Unprecedented property of 4,6-Di(substituted)amino-1,2-dihydro- 1,3,5-triazines: Formation of acid salts by simple treatment with alkali metal salts of protic acids

Article abstract of DOI:10.1246/bcsj.20160329

Treatment of 4,6-di(substituted)amino-1,2-dihydro-1,3,5-triazine derivative (5) with sodium chloride in a bilayer system consisting of toluene and water gave its hydrochloride (6) in the toluene layer and sodium hydroxide in the aqueous layer. Similarly, treatment of 5, 8 and 9 with an aqueous sodium acetate solution in organic solvents resulted in the formation of their corresponding acetates, 7, 12 and 13, respectively.

Full text of DOI:10.1246/bcsj.20160329

Unprecedented Property of 4,6-Di(substituted)amino-1,2-dihydro-
1,3,5-triazines: Formation of Acid Salts by Simple Treatment
with Alkali Metal Salts of Protic Acids
Shirou Maeda,1 Akihisa Maeda,1 Rika Hirose,2 Sayaka Tsuri,2 Takae Takeuchi,2 and Kanji Meguro*1
1Research and Development Division, Hamari Chemicals, Ltd., 1-4-29 Kunijima, Higashiyodogawa-ku, Osaka 533-0024
2Department of Chemistry, Faculty of Science, Nara Women’s University, Kitauoyanishi-machi, Nara 630-8506
E-mail: kanji-meguro@hamari.co.jp
Received: September 29, 2016; Accepted: November 17, 2016; Web Released: January 19, 2017
Kanji Meguro
Kanji Meguro joined Takeda Chemical Industries, Ltd. in 1965 and has been Managing Director of
Pharmaceutical Research Division (1995–1999). He received his Ph.D. degree from Tohoku University in
1974 and joined Imperial College of Science and Technology as a doctoral research fellow (Professor Sir
Derek H. R. Barton) during 1976–1977. He moved to Hamari Chemicals, Ltd. in 1999 as Executive
Managing Director, and currently he is Special Advisor to the company. He received Pharmaceutical Society
of Japan Award for Drug Research and Development in 2002.
Abstract
aqueous NaOH solution to completely rearrange 4 to the de-
sired 5. When the reaction mixture was extracted with toluene
and the extract washed with aqueous NaCl solution, we noted
that the hydrochloride (6) was unexpectedly produced in part in
the toluene layer during this process. The aqueous layer was
made strongly alkaline during the washing procedure. It was
also found that treatment of a solution of 5 in toluene with
aqueous sodium acetate (NaOAc) solution similarly gave rise
to the formation of its acetate (7) (vide infra). Compound 5
has strong basicity (pKa 12.7, Figure 1)5 and hydrolyzed ethyl
acetate to form the acetate 7 by extracting the above reaction
mixture with ethyl acetate in place of toluene. However, the
basicity of 5 alone could not explain the formation of 6 or 7 by
simple treatment of 5 with aqueous NaCl or NaOAc solutions,
respectively. It is generally assumed that a small molecule
organic base, even a so-called super base,6 would not have the
ability to build its acid salts by treatment with inorganic salts
such as NaCl, NaOAc, etc. with simultaneous formation of
NaOH.
Treatment of 4,6-di(substituted)amino-1,2-dihydro-1,3,5-tri-
azine derivative (5) with sodium chloride in a bilayer system
consisting of toluene and water gave its hydrochloride (6) in
the toluene layer and sodium hydroxide in the aqueous layer.
Similarly, treatment of 5, 8 and 9 with an aqueous sodium
acetate solution in organic solvents resulted in the formation of
their corresponding acetates, 7, 12 and 13, respectively.
We have reported that novel 4,6-di(substituted)amino-
1,2-dihydro-1,3,5-triazine derivatives with highly lipophilic
substituents exhibited potent antimicrobial activity against
both Gram-positive and Gram-negative bacteria includ-
ing methicillin-resistant Staphylococcus aureus (MRSA) and
vancomycin-resistant Enterococcus faecalis (VRE).1-3 Com-
pound 51 (Scheme 1) was selected for further development as a
novel antiseptic drug as its water-soluble D-gluconate (HM-
242).4 The mode of antibacterial action of HM-242 was eluci-
dated by Okunishi et al.2 as its bactericidal property induced by
disruption of the bacterial cell wall. During the study of the
chemical and physicochemical properties of this type of anti-
septic drug in terms of structure-activity relationships, we
found that these triazine derivatives formed acid salts upon
simple treatment with alkali metal salts, e.g., sodium chloride
(NaCl), sodium acetate (NaOAc), etc., in bilayer systems
consisting of organic solvents and water.
Results and Discussion
We investigated the above-mentioned unusual phenome-
non observed with 4,6-di(substituted)amino-1,2-dihydro-1,3,5-
triazine derivatives in more detail using 5 and some related
compounds (8, 9) having pKa5 and Clog P values shown in
Figure 1.
Although attempted X-ray crystallographic analyses with 5,
6 and 7 were unsuccessful due to the difficulty of obtaining
their single crystals, we succeeded in the analysis of HCl salt
of 8, namely, 1,2-dihydro-2,2-dimethyl-6-(morphorin-4-yl)-4-
dodecylamino-1,3,5-triazinium chloride (10, Figure 2),4 pre-
The process for synthesizing 5 is shown in Scheme 1.1
When biguanide compound (3) was cyclized with acetone,
triazine derivative (5) was obtained together with minor by-
product (4). The mixture of 4 and 5 was then treated with
178 | Bull. Chem. Soc. Jpn. 2017, 90, 178–187 | doi:10.1246/bcsj.20160329
© 2017 The Chemical Society of Japan

H3C
H3C
H
H
NC
N
N C8H17
H
N
H
H
NH2 .HCl
+
(a), (b)
(e)
N
N
C8H17
NH
.
2HCl
NH NH
3
1
2
H3C
H3C
(c)
H
H
H
N
H
N
N
N C8H17
N
N
C8H17
·HCl
HN
N
CH3
HN
N
H3C
H3C CH3
6
5
(f)
(d)
(e)
H3C
H
H
H
H2N
N
N C8H17
N
N
N
C8H17
H3C
N
N
HN
N
·CH3COOH
H3C CH3
4
H3C CH3
7
Scheme 1. Preparation of diamino-dihydrotriazine derivatives. Reagents and conditions: (a) xylene, reflux; (b) conc. HCl; (c) acetone,
MeOH, conc. HCl, reflux; (d) 5 M NaOH, aq. EtOH, reflux; (e) HCl or aq. NaCl/toluen; (f) AcOH or aq. NaOAc/toluene.
H3C
H3C
O
H
H
H
H
H
N
N
N
C12H25
N
N
N
C8H17
N
N
N CH3
HN
N
HN
N
HN
N
H3C CH3
H3C CH3
H3C CH3
5: pKa = 12.7a)
CLog P = 5.77b)
9: pKa = 11.3a)
CLog P = 2.07b)
8: pKa = 11.5a)
CLog P = 5.96b)
Figure 1. Physicochemical properties of dihydro-1,3,5-triazines 5, 8 and 9. a) pKa values of the conjugate acids are extrapolated
from those obtained in EtOH-H2O. b) CLog P values were obtained from ChemBioDraw Ultra 12.0.
O
Table 1. Bond lengths (¡) of C-N bonds of 10
H
N
N
N
CH3
+
N(1)-C(5)
C(5)-N(2)
N(2)-C(6)
C(6)-N(5)
1.336
1.346
1.336
1.327
C(6)-N(3)
N(3)-C(7)
C(7)-N(4)
N(4)-C(5)
1.347
1.455
1.457
1.349
· Cl-
HN
NH
CH3
H3C
10
R
H
N2
N1
N5
C5
C6
N
N
N
+
N
R1
R2
N4
H
N3
-
· X
N
C7
H
H3C
CH3
11
Figure 3. Resonance structure of diamino-dihydro-1,3,5-
triazine acid salts.
Figure 2. ORTEP drawing of 10 with thermal ellipsoids at
50% probability.
pared similarly. The results indicated that the bond lengths of
all the N-C bonds in the biguanide framework, i.e., N(1)-C(5),
C(5)-N(2), N(2)-C(6), C(6)-N(5), C(6)-N(3) and N(4)-C(5),
were shorter than the usual N-C single bonds and longer
than the usual N=C double bonds (Table 1). Therefore, these
strongly basic diamino-dihydro-1,3,5-triazines easily form their
acid salts with a stable resonance structure (11, Figure 3), as
has been reported for similar compounds.7,8 Hence, structures
of the acid salts in this paper are hereinafter depicted using the
resonance structure 11.
When a solution of 5 in toluene was treated with a large
excess amount (18.4 equivalent) of 30% (w/v) aqueous NaCl
solution at 25 °C (Table 2) for 1 week, the main 1H NMR
signals of 5 were found to be shifted downfield indicating that
something must have happened to 5 under the conditions
employed. Because the downfield shifts did not seem to prog-
ress anymore, we discarded the aqueous layer to exchange with
a fresh 30% (w/v) aqueous NaCl solution. This procedure
1
induced further changes in H NMR signals. It was therefore
thought that a chemical change of 5 occurred in toluene
Bull. Chem. Soc. Jpn. 2017, 90, 178–187 | doi:10.1246/bcsj.20160329
© 2017 The Chemical Society of Japan | 179

1
Table 2. Changes in H NMR and pH during conversion of 5 to 6
H3C
H3C
H
H
H
H
N
N
N
C8H17
N
N
+
N
C8H17
aq. NaCl
toluene
NaOH
+
-
HN
H3C
N
HN
H3C
NH
CH3
·Cl
CH3
5
6
1H NMR (CDCl3 + D2O)
δ (ppm)
Mol. equiv
NaCl
Reaction
period at 25 °C
pH of
aq. layer
CH3CCH3
NHCH2
Benzylic CH2
5
®
1 week
1.29a)
1.30
3.14a)
3.14
4.34a)
4.34
®
9.5
Blank experimentb)
18.4
1 weekc,d)
2 weeksc,d)
3 weeksc,d)
4 weeks
1.37
1.41
1.41
1.40
3.22
3.26
3.26
3.26
4.42
4.45
4.46
4.46
12.6
12.2
12.0
11.6
1
1
a) H NMR of standard 5 is shown for comparison. b) Without NaCl. c) H NMR was measured after an
aliquot of the toluene layer was evaporated to dryness in vacuo. d) Reaction was continued after the aq.
layer was replaced with fresh 30% (w/v) aq. NaCl.
depending on the quantity of NaCl. We eventually came up
with a unique concept that 5 was gradually converted into
the hydrochloride 6 upon contacting with NaCl, as though
hydrolysis of NaCl took place during the treatments. Hence,
the renewal of the aqueous layer was continued once a week
was employed (Entry 1 vs. 2, Table 3). We subsequently treat-
ed 5 at different temperatures in toluene with 13.1 equiv of 30%
(w/v) aqueous NaOAc solution for 4 weeks while changing
the aqueous layer with freshly prepared aqueous NaOAc solu-
tion every week. Although the conversion from 5 to 7 did not
complete even after 4 weeks at 35 °C (Entry 3, Table 3), com-
pletion of the reaction was observed after 2 weeks when the
reaction was conducted at 95 °C (Entry 4, Table 3), indicating
that the conversion reaction was influenced by temperature.
The product of Entry 4 was isolated after 3-week reaction and
unambiguously identified as 7.
1
for another 2 weeks until no more changes in H NMR were
observed (4 weeks in all, Table 2). We also noted that during
this reaction the pH of the aqueous layers increased from 9.5 to
11.6-12.6 over 1- to 4-week observation (Table 2), indicating
the generation of NaOH during the conversion.
Chemical shifts of the CH3-C-CH3 protons of the triazine
ring, the NHCH2 protons in the octylamino group and the
benzylic CH2 protons were gradually shifted downfield as the
above-presumed conversion progressed, and finally converged
with those of the authentic 6 which was prepared from 5 by
treatment with HCl. Changes of these signals were easily
followed because the downfield shifts of these protons were
distinct from others (Table 2). While conversion of 5 to 6
completed after 2- to 3-week treatment as seen in Table 2, the
product was isolated in quantitative yield from the toluene
layer after 4-week treatment and identified unambiguously as
the pure hydrochloride 6 by direct comparison with authentic 6.
In a blank experiment using 5 without adding aqueous NaCl
solution at 25 °C for 1 week, the 1H NMR spectrum of the
organic extract did not change, and the pH of the aqueous layer
remained at 9.5 (Table 2).
In view of the above-described observations (Tables 2, 3),
an equilibrium reaction across the organic and aqueous layers
as shown in Scheme 2 could be postulated. Water-solubility
and distribution coefficient (Log P) between toluene and the
aqueous phase used were measured for 5, 6 and 7 (Table 4)
to find their effects on such reaction. As might have been
expected, the free base 5, which is soluble in toluene, was only
sparingly soluble in water. The salts 6 and 7 which had high
water-solubility were sparingly distributed in toluene. Con-
versely, however, 6 and 7 were almost completely distributed
in toluene when 30% (w/v) aqueous NaCl and NaOAc solu-
tions were employed as the aqueous phases, respectively,
probably due to the contribution of their salting-out effect.
Based on these findings, it seems that a hydrated form of 5 or
a triazinium hydroxide intermediate (A) is initially generated
slightly in the aqueous layer. As soon as A is produced, the
Similarly, treatment of 5 with 30% (w/v) aqueous NaOAc
solution (Table 3) afforded the corresponding acetate 7, which
was isolated (Entry 4) and identified by comparing the
1H NMR with that of authentic 7. Since the acetate contents
¹
¹
OH is quickly ion-exchanged with anion X which is present
in large excess there, to produce 6 or 7 in the aqueous layer as
an ionic mixture (B). Because the toluene removes 6 or 7 from
the aqueous layer with the help of the salting-out effect caused
by the NaCl or NaOAc, the whole equilibrium moves forward.
The transformation thus proceeds efficiently along with the
¹
can be calculated from the integration of CH3COO , time
course of the formation of 7 from 5 was easily followed using
1H NMR (Table 3). Hence, conversion from 5 to 7 was inves-
tigated in more detail.
¹
increase of anion X , particularly by substituting the aqueous
We first carried out reactions at different molar ratios of
aqueous NaOAc solutions (5.7 and 13.1 equiv) to ascertain if
molar ratio is a key factor for this reaction. As expected, the
reaction was accelerated when a higher molar ratio of NaOAc
layer containing the salt with fresh solution occasionally. The
reverse conversion 6 or 7 to 5 cannot be observed during the
equilibrium reaction, because at most only one equivalent of
NaOH is formed in the system. Therefore, it is thought that
180 | Bull. Chem. Soc. Jpn. 2017, 90, 178–187 | doi:10.1246/bcsj.20160329
© 2017 The Chemical Society of Japan

1
Table 3. Monitoring of conversion of 5 to 7 by means of H NMRa)
H3C
H3C
H
H
H
H
N
N
N
C8H17
N
N
+
N
C8H17
aq. NaOAc
tolu
+
NaOH
CH3COO-
.
HN
H3C
N
HN
H3C
NH
CH3
ene
CH3
7
5
Mol.
equiv
(NaOAc/5)
Temp.
(°C)
Reaction
period
Acetate content
Entry
pH of aq. layerc)
(%)b)
1
2
5.7
25
25
24 h
1 week
24 h
23
24
33
32
34
34
33
52
65
76
92
100
99
nd
12.8
nd
nd
nd
13.1
12.9
12.8
12.8
12.5
11.5
11.1
10.9
13.1
13.1
13.1
1 week
2 weeks
3 weeks
1 weekd)
2 weeksd)
3 weeksd)
4 weeks
1 weekd)
2 weeksd)
3 weeks
3
35
95
4
1
a) H NMR was measured after an aliquot of the toluene layer at the respective reaction period was taken
¹
and evaporated to dryness in vacuo. b) Determined by 1H NMR (CH3COO ) of toluene layer. c) nd = Not
determined. d) Reaction was continued further after the aqueous layer was replaced with fresh 30% (w/v)
aq. NaOAc.
H3C
H3C
H
H
H
H
N
N
N C8H17
N
N
+
N
C8H17
-
HN
N
HN
NH
·X
H3C CH3
H3C CH3
6 (X=Cl)
7 (X=CH3COO)
5
organic layer
+ H2O
- H2O
extraction
H3C
H3C
NaX
H
H
H
H
N
N
+
N
C8H17
N
N
+
N C8H17
(large excess)
HN
NH
-
+ Na+
+ X-
+ OH-
HN
NH
-
· X
· OH
NaOH (< 1 equiv.)
H3C CH3
H3C CH3
A
B
aqueous layer
Scheme 2. Formation of the triazinium salts 6 and 7 from 5.
Table 4. Solubility of 5, 6, and 7
Log Pb)
Toluene/aq. NaClc)
Solubility in
Compd.
water (mg/mL, 20 °C)a)
Toluene/H2O
Toluene/aq. NaOAcc)
5
6
7
0.03
1.7
20
nd
¹1.05
¹0.97
nd
>4.0
nd
nd
nd
3.08
a) Determined by the method of Japanese Pharmacopeia 7th Edition. b) Distribution coefficient determined by HPLC
peak area integration: nd = Not determined. c) 30% (w/v) aqueous solution.
difference in solubility among 5, 6, and 7 in water or toluene
(Table 4) itself does not directly correlate to the above-
mentioned conversion reaction. Initial formation of A may
be crucial for this conversion to be initiated. The subsequent
reaction is understood as a kind of anion-exchange reaction
followed by effective extraction of the salts 6 or 7 in aqueous
Bull. Chem. Soc. Jpn. 2017, 90, 178–187 | doi:10.1246/bcsj.20160329
© 2017 The Chemical Society of Japan | 181

1
Table 5. Monitoring by H NMRa) of conversion from compounds 5, 8 and 9 to their acetates in CHCl3
H3C
H3C
O
H
H
H
H
H
N
N
N
C8H17
N
N
N
C12H25
N
N
N CH3
HN
N
HN
N
HN
N
H3C CH3
H3C CH3
H3C CH3
9
5
8
NaOAc/CHCl3
NaOAc/CHCl3
NaOAc/CHCl3
H3C
H3C
O
H
H
H
H
N
N
N
C8H17
N
N
N
C12H25
N
N
N
CH3
+
+
+
· CH3COO-
NH
· CH3COO-
· CH COO-
3
HN
HN
NH
HN
NH
H3C CH3
H3C CH3
H3C CH3
7
13
12
Mol. equiv
(NaOAc/Compd.)
Period
(Week, 25 °C)
Acetate content (%) of
CHCl3 layer
Entry
1
Compd.
5
13.1/5
1b)
2b)
3
87
98
100
63
84
93
97
91
100
100
2
3
8
9
13.9/8
13.1/9
1b)
2b)
3b)
4
1b)
2b)
3
1
a) H NMR was measured after an aliquot of the toluene layer was evaporated to dryness in vacuo. b) Reaction was
continued after the aqueous layer was replaced with fresh 30% (w/v) aq. NaOAc every week.
layer with toluene. Formation of A may be supported by the
fact that the aqueous layer in the blank experiment (Table 1)
was alkaline (pH 9.5).
acetate signal almost disappeared. After continuing the reac-
tion for 1 week (pH of the aqueous layer was 7.6 due to the
formation of NaOAc), the product in the toluene layer was
isolated to identify as 6. The ion-exchange reaction is thought
to proceed also through an ionic mixture such as C shown in
Scheme 3. Subsequent treatment of the hydrochloride 6 with
aqueous NaI solution (7.9 equiv) for 2 weeks at 25 °C, resulted
in a similar ion-exchange reaction to afford the pure 5¢HI.
Similar ion-exchange reaction has been reported by
Shapland et al.9 with an acetic acid salt of a lipophilic amine
they prepared. According to their report the amine acetate was
converted to the hydrochloride by treating with an aqueous
NaCl solution in organic solvents. They, however, failed to
obtain the hydrochloride by direct treatment of the free amine
with NaCl. However, our present paper describes the formation
of hydrochloride by direct treatment of the free amine with
NaCl.
In accordance with the transformation shown in Scheme 2,
similar reactions occurred when 5 was treated with other salts
such as aqueous CsOAc and NaI solutions in toluene at 25 °C
to result in the formation of the acetate 7 and the hydroiodide
(5¢HI), respectively. Pure 5¢HI was isolated in high yield
without any purification procedure, while direct treatment of
5 with aqueous HI solution in toluene afforded 5¢HI with a
slightly decreased purity (92.5%, HPLC) due to formation of
unknown by-products.
The generality of the reaction was examined using other
triazine derivatives 8 and 9 which were converted to their
acetates (12 and 13), respectively (Table 5). These reactions
were carried out in CHCl3 which was found to be a useful
solvent to dissolve the substrates and the salts well, and thereby
time to reach equilibrium was shortened. For comparison, data
with 5 to form 7 in CHCl3 are shown (Table 5, Entry 1). Even
when respective pKa5 or CLog P values of the substrates
(Figure 1) are taken in consideration, ease of reaching equi-
librium does not seem to correlate directly with either strength
of basicity or grade of lipophilicity of the substrates (Table 5,
Entries 1-3).
Conclusion
Treatment of 4,6-diamino-1,2-dihydro-1,3,5-triazine deriva-
tives (5, 8, and 9) with a large excess of alkali metal salts such
as NaCl, NaI, NaOAc, and CsOAc in bilayer systems consist-
ing of organic solvent and water gave rise to the formation
of the corresponding acid salts (hydrochloride, hydroiodide,
acetate) of the triazine derivatives in the organic layer. These
conversion reactions may be attributed to the initial formation
of triazinium hydroxide intermediate A, which is followed by
An anion-exchange reaction of the acetate 7 with aqueous
NaCl solution to form the hydrochloride 6 (Scheme 3) was
examined in detail. The progress of the reaction was monitored
¹
1
¹
¹
focusing on the chemical shifts of CH3COO under H NMR.
When 7 was treated with a large excess amount of aqueous
NaCl solution (7.4 equiv) in toluene at 25 °C for 24 hours, the
quick anion-exchange with the corresponding anion Cl , I ,
¹
or AcO to give the intermediate B. The acid salts are extract-
ed and concentrated into the organic layer by the salting-out
182 | Bull. Chem. Soc. Jpn. 2017, 90, 178–187 | doi:10.1246/bcsj.20160329
© 2017 The Chemical Society of Japan

H C
3
H C
3
H
H
N
H
H
N
N
+
C H
8 17
N
N
+
N C H
8 17
-
.
-
Cl
HN
NH
HN
NH .
CH COO
3
H C CH
3
H C CH
3
3
3
organic layer
6
7
NaCl
(large excess)
extraction
H C
3
H
H
N
N
+
N
C H
8 17
HN
NH
-
+
+ Cl
+ Na
-
H C CH
3
3
+ CH COO
3
C
aqueous layer
Scheme 3. Anion-exchange reaction.
effects of the inorganic salts used. Thus the equilibrium reac-
tions across the bilayer systems are thought to move forward
(Scheme 2).
in this report was prepared from its D-gluconate: To a stirred
suspension of 1,2-dihydro-2,2-dimethyl-6-(4-methylbenzyl)-
amino-4-octylamino-1,3,5-triazine D-gluconate4 (62.0 g, 0.11
mmol) in toluene (300 mL) was added 2 M NaOH (90 mL,
0.18 mmol) and the mixture was stirred at 50 °C for 1 h. The
toluene layer was washed with water four times and evaporat-
ed in vacuo. The residue was recrystallized from 2-butanone,
giving 5 (34.6 g) as colorless crystals, mp 83-84 °C. The HPLC
We conclude that the presence of a large excess amount of
the inorganic salts is important to promote the whole equi-
librium reaction. It is also necessary that the acid salts of the
diamino-dihydrotriazines are distributed preferentially in the
organic layer to be extracted out of B in the aqueous layer.
Intermediate A presumably acts as a critical bridge for the
equilibrium reaction between organic and aqueous phases. The
anion-exchange reactions exemplified by the acetate 7 to the
hydrochloride 6 occurred by similar reasons via intermediate C
(Scheme 3). These types of salt-forming and anion-exchanging
reactions may be possible with using other metal salts.
We have reported1 that appropriate lipophilicities would be
important for the diamino-dihydrotriazines to exert potent anti-
septic effects. The proper lipophilicity is thought to be a key for
these triazine derivatives to interact with bacterial cell wall and
disrupt it to kill bacteria. There may be a relationship between
the distinctive properties described above and the bactericidal
activities of the triazine derivatives.
1
purity was 99.4%. H NMR (300 MHz, CDCl3 + D2O): δ 0.87
(3H, t, J = 6.9 Hz, CH3), 1.24 (10H, m), 1.29 (6H, s, (CH3)2C),
1.35-1.50 (2H, m, NHCH2CH2), 2.31 (3H, s, ArCH3), 3.14
(2H, t, J = 7.2 Hz, NHCH2CH2), 4.34 (2H, s, ArCH2), 7.08
(2H, d, J = 8.1 Hz, ArH), 7.16 (2H, d, J = 8.1 Hz, ArH);
13C NMR (75 MHz, CDCl3): δ 157.7, 157.5, 136.6, 136.2,
129.1, 127.4, 66.7, 44.5, 40.8, 31.9, 30.4, 29.9, 29.4, 29.3,
27.0, 22.7, 21.1, 14.1; HRMS: Exact mass calcd. for C21H36N5
MH+: m/z 358.2971. found: m/z 358.2965. calcd. for
¹
C21H34N5 [M ¹ H] : m/z 356.2814. found: m/z 356.2825;
Anal. Calcd. for C21H35N5: C, 70.54; H, 9.87; N, 19.59. Found:
C, 70.65; H, 9.83; N, 19.66.
1,2-Dihydro-2,2-dimethyl-6-(4-methylbenzyl)amino-4-
octylamino-1,3,5-triazinium Chloride (6): To a solution of 5
(2.0 g, 5.59 mmol) in toluene (50 mL) was added dropwise 4 M
HCl/AcOEt (10 mL) with stirring and ice-cooling. The mixture
was evaporated in vacuo and the residue was recrystallized
from 2-butanone-ether, giving 6 (2.0 g) as colorless hydro-
scopic crystals, mp 137-145 °C. The HPLC purity as 5 was
99.0%. 1H NMR (300 MHz, CDCl3 + D2O): δ 0.87 (3H, t, J =
6.9 Hz, CH3), 1.24 (10H, m), 1.41 (6H, s, (CH3)2C), 1.35-1.50
(2H, m, NHCH2CH2), 2.30 (3H, s, ArCH3), 3.26 (2H, t, J =
6.6 Hz, NHCH2CH2), 4.46 (2H, s, ArCH2), 7.08 (2H, d, J =
8.1 Hz, ArH), 7.15 (2H, d, J = 8.1 Hz, ArH); 13C NMR (75
MHz, CDCl3): δ 156.8, 156.7, 137.0, 135.1, 129.2, 127.4, 64.8,
44.1, 40.8, 31.8, 29.7, 29.2, 29.1, 26.8, 22.7, 21.1, 14.1;
Experimental
All melting points were measured in open capillary tubes
on a METTLER TOLEDO MP70 melting point apparatus and
are uncorrected. NMR spectra were measured on a Bruker
1
spectrometer operating at 300 MHz for H and at 75 MHz for
1
13C or a Varian spectrometer operating at 200 MHz for H and
at 50 MHz for 13C. Electrospray ionization high-resolution
mass spectra (HRMS) were obtained on an Orbitrap XL,
Thermo Fisher Scientific Inc. and are reported as m/z (relative
intensity). Accurate masses are reported for the molecular ion
(M + H, M or M ¹ H) or a suitable fragment ion. Elemental
analyses were carried out on a MICRO CORDER JM10.
Materials. Compounds utilized as starting materials were
prepared as follows.
(12)
¹
HRMS: Exact mass calcd. for
C H35N5(35)Cl [M ¹ H] :
21
m/z 392.2581. found: m/z 392.2580 (the mass and isotope ratio
were proper); Anal. Calcd. for C21H36N5Cl: C, 64.01; H, 9.21;
N, 17.78. Found: C, 63.76; H, 9.29; N, 17.82.
1,2-Dihydro-2,2-dimethyl-6-(4-methylbenzyl)amino-4-
octylamino-1,3,5-triazine (5):1 Compound 5 which is used
Bull. Chem. Soc. Jpn. 2017, 90, 178–187 | doi:10.1246/bcsj.20160329
© 2017 The Chemical Society of Japan | 183

1,2-Dihydro-2,2-dimethyl-6-(4-methylbenzyl)amino-4-
octylamino-1,3,5-triazinium Acetate (7): A mixture of 5
(6.6 g, 18.46 mmol), AcOEt (100 mL) and AcOH (1.2 g, 20.0
mmol) was stirred for 1 h, and evaporated in vacuo. The
residue was recrystallized from 2-butanone, giving 7 (8.5 g) as
colorless crystals, mp 103-104 °C. The HPLC purity as 5 was
98.8%. Compound 7 was also prepared by stirring a mixture of
5, AcOEt and water without adding AcOH at room temperature
137 °C. The HPLC purity as 8 was 99.3%. 1H NMR (300 MHz,
CDCl3 + D2O): δ 0.88 (3H, t, J = 6.9 Hz, CH3), 1.25 (18H, m),
1.45-1.65 (2H, m, NHCH2CH2), 1.60 (6H, s, (CH3)2C), 3.25-
3.32 (2H, m, NHCH2CH2), 3.70-3.80 (4H, m, NCH2CH2O),
3.80-3.90 (4H, m, NCH2CH2O); 13C NMR (50 MHz, CDCl3):
δ 156.6, 155.2, 66.4, 65.9, 45.3, 40.9, 31.9, 29.6, 29.5,
29.3, 29.0, 26.8, 22.6, 14.1; Anal. Calcd. for C21H42N5OCl:
C, 60.62; H, 10.18; N, 16.83. Found: C, 60.71; H, 10.32; N,
16.92.
1
followed by treating as described above. H NMR (300 MHz,
CDCl3 + D2O): δ 0.87 (3H, t, J = 6.9 Hz, CH3), 1.24 (10H, m),
1.32 (6H, s, (CH3)2C), 1.35-1.50 (2H, m, NHCH2CH2), 1.91
(3H, s, CH3COOH), 2.30 (3H, s, ArCH3), 3.24 (2H, t, J = 7.2
Hz, NHCH2CH2), 4.43 (2H, s, ArCH2), 7.07 (2H, d, J = 8.1
Hz, ArH), 7.15 (2H, d, J = 8.1 Hz, ArH); 13C NMR (75 MHz,
CDCl3): δ 179.5, 157.1, 157.0, 136.7, 135.8, 129.1, 127.5,
64.4, 44.0, 40.8, 31.8, 29.9, 29.3, 29.0, 26.9, 24.9, 22.7, 21.1,
14.1; Anal. Calcd. for C23H39N5O2: C, 66.15; H, 9.41; N,
16.77. Found: C, 66.21; H, 9.44; N, 16.75.
1,2-Dihydro-2,2-dimethyl-6-(morphorin-4-yl)-4-dodecyl-
amino-1,3,5-triazinium Acetate (12): To a solution of 8
(4.5 g, 11.86 mmol) in toluene (100 mL) was added acetic acid
(0.85 g, 14.2 mmol) and the mixture was evaporated in vacuo.
The excess acetic acid was removed by azeotropic distillation
(three times) with toluene. The residue was recrystallized from
2-butanone, giving 12 (3.9 g) as colorless crystals, mp 123-
124 °C. The HPLC purity as 8 was 99.4%. 1H NMR (300 MHz,
CDCl3 + D2O): δ 0.88 (3H, t, J = 6.9 Hz, CH3), 1.25 (18H, m),
1.46 (6H, s, (CH3)2C), 1.45-1.60 (2H, m, NHCH2CH2), 1.93
1,2-Dihydro-2,2-dimethyl-6-(morphorin-4-yl)-4-dodecyl-
amino-1,3,5-triazine (8): 1) To a stirred suspension of N-
dodecylcyanoguanidine10 (23.6 g, 93.5 mmol) in a mixture of
morpholin (8.18 g, 93.9 mmol) and xylene (160 mL) was added
concentrated HCl (9.75 g, 117 mmol). The mixture was re-
fluxed for 6 h in a flask equipped with a Dean Stark apparatus.
After cooling, AcOEt was added thereto, and the precipitate
was collected by filtration, washed with AcOEt, and dried
in vacuo to give N1-(morpholin-4-yl)-N5-dodecyl-biguanide
hydrochloride (33.4 g) as colorless crystals, mp 203-208 °C.
¹
(3H, s, CH3COO ), 3.24 (2H, t, J = 7.2 Hz, NHCH2CH2),
3.65-3.80 (8H, m, NCH2CH2O © 2); 13C NMR (50 MHz,
CDCl3): δ 178.3, 156.8, 155.6, 66.5, 65.2, 44.8, 40.7, 31.9,
29.7, 29.6, 29.3, 28.8, 27.0, 25.2, 22.6, 14.1; HRMS: Exact
¹
mass calcd. for C23H44N5O3 [M ¹ H] : m/z 438.3444. found:
m/z 438.3445; Anal. Calcd. for C23H45N5O3: C, 62.83; H,
10.32; N, 15.93. Found: C, 62.52; H, 10.59; N, 15.95.
1,2-Dihydro-2,2-dimethyl-4-methylamino-6-(4-methyl-
benzyl)amino-1,3,5-triazine (9) and Its Acetate (13): 1) 4 M
HCl-AcOEt (125 mL) was added dropwise to a solution of
4-methylbenzylamine (50.0 g, 0.41 mol) in AcOEt (150 mL)
with stirring and ice-cooling. The precipitate was collected by
filtration, washed with AcOEt and dried in vacuo to give 4-
methylbenzylamine hydrochloride (64.1 g). A mixture of the 4-
methylbenzylamine hydrochloride (64.1 g, 0.41 mol) and so-
dium dicyanamide (40.0 g, 0.45 mol), n-butyl alcohol (400 mL)
and water (30 mL) was refluxed for 6 h. The reaction mixture
was evaporated to dryness in vacuo and water was added to
the residue. The precipitate was collected by filtration, washed
with water, and dried in vacuo, to give N-(4-methylbenzyl)-
cyanoguanidine (56.6 g), mp 139-140 °C. The HPLC purity
was 98.7%. 1H NMR (200 MHz, DMSO-d6): δ 2.28 (3H, s,
ArCH3), 4.22 (2H, d, J = 6.0 Hz, ArCH2), 6.6-6.9 (2H, br m,
NH © 2), 7.14 (4H, s, ArH).
1
The HPLC purity as the biguanide base was 97.5%. H NMR
(200 MHz, DMSO-d6): δ 0.86 (3H, t, J = 6.6 Hz, CH3), 1.24-
1.44 (20H, m), 3.05-3.07 (2H, m, HNCH2), 3.43-3.45 (4H, m,
NCH2CH2O © 2), 3.59-3.61 (4H, m, NCH2CH2O © 2), 6.7-
7.6 (5H, br.m, NH and NH+).
2) To a solution of the N1-(morpholin-4-yl)-N5-dodecylbi-
guanide hydrochloride (33.4 g, 88.8 mmol) obtained above in
ethanol (500 mL) were added acetone (134 mL) and concen-
trated H2SO4 (5.68 g, 57.9 mmol). The mixture was refluxed for
24 h and evaporated in vacuo. The residue was dissolved in
ethanol (200 mL) and to the solution were added water (200
mL) and 48% aq. NaOH (43 mL). The mixture was refluxed
for 2 h and concentrated in vacuo. The precipitate was collected
by filtration, washed with water, and dried in vacuo to give
colorless crystals, mp 117-118 °C. Recrystallization from 2-
butanone gave 8 (30.6 g) as colorless crystals, mp 120-121 °C.
2) A suspension of N-(4-methylbenzyl)cyanoguanidine
(27.0 g, 0.14 mol) obtained above and methylamine hydrochlo-
ride (9.66 g, 0.14 mol) in xylene (180 mL) was refluxed with
stirring in a flask equipped with a Dean Stark apparatus for 7 h.
After cooling, AcOEt was added thereto, and the precipitate
was collected by filtration, washed with AcOEt, and dried
in vacuo to give crude N1-(4-methylbenzyl)-N5-methyl-
biguanide hydrochloride (30.2 g). Recrystallization from aq.
acetonitrile gave colorless crystals (22.7 g), mp 116-119 °C.
1
The HPLC purity was 99.5%. H NMR (300 MHz, CDCl3 +
D2O): δ 0.88 (3H, t, J = 6.9 Hz, CH3), 1.25 (18H, m), 1.33 (6H,
s, (CH3)2C), 1.40-1.55 (2H, m, NHCH2CH2), 3.20 (2H, t, J =
7.2 Hz, NHCH2CH2), 3.40-3.55 (4H, m, NCH2CH2O © 2),
3.60-3.75 (4H, m, NCH2CH2O © 2); 13C NMR (75 MHz,
CDCl3): δ 157.7, 156.7, 68.1, 66.9, 44.9, 40.7, 31.9, 30.9,
29.8, 29.66, 29.64, 29.61, 29.56, 29.4, 27.0, 22.7, 14.1; HRMS:
Exact mass calcd. for C21H42N5O MH+: m/z 380.3389. found:
m/z 380.3380.
1
The HPLC purity as the biguanide base was 96.5%. H NMR
1,2-Dihydro-2,2-dimethyl-6-(morphorin-4-yl)-4-dodecyl-
amino-1,3,5-triazinium Chloride (10): To a stirred solution
of 8 (1.7 g, 4.48 mmol) in toluene (30 mL) was added dropwise
4 M HCl-AcOEt (10 mL) under ice-cooling and the mixture
was evaporated in vacuo. The residue was recrystallized from
2-butanone, giving 10 (1.7 g) as colorless crystals, mp 133-
(200 MHz, DMSO-d6): 2.28 (3H, s, ArCH3), 2.66 (3H, d, J =
4.4 Hz, NHCH3), 4.29 (2H, d, J = 5.8 Hz, ArCH2), 7.14 (2H, d,
J = 8.4 Hz, ArH), 7.21 (2H, d, J = 8.4 Hz, ArH), 6.7-7.9 (4H,
br. m, NH and NH+).
3) To a solution of N1-(4-methylbenzyl)-N5-methylbiguanide
hydrochloride (15.0 g, 58.7 mmol) obtained above in ethanol
184 | Bull. Chem. Soc. Jpn. 2017, 90, 178–187 | doi:10.1246/bcsj.20160329
© 2017 The Chemical Society of Japan

1
(225 mL) were added acetone (86 mL) and conc. H2SO4
(3.74 g). The mixture was refluxed for 17 h, and evaporated
in vacuo. The residue was dissolved in ethanol (165 mL) and a
solution of NaOH (40 g) in water (330 mL) was added thereto.
The whole mixture was refluxed for 2 h, concentrated in vacuo,
and extracted with dichloromethane. The dichloromethane
extract was filtered through a column of cellulose powder and
evaporated in vacuo. The residue was subjected to chromatog-
raphy on a silica gel column [CHCl3-MeOH (10:1), CHCl3-
MeOH (4:1), then CHCl3-MeOH-AcOH (4:1:1)] to give crude
9 (15.4 g) with HPLC purity of 96.2%. The crude 9 (15.4 g,
59.4 mmol) was then dissolved in a mixture of methanol
(150 mL) and acetic acid (3.9 g, 67.9 mmol), and the whole
mixture was evaporated in vacuo. The excess acetic acid was
removed azeotropically using toluene (three times), and the
residue was purified by chromatography on a basic silica gel
column [AcOEt, then AcOEt-MeOH (2:1)] to yield the ace-
tate 13. Recrystallization from AcOEt gave colorless crystals
(12.6 g), mp 159-160 °C. The HPLC purity as the free base
was 99.3%. 1H NMR (300 MHz, CDCl3 + D2O): δ 1.34 (6H, s,
12.2. H NMR (200 MHz, CDCl3 + D2O): δ 0.87 (3H, t, J =
6.9 Hz, CH3), 1.25 (10H, m), 1.41 (6H, s, (CH3)2C), 1.3-1.6
(2H, m, NHCH2CH2), 2.30 (3H, s, ArCH3), 3.26 (2H, t, J =
7.0 Hz, NHCH2CH2), 4.45 (2H, s, ArCH2), 7.07 (2H, d, J =
8.4 Hz, ArH), 7.15 (2H, d, J = 8.4 Hz, ArH).
The reaction was continued for another 2 weeks (total
4 weeks) in the same way as above, and pH of the aqueous
layer and H NMR of the toluene layers were measured at the
end of the each week to obtain the following results: After
3 weeks, pH 12.0; H NMR (200 MHz, CDCl3 + D2O): δ 0.87
(3H, t, J = 7.0 Hz, CH3), 1.24 (10H, m), 1.41 (6H, s, (CH3)2C),
1.3-1.6 (2H, m, NHCH2CH2), 2.30 (3H, s, ArCH3), 3.26 (2H, t,
J = 7.0 Hz, NHCH2CH2), 4.46 (2H, s, ArCH2), 7.08 (2H, d,
J = 8.4 Hz, ArH), 7.15 (2H, d, J = 8.4 Hz, ArH). After 4 weeks,
1
1
1
pH 11.6; H NMR (300 MHz, CDCl3 + D2O): δ 0.87 (3H, t,
J = 6.9 Hz, CH3), 1.24 (10H, m), 1.40 (6H, s, (CH3)2C), 1.35-
1.55 (2H, m, NHCH2CH2), 2.30 (3H, s, ArCH3), 3.26 (2H, t,
J = 6.9 Hz, NHCH2CH2), 4.46 (2H, s, ArCH2), 7.08 (2H, d,
J = 8.1 Hz, ArH), 7.15 (2H, d, J = 8.1 Hz, ArH). The toluene
layer of the above final reaction mixture was washed with
water twice and evaporated in vacuo, giving 6 as colorless
hydroscopic solid (3.2 g, 97%), mp 73-98 °C. The HPLC purity
¹
(CH3)2C), 1.91 (3H, s, CH3COO ), 2.30 (3H, s, ArCH3), 2.81
(3H, s, NHCH3), 4.45 (2H, s, ArCH2), 7.08 (2H, d, J = 8.1 Hz,
ArH), 7.15 (2H, d, J = 8.1 Hz, ArH); 13C NMR (50 MHz,
CDCl3): δ 179.5, 157.5, 157.0, 136.8, 135.6, 129.1, 127.5, 64.5,
44.0, 29.2, 27.1, 24.9, 21.0; Anal. Calcd. for C16H25N5O2: C,
60.16; H, 7.89; N, 21.93. Found: C, 59.80; H, 7.95; N, 22.03.
4) A mixture of the 13 (8.65 g, 27.1 mmol) obtained above
and 1 M NaOH (100 mL) was extracted with dichloromethane
three times. The dichloromethane extract was filtered through
a column of cellulose powder and evaporated in vacuo to
give 9 (7.0 g), mp 126-129 °C. The HPLC purity was 99.2%.
1H NMR (300 MHz, CDCl3 + D2O): δ 1.29 (6H, s, (CH3)2C),
2.31 (3H, s, ArCH3), 2.75 (3H, s, NHCH3), 4.33 (2H, s, ArCH2),
7.09 (2H, d, J = 8.1 Hz, ArH), 7.16 (2H, d, J = 8.1 Hz, ArH);
13C NMR (50 MHz, CDCl3): δ 157.4, 156.9, 136.8, 135.2,
129.2, 127.3, 64.8, 44.0, 29.3, 27.2, 21.0; HRMS: Exact mass
calcd. for C14H22N5 MH+: m/z 260.1875. found: m/z 260.1871.
¹
as 5 was 100.0%. HRMS: Exact mass calcd. for 5¢Cl
(12)
¹
(
C H35N5(35)Cl) [M ¹ H] : m/z 392.2581. found: m/z
21
392.2580 (the mass and isotope ratio were proper).
Reaction of 5 with aq. NaOAc in Toluene at 95 °C
(Table 3, Entry 4): A mixture of a solution of 5 (3.0 g, 8.39
mmol) in toluene (30 mL) and 30% (w/v) aq. NaOAc solu-
tion (30 mL, 109.7 mmol as NaOAc) were treated at 95 °C
(refluxed) under argon atmosphere while changing the aq. layer
with freshly prepared 30% aq. NaOAc every week. After
3 weeks, the toluene layer was separated, washed with water
twice and evaporated in vacuo, giving colorless solid (3.0 g),
mp 98-101 °C. The HPLC purity as 5 was 97.8%. The residue
(2.9 g) was recrystallized from 2-butanone to give colorless
crystals (7, 2.5 g), mp 103 °C. Anal. Calcd. for C23H39N5O2: C,
66.15; H, 9.41; N, 16.77. Found: C, 65.95; H, 9.23; N, 16.84.
Changes of pH of the aq. layers, 1H NMR signals and acetate
contents of the toluene layers are shown below. After 1 week,
pH 11.5; H NMR (200 MHz, CDCl3 + D2O): δ 0.87 (3H, t,
J = 6.8 Hz, CH3), 1.23 (10H, m), 1.33 (6H, s, (CH3)2C), 1.3-
1.6 (2H, m, NHCH2CH2), 1.91 (2.77H, s, CH3COO ), 2.30
(3H, s, ArCH3), 3.23 (2H, t, J = 7.0 Hz, NHCH2CH2), 4.42
(2H, s, ArCH2), 7.06 (2H, d, J = 8.0 Hz, ArH), 7.14 (2H,
d, J = 8.0 Hz, ArH). Acetate content: 92.3%. After 2 weeks,
pH 11.1; H NMR (200 MHz, CDCl3 + D2O): δ 0.87 (3H, t,
J = 6.8 Hz, CH3), 1.23 (10H, m), 1.33 (6H, s, (CH3)2C), 1.3-
¹
Calcd. for C14H20N5 [M ¹ H] : m/z 258.1719. Found: m/z
258.1721.
1
Treatment of 5, 8, and 9 with aq. NaCl or aq. NaOAc
(Tables 2, 3, and 5): Representative examples are shown
to illustrate the general procedure. Others are in Supporting
Information.5
¹
Reaction of 5 with aq. NaCl in Toluene (Table 2): To a
solution of 5 (3.0 g, 8.39 mmol) in toluene (30 mL) was added
30% (w/v) aq. NaCl solution (30 mL, 154.0 mmol as NaCl)
and the mixture was stirred at 25 °C under argon atmosphere.
After the reaction mixture was stirred for 1 week, the aqueous
layer was separated and its pH was measured to be 12.6. An
aliquot of the toluene layer was taken and evaporated in vacuo
to measure NMR of the residual substance. 1H NMR (200
MHz, CDCl3 + D2O): δ 0.87 (3H, t, J = 6.6 Hz, CH3), 1.24
(10H, m), 1.37 (6H, s, (CH3)2C), 1.3-1.6 (2H, m, NHCH2CH2),
2.30 (3H, s, ArCH3), 3.22 (2H, t, J = 6.8 Hz, NHCH2CH2),
4.42 (2H, s, ArCH2), 7.07 (2H, d, J = 8.4 Hz, ArH), 7.15 (2H,
d, J = 8.4 Hz, ArH). To the remaining toluene layer was added
freshly prepared 30% (w/v) aq. NaCl solution (30 mL) and
the stirring continued at 25 °C under argon atmosphere. After
another 1 week (total 2 weeks) the pH of the aqueous layer was
1
¹
1.6 (2H, m, NHCH2CH2), 1.91 (3.00H, s, CH3COO ), 2.30
(3H, s, ArCH3), 3.23 (2H, t, J = 7.0 Hz, NHCH2CH2), 4.42
(2H, s, ArCH2), 7.06 (2H, d, J = 8.0 Hz, ArH), 7.14 (2H,
d, J = 8.0 Hz, ArH). Acetate content: 100%. After 3 weeks,
pH 10.9; H NMR (300 MHz, CDCl3 + D2O): δ 0.87 (3H, t,
J = 7.2 Hz, CH3), 1.24 (10H, m), 1.32 (6H, s, (CH3)2C), 1.40-
1
¹
1.55 (2H, m, NHCH2CH2), 1.91 (2.98H, s, CH3COO ), 2.30
(3H, s, ArCH3), 3.24 (2H, t, J = 7.2 Hz, NHCH2CH2), 4.43
(2H, s, ArCH2), 7.07 (2H, d, J = 8.1 Hz, ArH), 7.14 (2H, d,
J = 8.1 Hz, ArH). Acetate content: 99.3%.
Reaction of 8 with aq. NaOAc in CHCl3 (Table 5,
Entry 2): A solution of 8 (3.0 g, 7.90 mmol) in CHCl3 (30
Bull. Chem. Soc. Jpn. 2017, 90, 178–187 | doi:10.1246/bcsj.20160329
© 2017 The Chemical Society of Japan | 185

mL) was treated with 30% (w/v) aq. NaOAc solution (30 mL,
109.7 mmol as NaOAc) at 25 °C under argon atmosphere for
4 weeks by changing the aq. layer with fresh 30% (w/v) aq.
NaOAc solution every week. The CHCl3 layer was separated,
washed with water twice and evaporated in vacuo, giving a
colorless solid (12, 3.0 g), mp 110-119 °C. The HPLC purity as
NHCH3), 4.45 (2H, s, ArCH2), 7.07 (2H, d, J = 7.8 Hz, ArH),
7.15 (2H, d, J = 7.8 Hz, ArH). Acetate content: 99.7%. The
rather low pH of the aqueous layers as above may be the result
of the buffering effect by HCl which formed by partial decom-
position of the CHCl3.
Conversion of 5 to Its Hydroiodide (5¢HI): A mixture
of 5 (3.0 g, 8.39 mmol) in toluene (30 mL) and 30% (w/v) aq.
NaI solution (30 mL, 60.0 mmol as NaI) was allowed to react
with stirring for 4 weeks by changing the aq. layer with fresh
30% (w/v) aq. NaI solution every week at 25 °C under argon
atmosphere. The toluene layer was washed with water twice
and evaporated in vacuo, giving 5¢HI as colorless solid (3.9 g),
mp 73-77 °C. The HPLC purity as 5 was 100.0%. Anal. Calcd.
for C21H36N5I: C, 51.95; H, 7.48; N, 14.43. Found: C, 51.96;
H, 7.54; N, 14.38.
1
8 was 98.7%. Changes of pH of aq. layers, H NMR signals
and acetate contents of the CHCl3 layers are shown below.
1
After 1 week, pH 10.1; H NMR (200 MHz, CDCl3 + D2O):
δ 0.88 (3H, t, J = 6.6 Hz, CH3), 1.25 (18H, m), 1.47 (6H, s,
(CH3)2C), 1.4-1.6 (2H, m, NHCH2CH2), 1.93 (1.90H, s,
¹
CH3COO ), 3.25 (2H, t, J = 7.2 Hz, NHCH2CH2), 3.71 (8H,
m, NCH2CH2O © 2). Acetate content: 63.3%. After 2 weeks,
pH 9.4: 1H NMR (200 MHz, CDCl3 + D2O): δ 0.88 (3H, t, J =
7.0 Hz, CH3), 1.25 (18H, m), 1.47 (6H, s, (CH3)2C), 1.4-1.6
¹
1
(2H, m, NHCH2CH2), 1.93 (2.52H, s, CH3COO ), 3.25 (2H, t,
Changes of pH of the aq. layers and H NMR signals of
J = 7.2 Hz, NHCH2CH2), 3.73 (8H, m, NCH2CH2O © 2).
Acetate content: 84.0%. After 3 weeks, pH 8.5; H NMR (200
the toluene layers are shown below. After 1 week, pH 12.9.
1H NMR (200 MHz, CDCl3 + D2O): δ 0.87 (3H, t, J = 7.0 Hz,
CH3), 1.25 (10H, m), 1.46 (8H, m, (CH3)2C and NHCH2CH2),
2.31 (3H, s, ArCH3), 3.27 (2H, t, J = 7.4 Hz, NHCH2CH2),
4.45 (2H, s, ArCH2), 7.09 (2H, d, J = 8.0 Hz, ArH), 7.16 (2H,
d, J = 8.0 Hz, ArH). After 2 weeks, pH 12.3. 1H NMR (200
MHz, CDCl3 + D2O): δ 0.87 (3H, t, J = 6.6 Hz, CH3), 1.25
(10H, m), 1.50 (8H, m, (CH3)2C and NHCH2CH2), 2.31 (3H, s,
ArCH3), 3.29 (2H, t, J = 7.2 Hz, NHCH2CH2), 4.48 (2H, s,
ArCH2), 7.10 (2H, d, J = 8.2 Hz, ArH), 7.17 (2H, d, J =
8.2 Hz, ArH). After 3 weeks, pH 11.6. 1H NMR (200 MHz,
CDCl3 + D2O): δ 0.87 (3H, t, J = 6.6 Hz, CH3), 1.25 (10H, m),
1.51 (8H, m, (CH3)2C and NHCH2CH2), 2.31 (3H, s, ArCH3),
3.30 (2H, t, J = 7.4 Hz, NHCH2CH2), 4.48 (2H, s, ArCH2),
7.10 (2H, d, J = 8.2 Hz, ArH), 7.17 (2H, d, J = 8.2 Hz, ArH).
1
MHz, CDCl3 + D2O): δ 0.88 (3H, t, J = 7.0 Hz, CH3), 1.25
(18H, m), 1.46 (6H, s, (CH3)2C), 1.4-1.6 (2H, m, NHCH2CH2),
¹
1.93 (2.79H, s, CH3COO ), 3.25 (2H, t, J = 7.0 Hz,
NHCH2CH2), 3.72 (8H, m, NCH2CH2O © 2). Acetate content:
1
93.0%. After 4 weeks, pH 8.7; H NMR (300 MHz, CDCl3 +
D2O): δ 0.88 (3H, t, J = 6.9 Hz, CH3), 1.25 (18H, m), 1.46 (6H,
s, (CH3)2C), 1.45-1.60 (2H, m, NHCH2CH2), 1.93 (2.90H, s,
¹
CH3COO ), 3.25 (2H, t, J = 7.2 Hz, NHCH2CH2), 3.72 (8H,
m, NCH2CH2O © 2). Acetate content: 96.7%. In addition, reac-
tion of 8 with sodium acetate (molar ratio of NaOAc/8 = 13.9)
in CHCl3 at 35 °C for 4 weeks gave a similar result to the above
Entry 2. The product was isolated from the CHCl3 layer of the
4-week reactant and recrystallized from 2-butanone to afford
12. Anal. Calcd. for C23H45N5O3: C, 62.83; H, 10.32; N, 15.93.
Found: C, 62.60; H, 10.38; N, 15.95.
1
After 4 weeks, pH 9.5. H NMR (300 MHz, CDCl3 + D2O): δ
0.87 (3H, t, J = 6.9 Hz, CH3), 1.25 (10H, m), 1.50 (8H, m,
(CH3)2C and NHCH2CH2), 2.32 (3H, s, ArCH3), 3.30 (2H, t,
J = 7.2 Hz, NHCH2CH2), 4.49 (2H, s, ArCH2), 7.10 (2H, d,
J = 8.1 Hz, ArH), 7.17 (2H, d, J = 8.1 Hz, ArH).
Reaction of 9 with aq. NaOAc in CHCl3 (Table 5,
Entry 3):
A stirred solution of 9 (2.17 g, 8.37 mmol) in
CHCl3 (30 mL) was treated with 30% (w/v) aq. NaOAc
solution (30 mL, 109.7 mmol as NaOAc) at 25 °C under argon
atmosphere for 3 weeks while changing the aqueous layer with
freshly prepared 30% (w/v) aq. NaOAc solution every week.
The CHCl3 layer was separated, washed twice with water and
evaporated in vacuo, giving colorless crystals (2.1 g), mp 121-
125 °C. The HPLC purity as 9 was 95.5%. The crude crys-
tals (1.8 g) were recrystallized from AcOEt to give colorless
crystals (13, 1.0 g), mp 160-161 °C. The HPLC purity as 9 was
99.4%. Anal. Calcd. for C16H25N5O2: C, 60.16; H, 7.89; N,
21.93. Found: C, 60.09; H, 7.94; N, 22.03.
Conversion of 7 to 6: To a solution of 7 (3.0 g, 7.18 mmol)
in toluene (30 mL) was added a solution of NaCl (3.1 g, 53.05
mmol) in H2O (27 mL) and the mixture was stirred at 25 °C
under argon atmosphere. After stirring for 24 h, an aliquot of
the toluene layer was taken and then evaporated in vacuo to
measure NMR. 1H NMR (200 MHz, CDCl3 + D2O): δ 0.87
(3H, t, J = 6.8 Hz, CH3), 1.24 (10H, m), 1.41 (6H, s, (CH3)2C),
¹
1.3-1.6 (2H, m, NHCH2CH2), 1.90 (0.15H, s, CH3COO ), 2.30
(3H, s, ArCH3), 3.26 (2H, t, J = 6.8 Hz, NHCH2CH2), 4.46
(2H, s, ArCH2), 7.07 (2H, d, J = 8.2 Hz, ArH), 7.15 (2H, d,
J = 8.2 Hz, ArH). Acetate content: 5.0%. After the reaction
mixture was stirred for additional 1 week, the residual acetate
content in the toluene layer was 4.3% with the pH of the aq.
layer being 7.6. The toluene layer was washed with water twice
and evaporated in vacuo, giving 6 as colorless solid (2.7 g),
mp 111-118 °C. The HPLC purity as 5 was 100.0%. HRMS:
1
Changes of pH of the aq. layers, H NMR signals and the
acetate contents of the CHCl3 layers are shown below. After
1 week, pH 8.9. 1H NMR (200 MHz, CDCl3 + D2O): δ 1.34
¹
(6H, s, (CH3)2C), 1.91 (2.73H, s, CH3COO ), 2.30 (3H, s,
ArCH3), 2.82 (3H, s, NHCH3), 4.45 (2H, s, ArCH2), 7.07 (2H,
d, J = 8.0 Hz, ArH), 7.14 (2H, d, J = 8.0 Hz, ArH). Acetate
content: 91.0%. After 2 weeks, pH 8.6. 1H NMR (200 MHz,
CDCl3 + D2O): δ 1.33 (6H, s, (CH3)2C), 1.91 (3.00H, s,
¹
(12)
¹
Exact mass calcd. for 5¢Cl
(
C H35N5(35)Cl) [M ¹ H] :
21
m/z 392.2581. found: m/z 392.2589 (the mass and isotope
¹
¹
CH3COO ), 2.30 (3H, s, ArCH3), 2.82 (3H, s, NHCH3), 4.45
ratio were proper). Calcd. for 5¢CH3COO (C23H38N5O2)
¹
(2H, s, ArCH2), 7.07 (2H, d, J = 8.0 Hz, ArH), 7.16 (2H, d, J =
8.0 Hz, ArH). Acetate content: 100.0%. After 3 weeks, pH 8.5.
1H NMR (300 MHz, CDCl3 + D2O): δ 1.34 (6H, s, (CH3)2C),
[M ¹ H] : m/z 416.3026. found: m/z 416.3034. calcd. for
5H+ (C21H36N5) MH: m/z 358.2971. Found: m/z 358.2961.
1H NMR (300 MHz, CDCl3 + D2O): δ 0.87 (3H, t, J = 6.9
Hz, CH3), 1.24 (10H, m), 1.41 (6H, s, (CH3)2C), 1.35-1.55
¹
1.91 (2.99H, s, CH3COO ), 2.30 (3H, s, ArCH3), 2.81 (3H, s,
186 | Bull. Chem. Soc. Jpn. 2017, 90, 178–187 | doi:10.1246/bcsj.20160329
© 2017 The Chemical Society of Japan

¹
(2H, m, NHCH2CH2), 1.90 (0.13H, s, CH3COO ), 2.30 (3H, s,
This work was supported in part by JSPS KAKENHI Grant
24619003. The authors thank Professor M. Node, Professor M.
Yamashita and Mr. S. Ogawa of Kyoto Pharmaceutical Univer-
sity for X-ray crystallographic analysis, and Mr. Y. Nishimura
of Hamari Chemicals, LTD., for measurements of pKa. Grati-
tudes are also due to Professor S. Murahashi and Dr. N. Tohnai
of Osaka University and to Professor K. Kitaura of Kobe
University for helpful discussions.
ArCH3), 3.26 (2H, t, J = 7.2 Hz, NHCH2CH2), 4.46 (2H, s,
ArCH2), 7.08 (2H, d, J = 8.1 Hz, ArH), 7.15 (2H, d, J = 8.1
Hz, ArH).
Conversion of 6 to 5¢HI: To a solution of 6 (3.0 g, 7.61
mmol) in toluene (30 mL) was added 30% (w/v) aq. NaI solu-
tion (30 mL, 60.0 mmol as NaI) and the mixture was stirred at
25 °C under argon atmosphere. After the reaction mixture was
stirred for 1 week, an aliquot of the toluene layer was taken and
evaporated in vacuo to measure NMR. 1H NMR (200 MHz,
CDCl3 + D2O): δ 0.87 (3H, t, J = 6.6 Hz, CH3), 1.25 (10H, m),
1.50 (8H, m, (CH3)2C and NHCH2CH2), 2.31 (3H, s, ArCH3),
3.29 (2H, t, J = 7.0 Hz, NHCH2CH2), 4.48 (2H, s, ArCH2),
7.10 (2H, d, J = 8.2 Hz, ArH), 7.17 (2H, d, J = 8.2 Hz, ArH).
After the reaction mixture was stirred for another 1 week (total
2 weeks), the toluene layer was washed with water twice and
evaporated in vacuo, giving 5¢HI as colorless solid (3.7 g), mp
Supporting Information
Additional experimental results, IR spectra of 5, 6 and 7,
1
as well as H NMR spectra are available on http://dx.doi.org/
10.1246/bcsj.20160329.
References
1
597.
2
S. Maeda, T. Kita, K. Meguro, J. Med. Chem. 2009, 52,
1
75-78 °C. The HPLC purity as 5 was 98.7%. H NMR (300
J. Okunishi, Y. Nishihara, S. Maeda, M. Ikeda, J. Antibiot.
MHz, CDCl3 + D2O): δ 0.87 (3H, t, J = 6.9 Hz, CH3), 1.25
(10H, m), 1.50 (8H, m, (CH3)2C and NHCH2CH2), 2.31 (3H, s,
ArCH3), 3.29 (2H, t, J = 6.9 Hz, NHCH2CH2), 4.48 (2H, s,
ArCH2), 7.10 (2H, d, J = 7.8 Hz, ArH), 7.17 (2H, d, J = 7.8
2009, 62, 489.
3
J. Okunishi, H. Nishimura, A. Takada, T. Inada, S. Maeda,
T. Maeda, T. Nishihara, S. Komemushi, Y. Sakagami, Biocontrol
Sci. 2010, 15, 7.
¹
Hz, ArH). HRMS: Exact mass calcd. for 5¢I (C21H35N5I)
4
5
6
S. Maeda, T. Kita, K. Meguro, US20060154928 A1.
For details see Supporting Information for this article.
T. Ishikawa, in Superbases for Organic Synthesis, John
¹
[M ¹ H] : m/z 484.1937. found: m/z 484.1942. The solid
obtained above (3.5 g) was recrystallized from 2-butanone to
give colorless crystals (0.6 g), mp 76-78 °C. Anal. Calcd. for
C21H36N5I: C, 51.95; H, 7.48; N, 14.43. Found: C, 51.73; H,
7.74; N, 14.42.
X-ray Crystallographic Analysis of 105: Crystallographic
data for 10 have been deposited with Cambridge Crystallo-
graphic Data Centre: Deposition number CCDC-1517153.
Copies of the data can be obtained free of charge via http://
www.ccdc.cam.ac.uk/conts/retrieving/html.
Wiley & Sons Ltd., United Kingdom, 2009, pp. 1-7.
7
8
W. J. Welsh, J. Comput. Chem. 1990, 11, 644.
G. D. Nigam, S. Karek, C. R. Saha, S. Srinivasan, Acta
Crystallogr., Sect. C: Cryst. Struct. Commun. 1994, 50, 613.
D. M. Caine, I. L. Paternoster, S. Sedehizadeh, P. D. P.
9
Shapland, Org. Process Res. Dev. 2012, 16, 518.
10 R. P. Welcher, D. W. Kaiser, V. P. Wystrach, J. Am. Chem.
Soc. 1959, 81, 5663.
Bull. Chem. Soc. Jpn. 2017, 90, 178–187 | doi:10.1246/bcsj.20160329
© 2017 The Chemical Society of Japan | 187