Synthesis of Bicyclic and Tricyclic Chiral Guanidinium Salts by an Intramolecular Alkylation Approach

Article abstract of DOI:10.1002/ejoc.201601154

Synthetic studies leading to three new types of chiral guanidinium scaffolds are described. Structures of interest include bicyclic guanidine derivatives with various substitution patterns around the guanidine core, as well as a highly rigid tricyclic scaffold. The challenging targets were accessed by application of mercury(II)-promoted guanylation of N-Boc-substituted thioureas and Ishikawa's desulfurative cyclization. In addition to our synthetic logics, a scalable synthesis of the tricyclic guanidinium salt is presented.

Full text of DOI:10.1002/ejoc.201601154

A Journal of
Accepted Article
Title: Synthesis of Bicyclic and Tricyclic Chiral Guanidinium Salts by an Intramolecu-
lar Alkylation Approach
ˇ
Authors: Aleksej Turockin; William Raven; Philipp Selig
This manuscript has been accepted after peer review and the authors have elected
to post their Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by using the Digital
Object Identifier (DOI) given below. The VoR will be published online in Early View as
soon as possible and may be different to this Accepted Article as a result of editing.
Readers should obtain the VoR from the journal website shown below when it is pu-
blished to ensure accuracy of information. The authors are responsible for the content
of this Accepted Article.
To be cited as: Eur. J. Org. Chem. 10.1002/ejoc.201601154
Link to VoR: http://dx.doi.org/10.1002/ejoc.201601154
Supported by

European Journal of Organic Chemistry
10.1002/ejoc.201601154
FULL PAPER
Synthesis of Bicyclic and Tricyclic Chiral Guanidinium Salts by
an Intramolecular Alkylation Approach
Aleksej Turočkin,*[a][b] William Raven[c] and Philipp Selig*[a][d]
Abstract: Synthetic studies leading to three novel types of chiral
guanidinium scaffolds are described. Structures of interest include
bicyclic guanidines with various substitution patterns around the
guanidine core as well as a highly rigid tricyclic scaffold. The
challenging targets were accessed by application of mercury(II)
promoted guanylation of N-Boc substituted thioureas and Ishikawa’s
desulfurative cyclization. In addition to our synthetic logics, scalable
synthesis of the tricyclic guanidinium salt is presented.
conformational rigidity, whereas acyclic substituted guanidines
and guanidinium salts are known to exist in various
conformations (Figure 1a). As
a
consequence, bicyclic
guanidines and their corresponding salts can undergo directed
interactions with hydrogen bond donors and transition metals
(Figure 1b).[8]
As can be seen from the works of Moyer[9] and Rozas,[10]
conformationally rigid guanidine scaffolds can be obtained by
construction of acyclic species stabilized by intramolecular
hydrogen bonds. Nevertheless, synthesis of saturated bicyclic
guanidines remains the most obvious approach towards the
molecules with desired geometric properties, and also enables
entry to chiral guanidines using readily available amino acids.
Among the noteworthy applications of chiral bicyclic guanidines,
the following developments deserve special attention (Figure 2):
a) novel HIV-protease inhibitors based on the chiral subunit 1;[11]
b) enantioselective recognition of chiral carboxylates using chiral
receptors based on the chiral subunit 1;[12] c) [3.3.0]-bicyclic
Introduction
The guanidine moiety represents an important motif
encountered in numerous molecules of biological significance
including the amino acid arginine and a great variety of marine
alkaloids.[1] Over several decades, this functionality has been
used for the design of small-molecule drugs,[2] imaging agents,[2]
organocatalysts,[3] metal ligands[4] and ion receptors.[5] It is
worthwhile to point out that syntheses of guanidines embedded
into bicyclic systems have been of particular interest, because
these molecules exhibit a range of chemical properties putting
them far ahead of their acyclic counterparts. Bicyclic guanidines
not only stand out as exceptionally strong Brønsted bases,[6] but
have also been identified as very strong nucleophiles.[7]
Furthermore, the species of interest are notable for their
organocatalysts
(2a
and
2b)
.
enabling
numerous
enantioselective transformations.[13]
Figure 2. Structures of the chiral bicyclic guanidine-receptor 1 developed by
Mendoza and Schmidtchen as well as bicyclic guanidine catalysts developed
by Corey (2a) and Tan (2b).
In fact, careful survey of the literature reveals that the number of
successful chiral bicyclic guanidine scaffolds is limited only to
several examples.[14] Owing to their valuable chemical properties
and utilities, such a scarcity of precedents is surprising. Slow
progress in the rational design of novel chiral bicyclic guanidines
can be attributed to the fact that syntheses of these targets are
very difficult due to great experimental challenges associated
with chemical manipulations and purification of basic and highly
water-soluble guanidine-containing intermediates. To date, the
vast majority of known chiral bicyclic guanidines are prepared by
the approach generalized in Figure 3: In the final stages of the
synthesis, a key triamine intermediate is converted to a bicyclic
Figure 1. a) Possible conformations of acyclic guanidinium cations. b)
Directed hydrogen bonding between a bicyclic guanidinium cation and a
carboxylate anion.
[a]
[b]
[c]
[d]
Dr. A. Turočkin and Dr. P. Selig
Institute of Organic Chemistry, RWTH Aachen University,
Landoltweg 1, 52074 Aachen, Germany
Current address: Laboratorium für Organische Chemie, ETH Zürich,
HCI E306, Vladimir-Prelog-Weg 3, 8093 Zürich, Switzerland
E-mail: aleksej.turockin@org.chem.ethz.ch
Dr. W. Raven
Institute of Inorganic Chemistry, RWTH Aachen University,
Landoltweg 1, 52074 Aachen, Germany
scaffold by
a condensation reaction with an appropriate
carbonate equivalent.[15] Among the main limitations of this
strategy is not only the need for careful choice of protecting
groups and occurrence of eventually unstable intermediates,[16]
but also the fact that the critical key guanylation step occurs at
the very end of the synthesis. Moreover, as we tried to
implement the twofold amine-cyclization approach for the
Current address: Patheon, St.-Peter-Str. 25, 4020 Linz, Austria
E-mail: philipp.selig@patheon.com
No http at all.
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved

European Journal of Organic Chemistry
10.1002/ejoc.201601154
FULL PAPER
was reacted with protected aminoalcohol 7 in the presence of
HgCl2 and NEt3, the key intermediate 8 was obtained in
excellent yield despite the significant steric bulk of the
benzhydryl group (Scheme 1). During our initial investigations,
we attempted the deprotection using a two-step procedure
consisting of desilylation with TBAF in THF followed by Boc-
deprotection under acidic conditions. We found, however, that
using TBAF led to unselective cleavage of Boc-groups on the
guanidine functionality resulting in overall lower yields and
experimentally demanding purification of partially deprotected
intermediates. We therefore decided to use more vigorous
conditions and performed a one-pot global deprotection by
treatment of 8 with an excess of aqueous HCl and KF.[20] From
our previous experience, we anticipated that the resulting
guanidinium alcohol would be a highly polar compound. It should
be also mentioned parenthetically that guanidinium salts
containing unprotected hydroxyl-groups frequently represent
considerable challenges with regard to their analysis and give
NMR spectra of very low resolution. Therefore, in order to avoid
the time consuming purification and handling of this species, we
did not isolate it in pure form but rather submitted the crude
intermediate directly to the chlorination step with thionyl chloride
as a crude mixture. Chlorinated guanidinium salt 9 (Figure 6),[21]
on the other hand, was well suitable for chromatographic
purification and, after additional recrystallization, was subjected
to the final cyclization step to provide the final product 3.
Figure 3. Retrosynthetic analysis of bicyclic guanidines via triamine
precursors.
syntheses of chiral bicyclic guanidines, the final guanylation step
frequently suffered from low reproducibility and dissatisfactory
scalability.In order to simplify the development of novel and
hitherto unknown chiral polycyclic guanidines, we propose a
largely neglected synthetic concept: The key guanylation step is
introduced before the assembly of the polycyclic guanidine
skeleton and followed by an intramolecular nucleophilic
substitution using a suitable leaving group (Figure 4).[17] The
advantage of this logic is the introduction of the critical key
transformation early in the synthesis thus providing a robust and
reliable access to target molecules. Furthermore, this approach
should give an alternative access to guanidines of interest, when
the triamine-based approach does not give the desired result.
Figure 4. Retrosynthesis of bicyclic and tricyclic guanidinium salts by
intramolecular nucleophilic substitution
In order to demonstrate the feasibility of the chosen task for the
synthesis of structurally diverse chiral bicyclic guanidines, we
prepared three target molecules presented in Figure 5. Our
study aimed toward the preparation of three distinct types of
guanidinium salts with the following structural features of
increasing complexity: a) one stereogenic center shielding one
of the nitrogen atoms (molecule 3);[18] b) two stereogenic centers
creating a chiral environment around the guanidine core of the
molecule (molecule 4); c) regioseletive construction of N-
substituted guanidines (molecules 4 and 5); d) synthesis of a
highly rigid chiral tricyclic guanidinium salt 5.
Scheme 1. Synthesis of guanidinium salt 3. Reagents and conditions: (a)
HgCl2, NEt3, MeCN, 85%; (b) HCl (aq.), KF, MeCN; (c) SOCl2, MeCN, 85%
over two steps; (d) NaH, MeCN, then HCl in MeOH, 65%.
Figure 5. Structures of the three target guanidinium salts.
Results and Discussion
The inspiration for our approach to 3 was the seminal work by
Kim et. al. on syntheses of guanidines from di-Boc-thiourea
using mercury salts as activating reagents.[19] When thiourea 6
Figure 6. X-Ray crystal structure of the guanidinium salt 9. Ellipsoids are
drawn at 80% probability. Solvent molecules have been omitted for clarity.
This article is protected by copyright. All rights reserved

European Journal of Organic Chemistry
10.1002/ejoc.201601154
FULL PAPER
We planned our approach to the bicycle 4 using the same
strategy presented above (Scheme 2). At first, we identified the
chiral diamine 10[22] as a viable starting material and transformed
it to the corresponding thiourea 12 by the reaction with carbon
disulfide and subsequent protection using Boc2O. Initially, we
expected the thiourea 12 to participate in the mercury(II)
promoted guanylation, as this type of reaction had been reported
with simpler acyclic mono-N-Boc substituted substrates.[23]
However, when 12 was reacted with the amine 7 in the presence
of HgCl2, the desired cyclic guanidine 13 was not formed.
In search of a different approach to 4, we considered the
construction of the cyclic guanidine intermediate by an
alternative C-N bond forming reaction. Thus, starting from the
corresponding acyclic thiourea intermediate, we planned to
implement Ishikawa’s guanylative cyclization using 2-chloro-1,3-
dimethylimidazolinium chloride (DMC, 16) (Scheme 3).[24] Upon
reacting the amine 10 and isothiocyanate 14,[22] the only isolated
product most likely possessed the structure 15.[25] Although in
the course of this reaction, the formation of an alternative
regioisomer of thiourea was expected as well, desulfurative
cyclization of either regioisomer would lead to the desired
intermediate 17. Cyclization of 15 led to the formation of the
protected guanidine 17 and, after subsequent desilylation using
polymer supported fluoride, we obtained the guanidinium alcohol
intermediate 18. Purification of this species appeared to be a
challenging task due to its very unfavorable chromatographic
behavior. When we attempted trituration of the crude reaction
mixture with Et2O/pentane in order to remove the non-polar silyl
side products, we observed a significant dissolution of the
guanidinium salt as well and were only able to isolate 18 in low
yield. In order to minimize the material losses, purification was
omitted at this stage and the crude reaction mixture was
chlorinated to 19, which was amenable to chromatographic
purification. Finally, after NaH-mediated cyclization, densely
decorated target bicyclic guanidinium salt 4 was obtained.[26]
We now turned to the synthesis of tricycle 5. Our first operations
were targeted towards the construction of guanidine 24, which
was regarded as a key intermediate en route to the structure of
interest (Scheme 4). Among the two available entries to such
monocyclic tetrasubstituted guanidines, we considered
Ishikawa’s intramolecular guanylation as the only suitable
strategy owing to its mild reaction conditions compatible with
functionalized intermediates.[27] At first, we prepared thiourea 22
by addition of the pyrrolidine 20[22] to the isothiocyanate 21. Our
Scheme 3. Synthesis of the guanidinium salt 4. Reagents and conditions: (a)
10, NEt3, MeCN, 51%; (b) 16, NEt3, MeCN, 68%; (c) fluoride on polymer,
MeCN/CH2Cl2; (d) SOCl2, MeCN, 69% over two steps; (e) NaH, MeCN, then
HCl in MeOH 84%.
attempts to obtain the free amine 23 failed, as we were not able
to remove the Cbz-protecting group under hydrogenolytic
conditions. The backup strategies to proceed further in the
synthesis through the corresponding carbamate precursor 22
were not successful either, as 22 could not be cyclized to 25
under Ishikawa-conditions. Despite the existence of very similar
Scheme 2. Attempted synthesis of the cyclic guanidine 13. Reagents and
conditions: (a) CS2, EtOH/H2O, 66%; (b) NaH, Boc2O, THF, 95%; (c) 7, HgCl2,
NEt3, MeCN.
Scheme 4. Attempted syntheses of the cyclic guanidinines 24 and 25.
Reagents and conditions: (a) NEt3, MeCN, 73%; (b) Pd/C or Pd(OH)2, H2; (c)
19, NEt3, MeCN
This article is protected by copyright. All rights reserved

European Journal of Organic Chemistry
10.1002/ejoc.201601154
FULL PAPER
precedent in the literature,[24c] our system suffered from rapid
cleavage of the TBS-protecting group leading to undesired side
reactions caused by the free hydroxyl group. During our further
attempts of desulfurative cyclizations we explored the use of
mercury(II) salts, but in every case observed decomposition of
22 or formation of undesired products.
based reagents and catalysts. For that purpose, we probed the
scalability of the synthesis of 5 and prepared 840 mg (3 mmol)
of this salt. We did face practical limitations in performing the
last three steps on a large scale, since the global deprotection of
28 using concentrated acid and subsequent workup procedures
put some constraints on operations in laboratory settings.
Nevertheless, end-game chemistry could be easily performed on
gram-scale thus giving us rapid access to practical amounts of
the guanidinium salt 5. The key intermediate 28, on the other
hand, was conveniently prepared on decagram scale (25 g, 30
mmol) from commercially available pyroglutamic acid in a single
run.
The solution to this synthetic puzzle was found, as we
envisioned that the target structure could be alternatively
accessed by a twofold cyclization of a monocyclic guanidine
precursor. As a possible key transformation, we considered the
N-alkylation at the guanidine core[28] enabling the introduction of
all carbon atoms and functionalities prerequisite for the
assembly of the guanidinium salt 5. Thus, the intermediate 27,
easily obtained after Kim-guanylation, was reacted with TBDPS-
protected bromopropanol in the presence of NaH in anhydrous
DMF (Scheme 5). To our great delight, all protecting groups
were well tolerated under the reaction conditions and the fully
substituted guanidine 28 was obtained in 53% yield. This key
intermediate was subsequently subjected to vigorous
deprotection by treatment with aqueous HCl and KF,[20] followed
by chlorination and final twofold cyclization. A special note
should be made about the purification of the tricyclic
guanidinium salt 5.[26] This compound possessed a very high
polarity and therefore was not suitable for chromatographic
purification. Our attempts at purification by recrystallization were
fruitless as well, as the aggregate state of 5 was an amorphous
foam. Therefore, owing to the good solubility of 5 in both CH2Cl2
and aqueous media, we applied a series of extractions using
various solvent systems in order to remove organic and
inorganic impurities present after the final cyclization step.
Conclusions
In summary, we have developed the syntheses of three novel
types of chiral guanidinium salts. Structures of interest are both
bicyclic and tricyclic systems incorporating dense substitution
patterns at the guanidine core and in direct proximity to it. This
study represents a rare example of strategic application of Kim-
guanylation for the syntheses of complex bi- and tricyclic
guanidines.[29] Moreover, our strategies took advantage of N-
alkylation of N-Boc substituted guanidines as well as Ishikawa’s
guanylative cyclization and enabled the preparation of every
target molecule in only 4 to 5 synthetic operations starting from
readily available chiral amines and thioureas. We would like to
point out that the established syntheses rely on simple and
robust experimental operations. The sequence towards 28 was
easily scaled up and afforded as much as 25 g of this
intermediate. Careful optimization of purification procedures
enabled the access to the polar target molecules by means of
orthodox chromatographic, crystallization and extraction
methods and obviated the use of high-priced HPLC-techniques
or reverse phase silica gel. In addition, due to commercial
availability of numerous chiral amine building blocks, the
presented approaches can be extrapolated to enable rapid
access to novel rigid guanidine-based reagents, organocatalysts
and transition metal ligands.
Among the major objectives of this study was also practical
applicability of our work for the development of novel guanidine-
Experimental Section
General: Experimental details are provided in the Supporting Information.
(S)-di-tert-butyl
2-((4-((tert-butyldiphenylsilyl)oxy)-1,1-diphenyl-
butan-2-yl)imino)di-hydropyrimidine-1,3(2H,4H)-dicarboxylate (8): To
a solution of protected aminoalcohol 7[30] (1.95 g, 4.06 mmol, 1.2 equiv.)
in MeCN (25 mL, HPLC-Grade) thiourea 6[31] (1.07 g, 3.38 mmol) was
added. To this solution NEt3 (1.45 mL, 10.48 mmol, 3.1 equiv.) and HgCl2
(1.83 g, 6.76 mmol, 2.0 equiv.) were sequentially added which resulted in
the formation of a precipitate. The resulting suspension was stirred at r.t.
and monitored by TLC (EtOAC/pentane = 2/8). After 11 h, and 2 d 11 h
respectively, two additional portions of HgCl2 (each consisted of 460 mg,
1.69 mmol, 0.5 equiv.) were added and the reaction mixture was further
stirred at the same temperature. After the total reaction time of 3.5 d,
TLC showed full consumption of the thiourea 6. The suspension was
filtered through a pad of celite and concentrated in vacuo. The resulting
crude material was purified by column chromatography (Si, 233 cm,
pentane (100%) → EtOAc/pentane = 1/9 → 2/8) to give the Boc-
protected guanidine 8 (2.19 g, 2.87 mmol, 85%) as a white foam. Rf = 0.6
(EtOAc/pentane = 2/8, [UV, bromocresol green]). Due to bulkiness of
protecting groups, an extreme signal broadening resulting from hindered
Scheme 5. Synthesis of the tricyclic guanidinium salt 5. Reagents and
conditions: (a) HgCl2, NEt3, DMF, 89%; (b) NaH, DMF, 53%; (c) HCl (aq.), KF,
MeCN; (d) SOCl2, MeCN, 80% over two steps; (e) NaH, MeCN, then HCl in
MeOH, 64%.
This article is protected by copyright. All rights reserved

European Journal of Organic Chemistry
10.1002/ejoc.201601154
FULL PAPER
rotation was observed in the NMR spectra. 1H NMR (600 MHz, CDCl3) δ
7.58 (t, J = 6.6 Hz, 4H), 7.44 – 7.40 (m, 4H), 7.34 (td, J = 7.3, 2.6 Hz, 4H),
7.22 – 7.09 (m, 8H), 4.33 (br s, 1H), 4.19 – 4.02 (br m, 1H), 3.81 – 3.70
(br m, 2H), 3.64 – 3.44 (br m, 1H), 3.42 – 3.28 (br m, 2H), 2.12 – 1.93 (br
m, 2H), 1.67 – 1.54 (br m, 3H), 1.50 (s, 9H), 1.45 (s, 9H), 1.01 (s, 9H).
13C NMR (151 MHz, CDCl3) δ 154.0, 142.8, 140.2, 135.7, 134.03, 134.00,
130.1, 129.6, 129.2, 128.3, 128.1, 127.74, 127.73, 126.34, 126.26, 83.8,
(S)-2-benzhydryl-2,3,4,6,7,8-hexahydro-1H-pyrimido[1,2-a]pyrimidine
hydrochloride (3): To a solution of guanidinium chloride 9 (310 mg, 819
µmol) in MeCN (20 mL, dry) NaH (132 mg, 60% dispersion in paraffin oil,
3.28 mmol, 4.0 equiv.) was added and the mixture was stirred at r.t. for
17.5 h. The reaction mixture was acidified with HCl (1M in MeOH) to pH
1-3 and all volatiles were removed under reduced pressure. The crude
product was subsequently purified by column chromatography (Si, 153
cm, MeOH/CH2Cl2 = 1/9 → MeOH/EtOAc = 3/7). Isolated material
(desired product contaminated with silica particles) was suspended in
MeCN and filtered. After concentration of the filtrate under reduced
pressure, the final product 3 (182 mg, 532 µmol, 65%) was obtained as a
white foam. Rf = 0.32 (EtOAc/MeOH = 7/3 [UV, bromocresol green,
KMnO4]). 1H NMR (600 MHz, CD3OD): δ 7.46 (d, J = 7.3 Hz, 2H), 7.39 –
7.34 (m, 4H), 7.32 – 7.28 (m, 2H), 7.28 – 7.24 (m, 1H), 7.23 – 7.19 (m,
1H), 4.37 (ddd, J = 10.8, 8.1, 3.9 Hz, 1H), 3.95 (d, J = 10.9 Hz, 1H), 3.40
– 3.35 (m, 5H), 3.30 – 3.21 (m, 2H), 2.05 – 1.95 (m, 2H), 1.90 – 1.82 (m,
1H), 1.79 – 1.70 (m, 1H). 13C NMR (151 MHz, CD3OD): δ 152.2, 142.3,
82.0, 81.3, 61.4, 55.0, 44.7, 44.2, 35.5, 28.4, 28.4, 27.0, 22.7, 19.3. Two
13C-signals were not included due to insufficient resolution. [α]D
=
21C
53.5 (CHCl3, c = 1). I cm-1 (w), 2933 (m), 2347 (w),
1698 (s), 1471 (m), 1363 (s), 1276 (m), 1138 (s), 945 (w), 852 (w), 739
(s), 700 (w). MS (EI, 70 eV) m/z (%) 504.1 (1), 394.1 (4), 166.9 (17), 57.1
(100). MS (ESI, Orbitrap) m/z 784.4 (M+Na+), 662.4 [(M-Boc+2H+)+],
562.3
[(M-2Boc+3H+)+].
HRMS
(ESI,
Orbitrap)
calcd.
for
C46H59N3NaO5Si+ (M+Na+): 784.41162; found: 784.41058.
(S)-4-chloro-1,1-diphenyl-N-(tetrahydropyrimidin-2(1H)-ylidene)bu-
tan-2-amine hydrochloride (9): Note: Mixing of the concentrated HCl
with KF produces highly toxic HF in the course of reaction. The produced
waste should therefore be handled with care and disposed accordingly.
To a solution of protected guanidine 8 (2.09 g, 2.74 mmol) in MeCN (25
mL, HPLC-Grade) concentrated aqueous hydrochloric acid (37%, 15 mL)
was added and the mixture was stirred at r.t. for 1 h 20 min. After
addition of KF (2.45 g, 42 mmol), the resulting suspension was vigorously
stirred at 75 °C for 11 h. Subsequently, all volatiles were carefully
removed under reduced pressure using high vacuum pump and cooling
trap for corrosive reaction components (heating to 50-80 °C). The
resulting clotted solid cake was mortared to a fine powder and further
dried at high vacuum (70 °C). The obtained fine powder was suspended
in pentane/Et2O = 5/1 (35 mL) and this suspension was stirred for 30
min.[32] The solvent was decanted and discarded. The remaining solids
were washed with pentane/Et2O once again and the solvent was
discarded. After the above washing with pentane/Et2O, solid material was
suspended in MeOH (50 mL), stirred for 30 min and filtered. Solid residue
remaining on the filter paper was washed with an additional portion of
MeOH (20 mL) and discarded. Combined methanolic filtrates were
concentrated and dried at high vacuum (50 °C, 2 h) to give the crude
deprotected monocyclic guanidinium alcohol intermediate. This material
was contaminated with inorganic salts (KF and KCl) and was subjected
to the subsequent chlorination step without further purification.[33] Crude
deprtected guanidinium alcohol (theoretical amount 2.74 mmol) was
suspended in MeCN (HPLC-Grade, 30 mL), treated with SOCl2 (2 mL,
27.4 mmol, 10 equiv.) and the resulting suspension was stirred at r.t. for
12.5 h. After addition of MeOH (20 mL) and stirring at r.t. for 40 min, all
volatiles were removed in vacuo to afford a sticky solid. This material was
suspended in MeCN (30 mL), filtered off, and remaining solids washed
with an additional portion of MeCN (10 mL). Combined filtrates were
concentrated under reduced pressure to give the crude product as a
brown foam. It was taken up with MeOH/CH2Cl2 (ca. 1/100) and applied
onto a column for chromatographic purification. Purification by column
chromatography (Si, 193 cm, MeOH/CH2Cl2 = 1/100 → 5/95) afforded
an off-white foam. This material was subsequently recrystallized from
MeOH/CH2Cl2 (slow evaporation at r.t.) to give 9 (870 mg, 2.3 mmol,
85% over two steps) as a white solid. This guanidinium salt is extremely
poorly soluble in pure MeOH or CH2Cl2, but shows very good solubility in
the mixtures of these two solvents. Rf = 0.4 (MeOH/CH2Cl2 = 1/10, [UV,
bromocresol green]). 1H NMR (400 MHz, CD3OD) δ 7.45 (d, J = 7.4 Hz,
2H), 7.40 – 7.34 (m, 4H), 7.32 – 7.18 (m, 4H), 4.51 (td, J = 10.5, 2.9 Hz,
1H), 3.94 (d, J = 10.8 Hz, 1H), 3.68 – 3.54 (m, 2H), 3.18 – 2.98 (m, 4H),
2.10 – 2.00 (m, 1H), 1.90 – 1.81 (m, 1H), 1.61 (br s, 2H). 13C NMR (151
MHz, CD3OD) δ 154.4, 142.9, 142.2, 130.0, 129.9, 129.4, 129.3, 128.2,
128.0, 59.7, 53.9, 42.1, 39.5, 37.6, 20.8. [α]D22C = +82.2 (MeOH, c = 1).
m. p.: 87-90 °C (MeOH/CH2Cl2). I cm-1 (w), 3226 (s),
2327 (w), 1636 (s), 1536 (w), 1309 (m), 1081 (m), 1023 (s), 903 (w), 804
(w), 749 (s), 697 (s). MS (EI, 70 eV) m/z (%) 164.9 (100), 151.9 (35),
137.9 (83), 109.9 (8). MS (ESI, Orbitrap) m/z 342.2 [(M-Cl-)+], 306.2 [(M-
141.8, 130.1, 129.9, 129.2, 128.9, 128.3, 128.1, 57.9, 52.5, 47.6, 46.5,
21C
38.9, 26.2, 21.4. [α]D
= +5.8 (MeOH, c = 1.0). This is a known
compound and was previously prepared in our group by a different
synthetic approach.[18]
(S)-1-benzyl-5-(tert-butyl)imidazolidine-2-thione (11): To a solution of
diamine 10[22] (500 mg, 2.42 mmol) in EtOH/H2O (1 mL + 1 mL) CS2 (161
µL, 2.66 mmol, 1.1 equiv.) was added and the resulting emulsion was
heated to reflux for 7 h. The reaction mixture was cooled to r.t. and left to
stand overnight (15 h). Upon standing at r.t., the first portion of product
11 (127 mg) precipitated as an off-white solid. It was isolated by filtration
and washing with MeOH (4 mL). The filtrate and mother liquor were
combined and concentrated under reduced pressure (to ca. 2-3 mL)
leading to precipitation of additional 230 mg of 11. Remaining
supernatant was purified by flash column chromatography (Si, 2.020 cm,
EtOAc/pentane = 1/4 → 1/3) to afford another 39 mg of 11. All crops
were combined to give the thiourea 11 (396 mg, 1.59 mmol, 66%) as an
off-white solid. Rf = 0.55 (EtOAc/pentane = 1/2 [UV, KMnO4]). 1H NMR
(600 MHz, CDCl3) δ 7.32 (t, J = 7.3 Hz, 2H), 7.26 (t, J = 8.2 Hz, 3H), 6.62
(br s, 1H), 5.84 (d, J = 15.6 Hz, 1H), 4.35 (d, J 1 .6 Hz, 1H , . 1 −
3.41 (m, 3H), 0.93 (s, 9H). 13C NMR (151 MHz, CDCl3) δ 185.6, 136.8,
128.8, 127.8, 127.6, 67.0, 51.6, 44.6, 36.0, 26.4. [α]D23C = 123.2 (CHCl3,
c = 1). m. p.: 130-133 °C (EtOAc). I cm-1 (w), 3265 (s),
2953 (s), 1739 (s), 1506 (s), 1358 (s), 1218 (s), 1034 (m), 966 (s), 809
(w), 700 (s). MS (EI, 70 eV) m/z (%) 248.2 (85), 191.1 (96), 91.1 (100),
57.1 (14). MS (ESI, Orbitrap) m/z 249.2 (M+H+). HRMS (ESI, Orbitrap)
calcd. for C14H21N2S+ (M+H+): 249.14200; found: 249.14171.
(S)-tert-butyl 3-benzyl-4-(tert-butyl)-2-thioxoimidazolidine-1-carbo-
xylate (12): A suspension of NaH (114 mg, 60% dispersion in paraffin oil,
2.84 mmol, 2.0 equiv.) in THF (dry, 15 mL) was cooled in an ice/water
bath. To this mixture, a solution of thiourea 11 (354 mg, 1.42 mmol) in
THF (dry, 6.5 mL) was added dropwise. After stirring for 20 min at the
same temperature, a solution of Boc2O (358 mg, 1.64 mmol, 1.15 equiv.)
in THF (dry, 4 mL) was added dropwise. The reaction mixture was kept in
the cooling bath for additional 20 min and then stirred at r.t. for 2.5 h.
Reaction was quenched by careful addition of saturated aqueous
NaHCO3 (5 mL) and the obtained mixture was poured into a separatory
funnel containing H2O (30 mL). Organic layer was separated, and
aqueous phase was extracted with EtOAc (220 mL). Combined organic
layers were dried over Na2SO4 and concentrated in vacuo. The obtained
solid material was carefully washed with pentane (1.5 mL) and after
decantation of the solvent and drying under reduced pressure, protected
thiourea 12 (470 mg, 1.35 mmol, 95%) was obtained as an off-white solid.
Rf = 0.73 (EtOAc/pentane = 1/4 [UV, KMnO4]). 1H NMR (400 MHz,
CDCl3) δ 7.27 – 7.16 (m, 5H), 6.01 (d, J = 15.5 Hz, 1H), 4.28 (d, J = 15.5
Hz, 1H), 3.83 (dd, J = 11.5, 3.1 Hz, 1H), 3.66 (dd, J = 11.5, 9.6 Hz, 1H),
3.23 (dd, J = 9.5, 3.1 Hz, 1H), 1.49 (s, 9H), 0.87 (s, 9H). 13C NMR (101
MHz, CDCl3) δ 180.5, 150.3, 136.0, 128.6, 127.6, 82.9, 63.9, 51.5, 48.1,
36.1, 28.1, 26.2. One 13C-signal was not listed due to insufficient
resolution. [α]D21C = 106.5 (CHCl3, c = 1). m. p.: 118-121 °C (EtOAc). I
2 cm-1 (s), 1739 (s), 1461 (s), 1378 (s), 1266 (s), 1147 (s),
+
H+-2Cl-)+]. HRMS (ESI, Orbitrap) calcd. for C20H25ClN3 [(M-Cl-)+]:
342.17315; found: 342.17315.
This article is protected by copyright. All rights reserved

European Journal of Organic Chemistry
10.1002/ejoc.201601154
FULL PAPER
968 (s), 857 (m), 709 (s). MS (EI, 70 eV) m/z (%) 348.1 (62), 292.1 (84),
191.0 (100), 90.9 (60), 57.1 (41). MS (ESI, Orbitrap) m/z 719.4 (2M+Na+),
371.2 (M+Na+), 271.1 (M-Boc+Na+). HRMS (ESI, Orbitrap) calcd. for
C19H28N2NaO2S+ (M+Na+): 371.17637; found: 371.17648.
for 12 h. Polymer beads were filtered off, washed with MeOH (20 mL)
and the solvents were removed under reduced pressure. In order to
remove non-polar side products, the obtained white solid was washed
with Et2O/pentane = 1/5 (420 mL). This procedure afforded the
deprotected guanidinium alcohol 18 in analytically pure form. The yield of
this material was, however, very low (ca. 40%), due to partial dissolution
of the desired product in the Et2O/pentane. Therefore, after complete
characterization of 18, it was combined with all contaminated
supernatants and quantitative yield was assumed. This mixture of 18 and
silyl-side products was then directly subjected to the chlorination step. Rf
= 0.0-0.38 (strong tailing on TLC; MeOH/CH2Cl2 = 1/9 [UV, bromocresol
green]). 1H NMR (400 MHz, CD3OD) δ 7.47 (d, J = 7.1 Hz, 2H), 7.44 (d, J
= 7.2 Hz, 2H), 7.33 – 7.16 (m, 9H), 6.55 (d, J = 6.8 Hz, 2H), 4.74 (td, J =
11.0, 2.5 Hz, 1H), 4.61 (d, J = 15.7 Hz, 1H), 4.18 (d, J = 15.8 Hz, 1H),
4.10 (d, J = 11.3 Hz, 1H), 3.69 – 3.62 (m, 1H), 3.58 – 3.51 (m, 2H), 3.37
– 3.31 (m, 1H), 3.14 (dd, J = 9.6, 3.4 Hz, 1H), 1.93 – 1.85 (m, 1H), 1.67 –
1.60 (m, 1H), 0.90 (s, 9H). 13C NMR (101 MHz, CD3OD) δ 160.8, 143.9,
143.1, 135.9, 130.12, 130.07, 130.0, 129.2, 129.0, 128.9, 128.8, 128.3,
128.0, 67.0, 59.3, 58.7, 55.6, 51.6, 44.3, 37.5, 37.0, 26.0. 19F NMR (376
MHz, CD3OD) δ 144.7. [α]D21°C = 62.6 (MeOH, c = 1). m. p.: 95-98 °C
(MeOH). I 2 6 cm-1 (m), 2869 (m), 2047 (w), 1652 (s), 1599
(s), 1490 (m), 1449 (s), 1260 (m), 1099 (m), 937 (w), 745 (s), 698 (s). MS
(EI, 70 eV) m/z (%) 456.3 (34), 288.1 (55), 258.1 (35), 167.0 (41), 90.9
(100). MS (ESI, Orbitrap) m/z 456.8 [(M-F-)+]. HRMS (ESI, Orbitrap)
calcd. for C30H38N3O+ [(M-F-)+]: 456.30094; found: 456.30078.
1-((S)-1-amino-3,3-dimethylbutan-2-yl)-1-benzyl-3-((S)-4-((tert-butyl-
diphenylsilyl)oxy)-1,1-diphenylbutan-2-yl)thiourea (15): Amine 10 (84
mg, 407 µmol, 1.2 equiv.) and NEt3 (23.5 µL, 170 µmol, 0.5 equiv.) were
dissolved in MeCN (HPLC-Grade, 2 mL). The resulting solution was
cooled in an ice/H2O bath and a solution of isothiocyanate 14[22] (177 mg,
339 µmol) in MeCN (HPLC-Grade, 2 mL) was added dropwise. After
stirring the reaction mixture at r.t. for 15 h, it was concentrated in vacuo.
The resulting crude material was purified by flash column
chromatography (Si, 3.015 cm, MeOH/CH2Cl2 = 1/100 → 2/100) to
afford the thiourea 15[25] (150 mg, 206 µmol, 51%) as a white foam. Rf =
0.53 (MeOH/CH2Cl2
= 1/10 [UV, ninhydrin]). Due to bulkiness of
protecting groups, an extreme signal broadening resulting from hindered
rotation was observed in the NMR spectra. 1H NMR (600 MHz, CDCl3) δ
7.68 (d, J = 6.8 Hz, 2H), 7.63 (d, J = 6.8 Hz, 2H), 7.45 – 7.07 (m, 20H),
6.26 (br s, 1H), 5.48 (br s, 1H), 4.30 – 4.16 (br m, 1H), 3.90 – 3.70 (br m,
2H), 3.70 – 3.38 (br m, 2H), 3.33 – 2.70 (br m, 1H), 2.47 – 1.90 (br m,
2H), 1.75 – 1.45 (br m, 1H), 1.09 (s, 9H), 0.82 (s, 9H). 13C NMR (151
MHz, CDCl3) δ 181.9, 142.0, 135.7, 135.6, 133.6, 133.5, 129.79, 129.75,
128.8, 128.7, 128.6, 128.5, 128.4, 127.85, 127.80, 127.4, 126.8, 126.6,
64.8, 61.2, 55.7, 54.4, 53.4, 44.2, 35.2, 34.5, 27.1, 19.3. Four 13C-signals
were not included due to insufficient resolution. [α]D24°C = +0.6 (CHCl3, c
= 1). I 2 2 cm-1 (m), 2860 (m), 1530 (m), 1460 (m), 1360 (w),
1224 (m), 1105 (s), 938 (m), 822 (m), 740 (s), 698 (s). MS (EI, 70 eV)
m/z (%) 464.0 (5), 239.9 (8), 207.0 (45), 176.0 (100), 128.9 (26), 90.0
(97). MS (ESI, Orbitrap) m/z 728.4 (M+H+). HRMS (ESI, Orbitrap) calcd.
for C46H58N3OSSi+ (M+H+): 728.40644; found: 728.40405.
(S)-N-((S)-1-benzyl-5-(tert-butyl)imidazolidin-2-ylidene)-4-chloro-1,1-
diphenylbutan-2-amine hydrochloride (19): To a flask containing a
crude solution of guanidinium salt 18 (354 µmol, quantitative yield of the
previous step assumed) in MeCN (HPLC-grade, 4 mL), thionyl chloride
(151 µL, 2.07 mmol, 6.0 equiv.) was added and the solution was stirred
at r.t. for 15 h. During this time, a white precipitate was formed. After
addition of MeOH (10 mL) all solids dissolved and the resulting mixture
was stirred for 1 h at r.t.. The solvents were removed under reduced
pressure and the crude product was purified by flash column
(S)-N-((S)-1-benzyl-5-(tert-butyl)imidazolidin-2-ylidene)-4-((tert-butyl-
diphenylsilyl)oxy)-1,1-diphenylbutan-2-amine (17): Thiourea 15 (266
mg, 365 µmol) was dissolved in MeCN (dry, 5.1 mL) under Ar. After
addition of 2-chloro-1,3-dimethylimidazolinium chloride (DMC, 16, 93 mg,
548 µmol, 1.5 equiv.; obtained from Sigma Aldrich, cat. No. 529249) and
NEt3 (151 µL, 1.1 mmol, 3.0 equiv.) the solution was stirred at 60-70 °C
for 12.5 h. The resulting white suspension was cooled to r.t., diluted with
Et2O (10 mL) and white solid was filtered off. The filtrate was
concentrated in vacuo and the residue was treated with Et2O (10 mL)
leading to formation of a white suspension again. White solid was filtered
off again. The isolated solids were combined,[34] dissolved in CH2Cl2 and
purified by flash column chromatography (Si, 3.018 cm, MeOH/CH2Cl2 =
1/100 → 2/100) to afford the guanidine 17 (178 mg, 250 µmol, 68%) as
an off-white solid. Rf = 0.55 (MeOH/CH2Cl2 = 1/10 [UV, bromocresol
green]). 1H NMR (600 MHz, CDCl3) δ 7.60 (d, J = 8.1 Hz, 2H), 7.57 (d, J
= 8.1 Hz, 2H), 7.47 – 7.25 (m, 17H), 7.21 (t, J = 7.4 Hz, 2H), 6.54 (d, J =
7.4 Hz, 2H), 4.60 (d, J = 15.9 Hz, 1H), 4.52 (td, J = 10.9, 2.4 Hz, 1H),
4.17 (d, J = 15.9 Hz, 1H), 4.12 (d, J = 11.3 Hz, 1H), 3.78 – 3.70 (m, 2H),
3.40 (dd, J = 10.8, 3.2 Hz, 1H), 3.35 – 3.30 (m, 1H), 3.13 (d, J = 12.7 Hz,
1H), 2.01 – 1.95 (m, 1H), 1.86 – 1.79 (m, 1H), 1.09 (s, 9H), 0.77 (s, 9H).
13C NMR (151 MHz, CDCl3) δ 160.1, 143.3, 142.8, 136.6, 136.5, 135.6,
134.3, 134.16, 131.14, 130.20, 130.16, 129.3, 129.00, 128.97, 128.9,
128.7, 128.4, 128.3, 67.0, 62.1, 59.0, 56.1, 51.4, 44.6, 38.0, 37.0, 27.3,
chromatography [Si, 3.011 cm, CH2Cl2 (100%) → MeOH/CH2Cl2
=
1/100 → 3/97] to afford the guanidinium salt 19 (124 mg, 243 µmol, 69%
over two steps) as an off-white solid. This compound possesses low
solubility in MeOH and MeCN. Rf = 0.58 (MeOH/CH2Cl2 = 1/9 [UV,
1
bromocresol green]). H NMR (600 MHz, CD3OD) δ 7.49 – 7.44 (m, 4H),
7.36 – 7.32 (m, 5H), 7.30 – 7.22 (m, 4H), 6.56 (d, J = 7.3 Hz, 2H), 4.74 –
4.70 (m, 2H), 4.28 (d, J = 16.0 Hz, 1H), 4.10 (d, J = 10.0 Hz, 1H), 3.70 –
3.66 (m, 1H), 3.61 (dd, J = 10.7, 3.4 Hz, 1H), 3.61 – 3.55 (m, 1H), 3.38
(virt t, J = 10.2 Hz, 1H), 3.20 (dd, J = 9.6, 3.2 Hz, 1H), 2.08 – 2.00 (m,
2H), 0.91 (s, 9H). 13C NMR (151 MHz, CD3OD) δ 160.0, 143.3, 142.6,
135.7, 130.2, 129.3, 129.1, 128.9, 128.6, 128.5, 128.3, 67.2, 59.4, 56.6,
51.3, 44.9, 41.9, 37.5, 37.0, 26.0. Two 13C-signals were not included due
20°C
to insufficient resolution. [α]D
= 40.9 (MeOH, c = 0.85). m. p.: 227-
230 °C (MeOH). I 16 cm-1 (w), 2963 (s), 2246 (s), 1645 (s),
1541 (m), 1451 (m), 1367 (m), 1297 (m), 1185 (m), 1077 (m), 942 (m),
748 (s), 701 (s). MS (EI, 70 eV) m/z (%) 270.2 (48), 212.2 (19), 167.0
(100), 152.0 (30), 122 (17), 91.0 (49), 57.2 (42). MS (ESI, Orbitrap) m/z
985.5 [(2M+K+-2HCl)+], 474.3 [(M-Cl-)+]. HRMS (ESI, Orbitrap) calcd. for
C30H37ClN3+ [(M-Cl-)+]: 474.26705; found: 474.26627.
26.0, 19.9. Three 13C-signals were not included due to insufficient
(2S,7S)-7-benzhydryl-1-benzyl-2-(tert-butyl)-1,2,3,5,6,7-hexahydro-
imidazo[1,2-a]pyrimi-dine hydrochloride (4): To a flask containing a
suspension of guanidinium salt 19 (91 mg, 179 µmol) in MeCN/CH2Cl2
(anhydrous, 10 mL + 5 mL), NaH (36 mg, 60% dispersion in paraffin oil,
895 µmol, 5.0 equiv.) was added. The resulting suspension was stirred
under Ar at r.t. for 22 h. The reaction mixture was acidified with HCl (1M
in MeOH) to pH 1-3 and the solvents were removed under reduced
pressure. The resulting crude material was purified by column
chromatography (Si, 3.011 cm, MeOH/CH2Cl2 = 2/100 → 1/9) to afford
the final product 4 (71 mg, 150 µmol, 84%) as a white foam. Rf = 0.43
(MeOH/CH2Cl2 = 1/12 [UV, bromocresol green]). 1H NMR (600 MHz,
CD3OD) δ 7.36 (t, J = 8.2 Hz, 4H), 7.32 – 7.28 (m, 7H), 7.25 – 7.20 (m,
2H), 7.06 (dd, J = 7.4, 1.8 Hz, 2H), 4.60 – 4.46 (m, 2H), 4.48 – 4.43 (m,
1H), 4.03 (d, J = 11.2 Hz, 1H), 3.79 (virt t, J = 9.7 Hz, 1H), 3.72 – 3.66 (m,
21°C
resolution. [α]D
= 48.8 (MeOH, c = 1). m. p.: decomposes above
200 °C (EtOH). I 2 2 cm-1 (m), 2858 (m), 2154 (m), 1650 (s),
1586 (m), 1454 (m), 1105 (s), 935 (w), 822 (m), 743 (s), 700 (s). MS (EI,
70 eV) m/z (%) 526.1 (3), 378.1 (1), 166.9 (50), 90.9 (100), 57.1 (33). MS
(ESI, Orbitrap) m/z 1424.8 (2M+K+), 694.4 (M+H+). HRMS (ESI, Orbitrap)
calcd. for C46H56N3OSi+ (M+H+): 694.41872; found: 694.41541.
(S)-3-(((S)-1-benzyl-5-(tert-butyl)imidazolidin-2-ylidene)amino)-4,4-
diphenylbutan-1-ol hydrofluoride (18): To a flask containing a solution
of protected guanidine 17 (252 mg, 354 µmol) in MeCN/CH2Cl2 (6 mL + 6
mL, HPLC grade), fluoride on polymer (295 mg, 885 µmol, 2.5 equiv.; 3.0
mmol/g loading on Amberlyst® A-26 resin; obtained from Sigma-Aldrich,
cat. No. 387789) was added and the resulting mixture was stirred at r.t.
This article is protected by copyright. All rights reserved

European Journal of Organic Chemistry
10.1002/ejoc.201601154
FULL PAPER
2H), 3.54 – 3.50 (m, 1H), 3.36 – 3.32 (m, 1H), 1.92 – 1.87 (m, 1H), 1.80 –
1.75 (m, 1H), 0.95 (s, 9H). 13C NMR (151 MHz, CD3OD) δ 157.2, 142.4,
141.7, 136.8, 130.2, 130.10, 130.07, 129.3, 129.1, 129.0, 128.30, 128.28,
128.0, 69.1, 56.6, 53.2, 52.5, 51.2, 41.8, 36.4, 25.8, 25.1. [α]D20°C = +14.6
(MeOH, c = 1). I cm-1 (w), 2953 (m), 1643 (s), 1598 (s),
1453 (s), 1348 (s), 1216 (m), 1037 (w), 921 (m), 705 (s). MS (EI, 70 eV)
m/z (%) 438.2 (3), 270.1 (100), 212.1 (20), 166.9 (89), 151.9 (24), 121.9
(21), 90.9 (34), 57.1 (21). MS (ESI, Orbitrap) m/z 438.3 [(M-Cl-)+]. HRMS
237.0 (53), 217.0 (64), 145.9 (43), 57.0 (58). MS (ESI, Orbitrap) m/z
534.3 (M+H+), 334.2 [(M-2Boc+3H+)+]. HRMS (ESI, Orbitrap) calcd. for
C28H48N3O5Si+ (M+H+): 534.33577; found: 534.33539.
tert-Butyl (((tert-butoxycarbonyl)imino)((2S,5R)-2-(((tert-butyldime-
thylsilyl)oxy)methyl)-5-phenylpyrrolidin-1-yl)methyl)(3-((tert-butyl-
diphenylsilyl)oxy)propyl)carbamate (28): Protected guanidine 27 (30.4
g, 56.9 mmol) was stirred in DMF (dry, 600 mL) under Ar for 1 h at r.t. to
achieve complete dissolution of the material. Subsequently, this solution
was cooled in an ice/H2O bath for 30 min. After addition of NaH (5.68 g of
60% dispersion in paraffin oil, 142.2 mmol, 2.5 equiv) the suspension
was cooled for further 5 min. After removing the cooling bath, the mixture
was stirred for 20 min (gas evolution). Afterwards, the mixture was
cooled in an ice/H2O bath for 20 min. A solution of TBDPS protected 3-
bromo-1-propanol[38] (79.7 g, 211.2 mmol, 3.7 equiv.) in DMF (dry, 200
mL) was transferred into the reaction flask with a canula. After 10 min,
the cooling bath was removed and the reaction mixture was stirred at r.t.
for 45 h. The reaction was quenched by careful addition of H2O (500 mL)
and the resulting mixture was poured into a separatory funnel. After
addition of H2O (300 mL) and EtOAc (700 mL) the mixture was extracted.
Organic layer was separated and washed with 5% aqueous LiCl solution
(11 L). Aqueous layer was washed with EtOAc (11 L) and combined
organic extracts were dried over MgSO4 and concentrated in vacuo. The
obtained crude was purified by column chromatography (Si, 8.035 cm,
100% pentane → EtOAc/pentane = 1/50 → 1/19 [elution of the unreacted
protected bromopropanol contaminated with minor impurities] → 100%
CH2Cl2 → MeOH/CH2Cl2 = 1/333 [elution of the product 28] → 1/10
[elution of the unreacted guanidine 27 contaminated with unknown side
products]). The guanidine 28 (25.1 g, 30.3 mmol, 53%) was isolated as a
yellow thick oil. Rf = 0.48 (EtOAc/pentane = 1/10 [UV, bromocresol
green]). Due to bulkiness of protecting groups and side chains, an
extreme signal broadening resulting from hindered rotation was observed
in the NMR spectra. 1H NMR (400 MHz, CDCl3) δ 7.74 – 7.63 (m, 2H),
7.55 (d, J = 7.8 Hz, 2H), 7.46 – 7.32 (m, 6H), 7.30 – 7.12 (m, 5H), 5.33 –
4.63 (br m, 1H), 4.31 – 3.58 (br m, 3H), 3.60 – 3.15 (br m, 2H), 3.03 –
2.57 (br m, 1H), 2.46 – 2.71 (br m, 5H), 1.65 – 1.33 (m, 18H), 1.17 – 0.83
(m, 20H), 0.20 – -0.04 (m, 6H). 13C NMR (101 MHz, CDCl3) δ 135.7,
135.6, 134.9, 133.8, 129.9, 129.8, 129.6, 128.8, 128.1, 127.8, 127.7,
127.5, 126.5, 64.1, 63.7, 63.0, 62.8, 62.1, 28.6, 28.3, 27.0, 26.9, 26.7,
26.2, 25.9, 19.3, 18.5, -5.1. Four 13C-signals were not included due to
insufficient resolution. [α]D21°C = +49.7 (CHCl3, c = 1). I
cm-1 (m), 3393 (w), 2936 (s), 2703 (m), 2309 (s), 2089 (s), 1723 (s), 1594
(s), 1449 (m), 1366 (m), 1244 (m), 1115 (s), 842 (m), 713 (s). MS (EI, 70
eV) m/z (%) 829.8 (3), 616.4 (14), 489.3 (15), 290.1 (11), 146.0 (46),
57.2 (100). MS (ESI, Orbitrap) m/z 830.5 (M+H+), 730.4 [(M-Boc+2H+)+].
(ESI, Orbitrap) calcd. for C30H36N3 [(M-Cl-)+]: 438.29037; found:
438.29025.
+
Benzyl (3-((2S,5R)-2-(((tert-butyldimethylsilyl)oxy)methyl)-5-phenyl-
pyrrolidine-1-car-bothioamido)propyl)carbamate (22): To a solution of
protected aminoalcohol 20[22] (190 mg, 652 µmol, 1.2 equiv.) and NEt3
(38 µL, 271 µmol, 0.5 equiv.) in MeCN (HPLC-grade, 1 mL) was added
isothiocyanate 21[35] (138 mg, 543 µmol) as a solution in MeCN (2 mL).
After stirring the reaction mixture under Ar for 17 h at r.t., the solvent was
removed under reduced pressure and the residue purified by column
chromatography (Si, 3.010 cm, MeOH/CH2Cl2 = 1/100) affording the
thiourea 22 (257 mg, 474 µmol, 73%) as a brown oil. Rf = 0.5
(EtOAc/pentane = 3/7 [UV, ninhydrin]). Due to bulkiness of protecting
groups and side chains, a significant signal broadening resulting from
hindered rotation was observed in the NMR spectra. 1H NMR (600 MHz,
CDCl3) δ 7.36 – 7.21 (m, 10H), 5.29 (br s, 1H), 5.07 (d, J = 12.5 Hz, 1H),
5.05 (d, J = 12.5 Hz, 1H), 4.59 (br s, 1H), 3.91 (br s, 1H), 3.88 (br s, 1H),
3.77 (br s, 1H), 3.49 – 3.43 (m, 1H), 3.06 (br s, 2H), 2.39 (br s, 1H), 2.10
– 2.00 (br m, 2H), 1.84 (br s, 1H), 1.68 – 1.52 (br m, 2H), 1.91 (s, 9H), -
1.12 (s, 6H). 13C NMR (151 MHz, CDCl3) δ 182.6, 156.6, 142.5, 136.8,
128.9, 128.6, 128.1, 127.4, 125.6, 66.5, 66.2, 65.6, 65.1, 42.1, 37.7, 34.9,
29.9, 27.3, 26.0, 18.4, -5.2. One 13C-signal was not included due to
insufficient resolution. [α]D21°C = 43.3 (CHCl3, c = 1). I 2
cm-1 (w), 2941 (s), 2870 (m), 2089 (m), 1708 (s), 1609 (s), 1515 (s), 1354
(s), 1247 (s), 1109 (s), 835 (s), 758 (s), 697 (s). MS (EI, 70 eV) m/z (%)
292.1 (84), 234.1 (12), 146.0 (100), 128.9 (20). MS (ESI, Orbitrap) m/z
564.3 (M+Na+), 542.3 (M+H+). HRMS (ESI, Orbitrap) calcd. for
C29H44N3O3SSi+ (M+H+): 542.28672; found: 542.28424.
tert-Butyl
(((tert-butoxycarbonyl)imino)((2S,5R)-2-(((tert-butyldi-
methylsilyl)oxy)methyl)-5-phenylpyrrolidin-1-yl)methyl)carbamate
(27): To a solution of N,N-di-Boc-thiourea 26[36] (27.4 g, 99 mmol, 1.1
equiv.) and protected aminoalcohol 20 (26.2 g, 90 mmol) in DMF (HPLC-
Grade, 450 mL) NEt3 (37.5 mL, 270 mmol, 3.0 equiv.) was added and
this solution cooled in an ice/H2O bath. Subsequently, HgCl2 (48.9 g, 180
mmol, 2.0 equiv.) was added which was accompanied by a heavy
precipitate formation. After 40 min, the suspension was filtered through a
Büchner-funnel and solids washed with EtOAc (2100 mL). The obtained
filtrate was filtered again and remaining solids washed with another
portion of EtOAc (100 mL). The filtrate was washed with H2O (11 L)
followed by extraction with 5% aqueous LiCl solution (11 L). Extraction
step was accompanied by a severe formation of emulsion and solid
precipitate. Therefore, waiting times of ca. 30 min were needed in order
to achieve a decent layer separation. Organic phase was dried over
MgSO4 and concentrated in vacuo to give the crude product as a yellow
foam. This material was split into 3 parts and each part was purified by
column chromatography (Si, 7.026 cm, MeOH/CH2Cl2 = 1/200 → 5/95)
affording the protected guanidine 27 (42.7 g, 80 mmol, 89%) as a white
foam.[37] Rf = 0.62 (EtOAc/pentane = 1/1 [UV, bromocresol green]). Rf =
0.53 (MeOH/CH2Cl2 = 1/15 [UV, bromocresol green]). Due to bulkiness of
protecting groups and side chains, an extreme signal broadening
resulting from hindered rotation was observed in the NMR spectra. 1H
NMR (400 MHz, CDCl3) δ 9.17 (br s, 1H), 7.33 – 7.16 (m, 5H), 5.56 (br s,
1H), 4.29 (br s, 1H), 3.84 – 3.71 (br m, 1H), 3.69 – 3.45 (br m, 1H), 2.45
– 2.32 (br m, 1H), 2.15 – 2.01 (br m, 1H), 2.00 – 1.85 (br m, 1H), 1.49 (s,
9H), 1.41 (s, 9H), 1.00 (s, 9H), 0.20 (d, J = 7.6 Hz, 6H). 13C NMR (101
MHz, CDCl3) δ 159.5, 150.4, 149.5, 142.7, 128.1, 126.5, 125.6, 81.1,
78.0, 67.8, 61.7, 32.7, 28.2, 27.3, 26.0, 18.5, -5.4. [α]D21°C = 28.2 (CHCl3,
c = 1). I 2 cm-1 (w), 2952 (m), 1758 (m), 1681 (m), 1614
(s), 1507 (m), 1367 (w), 1248 (s), 1137 (s), 1056 (s), 936 (w), 835 (s),
779 (s). MS (EI, 70 eV) m/z (%) 533.2 (100), 333.1 (11), 275.9 (51),
+
HRMS (ESI, Orbitrap) calcd. for C47H72N3O6Si2 (M+H+): 830.49542;
found: 830.49347.
(2S,5R)-2-(chloromethyl)-N-(3-chloropropyl)-5-phenylpyrrolidine-1-
carboximidamide hydrochloride (29): Note: Mixing of the concentrated
HCl with KF produces highly toxic HF in the course of reaction. The
produced waste should therefore be handled with care and disposed
accordingly. To a solution of guanidine 28 (1 g, 1.2 mmol) in MeCN (10
mL) was added aqueous HCl (37%, 10 mL) and the resulting mixture
stirred at r.t. for 1.5 h. After subsequent addition of solid KF (1.4 g, 24
mmol, 20 equiv.), the reaction mixture was stirred at 70 °C for 8.5 h. All
solvents were removed under reduced pressure (70 °C, using high
vacuum and cooling trap to protect the pump from corrosive components
of the reaction). The resulting clotted solid cake was mortared to a fine
powder and further dried at high vacuum (70 °C). The resulting solids
were suspended in a pentane/Et2O mixture = 5/1 (25 mL) to remove non-
polar silyl side products. This suspension was stirred for 15 min, the
organic phase was decanted and the remaining solid material was
washed with pentane/Et2O mixture = 5/1 once again. Solvents were
decanted and the remaning solids were suspended in MeOH (40 mL)
and stirred for 20 min and filtered. Solid material remaining on the filter
paper was washed with an additional portion of MeOH (15 mL).
Methanolic filtrates were combined and concentrated in vacuo. After
removing all solvents under reduced pressure, the crude deprotected
This article is protected by copyright. All rights reserved

European Journal of Organic Chemistry
10.1002/ejoc.201601154
FULL PAPER
monocyclic guanidinium alcohol intermediate (Rf = 0.0, MeOH/CH2Cl2
=
3.02 mmol, 64%). Rf = 0.0 (MeOH/EtOAc = 1/2, [UV, bromocresol green,
CAM]). In order to obtain good quality NMR spectra, we recommend to
use diluted solution of the product in deuterated solvent (ca. 2-6 mg 5 per
600 µL [i.e. 0.012-0.036 M] CD3OD). Otherwise, very strong line
broadening is observed. 1H NMR (400 MHz, CD3OD) δ 7.43 – 7.31 (m,
3H), 7.20 (d, J = 8.0 Hz, 2H), 4.54 – 4.43 (m, 1H), 3.94 (virt t, J = 12.0 Hz,
1H), 3.84 (virt t, J = 12.0 Hz, 1H), 3.49 – 3.35 (m, 2H), 3.17 (dt, J = 10.8,
5.2 Hz, 1H), 2.96 – 2.90 (m, 1H), 2.76 – 2.65 (m, 1H), 2.22 – 2.16 (m,
1H), 2.09 – 1.90 (m, 4H). Signal of the benzylic proton at ca. 4.85 ppm
was not listed due to overlap with H2O signal. 13C NMR (101 MHz,
CD3OD) δ 154.7, 141.2, 130.2, 129.3, 127.9, 63.3, 62.6, 55.2, 44.2, 39.9,
39.5, 28.9, 21.2. [α]D20°C = 45.2 (MeOH, c = 1). I 1 cm-1
(s), 2954 (s), 2315 (w), 2093 (w), 1908 (w), 1654 (s), 1591 (s), 1438 (m),
1370 (m), 1308 (s), 1182 (w), 1024 (w), 756 (m), 699 (s). MS (EI, 70 eV)
m/z (%) 240.1 (100) [(M-Cl-)+], 212.1 (23), 186.0 (12), 137.0 (14), 123.9
(26), 116.9 (18), 103.9 (21), 90.9 (19), 76.9 (18), 55.1 (20). MS (ESI,
Orbitrap) m/z 242.16 [(M-Cl-)+], 124.09. HRMS (ESI, Orbitrap) calcd. for
C15H20N3+ [(M-Cl-)+]: 242.16517; found: 242.16516..
1/10 [UV, bromocresol green]) was obtained as a white sticky solid. This
material was contaminated with inorganic salts (KF and KCl) and was
subjected to the subsequent chlorination step without further purification.
Crude deprotected guanidinium diol (theoretical amount 1.2 mmol) was
suspended in MeCN (HPLC-Grade, 15 mL) treated with SOCl2 (1.5 mL,
20 mmol, 17 equiv.) and the resulting suspension was stirred at 60 °C for
17 h. After addition of MeOH (35 mL) and stirring at r.t. for 20 min, all
volatiles were removed in vacuo to afford a sticky solid. This material was
suspended in MeCN (35 mL), stirred at r.t. for 15 min, filtered off, and
remaining solids washed with an additional portion of MeCN (20 mL).
Combined filtrates were concentrated under reduced pressure to give the
crude product as
a brown sticky mass. It was taken up with
MeOH/CH2Cl2 (ca. 1/6) and applied onto a column for chromatographic
purification. Purification by column chromatography (Si, 3.012 cm,
100% CH2Cl2 → MeOH/CH2Cl2 = 1/10 → 1/1) afforded a sticky brown
mass containing solids (product 29 contaminated with silica and
inorganic salts). It was taken up with MeCN (10 mL), filtered and
remaining solids washed with MeCN (10 mL). The filtrates were
concentrated in vacuo and the resulting syrup taken up with CH2Cl2 (20
mL) leading to further precipitation of solid impurities. This suspension
was stirred for 10 min, filtered and the solids washed with CH2Cl2 (10 mL).
The filtrate was treated with activated charcoal (200 mg), stirred for 5 min,
filtered and washed with CH2Cl2 (15 mL). All solvents were removed
under reduced pressure and guanidinium salt 29 (335 mg, 955 µmol,
80% over two steps) was obtained as a rose oil. Rf = 0.35 (MeOH/CH2Cl2
= 1/9 [UV, bromocresol green]). 1H NMR (400 MHz, CD3OD) δ 7.43 (t, J
= 7.4 Hz, 2H), 7.37 – 7.30 (m, 3H), 5.07 (virt t, J = 8.0 Hz, 1H), 4.44 –
4.38 (m, 1H), 3.93 (dd, J = 11.2, 5.4 Hz, 1H), 3.74 (dd, J = 11.2, 9.0 Hz,
1H), 3.41 – 3.22 (m, 4H), 2.64 – 2.56 (m, 1H), 2.30 – 2.13 (m, 2H), 2.09 –
1.99 (m, 1H), 1.89 (quint, J = 6.5, 2H). 13C NMR (101 MHz, CD3OD) δ
156.4, 141.4, 130.3, 129.3, 126.4, 65.2, 62.9, 45.0, 42.2, 40.7, 35.1, 32.3,
29.3. [α]D20°C = +10.0 (MeOH, c = 1). I 1 1 cm-1 (s), 1646 (s),
1594 (s), 1446 (s), 1357 (m), 1283 (w), 1218 (w), 1082 (m), 1042 (w),
751 (s), 698 (s). MS (EI, 70 eV) m/z (%) 314.0 (4), 242.1 (100), 159.9
(29), 116.9 (36), 103.9 (49), 90.9 (40), 76.9 (36), 55.0 (42). MS (ESI,
Orbitrap) m/z 314.1 [(M-Cl-)+], 278.1 [(M-2Cl-+H+)+], 242.2 [(M-3Cl-+2H+)+].
Acknowledgements
Generous financial support by the Fonds der Chemischen
Industrie (Liebig-Scholarship to P.S.) is gratefully acknowledged.
We thank Prof. Dr. Ulli Englert (Institute of Inorganic Chemistry,
RWTH Aachen University) for X-ray structure determinations.
We are grateful to Dr. Oleg Maltsev, Dr. Dmitry Mazunin and Dr.
Imants Kreituss for helpful discussions. Furthermore, we would
like to express our gratitude to anonymous referees from
previous submission of this article for invaluable remarks.
Special thanks go to Prof. Dr. Dieter Enders, Institute of Organic
Chemistry, RWTH Aachen University, for the generous provision
of laboratory space and chemicals.
HRMS (ESI, Orbitrap) calcd. for C15H22Cl2N3 [(M-Cl-)+]: 314.11853;
+
Keywords: cyclic guanidines • guanylation • N-alkylation •
found: 314.11850.
cyclization • chiral saturated heterocycles
(6aS,9R)-9-phenyl-2,3,4,6,6a,7,8,9-octahydropyrrolo[1',2':3,4]imidazo
[1,2-a]pyrimidine hydrochloride (5): Guanidinium salt 29 (1.65 g, 4.7
mmol) was suspended in MeCN (dry, 150 mL) under Ar. After addition of
NaH (60% dispersion in paraffin oil; 1.13 g, 28.2 mmol, 6.0 equiv.) the
resulting suspension was vigorously stirred at r.t. and reaction progress
was monitored by TLC using MeOH/EtOAc = 1/2 eluent. In order to
obtain reliable results, it is highly recommended to develop TLC plates by
using both CAM and bromocresol green staining reagents. After 2 h the
starting material was fully consumed and two new spots with Rf = 0.35
and Rf = 0.0 were formed. After additional 5 h, the less polar spot
disappeared and only the one with Rf = 0.0 could be detected. Reaction
was acidified with HCl (2 M in MeOH) to pH 1-3 and all volatiles were
removed under reduced pressure. The resulting brown sticky residue
containing many oily droplets was taken up with MeCN (150 mL),
washed with pentane (375 mL) and concentrated under reduced
pressure to give a brown sticky mass. This material was washed with
EtOAc (315 mL) by gentle mixing on rotary evaporator and decanting.
After additional drying on high vacuum, the crude product was taken up
with CH2Cl2 (120 mL) and the resulting heterogeneous mixture was
extracted with H2O two times (100 mL and 50 mL respectively). At this
point, long waiting times (up to 6 h) were needed to obtain sufficient layer
separation and avoid contamination of aqueous phase. Subsequently,
aqueous layers were combined and concentrated under reduced
pressure to give a light brownish residue. When this substance was
dissolved in CH2Cl2 (300 mL), additional fraction of solid impurities
precipitated. This suspension was treated with activated charcoal (2 g),
filtered through a pad of celite and washed with an ample amount of
CH2Cl2 (ca. 200-300 mL). After evaporation of the solvent, the desired
product 5 was obtained as an off-white foam in ca. 90% purity (840 mg,
[1]
a) R. G. S. Berlinck, A. C. B. Burtoloso, A. E. Trindade-Silva, S.
Romminger, R. P. Morais, K. Bandeira, C. M. Mizuno, Nat. Prod. Rep.
2010, 27, 1871; b) R, G. S. Berlinck, A. E. Trindade-Silva, M. F. C.
Santos, Nat. Prod. Rep. 2012, 29, 1382; c) R.Sakai, G. T. Swanson,
Nat. Prod. Rep. 2014, 31, 273.
[2]
[3]
a F. Sączewski, Ł. Balewski, Expert Opin. Ther. Patents 2009, 19,
1 1 ; b F. Sączewski, Ł. Balewski, Expert Opin. Ther. Patents 2013,
23, 965.
Selected reviews on guanidine organocatalysis: a) J. E. Taylor, S. D.
Bull, J. M. J. Williams, Chem. Soc. Rev. 2012, 41, 2109; b) X. Fu, C.-H.
Tan, Chem. Commun. 2011, 47, 8210; c) P. Selig, Synthesis 2013, 45,
703; d) D. Leow, C.-H. Tan, Chem. Asian J. 2009, 4, 488; e) M. Terada,
J. Synth. Org. Chem. Jpn. 2010, 68, 1159; f) T. Ishikawa, Chem. Pharm.
Bull. 2010, 58, 1555.
[4]
[5]
a) M. P. Coles, Dalton Trans. 2006, 985; b) M. P. Coles, Chem.
Commun. 2009, 3659.
Selected reviews on ion recognition a . Blondeau, . Segura, .
rez-Fern ndez, . de endoza, Chem. Soc. Rev. 2007, 36, 198; b) P.
S. Dieng, C. Sirlin, Int. J. Mol. Sci. 2010, 11, 3334.
[6]
For example, the pKa[H2O] of guanidine is 13.6: See a) D. D. Perrin,
Dissociation Constants of Organic Bases in Aqueous Solution,
Butterworths, London, 1965; Supplement, 1972; On the other hand, the
pKa[MeCN] of 1,5,7-triazabicyclo[4.4.0]dec-5-ene (TBD) is 28.0: See b)
I. Kaljurand, A. Kütt, L. Sooväli, T. Rodima, V. Mäemets, I. Leito, I. A.
Koppel, J. Org. Chem. 2005, 70, 1019.
[7]
[8]
B. Maji, D. S. Stephenson, H. Mayr, ChemCatChem 2012, 4, 993.
a) B. Cho, C.-H. Tan, M. W. Wong, Org. Biomol. Chem. 2011, 9, 4550;
b) R. Lee, X. Lim, T. Chen, G. K. Tan, C.-H. Tan, K.-W. Huang,
This article is protected by copyright. All rights reserved

European Journal of Organic Chemistry
10.1002/ejoc.201601154
FULL PAPER
Tetrahedron Lett. 2009, 50, 1560; c) P. W. Dyer, J. Fawcett, M. J.
Hanton, Organometallics 2008, 27, 5082.
as a less polar fraction, the desired monocyclic guanidine 17 could not
be obtained.
[9]
C. A. Seipp, N. J. Williams, V. S. Bryantsev, R. Custelcean, B. A. Moyer,
[26] The structural identity of the species 4 and 5 was assigned based on
the known absolute configurations of the starting chiral amines. Due to
the fact that we did not observe any diastereoisomeric mixtures in the
course of the syntheses, we exclude the possibility of racemization
leading to products possessing relative stereochemistries different from
those presented in the manuscript. Furthermore, racemization of the
corresponding intermediates upon treatment with NaH was excluded
based on comparison with previous publications: Various chiral
guanidines were not reported to undergo racemization in the presence
of such strong bases as KOH and MeONa: See Ref. 24c and M. T.
Allingham, A. Howard-Jones, M. Segura, V. Alcazar, P. Prados, J. de
Mendoza, Tetrahedron 1997, 53, 13119.
RSC Adv. 2015, 5, 107266.
[10] B. Kelly, M. McMullan, C. Muguruza, J. E. Ortega, J. J. Meana, L. F.
Callado, I. Rozas, J. Med. Chem. 2015, 58, 963.
[11] P. Breccia, N. Boggetto, R. Pérez-Fernández, M. Van Gool, M.
Takahashi, L. René, P. Prados, B. Badet, M. Reboud-Ravaux, J. de
Mendoza, J. Med. Chem. 2003, 46, 5196.
[12] V. D. Jadhav, F. P. Schmidtchen, J. Org. Chem. 2008, 73, 1077.
[13] a) E. J. Corey, M. J. Grogan, Org. Lett. 1999, 1, 157; b) D. Leow, C.-H.
Tan, Synlett 2010, 11, 1589.
[14] For additional examples, see: a) T. Misaki, G. Takimoto, T. Sugimura, J.
Am. Chem. Soc. 2010, 132, 6286; b) M. T. Allingham, A. Howard-Jones,
P. J. Murphy, D. A. Thomas, P. W. R. Caulkett, Tetrahedron Lett. 2003,
44, 8677.
[27] For instance, related structures were accessed by hydrogenation of
substituted 2-aminopyridines under acidic conditions. Also, the
syntheses of the corresponding 2-aminopyridines required vigorous
reflux in pentanol for 5 hours, which may put further constraints on the
choice of protecting groups: J. W. Liebeschuetz, R. B. Katz, A. D.
Duriatti, M. L. Arnold, Pestic. Sci. 1997, 50, 258.
[15] See the seminal work by McKay et al.: A. F. McKay, M.-E. Kreling, G. Y
Paris, R. O. Braun, D. J. Whittingham, Can. J. Chem. 1957, 35, 843.
For further examples, see also Refs. 13 and 16. In this context, a
slightly distinct approach by Schmidchen et al consisting of a one-pot
intramolecular alkylation and subsequent desulfurative cyclization
should be mentioned as well: H. Kurzmeier, F. P. Schmidtchen J. Org.
Chem. 1990, 55, 3749.
[28] S. Y. Ko, J. Lerpiniere, A. M. Christofi, Synlett 1995, 815.
[29] Examples can be found in works from several research groups dealing
with the syntheses of alkaloids from crambescidin and banzelladine
family. The described syntheses, however, aimed cyclic guanidines
architectonically different from the salts 3 and 5 described in this
manuscript: a) S. Louwrier, M. Ostendorf, A. Tuynman, H. Hiemstra,
Tetrahedron Lett. 1996, 37, 905; b) K. Nagasawa, A. Georgieva , H.
Koshino , T. Nakata , T. Kita , Y. Hashimoto, Org. Lett. 2002, 4, 177. In
the context of this guanylation reaction, brilliant contributions from
Looper et al. and Nagasawa et al. aiming syntheses of saxitoxin-family
alkaloids should also be mentioned: c) V. R. Bhode, R. E. Looper, J.
Am. Chem. Soc. 2011, 133, 20172; d) O. Iwamoto, R. Shinohara, K.
Nagasawa, Chem. Asian J. 2009, 4, 277.
[16] In this context, a brilliant work of Davis et. al. is an example of how
challenging syntheses of bicyclic guanidines can actually be: a) P. H.
Boyle, A. P. Davis, K. J. Dempsey, G. L. Hosken, Tetrahedron:
Asymmetry 1995, 6, 2819; b) A. P. Davis, K. J. Dempsey, Tetrahedron:
Asymmetry 1995, 6, 2829.
[17] The very first example of this approach towards bicyclic guanidines was
reported by McKay et. al.: a) A. F. McKay, M.-E. Kreling, Can. J. Chem.
1957, 35, 1438. Further noteworthy examples: b) A. Kosayama, T.
Konno, K. Higashi, F. Ishikawa, Chem. Pharm. Bull. 1979, 27, 848; c) F.
Ishikawa, M. Kitagawa, Y. Satoh, J. Saegusa, S. Tanaka, S. Shibamura,
T. Chiba, Chem. Pharm. Bull. 1985, 33, 2838. Furthermore, a similar
protocol was implemented by Schmidtchen (See. Ref. 12) and Ishikawa
(See Ref. 24).
[30] This is a known compound and was reported by our group in the
previous work. Please see Ref. 18.
[31] This is
a known compound and was prepared according to the
[18] In our previous work we prepared this guanidinium salt by a distinct
approach using di(imidazole-1-yl methanimine . uročkin, . Honeker,
W. Raven, P. Selig, J. Org. Chem. 2016, 81, 4516.
published procedure: B. B. Chen, H. Y. Chen, G. J. Gesicki, R. A.
Haack, J. W. Malecha, T. D. Penning, J. G. Rico, T. E. Rogers, P. G.
Ruminski, M. A. Russell, S. S. Yu, Patent US5952381 A1 1999.
[32] This washing step was performed in order to remove non-polar silyl
side products.
[19] K. S. Kim, L. Qian, Tetrahedron Lett. 1993, 34, 7677. Although we used
HgCl2 to promote the guanylation step, CuCl2 could likely be considered
as a less hazardous alternative for scale-up syntheses.
[33] ESI-MS analysis of this material showed no presence of starting
material or any partially protected intermediates.
[20] This step requires special care, as mixing of the concentrated HCl with
KF produces highly toxic HF in the course of the reaction. The
produced waste should be handled with care as well and disposed
accordingly. Alternatively, using HCl in methanol could be considered
as a less hazardous strategy for possible one-pot deprotection.
[21] CCDC 1504458 (9) contains the supplementary crystallographic data
for this compound. This data can be obtained free of charge from The
[34] This precipitation procedure was necessary to separate the cyclic
guanidine
product
17
from
dimethylethyleneurea
and
dimethylethylenethiourea. The latter two compound were formed in
course of the reaction as side products, and could not be separated by
column chromatography. In course of this work, we found out that the
Et2O filtrate contained the cyclic urea and thiourea side products,
whereas isolated solid contained the desired product 17.
Cambridge
Crystallographic
Data
Centre
via
www.ccdc.cam.ac.uk/data_request/cif.
[35] This is a known compound and was prepared according to the known
procedure: M. Kahn, M. Eguchi, S. Moon, J. Chung, Patent
US20040053331 A1 2004.
[22] See Supporting Information for the syntheses of the intermediates 10,
14 and 20.
[23] . ardonville, . oyaa, I. ozas, . lsasua, . I. art n, . .
Borrego, Bioorg. Med. Chem. 2000, 8, 1567.
[36] This is a known compound and was prepared according to the known
procedure: B. Kelly, I. Rozas, Tetrahedron Lett. 2013, 54, 3982.
[24] a) T. Isobe, K. Fukuda, T. Ishikawa, J. Org. Chem. 2000, 65, 7770; b) T.
Isobe, K. Fukuda, T. Tokunaga, H. Seki, K. Yamaguchi, T. Ishikawa, J.
Org. Chem. 2000, 65, 7774; c) T. Isobe, K. Fukuda, K. Yamaguchi, H.
Seki, T. Tokunaga, T. Ishikawa, J. Org. Chem. 2000, 65, 7779.
[25] In the course of this reaction, we expected the formation of two
regioisomeric thioureas with the terminal primary and secondary amino
groups accordingly. Although the obtained spectroscopic data do not
unambiguously prove the structure of the isolated product 15, we
propose the structure with the terminal primary amino group, as this
compound was isolated as a more polar fraction after chromatographic
purification. The species isolated as a less polar fraction, which also
showed amine-specific staining with ninhydrin, could not be sufficiently
characterized due to its low purity. Furthermore, as we attempted the
subsequent Ishikawa’s guanylative cyclization using material isolated
[37] We observed an unidentified impurity with an Rf = 0.23 using
MeOH/CH2Cl2 = 1/15 (UV). Removing this impurity was absolutely
necessary, as its presence fully inhibited the subsequent alkylation step.
Using eluent system consisting of EtOAc/pentane was not sufficient to
remove it (despite its Rf = 0.0 using EtOAc/pentane = 1/1). As we used
the eluent system consisting of MeOH/CH2Cl2, the undesired impurity
could be successfully removed. To our great surprise, in course of this
chromatographic purification, the impurity, despite its lower Rf-value,
was eluted before the guanidine 27!
[38] This is a known compound and was prepared according to the known
procedure: G. McGowan, E. J. Thomas, Org. Biomol. Chem. 2009, 7,
2576.
This article is protected by copyright. All rights reserved

European Journal of Organic Chemistry
10.1002/ejoc.201601154
FULL PAPER
FULL PAPER
Guanidine Synthesis*
Aleksej Turočkin,* William Raven and
Philipp Selig*
Page No. – Page No.
Synthesis of Bicyclic and Tricyclic
Chiral Guanidinium Salts by an
Intramolecular Alkylation Approach
Syntheses of three novel types of chiral guanidinium salts are described. The
structures of interest possess highly rigid bi- and tricyclic scaffolds. They were
assembled by means of two different guanylation methods followed by inter- or
intramolecular alkylation of the guanidine core.
This article is protected by copyright. All rights reserved