Synthesis, spectroscopy and computational studies of some novel phosphorylated derivatives of quinoline-5,8-diones

Article abstract of DOI:10.1016/j.molstruc.2010.11.032

The neutral phosphorus nucleophiles such as R₂P(Y)M [1a-f, YO or lone pair; R = Ph, tBu, OCH₂C(CH₃)₂CH ₂O, PrⁱO, EtO or MeO; M = H or SiMe₃] allowed the radical addition to 2-met

Full text of DOI:10.1016/j.molstruc.2010.11.032

Journal of Molecular Structure 986 (2011) 39–48
Contents lists available at ScienceDirect
Journal of Molecular Structure
journal homepage: www.elsevier.com/locate/molstruc
Synthesis, spectroscopy and computational studies of some novel
phosphorylated derivatives of quinoline-5,8-diones
a
c
d
b
Jacek E. Nycz a, , Grzegorz Malecki , Lukasz Ponikiewski , Markus Leboschka , Maria Nowak ,
⇑
Joachim Kusz b
a Institute of Chemistry, University of Silesia, ul. Szkolna 9, PL-40007 Katowice, Poland
b Institute of Physics, University of Silesia, ul. Uniwersytecka 4, PL-40007 Katowice, Poland
c Department of Inorganic Chemistry, Faculty of Chemistry, Gdan´sk University of Technology, 11/12 G. Narutowicz St., PL-80233 Gdan´sk, Poland
d Institut für Anorganische Chemie, Universität Stuttgart, Pfaffenwaldring 55, 70550 Stuttgart, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
The neutral phosphorus nucleophiles such as R2P(@Y)M [1a–f, Y@O or lone pair; R = Ph, tBu, OCH2C(CH3)2
CH2O, PriO, EtO or MeO; M = H or SiMe3] allowed the radical addition to 2-methyl-5,8-dioxo-5,8-dihydro-
quinoline-7-amine (2a) and N-(2-methyl-5,8-dioxo-5,8-dihydroquinolin-7-yl)acetamide (2b) giving
exclusively O-phosphorylated products, i.e.: 7-acetylamino-5-hydroxy-2-methylquinolin-8-yl diph-
enylphosphinate (3a), 7-acetylamino-5-hydroxy-2-methylquinolin-8-yl-tbutylphenylphosphinate (3b),
7-acetylamino-5-hydroxy-2-methylquinolin-8-yl diisopropyl phosphate (3c), 7-acetylamino-5-hydroxy-
2-methylquinolin-8-yl diethyl phosphate (3d), N-{8-[(5,5-dimethyl-2-oxido-1,3,2-dioxaphosphinan-2-
yl)oxy]-5-hydroxy-2-methylquinolin-7-yl}acetamide (3e), 7-amino-5-hydroxy-2-methylquinolin-8-yl
dimethyl phosphate (3f) and 7-amino-5-hydroxy-2-methylquinolin-8-yl-tbutylphenylphosphinate (3g),
respectively, with high yield. All products were quantitatively prepared and characterized by microanaly-
sis, and multinuclear NMR spectroscopy. Seven of them, i.e.: 2a, 2b, 3a, 3b, 3d, 3e and 3g have been char-
acterized by single crystal X-ray diffraction method. The geometries of the studied compounds were
optimized in singlet states using the density functional theory (DFT) method with B3LYP functional. The
redox properties of 2a and 2b have been studied using cyclic voltammetry. The reduction corresponds to
the electrochemical behaviour of naturally occurring quinones.
Received 15 September 2010
Accepted 11 November 2010
Available online 25 November 2010
Keywords:
Quinoline
Quinone
Phosphorylation
Single-Electron Transfer (SET)
DFT
X-ray
Ó 2010 Elsevier B.V. All rights reserved.
1. Introduction
transfer reaction is part of an electron transport chain embedded
in the mitochondrial inner membranes [8].
The quinones are a class of compounds of great interest from
the synthetic, theoretical and applications point of view [1–3].
Among them, the 1-(7-amino-5,8-dioxo-5,8-dihydroquinolin-2-
yl)-4-methyl-9H-b-carboline-3-carboxylic acid (shortly named
lavendamycin) has attracted special attention due to its broad
spectrum antitumor, antibacterial and antiviral activity [4–7].
The redox properties of biologically important hydroquinone/
p-quinone systems are well known [3]. Three quinones, viz., coen-
zyme Q (ubiquinone), plastoquinone and menaquinone are among
the most important biological electron carriers. One-electron re-
duced forms, the semiquinone anion radicals, are directly involved
in electron transfer reactions within photosynthesis and respira-
tion. They are strongly affected by the surrounding medium, by
pH and temperature, particularly if the corresponding electron
Phosphorus plays a crucial role in the biochemistry of all living
beings. Incorporation of phosphorus containing groups into biolog-
ically active compounds or their structural analogues could be of
interest for at least two important reasons. First, 31P NMR spectros-
copy offers good opportunities to better understand the biological
profile of potential activity, and it has been established as a useful,
non-perturbing tool for probing effects at the phosphorus atoms.
The NMR properties of the phosphorus-31 nucleus are well suited
for in vivo NMR because of its natural abundance (100%), reason-
able sensitivity, and adequate but not excessive chemical shifts.
Changes in 31P chemical shifts can be helpful to understand several
kinds of interactions such as hydrogen bonding or protonation,
without introducing any modifications that could alter the struc-
ture. Secondly, some key metabolites, particularly those involved
in the production of energy, contain phosphorus and are present
in substantial concentrations in living tissues [9,10].
The biological activity of lavendamycin may be connected with
the presence of a quinoline-5,8-dione subunit. To the best of our
⇑
Corresponding author. Tel.: +48 323591206; fax: +48 322599978.
E-mail address: jnycz@us.edu.pl (J.E. Nycz).
0022-2860/$ - see front matter Ó 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.molstruc.2010.11.032

40
J.E. Nycz et al. / Journal of Molecular Structure 986 (2011) 39–48
knowledge only few examples of phosphorylated derivatives of
lavendomycin have been reported so far [11]. Surprisingly, there
is no example of a phosphorylated quinoline-5,8-dione unit re-
ported. In consideration of the present knowledge of the reactivity
of quinoline-5,8-diones, it is conceivable that further explorations
in this field may lead to new applications, not only in the area of
introducing a phosphorus containing moiety.
C–H distances of 0.95 Å or 0.98 Å and Uiso(H) values of 1.2Ueq(C),
or 1.5Ueq(C), respectively. H atoms, which take part in hydrogen
bond, were permitted to ride at the positions deduced from the dif-
ference maps with Uiso(H) equal to 1.2Ueq(C), 1.2Ueq(N) or
1.5Ueq(O).
For compound 3g the X-ray measurements were performed
with Oxford Diffraction Xcalibur Gemini A diffractometer with
0
graphite-monochromated Mo K
crystal was measured at 265 K. During the data collection
a
radiation (k = 0.71073 ÅA). The
scans
x
2. Experimental
were performed and absorption corrections were used. The struc-
ture was first solved using the direct method with SHELXS-97 soft-
ware [15] and then the solution was refined with SHELXL-97 [16].
All non-hydrogen atoms were refined with anisotropic tempera-
ture factors. All H atoms bound to C atoms were refined using a rid-
ing model with C–H distances of 0.95 Å or 0.98 Å and Uiso(H)
values of 1.2Ueq(C), or 1.5Ueq(C), respectively. H atoms, which
take part in hydrogen bonds, were permitted to ride at the posi-
tions deduced from the difference maps with Uiso(H) equal to
1.2Ueq(C), 1.2Ueq(N) or 1.5Ueq(O).
2.1. General
All experiments were carried out in an atmosphere of dry argon,
using Schlenk techniques. Solvents were dried by usual methods
(benzene and THF over benzophenone ketyl, CHCl3 and CH2Cl2 over
P4O10, hexane over sodium–potassium alloy), distilled and de-
gassed. Chromatography was carried out on Silica Gel 60 (0.15–
0.3 mm) Machery Nagel. NMR spectra were obtained with Bruker
Avance 400 operating at 400.13 MHz (1H), 161.9 MHz (31P) and
100.4 MHz (13C) at 30 °C; chemical shifts referenced to ext. TMS
2.3. Synthesis of 3a, 3b and 3g
(1H, 13C), 85% H3PO4 = 40.480747 MHz, 31P); positive signs of
(N
chemical shifts denote shifts to lower frequencies, coupling con-
stants are given as absolute values. Elemental analysis: Perkin–
Elmer Analyzer 240. Mass spectra were obtained with a MASPEC
II system [II32/99D9], Varian MAT 711 and Finnigan MAT 95 (Finn-
igan MAT GmbH, Brema); in EI mode (70 eV) and, where necessary,
LSI technique was applied. EPR spectra in the X-band were re-
corded with a Bruker System EMX. Cyclic voltammetry was carried
out in 0.1 M Bu4NPF6 solutions using a three-electrode configura-
tion (glassy-carbon working electrode, Pt counter electrode, Ag
wire as pseudoreference) and a PAR 273 potentiostat and function
generator. The ferrocene/ferrocenium (Fc/Fc+) couple served as
internal reference. Melting points were taken in sealed capillaries
and are uncorrected. Compounds 2a, 2b [12] and 1a [13] were syn-
thesized according to procedures described in the literature. Re-
agents 1d and 1f were purchased from Aldrich, and 1c from
Acros Organics and were distilled before used.
2b (5 mmol) in THF (100 mL) was added to the solution of 1a or
1b (5 mmol), respectively, in THF (25 mL) at room temperature.
After complete addition the reaction mixture was heated at 60 °C
for 16 h. Next, a sample of the reaction mixture was taken,
DMSO-d6 was added, and the 31P{1H} NMR spectrum was re-
corded. Subsequently dry air was passed through the reaction mix-
ture at room temperature, for 1a. The volatiles were evaporated
and the residue was purified by chromatography and
crystallisation:
3a (CH2Cl2:MeOH = 5:1) 0.259 g (0.6 mmol, 12%); mp = 220–
221 °C (MeOH); 1H NMR (400.13 MHz, DMSO-d6, 30 °C) d = 2.11
(s, 3H, CH3), 2.64 (s, 3H, CH3), 7.23 (d, JH,H = 8.5 Hz, 1H, quin-H),
7.42 (ddt, JP,H = 7.3 Hz, JH,H = 3.7 Hz, JH,H = 1.2 Hz, 4H, m-arom-H),
7.49 (ddd, JP,H = 12.5 Hz, JH,H = 6.9 Hz, JH,H = 1.5 Hz, 2H, p-arom-H),
7.57 (s, 1H, quin-H), 8.10 (dd, JP,H = 12.5 Hz, JH,H = 7.1 Hz, 4H, o-
arom-H), 8.12 (d, JH,H = 8.5 Hz, 1H, quin-H), 9.80 (s, 1H, OH),
10.49 (s, 1H, NH); 13C{1H} NMR (100.4 MHz, DMSO-d6, 30 °C)
d = 24.59, 25.16, 103.00, 115.45, 120.29, 128.69 (d, JP,C = 13.3 Hz,
P–Ph), 129.73, 131.08, 131.12, 131.23 (d, JP,C = 3.4 Hz, P–Ph),
132.36 (d, JP,C = 10.6 Hz, P–Ph), 133.16 (d, JP,C = 2.8 Hz, P–Ph),
150.25 (d, JP,C = 1.5 Hz), 159.27, 168.69; 31P{1H} NMR (161.9 MHz,
DMSO-d6, 30 °C) d = 36.28; MS: (LSI) (M+H)+ 433; HRMS: m/z Calcd
for C24H22N2O4P (M+H)+: 433.13176 Found 433.13172; E.A. [Found
C, 64.67; H, 5.50; N, 6.03; C24H21N2O4P ꢀ CH3OH requires C,
64.65%; H, 5.43%; N, 6.03%].
3b (CH2Cl2:MeOH = 5:1) 1.112 g (2.7 mmol, 54%); mp = 265–
266 °C (MeOH); 1H NMR (400.13 MHz, DMSO-d6, 30 °C) d = 1.30
(d, JP,H = 16.3 Hz, 9H, tBu), 2.09 (s, 3H, CH3), 2.64 (s, 3H, CH3),
7.30 (d, JH,H = 8.5 Hz, 1H, quin-H), 7.43 (dt, JP,H = 7.6 Hz,
JH,H = 3.5 Hz, 2H, m-arom-H), 7.54 (dt, JP,H = 6.7 Hz, JH,H = 1.1 Hz,
1H, p-arom-H), 7.62 (s, 1H, quin-H), 8.08 (m, 2H, o-arom-H), 8.25
(d, JH,H = 8.5 Hz, 1H, quin-H), 10.43 (s, 1H, OH), 10.53 (s, 1H, NH);
13C{1H} NMR (100.4 MHz, DMSO-d6, 30 °C) d = 23.91, 24.04,
24.70, 33.34 (d, JP,C = 98.0 Hz, P-tBu), 102.84, 115.17, 119.89,
125.77, 126.92, 127.85 (d, JP,C = 12.0 Hz, P–Ph), 130.04 (d,
JP,C = 3.0 Hz, P–Ph), 130.74, 132.72 (d, JP,C = 2.0 Hz, P–Ph), 133.27
(d, JP,C = 9.7 Hz, P–Ph), 140.70 (d, JP,C = 3.2 Hz), 149.48, 158.71,
168.03; 31P{1H} NMR (161.9 MHz, DMSO-d6, 30 °C) d = 56.10;
MS: (LSI) (M+H)+ 413; HRMS: m/z Calcd for C22H26N2O4P (M+H)+:
413.16302 Found 413.16343; E.A. [Found C, 63.97; H, 6.12; N,
6.78; C22H25N2O4P requires C, 64.07%; H, 6.11%; N, 6.79%].
3g (MeOH) 1.166 g (3.15 mmol, 63%); mp = 252 °C (MeOH); 1H
NMR (400.13 MHz, DMSO-d6, 30 °C) d = 1.31 (d, JP,H = 16.9 Hz, 9H,
tBu), 2.78 (s, 3H, CH3), 6.71 (s, 1H, quin-H), 7.27 (d, JH,H = 8.2 Hz,
2.2. X-ray crystallographic studies
Data for compounds 2a, 2b and 3a, 3b, 3d were collected on a
KUMA KM4 diffractometer with graphite-monochromated Mo K
a
radiation using Sapphire-2 CCD detector. The apparatus was
equipped with an open flow thermostat (Oxford Cryosystems) with
enabled experiments at 120 K. The stream of nitrogen also pre-
vented the specimen from contact with moisture and oxygen.
The structure were solved by direct methods and refined by full-
matrix least-squares on F2 (all data) using the SHELXTL program
package [14]. All non-hydrogen atoms were refined anisotropi-
cally. Hydrogen atoms were refined in geometrically idealised
positions with isotropic temperature factors 1.2 times the equiva-
lent isotropic temperature factors Ueq of their attached atoms (1.5
for CH3 groups).
For compounds 3b and 3e the X-ray measurements were per-
formed with Oxford Diffraction Xcalibur diffractometer equipped
with a Sapphire3 CCD detector and graphite-monochromated
0
Mo Ka radiation (k = 0.71073 ÅA). The crystal was cooled down to
100 K by a cold dry nitrogen gas stream (Oxford Cryosystems
equipment). The temperature stability of the measurement was
1 K. During the data collection
x scans were performed. No
absorption corrections were used. The structure was first solved
using the direct method with SHELXS-97 software [15] and then
the solution was refined with SHELXL-97 [16]. All non-hydrogen
atoms were refined with anisotropic temperature factors. All H
atoms bound to C atoms were refined using a riding model with

J.E. Nycz et al. / Journal of Molecular Structure 986 (2011) 39–48
41
1H, quin-H), 7.55 (dt, JP,H = 3.4 Hz, JH,H = 7.6 Hz, 2H, P–Ph), 7.66 (t,
JH,H = 7.2 Hz, 1H, m-Ph), 7.92 (m, 2H, o-Ph), 8.62 (bs, 1H, quin-H),
11.42 (s, 1H, OH), 14.82 (s, 1H, NH); 13C{1H} NMR (100.4 MHz,
DMSO-d6, 30 °C) d = 20.54, 23.86, 33.13 (d, JP,C = 95.7 Hz, P-tBu),
100.87, 113.15, 115.20, 116.90, 123.53, 124.62, 128.69 (d,
JP,C = 12.0 Hz, P–Ph), 132.69 (d, JP,C = 9.9 Hz, P–Ph), 133.70,
139.57, 147.36, 152.27, 154.71; 31P{1H} NMR (161.9 MHz, DMSO-
d6, 30 °C) d = 62.96; MS: (LSI) (M+H)+ 371; HRMS: m/z Calcd for
2CH3), 2.12 (s, 3H, CH3), 2.62 (s, 3H, CH3), 4.36 (m, 4H, 2CH2),
7.31 (d, JH,H = 8.5 Hz, 1H, quin-H), 7.49 (s, 1H, quin-H), 8.32 (d,
JH,H = 8.5 Hz, 1H, quin-H), 9.51 (s, 1H, OH), 10.49 (s, 1H, NH);
13C{1H} NMR (100.4 MHz, DMSO-d6, 30 °C) d = 15.81 (d,
JP,C = 6.9 Hz), 23.74, 24.66, 64.25 (d, JP,C = 6.2 Hz), 103.73, 115.44,
120.10, 127.72 (d, JP,C = 8.7 Hz), 128.40 (d, JP,C = 7.7 Hz), 130.72,
140.89 (d, JP,C = 2.9 Hz), 149.74 (d, JP,C = 1.8 Hz), 158.89, 168.18;
31P{1H} NMR (161.9 MHz, DMSO-d6, 30 °C) d = ꢁ4.78; MS: (LSI)
(M+H)+ 369; HRMS: m/z Calcd for C16H22N2O6P (M+H)+:
369.12179 Found 369.12155; E.A. [Found C, 51.81; H, 5.71; N,
7.53; C16H21N2O6P requires C, 52.17; H, 5.75; N, 7.61%].
C
20H24N2O3P (M+H)+: 371.16302 Found 371.16343; E.A. [Found
C, 64.56; H, 6.11; N, 7.79; C20H23N2O3P requires C, 64.86%; H,
6.26%; N, 7.56%].
3e (THF) 1.349 g (3.55 mmol, 71%); mp = 245 °C (THF); 1H NMR
(400.13 MHz, DMSO-d6, 30 °C) d = 0.92 (s, 3H, CH3), 1.23 (s, 3H,
CH3), 2.11 (s, 3H, CH3), 2.66 (s, 3H, CH3), 4.05 (dd, JH,Hgem = 10.7 Hz,
JP,H = 23.5 Hz, 2Ha, CH2), 4.90 (d, JH,Hgem = 10.4 Hz, 2He, CH2), 7.35
(d, JH,H = 8.5 Hz, 1H, quin-H), 7.59 (s, 1H, quin-H), 8.34 (d,
JH,H = 8.5 Hz, 1H, quin-H), 9.77 (s, 1H, OH), 10.62 (s, 1H, NH);
13C{1H} NMR (100.4 MHz, DMSO-d6, 30 °C) d = 19.19, 20.95,
23.95, 24.81, 31.73 (d, JP,C = 6.0 Hz), 78.26 (d, JP,C = 6.9 Hz),
103.46, 115.40, 120.23, 126.77 (d, JP,C = 7.4 Hz), 130.70 (d,
JP,C = 3.4 Hz), 131.02, 140.65 (d, JP,C = 3.4 Hz), 150.09 (d,
JP,C = 1.7 Hz), 159.14, 168.12; 31P{1H} NMR (161.9 MHz, DMSO-d6,
30 °C) d = ꢁ11.61; 31P NMR (161.9 MHz, DMSO-d6, 30 °C)
d = ꢁ11.61 (t, JP,H = 23.4 Hz); MS: (LSI) (M+H)+ 381; HRMS: m/z
Calcd for C17H22N2O6P (M+H)+: 381.12155 Found 381.12069, E.A.
[Found C, 53.84; H, 5.63; N, 7.28; C17H21N2O6P requires C, 53.69;
H, 5.57; N, 7.37%].
2.4. Synthesis of 3c, 3d, 3e and 3f
2a or 2b (5 mmol) in THF (100 mL) was added to the solution of
1c, 1d, 1e or 1f (20 mmol), respectively, in THF (25 mL) at room
temperature. After complete addition the reaction mixture was
heated under reflux for 16 h. Next, a sample of the reaction mixture
was taken, DMSO-d6 was added, and the 31P{1H} NMR spectrum
was recorded. The volatiles were evaporated and the residue was
purified by chromatography and crystallisation:
3c (MeOH) 1.287 g (3.25 mmol, 65%); mp = 183 °C (MeOH); 1H
NMR (400.13 MHz, DMSO-d6, 30 °C) d = 1.19 (d, JH,H = 6.1 Hz, 6H,
2CH3), 1.30 (d, JH,H = 6.1 Hz, 6H, 2CH3), 2.12 (s, 3H, CH3), 2.62 (s,
3H, CH3), 4.99 (m, 2H, 2CH), 7.31 (d, JH,H = 8.5 Hz, 1H, quin-H),
7.59 (s, 1H, quin-H), 8.31 (d, JH,H = 8.5 Hz, 1H, quin-H), 9.42 (s,
1H, OH), 10.47 (s, 1H, NH); 13C{1H} NMR (100.4 MHz, DMSO-d6,
30 °C) d = 23.06 (d, JP,C = 5.6 Hz), 23.28 (d, JP,C = 4.6 Hz), 24.01,
24.63, 72.91 (d, JP,C = 6.2 Hz), 103.02, 115.24, 119.24, 127.72 (d,
JP,C = 8.7 Hz), 130.38 (d, JP,C = 4.0 Hz), 130.66, 140.96 (d,
JP,C = 3.1 Hz), 149.72 (d, JP,C = 1.9 Hz), 158.84, 168.12; 31P{1H}
NMR (161.9 MHz, DMSO-d6, 30 °C) d = ꢁ5.90; MS: (LSI) (M+H)+
397; HRMS: m/z Calcd for C18H26N2O6P (M+H)+: 397.15194 Found
397.15211; E.A. [Found C, 54.50; H, 6.37; N, 7.05; C18H25N2O6P re-
quires C, 54.54%; H, 6.36%; N, 7.07%].
3f (MeOH) 1.014 g (3.40 mmol, 68%); mp = 226 °C (MeOH); 1H
NMR (400.13 MHz, DMSO-d6, 30 °C) d = 2.68 (s, 3H, CH3), 3.52 (d,
JP,H = 11.1 Hz, 6H, CH3), 6.64 (s, 1H, quin-H), 7.15 (d, JH,H = 8.2 Hz,
1H, quin-H), 8.57 (d, JH,H = 8.2 Hz, 1H, quin-H), 10.94 (s, 1H, OH);
13C{1H} NMR (100.4 MHz, DMSO-d6, 45 °C) d = 19.75, 52.68 (d,
JP,C = 6.3 Hz), 100.59, 112.85, 113.84, 117.54 (d, JP,C = 7.7 Hz),
132.73 (d, JP,C = 2.1 Hz), 138.50, 146.62 (d, JP,C = 4.2 Hz), 150.70,
153.68; 31P{1H} NMR (161.9 MHz, DMSO-d6, 30 °C) d = ꢁ0.20; 31P
NMR (161.9 MHz, DMSO-d6, 30 °C) d = ꢁ0.20 (q, JP,H = 11.1 Hz);
MS: (LSI) (M+H)+ 299; HRMS: m/z Calcd for C12H16N2O5P (M+H)+:
3d (MeOH) 1.049 g (2.85 mmol, 57%); mp = 216 °C (MeOH); 1H
NMR (400.13 MHz, DMSO-d6, 30 °C) d = 1.26 (t, JH,H = 7.0 Hz, 6H,
Table 1
Selected bond lengths for 2a. Symmetry codes: (i) x + 1/2, ꢁy + 3/2, z + 1/2; (ii) ꢁx + 3/
2, y ꢁ 1/2, ꢁz + 1/2.
Exp.
Calc.
Bond lengths (Å)
C4–O1
C7–O2
1.233(2)
1.213(19)
1.341(2)
1.232
1.221
1.339
C14–O3
C17–O4
C16–N4
1.249(12)
1.250(19)
1.319(2)
C6–N2
Hydrogen-bond geometry (Å, °)
D–Hꢂ ꢂ ꢂA
D–H
0.99(3)
0.88
0.88
0.88
Hꢂ ꢂ ꢂA
Dꢂ ꢂ ꢂA
D–Hꢂ ꢂ ꢂA
176.0(3)
143.3(2)
155.2(3)
131.4(3)
142.3(4)
O5–H5ꢂ ꢂ ꢂO3
N2–H2Aꢂ ꢂ ꢂN3
N4–H4Aꢂ ꢂ ꢂN1i
N4–H4Aꢂ ꢂ ꢂO2i
N4–H4Bꢂ ꢂ ꢂO5ii
1.81(3)
2.30(1)
2.15(6)
2.63(3)
2.17(3)
2.8018(19)
3.0468(19)
2.971(2)
3.274(2)
2.9125(19)
0.88
Table 2
Selected bond lengths for 2b. Symmetry code: (i) ꢁx + 4, ꢁy + 1, ꢁz + 1.
Exp.
Calc.
Exp.
Calc.
Bond lengths (Å)
C5–O1
C8–O2
1.222(2)
1.214(2)
1.378(2)
1.228
1.222
1.381
N2–C11
C11–O3
1.386(2)
1.211(2)
1.388
1.218
C7–N2
Hydrogen-bond geometry (Å, °)
D–Hꢂ ꢂ ꢂA
D–H
0.82(2)
0.82(2)
Hꢂ ꢂ ꢂA
Dꢂ ꢂ ꢂA
D–Hꢂ ꢂ ꢂA
N2–H5ꢂ ꢂ ꢂO2A
2.23(2)
2.31(2)
2.6360(18)
3.1193(18)
110.8(17)
167.7(19)
N2A-H5Aꢂ ꢂ ꢂO2i
Fig. 1. Molecular structure diagrams of 2a (upper) and 2b (below). The thermal
ellipsoids are drawn at the 50% probability level.

42
J.E. Nycz et al. / Journal of Molecular Structure 986 (2011) 39–48
5,8-dione unit. The quinoline-5,8-dione derivatives 2-methyl-5,8-
dioxo-5,8-dihydroquinoline-7-amine (2a) and N-(2-methyl-5,8-di-
oxo-5,8-dihydroquinolin-7-yl)acetamide (2b) were chosen as
Scheme 2. Reagents: (i) P(OR)3; (ii) (RO)2P(@O)H; (iii) PPh3.
Fig. 2. Molecular structure of the ester 3a (upper) and the interaction diagram
between two molecules of 3a and methanol (below). The thermal ellipsoids are
drawn at 50% probability level. All hydrogen atoms other than H1, H1A and H1B
have been omitted for clarity.
299.07968 Found 299.07964; E.A. [Found C, 48.19; H, 5.17; N, 9.27;
C12H15N2O5P requires C, 48.33; H, 5.07; N, 9.39%].
3. Results and discussion
Our present study describes electron transfer properties of the
two structurally characterised compounds possesses quinoline-
Scheme 3. Synthesis of 3.
Scheme 1. Reaction of 1a with 2b. Proposal of mechanism.

J.E. Nycz et al. / Journal of Molecular Structure 986 (2011) 39–48
43
model compounds and characterised by crystal structure analysis
(Fig. 1; Tables 1 and 2).
also explain the differences between the C@O distances [C7–O2
1.173(12) Å and C4–O1 1.210(12) Å].
Compound 2a crystallises with methanol in space group P21/
n (No. 14), compound 2b in P-1 (No. 2). The molecular ring sys-
tems are both essentially planar. The packing of the molecules
The phosphorylation of lavendomycin and its structural build-
ing blocks is still not fully explored. To increase accessibility, the
reaction between neutral silyl derivative of phosphorus nucleo-
phile, i.e. Ph2P–SiMe3 (1a) and 2b was chosen. The reaction
between neutral silyl derivatives of phosphorus nucleophiles are
convenient reagents for creating C–P bonds during addition into
carbonyl compounds [17,18]. In constitution of 2b there are three
possible centres of attack in the quinolidione ring: (a) the carbonyl
carbon atoms (b) the carbonyl oxygen atoms, and (c) the activated
carbon–carbon double bond. Additionally, there are two centres in
its acyl group (a) the carbonyl carbon atom, and (b) the carbonyl
oxygen atoms.
2a and 2b in the structure is stabilised by
p–p stacking interac-
tions. Both compounds co-crystallised with CH3OH which is
associated through strong C–O–Hꢂ ꢂ ꢂO(C) hydrogen bonds. The
molecules of 2a are linked by weak bifurcated N2–H2Aꢂ ꢂ ꢂN3,
N1–H4AAꢂ ꢂ ꢂN4A and O5–H5ꢂ ꢂ ꢂO3 hydrogen bonds, the molecules
of 2b by strong N2–H2Bꢂ ꢂ ꢂO2A(C) and N2A–H2AB–O2(C) hydro-
gen bonds.
Compared to 2a, the C5–C8 distances are longer in 2b which
could be explained by the resonance structures in 2a which can
The reaction of 1a with 2b gave a complex reaction mixture.
Among the products we isolated and structurally characterised
the radical addition product 3a, i.e. compound with the construc-
tion P–O–C. To the best of our knowledge this is the first example
of the radical addition of silylated phosphane to carbonyl com-
pound (Fig. 2).
ꢀ
Compound 3a crystallises in space group P1 (No. 2). The quino-
line rings are practically planar. The packing of molecules in the
structure is stabilised by N–H1Aꢂ ꢂ ꢂO4(P) hydrogen bonds. The
compound co-crystallised with CH3OH in a dimer form which is
bound by (P)O4ꢂ ꢂ ꢂH1B–O5, (C)O1–H1ꢂ ꢂ ꢂO5A and N2–H1A–O4
hydrogen bonds.
A possible explanation of the formation of final product 3a is a
transfer of electron (SET) from the nucleophile 1a to 2b followed by
the migration of the silyl group from cation radical of 1a to the
oxygen atom of anion radical of 2b, giving the O- and P-centered
radicals (Scheme 1). Next, phosphinyl and oxygen radicals react
followed by oxygenation to 3a. Possibly, the process occurs with-
out formation of completely free radical intermediates, because
tetraphenyldiphosphane was not detected in the reaction mixture,
using 31P NMR techniques. The redox addition approach presented
here provides a direct route to a new class of compounds which are
potentially useful from a biological, synthetic and coordination
point of view.
However the reaction of quinones with tertiary phosphanes and
phosphites is well reported in the literature and provide to variety
of interesting products [19–24]. Ramirez et al. [19,20,22–25]
reported the reaction of chloranil with a range of phosphorus com-
pounds including trialkil phosphites, dialkil phosphates and tri-
phenylphosphane to give transient red solutions from which the
Fig. 3. Molecular structure diagram of 3b. The thermal ellipsoids are drawn at the
50% probability level, for the sake of clarity, hydrogen atoms not involved in
hydrogen bonds were omitted.
Fig. 4. Molecular structure diagram of 3d. All hydrogen atoms other than H2 and
H2B have been omitted for clarity. The thermal ellipsoids are drawn at the 50%
probability level.

44
J.E. Nycz et al. / Journal of Molecular Structure 986 (2011) 39–48
phenomenon would account for the appearance of deep-red colour
in the solution.
Due to reactive phosphorus–hydrogen bonds the secondary
phosphanes, their oxides and dialkyl (diaryl) phosphates are
widely used reagents and synthetic intermediates [25–27]. Acid-
catalyzed addition reactions into the carbonyl are presumed to
proceed by nucleophilic attack of the phosphorus atom at the
substrate which was previously activated by the acid, followed
by subsequent deprotonation. This type of catalysis occurs because
the conjugate acid of the carbonyl compound is much more elec-
trophilic than the neutral molecule. To compare the reactivity of
various three-coordinate phosphorus nucleophiles towards 2a
and 2b a reaction between 1b with 2b and 2a was performed. 1b
was used in the reaction with 2a and 2b in excess. The transforma-
tion leads to 3b and 3g, respectively with high yield (Scheme 3).
The secondary phosphanes and their oxides demonstrate higher
nucleophilicity than dialkyl phosphates or analogue compounds,
which possess alkoxy or aryloxy substituents at the phosphorus
atom. Because of this, commercially available dialkyl phosphates
1c, 1d, 1e and 1f were used in the reaction with 2a and 2b in ex-
cess. The conversion was unambiguous and gave a radical addition
products 3c, 3d, 3e and 3f with high yield (Scheme 3; Figs. 3–6).
3b crystallises in space group P21/c (No. 14). The quinoline
rings are practically planar. The packing of molecules in the struc-
Fig. 5. Molecular structure of the ester 3e. The thermal ellipsoids are drawn at 50%
probability level. All hydrogen atoms other than H1A, H2A, H1AA and H2AA have
been omitted for clarity.
quasi-phosphonium salt is obtained in the case of a phosphane,
and the corresponding alkylated quinol phosphates in the case of
alkyl phosphites, and hydroxyarylalkylphosphates in the case of
dialkil phosphates. They reported no chlorinated substitution
products (Scheme 2).
The mechanism of these reactions have been investigated in
some details, and the transient red colour attributed to free
radicals intermediates detected by the EPR spectrum. Lucken [21]
has shown that the spectrum is typical of that of a semiqui-
none. This may be formed of a charge-transfer complex. The
ture is stabilised by
p
–
p
stacking interactions and by O1–H1ꢂ ꢂ ꢂO3,
N2–H1Aꢂ ꢂ ꢂO4 and C10–H10Bꢂ ꢂ ꢂO3 hydrogen bonds.
3d crystallises in space group P-1 (No. 12). The quinoline rings
are practically planar. The packing of molecules in the structure is
stabilised by O2–H2ꢂ ꢂ ꢂO3, N2–H2Bꢂ ꢂ ꢂO1 and N2–H2Bꢂ ꢂ ꢂO5 hydro-
gen bonds.
3e crystallises in space group P2(1)/c (No. 14). The quinoline
rings is practically planar. The molecular structure is stabilized
by an intramolecular N–Hꢂ ꢂ ꢂO hydrogen bond. In addition, the
Fig. 6. Molecular structure of the 3g. The thermal ellipsoids are drawn at 50% probability level.

J.E. Nycz et al. / Journal of Molecular Structure 986 (2011) 39–48
45
Table 3
molecules are linked by intermolecular O–Hꢂ ꢂ ꢂO hydrogen bonds,
Selected bond lengths and angles for 3a. Symmetry code: (i) ꢁx + 2, ꢁy, ꢁz + 1.
the symmetry code is ꢁx, 1 ꢁ y, 1 ꢁ z.
3g crystallises in space group P2(1)/c (No. 14). The quinoline
rings is practically planar. The packing of molecules in the struc-
ture is stabilised by p–p stacking interactions and by an intramo-
lecular N–Hꢂ ꢂ ꢂO hydrogen bond. In addition, the molecules are
linked by intermolecular C–Hꢂ ꢂ ꢂO hydrogen bonds, the symmetry
code is ꢁx, 1 ꢁ y, 1 ꢁ z.
Exp.
Calc.
Exp.
Calc.
Bond lengths (Å)
P1–O4
P1–O2
O1–C4
O2–C7
1.492(14)
1.598(13)
1.359(2)
1.404(2)
1.222(2)
1.414(2)
1.504
1.636
1.366
1.391
1.225
1.398
N2–C11
C4–C5
C5–C6
C6–C7
C8–C9
1.362(2)
1.374(3)
1.416(3)
1.372(3)
1.418(3)
1.367
1.375
1.422
1.396
1.429
O3–C11
N2–C6
The neutral phosphorus nucleophiles 1a, 1b, 1c, 1d, 1e and 1f
allowed the radical addition to 2a and 2b giving exclusively
O(2)-phosphorylated products 3a, 3b, 3c, 3d, 3e, 3f and 3g, respec-
tively, with high yield. The smaller yield of 3a could be explained
by hydrolysis during isolation (chromatography). On the 31P NMR
spectra of the reaction mixtures, in all cases no signal from prod-
ucts with a P–P bond, such as the diphosphanes, diphosphane diox-
ides or hypophosphoric acid esters was detected. Presented here
radical addition reaction is chemoselective. We do not isolated re-
versed products, i.e. O1(C4)-phosphorylation, and addition prod-
ucts into acyl group, or into the activated carbon–carbon double
bond (Michael type products). All results support presents a
charge-transfer complex, rather than free radicals.
Bond angles (°)
Exp.
Calc.
Exp.
110.86(8)
N2–C11–O3 123.26(18) 124.80
Calc.
113.72
C6–C7–O2
C6–N2–C11
C7–O2–P1
120.41(16) 121.42
127.44(17) 128.64
124.09(11) 128.88
O2–P1–O4
C5–C4–O1
122.59(17) 122.40
Hydrogen-bond geometry (Å, °)
D–Hꢂ ꢂ ꢂA
D–H
Hꢂ ꢂ ꢂA
Dꢂ ꢂ ꢂA
D–Hꢂ ꢂ ꢂA
175(3)
163(2)
171(3)
O1–H1ꢂ ꢂ ꢂO5Ai 0.89(3)
1.78(3) 2.670(2)
2.13(3) 2.964(2)
1.95(3) 2.735(2)
N2–H1Aꢂ ꢂ ꢂO4
O5–H1Bꢂ ꢂ ꢂO4
0.86(2)
0.80(3)
3.2. DFT calculations
3.1. Cyclic voltammetry
In order to get further insight into the electronic structures and
bonding properties of the compounds, the DFT calculations were
carried out. First the geometries of the compounds were optimized
in singlet states using the DFT method with a B3LYP functional and
next the electronic structures of the compounds were calculated.
In general, the predicted bond lengths and angles are in a very good
agreement with the values based on the X-ray crystal structure
data, and the general trends observed in the experimental data
are well reproduced in the calculations as can be seen in Tables
1–7. Among the studied compounds in the first group belong qui-
none 2a and 2b. The natural atomic charges on the heteroatoms of
these compounds are taken from the Natural Population Analysis
and are close to each other; 2a: N(quinoline) ꢁ0.42, N(amine)
ꢁ0.79, O (1) ꢁ0.58, O(2) ꢁ0.51; 2b: N(quinoline) ꢁ0.42; N(amide)
ꢁ0.63, O(1) ꢁ0.53, O(2) ꢁ0.52. The atomic charge calculations can
give a feature for the relocation of the electron density of the com-
pounds, but the local concentration and local depletion of electron
charge density allow us to determine whether the nucleophile or
electrophile can be attracted. Since the electron distribution is
not apparent from the partial atomic charges, Fig. 8 shows the plots
of the electrostatic potentials for the compounds. The isoelectronic
contours are plotted at 0.05 a.u. (31 kcal/mol). The colour code of
Described here the redox properties of 2a and 2b provided
opportunities for advancing electroanalytical measurements. The
cyclic voltammetry studies show similar redox properties for both
2a and 2b (Fig. 7). They undergo reversible one-electron reductions
at ꢁ1.06 V (vs. Fc/Fc+) (2b) and ꢁ1.20 V (vs. Fc/Fc+) (2a). The sec-
ond reduction at ꢁ1.66 V is nearly reversible only for 2b. The sec-
ond reduction for 2a at around ꢁ1.79 V (vs. Fc/Fc+) is irreversible at
25 °C, at lower temperature (ꢁ60 °C) it becomes more reversible.
The first reduction potential of 2a is shifted to a more negative va-
lue in comparison with that of 2b which suggests that this precur-
sor is a poorer electron acceptor and that the electron-withdrawing
acyl group makes 2b a better electron acceptor.
The first reduction potential corresponds with the electrochem-
ical behaviour of natural occurring quinones [1–3,28]. In a broader
sense it also highlights the possible role of electron transfer in the
biological activity of naturally occurring quinolinediones such as
lavendamycin [8].
Table 4
Selected bond lengths and angles for 3b. Symmetry code: (i) ꢁx, ꢁy + 1, ꢁz; (ii) ꢁx,
ꢁy + 2, ꢁz.
Exp.
Calc.
Exp.
Calc.
Bond lengths (Å)
P1–O4
P1–O2
O1–C4
O2–C7
O3–C11
N2–C6
N2–C11
1.487 (6)
1.604 (6)
1.508
1.638
C11–C12
C4–C5
C5–C6
C6–C7
C7–C8
C8–C9
C1–C10
1.499(13) 1.521
1.370(12) 1.375
1.417(12) 1.422
1.379(12) 1.398
1.423(11) 1.422
1.415(12) 1.429
1.500(13) 1.510
1.357 (10) 1.366
1.399 (10) 1.392
1.236 (11) 1.225
1.418 (11) 1.400
1.351 (11) 1.381
Bond angles (°)
Exp.
Calc.
Exp.
Calc.
O4–P1–O2
C7–O2–P1
113.47 (3) 113.14
126.92 (5) 130.17
C6–C7–O2
O1–C4–C5
121.83 (7) 121.99
123.44 (8) 122.44
Hydrogen-bond geometry (Å, °)
D–Hꢂ ꢂ ꢂA
D–H
Hꢂ ꢂ ꢂA
Dꢂ ꢂ ꢂA
D–Hꢂ ꢂ ꢂA
O1–H1ꢂ ꢂ ꢂO3i
N2–H1Aꢂ ꢂ ꢂO4
0.853 (14) 1.818 (14) 2.6692 (10) 175.2 (13)
0.890 (12) 1.897 (12) 2.7512 (10) 160.3 (11)
Fig. 7. Cyclic voltammetry of 2b in CH2Cl2 (0.1 M n-Bu4NBF4 as electrolyte; Ag/Ag+
reference electrode, 20 °C, Fc/Fc+ (right) as internal standard).
C10–H10Bꢂ ꢂ ꢂO3ii 0.936 (14) 2.471 (14) 3.3928 (12) 168.4 (11)

46
J.E. Nycz et al. / Journal of Molecular Structure 986 (2011) 39–48
Table 5
Selected bond lengths and angles for 3d. Symmetry code: (i) ꢁx + 1, ꢁy ꢁ 1, ꢁz + 1;
(ii) ꢁx, ꢁy, ꢁz + 1.
Exp.
Calc.
Exp.
Calc.
Bond lengths (Å)
P1–O4
P1–O1
O2–C4
O1–C7
O3–C11
N2–C6
N2–C11
1.460(2)
1.556(2)
1.354(3)
1.404(3)
1.230(3)
1.424(3)
1.353(3)
1.478
1.624
1.365
1.394
1.223
1.402
1.378
C11–C12
C4–C5
C5–C6
C6–C7
C7–C8
C8–C9
C1–C10
1.496(4)
1.377(4)
1.415(4)
1.372(4)
1.414(4)
1.415(4)
1.502(4)
1.522
1.378
1.418
1.389
1.417
1.427
1.510
Bond angles (°)
Exp.
Calc.
Exp.
Calc.
O4–P1–O1
C7–O1–P1
114.24(11)
118.97(16)
117.63
119.54
C6–C7–O1
O2–C4–C5
119.80(2)
116.20(2)
118.66
116.42
Hydrogen-bond geometry (Å, °)
D–Hꢂ ꢂ ꢂA
D–H
0.84(4)
0.88
Hꢂ ꢂ ꢂA
Dꢂ ꢂ ꢂA
D–Hꢂ ꢂ ꢂA
174.0(4)
142.3(4)
136.5(4)
O2–H2ꢂ ꢂ ꢂO3i
N2–H2Bꢂ ꢂ ꢂO1ii
N2–H2Bꢂ ꢂ ꢂO5
1.81(4)
2.35(4)
2.56(3)
2.643(3)
3.090(3)
3.255(3)
0.88
Fig. 8. Electrostatic potential (ESP) surfaces of 2a (upper) and 2b (below). ESP
surface is shown both in space (with positive and negative regions shown in blue
and red, respectively) and mapped on electron densities (in the range of 0.05 a.u. –
deepest red-to-0.005 a.u. – deepest blue) of the molecule (ESP colour scale is such
that d+ ? dꢁ in the direction red ? blue). (For interpretation of the references to
colour in this figure legend, the reader is referred to the web version of this article.)
Table 6
Selected bond lengths and angles for 3e. Symmetry code: (i) ꢁx, 1 ꢁ y, 1 ꢁ z.
Exp.
Calc.
Exp.
Calc.
Bond lengths (Å)
P1–O4
P1–O2
O1–C4
O2–C7
1.461(3)
1.581(2)
1.360(4)
1.402(4)
1.221(4)
1.407(4)
1.480
1.597
1.365
1.399
1.224
1.398
N2–C11
C4–C5
C5–C6
C6–C7
C8–C9
1.349(4)
1.371(5)
1.411(5)
1.393(4)
1.421(5)
1.383
1.374
1.421
1.397
1.429
O3–C11
N2–C6
Bond angles (°)
C6–C7–O2
C6–N2–C11
C7–O2–P1
120.40(3)
129.20(3)
127.60(2)
121.33
128.49
127.49
O2–P1–O4
N2–C11–O3
C5–C4–O1
113.77(3)
123.80(3)
122.80(3)
113.96
124.79
122.48
Hydrogen-bond geometry (Å, °)
D–Hꢂ ꢂ ꢂA
D–H
0.84(3)
0.88(2)
Hꢂ ꢂ ꢂA
Dꢂ ꢂ ꢂA
D–Hꢂ ꢂ ꢂA
178.7(3)
146.0(2)
O1–H1Aꢂ ꢂ ꢂO5i
N2–H2Aꢂ ꢂ ꢂO4
1.83(3)
2.07(3)
2.670(3)
2.846(4)
Table 7
Selected bond lengths and angles for 3g. Symmetry code: (i) ꢁ1 ꢁ x, 1 ꢁ y, 1 ꢁ z.
Exp.
Calc.
Exp.
Calc.
Bond lengths (Å)
P1–O1
P1–O2
P1–C11
P1–C17
1.485(4)
1.618(4)
1.783(7)
1.810(6)
1.404(7)
1.4925
1.6570
1.8230
1.8690
1.3871
N1–C1
N2–C7
N1–C9
O3–C5
1.341(7)
1.351(7)
1.378(7)
1.342(7)
1.3210
1.3947
1.3572
1.3681
O2–C8
Bond angles (°)
O1–P1–C11
O2–P1–C11
O1–P1–C17
111.5(3)
106.0(3)
112.5(3)
113.93
105.44
110.34
O1–P1–O2
O2–P1–C17
C11–P1–C17
111.8(2)
102.0(3)
112.5(3)
119.13
97.80
108.74
Hydrogen-bond geometry (Å, °)
D–Hꢂ ꢂ ꢂA
D–H
0.93
0.857(19)
Hꢂ ꢂ ꢂA
2.55
2.41(5)
Dꢂ ꢂ ꢂA
D–Hꢂ ꢂ ꢂA
143.8
102(3)
C3–H3Aꢂ ꢂ ꢂO3i
N2–H2Aꢂ ꢂ ꢂO2
3.344(8)
2.713(6)
these maps is in the range of 0.05 a.u. (deepest red) to ꢁ0.005 a.u.
(deepest blue), where blue indicates the strongest attraction and
red indicates the strongest repulsion. Regions of negative V(r) are
usually associated with the lone pairs of electronegative atoms.
The negative potential in the studied compounds wraps the oxygen
atoms and as one can see in the Fig. 8, negative potentials on oxy-
gen atoms in position 8 of quinoline ring are smaller than the ones
on positions 5. The calculated values could explain the preferred
selectivity of phosphorylation.
Fig. 9. Electrostatic potential (ESP) surfaces of 3a (upper) and 3e (below).
The natural phosphorus charges on the phosphate derivatives
are about 2.3 in the phenyl phosphinate 3a, 3b, 3g and about 2.6
in the case of 3d, 3e phosphate. The EPS surfaces of 3a and 3e
are depicted on the Fig. 9 with the same parameters as on Fig. 8.

J.E. Nycz et al. / Journal of Molecular Structure 986 (2011) 39–48
47
The energy of highest occupied and lowest virtual molecular
orbitals of quinone type compounds are negative and close to:
ꢁ6.28, ꢁ6.83 eV (HOMO) and: ꢁ2.84, ꢁ3.17 eV (LUMO) for 2a,
2b, respectively. In the case of phosphinate and phosphate deriva-
tives the HOMO–LUMO gaps (3a, 3b, 3e, 3g: ꢃ4.16 eV; 3d: 4.27 eV)
are bigger, though the energy levels are less negative which sug-
gest strong acceptor affinity of these compounds. The partial den-
sity of states (PDOS) in terms of Mulliken population analysis were
calculated using the GaussSum program which provides a pictorial
representation of MOs compositions. The compounds were divided
in quinoline and phosphate fragments. Fig. 10 represents the PDOS
diagrams.
out in methanol solvent as more realistic than the calculation in
gas phase. The obtained energies: ꢁ212.58 (3a), ꢁ208.71 (3b),
208.29 (3g) and ꢁ230.96 (3d), ꢁ235.11 (3e) kcal/mol indicated
that the phosphate derivatives are more stable.
4. Conclusion
This study describes electron transfer properties of the structur-
ally characterised compounds 2a and 2b. All employed neutral
phosphorus nucleophiles 1b–f and the silylphosphine 1a reacted
in a similar fashion with the quinone 2a and 2b to produce the cor-
responding addition products 3a, 3b, 3c, 3d, 3e, 3f and 3g which
were characterised by spectroscopic and analytical techniques.
Furthermore, neither free radical intermediates nor the radical
coupling product with a P–P bond, such as the diphosphanes,
diphosphane dioxides or hypophosphoric acid esters was detected
in these reactions, which strongly suggests that the process occurs
without the formation of completely free radical species. The cyclic
voltammetry measurements showed similar redox properties for
both 2a and 2b. The first reduction potential corresponds with
the electrochemical behaviour of natural occurring quinones. The
calculated values could explain the preferred selectivity of
phosphorylation.
As can be seen the HOMO orbitals are mainly localised on the
quinoline fragments and the phosphate fragment plays a role in
lowering the HOMO and LUMO orbitals. In the case of compounds
3a, 3b containing phenyl phosphinate moiety with
p-acceptor
properties, which is connected to the lower LUMO orbitals, and
are stronger compared with the phosphate derivatives. To estimate
bonds between phosphinate/phosphate and quinoline fragments in
the compounds the Mayer bonds orders were calculated and have
values 1.78 (3a), 1.75 (3b, 3g), 1.98 (3d) and 1.90 (3e). As can be
seen from the bond orders, the 3d compound should be more sta-
ble in terms of P–O(quin) bond break. At the next step the energy
decomposition analyses of the studied compounds based on the
work of Morokuma [29] and the extended transition state (ETS)
partitioning scheme of Ziegler [30] have been carried out using
ADF program at the level of B3LYP/DZP. The binding energies of
the compounds were calculated as the difference between the en-
ergy of whole molecule with the optimized geometry and the ener-
gies of the optimized fragments phosphinate or phosphate chain
and quinoline parts, respectively. The calculations were carried
5. Supplementary material
CCDC – 726156 (for 2a), 726157 (for 2b), 726158 (for 3a),
734355 (for 3b), 760552 (for 3d), 767288 (for 3e) and 782763
(for 3g) contains the supplementary crystallographic data for this
paper. These data can be obtained free of charge from http://
www.ccdc.cam.ac.uk/conts/retrieving.html, or from the Cambridge
Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ,
UK; fax: +44 1223 336 033; or e-mail: deposit@ccdc.cam.ac.uk Cal-
culations have been carried out in Wroclaw Centre for Networking
and Supercomputing (http://www.wcss.wroc.pl).
Acknowledgments
We thank the DAAD for the award of a fellowship to J.E. Nycz.
This study was supported by Polish Ministry of Science No.
NN405178735. Calculations have been carried out in Wroclaw
Centre for Networking and Supercomputing (http://www.wcss.
wroc.pl).
References
[1] P.A. Frey, Chem. Rev. 90 (1990) 1343–1357.
[2] S.E. Wolkenberg, D.L. Boger, Chem. Rev. 102 (2002) 2477–2495.
[3] H. Ishikita, E.W. Knapp, J. Biol. Chem. 278 (2003) 52002–52011.
[4] T.W. Doyle, D.M. Balitz, R.E. Grulich, D.E. Net, S.J. Gould, C. Tann, A.E. Moews,
Tetrahedron Lett. 22 (1981) 4595–4598.
[5] M. Shibata, M. Uyeda, Y. Kido, N. Toya, R. Nakashima, R. Terazumi, J. Antibiot.
33 (1980) 1231–1235.
[6] D.M. Balitz, J.A. Bush, W.T. Bradner, T.W. Doyle, F.A. O’Herron, D.E. Nettleton, J.
Antibiot. 35 (1982) 259–264.
[7] D.L. Boger, M. Yasuda, L.A. Mitscher, S.D. Drake, P.A. Kitos, S.C. Thompson, J.
Med. Chem. 30 (1987) 1918–1928.
[8] R. Wolfenden, Y.-L. Liang, M. Matthews, R. Williams, J. Am. Chem. Soc. 109
(1987) 463–466.
[9] J.E. Nycz, Pol. J. Chem. 83 (2009) 589–594.
[10] J.E. Nycz, Pol. J. Chem. 83 (2009) 1637–1642.
[11] M. Behforouz, W. Cai, M.G. Stocksdale, J.S. Lucas, J.Y. Jung, D. Briere, A. Wang,
K.S. Katen, N.C. Behforouz, J. Med. Chem. 46 (2003) 5773–5780.
[12] M. Behforouz, J. Haddad, W. Cai, M.B. Arnold, F. Mohammadi, A.C. Sousa, M.A.
Horn, J. Org. Chem. 61 (1996) 6552–6555.
[13] S.E. Tunney, J.K. Stille, J. Org. Chem. 52 (1987) 748–753.
[14] G.M. Sheldrick, SHELXTL-97, University of Göttingen, 1997.
[15] G.M. Sheldrick, Acta Cryst. A 64 (2008) 112–122.
[16] L. Coudray, J.-L. Montchamp, Eur. J. Org. Chem. (2008) 3601–3613.
[17] M. Hayashi, Y. Matsuura, Y. Watanabe, Tetrahedron Lett. 45 (2004) 9167–
9169.
Fig. 10. Partial density of states of 3a (upper) and 3e (below).

48
J.E. Nycz et al. / Journal of Molecular Structure 986 (2011) 39–48
[18] D. Enders, A. Saint-Dizier,M.-I. Lannou,A. Lenzen,Eur.J. Org. Chem. (2006)29–49.
[19] F. Ramirez, S. Dershowitz, J. Am. Chem. Soc. 81 (1959) 587–590.
[20] F. Ramirez, E.H. Chen, S. Dershowitz, J. Am. Chem. Soc. 81 (1959) 4338–4342.
[21] E.A.C. Lucken, J. Chem. Soc. (1963) 5123–5127.
[22] F. Ramirez, S.B. Bhatia, A.V. Patwardhan, E.H. Chen, C.P. Smith, J. Org. Chem. 33
(1968) 20–24.
[23] R.O. Duthaler, P.A. Lyle, C. Heuberger, Helv. Chim. Acta 67 (1984) 1406–
1426.
[24] F. Ramirez, S. Dershowitz, J. Org. Chem. Soc. 22 (1957) 1282–1283.
[25] S. Burck, D. Gudat, M. Nieger, W.-W. du Mont, J. Am. Chem. Soc. 128 (2006)
3946–3955.
[26] R.B. King, J. Org. Chem. 41 (1976) 972–977.
[27] F. Wang, X. Liu, X. Cui, Y. Xiong, X. Zhoy, X. Feng, Chem. Eur. J. 15 (2009) 589–
592.
[28] J.E. Nycz, G. Malecki, M. Morag, G. Nowak, L. Ponikiewski, J. Kusz, A. Switlicka,
J. Mol. Struct. 980 (2010) 13–17.
[29] K.J. Morokuma, Chem. Phys. Lett. 9 (1971) 129–132.
[30] T. Ziegler, A. Rauk, Theor. Chim. Acta 46 (1977) 1–10.