Synthesis and examination of amine-cyanocarboxyboranes, the boron analogues of α-cyanocarboxylic acids: X-ray structural study of the first lactam containing a boron atom in the lactam ring

Article abstract of DOI:10.1016/j.jorganchem.2004.06.066

The first representatives of chiral boron atom-containing amine-cyanomethoxycarbonyl boranes (A·BH(CN)COOMe) have been synthesized either from the corresponding amine-bromocyanomethoxycarbonylborane complexes with [Bu₄ or from Me₃N·BH(CN)COOMe and an amine in a base-exchange reaction. Acid hydrolyses of methyl esters generated the free acids (A·BH(CN)COOH), which are isoelectronic to the α-cyano carboxylic acids. Their pK_a values and hydrolysis half-lives in acidic medium (that is rate of proton reduction) have been determined. Similarly to the alpha cyano carboxylic acids, the cyano group attached to the boron (in alpha position to the COOH group) increased the acid strength of carboxy boranes with 2.0-2.5 orders of magnitude. Independently from the type of the amine, pK_a values of the amine-cyanocarboxyboranes (6.34-5.82) decrease consistently with the increase of pK_b values of the amines. Hydrolytic decomposition rate of the alkylamine complexes increases with increasing pK_b values of the amines while the opposite was found for pyridine base complexes. When considering both types of the amines, hydrolysis half-lives of the complexes range over several orders of magnitude from 0.005 to 400 h. Based on these observations protonation of the amine nitrogen atom appears to be the rate determining step in the hydrolysis process. With loss of methanol, 2-NH₂-py·BH(CN)COOMe transformed into a five membered lactam derivative. X-ray diffraction revealed that the pyridine ring is coplanar with the five membered lactam ring. In the crystal two molecules are connected in a head to tail arrangement by strong intermolecular H-bonds between N(2)-H and the carbonyl oxygen (O1) with a donor and acceptor distance of 2.867(3) A?. Three new cyanomethoxycarbonylborates having the composition of K[BH_n (CN)_3-nCOOMe] (n = 1, 2) and K[B(OH)(CN)₂COOMe] have also been synthesized and their properties examined.

Full text of DOI:10.1016/j.jorganchem.2004.06.066

Journal of Organometallic Chemistry 689 (2004) 3567–3581
www.elsevier.com/locate/jorganchem
Synthesis and examination of amine-cyanocarboxyboranes, the
boron analogues of a-cyanocarboxylic acids: X-ray structural
study of the first lactam containing a boron atom in the lactam ring
a
Bela Gyori , Istvan Lazar , Zoltan Berente , Robert Kiraly , Attila Benyei
a,
b
a
c
*
}
´
´
´ ´
´
´
´
´
a
Department of Inorganic and Analytical Chemistry, University of Debrecen, Debrecen H-4010, Hungary
b
´
Institute of Biochemistry, Faculty of Medicine, University of Pecs, Debrecen H-7624, Hungary
c
Department of Chemistry, Laboratory for X-ray Diffraction, University of Debrecen, Debrecen H-4010, Hungary
Received 23 December 2003; accepted 18 June 2004
Available online 18 September 2004
Abstract
The first representatives of chiral boron atom-containing amine-cyanomethoxycarbonyl boranes (AÆBH(CN)COOMe) have been
synthesized either from the corresponding amine-bromocyanomethoxycarbonylborane complexes with [Bu4N]CN or from
Me3NÆBH(CN)COOMe and an amine in a base-exchange reaction. Acid hydrolyses of methyl esters generated the free acids
(AÆBH(CN)COOH), which are isoelectronic to the a-cyano carboxylic acids. Their pKa values and hydrolysis half-lives in acidic
medium (that is rate of proton reduction) have been determined. Similarly to the alpha cyano carboxylic acids, the cyano group
attached to the boron (in alpha position to the COOH group) increased the acid strength of carboxy boranes with 2.0–2.5 orders
of magnitude. Independently from the type of the amine, pKa values of the amine-cyanocarboxyboranes (6.34–5.82) decrease con-
sistently with the increase of pKb values of the amines. Hydrolytic decomposition rate of the alkylamine complexes increases with
increasing pKb values of the amines while the opposite was found for pyridine base complexes. When considering both types of the
amines, hydrolysis half-lives of the complexes range over several orders of magnitude from 0.005 to 400 h. Based on these obser-
vations protonation of the amine nitrogen atom appears to be the rate determining step in the hydrolysis process. With loss of meth-
anol, 2-NH2-pyÆBH(CN)COOMe transformed into a five membered lactam derivative. X-ray diffraction revealed that the pyridine
ring is coplanar with the five membered lactam ring. In the crystal two molecules are connected in a head to tail arrangement by
strong intermolecular H-bonds between N(2)–H and the carbonyl oxygen (O1) with a donor and acceptor distance of 2.867(3)
˚
A. Three new cyanomethoxycarbonylborates having the composition of K[BHn(CN)3ꢀnCOOMe] (n = 1, 2) and K[B(OH)(CN)2-
COOMe] have also been synthesized and their properties examined.
ꢀ 2004 Elsevier B.V. All rights reserved.
Keywords: Amine-carboxyboranes; Borates; Complexes; Reduction; Acidity; Lactam
1. Introduction
ated by the similarities to a-aminoacids, amine-
carboxyboranes and their precursor amine-cyanobora-
nes were tested in biological systems and significant
antitumor [3] and antiinflammantory [4] activities were
found.
Inspired by the initial results, several carboxylic acid
derivatives (AÆBH2COX) of amine-carboxyboranes
were synthesized including ester [5,6], amide [5,7], pep-
tide [8] and hydroxamic acid [9] derivatives. Extensive
Amine-carboxyboranes (AÆBH2COOH), on the basis
of the C+–B and CC–BN isoelectronic relationship, are
considered as the boron analogues of protonated a-ami-
noacids [1] and carboxylic acids [2], respectively. Initi-
*
Corresponding author. Tel.: +36 52 512900; fax: +36 52 489667.
´ ´
E-mail address: lazar@delfin.klte.hu (I. Lazar).
0022-328X/$ - see front matter ꢀ 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2004.06.066

3568
B. Gy}ori et al. / Journal of Organometallic Chemistry 689 (2004) 3567–3581
biological and pharmacological studies revealed remark-
able hypolipidemic [10,11], anticancer [8,10,12], antios-
teoporotic [13] and antiinflammatory [10,11] activities.
In addition, their mode of action has also been studied
[12,14]. Recently these molecules [15] as well as carbor-
anyl salts containing BH2X (X = CN, COOH, COOMe)
groups [16] have also been suggested as possible boron
carriers to tumor cells for boron neutron capture ther-
apy, although molecules with only one boron atom ap-
pear to be less suited for BNCT purposes.
carboxylic acids [28]. Encouraged by the successfull use
of NBS for brominations [2,20], the conversion of
AÆBH2COOMe into the cyano derivative with N-cyano-
succinimide was attempted. Unfortunately, all of the
above mentioned reactions failed and led to, instead of
the recovery of the starting materials, the rupture of
the B–N bond and/or the formation of inseparable
multicomponent reaction mixtures.
Finally, by using [Bu4N]CN as a cyanide source, the
trimethylamine and quinuclidine complex of BH(CN)-
COOMe have been successfully prepared from the cor-
responding BH(Br)COOMe complexes in acetonitrile
solution (Scheme 1). Progress of the reaction was mon-
itored by taking and analyzing consecutive samples with
1H, 11B, quantitative 13C NMRs and IR spectroscopy.
The NMR data indicated that, beside the main product
1d, the isomer Me3NÆBH(NC)COOMe and a small
amount of [Bu4N][BH(CN)2COOMe] have also been
formed, the latter one due to the replacement of the
amine by the cyanide ion. In spite of optimization of
the reaction conditions and performing several separa-
tion steps the isolated product (1d) was still contami-
nated with 2–3% of the starting material, 3–4% of
Me3NÆBH(NC)COOMe and 1–3% of [Bu4N][BH(CN)2-
COOMe]. The reaction that led to the formation of 1a
proceeded similarly but more slowly and the amine–
cyanide exchange was significantly slower. 1a could be
conveniently purified by adding water to the evapora-
tion residue of the reaction mixture as its water solubil-
ity is much lower than that of the side products. It was
also observed that no substitution reaction took place in
CDCl3 as the NMR samples remained unchanged for
more than two weeks. Thus the successful synthesis of
the two amine-cyanocarboxyboranes was likely due to
the presence of the polar aprotic solvent acetonitrile.
Preliminary experiments showed that QÆBBr
(COOMe)2, which was prepared from QÆBH(COOMe)2
[21] with NBS, also reacted with [Bu4N]CN in acetonit-
rile. The reaction was significantly slower than that of
QÆBH(Br)COOMe, probably because of the considera-
bly larger steric hindrance on the boron. After 50% con-
version it gave two products in ca. 1:1 molar ratio. These
products were identified as QÆBCN(COOMe)2 and
[Bu4N][B(CN)2(COOMe)2] by 1H and 11B NMR and
IR spectroscopy.
However, only a limited number of amine-carboxy-
boranescontaining anadditional substituentontheboron
atom [AÆBH(Y)COX] have been synthesized so far
[17]. Lately several new derivatives of amine-cyanobora-
nes and amine-carboxyboranes containing substituents
on the boron atom have been prepared in our labora-
tory, such as amine-bromocyanoboranes (AÆBH(Br)-
CN), amine-dibromocyanoboranes (AÆBBr2CN), salts
of boronium ions having the composition of [AA0B(H)-
CN]+ [18,19], amine-bromocarboxyboranes (AÆBH(Br)-
COOH) and their ester derivatives (AÆBH(Br)COOX,
X = Me, Et, i-Pr) [2,6,20], amine-dibromocarboxybo-
rane methyesters (AÆBBr2COOMe) [20], salts of diamin-
hydrocarboxyboronium cations [AA0B(H)COX]+ (X =
OH, OMe) [19], cyclic diazaborolidinium and diazabo-
rinanium ions [[CnN2B](Y)COX] (n = 2–4; Y = H, Br;
X = OH, OMe) [2], as well as amine-dicarboxyboranes
and their ester and amide derivatives AÆBH(COX)2
(X = OH, OMe, NHEt) [21]. Our efforts were driven
by the desire to find new or improved biological activi-
ties and to understand structure–activity relationships.
After a short preliminary communication [22], we now
report the synthesis and properties of a series of
amine-cyanocarboxyboranes that contain a chiral boron
atom and carry the two biologically active moieties
AÆB(H)CN and AÆB(H)COOX in the same molecule.
2. Results and discussion
2.1. Synthesis of amine-cyano(methoxycarbonyl)boranes
A considerable number of reactions have been at-
tempted for the synthesis of amine-cyanocarboxybora-
nes in our laboratory. One group of reactions was
based on earlier examples for the formation of B–CN
bond. These experiments included a variety of reactions
between AÆBH2COOR or AÆBH(Br)COOR (R = H,
Me) and either Hg(CN)2 [23], Me3SiCN [24], or AgCN
[25]. Considering the ability of trityl cation to abstract
hydride from boron [26], the reaction between AÆBH2-
COOMe and Ph3CCN [27] was also tested. Based on
the analogy between amine-carboxyboranes and the ali-
phatic carboxylic acids, we examined the reaction be-
tween AÆBH(Br)COOMe and acetone cyanohydrine, a
reagent that has been used for the synthesis of a-cyano-
Molecule 1d proved to be an excellent starting mate-
rial for the preparation of several AÆBH(CN)COOMe
complexes in base exchange reactions. These reactions
were performed in refluxing acetonitrile except for 1c,
where Et2NH acted both as reactant and solvent. 1c
was isolated as a colourless oil while the other com-
pounds were white crystalline solids. Unlike 1d, the
majority of the complexes prepared by this reaction were
practically pure. The isocyano isomer probably under-
goes isomerization during the 2–14 h reflux time, while
the majority of the other two impurities are removed

B. Gy}ori et al. / Journal of Organometallic Chemistry 689 (2004) 3567–3581
3569
Scheme 1. Schematic representation of the synthesis of compounds 1a–6. Reagents and conditions. (i) N-bromosuccinimide, methanol, room
temperature, 30 min; (ii) 1.5 mol equiv. [Bu4N]CN, acetonitrile, reflux, 1 h (1a), 1.4 mol equiv. [Bu4N]CN, acetonitrile, room temperature, 30 h (1d);
(iii) from 1d, 2–3 mol equiv. of A, acetonitrile (or Et2NH for 1c), reflux, 2–14 h; (iv) 0.45 mol equiv. HCl in 0.05–0.2 M HCl in water or water–acetone
mixture, 55–100 ꢁC, 0.50–2.5 h; (v) 3 mol equiv. 2-aminopyridine, acetonitrile, reflux 22 h, water, reflux, 1 h; (vi) from 1d, 2 mol equiv. of [Bu4N]CN,
acetonitrile, reflux, 2.5 h, ion exchange; (vii) 1 mol equiv. bromine, 2.8 mol equiv. KHCO3, water, 0 ꢁC – room temperature; (viii) methanol, room
temperature, 3 mol% HBr, 10 min; (ix) 3 mol equiv. LiCN, acetonitrile, reflux, 200 h, ion exchange; (x) 1 mol equiv. 4-CN-pyÆBr2 + 1 mol equiv.
4-CN-py, THF, 0 ꢁC, 2 h. [A or A0 = quinuclidine, Q (a); piperidine, pip (b); Et2NH (c); Me3N (d); 4-dimethylaminopyridine, DMAP (e);
4-aminopyridine, 4-NH2-py (f); morpholine, Morf (g); N-methylmorpholine, N-Me-Morf (h); 1-methylimidazole, Mimid (i); 2-aminopyridine,
2-NH2-py (j); 3-aminopyridine, 3-NH2-py (k); picoline, pic (l); pyridine, py (m)].
in the purification steps. Due to the presence of an asym-
metric centre in the molecule geminal CH2 protons in
1a–c, g, h, 2a, 2,6 and 3,5 carbons in 1b, g, h, 2b, g,
and methylene and methyl carbons in 1c, 2c are in dia-
stereotopic relationship. Thus, the observed splitting of
these signals indicates that the configuration of the chi-
rogenic boron atom is stable on the NMR time scale in
these compounds.
In acidic medium 3 is fairly stable, its decomposi-
tion half-life at 100 ꢁC in 0.5 M D2SO4 is 2.9 h as
determined from time dependent changes observed
in the 1H and 11B NMR spectra of the hydrolysis
mixture. Simultaneously with the protonated amine
and boric acid neither 2j nor other intermediary
products can be observed by NMR. For the hydrol-
ysis half-life of 2j an approximately 16 min value is
expected as calculated from the pKb value (7.07) of
2-NH2-py (see later). If the hydrolysis proceeded
through the splitting of the lactam ring, the amount
of 2j should be at the highest level (approximately
7% or more) in the hydrolysis mixture in the 0.8–
1.4 h range, and it would be well visible by 11B
NMR. As this was not observed, it is concluded that
the dominant hydrolysis pathway is probably the
breaking of the B–N bond rather than splitting of
the C–N bond.
2.2. Lactam formation with 2-NH2-py (3)
The base exchange reaction between 1d and 2-NH2-
py is very slow in acetonitrile. While the reaction with
threefold excess of 3-NH2-py or with twofold excess of
4-NH2-py at reflux temperature takes place quantita-
tively in 6 h, it reaches only 75–78% conversion in 10
h and 92–94% conversion in 19 h with 2-NH2-py. Simul-
taneously with the base exchange reaction metanol splits
slowly off from 1j and molecule 3 is formed (Scheme 1).
The conversion to 3 is still not complete in 40 h in aceto-
nitrile. In water, however, at reflux temperature the reac-
tion is complete in 1 h. It is interesting to note that 3 is
isoelectronic with the 3-substituted indolin-2-ones, a no-
vel class of tyrosine kinase inhibitors [29].
In NaOD/D2O solution the lactam ring is partially
cleaved and an equilibrium exists between 3 and the 2-
NH2-pyÆBH(CN)COOꢀ ion (2jN–H+). For example,
in a 0.046 M solution of lactam 3 (3:NaOD = 1:4) at
100 ꢁC the equilibrium is reached in 50–60 min and
the equilibrium mixture contains 52–55% of 2j–H+. At
room temperature it requires 16–18 weeks to reach the

3570
B. Gy}ori et al. / Journal of Organometallic Chemistry 689 (2004) 3567–3581
decomposition accompanied by H2 formation occurs
to various extents depending on the amines. The prod-
ucts 2 in the majority of cases were found to be consid-
erably more stable, showing hardly any decomposition
(gas evolution) under the hydrolysis conditions, so the
end or the considerable decrease of the effervesence
indicated completion of the ester hydrolysis. Although
most products crystallized upon evaporation of the
reaction mixtures at 0–5 ꢁC, some of them formed
oversaturated solutions in which initiation of the crys-
tallization was rather difficult. All 2 except for 2i were
obtained in pure form as no impurities could be de-
tected by NMR.
Since the preparation of their first representative,
amine-carboxyboranes have been generally considered
to be the boron analogues of the protonated a-amino
acids [1], based on the isoelectronic relationship between
CN+ and BN groups. On the other hand, as we recently
pointed out, experimental data suggests that it is more
appropriate to consider the BN M CC rather than
B M C+ isoelectronic analogy as the guiding rule and
to regard amine-carboxyboranes as boron analogs of
aliphatic carboxylic acids [2]. Potentiometric studies re-
vealed that amine-carboxyboranes are exceptionally
weak acids (pKa = 8.14–8.62, 12 compounds, A = NH3,
RNH2, R2NH) [33]. This fact suggests that the R3N–B
group has a marked electron-donating effect towards
its substituents. In order to explore the electronic effects
of the amines and the cyano group on the COO group
we have determined the pKa values of all 2 by pH
potentiometry.
Fig. 1. ORTEP view of 3 at 50% probability level with numbering
o
˚
scheme. Selected bond length (A), angle and torsion angle ( ) data
include B1–N10 1.564(3), B1–C2 1.569(4), B1–C1 1.616(4), B1–H1
1.08(4), N1–C2 1.142(4), C1–O1 1.222(3), N10–C20 1.349(4), C20–N2
1.372(4), N2–C1 1.384(4); N10–B1–C1 98.3(2), N10–B1–C2 112.5(2),
C20–N10–B1 110.7(2), C60–N10–B1 129.0(2); N1–C2–B1–C1 79.4, N1–
C2–B1–N10 ꢀ169.6, N1–C2–B1–H1 ꢀ46.9.
equilibrium and the relative concentration of 2j–H+ in
the equilibrium mixture is 87–89%.
The structure of 3 as well as selected bond length and
angle data are shown on Fig. 1.
The pyridine ring is coplanar with the five membered
ring (root mean square (rms) deviation from the plane is
˚
0.0240 A, distance of the B1 boron and C1 carbon
atoms from the plane is 0.0497 (0.0022) and 0.0446
˚
(0.0021) A, respectively). In the crystal two molecules
are connected in a head to tail arrangement by strong
intermolecular H-bonds between N(2)–H and the carbo-
nyl oxygen (O1) with a donor and acceptor distance of
It is plausible to assume that the electron density of
the hydrogen atom in the B–H bond (the hydride char-
acter) changes linearly with the electron density of the
COO group. Initiated by the assumption that electron
density of the hydrogen atom in the B–H bond shows
correlation with the rate of acidic hydrolysis
(reduction of protons) of amine-cyanocarboxyboranes
(AÆBH(CN)COOH), semiquantitative data were col-
˚
2.867(3) A. According to the generally accepted graph
set analysis [30] of hydrogen bond patterns developed
by Etter et al. the pattern is R22ð8Þ. The hydrogen bond
pattern is described as Gad ðnÞ where G stands for a des-
ignator letter (C for chains, R for rings, S for intramo-
lecular pattern and D for other finite pattern), d is the
number of H-donors while a is the number of acceptor
atoms and n the total number of atoms in the pattern
for example in the ring.
The latest edition of Cambridge Crystallographic
Database [31] (November, 2002 release) contains no
boron compound comparable to 3. However, the struc-
ture of 3-methylindolin-2-one, which contains the iso-
electronic ring system on the basis of the BN M CC
rule, is known [32], and the two structures show similar
bond length and angle values.
1
lected using H and 11B NMR spectroscopy to deter-
mine the half-lives of the hydrolysis reactions in 0.5 M
1
D2SO4/D2O solutions at 100 ꢁC. H NMR spectra indi-
cated the conversion of amine ligands to ammonium
ions while 11B NMR spectra showed the decomposition
of the complex (characterized by a doublett boron peak
coupled with hydrogen) to boric acid (Table 1).
A comparison of the acidity of the AÆBH2COOHs,
the carboxylic acids and their corresponding a-cyano
derivatives revealed that cyano substitution on both
the boron and the carbon atom results in the same 2–
2.5 units drop in the pKa values. Interestingly, the type
of the amine has only a small impact on the pKa values.
One unit change in the pKb value gives only 0.086 unit
change in the pKa value. A plot of pKas versus the pKbs
of the amines shows a nearly linear correlation for all
compounds except for 2i (Fig. 2).
2.3. Preparation, acidity and hydrolysis of 2
Hydrolysis of all esters 1 takes place in 0.1–0.2 M
aqueous HCl or in aqueous HCl–acetone mixture at
60–100 ꢁC. Simultaneously with ester hydrolysis,

B. Gy}ori et al. / Journal of Organometallic Chemistry 689 (2004) 3567–3581
3571
Table 1
pKa values of amine-cyanocarboxyboranes (2) (25 ꢁC, I = 1.0 M NaClO4) and their hydrolysis half-lives in 0.5 M D2SO4/D2O at 100 ꢁC
Comp.
pKb values of the amine
Hybridization state of donor N-atom
PKa
Half-life/houra
2a
2b
2c
2d
2e
2f
2.85
2.88
2.89
4.02
4.45
4.75
5.26
6.23
6.86
7.84
7.89
8.59
sp3
sp3
sp3
sp3
sp2
sp2
sp3
sp3
sp2
sp2
sp2
sp2
6.34 (0.001)b
6.28 (0.002)
6.26 (0.002)
6.24 (0.002)
6.29 (0.003)
6.23 (0.002)
6.17 (0.002)
6.06 (0.002)
6.26 (0.002)
5.89 (0.002)
5.92 (0.002)
5.82 (0.004)
>400
15
14
55
<0.005
<0.005
5
2g
2h
2i
3.2
0.023
0.75
1.2
2k
2l
2m
a
3.1
Calculated from the integral ratios of the amine peaks of AH+ and 2 in the 1H NMR spectra and from the integral ratios of the boron peaks of
B(OH)3 and 2 in the11B NMR spectra.
b
Standard deviations are in parentheses.
different slopes for all data points with one (i) or two
exceptions (b, c) for sp2 and sp3 amines, respecively
(Fig. 3).
The decomposition (proton reduction) rates of these
complexes can not be explained with the hypothesis that
the hydrolysis rates are basically determined by the hy-
dride character of the B–H hydrogen. Since pyridine
base complexes do not show higher than usual electron
densities on the carboxylate groups, it is quite unlikely
that high electron densities could appear on the B–H
hydrogens. Consequently, several orders of magnitude
faster hydrolysis rates of strong pyridine base complexes
can only be explained by an initial attack of protons on
the trigonal planar sp2 pyridine nitrogen atoms which
are sterically freely available in contrast to strong sp3
donor nitrogens, which are sterically fairly crowded
and may have a rigid structure. On the other hand,
Fig. 2. Plot of pKas of AÆBH(CN)COOHs (2) versus pKbs of amines
(j and d indicate amines with sp3 and sp2 hybridization state,
respectively.) Kb = [BH+][OHꢀ]/[B].
It clearly indicates that, in accordance with our
expectations, weaker amines donate less electrons to
the borane moiety and thus to the carboxylic group,
and therefore they form stronger acids.
We have found it quite unexpected that soft pyridine
bases with sp2 nitrogen donor atoms did not result in an
extra increase in the electron density on the soft cyan-
oborane moiety and did not result in an extra drop of
acidity of molecules 2e, f and 2i–m.
In contrast to acid stengths, hydrolysis rates in 0.5 M
D2SO4 showed exceptionally strong dependance on the
quality of the amine part. Plots of logarithms of hydrol-
ysis half-lives of complexes with either sp2 (j) or sp3 (d)
nitrogen donor atoms versus pKb values of the corre-
sponding amines resulted in two straight lines with
Fig. 3. Plots of logarithms of half-lives of amine-cyanocarboxybora-
nes versus pKb values of the amines (j and d indicate amines with sp3
and sp2 hybridization state, respectively).

3572
B. Gy}ori et al. / Journal of Organometallic Chemistry 689 (2004) 3567–3581
decrease in the basicity of the amine part results in a de-
crease in the virtual concentration of the transition state
[AHÆBH(CN)COOH]+ ions resulting in an increase in
the hydrolytic stabilities of complexes.The decomposi-
tion of [AHÆBH(CN)COOH]+ ions may occur by either
the cleavage of the B–N bond or directly by the elimina-
tion of an H2 molecule. Decomposition mechanism of
alkylamine complexes could not have been explained
so far mainly due to the very low number of amines
belonging to the same group of bicyclic, tertiary or sec-
ondaryalkyl amines.
In a recent publication Funke and Mayr [34] found
pyridine base complexes significantly stronger hydride
donors than trialkylamine complexes in reduction of
carbenium ions by amine-boranes following a polar
mechanism. In the light of our results the reduction of
carbenium ions more likely goes through a transition
state in which nitrogen atom of the pyridine base com-
plexes are involved rather than a direct attack by the
B–H bond.
Simple aliphatic carboxylic acids have pKa values
around 3–5. Strongly electron withdrawing subtituents
such as F, Cl or NO2 increase the acidity (for example,
the pKa of nitroacetic acid is 1.68, trifluoroacetic acid
0.5) while electron donating groups have the opposite ef-
fect. The pKa of amine-cyanocarboxyboranes (around 6)
at least partially fills the gap that exists between the pKa
values of monocarboxylic acids and amine-carboxybor-
anes (pKa = 8.1–8.6). We belive that the synthesis and
investigation of amine-cyanocarboxyboranes may be
the first step towards the systematic design, synthesis
and characterization of monocarboxylic compounds
whose pKas would cover the entire 0.5–8.6 pKa range.
in acetonitrile at reflux temperature, formation of an
intermediate was detected. When monitoring the raction
mixture by 11B NMR the disappearence of the bromobo-
rane doublet (d = ꢀ6.1 ppm) and simultaneous appea-
rence of a very broad peak of the intermediate
(d = ꢀ9.45 ppm, m1/2 ꢁ330 Hz) were observed. Peak inten-
sity of the intermediate then started to decrease and simul-
taneously QÆBH(CN)COOMe was formed. When more
than 90% of the total boron content was observed under
the peak of ꢀ9.45 ppm the reaction mixture was evapo-
rated and the residue was treated in methylene chloride
with H2S and then Ag2S was filtered off. Evaporation of
the filtrate resulted in a white solid whose main compo-
nent was QÆBH(NC)COOMe as indicated by its charac-
teristic IR absorbance (2131 cmꢀ1) and its quantative
13C NMR spectrum (sharp peak at 170.37 ppm and broad
peak at 185.12 ppm). Unfortunately all of our efforts to
isolate this compound in pure form failed. The intermedi-
ate was identified as QÆBH(NCAgCN)COOMe based on
our previous observations and literature data published
for Me3NÆBH2(NCAgCN) [35].
2.5. Cyano(methoxycarbonyl)borates
As we have already mentioned (Section 2.1) in the
reaction of Me3NÆBH(Br)COOMe with [Bu4N]CN in
acetonitrile at room temperature, in parallel to the bro-
mide to cyanide exchange, the exchange of the amine by
the cyanide also occurs leading to the formation of
[BH(CN)2COOMe]ꢀ ion. This type of exchange has al-
ready been published in the literature [38]. At reflux tem-
perature in the presence of a large excess of [Bu4N]CN
the reaction takes place quantitatively. After cation ex-
change the pure product K[BH(CN)2COOMe] (4) has
been isolated in high yield.
Salt 4 failed to produce the desired AÆB(CN)2-
COOMe complexes when treated with bromine in the
presence of an amine (Q, 4-CN-py, 4-Me2N-py) in
acetonitrile. In aqueous solution however, in the pres-
ence of KHCO3 the following reaction takes place
(Scheme 2) with both chlorine and bromine, that may
be considered an unexpected reaction in the meaning
that it yielded the hydroxyl group-containing compound
5, which is a unique representative of stable anionic
hydroxoborates:
2.4. Examinations of the synthesis of amine-isocy-
ano(methoxycarbonyl)borane
To the best of our knowledge only the trimethylamine
[35–37], pyridine and picoline [37] complexes of
BH2(NC) have been prepared so far. The synthesis of
the trimethylamine complex from trimethylamine-iod-
oborane has been achieved by several reagents including
silver cyanide [35], or [Bu4N][Ag(CN)2], which was pre-
pared from AgCN and [Bu4N]Br in methylenechloride
solution [36], or the cyanocomplexes [Bu4N][M(CN)4]
(M = Ni, Zn). The latter ones were used for the synthesis
of pyridine and picoline complexes as well.
Chemical composition of compound 5 has been con-
1
firmed by boron elemental analysis, H NMR (presence
In our hands, in reactions with AÆBH(Br)COOMe,
AgCN has failed to achieve B–Br to B–CN or B–NC sub-
stitution (see Section 2.1). Therefore, we decided to test
[Bu4N][Ag(CN)2] for this purpose. This reagent first was
prepared in pure solid form in our laboratory (see Section
3.7.4). In the reaction mixture of Me3NÆBH(Br)COOMe
and [Bu4N][Ag(CN)2] only Me3N BH(CN)COOMe
could be detected. When the reaction was run with the
corresponding quinuclidine complex QÆBH(Br)COOMe
of a OMe group), 11B NMR (a singlet peak only), quan-
titative 13C{1H} NMR(OMe:CN:COOMe = 1:2:1) and
IR spectroscopy (characteristic absorption of non-associ-
ated OH). Compound 5 decomposes in water in one day.
Reaction of Me3NÆBH(Br)COOMe with LiCN in
acetonitrile was examined at reflux temperature and the
1
process was monitored by H and 11B NMR spectros-
copy. The rate of Br–CN exchange reaction was approx-
imately two orders of magnitude lower than with

B. Gy}ori et al. / Journal of Organometallic Chemistry 689 (2004) 3567–3581
3573
Scheme 2.
[Bu4N]CN, while the rate of amine to cyanide exchange
decreased less significantly as compared to [Bu4N]CN.
Thus, after a 120 h of reaction time virtually the total
boron content was transformed into Li[BH(CN)2-
COOMe]. None of the monitoring samples taken over
the course of the reaction indicated the presence of any
Me3NÆBH(NC)COOMe in the reaction mixture. Re-
cently a succesful Br–CN exchange reaction has been re-
ported in the literature for the reaction of
Cy3PÆBH(Br)COOMe (Cy = cyclohexyl) and LiCN
without the replacement of the phosphine base with cya-
nide ions [39].
Me3NÆBH2COOMe reacted reluctantly with LiCN in
refluxing acetonitrile in an amine–CNꢀ ion exchange
reaction. The product has been isolated, after a cation
exchange procedure, in the form of its potassium salt
(6). Molecule 6 was synthesized in order to develop a
new reaction in which 6 and amine perbromides would
react in the presence of amines to produce molecules 1
since 1d could not be transformed with very weakly ba-
sic amines into their BH(CN)COOMe complexes. We
have attempted to synthesize 4-CN-pyÆBH(CN)-
COOMe in this way and found that, in addition to the
oxidation of the B–H hydrogen, the oxidation of the
carboxymethyl group had occured simultaneously and
the reaction product was contaminated with approxi-
mately 40% of 4-CN-pyÆBH2(CN). All of our efforts
to separate the individual components of the reaction
product have failed so far.
from Mg(OCH3)2. Pentane was fractionally distilled.
Water denotes twice distilled water. Methanolic HBr
solution was prepared by dropwise addition of concen-
trated aqueous HBr solution to P2O5 under vigorous stir-
ring and dissolving the liberated HBr gas, after drying
with Granusic A (J. T. Baker), in methanol. Molecular
sieves were activated by keeping under dynamic vacuum
for 3 h at 220 ꢁC, and stored under dry N2. Quinuclidine
was recrystallized from ether. 4-Aminopyridine was ex-
tracted with ether. Diethylamine, pyridine, 1-methylim-
idazole and picoline were distilled from KOH.
Morpholine and N-methylmorpholine was distilled from
KOH and then from Na. Piperidine was distilled from
CaH2. 3-Aminopyridine and 2-aminopyridine were
recrystallized from ether and ether-pentane mixture,
respectively. NBS was recrystallized from water and
dried in a N2 stream before use. Bromine (Ferak),
[Bu4N]I (Fluka) and 4-DMAP (Janssen) were used as
received. Me3NÆBH2COOH [3], QÆBH2COOH [41],
[Bu4N]CN [42] and LiCNÆnTHF [43] were prepared by
known procedures. 4-Cyanopyridine and 4-dimethyl-
aminopyridine perbromides were prepared by the proce-
dure published for pyridine perbromide [44]. Syntheses
of Me3NÆBH(Br)COOMe, QÆBH(Br)COOMe, LiCNÆ
MeCN and [Bu4N][Ag(CN)2] are given in Section 3.6.
NMR spectra were recorded on a Bruker AM 360
instrument in 5 mm o.d. tubes at room temperature.
1H (360.1 MHz) spectra were referred to internal DSS
in D2O and internal TMS in CDCl3, acetone-d6 and
DMSO-d6. Protons adjacent to boron generally gave
1
distinguishable but broad signals in the H NMR spec-
3. Experimental
tra and their chemical shifts are omitted. 13C (90.5 MHz)
spectra were referred to solvent signals (CDCl3: 77.0
ppm, acetone-d6: 29.9 ppm, DMSO-d6: 39.5 ppm) and
DSS in D2O as external reference. Ambiguities in assign-
3.1. General
1
All reactions, except those involving water or noted
otherwise, were performed under an oxygen and water
free N2 atmosphere using the general Schlenk techniques
in flamed or oven dried glassware. Extractions of solids
were carried out according to the standard Schlenk period
extraction techniques [40]. All solid products were dried
in a dry nitrogen stream to constant weight. Absolu-
tized solvents were freshly distilled prior to use. Acetone
was distilled from a 1.5 m Raschig packed column.
Acetonitrile was distilled from P2O5 after drying with
CaH2. Dichloromethane was distilled from CaH2, then
refluxed with NaBH4/diglyme and fractionally distilled.
Chloroform was shaken with cc. H2SO4, dried over
CaCl2 and distilled from P2O5. Ether and THF were dis-
tilled from Na-benzophenone. Methanol was distilled
ing H and 13C signals were cleared with homonuclear
decoupling and chemical shift correlation (1H–1H and
13C–1H) experiments. Carbons attached directly to
boron could be observed in a few complexes only, because
of the line-broadening due to the nuclear spin and quad-
rupolar effect of boron nuclei. 11B (115.5 MHz) spectra
were referred to Et2OÆBF3 in a capillary inserted into
the tube. In cases when multiplicities could only be re-
vealed by mathematical resolution enhancement, multi-
plets are marked ‘‘broad’’ and coupling constants are
not given. IR spectra were recorded on a Perkin–Elmer
Paragon PC 1000 FT-IR spectrometer. X-ray structure
analysis of 3 was performed on a colourless block crystal
(0.45 · 0.35 · 0.3 mm) prepared by slow evaporation of
an acetonitrile solution. Molecular formula C7H6BN3O,

3574
B. Gy}ori et al. / Journal of Organometallic Chemistry 689 (2004) 3567–3581
˚
M = 158.96, monoclinic, a = 7.5697(10) A, b =
tion. When all acid had been consumed by Me3N (as
shown by the color change of the indicator) another
5% portion of standardized HCl solution was added.
o
˚
3
˚
13.0452(10) A, c = 7.7453(10) A, b = 92.31(1) , V =
˚
764.21(15) A , Z = 4, space group: P21/n, qcalc = 1.382
g cmꢀ3. Data were collected at 293(1) K, Enraf Nonius
MACH3 diffractometer, Mo Ka radiation k = 0.71073
3.2.2. QÆBH(CN)COOMe (1a)
A, x–2h motion, hmax = 25.3o, 1387 measured, 1007
To an acetonitrile (20 ml) solution of [Bu4N]CN (3.74
g, 13.93 mmol) was added QÆBH(Br)COOMe (2.44 g,
9.31 mmol). The mixture was refluxed for 1 h then evap-
orated under reduced pressure to an oily residue that
was taken up and triturated in ether (40 ml) until it
solidified. The solid was filtered off, extracted with the
filtrate ten times, then the filtrate was evaporated to dry-
ness. The solid residue was stirred in water (10 ml) at 0
ꢁC, then the insoluble product was filtered off, washed
with icy water (3 · 5 ml). Yield: 1.37 g (71%). Anal.
Found: C, 57.45 (calc: 57.73%) H, 8.18 (calc: 8.24%)
N, 13.56 (calc: 13.46%) B, 5.13 (calc: 5.20%); 1H
NMR (CDCl3, d): 1.84 (m, 6H, CCH2), 2.12 (h, 1H,
CH), 3.33 (m, 6H, NCH2), 3.59 (s, 3H, OCH3).
13C{1H} NMR (CDCl3, d): 19.58 (CH), 24.03 (CC
H2), 48.93 (OCH3), 50.84 (NCH2), 128.7 (br BCN).
11B NMR (CDCl3, d): ꢀ14.6 (d, J = 96 Hz). IR (KBr,
cmꢀ1): 2397 m(B–H), 2204 m(CN), 1688 m(C@O).
˚
reflections were unique with I > 2r(I), decay: none.
, empirical
absorption correction was applied using the psi scan
method. The structure was solved using the SIR-92 soft-
ware [45] and refined on F2 using SHELX-97 program
[46], publication material was prepared with the
Absorption coefficient l = 0.095 mmꢀ1
WINGX-97
suite
[47],
R(F) = 0.0476
and
wR(F2) = 0.2107 for 1387 reflections, 115 parameters,
goodness of fit G = 0.836. Residual electron density:
0.23/ꢀ0.24 e Aꢀ3. Hydrogen atoms were placed into
˚
geometric position except H1 (B–H) and H2 (N–H)
atoms which were found at the difference electron den-
sity map and their coordinates were also refined.
The boron content of the samples was determined
with acid–base titration in the presence of mannitol,
after fusion with sodium hydroxide and potassium
hydroxide. Lithium content of LiCNÆMeCN has been
determined by acid–base titration. pKb values of amines
were determined at 25 ꢁC in 1.0 M ionic strength media
except for amines h and k where 0.5 M ionic strength
was applied. [48]. Protonation constants of the products
were determined by pH-potentiometric technique in 1.0
M ionic strength media (NaClO4) at 25 ꢁC. For the
measurements a RADIOMETER PHM85 pH-meter, a
pHG211 glass working electrode and K401 calomel ref-
erence electrode filled with saturated LiCl was used.
Standardized base solution (0.2 M) was added by a
ABU 80 automatic burette. Titrated solutions contained
the selected compounds at 0.002 M initial concentration.
Disturbing effect of carbon dioxide content of the air
was avoided by passing argon gas through the solution.
pH difference originated from the diffusion potential
appearing on the surface of the reference electrode has
been taken into account by the method of Irving et al.
[49]. Protonation constants were calculated by the com-
puter program PSEQUAD [50].
3.2.3. pipÆBH(CN)COOMe (1b)
1d (220 mg, 1.41 mmol) was dissolved in acetonitrile
(2 ml), piperidine (360 mg, 4.23 mmol) was added and
the solution was refluxed for 2 h. The reaction mixture
was evaporated under reduced pressure. The residue
was taken up and triturated with ether (2 ml) until it
solidified. It was then filtered off, washed with ether
(0.5 ml) then extracted with ether (5 ml). The extract
was concentrated to 0.5 ml volume to which pentane
(4 ml) was added. A crystalline solid formed that was fil-
tered off after 30 min, washed with pentane (2 · 1 ml).
Yield: 169 mg (66%). Anal. Found: C, 52.35 (calc:
52.79%) H, 8.27 (calc: 8.31%) N, 15.21 (calc: 15.39%)
1
B, 5.91 (calc: 5.94%); H NMR (CDCl3, d): 1.47 (m,
1H, 4-CH2 ax), 1.71 (m, 2H, 3,5-CH2 ax), 1.90 (m,
3H, 3,5-CH2 eq + 4-CH2 eq), 2.86 (m, 2H, 2,6-CH2
ax), 3.32 (m, 2H, 2,6-CH2 eq), 3.64 (s, 3H, OCH3),
4.68 (br, NH). 13C{1H} NMR (CDCl3, d): 22.10 (4-
CH2), 24.27, 24.32 (3,5-CH2), 49.42, 51.31 (2,6-CH2),
129.2 (br, BCN), 188.5 (br, BCOOCH3). 11B NMR
(CDCl3, d): ꢀ15.4 (d, J = 96 Hz). IR (KBr, cmꢀ1):
3154 m(N–H), 2396 m(B–H), 2212 m(C„N), 1654
m(C@O).
3.2. Preparations of amine-cyano(methoxycarbonyl)bor-
anes
3.2.1. General remarks
Compounds 1a and 1d were prepared from the corre-
sponding BH(Br)COOMe complexes with [Bu4N]CN,
while other complexes were synthesized from 1d by
amine exchange processes. Liberated Me3N was purged
from the reaction atmosphere by a slow N2 stream into a
gas-washing bottle containing a solution of methylor-
ange indicator and a known amount of standardized
aqueous HCl solution (5% of the calculated stoichiomet-
ric amount) in order to monitor the progress of the reac-
3.2.4. Et2NHÆBH(CN)COOMe (1c)
To 1d (217 mg, 1.39 mmol) was added diethylamine
(3 ml) and the solution was refluxed for 4.5 h. The reac-
tion mixture was evaporated under reduced pressure to
a viscous oil to which ether was added (4 ml) and was
allowed to stand for 5 h to produce a small amount of
a solid that was filtered off. The filtrate was evaporated
under reduced pressure to a colorless viscous oil which

B. Gy}ori et al. / Journal of Organometallic Chemistry 689 (2004) 3567–3581
3575
was then kept under dynamic vacuum until constant
weight was reached. Yield: 138 mg (64%). Anal. Found:
B, 6.40 (calc: 6.36%); 1H NMR (CDCl3, d): 1.31 (m, 6H,
Et-CH3), 2.96–3.18 (m, 4H, CH2), 3.64 (s, 3H, OCH3),
4.93 (br, 1H, NH).13C{1H} NMR (CDCl3, d): 10.89,
11.38 (Et-CH3), 46.51, 46.93 (Et-CH2), 49.55 (OCH3).
11B NMR (CDCl3, d): ꢀ16.7 (d, J = 96 Hz). IR (net,
cmꢀ1): 3164 m(N–H), 2426 m(B–H), 2209 m(C„N),
1682 m(C@O).
J = 96 Hz). IR (KBr, cmꢀ1): 2415 m(B–H), 2204
m(C„N), 1693, 1641 m(C@O).
3.2.7. 4-NH2-pyÆBH(CN)COOMe (1f)
To 1d (207 mg, 1.33 mmol) dissolved in acetonitrile (6
ml) was added 4-aminopyridine (254 mg, 2.70 mmol)
and the solution was refluxed for 6 h. After evaporation
of the reaction mixture to dryness the residue was sus-
pended in ether (10 ml), filtered and the solid was ex-
tracted with the filtrate eight times. The solid was
further extracted fifty times with a fresh portion of ether
(20 ml). The solid that formed and settled from the ether
during the extraction was filtered off. Yield: 114 mg
3.2.5. Me3NÆBH(CN)COOMe (1d)
Me3NÆBH(Br)COOMe (1.36 g, 6.48 mmol) was
added to the solution of [Bu4N]CN (2.42 g, 9.03 mmol)
in acetonitrile (9 ml) and the mixture was stirred for 30 h
at room temperature. The volatile components were
then evaporated in vacuo, without warming the flask.
The residue was suspended in ether (25 ml), the insoluble
parts were filtered off and extracted six times with the fil-
trate. A small volume of an oily layer had been formed
in the solution on standing. The upper solution phase
was removed and the oily residue was vigorously stirred
and washed with ten portions of fresh ether (15 ml each).
The upper phase and the washing portions were com-
bined and evaporated and the residue was triturated
with pentane (10 ml) until solidification. The solid prod-
uct was then filtered off, washed with pentane (2 · 4 ml).
Yield: 0.85 g (84%). (As determined by 1H and 11B
NMR, the product was still contaninated by 2, 4 and
3 mol% of Me3NÆBH(Br)COOMe, Me3NÆBH(NC)-
COOMe and [Bu4N][BH(CN)2COOMe], respectively.)
Anal. Found: C, 45.85 (calc: 46.20%) H, 8.45 (calc:
8.40%) N, 17.78 (calc: 17.96%) B, 6.81 (calc: 6.93%);
1H NMR (CDCl3, d): 2.90 (s, 9H, NCH3), 3.61 (s, 3H,
OCH3). 13C{1H} NMR (CDCl3, d): 48.77 (OCH3),
51.90 (NCH3), 128.9 (br q, BCN), 185.54 (br q, BC
OOCH3).11B NMR (CDCl3, d): ꢀ12.4 (d, J = 105 Hz).
IR (KBr, cmꢀ1): 2409, 2432 m(B–H), 2208 m(C„N),
1689 m(C@O).
1
(45%). Anal. Found: B, 5.65 (calc: 5.66%); H NMR
(DMSO-d6, d): 3.44 (s, 3H, OCH3), 6.73 (d, 2H, 3,5-
CH), 7.64 (br s, 2H, NH), 7.87 (d, 2H, 2,6-CH).13C{1H}
NMR (DMSO-d6, d): 48.32 (OCH3), 108.18 (3,5-CH),
146.22 (2,6-CH), 157.07 (ipso-C). 11B NMR (DMSO-
d6, d): ꢀ14.0 (br). IR (KBr, cmꢀ1): 3439, 3357, 3251
m(N–H), 2403 m(B–H), 2211 m(C„N), 1689, 1650
m(C@O).
3.2.8. MorfÆBH(CN)COOMe (1g)
To an acetonitrile solution (2 ml) of 1d (220 mg, 1.41
mmol) was added morpholine (370 mg, 4.25 mmol) and
the solution was refluxed for 3 h, after which it was
evaporated to a viscous oil. The residue was triturated
in ether (5 ml) and the solid formed was filtered off,
washed with ether (2 · 2 ml) and dried in a nitrogen
stream. This was then extracted with chloroform until
a gel-like substance remained on the filter disc. Evapora-
tion of the extract gave a solid which was then sus-
pended in ether (10 ml), filtered off, extracted six times
with the filtrate. Yield: 142 mg (55%). Anal. Found: B,
1
5.92 (calc: 5.88%); H NMR (CDCl3, d): 3.00–3.33 (m,
4H, NCHH ax + OCHH ax ), 3.65 (s, 3H, OCH3),
3.80 (m, 2H, NCHH eq), 4.01 (m, 2H, OCH H eq),
5.39 (br, 1H, NH). 13C{1H} NMR (CDCl3, d): 49.00,
50.04 (NCH2), 49.73 (OCH3), 65.16 (OCH2). 11B
NMR (CDCl3, d): ꢀ15.5 (d, J = 96 Hz). IR (KBr,
cmꢀ1): 3148 m(N–H), 2417 m(B–H), 2195 m(C„N),
1668 m(C@O).
3.2.6. 4-DMAPÆBH(CN)COOMe (1e)
1d (206 mg, 1.32 mmol) was dissolved in acetonitrile
(5 ml) and 4-dimethylaminopyridine (325 mg, 2.66
mmol) was added. The solution was refluxed for 8 h then
evaporated under reduced pressure to a gel-like mate-
rial, which was then triturated in ether (6 ml) and, after
solidification, filtered off. The solid was then extracted
three times with the filtrate. The extraction was contin-
ued with a fresh portion of ether (20 ml) until a sticky
solid remained on the filter disc. The etheral extract
was evaporated to 5 ml final volume, the solid was then
filtered off. Yield: 133 mg (46%). Anal. Found: B, 4.99
(calc: 4.94%); 1H NMR (CDCl3, d): 3.18 (s, 6H,
NCH3), 3.61 (s, 3H, OCH3), 6.64 (d, 2H, 3,5-CH),
8.03 (d, 2H, 2,6-CH). 13C{1H} NMR(CDCl3, d): 39.71
(NCH3), 49.26 (OCH3), 106.83 (3,5-CH), 145.96 (2,6-
CH), 155.86 (ipso-C). 11B NMR (CDCl3, d): ꢀ14.3 (d,
3.2.9. N-Me-MorfÆBH(CN)COOMe (1h)
To an acetonitrile solution (2 ml) of 1d (215 mg, 1.38
mmol) was added N-Me-morpholine (420 mg, 4.15
mmol), the solution was refluxed for 12 h and then it
was evaporated to a viscous oil, which solidified on
standing. Ether (5 ml) was added and the solid was fil-
tered off, washed with ether (2 · 2 ml) and dried in a
nitrogen stream. This was then washed down with chlo-
roform (4 ml) leaving a gel-like material behind on the
filter disc. Evaporation of the chloroform solution re-
sulted in a solid which was suspensed in 1:1 ether-pen-
tane mixture (5 ml), and filtered off. Yield: 173 mg
1
(63%). Anal. Found: B, 5.44 (calc: 5.46%); H NMR

3576
B. Gy}ori et al. / Journal of Organometallic Chemistry 689 (2004) 3567–3581
(CDCl3, d): 3.00 (s, 3H, NCH3), 3.05, 3.17, 3.48, 3.73
(multiplets, 4 · 1 H, NCH2), 3.62 (s, 3H, OCH3), 3.81–
4.03 (m, 4H, OCH2). 13C{1H} NMR (CDCl3, d): 45.00
(NCH3), 49.24 (OCH3), 56.04, 57.69 (NCH2), 60.51,
60.64 (OCH2). 11B NMR (CDCl3, d): ꢀ13.6 (d,
J = 105 Hz). IR (KBr, cmꢀ1): 2422 m(B–H), 2207
m(C„N), 1687 m(C@O).
oil. Ether (10 ml) was added and the mixture was stirred
for 10 min meanwhile the oil solidified. The solid was fil-
tered off, washed with ether (1 ml) then extracted with
fresh ether (5 ml) until a loose and slowly sedimenting
solid remained on the filter disc. The ether extract was
concentrated to a 3 ml volume and the fine needle crys-
tals formed were filtered off and washed with ether
(2 · 0.5 ml). Yield: 178 mg (69%). Anal. Found: C,
56.82 (calc: 56.89%) H, 5.77 (calc: 5.84%) N, 14.83 (calc:
14.74%) B, 5.67 (calc: 5.69%); 1H NMR (CDCl3, d): 2.61
(s, 3H, CCH3), 3.62 (s, 3H, OCH3), 7.56 (d, 2H, 3,5-
CH), 8.49 (d, 2H, 2,6-CH). 13C{1H} NMR (CDCl3, d):
21.26 (CC H3), 48.98 (OCH3), 126.89 (3,5-CH), 128.70
(br, BCN), 146.05 (2.6-CH), 155.92 (ipso-C), 186.6 (br,
BCOOCH3). 11B NMR (CDCl3, d): ꢀ13.3 (d, J = 96
Hz). IR (KBr, cmꢀ1): 2434 m(B–H), 2207 (C„N), 1684
m(C@O).
3.2.10. mimidÆBH(CN)COOMe (1i)
1d (292 mg, 1.87 mmol) was dissolved in acetonitrile
(4 ml) and 1-methylimidazole (456 mg, 5.55 mmol)
added and the solution was refluxed for 5 h. The solu-
tion was then evaporated to an oil under reduced pres-
sure and dried to a constant mass under vacuum. To
the oily residue ether was added (4 ml), magnetically
stirred for approximately 5 min then supernatant etheral
solution was removed. To the oil a fresh portion of ether
was added, stirred and removed. This procedure was re-
peated ten times until a solid material was obtained,
which was finally dried to a constant mass under vac-
uum. Yield: 266 mg (79%). Anal. Found: C, 46.78 (calc:
46.97%) H, 5.50 (calc: 5.63%) N, 23.41 (calc: 23.48%) B,
3.2.13. pyÆBH(CN)COOMe (1m)
A solution of 1d (216 mg, 1.38 mmol) and pyridine
(219 mg, 2.77 mmol) in acetonitrile (4 ml) was refluxed
for 14 h then evaporated under reduced pressure to an
oil, which was kept under dynamic vacuum to reach
constant weight. Ether was added to the viscous oily res-
idue and the mixture was stirred for 10 min. The ether
was then distilled off and the residue was kept under dy-
namic vacuum until the oil solidified completely. The so-
lid was suspended in pentane (8 ml), filtered off, and
dried in a nitrogen stream. The dry solid was then ex-
tracted with ether (10 ml) until a yellowish powder re-
mained on the filter disc. The extract was concentrated
to a 4 ml volume and the crystals formed were filtered
off and washed with ether (2 · 0.5 ml). Yield: 161 mg
1
5.91 (calc: 6.04%); H NMR (CDCl3, d): 3.61 (s, 3H,
OCH3), 3.90 (s, 3H, NCH3), 7.11 (s, 1H, 5-CH), 7.26
(s, 1H, 4-CH), 8.23 (s, 1H, 2-CH). 13C{1H} NMR
(CDCl3, d): 35.48 (NCH3), 49.23 (OCH3), 121.80 (4-
CH), 126.36 (5-CH), 137.46 (2-CH). 11B NMR (CDCl3,
d): ꢀ18.1 (d, J = 105 Hz). IR (KBr, cmꢀ1): 2398 m(B–H),
2205 m(C„N), 1685 m(C@O).
3.2.11. 3-NH2-pyÆBH(CN)COOMe (1k)
To 1d (210 mg, 1.35 mmol) dissolved in acetonitrile (2
ml) was added 3-aminopyridine (382 mg, 4.06 mmol)
freshly recrystallized from ether. The solution was ref-
luxed for 6 h, evaporated under reduced pressure and
then kept under dynamic vacuum (0.1 mbar). To the res-
idue ether (10 ml) was added and stirred for 30 min.
Ether was distilled out and the resultant resin-like sub-
stance was kept under vacuum for 30 min. When the
majority of the resin recrystallized water (1.5 ml) was
added, stirred for 10 min and the solid was filtered off
and washed with water (4 · 0.2 ml). Yield: 112 mg
1
(66%). Anal. Found: B, 6.17 (calc: 6.14%); H NMR
(CDCl3, d): 3.64 (s, 3H, OCH3), 7.80 (m, 2H, 3,5-CH),
8.25 (m, 1H, 4-CH), 8.69 (m, 2H, 2,6-CH). 13C{1H}
NMR (CDCl3, d): 49.69 (OCH3), 126.42 (3,5-CH),
142.53 (4-CH), 147.46 (2,6-CH), 11B NMR (CDCl3, d):
ꢀ12.9 (d, J = 105 Hz). IR (KBr, cmꢀ1): 2431 m(B–H),
2209 (C„N), 1677 m(C@O).
3.3. Preparations of amine-cyanocarboxyboranes
1
(44%). Anal. Found: B, 5.60 (calc: 5.66%); H NMR
(CDCl3, d): 3.62 (s, 3H, OCH3), 4.49 (br 2H, NH2),
7.36 (m, 1H, 4-CH), 7.42 (m, 1H, 5-CH), 7.94 (m, 1H,
6-CH), 8.04 (m, 1H, 2-CH).13C{1H} NMR (CDCl3, d):
49.61 (OCH3), 126.31, 126.40 (4,5-CH), 133.36 (2-CH),
136.61 (6-CH), 145.49 (ipso-C). 11B NMR (CDCl3, d):
ꢀ13.0 (d, J = 86 Hz). IR (KBr, cmꢀ1): 3418, 3346,
3244 m(N–H), 2451, 2427 m(B–H), 2210 m(C„N), 1680,
1655 m(C@O).
3.3.1. QÆBH(CN)COOH (2a)
0.1 M hydrochloric acid (5.0 ml) and acetone (1.5 ml)
were added to 1a (199 mg, 0.96 mmol) and the mixture
was heated for 3 h in a flask that was equipped with a
reflux condenser and immersed in an oil bath at a tem-
perature of 60–64 ꢁC. The acetone was evaporated with
a nitrogen stream meanwhile the temperature of the oil
bath was gradually increased to 75 ꢁC. Finally, the tem-
perature of the oil bath was increased to 100 ꢁC in 10
min, kept at this value for an additional 10 min and then
the reaction mixture was removed from the bath, al-
lowed to cool to room temperature and placed into an
ice-water bath for 1 h. The resulting crystals were fitered
3.2.12. picÆBH(CN)COOMe (1l)
A solution of 1d (213 mg, 1.37 mmol) and picoline
(3.82 mg, 4.10 mmol) in acetonitrile (2 ml) was refluxed
for 2.5 h then evaporated under reduced pressure to an

B. Gy}ori et al. / Journal of Organometallic Chemistry 689 (2004) 3567–3581
3577
off and washed with ice water (4 · 0.4 ml). Yield: 137 mg
(74%). Anal. Found: B, 5.55 (calc: 5.57%); H NMR
and extracted with the filtrate until a small amount of
a waxy solid remained on the filter disc. The extract
was concentrated to a 5 ml volume, the solid formed
was filtered off and washed with ether (3 · 0.5 ml).
Yield: 122 mg (66%). Anal. Found: B, 7.57 (calc:
7.62%); 1H NMR (CDCl3, d): 2.89 (s, 9H, NCH3),
9.87 (br, 1H COOH). 13C{1H} NMR (CDCl3, d):
48.77 (OCH3), 51.48 (NCH3). 11B NMR (CDCl3, d):
ꢀ12.6 (d, J = 105 Hz). IR (KBr, cmꢀ1): 2419 m(B–H),
2210 m(C„N), 1662 m(C@O).
1
(CDCl3, d): 1.86 (m, 6H, CCH2), 2.12 (h, 1H, CH),
3.35 (m, 6H, NCH2), 10.49 (br, 1H, COOH). 13C{1H}
NMR(CDCl3, d): 19.66 (CH), 24.13 (CC H2), 50.98
(NCH2). 11B NMR (CDCl3, d): ꢀ13.7 (d, J = 96 Hz).
IR (KBr, cmꢀ1): 2406 m(B–H), 2210 m(CN), 1653
m(C@O).
3.3.2. pipÆBH(CN)COOH (2b)
To 1b (101 mg, 0.55 mmol) was added 0.1 M hydro-
chloric acid (2.5 ml) and the solution was kept at 70 ꢁC
in a bath for 30 min followed by heating the reaction
mixture to boiling for a short period. It was cooled
and stored in a refrigirator overnight. The crystals
formed were filtered off and washed with ice water
(4 · 0.4 ml). Yield: 67 mg (73%). Anal. Found: B, 6.45
3.3.5. DMAPÆBH(CN)COOH (2e)
1e (170 mg, 0.78 mmol) was dissolved in a mixture of
0.1 M hydrochloric acid (4.0 ml) and acetone (4.0 ml).
The flask was equipped with a reflux condenser, placed
into a 50 ꢁC bath for 30 min, then the temperature
was increased to 60 ꢁC for 80 min. Then nitrogen gas
was bubbled through the solution with mild warming
to remove the acetone, which resulted in the precipita-
tion of a solid. Water (4 ml) was added and the solution
was heated to boiling to redissolve the solid. This solu-
tion was then allowed to crystallize at room temperature
for 2 h. The crystals formed were filtered off and washed
with water (3 · 0.5 ml). Yield: 128 mg (80%). Anal.
1
(calc: 6.43%); H NMR (D2O, d): 1.43 (m, 1H, 4-CH2
ax), 1.65 (m, 2H, 3,5-CH2 ax), 1.81 (m, 3H, 3,5-CH2
eq + 4-CH2 eq), 2.75 (m, 2H, 2,6-CH2 ax), 3.28 (m,
2H, 2,6-CH2 eq). 13C{1H} NMR (D2O, d): 24.49 (4-
CH2), 26.67 (3,5-CH2), 52.42, 53.61 (2,6-CH2). 11B
NMR (D2O, d): ꢀ16.3 (d, J = 96 Hz). IR (KBr, cmꢀ1):
3207 m(N–H), 2403 (B–H), 2215 m(C„N), 1695, 1639
(C@O).
1
Found: B, 5.25 (calc: 5.27%); H NMR (DMSO-d6, d):
3.13 (s, 6H, NCH3), 6.89 (d, 2H, 3,5-CH), 7.94 (d, 2H,
2,6-CH), 10.98 (s, 1H, COOH). 13C{1H} NMR
(DMSO-d6, d): 39.24 (NCH3), 49.26 (OCH3), 107.20
(3,5-CH), 145.47 (2,6-CH), 155.56 (ipso-C). 11B NMR
(DMSO-d6, d): ꢀ14.2 (br). IR (KBr, cmꢀ1): 2400 m(B–
H), 2204 m(CN), 1653 m(C@O).
3.3.3. Et2NHÆBH(CN)COOH (2c)
1c (119 mg, 0.70 mmol) was suspended in 0.1 M
hydrochloric acid (3.0 ml), stirred at 60 ꢁC in a water
bath, and the temperature of the bath was gradually in-
creased to 80 ꢁC over 70 min. The solution was evapo-
rated to a 1 ml volume under reduced pressure and
allowed to stand. When crystal formation started the
solution was stirred for a few minutes then placed in a
ice water bath. After 2 h the crystals were filtered off
and washed with ice water (3 · 0.2 ml). Yield: 44 mg
3.3.6. 4-NH2-pyÆBH(CN)COOH (2f)
1f (80 mg, 0.42 mmol) was dissolved in 0.1 M hydro-
chloric acid (2.0 ml), the solution was kept at 65 ꢁC for
18 min then cooled to room temperature. Crystal forma-
tion was initiated by scratching inner walls of the flask
with a glass rod. After several hours the crystals were fil-
tered off and washed with water (3 · 0.3 ml). Yield: 64
mg (86%). Anal. Found: B, 6.06 (calc: 6.11%); 1H
NMR (D2O, d): 6.77 (d, 2H, 3,5-CH), 7.89 (d, 2H,
2,6-CH).13C{1H} NMR (D2O, d): 112.50 (3,5-CH),
149.19 (2,6-CH), 160.91 (ipso-C). 11B NMR (D2O, d):
ꢀ14.8 (d, J = 96 Hz). IR (KBr, cmꢀ1): 3459, 3355
m(N–H), 2448 (B–H), 2203 m(C„N), 1643 m(C@O).
1
(40%). Anal. Found: B, 6.92 (calc: 6.93%); H NMR
(D2O, d): 1.24 (t, 6H, Et-CH3), 2.84–3.12 (m, 4H,
CH2).13C{1H} NMR (D2O, d): 12.90, 13.10 (Et-CH3),
48.79, 49.28 (Et-CH2). 11B NMR (D2O, d): ꢀ17.6 (d,
J = 96 Hz). IR (KBr, cmꢀ1): 3118 m(N—H), 2425 m(B–
H), 2217 m(C„N), 1661 m(C@O).
3.3.4. Me3NÆBH(CN)COOH (2d)
1d (204 mg, 1.31 mmol) was dissolved in a mixture of
water (2.6 ml) and 1 M hydrochloric acid (0.66 ml) and
the solution was heated at 60 ꢁC for 5 min in a water
bath, then bath temperature was increased to 100 ꢁC
for 24 min. The solution was evaporated to dryness un-
der reduced pressure, and the residue was co-evaporated
with methylene chloride twice (2 ml each) to remove
residual water. The solid was then redissolved in methyl-
ene chloride (5 ml) and filtered. The filtered-off solid was
washed with methylene chloride (3 · 0.4 ml). The filtrate
and washes were combined and evaporated to dryness.
The residue was triturated in ether (12 ml), filtered off
3.3.7. MorfÆBH(CN)COOH (2g)
To 1g (124 mg, 0.67 mmol) was added 0.1 M hydro-
chloric acid (2.9 ml) and the mixture was kept at 65 ꢁC
until the starting material completely dissolved. The
solution was gradually heated to 100 ꢁC in 1 h and kept
at this temperature for 15 min and then it was evapo-
rated under reduced pressure to an oily solid. This was
suspended in acetonitrile (2 ml), allowed to stand for 1
h and then the acetonitrile was distilled off. The solid
residue was re-suspended in acetonitrile (3 ml), filtered

3578
B. Gy}ori et al. / Journal of Organometallic Chemistry 689 (2004) 3567–3581
off, dried in a nitrogen stream and extracted ten times
with ether (8 ml). Yield: 66 mg (58%). Anal. Found:
B, 6.43 (calc: 6.36%); H NMR (D2O, d): 3.08 (m, 2H,
11B NMR (D2O, d): ꢀ13.7 (d, J = 96 Hz). IR (KBr,
cmꢀ1): 3433, 3347, 3228 m(N–H), 2428 m(B–H), 2216
m(C„N), 1646 m(C@O).
1
NCHH ax), 3.26 (m, 2H, OCHH ax ), 3.78 (m, 2H,
NCHH eq), 4.03 (m, 2H, OCHH eq). 13C{1H} NMR
(D2O, d): 51.00, 51.97 (NCH2), 67.58 (OCH2). 11B
NMR (D2O, d): ꢀ16.3 (d, J = 96 Hz). IR (KBr, cmꢀ1):
3106 m(N–H), 2425 m(B–H), 2213 m(C„N), 1678
m(C@O).
3.3.11. PicÆBH(CN)COOH (2l)
1l (117 mg, 0.62 mmol) was dissolved in a mixture
of 0.1 M hydrochloric acid (2.8 ml) and acetone (1.4
ml). The flask was equipped with a reflux condenser
and heated in a bath at 75 ꢁC for 2 h, then the acetone
was evaporated by bubbling nitrogen gas through the
solution. The solution was then heated at 80–85 ꢁC
for 30 min, cooled to room temperature. A few drops
of the solution was evaporated to dryness and then
the solution was seeded with the resulting solid. After
storing the solution in a refrigerator overnight the crys-
tals were removed from the walls of the flask and the
slurry was placed in a ice-water bath for a few hours
to complete the crystallization. The product was fil-
tered off and washed with ice water (3 · 0.2 ml). Yield:
3.3.8. N-Me-MorfÆBH(CN)COOH (2h)
Hydrochloric acid (0.1 M, 4.6 ml) was added to 1h
(136 mg, 0.69 mmol) and the solution was kept in a
100 ꢁC bath for 38 min. The solution was then cooled
to room temperature, evaporated to a 1 ml volume un-
der reduced pressure and stored in a refrigerator over-
night. The crystals formed were dispersed and the
solution was placed in an ice water bath for several
hours. The product was filtered off and washed with
ice water (3 · 0.2 ml). Yield: 48 mg (38%). Anal. Found:
1
1
57 mg (52%). Anal. Found: B, 6.09 (calc: 6.14%); H
B, 5.87 (calc: 5.88%); H NMR (D2O, d): 2.98 (s, 3H,
NMR (D2O, d): 2.58 (s, 3H, CCH3), 7.70 (d, 2H,
3,5-CH), 8.45 (d, 2H, 2,6-CH). 13C{1H} NMR (D2O,
d): 23.66 (CCH3), 130.31 (3,5-CH), 149.21 (2.6-CH),
160.46 (ipso-C). 11B NMR (D2O, d): ꢀ13.9 (d, J = 89
Hz). IR (KBr, cmꢀ1): 2429 m(B–H), 2209 m(C„N),
1656 m(C@O).
NCH3), 3.11–3.24, 3.39–3.59 (multiplets, 2 · 2 H,
NCH2), 3.88–4.05 (m, 4H, OCH2). 13C{1H} NMR(D2O,
d): 46.82 (NCH3), 58.85, 59.82 (NCH2), 63.15, 63.23
(OCH2). 11B NMR (D2O, d): ꢀ13.4 (d, J = 96 Hz). IR
(KBr, cmꢀ1): 2440 m(B–H), 2210 m(C„N), 1658 m(C@O).
3.3.9. MimidÆBH(CN)COOH (2i)
1i (142 mg, 0.79 mmol) was dissolved in 0.1 M hydro-
chloric acid (3.0 ml) and kept at 75 ꢁC for 50 min. The
solution was evaporated to a 1 ml volume and, after the
crystallization started, placed in a refrigerator overnight.
The slurry was then immersed in a ice-water bath for 1 h
then the crystals were filtered off and washed with ice
water (3 · 0.15 ml). Yield: 49 mg (38%). Anal. Found:
3.3.12. PyÆBH(CN)COOH (2m)
To 1m (116 mg, 0.66 mmol) was added 0.1 M
hydrochloric acid (2.9 ml), the solution was kept at
75 ꢁC for 2.5 h then evaporated to the half of its vol-
ume and cooled to room temperature. After the begin-
ning of the crystallization the solution was placed in
an ice-water bath for 2 h, then the crystals were fil-
tered off and washed with ice water (4 · 0.15 ml).
Yield: 69 mg (66%). Anal. Found: B, 6.64 (calc:
6.68%); 1H NMR (D2O, d): 7.91 (m, 2H, 3,5-CH),
8.39 (m, 1H, 4-CH), 8.67 (m, 2H, 2,6-CH). 13C{1H}
NMR (D2O, d): 129.86 (3,5-CH), 146.69 (4-CH),
150.24 (2.6-CH). 11B NMR (D2O, d): ꢀ13.6 (d,
J = 96 Hz). IR (KBr, cmꢀ1): 2432 m(B–H), 2208
m(CN), 1659 m(C@O).
1
B, 6.62 (calc: 6.55%); H NMR (D2O, d): 3.84 (s, 3H,
NCH3), 7.24 (s, 1H, 5-CH), 7.32 (s, 1H, 4-CH), 8.31
(s, 1H, 2-CH). 13C{1H} NMR (D2O, d): 37.62
(NCH3), 125.93 (4-CH), 127.89 (5-CH), 141.62 (2-CH).
11B NMR (D2O, d): ꢀ18.8 (d, J = 96 Hz). IR (KBr,
cmꢀ1): 2445 m(B–H), 2209 m(C„N), 1668 m(C@O).
3.3.10. 3-NH2-pyÆBH(CN)COOH (2k)
1k (98 mg, 0.51 mmol) was dissolved in a mixture of
0.1 M hydrochloric acid (2.2 ml) and acetone (1.5 ml).
The flask was equipped with a reflux condenser and
heated in a bath at 70 ꢁC for 20 min then the bath tem-
perature was increased to 80 ꢁC and kept at this temper-
ature for 30 min. The acetone was removed by bubbling
nitrogen gas through the solution at 50 ꢁC. After several
hours a crystalline product settled, which was filtered off
and washed with water (3 · 0.2 ml). Yield: 79 mg (88%).
3.4. Preparation and/or NMR data of compounds with 2-
NH2-py
3.4.1. 2-NH2-pyÆBH(CN)COOMe (1j)
1H NMR (CDCl3, d): 3.63 (s, 3H, OCH3), 5.93 (br
2H, NH2), 6.83 (m, 2H, 3,5-CH), 7.70 (m, 1H, 4-CH),
8.09 (m, 1H, 6-CH). 11B NMR (CDCl3, d): ꢀ16.08 (d,
1
1
Anal. Found: B, 6.11 (calc: 6.11%); H NMR (D2O, d):
J = 96 Hz). H NMR (D2O, d): 3.62 (s, 3H, OCH3),
7.58 (m, 2H 4,5-CH), 7.93 (m, 1H, 6-CH), 8.00 (m, 1H,
2-CH).13C{1H} NMR (D2O, d): 129.86 (4-CH), 130.99
(5-CH), 136.55 (2-CH), 139.46 (6-CH), 149.49 (ipso-C).
6.84 (m, 1H, 3-CH), 6.94 (m, 1H, 5-CH), 7.77 (m, 1H,
4-CH), 7.87 (m, 1H, 6-CH).11B NMR (D2O, d): ꢀ17.2
(d, J = 96 Hz).

B. Gy}ori et al. / Journal of Organometallic Chemistry 689 (2004) 3567–3581
3579
3.4.2. 2-NH2-pyÆBH(CN)COONa (Na salt of 2j)
1H NMR (D2O, d): 6.82 (m, 1H, 3-CH), 6.91 (m, 1H,
5-CH), 7.73 (m, 1H, 4-CH), 7.87 (m, 1H, 6-CH).
13C{1H} NMR(D2O, d): 116.07, 116.48 (3,5-CH),
144.34, 145.95 (4,6-CH), 170.81 (ipso-C). 11B NMR
(D2O, d): ꢀ16.6 (d, J = 96 Hz).
6.15 mmol) was added. The solution was refluxed for
2.5 h, then diluted to a 14 ml volume with acetonitrile
and this solution was chromatographed with acetonitrile
on a column containing 60 g of Silica gel 60 (particle
size: 0.04–0.063 mm). 5 ml fractions were collected and
all fractions were checked with TLC (HP-TLC Silica
60, eluent: 40% acetonitrile, 60% chloroform, spots were
visualized by iodine vapour and then with the Drag-
endorffꢀs reagent) and fractions containing the pure
product (Rf = 0.55) were combined and evaporated un-
der reduced pressure to yield a dense oil, which was dis-
solved in methanol (30 ml) and then water (60 ml) was
added under stirring. The moderately opalescent solu-
tion was treated with charcoal, filtered and passed
through a column containing potassium ion form of
Dowex 50X8-200 cation exchange resin, which was
washed with a mixture of water and methanol (2:1) prior
to separation. Water–methanol mixture (2:1 volume
ratio, 80 ml) was used for the elution of K[BH(CN)2-
COOMe] after loading of a solution of [Bu4N][BH-
(CN)2COOMe] on the column. The eluted solution
was evaporated under reduced pressure to a solid, which
was further dried in vacuum to constant weight. The
solid was then triturated in methylene chloride (5 ml),
filtered off and washed with methylene chloride (5 · 1
ml). Yield: 720 mg (72%). Anal. Found: B, 6.62 (calc:
3.4.3. –HN-C5H4N–BH(CN)CO–, (lactam) (3)
1d (206 mg, 1.32 mmol) and 2-NH2-py (398 mg, 4.22
mmol) were dissolved in acetonitrile (2 ml) and the solu-
tion was refluxed for 22 h then the solvent was evaporated
under reduced pressure. Ether (4 ml) was added to the oily
residue, stirred for a few minutes then the ether was re-
moved from the oil. This procedure was repeated three
times to afford a semi-solid residue that was redissolved
in water (10 ml) and refluxed for 1 h. The solution was
cooled to and kept at 40 ꢁC, treated with charcoal, stirred
for 30 min and then filtered. The filtrate was evaporated to
dryness under reduced pressure, then the residue was dis-
persed in methanol (3 ml) and evaporated to dryness
again. The resulting solid was triturated in chloroform
(3 ml), filtered off and washed with chloroform (3 · 0.4
ml). Yield: 82 mg (39%). Anal. Found: B, 6.84 (calc:
1
6.80%); H NMR (DMSO-d6, d): 7.17 (m, 1H, 3-CH),
7.29 (m, 1H, 5-CH), 8.14 (m, 1H, 4-CH), 8.41 (m, 1H, 6-
CH), 10.67 (br, 1H, NH).13C{1H} NMR(DMSO- d6, d):
110.22 (3-CH), 117.86 (5-CH), 141.92 (4-CH), 144.87
(6-CH), 155.55 (ipso-C). 11B NMR (DMSO-d6, d):
ꢀ17.3 (br).1H NMR (D2O, d): 7.27 (m, 1H, 3-CH), 7.36
(m, 1H, 5-CH), 8.14 (m, 1H, 4-CH), 8.32 (m, 1H,
6-CH).13C{1H} NMR(D2O, d): 113.59 (3-CH), 121.87
(5-CH), 144.80 (4-CH), 147.87 (6-CH), 157.40 (ipso-C).
11B NMR (D2O, d): ꢀ17.6 (d, J = 96Hz). 1H NMR
(0.183M NaOD, complex:NaOD = 1:4): 6.93 (m, 1H,
3-CH), 7.01 (m, 1H, 5-CH), 7.81 (m, 1H, 4-CH),
8.05 (m, 1H, 6-CH). 13C{1H} NMR(0.183M NaOD,
complex:NaOD = 1:4): 116.92 (3-CH), 117.71 (5-CH),
143.43 (4-CH), 145.60 (6-CH), 158.89 (ipso-C). 11B
NMR (0.183 M NaOD, complex:NaOD = 1:4): ꢀ17.7
(d, J = 96 Hz). IR (KBr, cmꢀ1): 3080 m(N–H), 2447
m(B–H), 2210 m(C„N), 1695, 1637 m(C@O).
1
6.67%); H NMR (D2O, d): 2.02 (q, 1H, BH, J = 94.3
Hz), 3.60 (s, 3H, OCH3).13C{1H} NMR(quantitative)
(D2O, d): 52.85 (1C, OCH3), 134.42 (2C, BCN,
JBC = 63.1 Hz), 193.63 (1C, BC OOCH3, JBC = 73.8
Hz). 11B NMR (D2O, ): ꢀ32.7 (d, J = 96 Hz). IR
(KBr, cmꢀ1): 2406 m(B–H), 2216 m(C„N), 1700 m(C@O).
3.6.2. K[B(OH)(CN)2COOMe] (5)
4 (203 mg, 1.25 mmol) and KHCO3 (350 mg, 3.50
mmol) were dissolved in water (5 ml) and the solution
was cooled to 0 ꢁC. Bromine (205 mg, 1.28 mmol) was
placed in a Schlenk-type flask and flushed completely
into the aqueous solution with nitrogen gas over a per-
iod of 1.5 h. When approximately 20% of the bromine
remained in the flask the cooling of the reaction mixture
was stopped. Half-an-hour after total consumption of
the bromine a sample was taken and checked with 11B
NMR to determine the degree of conversion of 4 and
then if it was necessary a calculated amount of bromine
was added to complete the reaction. The pH of the reac-
tion mixture was checked and set between pH 7 and 8
with KHCO3, then the solution was evaporated to dry-
ness under reduced pressure in a 40 ꢁC water bath. The
residue was kept under vacuum to reach a constant
weight. The solid was suspended in acetonitrile (3 ml),
filtered off, washed with acetonitrile (3 · 1.5 ml). The fil-
trate and washings were combined, decolorized with
charcoal and evaporated to dryness. The residue was
dissolved in methanol (4 ml), evaporated to dryness
3.5. A QÆBH(NC)COOMe, NMR data
1H NMR (CDCl3, d): 1.87 (m, 6H, CCH2), 2.15 (h,
1H, CH), 3.28 (m, 6H, NCH2), 3.59 (s, 3H, OCH3).
13C{1H} NMR(quantitative) (CDCl3, d): 19.32 (1C,
CH), 23.42 (3C, CCH2), 48.49 (1C, OCH3), 49.02 (3C,
NCH2), 170.37 (1C, BNC), 185.12 (1C, BCOOCH3).
11B NMR (CDCl3, d): ꢀ9.8 (d, J = 96 Hz).
3.6. Preparations of cyano(methoxycarbonyl)borates
3.6.1. K[BH(CN)2COOMe] (4)
[Bu4N]CN (3.38 g, 12.58 mmol) was dissolved in
acetonitrile (8.5 ml) and MeNÆBH(Br)COOMe (1.29 g,

3580
B. Gy}ori et al. / Journal of Organometallic Chemistry 689 (2004) 3567–3581
after 20 min, suspended in ether (5 ml) and filtered off.
Yield: 185 mg (83%). Anal. Found: B, 6.11 (calc:
6.07%); 1H NMR (D2O, d): 3.64 (s, OCH3).13C{1H}
NMR(quantitative) (D2O, d): 52.91 (1C, OCH3),
134.08 (2C, BCN, JBC = 64.1 Hz), 192.03 (1C, BC
OOCH3, JBC = 77.8 Hz). 11B NMR (D2O, ): ꢀ15.84
3H, OCH3). 13C{1H} NMR(CD3CN, d): 49.95
(OCH3). 11B NMR (CD3CN, d) ꢀ15.1. IR (KBr,
cmꢀ1): 3607 m(O–H), 2208, 2201 m(C„N), 1671 m(C@O).
pale yellow colour of the mixture disappeared. The
temperature was kept at 25 ꢁC by employing a cooling
bath. The volatile parts were then evaporated in vacuo,
the residue was triturated with ether which was then
evaporated. The solid residue was suspended in 0.01
M HBr solution (25 ml) at 0 ꢁC, the product was fil-
tered off, washed with 0.01 M HBr (2 · 10 ml), and
then with water (2 · 5 ml). Yield 8.02 g (87%). Anal.
1
(s). H NMR (CD3CN, d): 1.54 (s, 1H, BOH), 3.53 (s,
1
Found: B, 5.19 (calc: 5.15%). H, 11B NMR and IR-
spectra were identical with the previously published
values [20].
3.6.3. K[BH2(CN)COOMe] (6)
Me3NÆBH2COOMe (1.47 g, 11.22 mmol) was dis-
solved in acetonitrile (18 ml) and LiCNÆMeCN (2.47
g, 33.37 mmol) was added. The suspension was refluxed
for 200 h then the solid was filtered off and washed with
acetonitrile (3 · 3 ml). The filtrate and washings were
combined and evaporated under reduced pressure to a
viscous oil, which was kept in vacuum (0.01 mbar) until
constant weight was reached. The residue was redis-
solved in water (15 ml), treated with charcoal, filtered,
then loaded onto a column containing 80 ml of potas-
sium ion form of Dowex 50X8-200 cation exchange re-
sin. The product was eluted with water over a period
of 2 h at an approximate rate of 1 drop per second.
The presence of hydroborate ions in the downcoming
solution was identified from one drop of the solute by
adding silver nitrate solution followed by moderate
warming which produced finely dispersed silver as a
black precipitate. The precipitation of silver occurred
at room temperature from more concentrated solutions.
The eluted solution was evaporated to dryness, and the
residue was triturated in chloroform (5 ml), filtered off,
washed with chloroform (4 · 3 ml) and dried in a nitro-
gen stream. The crude product, which still contained a
significant amount of potassium cyanide, was washed
with acetonitrile (4 · 3 ml then 4 · 1 ml). The washings
were combined and evaporated under reduced pressure
to dryness. The solid residue was titurated in methylene
chloride (5 ml) and filtered off. Yield: 973 mg (63%).
Anal. Found: B, 7.93 (calc: 7.89%); 1H NMR (D2O,
d): 1.30 (q, 1H, BH, J = 90.4 Hz), 3.52 (s, 3H,
OCH3).13C{1H} NMR(quantitative) (D2O, d): 52.09
(1C, OCH3), 141.98 (1C, BCN, JBC = 57.2 Hz), 201.68
(1C, BCOOCH3, JBC = 69.1 Hz). 11B NMR (D2O, d):
ꢀ33.5 (t, J = 90 Hz). IR (KBr, cmꢀ1): 2402, 2371 m(B–
H), 2193 m(C„N), 1653 m(C@O).
3.7.2. QÆBH(Br)COOMe
The reaction between QÆBH2COOH (3.18 g, 18.81
mmol) and NBS (3.36 g, 18.88 mmol) and the workup
of the mixture was carried out analogously to that de-
scribed for Me3N complex under 3.7.1. Yield 4.31 g
(87%). Anal. Found: B, 4.11 (calc: 4.13%). 1H, 11B
NMR and IR-spectra were identical with the previously
published values [20].
3.7.3. LiCNÆMeCN
To crude LiCNÆ(0.05–0.1)THF (2.25 g) [43] was
added acetonitrile (60 ml) in portions and the mixture
was refluxed for 50 min. A microcrystalline solid
formed which was filtered off and extracted with the fil-
trate five times. Yield: 4.19 g. (The product slowly lost
MeCN at a rate of approximately 0.1 mol equivalent
per hour when kept in the nitrogen stream after dry-
ing). Anal. Found: Li, 9.30 (calc: 9.38%). 1H NMR
(D2O, d): 2.07 (CH3). 13C{1H} NMR(D2O, d): 122.07
(CH3), 167.98 (CN).
3.7.4. [Bu4N][Ag(CN)2]
Acetonitrile solution of [Bu4N]I (3.52 g, 9.53 mmol in
12.5 ml) was added to finely ground AgCN (2.55 g,
19.06 mmol) suspended in acetonitrile (7.5 ml). After
30 min stirring at room temperature the insoluble AgI
was filtered off and washed with acetonitrile (25 ml).
The filtrate was evaporated to dryness, the solid residue
was suspended in ether, filtered off and washed with
ether (2 · 5 ml). Yield 3.70 g (97%). 13C NMR (quanti-
tative) (CDCl3, d): 13.39 (4C, CH3), 19.36 (4C,
CH3CH2), 23.67 (4C, NCH2CH2), 58.57 (4C, NCH2),
143.78 (2C, CN).
3.7. Preparations of starting materials
4. Supplementary material
3.7.1. Me3NÆBH(Br)COOMe
Crystallographic data for the structural analysis has
been deposited with the Cambridge Crystallographic
Data Centre, CCDC No. 202629 for compound 1. Cop-
ies of this information may be obtained free of charge
from The Director, CCDC, 12 Union Road, Cambridge
CB2 1EZ, UK (fax: +44-1223-336033; e-mail: depos-
Me3NÆBH2COOH (5.14 g, 43.95 mmol) was dis-
solved in distilled methanol (85 ml). NBS (7.89 g,
44.33 mmol) was added to the solution in 15 portions.
The second portion was added 5 min after the first one
(ca. 5%) and subsequent portions were added when the

B. Gy}ori et al. / Journal of Organometallic Chemistry 689 (2004) 3567–3581
3581
}
[21] B. Gyori, Z. Berente, Inorg. Chem. 40 (2001) 1770.
it@ccdc.cam.ac.uk or www: http://www.ccdc.cam.
ac.uk).
}
´ ´ ´
[22] B. Gyori, Z. Berente, R. Kiraly, I. Lazar, J. Chem. Soc., Perkin
Trans. 1 (2002) 300.
[23] W.J. Mills, C.H. Sutton, E. Libby, L.J. Todd, Inorg. Chem. 29
(1990) 302.
Acknowledgements
[24] E.C. Evers, W.O. Freitag, W.A. Kriner, A.G. McDiarmid, S.
Sujishi, J. Inorg. Nucl. Chem. 13 (1960) 239.
[25] A. Meller, W. Maringgele, U. Sicker, J. Organomet. Chem. 141
(1977) 249.
This work was supported by the Hungarian Scientific
Research Found (OTKA T029060). I.L. thanks the
[26] L.E. Benjamin, D.A. Carvalho, S.F. Stafiej, E.A. Takacs, Inorg.
Chem. 9 (1970) 1844.
¨
Szechenyi Istvan Schoolarship (SzO 303/2003) for finan-
´
cial support.
´
[27] H.E. Zieger, W. Shiming, J. Org. Chem. 59 (1994) 3838.
[28] B.K. Wilk, Synth. Commun. 23 (1993) 2481.
[29] L. Sun, N. Tran, F. Tang, H. App, P. Hirth, G. McMahon, C.
Tang, J. Med. Chem. 41 (1998) 2588.
References
[30] M.C. Etter, Acc. Chem. Res. 23 (1990) 120;
J. Bernstein, R.E. Davis, L. Shimoni, N.L. Chang, Angew. Chem.
Int. Ed. Engl. 34 (1995) 1555.
[1] B.F. Spielvogel, M.K. Das, A.T. McPhail, K.D. Onan, I.H. Hall,
J. Am. Chem. Soc. 102 (1980) 6343.
´
[2] B. Gyo˜ri, Z. Berente, Z. Kovacs, Inorg. Chem. 37 (1998) 5131.
[31] F.H. Allen, O. Kennard, Chem. Design Autom. News 8 (1993) 1.
[32] J. Lipowski, R. Luboradzki, L. Stefaniak, J. Wojcik, J. Chem.
Cryst. 25 (1993) 299.
[3] I.H. Hall, C.O. Starnes, B.F. Spielvogel, P. Wisian-Neilson,
M.K. Das, L. Wojnowich, J. Pharm. Sci. 68 (1979) 685.
[4] I.H. Hall, C.O. Starnes, A.T. McPhail, P. Wisian-Neilson, M.K.
Das, F. Harchelroad Jr., B.F. Spielvogel, J. Pharm. Sci. 69 (1980)
1025.
[33] K.H. Scheller, R.B. Martin, Inorg. Chim. Acta 57 (1982)
227;
M.K. Das, P. Mukherjee, J. Chem. Res. (S) (1987) 368;
M.K. Das, P. Mukherjee, J. Chem. Res. (M) (1987) 2973.
[34] M.-A. Funke, H. Mayr, Chem. Eur. J. 3 (1997) 1214.
[35] J.L. Vidal, G.E. Ryschkewitsch, Inorg. Chem. 16 (1977) 1898.
[36] N.E. Miller, Inorg. Chem. 30 (1991) 2228.
[37] G.E. Harris, G.E. Ryschkewitsch, Proc. NOBCChE 22 (1995)
145, CA 124 (1996) 218656.
[5] M. Mittakanti, K.W. Morse, Inorg. Chem. 29 (1990) 554.
}
´ ´
[6] B. Gyori, Z. Berente, J. Emri, I. Lazar, Synthesis (1995) 191.
[7] M.K. Das, P. Mukherjee, S. Roy, Bull. Chem. Soc. Jpn. 63 (1990)
3658.
[8] S. Karthikeyan, A. Sood, J. Tomasz, B.F. Spielvogel, I.H. Hall,
Amino Acids 8 (1995) 323.
[9] V.M. Norwood III, K.W. Morse, Inorg. Chem. 27 (1988) 302.
[10] B.F. Spielvogel, A. Sood, I.H. Hall, WO Patent 93 09,123, 1993;
Chem. Abstr. 119 (1993) 203618f.
[38] H. No¨th, H. Beyer, Chem. Ber. 93 (1960) 928.
[39] T. Imamoto, H. Morishita, J. Am. Chem. Soc. 122 (2000)
6329.
[11] M.C. Miller III, A. Sood, B.F. Spielvogel, R.P. Shrewsbury, I.H.
Hall, Met. Based Drugs 3 (1996) 219.
[40] S. Herzog, J. Dehnert, K. Luhder, Tech. Inorg. Chem. 7 (1968)
119.
[12] I.H. Hall, A. Elkins, A. Sood, J. Tomasz, B.F. Spielvogel,
Anticancer Res. 16 (1996) 3709.
[41] B. Kemp, S. Kalbag, R.A. Geanangel, Inorg. Chem. 23 (1984)
3063.
[13] K.G. Rajendran, S.Y. Chen, A. Sood, B.F. Spielvogel, I.H. Hall,
Biomed. Pharmacother. 49 (1995) 131.
[42] V. Fedij, E.A. Lenoir III, M.J. Suto, J.R. Zeller, J. Wemple,
Tetrahedron: Asymmetry 5 (1994) 1131.
[14] M.C. Miller III, C.M. Woods, M.E. Murphy, A. Elkins, B.F.
Spielvogel, I.H. Hall, Biomed. Pharmacother. 52 (1998) 169.
[15] M. Takagaki, W. Powell, A. Sood, B.F. Spielvogel, N.S.
Hosmane, M. Kirihata, K. Ono, S.I. Masunaga, Y. Kinashi, S.I.
Miyatake, N. Hashimoto, Radiat. Res. 156 (2001) 118.
[16] B.F. Spielvogel, G. Rana, K. Vyakaranam, K. Grelck, K.E.
Dicke, B.D. Dolash, S.J. Li, C. Zheng, J.A. Maguire, M.
Takagaki, N.S. Hosmane, Collect. Czech. Chem. Commun. 67
(2002) 1095.
[43] K. Rossmanith, Monatsh. Chem. 96 (1965) 1690.
[44] M. Fieser, L.F. Fieser, Reagents for Organic Synthesis, vol. 5,
Wiley, New York, 1975, p. 568.
[45] A. Altomare, G. Cascarano, C. Giacovazzo, A. Guagliardi, J.
Appl. Crystallogr. 26 (1993) 343.
[46] G.M. Sheldrick, SHELXL-97, Universita¨t Go¨ttingen, Germany,
1997.
[47] L.J. Farrugia, WINGX-97 system, University of Glasgow, UK,
1996.
[17] C.H. Sutton, M.W. Baize, L.J. Todd, Inorg. Chem. 33 (1994)
4221, and references therein.
[48] A.E. Martell, R.M. Smith, Critical Stability Constants, Plenum
Press, New York, 1974–1989.
}
´ ´ ´
[18] B. Gyori, Z. Kovacs, J. Emri, I. Lazar, Inorg. Chim. Acta 218
[49] M.M. Irving, M.G. Milesand, L.D. Pettit, Anal. Chim. Acta 38
(1976) 475.
(1994) 21.
[19] J. Emri, Z. Kovacs, B. Gyo˜ri, Inorg. Chim. Acta 233 (1995) 119.
´
´
[20] B. Gyori, Z. Kovacs, J. Emri, Z. Berente, J. Organomet. Chem.
[50] L. Ze´ka´ny, I. Nagypa´l, in: D.J. Leggett (Ed.), Computational
Methods for Determination of Formation Constants, Plenum
Press, New York, 1985, p. 291.
}
484 (1994) 225.