Troubleshooting a molecular motor: a remarkably stable N-acyl pyridinium salt

Article abstract of DOI:10.1016/j.tet.2008.04.101

With the objective of establishing why reaction of the proposed molecular motors 7 and 22 with carbonyldiimidazole and phosgene does not result in unidirectional rotation, N-ethyl-2-[4-(N,N-dimethylamino)-2-pyridinyl]benzenamine [28, 2-(2-(ethylamino)phenyl)DMAP] was examined as a model substrate. The synthesis of 28 is described. Compound 28 was found to react with phosgene to give the unexpectedly stable N-acyl pyridinium salt 30. The latter (30) is so stable that it is effectively inert to reaction with methanol. At room temperature the two methyls in the Me₂N-group of 30 are nonequivalent (NMR) and the barrier to rotation around the Me₂N-pyridinium bond is 18.5 kcal/mol. To the authors' knowledge, that is the first quantitative determination of the barrier to rotation around the bond between a 4-(N,N-dimethylamino) group and an N-acyl pyridinium unit. The implications of the findings regarding 30 as to troubleshooting the proposed molecular motor 7, and possible strategies to follow, are discussed.

Full text of DOI:10.1016/j.tet.2008.04.101

Tetrahedron 64 (2008) 8381–8388
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Troubleshooting a molecular motor: a remarkably stable N-acyl pyridinium salt
*
Michael D. Markey, T. Ross Kelly
Boston College, E.F. Merkert Chemistry Center, Department of Chemistry, Chestnut Hill, MA 02467, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
With the objective of establishing why reaction of the proposed molecular motors 7 and 22 with car-
bonyldiimidazole and phosgene does not result in unidirectional rotation, N-ethyl-2-[4-(N,N-dimethyl-
Received 31 December 2007
Accepted 25 April 2008
Available online 30 April 2008
amino)-2-pyridinyl]benzenamine [28, 2-(2-(ethylamino)phenyl)DMAP] was examined as
a model
substrate. The synthesis of 28 is described. Compound 28 was found to react with phosgene to give the
unexpectedly stable N-acyl pyridinium salt 30. The latter (30) is so stable that it is effectively inert to
reaction with methanol. At room temperature the two methyls in the Me2N-group of 30 are nonequivalent
(NMR) and the barrier to rotation around the Me2N–pyridinium bond is 18.5 kcal/mol. To the authors’
knowledge, that is the first quantitative determination of the barrier to rotation around the bond between
a 4-(N,N-dimethylamino) group and an N-acyl pyridinium unit. The implications of the findings regarding
30 as to troubleshooting the proposed molecular motor 7, and possible strategies to follow, are discussed.
Ó 2008 Elsevier Ltd. All rights reserved.
Dedicated to Professor J. Fraser Stoddart in
honor of his unusually creative contribu-
tions to chemistry and on his receipt of the
Tetrahedron Prize
1. Introduction
Figure 1. The system shown in Figure 1 extends the prototype by: (i)
having an amino group on each blade of the triptycene; (ii) pro-
In 1999, we reported a prototype of the first rationally designed,
chemically powered, rotary molecular motor.1,2 The phosgene-
fueled system is shown in Scheme 1 and achieves 120ꢀ of unidi-
rectional rotation. More recently,3 we have sought to advance 1 to
a continuously rotating motor as summarized conceptually in
viding a means for delivering phosgene (or its equivalent) to the
amine in the ‘firing position’; (iii) achieving a phosgene-powered,
120ꢀ rotation of the triptycene by formation of an intramolecular
urethane in analogy to Scheme 1; (iv) affording a means for re-
moving the remains of the phosgene by cleavage of the urethane,
Scheme 1. Prototype of a chemically powered, rotary molecular motor.
* Corresponding author. Tel.: þ1 617 552 3621; fax: þ1 617 552 2705.
E-mail address: ross.kelly@bc.edu (T.R. Kelly).
0040-4020/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2008.04.101

8382
M.D. Markey, T.R. Kelly / Tetrahedron 64 (2008) 8381–8388
was to achieve the sequence of events given in Scheme 2, resulting
in a continuously rotating molecular motor.
A major synthetic effort eventually made available limited, but
sufficient, quantities of 7 to evaluate the design.3
2. Results and discussion
Reaction of 7 with phosgene and triethylamine under the same
conditions that were successful with the prototype (Scheme 1) led to
immediate polymerization of 7 by polyurea formation. Polymerization
could beavoided(atleastinitially)byconducting the reactionatꢁ78 ꢀC
under 100-fold more dilute conditions, but the phosgene seemed too
reactive toward the triptycylamines to give the DMAP an opportunity
to function as desired, since a mixture of mono-, di-, and tri-‘phosge-
nylation’ was observed. In order to give the DMAP time to accomplish
its role, the phosgene was replaced with the less reactive 1,10-car-
bonyldiimidazole. Reaction between 7 and 1,10-carbonyldiimidazole
gives only monoacylation of 7, as determined by quenching with
methanol to give 17, whose structure was determined unambiguously.
We inferred from the isolation of 17 from the methanol quench that
the reaction between 7 and 1,10-carbonyldiimidazole gave 18.
Figure 1. Schematic for a continuously rotating molecular motor involving selective
(and repeated) delivery of Cl2C]O to the amino group in the ‘firing position’ and
cleavage of the urethane only after each 120ꢀ of rotation has occurred.
Figure 2. Proposed repeatedly rotating motor.
thereby allowing repeated rotation by repetition of the three pre-
ceding steps.
Compound 7 was chosen3 as a molecular embodiment of the
concepts in Figure 1, specifically addressing the first three (i)–(iii)
items enumerated above, with the (dimethylamino)pyridine
(DMAP) unit intended to capture and deliver the phosgene to the
amino group (bold arrow in Fig. 2) in the firing position. The hope
Scheme 2. Proposed operation of a continuously rotating, chemically powered molecular motor.

M.D. Markey, T.R. Kelly / Tetrahedron 64 (2008) 8381–8388
8383
Imidazoyl ureas of primary anilines are reported4 by Staab to be
in equilibrium with the corresponding isocyanates (Eq. 1). If that
equilibrium occurred here and if the 7-based system behaved in
analogy to the prototype (Scheme 1), then the sequence of events in
Scheme 2 should have unfolded. But it did not.
either the isocyanate, the carbamoyl chloride 10, or carbamoyl
imidazole 18. Such an interaction would severely restrict rotation
around the triptycene–helicene bond, precluding access to ener-
getically excited rotamers analogous to 3. Furthermore, such an
interaction would attenuate the electrophilic nature of the phos-
gene-derived (or carbonyldiimidazole-derived) carbonyl in 25–27,
making reaction with the hydroxypropyl less likely (if geometric
constraints did not make it impossible).
ð1Þ
In order to determine if the interactions suggested in 25–27
were credible, we chose to examine the reaction of 28 with phos-
gene to see if there were any indications of analogous interactions,
such as those shown in 29 and 30.
The system in Scheme 2 differs from that in the prototype
(Scheme 1) in two ways. First, the motor molecules are different,
and second, the fuels are different (phosgene in Scheme 1; even-
tually 1,10-carbonyldiimidazole in Scheme 2). In order to determine
whether the failure of the system in Scheme 2 was due to a differ-
ent motor molecule or a different fuel, it was necessary to examine
a case where only one rather than two variables had been changed.
To that end, compound 22, where two of the triptylcylamines
are protected as trifluoroacetyl amides, was prepared.3 In this
instance, there is only one free amine, the one in the firing position,
so there is no need to enlist the DMAP to deliver the fuel to only one
amine. Consequently, it was possible to use phosgene rather than
1,10-carbonyldiimidazole as the fuel. In this situation, the system
differs from that in Scheme 1 by only one variable, the structure of
the motor molecule. The reaction of 22 with phosgene would either
lead to a functioning motor or indicate that the problem was with
the precise structure of the motor molecule. In fact,3 reaction of 22
with phosgene gave no intramolecular urethane (23 or 24) as
judged by mass spectrometry. Put briefly, despite the seemingly
excellent precedent provided by the prototype system in Scheme 1,
the completely developed versions (7 and 22) do not function as
motors.
It might be objected that 28 is not a good modely for 7/22
because the two nitrogens of interest (drawn in red) in 28 are
separated by only four bonds, while a dozen or so bonds separate
the analogous nitrogens in 7/22. But, because of 7/220s inherent
rigidity, there are no more degrees of freedom associated with 7/22
(apart from the triptycene–helicene bond, whose rotation is
severely constrained by a ca. 25 kcal/mol barrier1a,b to rotation)
than with 28.
Accordingly, the synthesis of 28 was undertaken and was
reduced to practice as shown in Scheme 3. Stille coupling6 of
y
Secondary amine 28 was chosen in preference to the unalkylated primary
amine 31 because if 31 was to form 32 (from 31 and phosgene), 32 could suffer loss
of a proton and electron reorganization to give pyridoquinazolinone 33, a com-
pound whose tricyclic ring system is known9 to be stable.
We suggested3 two explanations for why 7 and 22 do not behave
as motors. One explanation was that the hydroxypropyl group
adopts a conformation different from that in the prototype, possibly
due to hydrogen-bonding interactions with the DMAP or the added
substituents on the other two triptycene blades. The second
explanation was the intervention of a Bu¨rgi–Dunitz5 (25) or even
a covalent interaction, as in 26 and 27, between the DMAP and

8384
M.D. Markey, T.R. Kelly / Tetrahedron 64 (2008) 8381–8388
commercially available 1-bromo-2-nitrobenzene with stannane
347 gave biaryl 35. The nitro group was reduced8 and the resulting
aniline 31 was converted to its tert-butoxycarbonyl (BOC)
derivative 36. Treatment of the latter (36) with sodium hydride and
ethyl iodide and exposure of the crude product to acid cleaved the
BOC, affording 28 directly.
In order to preclude the possibility that 30 converts to 38 at
100 ꢀC in CH3OH but that 38 cyclizes back to 30 upon cooling, 1H
NMR spectra of 30 in CD3OD were recorded first at 25 ꢀC and then
at 100 ꢀC. The spectra were identical except that the two sharp
peaks for the N-methyl protons at 25 ꢀC had coalesced to a broad
singlet at 100 ꢀC (Fig. 3). Cooling of the sample back to 25 ꢀC
decoalesced the peak for the N-methyl protons and restored the
25 ꢀC spectrum. Prompted by this serendipitous observation of
coalescence, the actual coalescence temperature was measured and
determined to be 95 ꢀC (Fig. 3). That temperature, taken with
a difference in the frequencies of the two N-methyl resonances of
35.5 Hz (a 500 MHz spectrometer was employed) at 25 ꢀC, allows
one to calculate12 the barrier to rotation about the Me2N–pyri-
dinium bond in 30 as 18.5 kcal/mol, indicating that there is sub-
stantial double bond character (as in 39) to that bond. To our
knowledge, this is the first experimental determination of the
barrier to rotation around the Me2N–pyridinium bond in an N-acyl
(dimethylamino)pyridinium derivative. NMR spectra of un-
symmetrical but neutral DMAP derivatives such as 28, 31, 35, and
25 °C
50 °C
60 °C
70 °C
80 °C
Scheme 3. Synthesis of DMAP-substituted aniline 28.
In order to determine the behavior of 28 toward phosgene,
phosgenedas a solution in toluenedwas added to a mixture of
a CDCl3 solution of 28 and an excess of an insoluble and polymer-
bound tertiary amine.z A precipitate formed immediately. Addition
of CH3OH dissolved the precipitate. The CDCl3/CH3OH solution was
filtered through a plug of cotton to separate the insoluble polymer
(presumably a mixture of polymer-bound tertiary amine and the
corresponding amine hydrochloride), and the filtrate was evapo-
rated. The residue (a solid) was dissolved in CD3OD (it is only
slightly soluble in CDCl3), and the 1H and 13C NMR spectra were
recorded. The spectra of the solid show it to be a single compound,
but do not unequivocally distinguish between structures 29 and 30.
The spectra do exclude carbamoyl chloride 37 as a possible struc-
ture, because the two nitrogen-bound methyl groups are not
equivalent in the NMR spectra. Fortunately, diffraction-quality
crystals of the solid could be grown from ethanol/diethyl ether.
X-ray crystallography revealed the solid to be the cyclic acyl
pyridinium salt 30.
85 °C
90 °C
95 °C
100 °C
110 °C
It is remarkable to us10,11 that 30 is so stable that it can be
recrystallized from even initially hot ethanol or isopropanol. In fact,
30 is unchanged by heating in CH3OH at 100 ꢀC for 19 h, exhibiting
no detectable tendency to convert to carbamate 38.
25 °C
4.0 3.8 3.6 3.4 3.2 3.0 2.8 2.6 2.4 2.2 2.0 1.8 1.6 1.4 1.2
z
A polymer-bound tertiary amine rather than a ‘regular’ tertiary amine such as
Figure 3. 1H NMR spectra (500 MHz, CD3OD) of the two N-methyl resonances of 30 at
varying temperatures. The bottom 25 ꢀC spectrum was recorded after allowing the
110 ꢀC sample to cool. The peaks from the ethyl group serve as an internal reference.
Et3N was used to facilitate separation of amine hydrochloride salts and 28-derived
substances.

M.D. Markey, T.R. Kelly / Tetrahedron 64 (2008) 8381–8388
8385
36 show (see Section 4) the N-methyl protons to be equivalent at
room temperature.
27. The obvious question is: ‘how might one modify 7 to achieve
a functioning motor?’
The consensus mechanism for the catalytic role of DMAP in the
acylation of alcohols (and, presumably, other substrates) is shown
in Scheme 4.11g,14
The studies described above indicate that 30 is stable when
dissolved in CH3OH. Put differently, the equilibrium in Eq. 2 lies far
to the left.
Scheme 4. Generally accepted mechanism for the catalytic role of DMAP in the
acylation of alcohols by acylating agents.
ð2Þ
The second step, the reaction of the acyl pyridinium salt 44 with
R0OH, is the rate-determining step.11g,14 According to Hassner
et al.11b and Spivey and Arseniyadis,11g 4-amino-substituted pyri-
dines such as DMAP are more effective catalysts than pyridine itself
because they are better at stabilizing the intermediate acyl pyri-
dinium species 44 (by a contribution from resonance structure 47).
It may seem counterintuitive that stabilizing the starting material
for a rate-determining step would accelerate the rate for the overall
reaction. But stabilizing 44 results in more 44 being present,
because the position of the 42þ43#44 equilibrium is shifted to the
right. Evidently, the increased concentration of 44 plays a greater
role in accelerating the overall reaction than the stabilization of 44
does in decelerating the rate-limiting step.
The reactivity of 30 toward two nucleophiles other than CH3OH,
namely water and dimethylamine, was also examined. When a
solution of 30 in water is heated at 55 ꢀC for 16 h, about 20% is
converted to 28, presumably by way of an unfavorable equilibrium
to give carbamic acid 40 (Eq. 3) followed by decarboxylation13 of 40
to give 28. The decarboxylation provides a means for driving the
conversion of 30 to 28 to completion; no comparable driving force
is available to drive the conversion of 30 and CH3OH to 38 (Eq. 2).
ð3Þ
In contrast to the situation with 30 and CH3OH, where equilib-
rium (e.g., Eq. 2) favors N-acyl pyridinium species 30 and CH3OH
rather than urethane 38, reaction of 30 with dimethylamine in
CD3OD to give urea 41 proceeds virtually to completion in 3 h at
room temperature (Eq. 4), reflecting the greater stability of amides
(ureas) versus esters (urethanes). Here too, however, the proximity
of the DMAP unit to the urea carbonyl in 41 bestows reactivity on
the urea unlike that of typical ureas: the equilibrium in Eq. 4 can be
driven to the left at room temperature simply by several cycles of
dissolving the mixture of 30 and 41 in CD3OD and removing vola-
tiles under vacuum, with all of 41 being converted back to 30.
ð4Þ
From the standpoint of ‘repairing’ the molecular motor, it would
appear that replacing the DMAP in 7 with a less participatory unit,
perhaps just a pyridine, would be sufficient to attenuate the
interfering Bu¨rgi–Dunitz (25) or covalent (26 or 27) interaction. The
concern with adopting such an approach is that it might also
diminish the system’s ability to capture and deliver the carbonyl-
imidazole unit to the requisite aniline group. Perhaps a better way
to walk this molecular tightrope is to include a unit that can in-
terfere with the behavior of the DMAP unit, but only after the DMAP
3. Conclusions
As mentioned previously, the work described above was carried
out with the aim of troubleshooting the non-functioning molecular
motor of Scheme 2. The discovery of the unexpected stability of 30
leads us to conclude that the problem is the intrusion into Scheme 2
of the Bu¨rgi–Dunitz or covalent interactions depicted in 25, 26, and

8386
M.D. Markey, T.R. Kelly / Tetrahedron 64 (2008) 8381–8388
has delivered the carbonylimidazole group. Two possibilities are 48
and 49, where, after the carbonylimidazole moiety has been
delivered, one adds a metal ion (Mþþ), with metal-ion coor-
dinationdas in 50 and 51ddisengaging the dimethylamine group
from stabilizing the Bu¨rgi–Dunitz (25) or covalent (26 or 27)
interaction. Whether such a metal-coordination strategy would be
effective in the current instance remains to be determined, but the
use of metal-ion coordination stood us in good stead in effectively
engaging a molecular brake (Eq. 5).15
Crystallographic data (excluding structure factors) for the struc-
tural analysis has been deposited with the Cambridge Crystallo-
graphic Data Center as supplementary publication no. CCDC
670092. Copies of the data can be obtained, free of charge, on ap-
plication to CCDC, 12 Union Road, Cambridge CB2 1EZ, UK (fax: þ44
(0)1223 336033 or e-mail: deposit@ccdc.cam.ac.uk). Elemental
analyses were performed by Atlantic Microlab, Inc. (Norcross, GA).
4.2. Synthesis of 6H-9(N,N-dimethylamino)-5-(N-ethyl)-
6-oxopyrido[1,2-c]quinazolinium chloride (30)
4.2.1. 4-(N,N-Dimethylamino)-2-(2-nitrophenyl)pyridine (35)
Commercially available cuprous iodide (Aldrich, catalog number
20,554-0) was purified following Dieter’s procedure.19 For a typical
case, 5.00 g of cuprous iodide was added in one portion, open to the
air, to a saturated aqueous NaI solution (20 mL) immersed in a hot
oil bath (100 ꢀC). The homogenous solution was maintained in the
oil bath (100 ꢀC) for 30 min, then removed from the bath, diluted
with H2O (20 mL), and allowed to cool to room temperature. Upon
cooling, a precipitate formed. The precipitate was collected by
vacuum filtration and washed sequentially with H2O, ethanol, ethyl
acetate (EtOAc), diethyl ether (Et2O), and finally hexanes. The
purified cuprous iodide was both dried (ca. 24 h) and stored over
P2O5 in a vacuum desiccator (ca. 0.05 Torr).
ð5Þ
In conclusion, the foregoing studies appear to have identified
the source of the problem with the non-functioning of 7 and 22.
Unfortunately, however, for reasons that have been given else-
where,16 it is unlikely that the proposed cures will ever be
examined.
4. Experimental
4.1. General
For a related procedure see Ref. 7. A 100 mL airfreeÒ round-
bottomed flask (Chemglass, catalog no. AF-0528-02) was charged
with cesium fluoride (1.62 g, 10.7 mmol, purchased from Strem),
a rubber septum, and a stir bar. The flask and its contents were
flame-dried under vacuum and backfilled with Ar. To the flask were
added 1-bromo-2-nitrobenzene (1.08 g, 5.35 mmol), 347 (2.20 g,
5.35 mmol), and DMF (50 mL). The mixture was freeze–pump–
thaw degassed three times under vacuum and backfilled with Ar. To
the mixture was added tetrakis(triphenylphosphine)palladium(0)
(310 mg, 0.268 mmol, purchased from Strem) followed by purified
(see above) cuprous iodide (153 mg, 0.803 mmol), each in a single
portion. The flask was fitted with an Ar balloon and the mixture was
heated to 110 ꢀC. After stirring at 110 ꢀC for 19 h, the black mixture
was cooled to room temperature and passed through a plug of
CeliteÔ. The CeliteÔ was rinsed with EtOAc (100 mL). The com-
bined filtrate and rinse were added to 9 M NH4OH (150 mL) and
transferred to a separatory funnel. The aqueous layer was extracted
with EtOAc (3ꢂ50 mL). The organics were pooled, washed with 9 M
NH4OH (50 mL portions; washed until the blue/green color no
longer persisted in the aqueous layer), H2O (2ꢂ150 mL), and satu-
rated NaCl solution (1ꢂ100 mL), dried with MgSO4, and filtered.
Removal of the solvent in vacuo gave a red oil. The oil obtained was
purified by flash column chromatography (12 cmꢂ4 cm; Brockman
III neutral aluminum oxide) with 40% EtOAc in hexanes to afford
795 mg (61%) of 35 (Rf¼0.4; 1:1 hexanes/EtOAc; neutral aluminum
oxide) as a brown solid. Mp 110–112 ꢀC; 1H NMR (400 MHz, CDCl3)
All reactions were carried out under an atmosphere of argon in
flame-dried glassware. The reaction solvent dichloromethane
(CH2Cl2) was purified by passage through activated alumina col-
umns.17 Anhydrous N,N-dimethylformamide (DMF) was purchased
from ACROS and stored over activated 4 Å molecular sieves. All
other reagents and solvents were used as received from the man-
ufacturer unless otherwise stated. Flash column chromatography
was carried out using Bodman reagent silica gel 60 Å or Alfa Aesar
activated neutral aluminum oxide 58 Å. Prior to use, activated
Brockman I neutral aluminum oxide was adjusted18 to Brockman III
activity by adding 6% (w/w) water (H2O). Reactions were monitored
by thin layer chromatography (TLC) using Whatman 250 micron
silica gel plates with aluminum backing and UV254 nm fluorescent
indicator or Analtech 200 micron neutral aluminum oxide plates
with aluminum backing and UV254 nm fluorescent indicator. All
TLC plates were visualized by UV fluorescence quenching. All so-
lution pHs were measured using EM colorpHast pH strips (Fisher,
catalog number M95903). The phrase ‘removal of solvents in vacuo’
means that solvents were removed on a rotary evaporator using
a diaphragm pump (ca. 8 Torr) and that remaining traces of vola-
tiles were then removed on a high-vacuum oil pump (ca. 0.05 Torr).
Melting points were recorded on a Fisher-Johns melting point
apparatus or Lab Devices Mel-Temp II (mp>300 ꢀC) and are
uncorrected. Infrared spectra were recorded using a Nicolet Avatar
360 FT-IR spectrometer. 1H NMR spectra were recorded on a Varian
Gemini-400 (400 MHz) spectrometer or a Varian Unity INOVA-500
(500 MHz) spectrometer and are reported in parts per million using
TMS (0.00 ppm) or residual solvent protons (CDCl3¼7.26 ppm,
d
8.25 (d, J¼6.0 Hz, 1H), 7.85 (d, J¼8.0 Hz, 1H), 7.63–7.61 (m, 2H),
7.53–7.48 (m, 1H), 6.64 (d, J¼2.4 Hz,1H), 6.51 (dd, J¼6.0, 2.4 Hz,1H),
3.05 (s, 6H); 13C NMR (100 MHz, CDCl3)
155.6, 154.6, 149.5, 149.4,
136.3, 131.9, 131.0, 128.5, 123.8, 105.7, 105.2, 39.1; IR (KBr) 3447,
2917, 1601, 1539, 1456, 1374, 1303 cmꢁ1; HRMS (ESI) calcd for
13H14N3O2 (MHþ) 244.1086, found 244.1089.
d
n
CD3OD¼3.30 ppm) as an internal standard. Data are reported as: [
d
C
shift] ([s¼singlet, d¼doublet, ap d¼apparent doublet, dd¼doublet
of doublets, t¼triplet, q¼quartet, m¼multiplet, br¼broad], [J¼
coupling constant in Hz], and [integration]). Proton-decoupled 13C
NMR spectra were recorded on a Varian Gemini-400 (100 MHz)
spectrometer or a Varian Unity INOVA-500 (125 MHz) spectrome-
ter and are reported in parts per million using solvent (CDCl3¼
77.0 ppm and CD3OD¼49.0 ppm) as an internal standard. High
resolution mass spectra were obtained at the Boston College mass
spectrometry laboratory. X-ray crystallography and structure
analysis for 30 were performed at the Boston College X-ray facility.
4.2.2. 2-[4-(N,N-Dimethylamino)-2-pyridinyl]benzenamine (31)
For a related procedure see Ref. 7. To a 100 mL round-bottomed
flask fitted with a reflux condenser were added 35 (795 mg,
3.27 mmol) and methanol (45 mL). To the flask was added, in one
portion, a freshly prepared solution of sodium sulfide nonahydrate
(3.15 g, 13.1 mmol) and sodium hydroxide (1.30 g, 32.7 mmol) in
H2O (15 mL). The flask was heated to 85 ꢀC, stirred open to the air,
and maintained at 85 ꢀC as the progress of the reaction was mon-
itored by TLC (Rf of 31¼0.3; 1:1 Et2O/hexanes; neutral aluminum

M.D. Markey, T.R. Kelly / Tetrahedron 64 (2008) 8381–8388
8387
oxide). Once the reaction was complete (ca. 5 h), the solution was
cooled to room temperature and the solvents were removed in
vacuo to give a yellow residue. The residue obtained was dissolved
in a mixture of CH2Cl2 (50 mL) and 1 M NaOH (50 mL). The mixture
was transferred to a separatory funnel, the layers were separated,
and the aqueous layer was extracted with CH2Cl2 (3ꢂ50 mL). The
organics were pooled, washed with 1 M NaOH (2ꢂ75 mL), dried
with MgSO4, and filtered. Removal of the solvent in vacuo gave
a yellow residue. The residue was purified by flash column chro-
matography (12 cmꢂ4 cm; Brockman III neutral aluminum oxide)
with 55% EtOAc in hexanes to afford 527 mg (75%) of 31 as a white
neutral aluminum oxide). Once the reaction was complete (ca.
30 min), the mixture was quenched (CAUTION: residual NaH may
be present!) with H2O (50 mL) and extracted with EtOAc
(3ꢂ50 mL). The organics were pooled, washed with H2O
(4ꢂ100 mL) and saturated NaCl solution (1ꢂ200 mL), dried with
MgSO4, and filtered. Removal of the solvent in vacuo gave 650 mg of
a white solid. The white solid obtained was dissolved in EtOAc
(10 mL) and transferred to a 50 mL round-bottomed flask. To the
solution was added 3 M HCl (10 mL) and the biphasic mixture was
allowed to stir open to the air. The mixture was placed in a 40 ꢀC oil
bath and the progress of the reaction was monitored by TLC (stir-
ring was stopped and the EtOAc layer was spotted; Rf of 28¼0.5; 1:1
EtOAc/hexanes; silica). Once complete (ca. 4 h), 10 M NaOH was
slowly added to the mixture until (ca. 5 mL) the pH of the aqueous
layer was approximately 10. The contents of the flask were trans-
ferred to a separatory funnel, the EtOAc layer was separated, and
the aqueous layer was extracted with EtOAc (3ꢂ40 mL). The or-
ganics were pooled, washed with H2O (2ꢂ40 mL) and saturated
NaCl solution (1ꢂ50 mL), dried with MgSO4, and filtered. Removal
of the solvent in vacuo gave an oil. The oil obtained was purified by
flash column chromatography (12 cmꢂ3 cm; Brockman III neutral
aluminum oxide) with 15% EtOAc in CH2Cl2 to afford 316 mg (68%
from 36) of 28 as a white solid. Mp 94–96 ꢀC; 1H NMR (400 MHz,
solid. Mp 88–90 ꢀC; 1H NMR (400 MHz, CDCl3)
d
8.24 (d, J¼6.0 Hz,
1H), 7.43 (dd, J¼7.6,1.6 Hz,1H), 7.16–7.11 (m,1H), 6.78–6.72 (m, 3H),
6.44 (dd, J¼6.0, 2.6 Hz, 1H), 5.45 (br s, 2H), 3.03 (s, 6H); 13C NMR
(100 MHz, CDCl3)
d 159.5, 155.2, 148.2, 146.2, 129.3, 129.2, 124.3,
117.5, 116.7, 105.4, 104.8, 39.2; IR (KBr)
n 3451, 3326, 2922, 1597,
1537, 1492, 1445 cmꢁ1; HRMS (ESI) calcd for C13H16N3 (MHþ)
214.1344, found 214.1353.
4.2.3. 1,1-Dimethylethyl [2-[4-(N,N-dimethylamino)-2-
pyridinyl]phenyl]carbamate (36)
A 100 mL round-bottomed flask was charged with 31 (1.04 g,
4.88 mmol), triethylamine (3.45 mL, 24.4 mmol), and CH2Cl2
(35 mL). To the solution was added di-tert-butyl dicarbonate (1.13 g,
5.12 mmol) in six portions (ca. 200 mg each) over a 5 min period.
The solution was allowed to stir at room temperature. After stirring
for 20 h, the yellow precipitate, which had formed was collected by
vacuum filtration and washed with CH2Cl2 (3ꢂ10 mL). The com-
bined filtrate and washes were diluted with CH2Cl2 (50 mL),
transferred to a separatory funnel, washed with H2O (2ꢂ50 mL)
and saturated NaCl solution (1ꢂ50 mL), dried with MgSO4, and
filtered. Removal of the solvent in vacuo gave an oil. The oil
obtained was purified by flash column chromatography
(12 cmꢂ3 cm; silica gel) with 40% Et2O in hexanes to afford 600 mg
(39%) of 36 (Rf¼0.4; 2:3 hexanes/Et2O; silica) as a white solid. Mp
CDCl3)
d
8.24 (d, J¼5.6 Hz, 1H), 7.43 (dd, J¼7.6, 1.6 Hz, 1H), 7.39 (br s,
1H), 7.25–7.21 (m, 1H), 6.76 (d, J¼2.8 Hz, 1H), 6.71–6.67 (m, 2H),
6.44 (dd, J¼5.6, 2.8 Hz, 1H), 3.19 (q, J¼7.2 Hz, 2H), 3.04 (s, 6H), 1.27
(t, J¼7.2 Hz, 3H); 13C NMR (100 MHz, CDCl3)
d 159.9, 155.1, 147.9,
147.4, 129.6,129.3,123.6,115.4,110.8,105.6,104.6, 39.2, 37.9,14.7; IR
(KBr)
n
3314, 2970, 2849, 1894, 1609, 1506, 1437 cmꢁ1; HRMS (ESI)
calcd for C15H20N3 (MHþ) 242.1657, found 242.1666.
4.2.5. 6H-9-(N,N-Dimethylamino)-5-(N-ethyl)-6-oxopyrido-
[1,2-c]quinazolinium chloride (30)
Chloroform (CHCl3) was purified according to Perrin and
Armarego.20 For a typical case, CHCl3 (180 mL, ACS grade purchased
from Mallinckrodt) was washed with H2O (3ꢂ100 mL), dried over
anhydrous potassium carbonate (10 g), refluxed with CaCl2 (35 g),
and distilled prior to use.
114–116 ꢀC; 1H NMR (400 MHz, CDCl3)
d 10.99 (br s, 1H), 8.28–8.27
(ap d, 2H), 7.54 (dd, J¼8.0, 1.6 Hz, 1H), 7.36–7.32 (m, 1H), 7.07–7.03
(m, 1H), 6.79 (d, J¼2.4 Hz, 1H), 6.48 (dd, J¼6.4, 2.4 Hz, 1H), 3.05 (s,
6H), 1.50 (s, 9H); 13C NMR (100 MHz, CDCl3)
d
158.5, 155.3, 153.3,
To a 25 mL round-bottomed flask were added polymer-
147.8, 137.9, 129.2, 128.7, 127.0, 121.8, 119.9, 105.7, 105.0, 79.4, 39.2,
supported diisopropylamine (PS-DIA) beads (669 mg, 0.336–
1.00 mmol; 50–90 mesh, 1% cross-linked, 0.5–1.5 mmol/g,
purchased from Aldrich, catalog number 538469) and distilled
CHCl3 (6 mL, see above). After stirring for 10 min under Ar, the CHCl3
was removed with a syringe and a narrow (21 gauge) needle. This
process was repeated a total of three times after which the beads
weredriedundervacuum (ca. 0.05 Torr)for30 min. Totheflaskwere
added 28 (75.0 mg, 0.311 mmol) and distilled CHCl3 (10 mL, see
above). The mixture was allowed to stir for 30 min at room tem-
28.4; IR (KBr)
n 3227, 2974, 2560, 2252, 1917, 1715, 1600, 1538,
1435 cmꢁ1; HRMS (ESI) calcd for C18H23N3O2Na (MNaþ) 336.1688,
found 336.1685. Anal. Calcd for C18H23N3O2: C, 68.98; H, 7.40; N,
13.41. Found: C, 69.01; H, 7.50; N, 13.36.
The precipitate collected (239 mg, 21%) was tentatively identi-
fied as 9-(N,N-Dimethylamino)-6H-pyrido[1,2-c]quinazolin-6-one
(33). Compound 33 (Rf¼0.38; 1:5 methanol/DCM; neutral alumi-
num oxide) was obtained as a yellow solid. Mp 290 ꢀC (dec); 1H
NMR (400 MHz, CDCl3)
d
9.26 (d, J¼8.0 Hz, 1H), 7.82 (d, J¼8.0 Hz,
perature and thenphosgene (327 mL, 0.62 mmol; purchased as a 20%
1H), 7.52 (t, J¼6.8 Hz, 1H), 7.42 (d, J¼8.0 Hz, 1H), 7.07–7.03 (m, 2H),
6.68 (dd, J¼8.0, 2.8 Hz, 1H), 3.29 (s, 6H). The compound was too
w/w solution in toluene from Fluka, catalog number 79830; CAU-
TION: extremely toxic!) was added in one portion. The mixture
initially took on a deep yellow color and then quickly (ca. 2 min)
turned to a cloudy white color. After stirring for 30 min at room
temperature, the flask was quenched with methanol (10 mL), turn-
ing the reaction mixture from a cloudy white suspension to a clear
mixture. The mixture was filtered through a plug of cotton and the
beads were rinsed with methanol (3ꢂ5 mL). The combined filtrate
and rinses were transferred to a 50 mL round-bottomed flask and
removal of the solvents in vacuo gave 110 mg of a white solid. This
white solid was crystallized from isopropanol (ca. 10 mL) to give
76 mg (81%) of 30 as white prisms. X-ray quality crystals were
obtained by slow vapor diffusion using ethanol as the dissolving
solvent and diethyl ether as the precipitating solvent. Mp 306–
insoluble to obtain a 13C NMR spectrum. IR (KBr)
n 3421, 1638, 1607,
1555, 1522, 1468, 1437 cmꢁ1; HRMS (ESI) calcd for C14H14N3O
(MHþ) 240.1137, found 240.1131.
4.2.4. N-Ethyl-2-[4-(N,N-dimethylamino)-2-pyridinyl]-
benzenamine (28)
To a 50 mL round-bottomed flask equipped with an Ar mineral
oil bubbler were added 36 (600 mg, 1.92 mmol) and DMF (15 mL).
To the stirred homogenous solution was added (CAUTION: rapid
evolution of H2 gas!) NaH (60% dispersed in mineral oil, 112 mg,
2.87 mmol) in three portions (ca. 40 mg each) followed by heating
to 30 ꢀC. The solution was maintained at 30 ꢀC for 2 h and then
cooled to room temperature. To the solution was added ethyl iodide
309 ꢀC (dec); 1H NMR (500 MHz, CD3OD)
8.62 (d, J¼8.5 Hz, 1H), 7.86 (t, J¼8.0 Hz, 1H), 7.71 (s, 1H), 7.66 (d,
J¼8.5 Hz, 1H), 7.51 (t, J¼8.0 Hz, 1H), 7.34 (dd, J¼8.5, 2.5 Hz, 1H), 4.42
d
9.12 (d, J¼8.5 Hz, 1H),
(158
mL, 2.02 mmol) in one portion and the progress of the reaction
was monitored by TLC (desired product Rf¼0.2; 3:2 hexanes/Et2O;

8388
M.D. Markey, T.R. Kelly / Tetrahedron 64 (2008) 8381–8388
(q, J¼7.0 Hz, 2H), 3.52 (s, 3H), 3.45 (s, 3H), 1.43 (t, J¼7.0 Hz, 3H); 13
C
6. For a recent review of the Stille reaction, see: Espinet, P.; Echavarren, A. M.
Angew. Chem., Int. Ed. 2004, 43, 4704–4734; See also: Nicolaou, K. C.; Bulger, P.
G.; Sarlah, D. Angew. Chem., Int. Ed. 2005, 44, 4442–4489.
NMR (125 MHz, CD3OD) d 158.7,146.1,145.5,137.4,136.1,136.0,127.1,
125.9, 116.7, 114.6, 109.5, 100.2, 41.8, 41.3, 41.2, 12.4; IR (KBr) n 3404,
7. Kelly, T. R.; Cavero, M. Org. Lett. 2002, 4, 2635–2656.
8. Perry, P. J.; Reszka, A. P.; Wood, A. A.; Read, M. A.; Gowan, S. M.; Dosanjh, H. S.;
Trent, J. O.; Jenkins, T. C.; Kelland, L. R.; Neidle, S. J. Med. Chem. 1998, 41, 4873–
4884.
9. Molina, P.; Lorenzo, A.; Aller, E. Tetrahedron 1992, 48, 4601–4616.
10. For reviews of carbonic acid derivatives, see: Hegarty, A. F. Comprehensive
Organic Chemistry; Barton, D., Ollis, W. D., Sutherland, I. O., Eds.; Pergamon:
Oxford, 1979; Vol. 2, pp 1067–1118; See also: Methoden der Organischen Chemie
(Houben-Weyl); Hagemann, H., Ed.; Georg Thieme: Stuttgart, New York, NY,
1983; Vol. E4.
3084, 2972, 1728, 1705, 1651, 1645, 1607, 1574, 1556, 1504, 1470,
1446 cmꢁ1; HRMS (ESI) calcd for C16H18N3O (Mþ) 268.1444, found
268.1448. Anal. Calcd for C16H18N3OCl$1⁄2 H2O: C, 61.44; H, 6.12; N,
13.43. Found: C, 61.16; H, 6.04; N, 13.29.
Acknowledgements
11. For reviews of DMAP-catalyzed reactions, see: (a) Sheinkman, A. K.; Suminov, S.
I.; Kost, A. N. Russ. Chem. Rev. 1973, 42, 642–661; (b) Hassner, A.; Krespki, L. R.;
Alexanian, V. Tetrahedron 1978, 34, 2069–2076; (c) Ho¨fle, G.; Steglich, W.;
Vorbru¨ggen, H. Angew. Chem., Int. Ed. Engl. 1978, 17, 569–583; (d) Ragnarsson,
U.; Grehn, L. Acc. Chem. Res. 1998, 31, 494–501; (e) Spivey, A.; Maddaford, A.;
Redgrave, A. J. Org. Prep. Proced. Int. 2000, 32, 333–365; (f) Murugan, R.; Scriven,
E. F. V. Aldrichimica Acta 2003, 36, 21–27; (g) Spivey, A. C.; Arseniyadis, S. Angew.
Chem., Int. Ed. 2004, 43, 5436–5441; (h) France, S.; Guerin, D. J.; Miller, S. J.;
Lectka, T. Chem. Rev. 2003, 103, 2985–3012; (i) Fu, G. C. Acc. Chem. Res. 2004, 37,
542–547.
12. Abraham, R. J.; Loftus, P. Proton and Carbon-13 NMR Spectroscopy; Heyden:
London, 1978; p 165.
13. For a review of carbamic acids, see: Petersen, U. Methoden der Organischen
Chemie (Houben-Weyl); Hagemann, H., Ed.; Georg Thieme: Stuttgart, New York,
NY, 1983; Vol. E4, pp 142–293.
We thank Stephanie M. Ng (Boston College) for X-ray crystal-
lographic studies.
Supplementary data
Contains copies of 1H and 13C NMR spectra for compounds 28,
30, 31, 35, 36 and 1H NMR spectrum of 33. Also, X-ray diffraction
data for compound 30 is included (13 pages). Supplementary data
associated with this article can be found in the online version, at
doi:10.1016/j.tet.2008.04.101.
14. (a) Xu, S.; Held, I.; Kempf, B.; Mayr, H.; Steglich, W.; Zipse, H. Chem.dEur. J.
2005, 11, 4751–4757; (b) Singh, S.; Das, G.; Singh, O. V.; Han, H. Org. Lett. 2007, 9,
401–404.
References and notes
1. (a) Kelly, T. R.; De Silva, H.; Silva, R. A. Nature 1999, 400, 150–152; (b) Kelly, T. R.;
Silva, R. A.; De Silva, H.; Jasmin, S.; Zhao, Y. J. Am. Chem. Soc. 2000, 122, 6935–
6949; (c) Kelly, T. R. Acc. Chem. Res. 2001, 34, 514–522; (d) Kelly, T. R.; Sestelo,
J. P. Molecular Machines and Motors; Sauvage, J.-P., Ed.; Structure and Bonding;
Springer: Berlin, 2001; Vol. 99, pp 19–53.
15. Kelly, T. R.; Bowyer, M. C.; Bhaskar, K. V.; Bebbington, D.; Garcia, A.; Lang, F.;
Kim, M. H.; Jette, M. P. J. Am. Chem. Soc. 1994, 116, 3657–3658.
16. See final paragraph in Ref. 3.
17. Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.; Timmers, F. J.
Organometallics 1996, 15, 1518–1520.
2. For up-to-date coverage of the work of others, see the other papers in this issue
and references therein.
3. Kelly, T. R.; Cai, X.; Damkaci, F.; Panicker, S. B.; Tu, B.; Bushell, S. M.; Cornella, I.;
Piggott, M. J.; Salives, R.; Cavero, M.; Zhao, Y.; Jasmin, S. J. Am. Chem. Soc. 2007,
129, 376–386.
18. Furniss, B. S.; Hannaford, A. J.; Smith, P. W. G.; Tatchell, A. R. Vogel’s Textbook of
Practical Organic Chemistry, 5th ed.; Longman Scientific and Technical: Essex,
UK, 1989; p 212.
19. Dieter, R. K.; Silks, L. A., III; Fishpaugh, J. R.; Kastner, M. E. J. Am. Chem. Soc. 1985,
107, 4679–4692.
4. Stabb, H. A. Angew. Chem., Int. Ed. Engl. 1962, 1, 351–367.
5. Bu¨ rgi, H. B.; Dunitz, J. D. Acc. Chem. Res. 1983, 16, 153–161.
20. Perrin, D. D.; Armarego, W. L. F. Purification of Laboratory Chemicals, 3rd ed.;
Pergamon: Oxford, England, 1988; p 121.