Multiple organolithium generation in the continuous flow synthesis of amitriptyline

Article abstract of DOI:10.1002/adsc.201300614

A continuous flow protocol for the preparation of the tricyclic antidepressant (TCA) amitriptyline is reported. The advantages of flow chemistry when handling organometallic agents as well as when performing reaction with gases are demonstrated. Continuous multilithiation combined with carboxylation and the Parham cyclization, a Grignard addition and thermolytic water elimination by inductive heating are key features of the multistep protocol. Copyright

Full text of DOI:10.1002/adsc.201300614

COMMUNICATIONS
DOI: 10.1002/adsc.201300614
Multiple Organolithium Generation in the Continuous Flow
Synthesis of Amitriptyline
Lukas Kupracza and Andreas Kirschninga,*
a
Institut fꢀr Organische Chemie und Biomolekulares Wirkstoffzentrum (BMWZ) der Leibniz Universitꢁt Hannover,
Schneiderberg 1B, 30167 Hannover, Germany
E-mail: andreas.kirschning@oci.uni-hannover.de
Received: July 12, 2013; Revised: August 30, 2013; Published online: November 13, 2013
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201300614.
Abstract: A continuous flow protocol for the prepa-
ration of the tricyclic antidepressant (TCA) amitrip-
tyline is reported. The advantages of flow chemistry
when handling organometallic agents as well as
when performing reaction with gases are demon-
strated. Continuous multilithiation combined with
carboxylation and the Parham cyclization,
a Grignard addition and thermolytic water elimina-
tion by inductive heating are key features of the
multistep protocol.
tion after lithiation of the last aryl bromide moiety to
furnish dibenzosuberone (3).[10a]
Finally, Grignard addition and elimination of water
from the resulting carbinol complete the synthesis of
amitriptyline (1).[9]
As part of our research to expand the scope of syn-
thesis under continuous flow conditions, we devel-
oped a flow protocol that follows this principal route.
The use of microstructured flow reactors is one attrac-
tive technique, among other enabling technologies for
organic synthesis,[11] which shows great advantages
when handling highly reactive intermediates as well
as gases as reaction partners. Additionally, flow
chemistry is readily amenable to multistep synthe-
sis.[12] Often, these processes provide better yields or
selectivities compared to batch processes, because
only a small portion of the reactive intermediate is
formed at a given time which immediately reacts
Keywords: amitriptyline; antidepressants; carboxy-
lation; flow chemistry; inductive heating; lithiation;
Parham cyclization
Amitriptyline (Elavil) (1) is a tricyclic antidepressant under controlled conditions (control refers to reten-
(TCA), that among many other medical indications,
also serves against migraines, tension headaches, anxi-
ety attacks and some schizophrenic symptoms. in
spite of the fact that it has been established as a drug
for some time, it is as effective against depression as
the newer class of selective serotonin reuptake inhibi-
tors (SSRIs).[1] Amitriptyline (1) also shows strong ac-
tivities on the serotonin transporter and moderate ef-
fects on the norepinephrine transporter.[2,3] It acts as
a sodium, calcium, and potassium channel blocker.[4–6]
Some published syntheses[7–9] of amitriptyline ex-
ploit the inherent symmetry present in the tricyclic di-
benzosuberane backbone. 10,11-Dihydro-5H-dibenzo-
ACHTUNGTRENNUNG[a,d]cyclohepten-5-one (3, dibenzosuberone) can prin-
cipally be approached by a one-pot Parham cycliza-
tion[10] which is initiated by a Wurtz-type dimerization
of lithiated benzyl bromide 2 (Figure 1). A second
lithium halogen exchange sets the stage for a monocar-
boxylation using carbon dioxide as electrophile. The
resulting benzoic acid derivative will undergo cycliza-
À
Figure 1. Amitriptyline (1) and the key C C bond-forming
processes.[9,10]
Adv. Synth. Catal. 2013, 355, 3375 – 3380
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3375

Lukas Kupracz and Andreas Kirschning
COMMUNICATIONS
Table 1. Wurtz coupling under batch conditions.
tion time/flow rate and temperature). Pioneering con-
tributions from the Yoshida group demonstrated that
sequential lithium–halogen exchange protocols with
in situ trapping of aryllithium intermediates by differ-
ent electrophiles in microstructured flow devices are
superior to batch protocols.[13,14]
In the synthesis of amitriptyline (1), three different
lithium–halogen exchange reactions take place with
À
precise timing and each initiates a C C bond-forming
reaction. A second challenge of this synthesis is the
controlled introduction of carbon dioxide into the
flow system. This was achieved using Leyꢃs tube-in-
tube system.[15] The reaction takes place in the inter-
stitial liquid phase between the two tubes, with the
carbon dioxide gas being supplied by the outer tube,
[a]
Isolated yields.
which possesses a semipermeable membrane. This (0.2M in THF; flow rate=4 mLminÀ1) and of n-bu-
set-up allows many different gases, including carbon tyllithium (0.4M in hexane; flow rate=2 mLminÀ1)
monoxide,[16] hydrogen,[17] oxygen,[18] ammonia[19] and were independently fed into the reactor using two
ethylene[20] to be employed. Importantly, the gas flow HPLC pumps. The reaction mixture was hydrolyzed
is controlled by pressure and not by metered flow, with methanol at the reactor outlet. The transforma-
simplifying this type of gas–liquid phase reactions. tion proceeded with complete conversion irrespective
Bubbles of gas are avoided entirely.
of whether it was conducted at À908C or at room
The synthesis of the dibenzosuberone (3) under temperature (Table 2, part A) providing a mixture of
batch conditions was described and it requires very the desired 1-bromo-2-phenethylbenzene (5), 1,2-
low temperature (À1008C) for the lithiation step.[10a] bis(2-bromophenyl)ethane (4) as well as 6. Addition-
In order to be able to judge the efficiency of our flow ally, we detected 1-bromo-2-pentylbenzene (7) which
protocol we first established the batch protocol. [2-(2- resulted from the direct substitution reaction of n-bu-
Bromophenethyl)phenyl]lithium was generated from tyllithium with the starting material.
1-bromo-2-(bromomethyl)benzene (2) using one
We also investigated the influence of the inner di-
equivalent of n-butyllithium at À1008C for 1 h. A ameter of the micromixer on the outcome of lithiation
stream of carbon dioxide gas was bubbled through and Wurtz-type coupling (Table 2, part B). A smaller
the mixture for 1.5 h. Then the cooling was removed diameter (Ø=0.25 mm) favours formation of the de-
and at room temperature a stream of dry nitrogen sired aryl bromide 5, supposedly, because it guaran-
was bubbled through the mixture over 1.5 h. The solu- tees more rapid mixing of both reactants. Thus, it was
tion was again cooled to À1008C and n-butyllithium possible to achieve excellent conversions at À508C,
was added. After 30 min at À1008C and 6 h at room within 5 s, providing product 5 in 79% isolated yield.
temperature dibenzosuberone (3) was isolated in 38–
With these results in hand we commenced with the
continuous multistep synthesis of 10,11-dihydro-5H-
56% yield.[10a,21]
The key step of this process is the generation of [2- dibenzoACTHUNTGRNEUNG[a,d][7]annulen-5-one (3). Positioned directly
(2-bromophenethyl)phenyl]lithium. We found that behind R1, a tube-in-tube reactor allowed us to intro-
even at À1008C complete lithiation occurs resulting duce carbon dioxide into the stream of reactants. In
in the formation of the by-product 1,2-diphenylethane a second microtube reator R2 (Ø=1 mm, V=0.5 mL;
(6) in 31% yield (see Table 1, entry 4) besides the ex- PTFA) the carboxylation took place (Table 3: set-up
pected bromide 5. At higher temperatures this pro- A, bottom left). Then, a second solution of n-butyl-
cess prevails (Table 1, entries 1–3). In essence, the lithium (0.4M in hexane) was mixed in M2 (Ø=
practicability of this process has limitations, especially 0.4 mm) with the reaction stream that left R2 and the
when up-scaling is envisaged.
resulting mixture was passed through R3 (Ø=1 mm,
We assumed that flow conditions would improve V=0.5 mL; PTFA). A back-pressure regulator was in-
the handling of organolithium reagents and inter- stalled behind R3 to pressurize the system. All test
mediates as only a small portion of the n-BuLi is sub- samples were finally hydrolyzed in a flask that con-
jected to the reaction conditions at a given time at tained methanol.
a temperature that we hoped to be well above
Carboxylation in R2 was inefficient at À508C, pro-
À1008C.
viding ketone 3 in only low yield. Instead, double hal-
We devised a flow system that consists of a T- ogen–lithium exchange and formation of 1,2-diphenyl-
shaped micromixer M1 (Ø=0.4 mm; PEEK) and a mi- ethane (6) occurred (Table 3, entries 1–5). When the
crotube reator R1 (Ø=1 mm, V=0.5 mL; steel). Two reactor temperature was raised to room temperature,
solutions of 1-bromo-2-(bromomethyl)benzene (2) the yield of ketone 3 improved (entries 6–8). An in-
3376
asc.wiley-vch.de
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2013, 355, 3375 – 3380

Multiple Organolithium Generation in the Continuous Flow Synthesis of Amitriptyline
Table 2. Optimization of Wurtz coupling under flow condi-
Table 3. Multistep flow synthesis to ketone 3 without (set-up
tions.[a]
A) and with (set-up B) gas remover GR.[a]
[a]
T-Shaped micromixer M1 (if not otherwise noted Ø=
0.4 mm; PEEK), microtube reactor R1 (Ø=1 mm, V=
0.5 mL; steel); best results are printed in italics [T(1)=
temperature in R1].
Determined by GC-MS analysis.
Isolated yield.
[b]
[c]
creased flow rate for n-BuLi provided improved
yields of 3. However, an excess of n-butyllithium
[v(3)] was always required, likely due to the presence
of excess carbon dioxide which destroys some of the
reagent. Therefore, a set-up B with a gas remover
(GR) located behind R2 was employed. Here, the re-
action stream passed through a Teflon AF-2400 tube
(50 cm).
[a]
A gas remover GR: M1 (Ø=0.25 mm; steel), microtube
reactor R1 (Ø=1 mm, V=0.5 mL; steel), tube-in-tube
(Ø=2 mm/0.8 mm, V=1 mL, PTFE/Teflon AF-2400);
R2 (Ø=0.8 mm, V=0.5 mL; PTFE), R3 (Ø=0.8 mm,
V=0.5 mL, PTFE), T(1), T(2)=reactor temperatures,
GR (Ø=0.8 mm, V=1 mL, Teflon AF-2400); best results
are printed in italics.
We found that the back-pressure regulator should
be placed directly behind GR and not behind R3.
With this set-up the amount of n-BuLi required could
be reduced. This flow set-up allows the preparation of
ketone 3 from dibromide 2 with an overall residence
time of approximately 33 s at À508C, compared to
about 2 h at À1008C, under batch conditions.[10a] The
[b]
Determined by GC-MS analysis.
Isolated yield.
[c]
isolated yield was 76%, compared to the batch yield V=0.5 mL, PTFE). Hydrolysis took place at the
of 38–56%.
outlet by injecting the stream into a flask filled with
Next, the Grignard addition was investigated methanol. Best results were obtained when 1.5 equiv-
(Table 4).[9] A solution of ketone 3 and [3-(dimethyl- alents of the Grignard reagent in THF or toluene at
ACHTUNGTRENNUNGamino)propyl]magnesium chloride was mixed in a T- room temperature were employed. Under these con-
shaped micromixer M3 (Ø=0.4 mm, PEEK) and ditions the residence time was only about 30 s
pumped through a microtube reactor R4 (Ø=1 mm, (Table 4, entries 3 and 4).
Adv. Synth. Catal. 2013, 355, 3375 – 3380
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3377

Lukas Kupracz and Andreas Kirschning
COMMUNICATIONS
Table 4. Grignard addition.[a]
conductive materials. These materials may serve as
a fixed bed material in flow devices.[23] Consequently,
copper or steel reactors can directly be heated under
these conditions.
Here, we chose a cartridge reactor (60 mm length,
Ø=4 mm, V=0.3 mL, steel) filled with steel beads
(Ø=0.8 mm) which was incased in a high frequency
(hf) inductor (810 Hz).[23f] An injection loop was posi-
tioned between the pump and the reactor for injecting
amine 8.
The temperature of the reaction stream was mea-
sured with a K-element at the outlet of the flow reac-
tor. Next to the K-element we set up a heat-exchang-
er (Ø=1.0 mm, V=0.5 mL, steel) for rapid cooling to
room temperature. The pressure was controlled by
a back-pressure regulator. To our delight, elimination
of water proceeded smoothly (36 s residence time) in
ethanol at 2008C (810 MHz, 8% energy input) in the
absence of any acid and quantitatively yielded ami-
triptyline (1) (see the Supporting Information,
Table S2).[24] After having optimized the Grignard ad-
dition and the elimination, we focused on telescoping
these two steps (Scheme 1).[25]
[a]
M3 (Ø=0.25 mm, steel), R4 (Ø=1 mm, V=0.5 mL,
PTFE), T(3)=reactor temperature; the best result is
printed in italics.
Isolated yield.
[b]
Elimination of the tertiary alcohol with HCl (7M)
Thus, the Grignard product that left R4 was proton-
in ethanol provides amitriptyline hydrochloride in ated by diluting the reaction mixture with ethanol
63% yield.[9] These highly corrosive conditions are and the resulting carbinol was directly pumped into
deleterious for reactor and pump materials so we al- an inductively heated cartridge reactor R5 (V=
ternatively pursued a high temperature high/pressure 0.3 mL, residence time=30 s). At the outlet the crude
elimination using inductive heating (IH).[22]
elimination product was mixed in M5 with a 1 M solu-
During inductive heating an external oscillating tion of HCl in isopropyl alcohol to yield the HCl salt
electromagnetic field induces heat in superparamag- of amitriptyline. One recrystallization (ethanol-ether)
netic nanostructured iron oxide particles or in other provided pure 1·HCl in 71% yield. Salt formation has
Scheme 1. Continuous two-step Grignard addition, elimination and synthesis of 1·HCl. a) M3 (Ø=0.4 mm, PEEK), R4 (Ø=
1 mm, V=0.5 mL, PTFE); M4 (Ø=0.4 mm, PEEK), R5 (length=60 mm, Ø=4 mm, steel, filled with steel beads Ø=
0.8 mm, V=0.3 mL), IH(hf)=high frequency inductive heating (8%, 810 kHz), heat exchanger (V=0.5 mL), M5 (Ø=
0.4 mm, PEEK).
3378
asc.wiley-vch.de
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2013, 355, 3375 – 3380

Multiple Organolithium Generation in the Continuous Flow Synthesis of Amitriptyline
(5%), dried over MgSO4 and concentrated under reduced
several advantages: (i) it provides crystalline material
avoiding chromatographic purification steps and (ii)
the salt form is commonly employed in medical appli-
cations. The simple recrystallization procedure princi-
pally allows to us reach an impurity profile that corre-
sponds well with USP requirements.
pressure. Ketone 3 was obtained as a yellow oil; yield:
126.6 mgminÀ1 (0.61 mmolminÀ1, 76%); 1H NMR (400 MHz,
CDCl3,): d=8.02 (2H, dd, J=1.4, 7.8 Hz), 7.44 (2H, dt, J=
1.5, 7.4 Hz), 7.33 (2H, dt, J=1.3, 7.6 Hz), 7.23 (2H, dd, J=
0.9, 7.5 Hz), 3.22 (4H, s); 13C NMR (CDCl3, 100 MHz): d=
195.6, 141.9, 138.6, 132.3, 130.5, 129.2, 126.6, 34.9; HR-MS
(ESI): m/z=209.0944, calculated for C15H13O+ [M+H]+:
209.0961. All data were in accordance with published
values.[27]
In conclusion, we have disclosed a multistep flow
synthesis of the hydrochloride salt of the tricyclic anti-
depressant amitriptyline (1·HCl). The process in-
volves low temperature, as well as high temperature/
high pressure transformations. Intermediate ketone 3
was formed at a rate of about 127 mgminÀ1, while the
second sequence provided about 8.9 mgminÀ1 of ami-
triptyline (1·HCl). The flow protocols developed
clearly demonstrate the power of microreactor syn-
thesis when highly reactive intermediates such as or-
ganolithium species are generated. Metallations can
be operated at higher temperatures than for the cor-
responding batch processes and therefore proceed
more rapidly. This flow protocol reduces concerns on
safety issues compared to the corresponding batch
protocol because only a small portion of n-BuLi
needs to be subjected to the lithiation and coupling
conditions at a given time.
Synthesis of 5-[3-(Dimethylamino)propyl]-10,11-
dihydro-5H-dibenzoACTHUNGTRENNG[U a,d][7]annulen-5-ol (8) under
Flow Conditions
A solution of 10,11-dihydro-5H-dibenzoACTHNUTRGNEUNG[a,d][7]annulen-5-
one (3) [0.1M, v(4)=1 mLminÀ1] in toluene and a solution
of the Grignard reagent 9 [1M, v(5)=0.15 mLminÀ1] in
THF were mixed in M3 (Ø=0.25 mm, steel). The reaction
solution was passed through R4 (Ø=1 mm, V=0.5 mL;
steel) at room temperature. After reaching steady state the
product stream was collected over a period of 10 min. The
solvent was removed and the solid was hydrolyzed with an
aqueous solution of ammonium chloride (10%, 50 mL). The
mixture was extracted with dichloromethane (3ꢄ50 mL).
The combined, organic phases were dried over magnesium
sulfate, filtered and concentrated under vacuum. Alcohol 8
was obtained as colourless crystals without a detectable
Additionally, we have shown that high temperature
reactions with short residence times can be strategi-
cally implemented in a synthetic sequence. In this
context, inductive heating of flow reactors can be re-
garded as a powerful and safe[26] enabling technology.
trace
of
starting
material;
yield:
22.2 mgmLÀ1
(0.076 mmolmLÀ1, 77%); mp 1188C (ref.[9] 119–1208C);
1H NMR (400 MHz, CDCl3): d=8.04 (2H, dd, J=1.4,
7.9 Hz), 7.21–7.05 (6H, m), 3.52–3.42 (2H, m), 3.02–2.92
(2H, m), 2.52–2.47 (2H, m), 2.27–2.22 (2H, m), 2.22 (6H, s),
1.39–1.32 (2H, m); 13C NMR (CDCl3, 100 MHz): d=145.8,
137.4, 130.2, 126.8, 126.7, 125.9, 76.4, 59.6, 45.1, 44.2, 33.8,
22.3; HR-MS (ESI): m/z=296.2112, calculated for
C20H26NO+ [M+H]+: 296.2009. All data were in accordance
with published values.[9]
Experimental Section
Multistep Flow Synthesis to 10,11-Dihydro-5H-
dibenzoACHTUNGTRENNUNG[a,d][7]annulen-5-one (3)
A solution of 1-bromo-2-(bromomethyl)benzene (2) [0.20M,
v(1)=4 mLminÀ1] in THF and a solution of n-butyllithium
[0.4M, v(2)=2 mLminÀ1] in hexane were mixed in M1 (Ø=
0.25 mm, steel) and the mixture was passed through R1
(Ø=1 mm, V=0.5 mL; steel) at À508C. The reaction
stream was directed through a tube-in-tube reactor (Ø=
0.8 mm, V=1 mL, Teflon AF-2400) at room temperature to
saturate the reaction mixture with carbon dioxide. At the
outlet of R2 (Ø=0.8 mm, V=0.5 mL; PTFE) a gas remover
GR (Ø=0.8 mm, V=1 mL, Teflon AF-2400) was installed.
A back-pressure regulator (75 psi) was placed immediately
after the gas remover, which allowed a more rapid outgas-
sing of carbon dioxide. The resulting solution was mixed
with a solution of n-butyllithium [0.4M, v(3)=3 mLminÀ1]
in hexane at room temperature and passed through R3 (Ø=
0.8 mm, V=0.5 mL, PTFE). When including this microreac-
tor flow system, an overall residence time of 33 s was calcu-
lated. After a steady state had been reached, the product
was collected over a period of 30 min. After diluting the so-
lution with hydrochloric acid (5%, 60 mL), the organic layer
was separated and the aqueous layer was extracted with di-
chloromethane (3ꢄ60 mL). The combined organic phases
were washed with an aqueous sodium hydroxide solution
Two-Step Flow Synthesis towards Amitriptyline
Hydrochloride (1·HCl)
A solution of 10,11-dihydro-5H-dibenzoACTHNUTRGNEUNG[a,d][7]annulen-5-
one (3) [0.1M, v(4)=0.4 mLminÀ1] in toluene and a solution
of Grignard reagent 9 [0.5M, v(5)=0.1 mLminÀ1] in THF
were mixed in M3 (Ø=0.25 mm, steel). The resulting solu-
tion was passed through R4 (Ø=1 mm, V=0.5 mL; steel) at
room temperature and at the outlet it was mixed with etha-
nol [v(6)=0.1 mLminÀ1] in M4 (Ø=0.5 mm, PEEK). The
solution of reactants was pumped into cartridge reactor R5
(length=60 mm, Ø=4 mm, steel, filled with steel beads Ø=
0.8 mm, V=0.3 mL) and heated inductively at high frequen-
cies [8%, 800 kHz; T(4)=2108C]. The reaction stream was
cooled by an exchange heater (V=0.5 mL) before it reached
the back-pressure regulator (110 bar). Finally, the resulting
solution was mixed with a solution of HCl [1M, v(7)=
0.5 mLminÀ1] in isopropyl alcohol in M6 (Ø=0.5 mm,
PEEK) at room temperature. The product was collected
over a period of 30 min and filtered over CeliteTM. The sol-
vent was removed under reduced pressure. Recrystallization
from ethanol-ether solution (1:20) gave amitriptyline hydro-
chloride (1·HCl) as a colourless solid; yield: 8.91 mgminÀ1
Adv. Synth. Catal. 2013, 355, 3375 – 3380
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3379

Lukas Kupracz and Andreas Kirschning
COMMUNICATIONS
(0.028 mmolminÀ1, 71%); mp 1908C (ref.[S6] 193–1948C);
1H NMR (300 MHz, CDCl3): d=7.25–6.99 (8H, m), 5.75
(3H, t, J=7.1 Hz), 3.37–3.19 (2H, m), 3.18–3.03 (1H, m),
3.00–2.81 (2H, m), 2.80–2.71 (1H, m), 2.66 (8H, br s);
13C NMR (100 MHz, CDCl3): d=146.8, 139.6, 138.8, 138.6,
136.7, 129.8, 128.0, 127.8, 127.3, 127.1, 126.8, 126.7, 123.0,
56.5, 42.6, 41.9, 33.2, 31.6, 23.9. All data were in accordance
with published values.[9]
2010, 39, 1183–1202; i) T. Illg, P. Lçb, V. Hessel, Bioorg.
Med. Chem. 2010, 18, 3707–3719.
[13] a) H. Usutani, Y. Tomida, A. Nagaki, H. Okamoto, T.
Nokami, J.-i. Yoshida, J. Am. Chem. Soc. 2007, 129,
3046–3047; b) A. Nagaki, N. Takabayashi, Y. Tomida,
J.-i. Yoshida, Org. Lett. 2008, 10, 3937–3940; c) A.
Nagaki, Y. Uesugi, Y. Tomida, J.-i. Yoshida, Beilstein J.
Org. Chem. 2011, 7, 1064–1069; d) A. Nagaki, Y. Taka-
hashi, S. Yamada, C. Matsuo, S. Haraki, Y. Moriwaki,
S. Kim, J.-i. Yoshida, J. Flow Chem. 2012, 2, 70–72.
[14] J.-i. Yoshida, Chem. Rec. 2010, 10, 332–341.
[15] A. Polyzos, M. OꢃBrien, T. P. Petersen, I. R. Baxendale,
S. V. Ley, Angew. Chem. 2011, 123, 1222–1225; Angew.
Chem. Int. Ed. 2011, 50, 1190–1193.
Acknowledgements
We thank S. Ley and T. Polyzos for generous technical sup-
port in the use of tube-in-tube reactors and especially J.
Wegner for scientific advice.
[16] P. Koos, U. Gross, A. Polyzos, M. OꢃBrien, I. Baxen-
dale, S. V. Ley, Org. Biomol. Chem. 2011, 9, 6903–6908.
[17] S. Newton, S. V. Ley, E. C. Arcꢆ, D. M. Grainger, Adv.
Synth. Catal. 2012, 354, 1805–1812.
[18] T. P. Petersen, A. Polyzos, M. OꢃBrien, T. Ulven, I. R.
Baxendale, S. V. Ley, ChemSusChem 2012, 5, 274–277.
[19] P. B. Cranwell, M. OꢃBrien, D. L. Browne, P. Koos, A.
Polyzos, M. PeÇa-Lꢇpez, S. V. Ley, Org. Biomol. Chem.
2012, 10, 5774–5779.
References
[1] C. Barbui, M. Hotopf, Br. J. Psychiatry: J. Mental Sci.
2001, 178, 129–144.
[2] G. K. Tatsumi, R. D. Blakely, E. Richelson, Eur. J.
[20] S. L. Bourne, P. Koos, M. OꢃBrien, B. Martin, B. Schen-
kel, I. R. Baxendale, S. V. Ley, Synlett 2001, 2643–2647.
[21] We obtained up to 38% yield of 3 when repeating
ref.[10a]
Pharmacol. 1979, 340, 249–258.
[3] F. Hall, J. Schwarzbaum, M. Perona, J. Templin, M.
Caron, K.-P. Lesch, D. Murphy, G. Uhl, Neuroscience
2011, 175, 315–327.
[22] a) A. Kirschning, L. Kupracz, J. Hartwig, Chem. Lett.
2012, 41, 562–570; b) A.-H. Lu, E. L. Salabas, F.
Schꢀth, Angew. Chem. 2007, 119, 1242–1266; Angew.
Chem. Int. Ed. 2007, 46, 1222–1244.
[4] J. J. Pancrazio, G. L. Kamatchi, R. A. K. Roscoe, C.
Lynch, J. Pharmacol. Exp. Therap. 1998, 284, 208–214.
[5] I. Zahradnik, I. Minarovic, A. Zahradnikova, J. Phar-
macol. Exp. Therap. 2008, 324, 977–984.
[23] a) S. Ceylan, C. Friese, C. Lammel, K. Mazac, A.
Kirschning, Angew. Chem. 2008, 120, 9083–9086;
Angew. Chem. Int. Ed. 2008, 47, 8950–8953; b) J.
Wegner, S. Ceylan, C. Friese, A. Kirschning, Eur. J.
Org. Chem. 2010, 4372–4375; c) S. Ceylan, L. Coutable,
J. Wegner, A. Kirschning, Chem. Eur. J. 2011, 17, 1884–
1893; d) L. Kupracz, J. Hartwig, J. Wegner, S. Ceylan,
A. Kirschning, Beilstein J. Org. Chem. 2011, 7, 1441–
1448; e) L. Kupracz, A. Kirschning, J. Flow Chem.
2013, 3, 11–16; f) J. Hartwig, S. Ceylan, L. Kupracz, L.
Coutable, A. Kirschning, Angew. Chem. 2013, 125,
9995–9999; Angew. Chem. Int. Ed. 2013, 52, 9813–9817.
[24] The corresponding batch processes did not provide
a substantial amount of converted product; for details
see the Supporting Information.
[25] Efforts to combine both parts of the synthesis failed.
The lithium analogue of 9 only provided low yields of
addition product 8. The lithium salts present in the
mixture leaving R3 most likely led to transmetallation
when the Grignard reagent 9 was injected at the outlet
of R3.
[6] M. A. Punke, P. Friederich, Anesth. Analg. 2007, 104,
1256–1264.
[7] R. D. Hoffsommer, D. Taub, N. L. Wendler, J. Org.
Chem. 1963, 28, 1751–1753.
ˇ
ˇ
[8] M. Protiva, V. Hnevsovꢅ-Seidlovꢅ, Z. J. Vejdelek, I. Jir-
´
ˇ
kovsky, Z. Votava, J. Metysovꢅ, J. Med. Chem. 1961, 4,
411–415.
[9] D. P. Hudgens, C. Taylor, T. W. Batts, M. K. Patel,
M. L. Brown, Bioorg. Med. Chem. 2006, 14, 8366–8378.
[10] a) D. C. Reames, D. A. Hunt, C. K. Bradsher, Synthesis
1980, 454–456; b) first report: W. E. Parham, L. D.
Jones, Y. Sayed, J. Org. Chem. 1976, 41, 1184–1186.
[11] A. Kirschning, W. Solodenko, K. Mennecke, Chem.
Eur. J. 2006, 12, 5972–5990.
[12] a) I. R. Baxendale, C. Hornung, S. V. Ley, J. d. M. M.
Molina, A. Wikstrçm, Aust. J. Chem. 2013, 66, 131–
144; b) J. Wegner, S. Ceylan, A. Kirschning, Adv.
Synth. Catal. 2012, 354, 17–57; c) R. Yuryev, S. Stromp-
en, A. Liese, Beilstein J. Org. Chem. 2011, 7, 1449–
1467; d) J. Wegner, S. Ceylan, A. Kirschning, Chem.
Commun. 2011, 47, 4583–4592; e) J. P. McMullen, K. F.
Jensen, Annu. Rev. Anal. Chem. 2010, 3, 19–42; f) J.-i.
Yoshida, H. Kim, A. Nagaki, ChemSusChem. 2011, 4,
331–340; g) D. Webb, T. F. Jamison, Chem. Sci. 2010, 1,
675–680; h) S. Marre, K. F. Jensen, Chem. Soc. Rev.
[26] International Commission on Non-Ionizing Radiation
Protection, Health Physics 1998, 74, 494–522.
ˇ ˇ
[27] M. Jereb, D. Vrazic, Org. Biomol. Chem. 2013, 11,
1978.
3380
asc.wiley-vch.de
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2013, 355, 3375 – 3380