†† CCDC reference number(s) 167260–167264. See http://ww w .rsc.org/

suppdata/p1/b1/b105917n for crystallographic ﬁles in .cif or other

electronic format.

References

1 R. S. Atkinson, E. Barker and M. J. Sutcliﬀe, Chem. Commun., 1996,

1051.

2 Y. Murakami, K. Kondo, K. Miki, Y. Akiyama, T. Watanabe and

Y. Yokoyama, Tetrahedron Lett., 1997, 38, 3751.

3 For preliminary communications of this work see (a) A. G.

Al-Sehemi, R. S. Atkinson, J. Fawcett and D. R. Russell, Chem.

Commun., 2000, 43; (b) A. G. Al-Sehemi, R. S. Atkinson, J. Fawcett

and D. R. Russell, Tetrahedron Lett., 2000, 41, 2239; (c) A. G.

Al-Sehemi, R. S. Atkinson, J. Fawcett and D. R. Russell,

Tetrahedron Lett., 2000, 41, 2243.

4 R. S. Atkinson, E. Barker, P. J. Edwards and G. A. Thomson,

J. Chem. Soc., Perkin Trans. 1, 1996, 1047.

5 A. G. Al-Sehemi, R. S. Atkinson, J. Fawcett and D. R. Russell,

J. Chem. Soc., Perkin Trans. 1., 2000, 4413.

6 G. A. El-Hiti, Spectrosc. Lett., 1999, 32, 671.

7 J. C. Craig and A. R. Pindlar, J. Org. Chem., 1971, 36, 3649.

8 E. Vedejs and X. Chen, J. Am. Chem. Soc., 1997, 119, 2584.

9 H. Sliwka and H. Hansen, Helv. Chim. Acta, 1984, 67, 434.

10 J. M. Andres, I. Herraiz-Sierra, R. Pedrosa and A. Perez-Enoabo,

Eur. J. Org. Chem., 2000, 1719.

11 R. S. Atkinson and P. J. Williams, J. Chem. Soc., Perkin Trans. 2,

1995, 205.

12 N. Brosse, M. F. Pinto and B. Jamart-Grégoire, J. Chem. Soc., Perkin

Trans. 1, 1998, 3685.

13 T. H. Applewhite, H. Wate and C. Niemann, J. Am. Chem. Soc.,

1958, 80, 1465.

14 D. Crich and J. W. Davies, Tetrahedron, 1989, 45, 5641.

15 G. Pandey and P. Das, Tetrahedron Lett., 1997, 38, 9073.

16 T. Kohara, Y. Hashimoto and K. Saijo, Tetrahedron, 1999, 55, 6453.

17 G. L. Milligan, C. J. Mossman and J. Aube, J. Am. Chem. Soc., 1995,

117, 10449.

18 S. Tsuboi, Y. Noada and A. Takeda, J. Org. Chem., 1984, 49, 1204.

19 E. Alonso, D. J. Ramm and M. Yul, Tetrahedron, 1998, 45, 13629.

20 R. M. Lawrence, S. A. Biller, O. M. Fryszman and M. A. Poss,

Synthesis, 1997, 553.

21 Y. Nagao, M. Yagi, T. Ikede and E. Fujita, Tetrahedron Lett., 1982,

23, 201.

22 R. A. Johnson, J. Org. Chem., 1968, 33, 3627.

23 E. Coderc, P. Cerutti, J. Vignan, J. F. Ronayrenc and J. M. Kamenka,

Eur. J. Med. Chem. Chim. Theor., 1995, 30, 463.

24 A. Thurkauf, P. Hillery, A. E. Jacobson and K. C. Rice, J. Org.

Chem., 1987, 52, 5466.

Crystallography

All data were measured on a Bruker P4 diﬀractometer and

collected using graphite monochromated Mo-Kα radiation

(λ = 0.71073 Å) (Table 1). Absorption corrections were not

applied to the data sets. The structures were solved by direct

methods and reﬁned by full-matrix least squares cycles on

F²for all data, using SHELXTL [SHELXTL–an integrated

system for solving, reﬁning and displaying crystal structures.

Version 5.10, Bruker Analytical X-ray Systems, Madison,

25 A. Karim, A. Mortreux, F. Petit, G. Buono, G. Pfeiﬀer and C. Siv,

J. Organomet. Chem., 1986, 317, 93.

26 E. Barker, Thesis, 1995, Leicester University, UK.

274

J. Chem. Soc., Perkin Trans. 1, 2002, 257–274

Article Doi

Kinetic resolution of amines with enantiopure 3-N,N-diacylaminoquinazolin-4(3H)-ones

DOI: 10.1039/b105917n

Source and publish data:

Authors:

Article abstract of DOI:10.1039/b105917n

Full text of DOI:10.1039/b105917n

Products guided by the article

R&D Labs maybe for 20918-90-5

Relevant to this article

Hot Product