On the Regioselectivity of the Gould–Jacobs Reaction: Gas-Phase Versus Solution-Phase Thermolysis

Article abstract of DOI:10.1002/ejoc.202001110

A detailed investigation of the regioselectivity in the thermal cyclization of (pyridyl)aminomethylenemalonates both in the gas- and solution phase is presented. Flash vacuum pyrolysis (FVP) as a gas-phase thermolysis technique is used to study the Gould–Jacobs reaction at temperatures between 450–650 °C, while different solution-phase heating techniques (reflux, microwave, and continuous flow) were employed at 260–350 °C. Depending on the position of the substituent in the pyridine moiety and the applied thermolysis technique, the regioselectivity of the cyclization can be controlled either in favor of the kinetic (pyridopyrimidinone) or the thermodynamic (naphthyridinone) product. Under FVP conditions, 6-substituted pyridopyrimidinones were obtained in high regioselectivity, which was not demonstrated before under standard Gould–Jacobs reaction conditions. DFT calculations have been additionally performed to provide further insights into the mechanistic pathways of this specific Gould–Jacobs reaction.

Full text of DOI:10.1002/ejoc.202001110

European Journal of Organic Chemistry
Accepted Article
Accepted Article
Title: On the Regioselectivity of the Gould−Jacobs Reaction: Gas-
phase versus Solution-phase Thermolysis
Authors: Michaela Wernik, Peter E. Hartmann, Gellért Sipos, Ferenc
Darvas, A. Daniel Boese, Doris Dallinger, and C. Oliver
Kappe
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Eur. J. Org. Chem. 10.1002/ejoc.202001110
Link to VoR: https://doi.org/10.1002/ejoc.202001110
01/2020

10.1002/ejoc.202001110
European Journal of Organic Chemistry
FULL PAPER
On the Regioselectivity of the Gould−Jacobs Reaction: Gas-
phase versus Solution-phase Thermolysis
Michaela Wernik,[a],[b] Peter E. Hartmann,[a] Gellért Sipos,[c] Ferenc Darvas,[c] A. Daniel Boese,[a] Doris
Dallinger,*[a],[b] and C. Oliver Kappe*[a],[b]
[a]
M. Wernik, P. E. Hartmann, Prof. Dr. A. D. Boese, Dr. D. Dallinger, Prof. Dr. C. O. Kappe
Institute of Chemistry
University of Graz, NAWI Graz
Heinrichstrasse 28, 8010 Graz, Austria
E-mail: do.dallinger@uni-graz.at
oliver.kappe@uni-graz.at
http://goflow.at
[b]
[c]
M. Wernik, Dr. D. Dallinger, Prof. Dr. C. O. Kappe
Center for Continuous Flow Synthesis and Processing (CCFLOW)
Research Center Pharmaceutical Engineering GmbH (RCPE)
Inffeldgasse 13, 8010 Graz, Austria
Dr. G. Sipos, Dr. F. Darvas
ThalesNano Inc.
Zahony u 7, 1031 Budapest, Hungary
Supporting information for this article is given via a link at the end of the document.
Abstract: A detailed investigation on the regioselectivity in the
thermal cyclization of (pyridyl)aminomethylenemalonates both in the
gas- and solution phase is presented. Flash vacuum pyrolysis (FVP)
as gas-phase thermolysis technique is used to study the
Gould−Jacobs reaction at temperatures between 450−650 °C, while
different solution-phase heating techniques (reflux, microwave and
continuous flow) were employed at 260−350 °C. Depending on the
position of the substituent in the pyridine moiety and the applied
thermolysis technique, the regioselectivity of the cyclization can be
controlled either in favor of the kinetic (pyridopyrimidinone) or the
thermodynamic (naphthyridinone) product. Under FVP conditions, 6-
reaction, which is generally controlled by both steric and
electronic factors. The cyclization can occur on either of the ortho
positions of the aniline moiety. This is a crucial aspect, particularly
when unsymmetrically substituted anilines are employed, and
often a mixture of products is obtained.[7,8] The regioselectivity of
the thermal cyclization of (pyridyl)aminomethylenemalonates 3
(Scheme 1) is also highly dependent on the substituent on the
pyridine ring.[9,10] In solution-phase, the cyclization generally takes
place at the more nucleophilic nitrogen and exclusively provides
pyridopyrimidinone 6. However, the ring closure of the 6-methyl
derivative of 3 is an exception: it preferentially leads to
substituted pyridopyrimidinones were obtained in high regioselectivity, naphthyridinone 5, presumably because of the steric hindrance of
which was not demonstrated before under standard Gould−Jacobs
reaction conditions. DFT calculations have been additionally
performed to provide further insights into the mechanistic pathways of
this specific Gould−Jacobs reaction.
the methyl group.[9–12] Also in this case, pyridopyrimidinone 6 is
formed initially, but then rearranges to the thermodynamically
more stable naphthyridinone 5, most likely via ketene
intermediate 4.[12] The same effect of a sterically hindering 6-
methyl group in 2-hydroxy-pyridopyrimidin-4-ones was also
reported in the rearrangement to the corresponding
naphthyridinones involving
analogous to 4.[13]
a carbonyl ketene intermediate
Introduction
The Gould−Jacobs reaction is an important thermal cyclization
method for the construction of the 4-quinolone scaffold,[1] which is
an essential heterocyclic motif in antibacterial, anticancer or anti-
inflammatory drugs.[2] The original approach involves the
condensation
ethoxymethylenemalonate
anilinomethylene
electrocyclization via
of
an aniline derivative and diethyl
(Scheme 1). The resulting
malonate readily undergoes an
ketene intermediate upon further
a
heating.[3] Depending on the substitution patterns of the employed
anilines,
various
fused
ring
structures
such
as
diazoloquinolones,[4]
pyrimidopyrrolopyrimidines,[5]
Scheme 1. Regioselectivity of the Gould−Jacobs Reaction.
pyridopyrimidones,[6] etc. can be generated.
An important issue is the regioselectivity of the Gould−Jacobs
1
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202001110
European Journal of Organic Chemistry
FULL PAPER
In general, temperatures >250 °C are necessary for the
cyclization step, which typically requires high-boiling solvents,
such as diphenyl ether (Ph2O) or Dowtherm, when performed
under conventional reflux conditions.[8–11] However, long exposure
to high temperatures can be detrimental, because side-reactions
can be induced. Recently, Gould−Jacobs reactions were
demonstrated under process intensified conditions by employing
high-temperature and high-pressure (high-T/p) continuous flow
reactors.[14,15] This approach allowed the use of low-boiling
solvents, such as MeCN or THF, under supercritical (sc)
conditions which facilitated product isolation. [14–16] A solvent-free
Gould−Jacobs reaction in flow was also reported by Strauss.[17]
Another technique that can be applied for reactions possessing
extreme activation barriers is flash vacuum pyrolysis (FVP), which
is a special form of gas-phase thermolysis.[18–20] In FVP, a
substrate is sublimed through a hot tube under high vacuum and
elevated temperatures (<10-3 mbar and 300−1100 °C). Thermal
activation in such a reduced atmosphere occurs mainly by
molecule-wall collisions with very short contact times that are
typically in the range of ms−s. FVP is primarily used for
We commenced our studies toward the regioselectivity of the
Gould−Jacobs reaction with the thermolysis of 6-methyl derivative
3a in the gas phase using FVP and in solution phase employing
different heating techniques (reflux, microwave and continuous
flow). Malonates 3 were generally synthesized by heating the
respective 2-amino-pyridine derivative
1
and diethyl
ethoxymethylenemalonate (2) (see Scheme 1) under sealed-
vessel microwave batch heating (see SI).[14]
FVP optimizations (see Figure S1 for the set-up) revealed that
a temperature of 130 °C was necessary to fully sublime the
substrate, and that at 450 °C oven temperature full conversion of
3a was achieved (Table 1, entry 2). In contrast to previously
reported results for the solution-phase cyclization,[9–11] derivative
3a did not cyclize to naphthyridinone 5a, but exclusively furnished
pyridopyrimidinone 6a, which is generally considered as kinetic
product, as it is usually formed at lower temperatures (vide
supra).[11] Because of the extremely short contact times of ms−s
in FVP,[18–20] it was not surprising to generate the kinetic product,
although the reaction was performed at nominal temperatures
much higher than in solution phase (450 °C vs 260 °C). In order
to direct the cyclization to the thermodynamic product, the oven
temperature was increased to 650 °C, but only decarboxylated
intramolecular
transformations
such
as
eliminations,
rearrangements or cyclizations. Because of the extremely short
residence times and the absence of solvent/reagents, secondary
reactions are minimized. Thus, often one regioisomer is favored,
and considerably cleaner reaction profiles are obtained compared
to other thermolysis techniques.[18–20]
pyridopyrimidinone
7a
(66%)
and
decarboxylated
naphthyridinone 8a (34%) were identified (Table 1, entry 3).
Another option to obtain the thermodynamic product is to increase
the contact time inside the heated zone. In FVP this can be done
either by reducing the vacuum (Table 1, entries 4,5), packing the
tube with e.g. quartz chips or quartz wool, or reducing the inner
diameter of the hot tube (Table 1, entry 6).[18–20] None of those
approaches were successful in generating the thermodynamic
product 5a, as only decarboxylated derivatives were detected, in
addition to 6a. The formation of such decarboxylated quinolones
upon FVP of related (arylamino)methylenemalonates was also
Because of its apparent applicability to access novel
heterocycles and being a reaction requiring high levels of energy,
we considered the Gould−Jacobs reaction as a particularly
suitable transformation to be studied under FVP conditions.
Another asset that renders the thermal cyclization attractive to be
investigated in FVP, is the involvement of ketene intermediate 4
in the reaction pathway. Ketenes are common intermediates in
many FVP reactions and are typically generated by a clean
thermal decomposition of suitable precursor molecules.[18–20]
Whereas ample studies on the generation of imidoylketenes
similar to 4 from various precursors and their cyclization in FVP
exist,[21] there is comparatively little literature precedence
conducting Gould−Jacobs reactions using FVP conditions, and
questions relating to the regioselectivity were not addressed in
those studies.[22–24]
Table 1. Gould−Jacobs Reaction of 3a under FVP Conditions.a
Herein, we present an in-depth investigation on the
Gould−Jacobs reaction with
a
particular focus on the
T
p
t
ID QT
[cm][d]
3a
6a
7a
8a
Entry
regioselectivity of the thermal cyclization of substituted
(pyridyl)aminomethylenemalonates 3 under various conditions in
solution- and gas phase. By performing the reaction under FVP
conditions, we were able to selectively generate and isolate the
kinetic pyridopyrimidinone product 6 from the 6-methyl derivative
of 3, which has not been reported previously using standard
Gould-Jacobs conditions. In addition to the experimental studies,
differential scanning calorimetry (DSC) measurements and DFT
calculations have been performed to provide further insight to the
specific mechanistic pathways of the Gould−Jacobs reaction
involving the 5- and 6-methyl substituted (pyridyl)aminomethylene
derivatives of 3.
[°C][b] [mbar] [h][c]
[%][e]
[%][e]
[%][e]
[%][e]
1
2[f]
3
350
450
650
450
450
550
10-3
10-3
10-3
10-1
1
2
2
2
4
6
2
3.5
3.5
3.5
3.5
3.5
1
16
-
84
100
-
-
-
-
-
-
66
3
34
1.5
13
-
4
-
95.5
59
5[g]
6
-
28
13
10-3
-
87
[a] Conditions: 100 mg of 3a, preheater: 130 °C. [b] Temperature refers to the
set oven temperature in the main zone (zones 2: + 20 °C), see SI for further
details and screening conditions. [c] Time refers to sublimation time. [d] ID QT:
inner diameter of a quartz tube with 60 cm length. [e] Selectivity determined by
1H NMR. [f] 93% isolated yield of 6a + 4% recovered 3a from QT and sublimation
flask. [g] 19% isolated yield of product mixture 6a, 7a, 8a + 63% recovered 3a
+ 6a from QT and sublimation flask.
Results and Discussion
Thermolysis of (6-Methylpyridyl)malonate 3a
2
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202001110
European Journal of Organic Chemistry
FULL PAPER
reported by Vera and co-workers.[24] In general, the products were
isolated by simply rinsing the U-tube cooling trap (Figure S2) with
DCM and evaporation of the solvent. At the optimized FVP
conditions (450 °C, 10-3 mbar and 2 h sublimation time), 6a was
isolated in 93% yield (>98% purity by 1H NMR), which
corresponds to a 97% mass balance. It has to be noted, that
pyridopyrimidinone 6a was only isolated previously by performing
or by the addition of small amounts of water.[14,15] In any case, the
use of dry Ph2O proved to be beneficial for microwave
experiments at 290 °C to reduce the amount of decarboxylated
naphthyridinone to 5% (Table 2, entry 2). In both heating
approaches, 5a was isolated in 77% yield and >98% purity after
precipitation with cyclohexane.
a
Gould−Jacobs type reaction under acidic conditions
(POCl3/PPA) which follows a different mechanistic pathway.[25]
We next turned our attention to the solution-phase thermolysis
of 3a, where the thermodynamic naphthyridinone 5a is the
expected product.[8–11] Indeed, under standard open-vessel reflux
conditions employing Ph2O as solvent (bp 259 °C) and a reaction
time of 30 min, full conversion of 3a was achieved to furnish 5a
as the main product (87%), together with 8% of 6a and 1% of
decarboxylated naphthyridinone 8a (Table 2, entry 1). For the first
10 min, the initial reaction rates for the generation of 5a and 6a,
respectively, were similar, but as the reaction progressed, the
amount of 6a decreased over time while 5a rapidly increased up
to 87% (Figure 1a). The reaction stalled after 30 min, as the
transformation of isomer 6a to the thermodynamically more stable
isomer 5a (via intermediate I, see Scheme 2) did not proceed
further. In order to investigate the transformation at higher
temperatures, we performed the Gould−Jacobs reaction of 3a
employing sealed-vessel microwave heating,[9] which allowed to
superheat Ph2O up to 290 °C (Figure S3). Although initially being
faster at 290 °C compared with standard reflux conditions (48%
vs 27% 5a after 10 min), full conversion could only be
accomplished after 30 min (Figures 1b and S3). Interestingly,
under the same conditions as described above (260 °C, 30 min),
Figure 1. Reaction time screening for the solution-phase thermolysis of 3a
under a) open-vessel reflux heating and b) sealed-vessel microwave heating at
290 °C employing Ph2O as solvent (bp 259 °C). The selectivity was determined
by HPLC UV-Vis area% at 320 nm.
Table 2. Gould−Jacobs Reaction of 3a in Solution Phase.a
In order to mimic the FVP conditions more closely in solution
phase with respect to temperature and reaction time, additional
investigations were performed in
a
dedicated high-
T
[°C]
5a
6a
8a
Others
[%][b]
Entry

[%][b]
[%][b]
[%][b]
temperature/pressure continuous flow reactor (Figure S5). This
high-T/p solution-phase method termed “flash flow pyrolysis”
(FFP) usually leads to similar results as obtained with FVP.[15,16]
Since the product-Ph2O mixture would solidify upon cooling in the
heat exchanger, we selected MeCN as a suitable solvent for the
FFP. MeCN becomes a supercritical fluid under the envisaged
1
2
OV
SV
260
290
87
87
8
5
1
5
4
3
[a] Conditions: 100 mg of 3a in 3 mL (0.1 M) of dry Ph2O, OV: open-vessel
reflux, SV: sealed-vessel microwave heating. [b] Selectivity determined by
HPLC UV-Vis area% at 320 nm.
high-T/p conditions, therefore
a decrease of the nominal
residence time can be expected (Figure S6).[16,27] An 8 mL
stainless steel coil was employed and the back pressure was set
to 100 bar. By pumping a 0.05 M solution of 3a in MeCN with 8
mL/min (ca. 25 s actual residence time) at 350 °C, nearly full
conversion of 3a was achieved, but a complex mixture containing
all four products 5a−8a was obtained (Figure 2, Figure S7).
Unexpectedly, along with decarboxylated pyridopyrimidinone 7a
(33%), decarboxylated malonate 9a (35%) was identified as the
other main product. Any solvent effect was ruled out, as a similar
outcome as in Ph2O was obtained when the reaction was
performed under microwave heating.[28] The presence of an
acrylate similar to 9a was also reported by Vera and co-workers
in their FVP studies on the Gould−Jacobs reaction,[24] but never
has been observed in solution before. A mechanistic pathway to
decarboxylated quinolones involving acrylates was ruled out by
Djuric and co-workers in their continuous flow studies. Because
only a 14% conversion of 3a was achieved (Figure S4). We
propose that under sealed-vessel conditions, the formed EtOH
cannot be efficiently removed from the reaction mixture and thus
prevents progression of the ketene cyclization.[26] As a result, a
higher temperature was necessary to shift the equilibrium to the
product side. Furthermore, a higher amount of 8a compared to
open-vessel reflux was detected (10% vs 1%, see also Figure 1).
This difference could be either attributed to the higher
temperatures and/or the presence of water in the solvent, as a
detectable amount of decarboxylated 8a was only noticed at
temperatures ≥280 °C. Similar observations have been reported
for the thermal cyclization of related malonates by Djuric and
Lengyel in continuous flow, where the decarboxylation was
favored either at higher temperatures and longer reaction times
3
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202001110
European Journal of Organic Chemistry
FULL PAPER
Figure 2. Gould−Jacobs reaction of 3a in continuous flow mode (FFP). Conditions: 0.05 M solution of 3a in anhydrous MeCN; flow rate: 8 mL/min (actual residence
time: 25 s); 8 mL stainless steel coil (SS, inner diameter 1 mm) inside the high-T/p flow reactor; the heat exchanger was submerged in a water bath to keep the
temperature constant. BPR: back pressure regulator. See Figure S5 for the set-up.
anhydrous MeCN (30 ppm H2O) was employed in our study, we
propose that 9a is generated by loss of ethylene and CO2, as
described by Vera and co-workers.[24] It has been reported that
under highly diluted conditions, acrylates such as 9a cyclize to
give the two isomeric structures of substituted 4-
hydroxyquinolines.[29] To confirm this, 9a was subjected to
microwave heating in Ph2O at 290 °C, and indeed, the
decarboxylated derivatives 7a and 8a were observed in a ratio of
1:2.7 after 30 min (Figure S8). For the thermolysis of 3a in FFP
we assume that the higher amount of 9a was detected in the
reaction mixture, because of the short reaction time of only 25 s.
Under microwave heating for 30 min on the other hand, 9a was
already converted to 8a (presumably via rearrangement 7a → 8a,
see also Table 5) and thus was only traceable in minor amounts
(1‒8%).[28]
interactions, therefore electron donating (EDG) and electron
withdrawing groups (EWG) were installed on malonates 3 (Table
3). All halides, except the iodo derivative, provided mixtures in
FVP with a preference for the corresponding pyridopyrimidinones
6d−f. The 6-iodo derivative 3g behaved like the 6-methyl analog
3a and exclusively furnished 6g.[33] We assume that the 6-iodo
derivative 6g is more stable while the other halogen derivatives
seem to have a lower energy barrier for 6 → 5 and thus are more
Table 3. Substrate Scope of the Thermolysis of Pyridylmalonate Derivatives 3
in Gas Phase (FVP) and Solution Phase.
Thermolysis of Other Substituted Pyridylmalonates
To further study the substitution effect on the regioselectivity, the
cyclization of 5-methyl derivative 3b and 4,6-dimethyl derivative
3c were investigated in more detail in both gas- and solution
phase. Not surprisingly, pyridopyrimidinone 6b was selectively
obtained when 3b was processed under the same FVP conditions
as described above (Table 3, entry 2). In agreement with literature
data, and in contrast to 3a, solely 6b was formed when 3b was
refluxed in Ph2O for 15 min.[9,10] The corresponding
naphthyridinone was not observed in any of the experiments. A
0.3 M solution of 3b in MeCN was also subjected to continuous
flow conditions at 370 °C with a nominal residence time of 1 min.
A much cleaner reaction profile compared with 3a was obtained,
and in accordance with the literature, 6b was the main product
(79%), together with decarboxylated pyridopyrimidinone 7b
(14%).[14] For the 4,6-dimethyl derivative 3c it was reported that
the corresponding Meldrum´s acid derivative gave the
pyridopyrimidinone isomer when heated at 240 °C, but the
naphthyridinone at 260 °C.[30–32] Also in this case,
pyridopyrimidinone 6c was the exclusive product in FVP (Table 3,
entry 3). In solution phase on the other hand, a predominant
preference for 6c was not observed at any evaluated temperature
(Table S1). Similarly, as for 3a, naphthyridinone 5c was the major
regioisomer. However, malonate 3c demands temperatures
≥260 °C to favor the thermodynamic product 5c (Table 3), as in
refluxing Ph2O only a ca 2:1 mixture of 5c:6c could be obtained
(Table S1).
FVP[a]
OV[b]
3
5
6
t
3
5
6
Others[
%]
Entry
R
[%] [%] [%] [min] [%] [%] [%]
1
2
3a: 6-Me
3b: 5-Me
3c: 4,6-diMe
3d: 6-F
-
-
-
-
100
100
100
45
30
15
15[c]
10
10
10
10
10
90
20
-
-
87
-
8
98
3
-
5
2
3
-
-
-
85
20
14
11
15
93
80
-
12
-
4
16
10
28
-
39
23
12
-
80
86
86
79
-
5
3e: 6-Cl
67
-
-
6
3f: 6-Br
60
-
1
7
3g: 6-I
100
86
-
6
8
3h: 6-OMe
-
14
15
-
-
7
9
3i: 6-CF
7
78
-
-
20
2
3
10
3j: H
-
100
3
95
[a] FVP Conditions: 100 mg of 3, preheater: 75-140 °C, oven temperature in the
main zone: 450 °C (zones 2: + 20 °C, see SI for details), 3-5×10-3 mbar, 2 h
sublimation time, quartz tube with 30 cm length and 3.5 cm ID, selectivity
determined by 1H NMR. Only products with 100% selectivity were isolated (see
Experimental Section). [b] OV Conditions: 100 mg of 3 in 3 mL of dry Ph2O,
open-vessel reflux at 260 °C, selectivity determined by HPLC UV-Vis area%.
Products 5d-5g were not isolated. [c] Sealed-vessel microwave heating at
290 °C.
In addition, we also wanted to investigate the influence of the
substituent in the 6-position with respect to steric and electronic
4
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202001110
European Journal of Organic Chemistry
FULL PAPER
susceptible to this transformation. Under open-vessel reflux
heating, all halogenated derivatives 3d−g showed decomposition
after 10 min. This phenomenon was also reported by Lappin for
the 6-bromo derivative 3f.[9] Malonate 3h with an OMe as
EDGgroup furnished 6h as the main isomer in FVP and
naphthyridinone 5h in solution after 10 min. The electron
withdrawing 6-CF3 derivative 3i behaved similarly in both FVP and
solution furnishing 6i and 5i as the main isomers, respectively.
Unsubstituted pyridylmalonate 3j performed exactly as the 5-
methyl derivative 3b (Table 3, entry 10). This study led us to the
assumption that the electronic nature of the substituent in position
6 seems not to have such a great impact on the regioselectivity
and that it is rather the steric hindrance of the 6-substituent that
facilitates the rearrangement of the kinetic (6) to the
thermodynamic (5) product (vide infra).[11,32] It has to be stressed
again, that under standard Gould−Jacobs reaction conditions, 6-
substituted pyridopyrimidinones 6 were never obtained in this
high regioselectivity as demonstrated herein using FVP.
of 6a could be achieved at 290 °C after 10 min whereby 5a was
obtained with 96% selectivity (Table 4, entry 2). Here, no EtOH is
being released that could inhibit an equilibrium shift under sealed-
vessel conditions (see also Scheme 2). According to our findings
reported above, this rearrangement only occurred when 6a was
heated in solution. Subjecting 6a to FVP at 450 °C only furnished
recovered starting material. At 550 °C, an additional amount of
20% of decarboxylated 7a was detected (Table 4). The
rearrangement 7a → 8a required the higher temperature of
290 °C in a preparative experiment in order to achieve an 88%
conversion to the naphthyridinone 8a (Table 5). It seems that an
even higher temperature would be needed in order to fully
rearrange 7a to 8a by shifting the equilibrium toward intermediate
V (see Scheme 3). As anticipated from the DSC analyses, and in
accordance with the literature,[35–37] 5-substituted 6b and
unsubstituted 6j did not undergo rearrangement to the
corresponding naphthyridinones. This observation explains why
exclusively pyridopyrimidinones are provided as products in
solution-phase thermolysis (see Table 3, entries 2 and 10).
Importantly, heating of pure naphthyridinone 5a under similar
conditions provided only recovered 5a.
Rearrangement Studies
In order to gain more insight into the generation of
naphthyridinone derivatives 5, we examined the rearrangement of
pyridopyrimidinones 6a, 6b, 6j and 7a, which were isolated from
FVP experiments,[34] in more detail. In addition to experimental
studies, DSC measurements were also performed. DSC analyses
displayed exotherms for the 6-substituted derivatives 6a and 7a
in similar ranges from 235−290 °C (Figure 3a,b), indicating the
rearrangement to take place. For 5-substituted 6b and
unsubstituted 6j, no exotherms were detected (Figure 3c,d),
therefore, also no rearrangement was expected.
Table 4. Rearrangement Studies of Pyridopyrimidinone 6a.a
T
[°C]
t
6a
[%]
5a
[%]
7a
[%]
8a
[%]
Others
[%]
Entry

[min]
1
2
3
4
OV
SV
260
290
450
550
40
10
6
-
90
96
-
-
-
2
2
-
2
2
-
FVP
FVP
120
120
100
80
-
-
20
-
-
[a] Solution-phase conditions: 100 mg of 6a in 3 mL of dry Ph2O, OV: open-
vessel reflux, SV: sealed-vessel microwave heating, selectivity determined by
HPLC UV-Vis area% at 320 nm. FVP conditions: 100 mg of 6a, preheater:
130 °C, temperature refers to the set oven temperature in the main zone (zones
2: + 20 °C, see SI for details), 3-5×10-3 mbar, quartz tube with 60 cm length and
3.5 cm ID, selectivity determined by 1H-NMR.
Table 5. Rearrangement Studies of Pyridopyrimidinone 7a.a
T
[°C]
7a
[%]
8a
[%]
Others
[%]
Figure 3. DSC measurements for the rearrangement of (a) pyridopyrimidinone
6a (exotherm: 235-290 °C, max. at 266 °C), (b) pyridopyrimidinone 7a
(exotherm: 236-273 °C, max. at 258 °C), (c) pyridopyrimidinone 6b, and (d)
pyridopyrimidinone 6j
Entry

1
2
OV
SV
260
290
81
8
19
88
-
4
The rearrangement of 6a was confirmed in a preparative
experiment by heating 6a in Ph2O at 260 °C and 290 °C,
respectively. At 260 °C under open-vessel reflux conditions,
similar results as in the solution-phase thermolysis of 3a were
obtained (6% vs 8% of 6a remaining). In contrast, full conversion
[a] Conditions: 100 mg of 7a in 3 mL of dry Ph2O, OV: open-vessel reflux, SV:
sealed-vessel microwave heating, selectivity determined by HPLC UV-Vis
area% at 320 nm.
5
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202001110
European Journal of Organic Chemistry
FULL PAPER
Theoretical Mechanistic Investigations
the experimental solution-phase thermolysis observations made
by us and Hermecz.[11,35]
To shed light on the diverging regioselectivity of the
Gould−Jacobs reaction in gas phase and in solution,
computational DFT studies were performed employing various
functionals and basis sets (for details see the Experimental
Section and the SI).[38] The data displayed were calculated at the
BMK[39]-D3[40]/def2-TZVPPD[41] level of theory and are given for
450 °C in gas-phase except where otherwise noted.
The formation of intermediate I is followed by intramolecular
proton transfer steps toward 5a (Scheme 2). These exhibit a
considerable energy barrier (Gǂgas = 237.9 kJ/mol) and, hence,
constitutes the rate determining step. The lowest energy barrier
for the rearrangement of 6a to 5a follows a pathway via
intermediate II2 and leads to the less stable enol-form of 5a2 (see
Scheme 2 and Figure S9). The 1,5-H shift in I  II2 was reported
previously by Wentrup and co-workers with similar activation
barriers for analogous 1,5- and 1,7-naphthyridinone derivatives.[8]
Tautomerization of the enol-form 5a2 to the more stable keto-form
5a3 is probably facilitated only once the FVP-products are
collected in a solvent. In solution, the participation of bimolecular
tautomerization steps has been proposed,[24] which would make
(6-Methylpyridyl)malonate 3a can easily generate ketene 4a
via TS1by releasing EtOH (Scheme 2). Once 4a is formed, it can
undergo a bond rotation around the imine double-bond (TS2),
after which the ketene moiety is directly captured by the nitrogen
of the pyridyl-moiety, forming the kinetic product 6a. In contrast to
DFT calculations performed by Vera and co-workers on
(arylamino)methylenemalonates,[24] we did not observe any
isomeric ketene intermediate between 4a and 6a. We ascribe this
to the presence of the pyridyl substituent, as in their studies a p-
substituted phenyl moiety was in place, which concomitantly also
ruled out any regioselectivity issues. Subsequently, kinetic
product 6a can rearrange via TS3 to the high-energy intermediate
I (Scheme 2), which tautomerizes to the thermodynamic product
5a. Importantly, our calculations indicate that ketene 4a cannot
directly cyclize to intermediate I, as we could not locate a
corresponding intermediate or transition state. Regardless of the
basis set or functional employed for the geometry optimization,
even for B3LYP[43]-D3/def2-TZVPPD, the corresponding
structures always revert to the stationary points belonging to the
pathway leading to 6a. This observation provides a strong
indication that 5a can only be formed from 6a via TS3 in a ring
opening–ring closure reaction, being in perfect agreement with
the generation of intermediate I (Gǂ = 159.4 kJ/mol; 260 °C
sol
in Ph2O) the rate determining step. The involvement of the solvent
in intermolecular proton transfers would therefore promote the
formation of 5a. However, our experimental findings in the
solution-phase thermolysis of 3a do not support solvent
participation to play an important role, as the reaction outcome
employing biphenyl as solvent was comparable to Ph2O.[44] We
therefore deduce that the reaction time is the governing factor for
the formation of 5a. Because of the short contact times of ms–s
in FVP and the equilibrium that forms between 6a and high-
energy intermediate I (Scheme 2), the high energy barriers of the
intramolecular proton transfer steps from 6a toward 5a cannot be
overcome. Thus, intermediate I falls apart and regenerates 6a
rather than forming 5a.
Scheme 2. Kinetic versus thermodynamic pathways in the Gould−Jacobs reaction of 3a.[41] The data given are G values [kJ/mol] calculated with BMK[39]-D3/def2-
TZVPPD (gas phase, 450 °C) and are referenced to the lowest energy conformer of 3a (3a1); See Figures S9 and S13 for more details.
6
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202001110
European Journal of Organic Chemistry
FULL PAPER
The reaction pathways toward decarboxylated products 7a, 8a
and acrylate 9a were also examined. The two pathways depicted
in Scheme 3 provided an energetically reasonable mechanistic
proposal. As previously reported,24 9a could be formed via initial
decarboxylation of malonate 3a (pathway A). Additionally, we
considered the formation of 7a and 8a via decarboxylation of 6a
and 5a (pathway B). Regardless at which stage the
decarboxylation occurs, the corresponding rate-determining step
is the detachment of ethene, yielding energetic barriers in the
range of 240-260 kJ/mol. The initial decarboxylation of 3a is
irreversible because of the loss of ethene (barrier of reverse
reaction: Gǂ = 303.7 kJ/mol), leading to low-energy acrylate 9a,
which, upon loss of EtOH, forms ketene 10a. EtOH can easily be
4a exhibits the lowest energetic barrier (Gǂgas = 124.1 kJ/mol,
Scheme 3). At 550 °C, 7a can be formed in principle either via
initial decarboxylation of 3a (pathway A) or decarboxylation of the
kinetic product 6a (pathway B). Compared to the transformation
6a  5a (Gǂ
= 237.9 kJ/mol) the decarboxylation to 7a
gas
exhibits a higher energetic barrier (Gǂ
= 262.9 kJ/mol).
gas
However, the former is hindered by the abovementioned
equilibrium involving intermediate I, while the latter is irreversible
because of the loss of ethene and CO2 and, hence, requires much
shorter contact times. Therefore, at higher temperatures in FVP
7a is generated rather than 5a. At more elevated temperatures
(650 °C, Table 1), any formed 6a decarboxylates to 7a.
Additionally, 8a1 is obtained. This can be attributed to an
analogous equilibrium between 7a and the decarboxylated
congener of intermediate I (intermediate V, Scheme 3), which,
given the short contact times, can be overcome only at 650 °C.
Finally, the key steps of our proposed mechanism (Scheme
2) were reinvestigated also for the 5-methyl (3b) and 6-fluoro (3d)
congeners of malonate 3a. Under FVP conditions, both 3a and 3b
only lead to their corresponding kinetic products 6a and 6b, while
the 6-fluoro congener 3d forms 6d and the thermodynamic
product 5d in an approximate 1:1 ratio (Table 3). This can be
attributed to the energy barriers for the rearrangement of 6 to 5
recaptured by ketene 10a (barrier of reverse reaction: Gǂ
=
gas
42.5 kJ/mol), leading to an equilibrium. The same equilibrium
exists for the formation of ketene 4a from 3a (barrier of reverse
reaction: Gǂgas = 66.5 kJ/mol, pathway B). These two equilibria
render a rationale why 9a is a main product in FFP (Figure 2) and
formation of 5a is slower in sealed vessel conditions at lower
temperatures (Figure S4).
The computational data also provided an explanation for the
different outcome of the FVP of 3a at different temperatures (see
Table 1). Only 6a is formed at 450 °C, because the pathway via
Scheme 3. Decarboxylation pathways in the Gould−Jacobs reaction of 3a. The data given are G values [kJ/mol] calculated with BMK[39]-D3/def2-TZVPPD (gas
phase, 450 °C) and are referenced to the lowest energy conformer of 3a (3a1); See Figures S9, S12 and S14 for more details.
7
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202001110
European Journal of Organic Chemistry
FULL PAPER
(Figure S17): While the 6-fluoro congener 6d exhibits a Gǂ
MHz. Chemical shifts (δ) are expressed in ppm downfield from TMS as
internal standard. The letters s, d, t, q, and m are used to indicate singlet,
doublet, triplet, quadruplet, and multiplet. Analytical HPLC analysis was
carried out on Shimadzu instrument using a C18 reversed-phase (RP)
analytical column (150 mm × 4.6 mm, particle size 5 μm) at 37 °C using
mobile phases A (H2O/MeCN (90:10 v/v) + 0.1% TFA) and B (MeCN +
0.1% TFA) at a flow rate of 1.5 mL/min. The following gradient was applied:
linear increase from 30% B to 100% B within 10 min. Melting points were
obtained on a Stuart melting point apparatus in open capillary tubes. High-
resolution mass spectrometry was performed on an Agilent TOF LC/MS
instrument using ESI (Dual AJS ESI) in positive mode. The ionization
parameters were set as following: Gas temp. (N2): 300 °C, drying gas: 5 L
min-1, nebulizer: 40 psig, fragmentor: 150 V, skimmer: 65 V, OCT 1 RF
Vpp: 750 V, Vcap: 1600 V. The data acquisition window was set to 100-
1,600 m/z with 1 spectra/s. The reference masses of 121.050873 and
922.009798 were continuously injected for lock mass calibration. Data was
acquired with MassHunter Workstation Rev.B.05.01SP2. Differential
scanning calorimeter (DSC) measurements were performed on a Netzsch
instrument between 20 and 400 °C with a heating rate of 5 °C/min using
aluminum crucibles. N2 at a flow rate of 50 mL/min was used as purge gas
and inert gas. FVP experiments were performed in the Flash Pyrolysis
Platform from ThalesNano. The Phoenix Flow Reactor was employed for
the high-T/p continuous flow experiments. Microwave irradiation
experiments were carried out in a Monowave 400 single-mode microwave
reactor from Anton Paar using 10 mL Pyrex vials. The reaction
temperature was controlled by an external infrared sensor while the
internal temperature was simultaneously monitored by a fiber optic probe.
The stated temperatures in the main text and Tables correspond to internal
temperatures. Reaction times refer to hold times at the temperature
indicated. The stirring speed was set to 600 rpm. Column chromatography
was carried out using a Biotage Isolera automated flash chromatography
system. All compounds synthesized herein, except for compounds 3d, 3g,
3h, 5h, 6g and 9a, are known in the literature. Products were characterized
by 1H NMR and identified by comparison of the spectra with those reported
in the literature. Proof of purity was obtained by 1H NMR, and if not
otherwise noted, the products have a purity of ≥95%.
gas
of only 198.1 kJ/mol and thus readily rearranges to 5d, the
pathways of 6a and 6b show considerably higher energy barriers
of 237.9 kJ/mol and 275.2 kJ/mol, respectively, preventing its
rearrangement in FVP. The difference in energy for the latter two
originates from the relative stability of 6a (Ggas= -0.2 kJ/mol) and
6b (Ggas = -34.9 kJ/mol), whereas the transition states are similar
in energy. However, in solution, 6b remains with respect to
rearrangement to 5b (vide supra and Figure 3), which can be
attributed to the different energy barriers (6a  5a: Gǂ
=
sol
238.2 kJ/mol; 6b  5b: Gǂsol = 276.1 kJ/mol; 260 °C in Ph2O).
Therefore, as anticipated, we conclude the controlling factors for
the rearrangement 6  5 to be the steric hindrance stemming
from the substituent in 6-position of the pyridyl moiety and the
accompanying different energy barriers.
Conclusion
We have performed an in-depth study on the regioselectivity of
the
Gould–Jacobs
reaction
of
substituted
(pyridyl)aminomethylenemalonates 3 both in gas- and solution
phase. The thorough experimental work was also supported by
mechanistic studies employing DFT calculations. For the methyl
substituted malonates 3a–c, the gas-phase thermolysis at 450 °C
using FVP exclusively led to the formation of pyridopyrimidinones
6 and is thus independent of the position of the substituent on the
pyridyl moiety. In solution phase, on the other hand, the
regioselectivity is highly dependent on the substituent position, as
the cyclization is controlled by steric characteristics. The batch
thermolysis of 5-methyl substituted 3b exhibited a remarkably
clean reaction profile with a 98% selectivity for the formation of
pyridopyrimidinone 6b, exactly as in gas phase. However, an
important difference was observed for 6-methyl-pyridyl malonate
3a: the cyclization proceeded with a major preference for the
naphthyridinone 5a. Pyridopyrimidinones 6 as kinetic products
are favored in FVP because of the extremely short contact times
(0.3 s). However, in solution at typical reaction times of 30 min
DFT calculations were performed with TURBOMOLE 7.4[45] and ORCA
4.2.1.[46] Geometries were optimized with RI-PBE-D3[47]/def2-SVP[48]
(+COSMO solvation model). Single points were calculated with B3LYP[43]
D3/def2-TZVPPD[48]//RI-PBE-D3/def2-SVP BMK[39]-D3/def2-
and
-
TZVPPD//RI-PBE-D3/def2-SVP, where solvent and temperature effects
were obtained from RI-PBE-D3/def2-TZVPPD//RI-PBE-D3/def2-SVP
calculations. For further details, see the Supporting Information. The data
presented are the BMK-D3/def2-TZVPPD//RI-PBE-D3/def2-SVP
G(+COSMO) values except where stated otherwise.
and temperatures ≥260 °C,
thermodynamically more stable naphthyridinone 5, but solely
when substituent is present at the 6-position. This
6 then rearranges to the
a
rearrangement pathway was also corroborated by DSC analysis.
DFT calculations provided a mechanistic rationale for the
formation of the kinetic and the thermodynamic product and the
distinct behavior of differently substituted substrates in both gas-
and solution phase.
General Procedure for the Preparation of Malonates 3a−j. Into a 10 mL
microwave process vial, equipped with a stir bar, were placed the
corresponding
2-aminopyridines
(1
mmol),
diethyl
ethoxymethylenemalonate (1.05 equiv, 1.05 mmol) and 1.2 mL of EtOH.
The vial was sealed with a snap cap and heated for 10−40 min at 130 °C
in a microwave reactor. Upon cooling to room temperature or in an ice bath,
the product precipitated. After filtration, washing with ice cold EtOH and
drying at 40 °C under vacuum, the products were obtained in 46−90% yield.
Overall, it was demonstrated that the regioselectivity of the
Gould–Jacobs reaction can be controlled either in favor of the
kinetic
(pyridopyrimidinone)
or
the
thermodynamic
(naphthyridinone) product, whereby the direction of the cyclization
is dependent on the position of the substituent in the pyridine
moiety and the applied thermolysis technique. In particular, 6-
substituted pyridopyrimidinones were generated in high
regioselectivity using FVP, which was not evidenced previously
under standard Gould−Jacobs reaction conditions.
Diethyl 2-(6-methylpyridin-2-yl)aminomethylenemalonate (3a). 10 min
heating. 90% yield (251 mg). mp 106−112 °C (lit.[9] 113−114 °C). 1H NMR
(300 MHz, DMSO-d6) δ 10.75 (d, J = 13.3 Hz, 1H), 9.05 (d, J = 13.3 Hz,
1H), 7.68 (t, J = 7.8 Hz, 1H), 7.18 (d, J = 8.1 Hz, 1H), 7.01 (d, J = 7.4 Hz,
1H), 4.26 – 4.11 (m, 4H), 2.44 (s, 3H), 1.26 (q, J = 7.0 Hz, 6H). The spectral
data are in agreement with the previously published values.[10]
Diethyl 2-(5-methylpyridin-2-yl)aminomethylenemalonate (3b). 15 min
heating. 69% yield (193 mg). mp 112−115 °C (lit.[9] 112−113 °C). 1H NMR
(300 MHz, DMSO-d6) δ 10.76 (d, J = 13.3 Hz, 1H), 9.03 (d, J = 13.3 Hz,
1H), 8.19 (d, J = 2.1 Hz, 1H), 7.63 (dd, J = 8.3, 1.9 Hz, 1H), 7.30 (d, J =
Experimental Section
General Methods. 1H NMR spectra were recorded on a Bruker 300 MHz
instrument. 13C NMR spectra were recorded on the same instrument at 75
8
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202001110
European Journal of Organic Chemistry
FULL PAPER
8.3 Hz, 1H), 4.25 – 4.11 (m, 4H), 2.26 (s, 3H), 1.29 – 1.21 (m, 6H). The
Synthesis of Ethyl 7-methyl-4-oxo-1,4-dihydro-1,8-naphthyridine-3-
carboxylate (5a). 3a (100 mg, 0.36 mmol) was added to 3 mL of refluxing
diphenyl ether. After refluxing for 30 min, the solution changed from
colorless to dark brown, and upon rapid cooling to room temperature and
addition of 15 mL of cyclohexane a brown precipitate formed. After filtration,
the precipitate was dissolved in 15 mL of DCM and washed with a
saturated NaHCO3 (2 × 20 mL) solution. Drying over Na2SO4 and removal
of the solvent under reduced pressure led to compound 5a as a beige
powder in 77% yield (64 mg). mp 255−260 °C (lit.[35] 274−277 °C). 1H NMR
(300 MHz, DMSO-d6) δ 12.64 (brs, 1H), 8.45 (s, 1H), 8.39 (d, J = 8.1 Hz,
1H) 7.36 (d, J = 8.1 Hz, 1H), 4.21 (q, J = 7.1 Hz, 2H), 2.59 (s, 3H), 1.27 (t,
J = 7.1 Hz, 3H). 13C NMR (75 MHz, DMSO-d6) δ 173.7, 164.3, 163.0, 149.4,
145.5, 135.5, 121.2, 119.7, 111.0, 59.7, 24.3, 14.3. The spectral data are
in agreement with the previously published values.[10]
spectral data are in agreement with the previously published values.[10]
Diethyl 2-(4,6-dimethylpyridin-2-yl)aminomethylenemalonate (3c). 20
min heating. 51% yield (148 mg). mp 96−98 °C (lit.[49] 96.8−98.6 °C). 1H
NMR (300 MHz, DMSO-d6) δ 10.67 (d, J = 13.3 Hz, 1H), 9.02 (d, J = 13.3
Hz, 1H), 6.99 (s, 1H), 6.85 (s, 1H), 4.25 – 4.10 (m, 4H), 2.39 (s, 3H), 2.25
(s, 3H), 1.25 (q, J = 7.1 Hz, 6H). 13C NMR (75 MHz, DMSO-d6) δ 167.0,
165.0, 156.7, 150.2, 149.9, 148.3, 120.3, 110.0, 94.6, 59.8, 59.6, 23.8,
20.5, 14.3, 14.2.
Diethyl 2-(6-fluoropyridin-2-yl)aminomethylenemalonate (3d). 30 min
heating. 59% yield (166 mg). mp 121−124 °C. 1H NMR (300 MHz, DMSO-
d6) δ 10.80 (d, J = 13.0 Hz, 1H), 8.79 (d, J = 13.0 Hz, 1H), 7.96 (q, J = 8.1
Hz, 1H), 7.34 (dd, JH,H = 7.9, JH,F = 2.0 Hz, 1H), 6.86 (dd, JH,H = 7.9, JH,F
2.3 Hz, 1H), 4.26 – 4.12 (m, 4H), 1.25 (q, J = 7.2 Hz, 6H). 13C NMR (75
MHz, DMSO-d6) δ 166.5, 164.7, 162.0 (d, JC,F = 238.9 Hz), 149.8 (d, JC,F
14.2 Hz), 147.0, 144.5 (d, JC,F = 8.3 Hz), 109.8 (d, JC,F = 4.1 Hz), 103. 7 (d,
JC,F = 35.5 Hz), 96.6, 60.1, 59.9, 14.3, 14.1. HRMS (ESI-TOF): m/z [M +
H]+ Calcd for [C13H15FN2O4 +H]+ 283.108862; Found 283.108811.
=
Synthesis of Ethyl 5,7-dimethyl-4-oxo-1,4-dihydro-1,8-naphthyridine-
3-carboxylate (5c). Into a 10 mL microwave process vial, equipped with a
stir bar, were placed 3c (50 mg, 0.17 mmol) and 1.5 mL of diphenyl ether.
Diphenyl ether was previously made water free by heating to 240 °C. The
vial was sealed with a snap cap and heated for 15 min at a set temperature
(IR) of 300 °C in a microwave reactor. Upon rapid cooling to room
temperature and addition of 8 mL of cyclohexane a brown precipitate
formed. After filtration, the precipitate was dissolved in 8 mL of DCM and
washed with a saturated NaHCO3 (2 × 10 mL) solution. Drying over
Na2SO4 and removal of the solvent under reduced pressure led to
compound 5c as a beige powder in 46 % yield (20 mg). mp 232−234 °C
(lit.[35] 232−234 °C). 1H NMR (300 MHz, DMSO-d6) δ 12.34 (d, J = 6.3 Hz,
1H), 8.31 (d, J = 6.6 Hz, 1H), 7.05 (s, 1H), 4.18 (q, J = 7.1 Hz, 2H), 2.75
(s, 3H), 2.48 (s, 3H), 1.26 (t, J = 7.1 Hz, 3H). 13C NMR (75 MHz, DMSO-
d6) δ 176.0, 164.4, 161.3, 151.3, 150.7, 144.0, 123.4, 118.3, 112.2, 59.6,
23.8, 22.4, 14.3.
=
Diethyl 2-(6-chloropyridin-2-yl)aminomethylenemalonate (3e). 30 min
heating. 68% yield (202 mg). mp 131−134 °C (lit.[50]131.5−132.0 °C). 1H
NMR (300 MHz, DMSO-d6) δ 10.79 (d, J = 9.4 Hz, 1H), 8.84 (d, J = 9.9 Hz,
1H), 7.84 (t, J = 7.9 Hz, 1H), 7.41 (d, J = 7.8 Hz, 1H), 7.22 (d, J = 7.6 Hz,
1H), 4.27 – 4.13 (m, 4H), 1.26 (q, J = 7.2 Hz, 6H). The spectral data are in
agreement with the previously published values.[50]
Diethyl 2-(6-bromopyridin-2-yl)aminomethylenemalonate (3f). 30 min
heating. 46% yield (157 mg). mp 127−132 °C (lit.[9] 114−115 °C). 1H NMR
(300 MHz, DMSO-d6) δ 10.78 (d, J = 12.9 Hz, 1H), 8.82 (d, J = 13.0 Hz,
1H), 7.73 (t, J = 7.9 Hz, 1H), 7.44 (d, J = 7.7 Hz, 1H), 7.35 (d, J = 7.3 Hz,
1H), 4.26 – 4.13 (m, 4H), 1.26 (q, J = 7.1 Hz, 6H). 13C NMR (75 MHz,
DMSO-d6) δ 166.5, 164.8, 151.2, 146.9, 141.9, 139.4, 123.1, 111.9, 96.8,
60.1, 59.9, 14.2, 14.1.
Synthesis of Ethyl 7-methoxy-4-oxo-1,4-dihydro-1,8-naphthyridine-3-
carboxylate (5h). 3h (100 mg, 0.32 mmol) was added to 3 mL of refluxing
diphenyl ether. The reaction mixture was heated for 10 min. Upon rapid
cooling to room temperature and addition of 5 mL of cyclohexane a brown
precipitate formed. After filtration and washing with cold cyclohexane, the
precipitate was dried in a vacuum drying oven over night at 40 °C.
Compound 5h was obtained as a beige powder in 68% yield (57 mg). mp
235-240 °C. 1H NMR (300 MHz, DMSO-d6) δ 12.57 (brs, 1H), 8.38 – 8.30
(m, 2H), 6.88 (d, J = 8.7 Hz, 1H), 4.20 (q, J = 7.1 Hz, 2H), 3.98 (s, 3H),
1.27 (t, J = 7.1 Hz, 3H). 13C NMR (75 MHz, DMSO-d6) δ 173.1, 165.4,
164.2, 149.1, 144.5, 138.0, 116.7, 111.7, 109.7, 59.8, 54.2, 14.3. HRMS
(ESI-TOF): m/z [M + H]+ Calcd for [C12H12N2O4 +H]+ 249.086983; Found
249.087112.
Diethyl 2-(6-iodopyridin-2-yl)aminomethylenemalonate (3g). 30 min
heating. 55% yield (215 mg). mp 117−119 °C. 1H NMR (300 MHz, DMSO-
d6) δ 10.72 (d, J = 13.0 Hz, 1H), 8.84 (d, J = 13.0 Hz, 1H), 7.57 (dd, J =
7.3, 1.0 Hz, 1H), 7.52 – 7.39 (m, 2H), 4.26 – 4.12 (m, 4H), 1.29 – 1.22 (m,
6H). 13C NMR (75 MHz, DMSO-d6) δ 166.0, 165.3, 151.7, 147.5, 141.1,
130.6, 116.9, 112.5, 96.9, 60.5, 60.3, 14.7, 14.6. HRMS (ESI-TOF): m/z
[M + H]+ Calcd for [C13H15IN2O4 +H]+ 391.014926; Found 391.014502.
Diethyl 2-(6-methoxypyridin-2-yl)aminomethylenemalonate (3h). 10
min heating. 78% yield (226 mg). mp 58−61 °C. 1H NMR (300 MHz,
DMSO-d6) δ 10.75 (d, J = 13.2 Hz, 1H), 9.06 (d, J = 13.2 Hz, 1H), 7.70 (t,
J = 7.9 Hz, 1H), 6.95 (d, J = 7.6 Hz, 1H), 6.56 (d, J = 8.0 Hz, 1H), 4.22 (q,
J = 7.1 Hz, 2H), 4.13 (q, J = 7.1 Hz, 2H), 3.88 (s, 3H), 1.29 – 1.21 (m, 6H).
13C NMR (75 MHz, DMSO-d6) δ 166.8, 164.7, 162.9, 148.8, 148.1, 141.7,
105.5, 104.3, 94.8, 59.9, 59.6, 53.0, 14.2, 14.1. HRMS (ESI-TOF): m/z [M
+ H]+ Calcd for [C14H18N2O5 +H]+ 295.128848; Found 295.128627.
Synthesis
of
Ethyl
4-oxo-7-(trifluoromethyl)-1,4-dihydro-1,8-
naphthyridine-3-carboxylate (5i). 3i (50 mg, 0.16 mmol) was added to
1.5 mL of refluxing diphenyl ether. The reaction mixture was heated for 90
min. Upon rapid cooling to room temperature and addition of 5 mL of
cyclohexane a brown precipitate formed. After filtration and washing with
cold cyclohexane, the precipitate was dried in a vacuum drying oven over
night at 40 °C. Compound 5i was obtained as a beige powder in 74% yield
(24 mg) and 91% purity (1H NMR). mp 235-240 °C (lit.[53] >250 °C). 1H NMR
(300 MHz, DMSO-d6) δ 13.21 (brs, 1H), 8.77 (d, J = 7.7 Hz, 1H), 8.58 (d,
J = 5.4 Hz, 1H), 7.93 (d, J = 8.2 Hz, 1H), 4.23 (q, J = 7.1 Hz, 2H), 1.28 (t,
J = 7.1 Hz, 3H). 13C NMR (75 MHz, DMSO-d6) δ 173.1, 163.8, 149.6, 148.8,
148.4, 147.0, 139.0, 120.5, 119.2, 111.8, 60.0, 14.2.
Diethyl 2-(6-trifluoromethylpyridin-2-yl)aminomethylenemalonate (3i).
40 min heating. 51% yield (168 mg). mp 129−136 °C. 1H NMR (300 MHz,
DMSO-d6) δ 10.88 (d, J = 13.0 Hz, 1H), 8.94 (d, J = 13.0 Hz, 1H), 8.06 (t,
J = 7.8 Hz, 1H), 7.72 (d, J = 8.3 Hz, 1H), 7.60 (d, J = 7.4 Hz, 1H), 4.24 (q,
J = 7.1 Hz, 2H), 4.16 (q, J = 7.1 Hz, 2H), 1.30 – 1.22 (m, 6H). The spectral
data are in agreement with the previously published values.[51]
General FVP Procedure for the Preparation of Pyridopyrimidinones 6.
FVP reactions were carried out in unpacked quartz tubes (60 cm × 3.5 cm
ID) at a temperature of 450 °C and a vacuum of 3-5×10-3 mbar (see Figure
S1 for further details). 100 mg of sample (3a–j) was used for each
experiment. The product was collected in an acetone/liquid N2 trap and
was then rinsed with DCM. The products were isolated after evaporation
of the solvent under reduced pressure.
Diethyl 2-(pyridin-2-yl)aminomethylenemalonate (3j). 10 min heating.
46% yield (122 mg). mp 67−70 °C (lit.[50] 68−69 °C). 1H NMR (300 MHz,
DMSO-d6) δ 10.78 (d, J = 13.2 Hz, 1H), 9.06 (d, J = 13.2 Hz, 1H), 8.40 –
8.32 (m, 1H), 7.81 (ddd, J = 8.2, 7.4, 1.9 Hz, 1H), 7.39 (d, J = 8.2 Hz, 1H),
7.15 (ddd, J = 7.3, 4.9, 0.9 Hz, 1H), 4.26 – 4.11 (m, 4H), 1.29 – 1.22 (m,
6H). The spectral data are in agreement with the previously published
values.[52]
9
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202001110
European Journal of Organic Chemistry
FULL PAPER
Ethyl 6-methyl-4-oxo-4H-pyrido[1,2-a]pyrimidine-3-carboxylate (6a).
Preheater: 125 °C; 93% yield (78 mg); off-white powder. mp 96−105 °C
(lit.[51] 99−100 °C). 1H NMR (300 MHz, CDCl3) δ 8.84 (s, 1H), 7.67 – 7.62
(m, 1H), 7.52 – 7.49 (m, 1H), 6.88 – 6.86 (m, 1H), 4.38 (q, J = 7.1 Hz, 2H),
3.09 (s, 3H), 1.38 (t, J = 7.1 Hz, 3H). 13C NMR (75 MHz, CDCl3) δ 164.6,
158.7, 158.1, 156.0, 146.4, 138.2, 125.6, 120.6, 107.2, 61.0, 25.1, 14.6.
The spectral data are in agreement with the previously published values.[54]
yield 0.7 g (65%) of 10a. 100 mg of 10a was sublimed. The reaction was
carried out in an unpacked quartz tube (60 cm × 3.5 cm ID) at a
temperature of 450 °C and a vacuum of 3-5×10-3 mbar. The preheater was
set to 170 °C. The product was collected in an acetone/liquid N2 trap over
a period of 2 h and was then rinsed with DCM. The product was isolated
after evaporation of the solvent under reduced pressure in 95% yield (59
mg). mp 116−120 °C (lit.[55]111−114 °C). 1H NMR (300 MHz, DMSO-d6) δ
8.07 (d, J = 6.2 Hz, 1H), 7.68 – 7.62 (m, 1H), 7.41 – 7.38 (m, 1H), 6.92 (d,
J = 6.9 Hz, 1H), 6.20 (d, J = 6.2 Hz, 1H), 2.93 (s, 3H). 13C NMR (75 MHz,
DMSO-d6) δ 160.9, 153.8, 153.2, 143.3, 136.1, 124.9, 118.7, 106.3, 24.1.
The spectral data are in agreement with the previously published values.[55]
Ethyl 7-methyl-4-oxo-4H-pyrido[1,2-a]pyrimidine-3-carboxylate (6b).
Preheater: 122 °C; 83% yield (70 mg); off-white powder. mp 136−139 °C
(lit.[9] 135−136 °C). 1H NMR (300 MHz, DMSO-d6) δ 8.98 (s, 1H), 8.81 (s,
1H), 8.08 (dd, J = 8.9, 2.1 Hz, 1H), 7.77 (d, J = 8.9 Hz, 1H), 4.26 (q, J =
7.1 Hz, 2H), 2.48 – 2.44 (m, 3H), 1.30 (t, J = 7.1 Hz, 3H). 13C NMR (75
MHz, DMSO-d6) δ 164.6, 158.1, 154.2, 152.2, 143.5, 128.9, 126.6, 126.3,
104.0, 60.5, 18.3, 14.7. The spectral data are in agreement with the
previously published values.[10]
Synthesis of 7-Methyl-1,8-naphthyridin-4(1H)-one (8a). Into a 10 mL
microwave process vial, equipped with a stir bar, were placed 7a (50 mg,
0.31 mmol) and 1.5 mL of diphenyl ether. Diphenyl ether was prior heated
to 240 °C in order to remove traces of H2O. The vessel was sealed with a
snap cap and heated for 40 min at a set temperature (IR) of 300 °C in a
microwave reactor. Upon rapid cooling to room temperature and addition
of 5 mL of n-hexane a brown precipitate formed, which was filtered and
washed with 50 mL of n-hexane. After drying in a vacuum oven at 40 °C
overnight, 27 mg (54%) of 8a were obtained in 95% purity (5% residual
diphenylether). mp 235−240 °C (lit.[56] 235−237 °C). 1H NMR (300 MHz,
DMSO-d6) δ 12.04 (brs, 1H), 8.31 (d, J = 8.1 Hz, 1H), 7.89 – 7.84 (m, 1H),
7.25 (d, J = 8.1 Hz, 1H), 6.05 (d, J = 7.4 Hz, 1H), 2.56 (s, 3H). 13C NMR
(75 MHz, DMSO-d6) δ 177.4, 162. 5, 150.2, 140.0, 134.8, 119.9, 118.2,
109.7, 24.4. The spectral data are in agreement with the previously
published values.[14]
Ethyl 6,8-dimethyl-4-oxo-4H-pyrido[1,2-a]pyrimidine-3-carboxylate
(6c). Preheater: 110 °C; 96% yield (81 mg); off-white powder. mp
145−148 °C (lit.[35] 146−147 °C). 1H NMR (300 MHz, DMSO-d6) δ 8.61 (s,
1H), 7.36 (s, 1H), 7.05 (s, 1H), 4.22 (q, J = 7.1 Hz, 2H), 2.91 (s, 3H), 2.38
(s, 3H), 1.27 (t, J = 7.1 Hz, 3H). 13C NMR (75 MHz, DMSO-d6) δ 164.6,
158.1, 158.0, 155.8, 152.0, 145.1, 123.9, 123.8, 105.4, 60.3, 24.4, 20.9,
14.8.
Ethyl 6-iodo-4-oxo-4H-pyrido[1,2-a]pyrimidine-3-carboxylate (6g).
Preheater: 130 °C; 74% yield (68 mg); yellow powder. mp 136−145 °C. 1H
NMR (300 MHz, DMSO-d6) δ 8.78 (s, 1H), 8.00 (dd, J = 6.9, 1.7 Hz, 1H),
7.68 – 7.57 (m, 2H), 4.26 (q, J = 7.1 Hz, 2H), 1.29 (t, J = 7.1 Hz, 3H). 13
C
NMR (75 MHz, DMSO-d6) δ 163.8, 156.5, 155.9, 155.4, 139.2, 135.2,
126.9, 104.5, 91.8, 60.3, 14.3. HRMS (ESI-TOF): m/z [M + H]+ Calcd for
[C11H9IN2O3 +H]+ 344.973062; Found 344.972835.
Acknowledgements
The CCFLOW Project (Austrian Research Promotion Agency
FFG No. 862766) is funded through the Austrian COMET
Program by the Austrian Federal Ministry of Transport, Innovation
and Technology (BMVIT), the Austrian Federal Ministry of
Science, Research and Economy (BMWFW), and by the State of
Styria (Styrian Funding Agency SFG). We thank Gerfried Luschin-
Ebengreuth (RCPE) for performing DSC measurements and
Philipp Neu (University of Graz) for the HRMS measurements.
Ethyl 4-oxo-4H-pyrido[1,2-a]pyrimidine-3-carboxylate (6j). Preheater:
70 °C; 77% yield (54 mg); off-white powder. mp 109−112 °C (lit.[52]
111−113 °C). 1H NMR (300 MHz, DMSO-d6) δ 9.15 (ddd, J = 7.1, 1.6, 0.8
Hz, 1H), 8.84 (s, 1H), 8.20 (ddd, J = 8.6, 6.9, 1.6 Hz, 1H), 7.83 (ddd, J =
8.7, 1.4, 0.8 Hz, 1H), 7.56 (td, J = 6.9, 1.4 Hz, 1H), 4.27 (q, J = 7.1 Hz, 2H),
1.30 (t, J = 7.1 Hz, 3H). 13C NMR (75 MHz, DMSO-d6) δ 164.1, 158.2,
153.9, 153.1, 140.7, 128.7, 126.3, 118.3, 103.9, 60.1, 14.3.The spectral
data are in agreement with the previously published values.[15]
Keywords: Gould−Jacobs reaction • flash vacuum pyrolysis •
pyridopyrimidinones • naphthyridinones • regioselectivity
Continuous Flow Synthesis of Ethyl (E)-3-(6-methylpyridin-2-
yl)aminoacrylate (9a). The flow rate of the flow reactor (see Figure S5 for
the set-up) was set to 8 mL/min, the back pressure regulator to 100 bar,
and the temperature to 350 °C. An 8 mL stainless steel coil was used. The
reactor was equilibrated with pure MeCN for ca. 40 min, then the three-
way valve was switched to a 0.1 M solution of 3a in MeCN and the solution
was injected into the Phoenix reactor. After 1 min, the feed was switched
back to pure MeCN. The product solution was collected for 4 min. The
organic solvent was evaporated and the crude product was purified via
flash column chromatography on silica gel using PE and EA with 1%
triethyl amine as eluent to afford product 9a as a colorless oil in 33% yield
(54 mg). 1H NMR (300 MHz, CDCl3) δ 10.10 (d, J = 11.4 Hz, 1H), 8.01 (dd,
J = 12.1, 8.6 Hz, 1H), 7.43 (t, J = 7.8 Hz, 1H), 6.71 (d, J = 7.4 Hz, 1H), 6.48
(d, J = 8.1 Hz, 1H), 4.92 (d, J = 8.6 Hz, 1H), 4.18 (q, J = 7.1 Hz, 2H), 2.43
(s, 3H), 1.30 (t, J = 7.1 Hz, 3H). 13C NMR (75 MHz, CDCl3) δ 170.5, 157.6,
151.5, 141.5, 138.5, 117.0, 107.4, 89.1, 59.5, 24.5, 14.6. HRMS (ESI-
TOF): m/z [M + H]+ Calcd for [C11H14N2O2 +H]+ 207.112804; Found
207.112906.
[1]
[2]
[3]
[4]
R. G. Gould, W. A. Jacobs, J. Am. Chem. Soc. 1939, 61, 2890–
2895.
C. Sissi, M. Palumbo, Curr. Med. Chem. - Anti-Cancer Agents 2003,
3, 439–450.
J. J. Li in Name Reactions: A Collection of Detailed Mechanisms
and Synthetic Applications, Springer, Berlin, 2009, pp 263–265.
M. Bella, M. Schultz, V. Milata, K. Koňriková, M. Breza, Tetrahedron
2010, 66, 8169–8174.
[5]
[6]
N. D. Desai, J. Heterocycl. Chem. 2006, 43, 1343–1348.
G. Y. Lesher, Preparation of 4H-Pyrido[1,2-a]Pyrimidin-4-Ones from
Cyclic Alkylidene 2-Pyridylaminomethylenemalonates, 1975,
US3907798.
[7]
[8]
R. Morgentin, G. Pasquet, P. Boutron, F. Jung, M. Lamorlette, M.
Maudet, P. Plé, Tetrahedron 2008, 64, 2772–2782.
D. Lecoq, B. A. Chalmers, R. N. Veedu, D. Kvaskoff, P. V.
Bernhardt, C. Wentrup, Aust. J. Chem. 2009, 62, 1631–1638.
G. R. Lappin, J. Am. Chem. Soc. 1948, 70, 3348–3350.
R. B. Smith, H. Faki, R. Leslie, Synth. Commun. 2011, 41, 1492–
1499.
FVP Synthesis of 6-Methyl-4H-pyrido[1,2-a]pyrimidin-4-one (7a).
Meldrum’s acid (0.72 g, 5 mmol) and trimethyl orthoformate (1.09 mL, 10
mmol) were heated under reflux for 3.5 h. After cooling to room
temperature, the resulting orange solid was dissolved in 8 mL of EtOH. 2-
Amino-6-methyl pyridine (0.45g, 4.2 mmol) was added, and the reaction
mixture was stirred at ambient temperature overnight. The orange
precipitate was filtered, washed with ice cold EtOH and dried at 40 °C to
[9]
[10]
10
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202001110
European Journal of Organic Chemistry
FULL PAPER
[11]
[12]
Z. Mészáros, I. Hermecz, Tetrahedron Lett. 1975, 12, 1019–1020.
L. Debreczy-Vasvári, I. Hermecz, Z. Mészáros, P. Dvortsák, G.
Tóth, J. Chem. Soc. Perkin Trans. 1 1980, 227–232.
a) B. D. Schober, T. Kappe, J. Heterocycl. Chem. 1988, 25, 1231–
1236; b) T. Kappe, W. Lube, Chem. Ber. 1979, 112, 3424–3431.
J. Tsoung, A. R. Bogdan, S. Kantor, Y. Wang, M. Charaschanya, S.
W. Djuric, J. Org. Chem. 2017, 82, 1073–1084.
decarboxylated equivalent of pyridopyrimidinone 6a can be
synthesized that way.
[35]
I. Hermecz, Z. Mészáros, L. Vasvári-Debreczy, À. Horváth, G.
Horváth, M. Pongor-Csákvári, J. Chem. Soc., Perkin Trans. 1 1977,
789–795.
[13]
[14]
[15]
[16]
[17]
[18]
[36]
[37]
[38]
I. Hermecz, L. Vasvári-Debreczy, K. Simon, J. Chem. Soc. Perkin
Trans. 2 1988, 1287–1289.
L. Lengyel, T. Z. Nagy, G. Sipos, R. Jones, G. Dormán, L. Ürge, F.
Darvas, Tetrahedron Lett. 2012, 53, 738–743.
P. L. Ferrarini, C. Mori, O. Livi, G. Biagi, A. M. Marini, J. Heterocycl.
Chem. 1983, 20, 1053–1057.
D. Cantillo, H. Sheibani, C. O. Kappe, J. Org. Chem. 2012, 77,
2463–2473.
For all calculations, the qualitative energetic picture obtained does
neither depend on the functional or basis set employed nor on a
change in reaction temperature or the application of the COSMO[57]
solvation model.
T. Cablewski, P. A. Gurr, P. J. Pajalic, C. R. Strauss, Green Chem.
2000, 25–27.
R. F. C. Brown in Pyrolytic Methods in Organic Chemistry:
Application of Flow and Flash Vacuum Pyrolytic Techniques,
Academic Press, New York, 1980.
[39]
BMK is known to perform reliably for transition state barriers without
sacrificing accuracy of equilibrium properties: A. D. Boese, J. M. L.
Martin, J. Chem. Phys. 2004, 121, 3405–3416.
S. Grimme, J. Antony, S. Ehrlich, H. Krieg, J. Chem. Phys. 2010,
132, 154104.
[19]
[20]
[21]
H. McNab, Aldrichimica Acta 2004, 37, 19–26.
[40]
[41]
[42]
[43]
C. Wentrup, Angew. Chemie - Int. Ed. 2017, 56, 14808–14835.
a) M. Cox, F. Heidarizadeh, R. H. Prager, Aust. J. Chem. 2000, 53,
665–671; c) G. Kollenz, Product Class 10: Imidoylketenes in
Science of Synthesis, Georg Thieme Verlag, Stuttgart, 2006, Vol.
23, pp 351–380; d) C. Wentrup, J. J. Finnerty, R. Koch, Curr. Org.
Chem. 2010, 14, 1586–1599.
F. Weigend, R. Ahlrichs, Phys. Chem. Chem. Phys. 2005, 7, 3297–
3305.
W. Humphrey, A. Dalke, K. Schulten, J. Mol. Graph. 1996, 14, 33–
38.
a) A. D. Becke, Phys. Rev. A 1988, 38, 3098–3100; b) C. Lee, W.
Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785–789; c) A. D. Becke,
J. Chem. Phys. 1993, 98, 5648–5652.
[22]
[23]
R. H. Prager, Y. Singh, Aust. J. Chem. 1994, 47, 1263–1270.
L. C. Lengyel, G. Sipos, T. Sipöcz, T. Vágó, G. Dormán, J.
Gerencsér, G. Makara, F. Darvas, Org. Process Res. Dev. 2015,
19, 399–409.
[44]
[45]
After 30 min open-vessel reflux heating the selectivity 5a:6a was
81:15 in Ph2 (bp: 255 °C) and 87:8 in Ph2O (bp 259 °C).
TURBOMOLE V7.4 2019, a development of the University of
Karlsruhe and Forschungszentrum Karlsruhe GmbH, 1989-2007,
TURBOMOLE GmbH, since 2007; available from
http://www.turbomole.com.
[24]
[25]
[26]
I. Malvacio, E. L. Moyano, D. M. A. Vera, RSC Adv. 2016, 6, 83973–
83981.
I. Hermecz, À. Horváth, L. Vasvári-Debreczy, Z. Mészáros,
Synthesis 1984, 152–158.
For a literature precedent of this effect involving the high-
temperature cyclization of malonates with anilines, see: T. Razzaq,
C. O. Kappe, Tetrahedron Lett. 2007, 48, 2513–2517.
R. E. Martin, F. Morawitz, C. Kuratli, A. M. Alker, A. I. Alanine,
European J. Org. Chem. 2012, 47–52.
[46]
[47]
F. Neese, Wiley Interdiscip. Rev. Comput. Mol. Sci. 2012, 2, 73–78.
J. P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 1996, 77,
3865–3868.
[27]
[28]
[48]
[49]
F. Weigend, Phys. Chem. Chem. Phys. 2006, 8, 1057–1065.
G. Y. Lesher, M. D. Gruett, 3-X-Oxo-1,8-Naphtyridines et Leur
Préparation, 1962, BE612258.
While performing the reaction screening under microwave heating,
we were not aware that decarboxylated malonate 9a was
generated, since only small amounts (1-8%) of an unidentified side
product were detected. Only after identification from the continuous
flow experiments, we assigned this side product to 9a.
C. C. Price, N. J. Leonard, R. H. Reitsema, J. Am. Chem. Soc.
1946, 68, 1256–1259.
[50]
[51]
T. Hirose, S. Mishio, J. Matsumoto, S. Minami, Chem. Pharm. Bull.
1982, 30, 2399–2409.
W. R. Carling, A. Mitchinson, M. G. Russell, L. J. Street, (1,8)
Naphthyridines as Gaba Ligands, Their Pharmaceutical
Compositions and Uses, 2003, WO03/006464A1.
U. R. Mane, H. Li, J. Huang, R. C. Gupta, S. S. Nadkarni, R.
Giridhar, P. P. Naik, M. R. Yadav, Bioorganic Med. Chem. 2012, 20,
6296–6304.
[29]
[30]
[31]
[32]
[33]
[52]
[53]
L. E. da Silva, A. C. Joussef, R. J. Nunes and K. N. de Oliveira,
Synth. Commun. 2008, 38, 15–20.
A. Molnár, F. Faigl, B. Podányi, Z. Finta, L. Balázs, I. Hermecz,
Heterocycles 2009, 78, 2477–2488.
B. J. Armitage, J. G. Bowen, M. J. Crossley, B. W. Leslie, T. K.
Miller, M. Spowage, Amino Naphthyridine Compounds as Anti-
Rhoumatic Agents, 1995, US5464781A.
A. Molnár, Z. Mucsi, G. Vlád, K. Simon, T. Holczbauer, B. Podányi,
F. Faigl, I. Hermecz, J. Org. Chem. 2011, 76, 696–699.
It has been reported by Hermecz et al. that for the cyclization of 6-
halo substituted pyridylaminomethylene Meldrum´s acid derivatives,
electrostatic interaction was the main factor for the 6-F derivative,
while for the 6-Cl and 6-Br derivatives both steric and electrostatic
interactions should be considered (see Ref 32).
[54]
[55]
[56]
[57]
Z. Mészáros, J. Knoll, P. Szentmiklósi, Á. Dávid, G. Horvath, I.
Hermecz, Arzneim.-Forsch. 1972, 22, 815–824.
I. Ravina, D. Zicane, M. Petrova, E. Gudriniece, U. Kalejs, Chem.
Heterycyclic Compd. 2002, 38, 945–946.
G. Y. Lesher, M. D. Gruett, 1,4-Dihydro-4-Oxo-1,8-Naphthyridine-3-
Carboxaldehydes, 1974, US3857851.
[34]
Pyridopyrimidinone 7a was obtained from the corresponding
Meldrum´s acid derivative 10a (see Scheme S1). Because of the
loss of acetone and CO2 from the Meldrum´s acid moiety, the
A. Klamt, G. Schüürmann, J. Chem. Soc. Perkin Trans. 2 1993,
799–805.
11
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202001110
European Journal of Organic Chemistry
FULL PAPER
Entry for the Table of Contents
The regioselectivity of the Gould–Jacobs reaction of (pyridyl)aminomethylenemalonates can be controlled either in favor of the kinetic
(pyridopyrimidinone) or the thermodynamic (naphthyridinone) product, depending on the position of the substituent in the pyridine
moiety and the applied thermolysis technique.
12
This article is protected by copyright. All rights reserved.