Expeditious one-pot, four-step preparation of 2-t-butoxycarbonyl-6-hydroxy- 1,2,3,4-tetrahydroisoquinoline and an improved conversion to its 6-cyano derivative

Article abstract of DOI:10.1080/00397910903318682

3-Methoxy-phenethylamine was subjected to a sequential imination/Pictet- Spengler/demethylation/Boc protection sequence that allowed a one-pot preparation of N-Boc-6-hydroxy-1,2,3,4-tetrahydroisoquinoline 1. This intermediate was elaborated almost quantit

Full text of DOI:10.1080/00397910903318682

This article was downloaded by: [Washington University in St Louis]
On: 22 April 2013, At: 10:19
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Synthetic Communications: An
International Journal for Rapid
Communication of Synthetic Organic
Chemistry
Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/lsyc20
Expeditious One-Pot, Four-Step
Preparation of 2-t-Butoxycarbonyl-6-
hydroxy-1,2,3,4-tetrahydroisoquinoline
and an Improved Conversion to Its 6-
Cyano Derivative
David W. Gordon a & Carol A. Bains a
a Research Chemistry Department, Pfizer Global Research and
Development, Sandwich, UK
Version of record first published: 07 Apr 2010.
To cite this article: David W. Gordon & Carol A. Bains (2010): Expeditious One-Pot, Four-Step
Preparation of 2-t-Butoxycarbonyl-6-hydroxy-1,2,3,4-tetrahydroisoquinoline and an Improved
Conversion to Its 6-Cyano Derivative, Synthetic Communications: An International Journal for Rapid
Communication of Synthetic Organic Chemistry, 40:9, 1399-1404
To link to this article: http://dx.doi.org/10.1080/00397910903318682
PLEASE SCROLL DOWN FOR ARTICLE
Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions
This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden.
The publisher does not give any warranty express or implied or make any representation
that the contents will be complete or accurate or up to date. The accuracy of any
instructions, formulae, and drug doses should be independently verified with primary
sources. The publisher shall not be liable for any loss, actions, claims, proceedings,
demand, or costs or damages whatsoever or howsoever caused arising directly or
indirectly in connection with or arising out of the use of this material.

Synthetic Communications1, 40: 1399–1404, 2010
Copyright # Taylor & Francis Group, LLC
ISSN: 0039-7911 print=1532-2432 online
DOI: 10.1080/00397910903318682
EXPEDITIOUS ONE-POT, FOUR-STEP PREPARATION
OF 2-t-BUTOXYCARBONYL-6-HYDROXY-1,2,3,4-
TETRAHYDROISOQUINOLINE AND AN IMPROVED
CONVERSION TO ITS 6-CYANO DERIVATIVE
David W. Gordon and Carol A. Bains
Research Chemistry Department, Pfizer Global Research and Development,
Sandwich, UK
3-Methoxy-phenethylamine was subjected to
a sequential imination/Pictet–Spengler/
demethylation/Boc protection sequence that allowed a one-pot preparation of N-Boc-
6-hydroxy-1,2,3,4-tetrahydroisoquinoline 1. This intermediate was elaborated almost
quantitatively via an improved triflation/cyanation procedure to its 6-cyano analog 2.
Keywords: Cyanation; one-pot; Pictet–Spengler; tetrahydroisoquinoline; triflation
INTRODUCTION
1,2,3,4-Tetrahydroisoquinoline (THIQ) alkaloids are widespread in nature,[1]
and myriad synthetic THIQs have been manufactured during various medicinal
chemistry programs.[2] While pursuing a medicinal chemistry project, we urgently
required a quantity of N-Boc-6-cyano-THIQ 2. The earliest reported literature synth-
eses of this compound by Selnick et al.[3] described its preparation from N-Boc-6-
hydroxy-THIQ 1. This, in turn, was prepared by an unspecified modification to the
procedure attributed to Buck[4] as illustrated in Scheme 1.
Thus, 3-methoxy-phenethylamine 3 and formaldehyde were condensed to form
imine 4. The water of condensation was first removed before treatment with hydro-
chloric acid to effect Pictet–Spengler cyclization. The bicyclic ether 5 was recrystallized
before separate demethylation to phenol 6 by heating with concentrated hydrochloric
acid in a sealed tube at 170 ꢀC. Boc protection under conventional conditions of
di-t-butyl-dicarbonate and triethylamine in tetrahydrofuran (THF) then furnished
key intermediate 1. Selnick et al.[3] suggested that this route to 1 was amenable to a
large scale, and so it represented a good basis for our requirements. (Compound 1
is not readily commercially accessible in large quantities.)
Received June 19, 2009.
Address correspondence to David W. Gordon, Research Chemistry Department, Pfizer Global
Research and Development, Sandwich, UK. E-mail: david.gordon@pfizer.com
1399

1400
D. W. GORDON AND C. A. BAINS
Scheme 1. (i) 40% aq. formaldehyde, rt; (ii) 23% HCl, evaporated; (iii) conc. HCl, sealed tube, 170 ꢀC; (iv)
Boc2O, NEt3, THF, rt; (v) triflic anhydride, NEt3, DCM, rt; and (vi) Zn(CN)2, Pd(PPh3)4, DMF, 80 ꢀC.
RESULTS AND DISCUSSION
Despite providing a good basis for our needs, the existing route contained
certain undesirable impediments. For instance, there appeared to be opportunities
to use milder conditions (alternatives to sealed-tube methodology), to streamline
the number of unit operations (in particular any unnecessary isolation or purification
of intermediates), and to optimize yields (the overall yield from 6 to 2 was reported
as being only 42%). In seeking an accelerated synthesis, we surveyed various options.
For instance, the Pictet–Spengler reaction of the parent phenol using formaldehyde
in hydrochloric acid initially seemed attractive. However, such reactions have been
reported as being capricious,[5–8] both in terms of regioselectivity and yield, because
of side reactions, including phenol-formaldehyde polymerization, which can prove
dangerous.[9] We therefore elected to persist with the existing protocol of Pictet–
Spengler cyclization of the methoxy derivative and search for a practically simpler
demethylation.
We hypothesized that refluxing 48% (aqueous) hydrobromic acid would be an
ideal reagent for the demethylation step.[10] Additionally, we proposed that this acid
could be used in place of hydrochloric acid to effect the preceding Pictet–Spengler
cyclization. We also reasoned that the water of condensation from the first step
should not unduly interfere in the second step, which uses aqueous conditions
anyway. Furthermore, by making the postdemethylation reaction medium alkaline
and adding a cosolvent, the Boc protection could potentially be performed under
Schotten–Baumann conditions, therefore opening up the possibility of implementing
the entire four-step sequence in a single flask (Scheme 2).
In practice, 3-methoxyphenethylamine 3 was stirred with 1.1 equivalents of
37% aqueous formaldehyde and briefly heated to complete imine formation. The
48% HBr solution was then introduced, and the flask was immersed in a preheated
bath at 100 ꢀC to encourage Pictet–Spengler cyclization versus premature hydrolysis.
After a brief period at this temperature, the solution was stirred under reflux until
demethylation was established. The reaction mixture was then cooled to ambient
temperature, neutralized with 2 N aqueous NaOH, adjusted to pH 8 with NaHCO3,
diluted with 2-methyltetrahydrofuran, treated with di-t-butyldicarbonate, and
stirred overnight. N-Boc-6-hydroxy-THIQ 1 could then be isolated simply by
evaporation of the separated organic layer. In this way, multigram quantities of 1

THIQ PREPARATION
1401
Scheme 2. One-pot conversion of 3-methoxyphenethylamine 3 to N-Boc-6-hydroxy-1,2,3,4-tetrahydroiso-
quinoline 1.
can be furnished within 24 h from inexpensive raw material. On scale-up to 25 g of
starting material 3, this provided a 65% yield for the one-pot, four-step process after
chromatographic purification.
We then turned our attention to conversion of phenol 1 to its triflate and
subsequent cyanation. Literature procedures[11] for aryl alcohol triflation routinely
employ triflic anhydride, a corrosive and water-sensitive reagent that generally works
best at subambient temperatures using an organic base such as pyridine or triethyla-
mine. Selnick[3] quoted 88% yield for the room-temperature triflation of 1 to crude
triflate 7. This was followed by a zinc cyanide–mediated cyanation using Pd(PPh3)4
catalysis with an overall yield of 69% after chromatography. We also found that this
protocol returned only modest yields of 6 (and even then only through repeated
catalyst loading).
Interestingly, Frantz et al.[12] showed that such triflation reactions generally
produce a few percent triethylammonium triflate that is not completely washed free
from the crude triflate (confirmed by close inspection of 1H and 19F NMR), and it is
this that interferes with the subsequent catalytic cyanation process. To circumvent
this, the authors described an alternative Schotten–Baumann procedure for trifla-
tion. This was intriguing to us, as in theory it could allow the continuation of our
one-pot sequence even further. Sadly, yields were significantly inferior to those from
the procedure eventually used (Scheme 3).
Reaction of 1 with 1 equivalent of the much milder alternative triflating reagent
N-phenyltriflimide[13] at ambient temperature gave complete conversion to triflate 7,
Scheme 3. (i) PhN(Tf)2, NEt3, DCM, rt; (ii) Zn(CN)2, Pd(PPh3)4, DMF, 80–100 ꢀC, 86–97% overall yield.

1402
D. W. GORDON AND C. A. BAINS
which was isolated as a crude product after an acid=base wash cycle. To begin with,
cyanation again required multiple reloads of catalyst to achieve good conversions
(longer reaction times were ineffective), and we initially ascribed this to interference
from small amounts of by-product N-phenyltriflamide 8 that had not been
completely washed away from the crude triflate. Nevertheless, an 86% isolated yield
of 100% HPLC-pure material could thus be realized on a 30-g scale, representing a
considerable improvement over literature precedence.[3] Subsequent to the com-
pletion of this piece of larger-scale work, it was found that if the Pd catalyst (even
though recently purchased) was simply washed with ethanol to ensure bright yellow,
crystalline material before transfer to the reaction vessel, >97% HPLC conversion of
the crude triflate could be realized with the standard 4 mol% catalyst by stirring in
technical-grade dimethylformamide (DMF) at 80 ꢀC for 45 min.
In summary, we have demonstrated how a novel, one-pot, four-step procedure
has allowed 25-g scale preparation of 2-t-butoxycarbonyl-6-hydroxy-1,2,3,4-tetrahy-
droisoquinoline 1 within 24 h in 65% yield. This material has been converted almost
quantitatively to its 6-cyano derivative via a milder triflation process, followed by
cyanation using the simple expediency of a prior catalyst wash. The overall 63% yield
for the two-pot, six-step sequence represents a marked improvement on previous
reports that prepare this privileged fragment.
EXPERIMENTAL
General Methods
All commercial reagents and solvents were used directly without purification,
with the exception of tetrakis(triphenylphosphine)palladium. Analytical thin-layer
chromatography (TLC) was performed with silica-gel 60 F254 plates and preparative
flash-column chromatography using Merck Kieselgel 60. Melting points were
determined on a Gallenkamp (Electronic) melting-point apparatus and are uncor-
rected. 1H NMR data for 1 and 2 were recorded on a Varian 400-M Hz instrument,
referenced to internal solvent signals, and conformed to literature values.[3] Mass
spectral (MS) data were recorded on a MassLynx instrument with electrospray
(ES) ionization. For high-performance liquid chromatographi (HPLC) determi-
nation of 2, an Agilent 1200 series machine was used whereby samples were dissolved
in acetonitrile (4 mg mlꢁ1), and 2-ml portions were injected into a Waters XBridge
C18 column of 2.1 ꢂ 50 mm and eluted at a flow rate of 1 ml minꢁ1 with an aqueous
acetonitrile mobile phase (20% to 95% MeCN over 1.35 min, returning to 20% at
1.4 min, in a total run time of 1.5 min) using detection at 210 nm. Nitrile product
2 eluted at 0.558 min, and unreacted triflate 7 eluted at 0.824 min.
Synthetic Chemistry
tert-Butyl 6-hydroxy-3,4-dihydroisoquinoline-2(1H)-carboxylate 1.
Formalin (37 wt.% formaldehyde solution, 15 ml, 185 mmol) was added slowly to
the stirred 3-methoxyphenylethylamine (3, 25 g, 165 mmol) using an ice bath to
maintain an internal temperature of 30 ꢃ 5 ꢀC. Upon complete addition, the mixture
was heated to 100 ꢀC for 1 h and then allowed to cool to ambient. The residue was

THIQ PREPARATION
1403
suspended in 100 ml 48% aq. HBr and immersed in a preheated oil bath at 100 ꢀC.
The reaction mixture was stirred for 1 h at this temperature to effect Pictet–Spengler
cyclization, then at reflux for an additional 2 h to effect methyl ether demethylation.
On cooling to ambient, the solution was neutralized with 2 N aq. NaOH, adjusted to
pH 8 with solid NaHCO3, diluted with an equal volume of 2-methyltetrahydrofuran,
and treated with di-t-butyl dicarbonate (36 g, 165 mmol). The solution was vigor-
ously stirred at ambient temperature overnight. The two phases were separated,
and the organic extract was concentrated and chromatographed on silica, eluting
with 10–50% EtOAc in hexane. Product-bearing fractions were combined and
concentrated to give an oil that solidified on standing (26.7 g, 65% yield). Mp
1
110–111.5 ꢀC (lit.[3a] 110–112 ꢀC). MS (ES) m=z 248 (M – H). H NMR CDCl3 d
6.94 (1H, d, J ¼ 8 Hz), 6.68 (1H, dd, J ¼ 2.4, 8 Hz), 6.63 (1H, d, J ¼ 2.4 Hz), 5.4
(1H, br s), 4.49 (2H, s), 3.61 (2H, t, J ¼ 6 Hz), 2.76 (2H, t, J ¼ 6 Hz), 1.49 (9H, s).
tert-Butyl 6-{[(trifluoromethyl)sulfonyl]oxy}-3,4-dihydroisoquinoline-2(1H)-
carboxylate 7. N-Phenyl triflimide (27 g, 75 mmol) was added in portions to a stir-
red solution of the tert-butyl 6-hydroxy-3,4-dihydroisoquinoline-2(1H)-carboxylate
(1, 17.5 g, 70 mmol) and triethylamine (11 ml, 77 mmol) in dichloromethane (200 ml)
under nitrogen, and the clear, colorless solution was stirred at ambient temperature
overnight. TLC suggested complete conversion; therefore the solution was con-
centrated and the residue was dissolved in 200 ml diethylether before being washed
successively with equal volumes of 10% aq. citric acid, water, 2 N aq. NaOH, and
brine. The organic extract was dried over MgSO4 and concentrated to a waxy solid,
which was used directly for cyanation.
tert-Butyl 6-cyano-3,4-dihydroisoquinoline-2(1H)-carboxylate 2. Crude
tert-butyl 6-{[(trifluoromethyl)sulfonyl]oxy}-3,4-dihydroisoquinoline-2(1H)-carboxy-
late (7, 31.3 g, 82 mmol), zinc cyanide (10.6 g, 90 mmol), and tetrakis(triphenylpho-
sphine)palladium (3.79 g, 4 mol%) were suspended in dimethylformamide (250 ml)
and heated to 100 ꢀC for 2 h. The solution had decolorized, but TLC predicted only
a modest reaction. A further 3 g catalyst were added at ambient temperature before
heating for an additional 2 h, with only marginal improvement by TLC. Another 3 g
catalyst were added, and heating continued for an additional 1 h, by which time the
reaction medium was a persistent orange color and TLC showed complete conversion
of triflate to nitrile. On cooling, the reaction was quenched with an equal volume of
water and EtOAc, the aqueous layer basified with NaHCO3, and the suspension fil-
tered through celite to remove insoluble matter. The aqueous layer was run off from
the filtrate and further extracted with EtOAc. The combined organic extracts were
washed (three times with brine), dried (MgSO4), concentrated, and triturated with hex-
anes to provide 2 aþs an off-white solid (18.3 g, 86%). Mp 142–145 ꢀC (lit.[3a]
1
143–145 ꢀC). MS (ES ) m=z 259 (M þ H). H NMR CDCl3 d 7.49–7.42 (2H, m),
7.20 (1H, d, J ¼ 7.8 Hz), 4.62 (2H, s), 3.66 (2H, t, J ¼ 5.9 Hz), 2.86 (2H, t, J ¼ 5.9 Hz),
1.49 (9H, s). HPLC rt 0.558 min (100%).
Note: Commercially obtained catalyst can be washed by stirring in absolute
ethanol for a minute, filtering, washing the precipitate with a little ether, and briefly
sucking dry on the sinter. The process is repeated, if necessary, to give brilliant
yellow crystalline material that shines brightly under laboratory lights. When
used as such, only 4 mol% catalyst is sufficient (without reload) to achieve

1404
D. W. GORDON AND C. A. BAINS
near-quantitative cyanation (>97% HPLC conversion) in DMF at 80 ꢀC, typically in
less than 1 h.
REFERENCES
1. Bentley, K. W. b-Phenylethylamines and the isoquinoline alkaloids. Nat. Prod. Rep. 2006,
23(3), 444–463.
2. For recent examples, see (a) Alper, P.; Azimioara, M.; Cow, C.; Epple, R.; Jiang, S.; Lelais,
G.; Michellys, P.-Y.; Nguyen, T. N.; Westscott-Baker, L.; Wu, B. Tetrahydroisoquinoline
compounds and compositions as modulators of GPR119 activity and their preparation.
WO Patent 2008097428, 2008; (b) Cowan, D. J.; Larkin, A. L.; Zhang, C.; Musso, D.
L.; Green, G. M.; Cadilla, R.; Spearing, P. K.; Bishop, M. J.; Speake, J. D. Preparation
of di(hetero)aryls as antagonists or inverse agonists at opioid receptors. WO Patent
2008021849, 2008; (c) Habashita, H.; Nakade, S.; Preparation of heterocyclic compounds
having sphingosine-1-phosphate (S1P) receptor binding potency. WO Patent 2006001463,
2006; (d) Ueno, H.; Yokota, K.; Hoshi, J.-I.; Yasue, K.; Hayashi, M.; Hase, Y.; Uchida, I.;
Aisaka, K.; Katoh, S.; Cho, H. Synthesis and structure–activity relationships of novel selec-
tive factor Xa inhibitors with a retrahydroisoquinoline ring. J. Med. Chem. 2005, 48(10),
3586–3604; (e) Babin, D.; Benedetti, Y.; Chatreaux, F.; Weston, J. B. Preparation of novel
1,2,3,4-tetrahydrosioquinolines for use as fungicides. WO Patent 2001074808, 2001.
3. (a) Selnick, H. G.; Smith, G. R.; Tebben, A. J. An improved procedure for the cyanation
of aryl triflates: A convenient synthesis of 6-cyano-1,2,3,4-tetrahydroisoquinoline. Synth.
Commun. 1995, 25(20), 3255–3261; (b) Baldwin, J. J.; Claremon, D. A.; Ponticello, G. S.;
Selnick, H. G. Spirobenzopyrancyclohexanes as class III antiarrhythmic agents. US
Patent 5403846, 1995.
4. Buck, J. S. Some substituted tetrahydroisoquinoline hydrochlorides. J. Am. Chem. Soc.
1934, 56, 1769–1771.
5. Bates, H. A.; Bagheri, K.; Vertino, P. M. Effect of pH on the regioselectivity of Pictet–
Spengler reactions of 3-hydroxyphenethylamine with formaldehyde and acetaldehyde. J.
Org. Chem. 1986, 51, 3061–3063.
6. Ornstein, P. L.; Arnold, M. B.; Augenstein, N. K.; Paschal, J. W. Syntheses of
6-oxodecahydroisoquinoline-3-carboxylates: Useful intermediates for the preparation of
conformationally defined excitatory amino acid antagonists. J. Org. Chem. 1991, 56,
4388–4392.
7. Verschueren, K.; Toth, G.; Tourwe, D.; Lebl, M.; Van Binst, G.; Hruby, V. A facile syn-
thesis of 1,2,3,4-tetrahydro-7-hydroxyisoquinoline-3-carboxylic acid, a conformationally
constrained tyrosine analogue. Synthesis 1992, 458.
8. Liu, C.; Thomas, J.; Brieaddy, L.; Berrang, B.; Carroll, I. An improved synthesis of
(3R)-2-(tert-butoxycarbonyl)-7-hydroxy-1,2,3,4-tetrahydroisoquinoline-3-carboxylic acid.
Synthesis 2008, 6, 856.
9. Consult MSDS for formaldehyde and Bretherick’s Handbook of Reactive Chemical
Hazards, 6th ed.; P. G. Urben (Ed.); Butterworth-Heinemann: Oxford, 1999.
10. Fieser, L. F.; Fieser, M. Reagents for Organic Synthesis; Wiley: New York, 1967; vol. 1,
p. 450.
11. Echavarren, A. M.; Stille, J. K. Palladium-catalyzed coupling of aryl triflates with
organostannanes. J. Am. Chem. Soc. 1987, 109, 5478.
12. Frantz, D. E.; Weaver, D. G.; Carey, J. P.; Kress, M. H.; Dolling, U. H. Practical
synthesis of aryl triflates under aqueous conditions. Org. Lett. 2002, 4(26), 4717.
13. Hendrickson, J. B.; Bergeron, R. Triflamides: New acylating and triflating reagents.
Tetrahedron Lett. 1973, 46, 4607.