Hydroamination reactions of alkynes with ortho-substituted anilines in ball mills: Synthesis of benzannulated N-heterocycles by a cascade reaction

Article abstract of DOI:10.1002/chem.201406306

It was demonstrated that ortho-substituted anilines are prone to undergo hydroamination reactions with diethyl acetylenedicarboxylate in a planetary ball mill. A sequential coupling of the intermolecular hydroamination reaction with intramolecular ring cl

Full text of DOI:10.1002/chem.201406306

DOI: 10.1002/chem.201406306
Full Paper
&
Solid-State Reactions
Hydroamination Reactions of Alkynes with ortho-Substituted
Anilines in Ball Mills: Synthesis of Benzannulated N-Heterocycles
by a Cascade Reaction
Maik Weiße, Markus Zille, Katharina Jacob, Robert Schmidt, and Achim Stolle*[a]
Abstract: It was demonstrated that ortho-substituted ani-
lines are prone to undergo hydroamination reactions with
diethyl acetylenedicarboxylate in a planetary ball mill. A se-
quential coupling of the intermolecular hydroamination re-
action with intramolecular ring closure was utilized for the
syntheses of benzooxazines, quinoxalines, and benzothia-
zines from readily available building blocks, that is, electro-
philic alkynes and anilines with OH, NH, or SH groups in the
ortho position. For the heterocycle formation, it was shown
that several stress conditions were able to initiate the reac-
tion in the solid state. Processing in a ball mill seemed to be
advantageous over comminution with mortar and pestle
with respect to process control. In the latter case, significant
postreaction modification occurred during solid-state analy-
sis. Cryogenic milling proved to have an adverse effect on
the molecular transformation of the reagents.
Introduction
The reactions of primary or secondary amines with CꢀC
double or triple bonds are known as hydroaminations and are
one possibility for the construction of CꢀN bonds from readily
available building blocks like alkenes or alkynes. Whereas the
reaction of amines with asymmetric alkenes yields two isomer-
ic amines from Markovnikov and anti-Markovnikov addition to
the double bond, the situation is rather more complex in the
case of alkynes as the reaction partner.[1] Not only does the re-
giochemistry of the amine addition to the triple bond have to
be considered in the case of terminal or other asymmetrical al-
kynes, but the utilization of metal catalysts can also initiate the
formation of imines, instead of enamines, by the reaction of
primary amines (Scheme 1).[1,2] Depending on the nature of the
alkynes, (E/Z)-enamines can originate from the hydroamination
of them. Several catalysts have been disclosed that initiate the
hydroamination of olefins and alkynes.[1–3]
Scheme 1. Possible products from addition of a primary amine to an alkyne.
described.[9] The highest activity of a catalyst for solvent-free
hydroamination was reported by Mizushima et al. with
0.1 mol% of [Au(PPh3)(CH3)] and 0.5 mol% H3TPA: A turnover
frequency of 3600 hꢀ1 was found for the reaction of p-bromoa-
niline and 4-methoxyethynylbenzene, and the reaction resulted
in 90% yield.[7] An intermolecular/intramolecular double hydro-
amination sequence via imine intermediates was suitable for
the solvent-free synthesis of a pyrrole derivative.[10] Utilization
of Au nanoparticles immobilized on an ordered mesoporous
organosilica network functionalized with thiol and sulfonate
groups for anchoring of the Au as the catalyst for the reaction
between hexa-1,5-diyne and aniline (2 equiv) yielded 87% of
2,5-dimethyl-1-phenyl-1H-pyrrole. A solvent-free tandem hy-
droamination–reduction sequence was proposed for the reac-
tion of ethynyl arenes with aniline derivatives to yield N-aryloc-
tan-2-amines (5 mol% AuCl3 +15 mol% AgOTf).[11]
With respect to advancement in the sustainability of hydro-
aminations, procedures have been introduced that avoid the
utilization of solvents in the reaction stage. For the synthesis
of ketimines, catalysts like Tl(OAc)3,[4] AuCl3,[5] Ru3(CO)12,[6]
([Au(PPh3)(CH3)]),[7] or anionic gold(I) complexes such as [AuS-
Phos][NTf2] (SPhos: chloro(2-dicyclohexylphosphino-2’,6’-dime-
thoxy-1,1’-biphenyl)[2-(2’-amino-1,1’-biphenyl)]palladium(II); Tf:
trifluoromethanesulfonyl) and [AuPPh3][NTf2],[8] as well as Cu-
and Ag-exchanged tungstophosphoric acid (TPA), have been
The noncatalyzed synthesis of enamines by addition of
amines to alkynes requires electron-deficient alkynes like prop-
argylic esters or diesters of acetylenedicarboxylic acid. The re-
action of anilines and secondary (cyclo)alkyl amines with those
substrates in the solid state by using a ball mill[12] and silica as
the milling auxiliary afforded the corresponding enamines in
80–96% yield after only 5 min of milling.[13] The reaction car-
ried out by mixing the reagents in a flask at 408C afforded the
product in only 66% yield, in comparison with 91% from the
ball-mill procedure. It has also been demonstrated that simple
[a] M. Weiße, M. Zille, K. Jacob, R. Schmidt, A. Stolle
Institute for Technical Chemistry and Environmental Chemistry
Friedrich Schiller University Jena
Lessingstrasse 12, 07743 Jena (Germany)
E-mail: Achim.Stolle@uni-jena.de
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201406306.
Chem. Eur. J. 2015, 21, 1 – 13
1
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
mixing of electron-deficient alkynes with a spatula is sufficient
for the preparation of enamines based on aniline(s), morpho-
line, n-PrNH2, or piperidine.[14] With respect to reproducibility,
the advantages of ball milling over mixing the reaction with
mortar and pestle have been described already.[15] Application
of K2HPO4 as the catalyst allowed the hydroamination of succi-
nimide and phthalimide with similar electron-deficient al-
kynes.[16] Reaction times were 4 and 45 min for the microwave-
assisted and classical heated procedures, respectively, whereas
the yields ranged from 50–70%, independently of the proce-
dure used.
lized Pd complexes,[25c] [(dippe)Ni(m-H)]2 (dippe: 1,2-bis(diiso-
propylphosphinoethane)),[25d] or AuCl3 have to be utilized.[5]
A
solvent-free ZnBr2-mediated version of such intramolecular hy-
droamination reactions that led to 2-substituted 1H-indoles in
a ball mill has been disclosed recently.[26] Tullberg et al. report-
ed a synthetic sequence for the construction of the 1H-indole
network by starting with a solvent-free Heck cross-coupling of
amino acid derivatives and aryl halides in a ball mill followed
by a cyclization via the corresponding hydrazine derivatives
(Fischer route) by refluxing in the presence of acetic acid anhy-
dride.[27] In a continuation of our previous work on intermolec-
ular hydroamination reactions in ball mills,[13] we would like to
present our efforts for the reaction of various ortho-substituted
anilines with electron-deficient alkynes in ball mills or under
similar conditions.
The mentioned procedures have in common that the hydro-
amination of ortho-substituted anilines failed or has not been
investigated at all. Inter- and intramolecular variants of those
hydroaminations are important for the solvent-free synthesis
of 1H-indoles. Indoles generally show a high biological activity
because they are selective for many receptors,[17] play a role in
the control of the day–night rhythm,[18] and are supposed to Results and Discussion
have anticarcinogenic properties.[19] Additionally, indole deriva-
Hydroamination of ortho-substituted anilines
tives possess huge potential for technical applications like sen-
sors, organic light-emitting diodes, and organic field-effect
transistors.[18,20] Beside traditional synthetic methods, such as
the Fischer[21] and Bischler indole syntheses,[22] the Larock
indole synthesis has attracted particular attention in research
throughout the past years.[23] In a retrosynthetic analysis, this
particular synthesis can be subdivided into a Sonogashira reac-
tion of o-haloanilines with terminal alkynes followed by an in-
tramolecular cyclization of the resulting o-alkynylanilines
(route 2 in Scheme 2). A second possibility for the synthesis of
In order to evaluate the ball-milling technology for the synthe-
sis of o-halo-N-vinylanilines (3) as the first step in the synthetic
sequence outlined as route 1 in Scheme 2, o-iodoaniline (1a)
and diethyl acetylenedicarboxylate (2b) were comminuted
under different conditions by using a planetary ball mill
(Scheme 3). Basic experiments were conducted in milling beak-
ers made from magnesia-stabilized zirconia (MSZ) with milling
balls of the same material and in the presence of NaCl as a mill-
ing auxiliary that could be removed after the synthesis by
aqueous workup. The results for the experiments with 1a are
summarized in Table 1 (entries 1–9). Milling at rpm=800 minꢀ1
in the absence of an additive and in an air atmosphere yielded
20% of 3a after 5 min without noticeable formation of side
products. Utilization of an Ar atmosphere, a reduced energy
density by decreasing the rotation frequency to 500 minꢀ1
(60 minꢀ1 =1 Hz), and an increase in the reaction time to 4 h
resulted in 90% yield. Replacement of NaCl by quartz sand in
an Ar atmosphere resulted in 88% of 3a within 10 min. Al-
though quartz sand is advantageous with regard to reactivity
and yield, it is less favorable with respect to workup because
solid–liquid extraction is required to recover adsorbed starting
materials and products. In a comparison to previous studies on
solvent-free intermolecular hydroamination,[13,14,16] it was possi-
ble to demonstrate the general applicability of ortho-substitut-
ed anilines for such reactions.
Scheme 2. Retrosyntheses for 1H-indoles with the incorporation of hydro-
amination steps. ewg: electron-withdrawing group.
indoles is shown in route 1 in Scheme 2 as a combination of
an intermolecular hydroamination followed by an intramolecu-
lar Heck cross-coupling. Both routes start with o-haloanilines,
but the substitution pattern of the alkyne as the second build-
ing block differs. The Larock type synthesis utilizes terminal al-
kynes, whereas electrophilic alkynes (bearing elec-
tron-withdrawing groups, for example, C(O)R with R:
alkyl, hydroxy, or alkoxy) have to be applied for the
uncatalyzed intermolecular hydroamination reactions.
Solvent-free variants of the second route incorpo-
In addition, several metal salts were explored for their ap-
plicability to accelerate this solvent-free hydroamination.[1,2]
rating protocols for the Sonogashira reaction[24] and
the intramolecular hydroamination are well known in
the literature.[5,25] Owing to the low activity of o-alky-
nylanilines with respect to the addition of the amine
Scheme 3. Hydroamination studies in a planetary ball mill with ortho-substituted anilines.
to the triple bond, catalysts like Pd–CuI,[25b] immobi- The R and R1 groups are given in Table 1. MSZ: magnesia-stabilized zirconia.
&
&
Chem. Eur. J. 2015, 21, 1 – 13
www.chemeurj.org
2
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
in 1c is illustrated by the reac-
tions with the corresponding m-
(1e) and p-chloroanilines (1 f),
which yielded >95% of the N-
vinyl derivatives, compared with
a maximum of only 46% of the
ortho isomer 3c. Hydroamination
of o-ethylaniline (1h) and 2-ami-
nobiphenyl (1i) yielded the
products in >95 and 77% yield,
respectively. It should be point-
ed out that, for the reaction of
anilines with a high nucleophilic-
ity like 1b, o-toluidine (R1: Me),
or 1h, spontaneous reactions
were observed when the re-
agents were unified in a sample
vial, mixed gently by shaking,
and analyzed immediately. Con-
trarily, the presence of electron-
withdrawing substituents in the
ortho position, like acyl (1j),
nitro (1l), or cyano (1m) groups,
reduces the nucleophilicity of
the amino group. Thus, the
yields were considerably lower
or no conversion was observed
(Table 1, entries 23–26). The reac-
tion of N-Tf-protected o-iodoani-
line (1n)[29] required a prolonged
reaction time and led to a mix-
ture of unprotected and triflated
products (3a and 3n, respective-
ly; Table 1, entry 27).
Table 1. Hydroamination studies in a planetary ball mill with o-substituted anilines (1, R: H) and diethyl acety-
lenedicarboxylate (2b; Scheme 3).[a]
Entry
Substrate Product R1
Additive
t [min] rpm [minꢀ1
]
Yield [%][b]
1
2[c]
3
4
5
1a
3a
I
none
none
5
240
30
30
15
30
10
30
15
5
800
500
600
600
800
600
800
600
800
800
20
90
0
0
66
0
88
58
72
89
99
46
38
15
27
>95
>95
>95
0.1 mmol ZnBr2
0.1 mmol ZnCl2
0.1 mmol TPA
0.1 mmol Cu–TPA
none
0.1 mmol ZnCl2
0.1 mmol TPA
none
none
none
none
none
6
7[d]
8[d]
9[d]
10
11[d]
12
13
14
15
16
17
18
1b
1c
1d
3b
3c
3d
H
Cl
Br
30
6
30
6
600
800
600
800
800
none
1e
1 f
1g
3e
3 f
4b
4a
4b
3h
3i
3j
3k
3l
m-Cl none
p-Cl none
OH
OH
OH
Et
10
none
none
none
none
none
none
19 (mixer ball mill)[e] 1g
20 (mixer ball mill)[f] 1g
5
5
10
900 (=^ f=15 Hz) >95
900 (=^ f=15 Hz) >95
800
21
22
23
24
25
26
27
1h
1i
1j
1k
1l
1m
1n
>95
77
21
0
Ph
Ac
OPh none
NO2 none
CN
I[g]
0
0
–
3m
3n
none
none
[h]
[a] Reaction conditions unless otherwise stated: 1 (1 mmol), 2b (1 mmol), NaCl (5 g); planetary ball mill, MSZ
beakers (V=45 mL) and MSZ milling balls (d=15 mm, n=6). [b] Analytical yield determined with GC-FID and
GC-MSD (uncorrected). [c] In an Ar atmosphere. [d] Quartz sand (5 g) was used instead of NaCl. [e] Reaction
conditions: 1g (2 mmol), 2a (2 mmol); beakers (stainless steel, V=50 mL) and milling balls (stainless steel, d=
5 mm, n=15). [f] Reaction conditions: 1g (2 mmol), 2b (2 mmol); beakers (stainless steel, V=50 mL) and mill-
ing balls (stainless steel, d=5 mm, n=15). [g] R: Tf; see Scheme 3. [h] Mixture of unprotected (3a) and triflated
product 3n.
Lewis acid catalysts like ZnX2 (X: Cl, Br) or TPA-derived catalysts
are well known for the reaction in solution,[28] and the suitabili-
Benzoxazine derivatives from o-aminophenol and dialkyl
ty of the last class of compounds has already been proven for
intermolecular hydroamination leading to imines in the ab-
sence of solvent.[9] For the ball-milling reaction, neither the ad-
dition of 10 mol% ZnBr2 (Table 1, entry 3), ZnCl2 (Table 1, en-
tries 4 and 8), or TPA (Table 1, entries 5 and 9) nor the applica-
tion of Cu-exchanged TPA (Table 1, entry 6) increased the yield
relative to that of the noncatalyzed reaction, regardless of
which type of milling auxiliary was applied.
acetylenedicarboxylates
Comminution of o-aminophenol (1g) with 2b in a planetary
ball mill together with NaCl (5 g) yielded benzoxazine (4b)
rather than the corresponding N-vinylated product 3g
(Scheme 3; Table 1). No side products were detected. Milling of
dimethyl acetylenedicarboxylate (2a) or 2b and 1g in a mixer
ball mill without further auxiliaries yielded the corresponding
alkyl esters 4a and 4b, respectively, as single products
(Table 1, entries 19 and 20). From the literature, it is known
that mixing of those reagents with a spatula on a Petri dish for
less than 5 min leads to the same result.[14] Beside this solvent-
free protocol, synthesis can be carried out in MeOH at room
temperature,[30] in refluxing EtOH,[31] or in ice-cold EtOH.[32]
Compound 4a is also available by cyclocondensation of 1g
with methyl 3-hydroxy-4-oxo-4-(phenylamino)but-2-enoate.[33]
The formation of 4a can be rationalized by a hydroamination
starting with the nucleophilic attack of the amino group in 1g
onto the electrophilic triple bond of 2a, which leads to di-
methyl 2-(2-hydroxyphenylamino)but-2-enedioate (3g) as an
intermediate (Scheme 4). In a subsequent transesterification,
Next, anilines with different substituents in the ortho posi-
tion were tested in the reaction with 2b (Table 1). The reaction
of aniline 1b with 2b resulted in 89 or 99%, respectively, of
the corresponding diethyl ester 3b by using NaCl or quartz
sand as the milling auxiliary (Table 1, entries 10 and 11). Appli-
cation of o-chloro- (1c; Table 1, entry 12) and o-bromoaniline
(1d; Table 1, entry 14) demonstrated the rapidly decreasing re-
activity of the halides with regard to the hydroamination reac-
tion with 2b: 1a (R1: I)!1d (R1: Br)<1c (R1: Cl). An increase of
the rpm from 600 to 800 minꢀ1, accompanied with a reduction
of the reaction time to 6 min to avoid thermal stress, did not
lead to considerable improvements (Table 1, entries 13 and 15).
The decreased reactivity as a result of the high steric demand
Chem. Eur. J. 2015, 21, 1 – 13
www.chemeurj.org
3
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
with release of one equivalent of MeOH, 4a is formed almost
quantitatively. A critical analysis of the mechanisms outlined in
Scheme 4 is provided in the Supporting Information, with con-
sideration of the formation of the alternative product 6a, on
the basis of IR and NMR spectroscopy analyses.
reaction (conversion>95%=^ >95% yield). As can be seen in
the Supporting Information (Figure S1), the product is indicat-
ed by a strong signal at n˜ =1762 cmꢀ1 for the C=O stretching
mode of the lactone, whereas the C=O stretching mode signal
at n˜ =1720 cmꢀ1 is typical for the presence of the two ester
groups in 2a. An increase of the first and decrease of
the latter with increasing reaction time is an unmis-
takable sign for the reaction progress (Figure 1A).
However, for semiquantitative analysis, the transmis-
sion from those signals alone might be misleading
because material properties like powder density, hu-
midity, and viscosity (release of MeOH during the re-
action) influence the quality of the spectra. In order
to monitor the reaction progress, the ratio of the
transmission minima (Tmin) from the lactone C=O
stretching mode and that of the absorption signal as-
signed to aromatic ring bending (n˜ =750 cmꢀ1) is cal-
culated to give the dimensionless number Av [Eq. (1)].
ꢀ
ꢀ
ꢀ
ꢀ
logðTmin;C¼Ostretching
Þ
Amax;C¼Ostretching
Amax;ringbending
ꢀ
ꢀ
ꢀ
ꢀ
À
Á
Av ¼
¼
ð1Þ
log Tmin;ringbending
Scheme 4. Reaction of o-aminophenol (1g) with dimethyl acetylenedicarboxylate (2a)
and tautomerization of 4a in protic solvents to form imine 7. HA: acid.
Repetitive measurements of the sample after tg =
20 min and calculation of AV leads to a value of
>0.46, which indicates no further change in the
Investigation of the reaction progress with attenuated total
reflectance FTIR spectroscopy
amount of 4a. A similar value was obtained for experiments in
a ball mill. In combination with the absence of an absorption
at n˜ =1720 cmꢀ1, full conversion is indicated.[35]
Our results indicate that the formation of benzoxazines 4a and
4b is possible in the absence of a solvent. Analysis of the
crude reaction mixture with attenuated total reflectance (ATR)
FTIR spectroscopy (5 min milling) indicates complete conver-
sion of the substrates to a single product. Dissolution of the re-
action mixture in CDCl3 or EtOAc and analysis by NMR spec-
troscopy and GC with flame ionization detection (GC-FID), re-
spectively, did not reveal further products. The FTIR spectra of
recrystallized products were almost superposable to those of
the crude reaction mixture. However, a subsequent reaction
during sample preparation by dissolution cannot be excluded,
although the literature reports mainly water or alcohols as the
preferred solvents.[34] The fact that simple mixing of the re-
agents on a Petri dish results in almost quantitative yields after
only a few minutes[14] indicates that a fast and direct analysis is
required for the product mixtures, ideally without pretreat-
ment (like extraction). The use of ATR FTIR spectroscopy is one
option because no sample workup is necessary and the time
gap between sampling and measurement can be reduced to
less than 40 s.
Experiments in a mixer ball mill (MBM) and those in the liter-
ature show that the reaction itself is apparently very fast.[14]
One might argue that the applied contact force induced by
pressurizing the powdered sample over the ZnSe crystal of the
ATR FTIR may induce a subsequent reaction. At least, the litera-
ture reports that similar experiments with a hydraulic press
(preheated anvil, P=19.6 MPa, Heck reaction) resulted in signif-
icantly lower yields than high-energy ball milling in a planetary
ball mill.[36] A pressure-induced reaction cannot be excluded,
which might interfere with the method for reaction monitor-
ing. Thus, the reaction time (tr) should be the considered as
the sum of the grinding or milling time (tg), the sample transfer
time (tst), and the measurement time (tm), that is, the time on
the ATR crystal [Eq. (2)].
tr ¼ tg þ tst þ tm
ð2Þ
With respect to the tg (variable) and tm (>1 min, see the Ex-
perimental Section) values, the tst value can be reduced to 40 s
from opening of the milling beakers to placement on the ATR
crystal and the start of the measurement.[37] Figure 2A shows
the reaction progress for the synthesis of 4a (MBM), which fol-
lows an exponential curve (each data point in the MBM curve
represents a single ball-milling experiment). In a second set of
experiments, samples withdrawn from the ball-milled reaction
mixture were pressurized on the ATR crystal and data record-
ing was continued. The results indicate that, after measure-
ment started, the reaction went on until approximately tm =
7.5 min before reaching equilibrium with regard to the Av
The FTIR spectra of the products from the reaction of 1g
and 2a or 2b contain several signals, which are suitable for
semiquantitative analysis of the reaction progress. Figure 1A
shows two different regions from the FTIR spectra recorded
after different grinding times, tg (1–20 min), during the commi-
nution of 1g and 2a at room temperature (238C) with a porce-
lain mortar and pestle. In the present case, only those spectra
recorded at the beginning of the reaction (tg =1 min) and after
tg =20 min should be considered. Those spectra exemplify the
beginning of the reaction (no conversion) and the end of the
&
&
Chem. Eur. J. 2015, 21, 1 – 13
www.chemeurj.org
4
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Figure 1. ATR FTIR spectra of the reaction products from comminution of 1g (2 mmol) and 2a (2 mmol) in a mortar with pestle (porcelain, room tempera-
ture). A) t=^ grinding time, tg. B) 1 min grinding, then the ground mixture (100 mg) was placed on the ZnSe crystal of the ATR FTIR spectrometer; t=^ duration
on the ZnSe crystal, tm (tm =0 min in B is equivalent to tg =1 min in A). C) 5 min grinding, then the ground mixture (100 mg) was placed on the ZnSe crystal
of the ATR FTIR spectrometer; t=^ duration on ZnSe crystal, tm (tm =0 min in C is approximately equivalent to tg =5.5 min in A). A colored version of this figure
is part of the Supporting Information (Figure S2).
values. The reaction progress of these subsequent reactions
did not reach full conversion, and the final Av values are 0.2 for
MBM-1 (1 min grinding) and MBM-2 (2 min grinding) and 0.35
for MBM-3 (4 min grinding) within tm =30 min. The increase of
Av values from the start of the measurement to equilibrium
(DAv =0.1) is independent of the tg value. This may be a result
of diffusion limitations in the powder fixed at the sample
holder. The reaction proceeds at the crystal boundaries of the
preground and well-mixed reagents but cannot take place in
the bulk powder. The reaction rates (r), as determined from
Equation (3), for reactions in the ball mill and the subsequent
reactions after 1, 2, and 4 min pregrinding are 0.0950, 0.0094,
0.0102, and 0.0191 minꢀ1, respectively (Table 2).
À
Á
À
Á
Av ti;2 ꢀ Av ti;1
ti;2 ꢀ ti;1
DAv;linear
Dti;linear
ð3Þ
r ¼
¼
Intense mixing and constant restoration of the active surfaces
in the ball mill are responsible for the high rate. The elaborat-
ed particle refinement after 4 min pregrinding (MBM-3) result-
Chem. Eur. J. 2015, 21, 1 – 13
www.chemeurj.org
5
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
ed in smaller particles, which leads to doubling of the reaction
rate relative to those of samples MBM-1 and -2.
Figure 2B displays the dependence of the Av values on the
measurement time for different powder treatment methods.
Mixing of compounds 1g and 2a with a spatula in a small
beaker (SP),[14] 1 min treatment in a mortar with a pestle (MP-
1), and cogrinding of the reagents together with KBr in a vibra-
tion ball mill (VBM)[38] resulted in similar Av,0 values to those ob-
served for MBM-1 (Table 2). This indicates a similar reaction
progress before placement on the ATR crystal (tm =0 min). Re-
crystallization (CHCl3) of the product from MBM-3 did not
affect the Av value (RC); thus, a subsequent reaction in solution
can be excluded for those samples reaching Av >0.46 (full con-
version). The behavior of the MBM-1, SP, and VBM reactions
during measurements is similar and results in equilibrium
values of Av =0.20–0.25 after tm =7.5 min (0.0094–0.017 minꢀ1).
Contrarily, samples generated by mixing in a mortar with
a pestle for 1 min (MP-1) revealed a slow but steady increase
to reach Av =0.390 after 15 min (0.0254 minꢀ1). Apparently,
energy entry due to friction, instead of mainly impact (MBM,
VBM), leads to a homogeneous activation of the reagents that
allows a reaction despite the diffusion limitations in the solid
state. Additionally, during ball milling of the samples with high
ball-to-powder-ratios, partial liquefaction took place. Spontane-
ous recrystallization after stopping the energy entry yields
a less homogeneous powder, which disfavors subsequent reac-
tion on the ATR crystal. Pregrinding of 1g and 2a with
a mortar and pestle with tg =1, 3, 5, and 7 min led to increas-
ing Av,0 values (Table 2). The MP reaction reaches Av values of
>0.390 after 15 min, whereas sample MP-4 (tg =7 min) re-
mained unchanged, which resulted in a low reaction rate (Fig-
ure 3A). The high Av value of >0.53 (Table 2) is due to severe
volatilization of the MeOH of the sample, whereas the alcohol
is part of the reaction mixture in all of the other cases. The
rate constants for the samples treated with mortar and pestle
are similar (0.0182–0.0254 minꢀ1), with the exceptions of MP-3
and MP-4. Consideration of the tg, tst, and tm values [Eq. 2] for
samples MP and MP-1 to MP-3 indicates a high overlap of the
reaction progress at similar tr values (Figure 3B).
Figure 2. Reaction progress for the reaction of 1g (2 mmol) and 2a
(2 mmol) under different reaction conditions measured with ATR FTIR spec-
troscopy: A) with respect to the overall reaction time, tr ([Eq. (2)]; see
Table 2); B) with respect to the measurement time, tm (see Table 2).
Table 2. Effect of the measurement time tm on the reaction progress
during measurement of crude samples from the reaction of 1g and 2a
under different reaction conditions measured with ATR FTIR spectrosco-
py.[a]
Conditions before placement
on ATR crystal
Av,0
ti,1/ti,2
[min][a]
DAv/Dti
[a]
[minꢀ1
]
grinding in mixer ball mill at f=15 Hz (see Figure 2A)
The conclusion can be drawn that mixing the reagents with
mortar and pestle yielded reaction mixtures that, upon pres-
surization, willingly react with the same rate constant to com-
pletion. In the case of other grinding and/or mixing methods,
subsequent reaction during ATR measurement resulted in equi-
librium values that were different from full conversion. Thus,
for the investigation of the reaction progress in the case of
fast solvent-free reactions, it has to be ensured that the
sample preparation time and measurement time are restricted
to a minimum. Also, solvent-based workup procedures and an-
alytical methods for the assessment of reaction progress and/
or purity should be avoided because of possible subsequent
reaction during solvation or in solution.
MBM
0.09
0.15
0.14
0.26
–
0.0950
0.0094
0.0102
0.0191
MBM-1
MBM-2
MBM-3
tg =1 min
tg =2 min
tg =4 min
0/6.5
0/6.5
0/4.0
grinding with mortar and pestle (see Figure 3)
MP
0.07
–
0.0229
0.0254
0.0182
0.0527
<0.0074
MP-1
MP-2
MP-3
MP-4
tg =1 min[b]
tg =3 min
tg =5 min[c]
tg =7 min
0.09
0.16
0.15
0.54
2.5/12.5
0/12.5
0/5.0
0/2.5
other conditions (see Figure 2B)
SP
RC
mixing with spatula for 1 min
as MBM-3 with recrystallization
from CHCl3
0.18
0.43
0/6.0
0/30
0.0132
<0.0018
VBM
grinding in a vibration ball mill
with KBr (tg =2 min)[d]
0.11
0/5.0
0.0170
Investigation of reaction conditions
[a] In accordance with Equation (3). [b] See Figure 1B. [c] See Figure 1C.
[d] 1g+2a/KBr=0.2 g/1 g.
The influence of the oscillation frequency was investigated
with the synthesis of 4a (Scheme 3) as a model reaction. For
&
&
Chem. Eur. J. 2015, 21, 1 – 13
www.chemeurj.org
6
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Figure 4. Reaction progress for reaction mixtures after cryomilling. Reaction
conditions: 1g (2 mmol), alkyne (2a or 2b; 2 mmol), mixer ball mill, milling
beakers (stainless steel, V=50 mL), milling balls (stainless steel, dMB =5 mm,
n
MB =14), 15 Hz, tm =5 min; precooled to ꢀ788C prior to milling.
higher reactivity of 2a in comparison to 2b. The temperature
increase after milling depends linearly on the defrosting time
(tdf). Reactions reach completion after 20 and 45 min with reac-
tion rates [Eq. (3)] of 0.036 and 0.0192 minꢀ1, respectively. A
comparison with the rate for milling of 2a and 1g from room
temperature (0.0950 minꢀ1; Table 2) indicates the retarding
effect of the powder temperature on the rate constant of the
reaction. In contrast to these conditions, the MBM-1 to MBM-3
reactions reach completion despite the diffusion limitation in
the solid state.
Figure 3. Reaction progress for the reaction of 1g (2 mmol) and 2a
(2 mmol) by using a mortar and pestle measured with ATR FTIR spectrosco-
py: A) with respect to the measurement time, tm (see Table 2); B) with re-
spect to the overall reaction time, tr ([Eq. (2)]; see Table 2).
the influence of the reaction time, see Figure 2A (MBM condi-
tions). An increase of the oscillation frequency from 15 to 20,
25, or 30 Hz indicated that further reduction of the milling
time to <5 min was not possible, because the ratio of side re-
actions increased considerably. Thus, a milling time of 5 min
and a frequency of 15 Hz were chosen for further experiments.
One option to avoid partial liquefaction of the reagents
during ball milling is reduction of the energy intensity by
either reducing the oscillation frequency or by working in a cy-
clized mode with milling pauses. Both options are inapplicable
because of inefficient mixing at f<15 Hz[15] and the already
short reaction times of 5 min. The third option is cryogenic
milling. For this purpose, stainless steel milling beakers are
filled with milling balls and the reagents and are cooled down
to ꢀ788C with solid CO2. After milling of the reaction mixture
for 5 min at 15 Hz, the reaction progress was measured and Av
values of 0.050 and 0.078 were immediately obtained for the
reaction of 2a and 2b, respectively, with 1g. Obviously, no re-
action occurred at the reaction temperatures, which had risen
to ꢀ17 and ꢀ108C. Presumably, the low reaction rate due to
“frozen” molecular movement is responsible for this observa-
tion. Similar to studies on cryogenic homocoupling of phenyl-
acetylene in a planetary ball mill,[39] the reaction proceeds as
soon as defrosting of the reaction mixture starts and powder
temperatures of 3 (2a) or 198C (2b) are reached (Figure 4).
The difference in reaction initiation summarized in Figure 4 for
the methyl (4a) and ethyl derivatives (4b) may be due to the
Structure flexibility for the reaction of o-aminophenols with
electrophilic alkynes
Table 3 summarizes the data for the reaction of different o-ami-
nophenol derivatives (1g, 8–12) with alkynes 2a and 2b. The
reported yields are yields of the isolated recovered product di-
rectly after the milling process without further workup (mass
yield). The purity of the products and the completeness of the
reactions were checked by immediate ATR FTIR spectroscopy
measurements. Relative to the results of the protocol de-
scribed by Choudhary and Peddinti (Petri dish),[14] the yields
are lower. This can be attributed to the fact that product isola-
tion from the milling beakers and milling balls is complex at
such a small reaction scale. Thus, the lower yields are rather
a sign of the nonoptimized isolation procedure than of a disad-
vantage of the method itself. The higher yields from reactions
with 2b compared with those resulting from 2a are due to
the higher crystallinity of the products. The reaction of 2-
amino-5-nitrophenol with both 2a and 2b failed under the
gentle milling conditions (f=15 Hz, t=5 min). An increase of
the oscillation frequency to f=30 Hz and elongation of the re-
action time to 20 min did not allow the yield to be increased.
Chem. Eur. J. 2015, 21, 1 – 13
www.chemeurj.org
7
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Other solvent-free reactions that start with 19 incorporate con-
densation steps for the formation of the heterocycle.[44] The re-
action with diethyl bromomalonate resulted in ethyl 3-
hydroxyquinoxaline-2-carboxylates (6–8 h under vacuum),[45]
whereas condensation with a-ketoacids yielded 3-substituted
Table 3. Results for the synthesis of benzoxazines from o-aminophenols
and alkynes with a mixer ball mill.[a]
Alkyne
o-Aminophenol
Product
Yield[a] [%]
Scheme 5. Synthesis of quinoxaline derivatives from o-phenylenediamine
(19; 2 mmol) and 2 (2 mmol) in a mixer ball mill (beakers: V=50 mL; balls:
d=5 mm, n=15).
2a
1g
8
9
10
11
1g
8
4a
70
82
89
82
60
94
86
94
85
98
88
79
13a
14a
15a
16a
4b
13b
14b
15b
16b
17b
18
quinoxalin-2-(1H)-ones in 70–91% yield after microwave irradi-
ation for 3–13 min.[46] Thus, the ball-milling method is not only
a fast and robust variant but also results in substitution pat-
terns that are not yet available by other solvent-free methods.
2b
9
10
11
12
1g
2c[b]
Synthesis of benzothiazine derivatives
[a] Reaction conditions: o-aminophenol (2 mmol), alkyne (2 mmol); mixer
ball mill, milling beakers (stainless steel, V=50 mL), milling balls (stainless
steel, dMB =5 mm, nMB =14); f=15 Hz, t=5 min. [b] Yield of isolated prod-
uct. [c] With addition of PPh3 (2 mmol), f=30 Hz, t=60 min.
Scheme 6 illustrates the formation of benzothiazine derivatives
22a and 22b by the reaction of o-aminothiophenol (21) with
esters 2a and 2b in a mixer ball mill. Reagents were ball
milled for 5 min at f=15 Hz in the absence of any other addi-
tive. Products were formed in almost quantitative yields. In
contrast to the reaction of 1g (Scheme 4) and other o-amino-
phenol derivatives (8–12; Table 3), the reaction sequence leads
to a product that is comparable with compound 6 (Scheme 4).
The higher nucleophilicity of the SH group over the NH2 group
in 21 favors hydrothiolation (thiol–yne click chemistry)[47] in-
stead of hydroamination. Furthermore, the resulting lactam
structure has a higher stability than that of the alternative thio-
ester. These findings are in accordance with many literature
protocols discussing the synthesis of benzothiazines in solu-
tion[34,41c,d,48] or in a solvent-free manner.[14,42] In some cases,
the formation of products with a similar structure to 4 was
postulated.[30d,42,49] However, with regard to the structure of 22,
there is a higher consistency than with the case outlined in
Scheme 4. As a result of the absence of stabilizing hydrogen
bonding between the sulfide and the carbonyl group, like the
amine and the ester carbonyl group in (Z)-methyl 2-(2-oxo-2H-
benzo[b][1,4]oxazin-3(4H)-ylidene)acetate (4a),[14,31,40a] the ben-
The reaction of 1g with 3-phenylpropiolate (2c) failed under
the aforementioned milling conditions. Due to the lower elec-
trophilicity of 2c relative to 2a and 2b, one equivalent of PPh3
has to be applied as mediating reagent. Under harsh condi-
tions (f=30 Hz, t=60 min), the corresponding product could
be isolated in 79% yield. The product adopts an imine confir-
mation comparable to that of 7 in Scheme 4.[40] This is the first
example for such a synthesis under solvent-free conditions
with a ball mill. Other procedures comprise reactions in tolu-
[40a]
ene heated to reflux with two equivalents of 2a and PPh3
and the reaction of equimolar amounts of the reagents in di-
chloromethane at ꢀ108C for 24 h.[40b]
Synthesis of quinoxaline derivatives
Comminution of o-phenylenediamine (19) with 2a or 2b
under standard reaction conditions in a mixer ball mill afforded
the corresponding quinoxalines in 91 and 93% yields after iso-
lation (Scheme 5). Similarly to the formation of benzoxazines
from o-aminophenols (1g, 8–12; Table 3), products 20a and
20b prefer to adopt a Z configuration due to the presence of
hydrogen bonding between the NH group and the ester car-
bonyl group. In comparison with other methods available for
the synthesis of 20 (in accordance with Scheme 5),[30a–c,31,34,41]
the present method and also other solvent-free methods are
fast and reduce the reaction time to less than 10 min.[14,42]
A
Scheme 6. Synthesis of benzothiazine derivatives from o-aminothiophenol
(21; 2 mmol) and 2 (2 mmol) in a mixer ball mill (beakers: V=50 mL; balls:
d=5 mm, n=15).
solvent-free reaction in the presence of Yb(OTf)3 resulted in
similar yields but double the reaction time was required.[43]
&
&
Chem. Eur. J. 2015, 21, 1 – 13
www.chemeurj.org
8
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
zothiazines are formed in two stereoisomeric forms: (E)- and
(Z)-22. The ratio of the two forms was determined by using
the difference in chemical shifts of the vinylic proton. As
shown in Table 4, the ratio of the stereoisomers is independent
from the alkyl group, with the formation of the E over the Z
form of 22 slightly favored.
can be initialized by defrosting the finely ground and quasi-
homogeneously mixed powders.
Experimental Section
General information
All chemicals were purchased from Sigma–Aldrich or Alfa Aesar
and were used as received. Milling experiments were accomplished
with a Fritsch Pulverisette 7 classic-line planetary ball mill (Table 1)
or a Retsch MM400 mixer ball mill. Experiments in a vibration ball
mill were carried out with a Fritsch Pulverisette 23 equipped with
one stainless steel ball (10 mm). GC-FID measurements were per-
formed on a 6890 GC instrument, and GC with mass selective de-
tection (GC-MSD) measurements were recorded on a 6890N instru-
ment with MS detector 5973. Conditions for GC-FID: HP 5 column,
30 mꢁ0.32 mmꢁ0.25 mm; H2: 10 psi; temperature program: 708C
(hold for 1 min), 10 Kminꢀ1 up to 2808C (hold for 7 min); injector
temperature: 2808C; detector temperature: 3008C. Conditions for
GC-MSD: HP 5 column, 30 mꢁ0.32 mmꢁ0.25 mm; He: 10 psi; tem-
perature program: 708C (hold for 3 min), 10 Kminꢀ1 up to 2808C
(hold for 7 min); injector temperature: 2808C; detector: electron
impact (70 eV). ESI HRMS spectrometry was performed with an ESI-
(Q)-TOFMS MICROTOF II (Bruker Daltonics GmbH) mass spectrome-
ter. ATR-FTIR spectroscopy measurements were recorded on
a Perkin–Elmer Spectrum 100 instrument equipped with a ZnSe
crystal for ATR measurements (step width: 1 cmꢀ1; measurement
time for one spectrum: 20 s; spectra accumulation: 16; overall
measuring time for one sample: 80 s; background correction
before each sample). NMR spectroscopy measurements were per-
formed on a Bruker Avance AC200 or AC400 spectrometer with
deuterated solvents as the internal standard. Melting points were
Table 4. Product distribution from 1H NMR spectroscopy experiments
with a solution of the product from the reaction of 21 with 2a or 2b
(Scheme 6).
22a
22b
d(E isomer) [ppm]
d(Z isomer) [ppm]
integral for E isomer
integral for Z isomer
E:Z
5.29
5.55
0.2026
0.1468
58:42
5.30
5.54
0.2569
0.1868
58:42
Conclusion
The reaction of activated ortho-substituted anilines with diethyl
acetylenecarboxylate in a planetary ball mill allows the forma-
tion of the corresponding N-vinyl derivatives by intermolecular
hydroamination. Apart from NaCl as a milling auxiliary, no fur-
ther additive or catalyst was necessary. In the same reaction
with o-hydroxyaniline, the corresponding 2-vinylidene benzox-
azine derivative was formed exclusively. Structural assignment
with FTIR and NMR spectroscopy indicated a Z configuration
of the exocyclic double bond, and a tautomeric shift observed
in protic deuterated solvent proved the suspected structure.
Based on these findings, it is supposed that the heterocycle is
formed sequentially, with an initial intermolecular hydroamina-
tion yielding the N-vinyl derivative, which subsequently under-
goes amidation to close the annulated cycle. By using this se-
quential reaction procedure, several benzoxazines, quinoxa-
lines, and benzothiazines have been synthesized from o-hy-
droxyanilines, o-phenylenediamine, and o-aminothiophenol,
respectively.
measured on
uncorrected.
a Stuart Scientific SMP10 apparatus and are
Synthesis of N-(2-iodophenyl)trifluoromethanesulfonamide
(1n)[29]
A solution of dried THF (50 mL), o-iodoaniline (22 mmol, 4.82 g),
and Et3N (25 mmol, 2.53 g) was cooled to ꢀ108C. With continuous
stirring, a solution of dried THF (30 mL) and Tf2O (24 mmol, 5.04 g)
was added slowly without exceeding a temperature of 08C. The re-
sulting reaction mixture was stirred for 1 h at 08C and then slowly
heated to room temperature. After being stirred for 16 h, the reac-
tion mixture was poured into water (100 mL) and extracted with
EtOAc (3ꢁ100 mL). The combined organic phases were dried over
MgSO4, the solvent was removed under vacuum, and the resulting
solid was purified with column chromatography (n-hexane/EtOAc,
1:1) and washed with n-hexane and water: white powder; 85%;
1H NMR (CDCl3, 600 MHz): d=6.96 (t, J=6.0 Hz, 1H), 7.40 (t, J=
6.0 Hz, 1H), 7.83 (d, J=6.0 Hz, 1H), 8.18 (d, J=6.0 Hz, 1H),
8.30 ppm (s, 1H); 13C NMR (CDCl3, 150 MHz): d=90.3, 115.6, 122.2,
127.9, 129.6, 135.7, 139.2, 154.8 ppm.
The reaction of o-hydroxyaniline with dialkyl acetylenedicar-
boxylates is very fast and may proceed in solution, so solid-
state FTIR spectroscopy was used for the semiquantitative re-
action monitoring. The reagents were applied under different
stress conditions that ranged from pressurization of gently pre-
mixed reagents through manual grinding with a mortar and
pestle to ball milling. Assessment of the subsequent reaction
during measurement indicated that manually ground mixtures
tend to undergo reaction to completeness, whereas reaction
mixtures from other stress conditions revealed only slow sub-
sequent reactions to reach equilibrium values. The conclusion
can be drawn that solid-state FTIR spectroscopy is a powerful
technique for the reaction monitoring of fast reactions without
the need for sample preparation under the condition of fast
and instantaneous measurements. Thus, solution-based
workup for reaction assessment can be avoided, which elimi-
nates the possibility for subsequent reactions in solution. Fur-
thermore, it was demonstrated that cryogenic milling has an
adverse effect for the synthesis in ball milling, but the reaction
Representative procedure for hydroamination reactions in
a planetary ball mill (Table 1)
The milling beakers (V=45 mL, MSZ) were filled with six milling
balls (d=15 mm, MSZ), and then NaCl (5 g), the aniline derivative
(1a–n; 1 mmol), and diethyl acetylenedicarboxylate (2b; 1 mmol,
170 mg) were added sequentially. Milling was performed at
800 minꢀ1 for 10 min. After being allowed to cool to room temper-
ature, the crude products were extracted with ethyl acetate
Chem. Eur. J. 2015, 21, 1 – 13
www.chemeurj.org
9
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
(50 mL). The resulting filtrate was analyzed by GC-FID and GC-MS.
Except for 4a and 4b, the products were not isolated.
51.7 ppm; IR (ATR): n˜max =3239, 1776, 1653, 1608, 898, 763 cmꢀ1
HRMS: m/z calcd for [C11H8ClNO4]Na+: 276.0034; found: 275.9994.
;
(Z)-Methyl
2-(2-oxo-2H-benzo[b][1,4]oxazin-3(4H)-ylidene)ace-
(Z)-Ethyl 2-(6-chloro-2-oxo-2H-benzo[b][1,4]oxazin-3(4H)-ylide-
ne)acetate (15b):[14] yellow solid; 1H NMR (CDCl3, 400 MHz): d=
10.66 (s, 1H), 7.24 (s, 1H), 7.05 (d, J=8 Hz, 1H), 6.94 (d, J=4 Hz,
1H), 5.94 (s, 1H), 4.22 (q, J=8 Hz, 2H), 1.31 ppm (t, J=6 Hz, 3H);
13C NMR (CDCl3, 400 MHz): d=169.7, 155.5, 138.5, 137.3, 130.8,
125.2, 122.5, 118.1, 114.7, 92.7, 60.6, 14.2 ppm; IR (ATR): n˜max =3236,
1760, 1655, 1628, 864, 802, 762 cmꢀ1; HRMS: m/z calcd for
[C12H10ClNO4]Na+: 290.0191; found: 290.0183.
tate (4a):[14] yellow solid; 1H NMR (CDCl3, 400 MHz): d=10.69 (s,
1H), 7.00–7.02 (m, 2H), 6.97 (m, 2H), 5.96 (s, 1H), 3.78 ppm (s, 3H);
13C NMR (CDCl3, 400 MHz): d=170.3, 155.9, 140.0, 138.2, 152.7,
124.2, 122.8, 117.1, 114.9, 90.7, 51.5 ppm; IR (ATR): n˜max =3338,
1762, 1663, 1611, 754 cmꢀ1
.
(Z)-Ethyl 2-(2-oxo-2H-benzo[b][1,4]oxazin-3(4H)-ylidene)acetate
(4b):[14] yellow solid; 1H NMR (CDCl3, 400 MHz): d=10.65 (s, 1H),
7.17–7.10 (m, 2H), 6.97–7.0 (m, 2H), 4.20 (q, J=7.3 Hz, 1H),
1.30 ppm (t, J=7 H7, 3H); 13C NMR (CDCl3, 400 MHz): d=169.6,
155.6, 139.7, 137.7, 125.3, 123.9, 122.4, 116.7, 114.5, 90.8, 60.0,
13.9 ppm; IR (ATR): n˜max =3291, 1767, 1660, 1627, 755 cmꢀ1; HRMS:
m/z calcd for [C12H11NO4]Na+: 256.0580; found: 256.0610.
(Z)-Methyl
2-(6,8-dichloro-7-methyl-2-oxo-2H-benzo[b]
[1,4]oxazin-3(4H)-ylidene)acetate (16a): yellow solid; 1H NMR
(CDCl3, 400 MHz): d=10.59 (s, 1H), 6.92 (s, 1H), 5.95 (s, 1H), 3.77 (s,
3H), 2.41 ppm (s, 3H); 13C NMR (CDCl3, 400 MHz): d=170.0, 154.7,
136.9, 135.2, 131.0, 129.1, 123.3, 123.1, 113.0, 92.3, 51.7, 17.0 ppm;
IR (ATR): n˜max =3391, 1789, 1687, 1623 cmꢀ1; HRMS: m/z calcd for
[C12H9Cl2NO4]Na+: 232.9801; found: 323.9832.
(Z)-Ethyl 2-(6,8-dichloro-7-methyl-2-oxo-2H-benzo[b][1,4]oxazin-
3(4H)-ylidene)acetate (16b): yellow solid; 1H NMR (CDCl3,
400 MHz): d=10.62 (s, 1H), 6.91 (s, 1H), 5.94 (s, 1H), 4.22 (q, J=
8 Hz, 2H), 2.41 (s, 3H), 1.31 ppm (t, J=6 Hz, 3H); 13C NMR (CDCl3,
400 MHz): d=169.7, 154.8, 136.8, 135.2, 131.0, 129.1, 123.3, 123.1,
113.0, 92.3, 51.7, 17.0 ppm; IR (ATR): n˜max =3225, 1767, 1669,
1616 cmꢀ1; HRMS: m/z calcd for [C13H11Cl2NO4]Na+: 337.9957; found:
337.9925.
Representative procedure for the formation of benzoxazines
13–17 in a mixer ball mill
The milling beakers (V=50 mL, stainless steel) were filled with 14
milling balls (d=5 mm, stainless steel), and then o-aminophenol
(1g; 2 mmol, 218 mg) and dimethyl acetylenedicarboxylate (2a;
2 mmol, 284 mg) were added sequentially. Milling was performed
at 15 Hz for 5 min. After being allowed to cool to room tempera-
ture, the crude products were removed and characterized with ATR
FTIR spectroscopy, NMR spectroscopy, and HRMS.
(Z)-Methyl 2-(6-tert-butyl-2-oxo-2H-benzo[b][1,4]oxazin-3(4H)-yli-
1
dene)acetate (17a):[14] yellow solid; H NMR (CDCl3, 400 MHz): d=
10.67 (s, 1H), 7.24 (s, 1H), 7.05 (d, J=8 Hz, 1H), 7.02 (d, J=12 Hz,
1H), 5.89 (s, 1H), 3.76 (s, 3H), 1.28 ppm (s, 9H); 13C NMR (CDCl3,
400 MHz): d=170.4, 156.1, 149.4, 138.4, 137.9, 123.4, 120.0, 116.4,
111.9, 90.2, 51.4, 34.6, 31.3 ppm; IR (ATR): n˜max =3245, 1757, 1667,
(Z)-Methyl 2-(6-methyl-2-oxo-2H-benzo[b][1,4]oxazin-3(4H)-ylide-
ne)acetate (13a):[14] yellow solid; 1H NMR (CDCl3, 400 MHz): d=
10.57 (s, 1H), 7.24 (s, 1H), 6.99 (d, J=12 Hz, 1H), 6.77 (d, J=24 Hz),
5.88 (s, 1H), 3.75 (s, 3H), 2.30 ppm (s, 3H); 13C NMR (CDCl3,
400 MHz): d=170.3, 156.1, 138.2, 138.0, 135.7,123.7, 123.5, 116.7,
115.1, 90.4, 51.4, 20.9 ppm; IR (ATR): n˜max =3234, 1756, 1655, 1611,
1613, 909, 818, 765 cmꢀ1; HRMS: m/z calcd for [C15H17NO4]Na+
:
298.1050; found: 298.1059.
871, 813, 764 cmꢀ1
.
(Z)-Ethyl 2-(6-tert-butyl-2-oxo-2H-benzo[b][1,4]oxazin-3(4H)-yli-
1
dene)acetate (17b):[14] yellow solid; H NMR (CDCl3, 400 MHz,): d=
(Z)-Ethyl 2-(6-methyl-2-oxo-2H-benzo[b][1,4]oxazin-3(4H)-ylide-
ne)acetate (13b):[14] yellow solid; 1H NMR (CDCl3, 400 MHz): d=
10.60 (s, 1H), 7.24 (s, 1H), 6.99 (d, J=8 Hz, 1H), 6.76 (d, J=20 Hz,
1H), 5.88 (s, 1H), 4.21 (q, J=8 Hz, 2H), 2.30 (s, 3H), 1.30 ppm (t, J=
8 Hz, 3H); 13C NMR (CDCl3, 400 MHz): d=169.9, 156.1, 138.1, 138.0,
135.7, 123.8, 123.4, 116.7, 115.0, 90.9, 60.3, 20.9, 14.3 ppm; IR (ATR):
10.70 (s, 1H), 7.24 (s, 1H), 7.03 (d, J=4 Hz, 1H), 6.95 (d, J=4 Hz,
1H), 5.89 (s, 1H), 4.23 (q, J=8 Hz, 2H), 1.30 ppm (m, 12H);
13C NMR (CDCl3, 400 MHz): d=170.1, 156.2, 149.3, 138.3, 137.9,
131.0, 119.9, 116.4, 111.9, 90.7, 60.3, 31.3, 14.3 ppm; IR (ATR): n˜max
=
3242, 1756, 1657, 1632, 856, 814, 762 cmꢀ1; HRMS: m/z calcd for
[C16H19NO4]Na+: 312.1206; found: 312.1219.
n˜max =3239, 1663, 1619, 876, 809, 765 cmꢀ1
.
(Z)-Methyl 2-(7-methyl-2-oxo-2H-benzo[b][1,4]oxazin-3(4H)-ylide-
ne)acetate (14a):[14] yellow solid; 1H NMR (CDCl3, 400 MHz): d=
10.60 (s, 1H), 7.24 (s, 1H), 6.92 (d, J=8 Hz, 1H), 6.82 (d, J=8 Hz,
1H), 5.86 (s, 1H), 3.75 (s, 3H), 2.30 ppm (s, 3H); 13C NMR (CDCl3,
400 MHz): d=170.4, 156.1, 139.8, 138.2, 133.1, 126.3, 121.6, 117.3,
114.5, 89.9, 51.4, 20.8 ppm; IR (ATR): n˜max =3288, 1747, 1658, 1632,
883, 813, 764 cmꢀ1; HRMS: m/z calcd for [C12H11NO4]Na+: 256.0580;
found: 256.0581.
Synthesis of 3-benzylidene-3,4-dihydrobenzo[b][1,4]oxazin-
2-one (18)[40b]
The milling beakers (V=50 mL, stainless steel) were filled with 14
milling balls (d=5 mm, stainless steel), and then o-aminophenol
(1g; 2 mmol, 218 mg), methyl 3-phenylpropiolate (2c; 2 mmol,
320 mg), and PPh3 (2 mmol, 525 mg) were added sequentially. Mill-
ing was performed at 30 Hz for 60 min. After being allowed to cool
to room temperature, the crude product was removed and charac-
terized with ATR FTIR spectroscopy and HRMS: yellow powder,
79%; IR (ATR): n˜max =1742, 1620, 1435 cmꢀ1; HRMS: m/z calcd for
[C15H21NO]H+: 232.1696; found: 232.1717.
(Z)-Ethyl 2-(7-methyl-2-oxo-2H-benzo[b][1,4]oxazin-3(4H)-ylide-
ne)acetate (14b):[14] yellow solid; 1H NMR (CDCl3, 400 MHz): d=
10.63 (s, 1H), 7.24 (s, 1H), 6.92 (d, J=8 Hz, 1H), 6.83 (d, J=8 Hz,
1H), 5.87 (s, 1H), 4.21 (q, J=8 Hz, 2H), 2.30 (s, 3H), 1.31 ppm (t, J=
6 Hz, 3H); 13C NMR (CDCl3, 400 MHz): d=170.0, 156.2, 139.9, 138.1,
133.0, 126.3, 121.7, 117.3, 114.5, 90.4, 60.3, 20.8, 14.3 ppm; IR (ATR):
n˜max =3230, 1754, 1656, 1622, 864, 820, 763 cmꢀ1; HRMS: m/z calcd
for [C13H13NO4]Na+: 270.0730; found: 270,0737.
Representative procedure for the formation of quinoxalines
in a mixer ball mill
(Z)-Methyl 2-(6-chloro-2-oxo-2H-benzo[b][1,4]oxazin-3(4H)-ylide-
ne)acetate (15a):[14] yellow solid; 1H NMR (CDCl3, 400 MHz): d=
10.62 (s, 1H), 7.24 (s, 1H), 7.05 (d, J=8 Hz, 1H), 6.95 (d, J=8 Hz,
1H), 5.94 (s, 1H), 3.76 ppm (s, 3H); 13C NMR (CDCl3, 400 MHz): d=
170.0, 155.4, 138.5, 137.4, 130.9, 125.1, 122.6, 118.1, 114.7, 92.2,
The milling beakers (V=50 mL, stainless steel) were filled with 14
milling balls (d=5 mm, stainless steel), and then o-phenylenedia-
mine (19; 2 mmol, 216 mg) and dimethyl acetylenedicarboxylate
(2a; 2 mmol, 284 mg) were added sequentially. Milling was per-
formed at 15 Hz for 5 min. After being allowed to cool to room
&
&
Chem. Eur. J. 2015, 21, 1 – 13
www.chemeurj.org
10
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
[6] M. Tokunaga, Y. Wakatsuki, J. Synth. Org. Chem. Jpn. 2000, 58, 587.
[7] E. Mizushima, T. Hayashi, M. Tanaka, Org. Lett. 2003, 5, 3349.
[8] A. Leyva, A. Corma, Adv. Synth. Catal. 2009, 351, 2876.
[9] a) N. Lingaiah, N. S. Babu, K. M. Reddy, P. S. S. Prasad, I. Suryanarayana,
Chem. Commun. 2007, 278; b) N. Pasha, N. S. Babu, K. T. V. Rao, P. S. S.
Prasad, N. Lingaiah, Tetrahedron Lett. 2009, 50, 239.
[10] F.-X. Zhu, W. Wang, H.-X. Li, J. Am. Chem. Soc. 2011, 133, 11632.
[11] Y. Luo, Z. Li, C.-J. Li, Org. Lett. 2005, 7, 2675.
[12] a) A. Stolle, T. Szuppa, S. E. S. Leonhardt, B. Ondruschka, Chem. Soc. Rev.
2011, 40, 2317; b) S.-E. Zhu, F. Li, G.-W. Wang, Chem. Soc. Rev. 2013, 42,
7535; c) G.-W. Wang, Chem. Soc. Rev. 2013, 42, 7668.
temperature, the crude products were removed and characterized
with ATR FTIR spectroscopy, NMR spectroscopy, and HRMS.
(Z)-Methyl 2-(3-oxo-3,4-dihydroquinoxalin-2(1H)-ylidene)acetate
(20a):[14] yellow solid; 1H NMR (CDCl3, 400 MHz): d=11.73 (s, 1H),
11.02 (s, 1H), 7.38 (d, J=8 Hz, 1H), 7.30 (d, J=8 Hz, 1H), 7.00 (m,
2H), 5.51 (s, 1H), 3.34 ppm (s, 3H); 13C NMR (CDCl3, 400 MHz): d=
169.5, 155.6, 144.0, 125.2, 124.8, 123.5, 122.6, 115.4, 115.2, 83.4,
50.7 ppm; IR (ATR): n˜max =1737, 1688, 1622, 742 cmꢀ1; HRMS: m/z
calcd for {2[C12H12N2O3]}Na+: 459.1275; found: 459.1266.
(Z)-Ethyl
2-(3-oxo-3,4-dihydroquinoxalin-2(1H)-ylidene)acetate
[13] R. Thorwirth, A. Stolle, Synlett 2011, 2200.
(20b):[14] yellow solid; H NMR (CDCl3, 400 MHz): d=11.71 (s, 1H),
11.04 (s, 1H), 7.37 (d, J=8 Hz, 1H), 7.30 (d, J=8 Hz, 1H), 7.0 (m,
2H), 5.48 (s, 1H), 4.14 (q, J=8 Hz, 2H), 1.22 ppm (t, J=8 Hz, 3H);
13C NMR (CDCl3, 400 MHz): d=169.2, 155.6, 144.0, 125.1, 124.9,
123.5, 122.5, 115.4, 115.2, 83.7, 59.2, 14.4 ppm; IR (ATR): n˜max =1721,
1
[14] G. Choudhary, R. K. Peddinti, Green Chem. 2011, 13, 3290.
[15] F. Schneider, T. Szuppa, A. Stolle, B. Ondruschka, H. Hopf, Green Chem.
2009, 11, 1894.
[16] A. Ramazani, M. Kardan, N. Noshiranzadeh, Synth. Commun. 2008, 38,
383.
[17] B. E. Evans, K. E. Rittle, M. G. Bock, R. M. Di Pardo, R. M. Freidinger, W. L.
Whitter, G. F. Lundell, D. F. Veber, P. S. Anderson, J. Med. Chem. 1988, 31,
2235.
1682, 1614, 742 cmꢀ1
; :
HRMS: m/z calcd for [C12H12N2O3]Na+
255.0710; found: 255.0744.
[18] T. Janosik, N. Wahlstrçm, J. Bergman, Tetrahedron 2008, 64, 9159.
[19] J. A. Salas, C. Mꢃndez, Curr. Opin. Chem. Biol. 2009, 13, 152.
[20] P. A. Gale, Chem. Commun. 2008, 4525.
Representative procedure for the formation of benzothia-
zines in a mixer ball mill
[21] E. Fischer, O. Hess, Ber. Dtsch. Chem. Ges. 1884, 17, 559.
[22] A. Bischler, Ber. Dtsch. Chem. Ges. 1892, 25, 2860.
The milling beakers (V=50 mL, stainless steel) were filled with 14
milling balls (d=5 mm, stainless steel), and then o-aminothiophe-
nol (21; 2 mmol, 250 mg) and dimethyl acetylenedicarboxylate
(2a; 2 mmol, 284 mg) were added sequentially. Milling was per-
formed at 15 Hz for 5 min. After being allowed to cool to room
temperature, the crude products were removed and characterized
with ATR FTIR spectroscopy, NMR spectroscopy, and HRMS.
[23] R. C. Larock, E. K. Yum, M. D. Refvik, J. Org. Chem. 1998, 63, 7652.
[24] A. Stolle, B. Ondruschka, Pure Appl. Chem. 2011, 83, 1343.
[25] a) G. W. Kabalka, L. Wang, V. Namboodiri, R. M. Pagni, Tetrahedron Lett.
2000, 41, 5151; b) G. W. Kabalka, L. Wang, R. M. Pagni, Tetrahedron
2001, 57, 8017; c) E. Tyrrell, L. Whiteman, N. Williams, Synthesis 2009,
829; d) A. Reyes-Sꢄnchez, F. Canavera-Buelvas, R. Barrios-Francisco, O. L.
Cifuentes-Vaca, M. Flores-Alamo, J. J. Garcia, Organometallics 2011, 30,
3340.
[26] M. Zille, A. Stolle, A. Wild, U. S. Schubert, RSC Adv. 2014, 4, 13126.
[27] E. Tullberg, F. Schacher, D. Peters, T. Frejd, Synthesis 2006, 1183.
[28] a) M. P. Kumar, R.-S. Liu, J. Org. Chem. 2006, 71, 4951; b) N. Seshu Babu,
K. Mohan Reddy, P. S. Sai Prasad, I. Suryanarayana, N. Lingaiah, Tetrahe-
dron Lett. 2007, 48, 7642.
Methyl 2-(3-oxo-3,4-dihydrobenzo[b][1,4]thiazin-2-ylidene)ace-
1
tate (22a):[14] yellow solid; H NMR ([D6]DMSO, 400 MHz): d=11.56
(s, 1H), 7.10–6.98 (m, 4H), 6.93 (s, 1H), 3.69 ppm (s, 3H); 13C NMR
([D6]DMSO, 400 MHz): d=165.9, 154.2, 150.7, 141.0, 137.0, 132.7,
125.2, 123.4, 52.8 ppm; IR (ATR): n˜max =1721, 1682, 1614, 742 cmꢀ1
HRMS: m/z calcd for [C11H9NO3S]Na+: 258.0214; found: 258.0195.
;
[29] Y. Chen, N. A. Markina, R. C. Larock, Tetrahedron 2009, 65, 8908.
[30] a) Y. Iwanami, Bull. Chem. Soc. Jpn. 1971, 44, 1311; b) N. Kawahara, T. Na-
kajima, T. Itoh, H. Takayanagi, H. Ogura, Chem. Pharm. Bull. 1984, 32,
1163; c) I. Yavari, S. Souri, M. Sirouspour, M. J. Bayat, Synlett 2009, 1921;
d) X. Li, N. Liu, H. Zhang, S. E. Knudson, R. A. Slayden, P. J. Tonge, Bioorg.
Med. Chem. Lett. 2010, 20, 6306.
Ethyl 2-(3-oxo-3,4-dihydrobenzo[b][1,4]thiazin-2-ylidene)acetate
(22a):[14] yellow solid; 1H NMR ([D6]DMSO, 400 MHz): d=11.56 (s,
1H), 7.10–6.90 (m, 4H), 1.1–1.4 ppm (m, 3H); 13C NMR ([D6]DMSO,
400 MHz): d=165.4, 154.2, 150.8, 140.9, 137.1, 132.7, 127.2, 125.2,
123.4, 60.4, 14.1 ppm; IR (ATR): n˜max =3373, 1723, 1664, 1595,
748 cmꢀ1; HRMS: m/z calcd for [C12H11NO3S]Na+: 272.0352; found:
272.0349.
[31] A. V. Babenysheva, N. A. Lisovskaya, I. O. Belevich, N. Y. Lisovenko,
Pharm. Chem. J. 2006, 40, 611.
[32] G. W. Danswan, P. W. Hairsine, D. A. Rowlands, J. B. Taylor, R. Westwood,
J. Chem. Soc. Perkin Trans. 1 1982, 1049.
[33] M. Augustin, P. Jeschke, Z. Chem. 1987, 27, 211.
[34] Q.-Y. Zhang, B.-K. Liu, W.-Q. Chen, Q. Wu, X.-F. Lin, Green Chem. 2008,
10, 972.
Acknowledgements
The study was partially funded by the Deutsche Bundesstif-
tung Umwelt (DBU; AZ 29622-31). A.S. and M.Z. are thankful to
Esra Altuntas (IOMC, FSU Jena) for performing HRMS analyses.
A.S. acknowledges the work done by Sascha Raufeisen and
Antje Tied with regard to postreaction experiments during ATR
measurements.
[35] It should be pointed out that Av,max >0.46 as a sign for full conversion/
quantitative yield cannot be commonly transferred to other products.
But, for an evaluation of some basic parameters with the reaction of
1g with 2a or 2b as a model, it is a reliable and robust tool.
[36] E. Tullberg, D. Peters, T. Frejd, J. Organomet. Chem. 2004, 689, 3778.
[37] X. Ma, W. Yuan, S. E. J. Bell, S. L. James, Chem. Commun. 2014, 50, 1585.
[38] Often used for analytical sample preparation in order to prepare KBr
discs for IR measurements.
[39] R. Schmidt, R. Thorwirth, T. Szuppa, A. Stolle, B. Ondruschka, H. Hopf,
Chem. Eur. J. 2011, 17, 8129.
Keywords: cascade reactions
spectroscopy · nitrogen heterocycles · solid-state reactions
·
hydroamination
·
IR
[40] a) I. Yavari, M. Adib, L. Hojabri, Tetrahedron 2002, 58, 6895; b) I. Yavari, S.
Souri, M. Sirouspour, H. Djahaniani, Synthesis 2006, 3243.
[41] a) V. Iwanami, Y. Kenjo, K. Nishibe, M. Kajiura, S. Isoyama, Bull. Chem.
Soc. Jpn. 1964, 37, 1740; b) N. Kawahara, T. Nakajima, T. Itoh, H. Ogura,
Heterocycles 1983, 20, 1721; c) S. Sadjadi, R. Hekmatshoar, S. J. Ahmadi,
M. Hosseinpour, M. Outokesh, Synth. Commun. 2010, 40, 607; d) S. Sad-
jadi, S. Sadjadi, R. Hekmatshoar, Ultrason. Sonochem. 2010, 17, 764.
[42] M. M. Heravi, N. Nami, H. A. Oskooie, R. Hekmatshoar, Phosphorus Sulfur
Silicon Relat. Elem. 2005, 180, 1873.
[1] T. E. Mꢂller, M. Beller, Chem. Rev. 1998, 98, 675.
[2] R. Severin, S. Doye, Chem. Soc. Rev. 2007, 36, 1407.
[3] a) J. F. Hartwig, Pure Appl. Chem. 2004, 76, 507–516; b) K. C. Hultzsch,
Org. Biomol. Chem. 2005, 3, 1819.
[4] a) J. Barluenga, F. Aznar, Synthesis 1975, 704; b) J. Barluenga, F. Aznar,
Synthesis 1977, 195.
[5] Y. Zhang, J. P. Donahue, C.-J. Li, Org. Lett. 2007, 9, 627.
Chem. Eur. J. 2015, 21, 1 – 13
www.chemeurj.org
11
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
[43] J. Tang, L. Wang, D. Mao, W. Wang, L. Zhang, S. Wu, Y. Xie, Tetrahedron
2011, 67, 8465.
[44] M. A. P. Martins, C. P. Frizzo, D. N. Moreira, L. Buriol, P. Machado, Chem.
Rev. 2009, 109, 4140.
[45] P. Haldar, B. Dutta, J. Guin, J. K. Ray, Tetrahedron Lett. 2007, 48, 5855.
[46] J. Gris, R. Glisoni, L. Fabian, B. Fernꢄndez, A. G. Moglioni, Tetrahedron
Lett. 2008, 49, 1053.
[48] a) D. R. Patil, S. M. Salunkhe, M. M. Aitawade, M. B. Deshmukh, G. B. Ko-
lekar, P. V. Anbhule, Pharma Chem. 2011, 3, 207; b) S. M. Kalbag, M. D.
Nair, P. Rajagopalan, C. N. Talaty, Tetrahedron 1967, 23, 1911; c) Y. Maki,
M. Suzuki, Chem. Pharm. Bull. 1972, 20, 832; d) M. R. Islami, F. Mollazehi,
A. Badiei, H. Sheibani, ARKIVOC (Gainesville, FL, USA) 2005, 25.
[49] R. M. Acheson, M. W. Foxton, G. R. Miller, J. .Chem. Soc. 1965, 3200.
[47] a) C. Cao, L. R. Fraser, J. A. Love, J. Am. Chem. Soc. 2005, 127, 17614;
b) A. Corma, C. Gonzꢄlez-Arellano, M. Iglesias, F. Sꢄnchez, Appl. Catal. A
2010, 375, 49.
Received: December 1, 2014
Published online on && &&, 0000
&
&
Chem. Eur. J. 2015, 21, 1 – 13
www.chemeurj.org
12
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
FULL PAPER
On the ball: The formation of heterocy-
cles by cascade reactions in ball mills is
investigated (see scheme). In addition
to a study of the reaction conditions
(such as stress mode or reaction tem-
perature) and the suitability of several
reagents, the applicability of attenuated
total reflectance FTIR spectroscopy as
a semiquantitative analytical method is
assessed. This allows direct analysis
without sample pretreatment with sol-
vents, which might have induced un-
wanted subsequent reactions.
&
Solid-State Reactions
M. Weiße, M. Zille, K. Jacob, R. Schmidt,
A. Stolle*
&& – &&
Hydroamination Reactions of Alkynes
with ortho-Substituted Anilines in Ball
Mills: Synthesis of Benzannulated N-
Heterocycles by a Cascade Reaction
Chem. Eur. J. 2015, 21, 1 – 13
www.chemeurj.org
13
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ