+ All Categories
Home > Documents > Phosphorus–nitrogen compounds: Part 26. Syntheses, spectroscopic and structural investigations,...

Phosphorus–nitrogen compounds: Part 26. Syntheses, spectroscopic and structural investigations,...

Date post: 01-Jan-2017
Category:
Upload: yagmur
View: 213 times
Download: 1 times
Share this document with a friend
12
Phosphorus–nitrogen compounds: Part 26. Syntheses, spectroscopic and structural investigations, biological and cytotoxic activities, and DNA interactions of mono and bisferrocenylspirocyclotriphosphazenes Nuran Asmafiliz a , Zeynel Kılıç a,, Tuncer Hökelek b , L. Yasemin Koç c , Leyla Açık d , Yasemin Süzen e , Yag ˘mur Öner d a Department of Chemistry, Ankara University, 06100 Ankara, Turkey b Department of Physics, Hacettepe University, 06800 Ankara, Turkey c Department of Biology, Ankara University, 06100 Ankara, Turkey d Department of Biology, Gazi University, 06500 Ankara, Turkey e Department of Chemistry, Anadolu University, 26470 Eskis ßehir, Turkey article info Article history: Received 19 November 2012 Received in revised form 26 February 2013 Accepted 2 March 2013 Available online 13 March 2013 Keywords: Ferrocenylphosphazenes Syntheses Spectroscopy Crystal structure DNA interactions HeLa cell line abstract The condensation reactions of the tetrachloro mono (1 and 2) and bisferrocenylspirocyclotriphosphaz- enes (35) with 1,4-dioxa-8-azaspiro[4,5]decane (DASD) resulted in the formation of the partly and fully DASD-substituted phosphazenes. The reactions of equal amounts of 15 and DASD produced the mono- DASD-substituted ferrocenylphosphazenes (1a5a), as the major product. When the reactions were car- ried out with 1 equiv of 15 and 2 equiv of DASD, corresponding geminal-phosphazenes (1b5b) were isolated. Moreover, the reactions of 1 equiv of 15 and 3 equiv of DASD gave the tri- (1c4c) and tetra-substituted (1d5d) phosphazenes. When the excess DASD was used, the fully-substituted phosp- hazenes (1d5d) were obtained. The chirality of 3a was evaluated using chiral HPLC column. The struc- tures of all the phosphazenes were verified by FTIR, MS, 1 H, 13 C and 31 P NMR, and HSQC spectral data. The crystal structures of 4a, 2b, 5b, and 1d were determined by X-ray diffraction techniques. The 10 phospha- zene derivatives were screened for antimicrobial activity. Meanwhile, interactions between the com- pounds and pBR322 plasmid DNA were presented by agarose gel electrophoresis. The compounds 2b, 1d, 2d, and 4d were tested against HeLa cancer cell lines. Among these compounds, 4d had cytotoxic effect on HeLa cell after 24 h treatment. Ó 2013 Elsevier B.V. All rights reserved. 1. Introduction Hexachlorocyclotriphosphazene (trimer), N 3 P 3 Cl 6 , has been a useful molecule for the preparation of the substituted phosphazene frameworks with mono-, di-, tri-, and tetra- functional reagents [1–11]. The monodentate substituents may be replaced with all the Cl atoms, gradually, in different solvents [12]. When the 2 equiv of primary amines exchange with two Cl atoms, cis/trans and/or geminal cyclotriphosphazene derivatives can form via the path- ways of SN 1 (P), SN 2 (P) and/or proton abstraction–chloride elimina- tion reaction mechanisms depending on the solvent polarities, base strength and nucleophiles [13–15]. The reactions of N 3 P 3 Cl 6 with primary amines are much more complex than the secondary amines [15]. The Cl replacement reactions of N 3 P 3 Cl 6 with the sec- ondary amines afford mainly non-geminal phosphazenes with usu- ally trans products predominating. Aziridine is only one exception to this trend in the literature [15,16]. It reacts to produce both the geminal and the non-geminal isomers in approximately equal amounts [17]. In contrast to the reactions of trimer with secondary amines, only the geminal products could have been isolated from the reactions of tetrachloro bulky-crypta [18,19] and bulky-monof- errocenyl spirocyclotriphosphazenes [20,21] with pyrrolidine, mor- pholine, and 1-aza-12-crown-4, not depending on the bulkiness of these amines. Besides, there are some papers in the literature com- paring the reactivity of trimer and octachlorocyclotetraphospha- zene (tetramer) [22,23]. On the other hand, cyclotriphosphazene derivatives with pendant mono and bisferrocenyl groups have aroused a considerable amount of interest recently, not only from their synthetic, mechanistic, and electrochemical point of view but also with respect to their stereogenic properties and unusual structural characteristics [24–27]. We report herein: (i) the synthesis of mono- (1a5a), gem- (1b5b), tri- (1c4c), and tetra-DASD-substituted (1d5d) mono and bisferrocenylphosphazenes obtained from the gradually Cl replace- ment reactions of the tetrachlorocyclotriphosphazenes containing 0020-1693/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.ica.2013.03.001 Corresponding author. Tel.: +90 3122126720x1043; fax: +90 3122232395. E-mail address: [email protected] (Z. Kılıç). Inorganica Chimica Acta 400 (2013) 250–261 Contents lists available at SciVerse ScienceDirect Inorganica Chimica Acta journal homepage: www.elsevier.com/locate/ica
Transcript
Page 1: Phosphorus–nitrogen compounds: Part 26. Syntheses, spectroscopic and structural investigations, biological and cytotoxic activities, and DNA interactions of mono and bisferrocenylspirocyclotriphosphazenes

Inorganica Chimica Acta 400 (2013) 250–261

Contents lists available at SciVerse ScienceDi rect

Inorgan ica Chimi ca Acta

journal homepage: www.elsevier .com/locate / ica

Phosphorus–nitrogen compounds: Part 26. Syntheses, spectroscopic and structural investigations, biological and cytotoxic activities, and DNA interactions of mono and bisferrocenylspirocyclotriphosphazenes

0020-1693/$ - see front matter � 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.ica.2013.03.001

⇑ Corresponding author. Tel.: +90 3122126720x1043; fax: +90 3122232395. E-mail address: [email protected] (Z. Kılıç).

Nuran Asmafiliz a, Zeynel Kılıç a,⇑, Tuncer Hökelek b, L. Yasemin Koç c, Leyla Açık d, Yasemin Süzen e,Yag ˘mur Öner d

a Department of Chemistry, Ankara University, 06100 Ankara, Turkey b Department of Physics, Hacettepe University, 06800 Ankara, Turkey c Department of Biology, Ankara University, 06100 Ankara, Turkey d Department of Biology, Gazi University, 06500 Ankara, Turkey e Department of Chemistry, Anadolu University, 26470 Eski s�ehir, Turkey

a r t i c l e i n f o

Article history: Received 19 November 2012 Received in revised form 26 February 2013 Accepted 2 March 2013 Available online 13 March 2013

Keywords:FerrocenylphosphazenesSynthesesSpectroscopyCrystal structure DNA interactions HeLa cell line

a b s t r a c t

The condensation reactions of the tetrachloro mono (1 and 2) and bisferrocenylspirocyclotriphosphaz- enes (3–5) with 1,4-dioxa- 8-azaspiro[4,5]decane (DASD) resulted in the formation of the partly and fully DASD-subst ituted phosphaze nes. The reactions of equal amounts of 1–5 and DASD produced the mono- DASD-subst ituted ferrocenylphosphazenes (1a–5a), as the major product. When the reaction s were car- ried out with 1 equiv of 1–5 and 2 equiv of DASD, corresp onding gemina l-phosphazenes (1b–5b) were isolated. Moreover, the reactions of 1 equiv of 1–5 and 3 equiv of DASD gave the tri- (1c–4c) and tetra-sub stituted (1d–5d) phosphazenes. When the excess DASD was used, the fully-substituted phosp- hazenes (1d–5d) were obtained. The chirality of 3a was evaluated using chiral HPLC column. The struc- tures of all the phosphazenes were verified by FTIR, MS, 1H, 13C and 31P NMR, and HSQC spectral data. The crystal structures of 4a, 2b, 5b, and 1d were determined by X-ray diffraction technique s. The 10 phospha- zene derivatives were scre ened for antimicrobial activity. Meanwhile, interactio ns between the com- pounds and pBR322 plasmid DNA were presented by agarose gel electrophoresis. The compounds 2b,1d, 2d, and 4d were tested against HeLa cancer cell lines. Among these compounds, 4d had cytotoxic effect on HeLa cell after 24 h treatment.

� 2013 Elsevier B.V. All rights reserved.

1. Introduction

Hexachlorocy clotriphosphazen e (trimer), N3P3Cl6, has been auseful molecule for the preparation of the substitut ed phosphazene frameworks with mono-, di-, tri-, and tetra- functional reagents [1–11]. The monodenta te substituents may be replaced with all the Cl atoms, gradually, in different solvents [12]. When the 2 equiv of primary amines exchange with two Cl atoms, cis/trans and/or geminal cyclotriphos phazene derivatives can form via the path- ways of SN 1(P), SN 2(P) and/or proton abstracti on–chloride elimina- tion reaction mechanism s depending on the solvent polarities, base strength and nucleophiles [13–15]. The reactions of N3P3Cl6 withprimary amines are much more complex than the secondary amines [15]. The Cl replacemen t reactions of N3P3Cl6 with the sec- ondary amines afford mainly non-geminal phosphazenes with usu- ally trans products predominating. Aziridine is only one exception

to this trend in the literature [15,16]. It reacts to produce both the geminal and the non-gemi nal isomers in approximat ely equal amounts [17]. In contrast to the reactions of trimer with secondar yamines, only the geminal products could have been isolated from the reactions of tetrachloro bulky-crypta [18,19] and bulky-mo nof- erroceny l spirocyclotri phosphazenes [20,21] with pyrrolidine, mor- pholine, and 1-aza-12- crown-4, not depending on the bulkiness of these amines. Besides, there are some papers in the literature com- paring the reactivity of trimer and octachloroc yclotetraph ospha- zene (tetramer) [22,23]. On the other hand, cyclotrip hosphazene derivatives with pendant mono and bisferrocenyl groups have aroused a considerable amount of interest recently, not only from their synthetic, mechanistic , and electrochemi cal point of view but also with respect to their stereogenic properties and unusual structura l characterist ics [24–27].

We report herein: (i) the synthesis of mono- (1a–5a), gem- (1b–5b), tri- (1c–4c), and tetra-DASD- substituted (1d–5d) mono and bisferroceny lphosphazenes obtained from the gradually Cl replace- ment reactions of the tetrachlorocy clotriphosphazen es containing

Page 2: Phosphorus–nitrogen compounds: Part 26. Syntheses, spectroscopic and structural investigations, biological and cytotoxic activities, and DNA interactions of mono and bisferrocenylspirocyclotriphosphazenes

N. Asmafiliz et al. / Inorganica Chimica Acta 400 (2013) 250–261 251

mono and bisferrocenyl pendant arms (1–5) with DASD (Scheme 1); (ii) the structure determination s of the compounds by elemental analyses, mass spectrometry, Fourier transform (FTIR), one-dimensional (1D) 1H, 13C, and 31P NMR, and two- dimensional (2D) heteronucle ar single quantum coherenc e (HSQC)techniques; (iii) the solid-state and molecular structures of 4a, 2b,5b and 1d; (iv) as an example, the stereogenic propertie s of 3a isfound out by chiral high pressure liquid chromatography (HPLC)column; (v) the investigations of antibacterial and antifungal activ- ity of 10 phosphazene derivatives; (vi) interactions between the compounds 3a–5a, 2b, 4b, 3c, 4c and 1d–5d and pBR322 plasmid DNA, and (vii) the evaluations of the compounds 2b, 1d, 2d and4d for cytotoxic activity against HeLa cancer cell lines.

2. Experimental

2.1. Reagents

Hexachlorocy clotriphosphazatr iene (Aldrich), ferroceneca rbox- aldehyde (Aldrich), aliphatic amines (Fluka) and 1,4-dioxa-8- aza- spiro[4,5]decan e (DASD) (Merck) were purchase d and used without further purification. All reactions were monitored using thin-layer chromatography in different solvents and chromato- graphed using silica gel. All experiments were carried out in an ar- gon atmosphere .

2.2. Instruments

The 1H, 13C, and 31P NMR spectra were recorded on a Bruker DPX FT-NMR spectrometer (SiMe4 as an internal standard and 85% H3PO4 as an external standard), operating at 499.94, 125.72, and 202.38 MHz. The spectromete r was equipped with a 5 mm

mono-substituted ferrocenylphosphazenes

bis-substituted ferrocenylphosphazenes

( ) ( )nn

H H FcCH2

FcCH2

FcCH2

12012

1b 2b 3b 4b 5b

R1 n Compound

Fc CH2 :CH2

Fe

5

2

23

3

4

1

O

ON

O

ON

Fc CH2 NR1 N

ClP

N

NP

NP

Cl

Fc

O

O

Cl

PN

PN

N

PCl

N NR1 Fc CH2

O

ON

Cl

Compound nR1

12012

1a 2a 3a 4a 5a

2

2

HHFcCHFcCHFcCH2

O

ONH

N

Cl

Cl Cl

ClP

N

P

N

NR1 Fc CH2 N

P

( )n

Scheme 1. The formulae of DASD-substitute

PABBO BB inverse-gradie nt probe. Standard Bruker pulse programs [28] were used. The IR spectra were recorded on Jasco FT/IR-430 spectromete r in KBr disks and were reported in cm �1 units. APIES mass spectrometric analyses were performed on an AGILENT 1100 MSD spectromete r. The melting points were measured on aGallenka mp apparatus using a capillary tube.

2.3. X-ray crystallog raphy

The suitable crystals of compounds 4a, 2b, 5b, and 1d werecrystallized from acetonitrile at room temperature. Crystallo -graphic data were recorded on a Bruker Kappa APEXII CCD area- detector diffractome ter using Mo Ka radiation (k = 0.71073 Å) at T = 100(2) K. Absorption corrections by multi-sca n [29] were ap- plied. Structures were solved by direct methods and refined by full-matr ix least squares against F2 using all data [30]. All non-H atoms were refined anisotropi cally. H atom positions were calcu- lated geometrical ly at distances of 0.95 Å (CH) and 0.99 Å (CH2)Å from the parent C atoms; a riding model was used during the refinement process and the Uiso(H) values were constrained to 1.2Ueq(carrier atom) for CH and CH 2 groups. In compound s 1dand 2b, H atoms of NH groups were located in difference Fourier maps and were freely refined. In compound 4a, the two Cl atoms of the phosphazene ring attached at P2 were disordered over two orientati ons. During the refinement process, the disordered Cl2, Cl20, Cl3 and Cl3 0 atoms were refined with occupancies of 0.62, 0.38, 0.74 and 0.26, respectively .

2.4. Preparatio n of compoun ds

The tetrachloro mono and bisferrocenyl phosphazenes (1–5)were obtained from the reactions of mono and bisferrocenyl amines

tris-substituted ferrocenylphosphazenes

tetrakis-substitutedferrocenylphosphazenes

( ) ( )nn

O

ON

O

ON

CH2 NR1 N

P

N

NP

NP

Cl

N

NO

O O

ON

O

ON

Fc CH2 NR1 N

P

N

NP

NP

N

O

O

H HFcCH2

FcCH2

FcCH2

12012

1c 2c 3c 4c 5c (not obtained)

R1 n Compound

H H FcCH2

FcCH2

FcCH2

12012

1d 2d 3d 4d 5d

R1 n Compound

H H FcCH2

FcCH2

FcCH2

12012

1 2 3 4 5

R1 n Compound

d mono and bisferrocenylphosphazenes.

Page 3: Phosphorus–nitrogen compounds: Part 26. Syntheses, spectroscopic and structural investigations, biological and cytotoxic activities, and DNA interactions of mono and bisferrocenylspirocyclotriphosphazenes

252 N. Asmafiliz et al. / Inorganica Chimica Acta 400 (2013) 250–261

with trimer according to the methods reported in the literature [21].

2.4.1. 7-Ferrocenyl-2,2 ,4-trichloro-4-(1,4-dioxa-8-azas piro[4.5]dec-8- yl)-1,3,5,7,11-penta aza-2 k5,4k5,6k5-triphosphas piro[5.5]undeca- 1,3,5-triene (1a)

A solution of DASD (0.23 mL, 1.83 mmol) in dry THF (50 mL)was added to a solution of 1 (1.00 g, 1.83 mmol) and triethylamin e(0.77 mL) in dry THF (100 mL) with stirring and refluxing for 12 h. The reaction was followed on TLC silica gel plates using toluene–THF (5:1), giving two new products. After the solvent was evapo- rated, the product was purified by column chromatograp hy with toluene–THF (5:1). The first product eluted was the mono-DASD- substituted derivative (1a). Yield: 0.52 g (43%), mp 93 �C. Anal.Calc. for C21H30N6O2FeP3Cl3: C, 38.59; H, 4.63; N, 12.86. Found: C, 38.98; H, 4.81; N, 12.76%. APIES-MS (fragments were based on 35Cl and 56Fe, Ir %): m/z 652 ([M]+, 71.4). FTIR (KBr, cm �1): 3392 (N–H), 3093 (C–H arom.), 2958, 2854 (C–H aliph.), 1222 (asymm.),1180 (symm.) (P@N), 1066 (COC), 561 (asymm.), 543 (symm.)(PCl). The second product was the gem-DASD-sub stituted deriva- tive (1b). Yield: 0.28 g (20%) (the analytical data of 1b are given in (part 2.3.6.).

2.4.2. 7-Ferrocenyl-2,2 ,4-trichloro-4-(1,4-dioxa-8-azas piro[4.5]dec-8- yl)-1,3,5,7,12-penta aza-2 k5,4k5,6k5-triphosphas piro[6.5]dodeca- 1,3,5-triene (2a)

The work-up procedure was similar to that of compound 1a,using 2 (0.80 g, 1.40 mmol), DASD (0.18 mL, 1.40 mmol) and trieth- ylamine (0.60 mL). The first product eluted was the mono-DASD- substituted derivative (2a). Yield: 0.41 g (44%), mp 169 �C. Anal. Calc.for C22H32N6O2FeP3Cl3: C, 39.58; H, 4.83; N, 12.59. Found: C, 39.83; H, 4.75; N, 12.42%. APIES-MS (fragments were based on 35Cl and 56Fe,Ir %): m/z 666 ([M]+, 78.9). FTIR (KBr, cm �1): 3385 (N–H), 3090 (C–Harom.), 2962, 2874 (C–H aliph.), 1220 (asymm.), 1176 (symm.)(P@N), 1052 (COC), 561 (asymm.) 543 (symm.) (PCl). The second product was the gem-DASD-sub stituted derivative (2b). Yield: 0.25 g (23%) (the analytica l data of 2b are given in part 2.3.7.).

2.4.3. 7,11-Diferrocenyl- 2,2,4-trichloro- 4-(1,4-dioxa-8-azaspiro[4.5 ]dec-8-yl)-1,3,5,7,10-pentaaza-2 k5,4k5,6k5-triphosphasp iro[4.5]deca-1,3,5 -triene (3a)

The work-up procedure was similar to that of compound 1a,using 3 (0.80 g, 1.09 mmol), DASD (0.14 mL, 1.09 mmol) and trieth- ylamine (0.70 mL). The first product eluted was the mono-DASD- substituted derivative (3a). Yield: 0.38 g (42%), mp 136 �C. Anal.Calc. for C31H38N6O2Fe2P3Cl3: C, 44.45; H, 4.57; N, 10.03. Found: C, 44.60; H, 4.56; N, 10.00%. APIES-MS (fragments were based on 35Cl and 56Fe, Ir %): m/z 836 ([M]+, 65.7). FTIR (KBr, cm �1): 3093 (C–H arom.), 2974, 2865 (C–H aliph.), 1224 (asymm.), 1170 (symm.) (P@N), 1062 (COC), 547 (asymm.) 519 (symm.) (PCl).The second product was the gem-DASD- substituted derivative (3b). Yield: 0.28 g (27%) (the analytical data of 3b are given in part 2.3.8.).

2.4.4. 7,11-Diferrocenyl- 2,2,4-trichloro- 4-(1,4-dioxa-8-azaspiro[4.5 ]dec-8-yl)-1,3,5,7,11-pentaaza-2 k5,4k5,6k5-triphosphasp iro[5.5]undeca-1 ,3,5-triene (4a)

The work-up procedure was similar to that of compound 1a,using 4 (1.00 g, 1.34 mmol), DASD (0.17 mL, 1.34 mmol) and trieth- ylamine (0.70 mL). The first product eluted was the mono-DASD- substituted derivative (4a). Yield: 0.44 g (39%), mp 137 �C. Anal.Calc. for C32H40N6O2Fe2P3Cl3: C, 45.13; H, 4.73; N, 9.87. Found: C, 45.39; H, 5.07; N, 9.65%. APIES-MS (fragments were based on 35Cl and 56Fe, Ir %): m/z 850 ([M]+, 62.4). FTIR (KBr, cm �1): 3077 (C–H arom.), 2985, 2844 (C–H aliph.), 1230 (asymm.), 1178 (symm.) (P@N), 1066 (COC), 557 (asymm.), 514 (symm.) (PCl).

The second product was the gem-DASD -substituted derivative (4b). Yield: 0.30 g (23%) (the analytical data of 4b are given in part 2.3.9.).

2.4.5. 7,11-Diferro cenyl-2,2,4-tric hloro-4-(1,4-dioxa-8- azaspiro[4 .5]dec-8-yl)-1,3,5,7,12-pentaaza- 2k5,4k5,6k5-triphosph aspiro[6.5]dodec a-1,3,5-triene (5a)

The work-up procedure was similar to that of compound 1a,using 5 (0.90 g, 1.18 mmol), DASD (0.15 mL, 1.18 mmol) and trieth- ylamine (0.50 mL). The first product eluted was the mono-DASD- substitut ed derivative (5a). Yield: 0.41 g (40%), mp 185 �C. Anal.Calc. for C33H42N6O2Fe2P3Cl3: C, 45.79; H, 4.89; N, 9.71. Found: C, 45.97; H, 4.67; N, 9.62%. APIES-MS (fragments were based on 35Cl and 56Fe, Ir %): m/z 865 ([MH]+, 98.4). FTIR (KBr, cm �1): 3095 (C–H arom.), 2964, 2863 (C–H aliph.), 1226 (asymm.), 1188 (symm.) (P@N), 1066 (COC), 561 (asymm.), 545 (symm.) (PCl).The second product was the gem-DASD -substituted derivative (5b). Yield: 0.26 g (23%) (the analytical data of 5b are given in part 2.3.10.).

2.4.6. 7-Ferroceny l-2,2-dichloro-4 ,4-di-1,4-dioxa-8 -azaspiro[4.5]d ec- 8-yl-1,3,5,7 ,11-penta aza-2 k5,4k5,6k5-triphosphas piro[5.5]und eca- 1,3,5-trie ne (1b)

A solution of DASD (0.47 mL, 3.66 mmol) in dry THF (50 mL)was added to a solution of 1 (1.00 g, 1.83 mmol) and triethylam ine (0.77 mL) in dry THF (100 mL) with stirring and refluxing for 16 h. The reaction was followed on TLC silica gel plates using toluene–THF (5:1). After the solvent was evaporated , the product was puri- fied by column chromatogr aphy with toluene–THF (5:1). Yield: 0.93 g (67%), mp 84 �C. Anal. Calc. for C28H42N7O4FeP3Cl2: C, 44.23; H, 5.57; N, 12.90. Found: C, 44.24; H, 5.80; N, 12.65%. APIES-MS (fragments were based on 37Cl and 56Fe, Ir %): m/z 761([M]+, 100.0). FTIR (KBr, cm �1): 3245 (N–H), 3087 (C–H arom.),2956, 2853 (C–H aliph.), 1222 (asymm.), 1184 (symm.) (P@N),1056 (COC), 543 (asymm.), 520 (symm.) (PCl).

2.4.7. 7-Ferroceny l-2,2-dichloro-4 ,4-di-1,4-dioxa-8 -azaspiro[4.5]d ec- 8-yl-1,3,5,7 ,12-penta aza-2 k5,4k5,6k5-triphosphas piro[6.5]dod eca- 1,3,5-trie ne (2b)

The work-up procedure was similar to that of compound 1b,using 2 (0.80 g, 1.40 mmol), DASD (0.46 mL, 3.59 mmol) and trieth- ylamine (0.60 mL). Yield: 0.74 g (68%), mp 132 �C. Anal. Calc. for C29H44N7O4FeP3Cl2: C, 44.98; H, 5.73; N, 12.66. Found: C, 44.74; H, 5.85; N, 12.78%. APIES-MS (fragments were based on 37Cl and 56Fe, Ir %): m/z 775 ([M]+, 100.0). FTIR (KBr, cm �1): 3401 (N–H),3090 (C–H arom.), 2969, 2867 (C–H aliph.), 1216 (asymm.), 1176 (symm.) (P@N), 1052 (COC), 555 (asymm.), 526 (symm.) (PCl).

2.4.8. 7,11-Diferro cenyl-2,2-dichloro -4,4-di-1,4-dioxa -8- azaspiro[4 .5]dec-8-yl-1,3,5 ,7,10-pentaaza- 2k5,4k5,6k5-triphosph aspiro[4.5]deca- 1,3,5-triene (3b)

The work-up procedure was similar to that of compound 1b,using 3 (0.80 g, 1.09 mmol), DASD (0.31 mL, 2.18 mmol) and trieth- ylamine (0.70 mL). Yield: 0.41 g (40%), mp 185 �C. (0.68 g, 66%, mp 207 �C). Anal. Calc. for C38H50N7O4Fe2P3Cl2: C, 48.33; H, 4.27; N, 10.38. Found: C, 48.48; H, 4.20; N, 10.15%. APIES-MS (fragmentswere based on 35Cl and 56Fe, Ir %): m/z 944 ([MH]+, 100.0). FTIR (KBr, cm �1): 3093 (C–H arom.), 2956, 2878 (C–H aliph.), 1205 (asymm.), 1151 (symm.) (P@N), 1056 (COC), 561 (asymm.), 548 (symm.) (PCl).

2.4.9. 7,11-Diferro cenyl-2,2-dichloro -4,4-di-1,4-dioxa -8- azaspiro[4 .5]dec-8-yl-1,3,5 ,7,11-pentaaza- 2k5,4k5,6k5-triphosph aspiro[5.5]undec a-1,3,5-triene (4b)

Th e wo rk -u p pr oc ed ure wa s si mi la r to th at of co mp ou nd 1b, us ing 4 (1.0 0 g, 1. 34 mm ol), DAS D (0.38 mL , 2.6 8 mm ol ) an d tr iet hy lam ine

Page 4: Phosphorus–nitrogen compounds: Part 26. Syntheses, spectroscopic and structural investigations, biological and cytotoxic activities, and DNA interactions of mono and bisferrocenylspirocyclotriphosphazenes

N. Asmafiliz et al. / Inorganica Chimica Acta 400 (2013) 250–261 253

(0.70 mL ). Yi eld : 0. 81 g (63% ), mp 21 2 �C. Ana l. Ca lc. fo r C39H52N7O4-

Fe2P3Cl2: C, 48 .8 8; H, 5. 47 ; N, 10. 23 . Fo un d: C, 49 .0 4; H, 5. 41 ; N, 9.9 7% .API ES -M S (fra gm en ts we re ba se d on 35Cl an d 56Fe , Ir %): m/z 89 5([M�(Cl+ C2H4)]+, 42 .0 ). FT IR (KBr, cm �1): 30 93 (C–H aro m. ), 29 54 ,28 73 (C–H al ip h.), 12 20 (asy mm .), 11 74 (sym m. ) (P@N), 10 52 (COC ), 56 3 (asy mm. ), 53 4 (symm .) (PCl ).

2.4.10. 7,11-Diferro cenyl-2,2-dichloro -4,4-di-1,4-dioxa -8- azaspiro[4.5 ]dec-8-yl-1,3,5,7 ,12-pentaaza-2 k5,4k5,6k5-triphosphasp iro[6.5]dodeca-1 ,3,5-triene (5b)

The work-up procedure was similar to that of compound 1b,using 5 (0.90 g, 1.18 mmol), DASD (0.33 mL, 2.36 mmol) and trieth- ylamine (0.60 mL). Yield: 0.70 g (61%), mp 225 �C. Anal. Calc. for C40H54N7O4Fe2P3Cl2: C, 49.41; H, 5.60; N, 10.08. Found: C, 49.64; H, 5.52; N, 10.02%. APIES-MS (fragments were based on 35Cl and 56Fe, Ir %): m/z 972 ([MH]+, 91.8). FTIR (KBr, cm �1): 3089 (C–Harom.), 2950, 2856 (C–H aliph.), 1213 (asymm.), 1169 (symm.)(P@N), 1066 (COC), 562 (asymm.), 543 (symm.) (PCl).

2.4.11. 7-Ferroceny l-2-chloro-2,4,4- tri-1,4-dioxa-8- azaspiro[4.5]dec -8-yl-1,3,5,7,1 1-penta aza-2 k5, 4k5,6k5-triphosphas piro[5.5]undeca- 1,3,5-triene (1c)

A solution of DASD (0.56 mL, 4.39 mmol) in dry THF (50 mL)was added to a solution of 1 (0.80 g, 1.46 mmol) and triethylamin e(0.85 mL) in dry THF (100 mL) with stirring and refluxing for 20 h. The reaction was followed on TLC silica gel plates using toluene–THF (3:1), giving two new products. After the solvent was evapo- rated, the tri-DASD-su bstituted product was purified by column chromatograp hy with toluene–THF (3:1). Yield: 0.50 g (40%), mp 103 �C. Anal. Calc. for C35H54N8O6FeP3Cl: C, 48.48; H, 6.28; N, 12.92. Found: C, 48.12; H, 6.28; N, 12.65%. APIES-MS (fragmentswere based on 35Cl and 56Fe, Ir %): m/z 837 ([M�(CH2OH)]+,100.0). FTIR (KBr, cm �1): 3280 (N–H), 3083 (C–H arom.), 2956, 2867 (C–H aliph.), 1232 (asymm.), 1171 (symm.) (P@N), 1054 (COC), 551 (PCl). The second product was the tetra-DASD -substi- tuted derivative (1d). Yield: 0.41 g (29%) (the analytical data of 1d are given in part 2.3.16.).

2.4.12. 7-Ferroceny l-2-chloro-2,4,4- tri-1,4-dioxa-8- azaspiro[4.5]dec -8-yl-1,3,5,7,1 2-penta aza-2 k5, 4k5,6k5-triphosphas piro[6.5]dodeca- 1,3,5-triene (2c)

The work-up procedure was similar to that of compound 1c,using 2 (0.80 g, 1.40 mmol), DASD (0.54 mL, 4.20 mmol) and trieth- ylamine (1.00 mL). The first product eluted was the tri-DASD-subs ti- tuted derivative (2c). Yield: 0.52 g (42%). Anal. Calc. for C36H56N8O6FeP3Cl: C, 49.08; H, 6.41; N, 12.72. Found: C, 48.89; H, 6.41; N, 12.61%. APIES-MS (fragments were based on 35Cl and 56Fe,Ir %): m/z 880 ([M]+, 65.0). FTIR (KBr, cm �1): 3396 (N–H), 3093 (C–H arom.), 2962, 2850 (C–H aliph.), 1230 (asymm.), 1176 (symm.)(P@N), 1051 (COC), 576 (PCl). The second product was the tetra- DASD-subst ituted derivative (2d). Yield: 0.39 g (28%) (the analytical data of 2d are given in part 2.3.17.).

2.4.13. 7,11-Diferro cenyl-2-chloro- 2,4,4-tri-1,4-diox a-8- azaspiro[4.5 ]dec-8-yl-1,3,5,7 ,10-pentaaza-2 k5,4k5,6k5-triphosphasp iro[4.5]deca-1,3,5 -triene (3c)

The work-up procedure was similar to that of compound 1c,using 3 (1.00 g, 1.37 mmol), DASD (0.53 mL, 4.10 mmol) and trieth- ylamine (1.00 mL). The first product eluted was the tri-DASD- substituted derivative (3c). Yield: 0.56 g (39%), mp 211 �C. Anal.Calc. for C45H62N8O6Fe2P3Cl: C, 51.42; H, 5.95; N, 10.66. Found: C, 51.78; H, 6.15; N, 10.41%. APIES-MS (fragments were based on 35Cl and 56Fe, Ir %): m/z 1033 ([M�OH]+, 100.0). FTIR (KBr, cm �1):3095 (C–H arom.), 2952, 2850 (C–H aliph.), 1247 (asymm.), 1188 (symm.) (P@N), 1056 (COC), 565 (PCl). The second product was

the tetra-DASD- substituted derivative (3d). Yield: 0.38 g (24%)(the analytica l data of 3d are given in part 2.3.18.).

2.4.14. 7,11-Dife rrocenyl-2-chloro -2,4,4-tri-1,4-d ioxa-8- azaspiro[4 .5]dec-8-yl-1,3,5 ,7,11-penta aza-2 k5,4k5,6k5-triphosph aspiro[5.5]undec a-1,3,5-triene (4c)

The work-up procedure was similar to that of compound 1c,using 4 (0.90 g, 1.21 mmol), DASD (0.46 mL, 3.62 mmol) and trieth- ylamine (0.95 mL). The first product eluted was the tri-DASD- substitut ed derivative (4c). Yield: 0.51 g (40%), mp 286 �C. Anal.Calc. for C46H64N8O6Fe2P3Cl: C, 51.87; H, 6.06; N, 10.52. Found: C, 51.80; H, 6.36; N, 10.62%. APIES-MS (fragments were based on 35Cl and 56Fe, Ir %): m/z 1047 ([M�OH]+, 100.0). FTIR (KBr, cm �1):3092 (C–H arom.), 2956, 2854 (C–H aliph.), 1220 (asymm.), 1166 (symm.) (P@N), 1052 (COC), 538 (PCl). The second product was the tetra-DASD-su bstituted derivative (4d). Yield: 0.42 g (30%)(the analytica l data of 4d are given in part 2.3.19.).

2.4.15. 7-Ferroceny l-2,2,4,4-tetr a-1,4-dioxa-8-az aspiro[4.5]dec-8 -yl- 1,3,5,7,11-pe ntaaza-2 k5,4k5,6k5-triphosphaspiro [5.5]undeca-1,3,5 -triene (1d)

A solution of DASD (0.82 mL, 5.84 mmol) in dry THF (50 mL)was added to a solution of 1 (0.80 g, 1.46 mmol) and triethylamin e(0.85 mL) in dry THF (100 mL) with stirring and refluxing for 25 h. The reaction was followed on TLC silica gel plates using toluene–THF (3:1). After the solvent was evaporated , the product was puri- fied by column chromatograp hy with toluene–THF (3:1). Yield: 0.96 g (68%), mp 230 �C. Anal. Calc. for C42H66N9O8FeP3: C, 51.80; H, 6.83; N, 12.95. Found: C, 52.13; H, 6.78; N, 12.58%. APIES-MS (fragments were based on 35Cl and 56Fe, Ir %): m/z 974 ([MH]+,100.0). FTIR (KBr, cm �1): 3286 (N–H), 3083 (C–H arom.), 2960, 2873 (C–H aliph.), 1224 (asymm.), 1191 (symm.) (P@N), 1056 (COC).

2.4.16. 7-Ferroceny l-2,2,4,4-tetr a-1,4-dioxa-8-az aspiro[4.5]dec-8 -yl- 1,3,5,7,12-pe ntaaza-2 k5,4k5,6k5-triphosphaspiro [6.5]dodeca-1,3,5 -triene (2d)

The work-up procedure was similar to that of compound 1d,using 2 (0.80 g, 1.40 mmol), DASD (0.78 mL, 5.60 mmol) and trieth- ylamine (1.00 mL). Yield: 0.95 g (69%), mp 271 �C. Anal. Calc. for C43H68N9O8FeP3: C, 52.28; H, 6.94; N, 12.76. Found: C, 52.05; H, 6.67; N, 12.46%. APIES-MS (fragments were based on 35Cl and 56Fe, Ir %): m/z 988 ([MH]+, 100.0). FTIR (KBr, cm �1): 3266 (N–H),3091 (C–H arom.), 2972, 2846 (C–H aliph.), 1226 (asymm.), 1191 (symm.) (P@N), 1050 (COC).

2.4.17. 7,11-Dife rrocenyl–2,2,4,4-tetra-1,4-dioxa -8-azaspiro[4.5 ]dec- 8-yl-1,3,5,7,1 0-penta aza-2 k5,4k5,6k5-triphosphas piro[4.5]deca -1,3,5- triene (3d)

The work-up procedure was similar to that of compound 1d,using 3 (1.00 g, 1.37 mmol), DASD (0.77 mL, 5.48 mmol) and trieth- ylamine (1.00 mL). Yield: 1.08 g (68%), mp 200 �C. Anal. Calc. for C52H74N9O8Fe2P3: C, 53.94; H, 6.44; N, 10.89. Found: C, 54.25; H, 6.75; N, 10.57%. APIES-MS (fragments were based on 35Cl and 56Fe, Ir %): m/z 959 ([M�(FcCH2]+, 3.4). FTIR (KBr, cm �1): 3091 (C–H arom.), 2954, 2846 (C–H aliph.), 1220 (asymm.), 1195 (symm.) (P@N), 1066 (COC).

2.4.18. 7,11-Dife rrocenyl-2,2,4,4- tetra-1,4-dioxa- 8-azaspiro[4.5] dec- 8-yl-1,3,5,7,1 1-penta aza-2 k5,4k5,6k5-triphosphas piro[5.5]und eca- 1,3,5-trie ne (4d)

The work-up procedure was similar to that of compound 1d,using 4 (0.90 g, 1.21 mmol), DASD (0.68 mL, 4.84 mmol) and trieth- ylamine (0.95 mL). Yield: 0.94 g (66%), mp 220 �C. Anal. Calc. for C53H76N9O8Fe2P3: C, 54.32; H, 6.54; N, 10.76. Found: C, 54.68; H, 6.69; N, 10.41%. APIES-MS (fragments were based on 35Cl and

Page 5: Phosphorus–nitrogen compounds: Part 26. Syntheses, spectroscopic and structural investigations, biological and cytotoxic activities, and DNA interactions of mono and bisferrocenylspirocyclotriphosphazenes

254 N. Asmafiliz et al. / Inorganica Chimica Acta 400 (2013) 250–261

56Fe, Ir %): m/z 1172 ([MH]+, 52.4). FTIR (KBr, cm �1): 3091 (C–Harom.), 2958, 2842 (C–H aliph.), 1220 (asymm.), 1186 (symm.)(P@N), 1056 (COC).

2.4.19. 7,11-Diferro cenyl-2,2,4,4-tetra -1,4-dioxa-8-aza spiro[4.5]dec- 8-yl-1,3,5,7,1 2-penta aza-2 k5,4k5,6k5-triphosphas piro[6.5]dodeca- 1,3,5-triene (5d)

The work-up procedure was similar to that of compound 1d,using 5 (1.00 g, 1.32 mmol), DASD (0.74 mL, 5.28 mmol) and trieth- ylamine (0.90 mL). Yield: 1.06 g (68%), mp 274 �C. Anal. Calc. for C54H78N9O8Fe2P3: C, 54.69; H, 6.63; N, 10.63. Found: C, 54.38; H, 6.54; N, 10.32%. APIES-MS (fragments were based on 35Cl and 56Fe, Ir %): m/z 1185 ([M]+, 65.3). FTIR (KBr, cm �1): 3089 (C–Harom.), 2952, 2840 (C–H aliph.), 1220 (asymm.), 1187 (symm.)(P@N), 1052 (COC).

3. Results and discussion

3.1. Syntheses

The tetrachloro mono (1 and 2) and bisferrocenyl (3–5) spirocy- clotriphosph azenes, as the starting compounds, are obtained according to the published procedure [21]. The compounds 1–5have four reactive Cl atoms which can be substitut ed readily with secondary amines. Utilizing these features the mono- (1a–5a),gem- (1b–5b), tri- (1c–4c), and tetra-DASD- substituted (1d–5d)mono and bisferrocenyl phosphazenes have been synthesized in the presence of Et 3N in refluxing dry THF (Scheme 1).

The proposed reaction mechanism s (Scheme 2), SN 1(P) or SN2(P), depicts tentatively the reaction routes of the mono and bisferrocenyl phosphazenes with DASD for providing a better understand ing of the nucleophilic replacement reactions. The reac- tions of equal amounts of 1–5 and DASD produce correspond ing mono-DASD- substituted ferrocenylphosp hazenes (1a–5a) as major and 1b–5b as by-produ cts. The geminal phosphaz enes (1b–5b) are only isolated from the Cl replacemen t reactions of 1 equiv of 1–5with 2 equiv of DASD. All attempts to obtain cis and/or trans prod- ucts have been failed. Afterwards, the reactions of 1 equiv of 1–5and 3 equiv of DASD afford tri- (1c–4c) and tetra-DASD -substituted (1d–5d) phosphazenes . But, the expected tri-DASD-su bstituted bisferrocenyl phophazene (5c) could not have been obtained.

NHR2

-Et3NHCl

N

( )n

P

NFc CH2 NR1

N

P

N

PCl

ClCl

Cl

-Et3NHCl

(i) SN1(P) NHR2

-Cl-

slowCl

Cl

Cl

R2NP

NP

N

P

N

Cl

δ−

δ+

mono-substituted products

Cl

NR2

P

NP

N

P

N

NR2

Cl Cl

(cis- or trans-)

P

NP

N

P

N

non-geminal products

Cl

NR2R2N

( )n

NFc CH2 NR1

SN2(P) or SN1(P)

NHR2/THFEt3N

NHR2

NCl

Cl

Fc CH2

( )n

NFc CH2 NR1

( )n

NFc CH2 NR1

(ii)SN2(P) or SN1(P)

Scheme 2. The chloride replacement reaction pathway of tetr

Moreove r, when the excess DASD is used, the fully-substitut ed phosphaz enes (1d–5d) have only formed. The yields of gem- and tetra-DASD -substituted mono and bisferrocenyl phosphazenes are considerabl y higher than mono and tri-DASD-su bstituted ones.

In this study, the isolation of only the geminal products (1b–5b)are considerably noteworthy , since the Cl replacement reactions of N3P3Cl6 with the secondar y amines, e.g. pyrrolidine and diethyl amine, afford mainly non-geminal phosphaz enes with usually trans products predominati ng in the literature [1,15]. Whilst, dur- ing the our ongoing studies, the only geminal products are isolated from the reactions of bulky-cryp ta and pendant-mo noferrocenyl spirocyclotr iphosphazenes with the pyrrolidine, morpholine, and 1-aza-12- crown-4-subst ituted, not depending on the bulkiness of these amines [18–21]. Addition ally, it is concluded that, when the different secondary amines were used and more amino substit- uents are added to the cyclotrip hosphazene skeleton, there could be a change in the mechanistic pathway [SN 2(P) to SN 1(P)]. The proposed mechanistic feature may also be contrasted with the behavior of aziridine as nucleophi le in analogous reactions. In this paper, from the chloride replacemen t reactions of tetrachlor o pen- dant-ferroce nyl spirocyclotr iphosphazenes with DASD which is asecondar y amine, the geminal products (1b–5b) have also been isolated by the way of SN 1(P) mechanism . The geminal structure sof 2b and 5b are verified by X-ray crystallo graphy. The structures of 1b–5b are also determined using 1H-coupled 31P NMR spectra. Conseque ntly, the formation of geminal products may be rational- ized with the depending on pendant ferrocenyl arms or bulky- crypta substituents . In addition, tri-subst ituted products (1c–4c)are formed according to the (i) geminal or (ii) possibly non-gemi- nal reaction pathways (Scheme 2).

In addition, some of the cyclotriphos phazenes obtained in this study have stereogeni c phosphorus atoms, and Table 1 outlinesthe optical isomers of the compounds. The mono- (1a and 2a)and tri-DASD- substituted (1c and 2c) monoferr ocenylphospha z- enes have two stereogenic P-centers and they are expected to be in diastereomeric mixtures. But, the mono- (3a–5a) and tri- DASD-su bstituted (3c–4c) bisferrocenyl , and the geminal-DA SD substitut ed monoferroceny lphosphazenes (1b and 2b) have one stereogeni c P-centers and they are expected to be in racemic forms. The 1H, 13C and 31P NMR spectral data indicate that the com- pounds 1a and 2a are the mixtures of diastereoi somers (Fig. 1) and the diastereoiso mers of these compound s were not separated

NHR2

NHR2

SN1(P) -HCl

NHR2

NHR2

fast

SN1(P)

Cl NR2

NR2

tri-substituted products

P

NP

N

P

N

R2N

( )n

NFc CH2 NR1

NPP

NR2

NR2

R2N

R2N

NNP

( )n

NFc CH2 NR1

fully-substituted products

P

P

N

P

N

N

NR2

Cl

Cl

NR2

( )n

NFc CH2 NR1

geminal di-substituted products

NR2

P

NP P

N

( )n

N NR1

+

-Et3NHCl

SN2(P) or SN1(P)

achloro mono and bisferrocenylphosphazenes with DASD.

Page 6: Phosphorus–nitrogen compounds: Part 26. Syntheses, spectroscopic and structural investigations, biological and cytotoxic activities, and DNA interactions of mono and bisferrocenylspirocyclotriphosphazenes

Table 1The expected geometrical and optical isom er distributio ns of the chiral phosphazenes.

Compound Stereogenic P atoms (n) Stereoisomers (2n) (expected) Chirality (expected) Geometrical isomer

3a–5a and 3c–4c 1 1 R Racemic (lines 1/2) –2 S

1b and 2b 1 1 R Racemic (lines 1/2) Geminal 2 S

1a, 2a, 1c and 2c 2 1 RR Racemic (lines 1/4) –2 RS Racemic (lines 2/3)3 SR 4 SS

N. Asmafiliz et al. / Inorganica Chimica Acta 400 (2013) 250–261 255

using column chromatograp hy. The NMR data of the diastereoi so- mers are presente d in Tables 3, S1 and S2 , and compared with the each others.

As an example, the chirality of 3a is unambiguou sly confirmedusing chiral HPLC column. The sample is dissolved in n-hexane–THF (1:2) at a concentration of 10 lg/mL. The experiment is made at room temperature. The peak of the solvent front is considered to be equal to the dead time. It is about 3.04 min at a flow-rate of 1.5 mL/min. There is resolution of 3a with separation factor 0.95

Fig. 1. The 31P NMR spectra of 1a and 1a0 , and 2a and 2a0 . Compounds 1a and 1a0 , and 2a a

in using a mobile phase containing 99% hexane–1% THF at aflow-rate of 1.0 mL/min. As expected, the peak for 3a separates into two peaks of the ratio of ca. 1:2 intensity correspondi ng to the two enantiomers . The peak area of one isomer is 34.5% and the other one is 65.5%, imply that both isomers (R and S) do not form equally. The HPLC chromatogram of 3a is illustrate d in Fig. 2. The two enan- tiomers of 3a are eluted at 25.04 and 26.26 min. The chromato- graphic conditions for HPLC resolution of 3a and results are shown in Table 2.

nd 2a0 are diastereoisomers and 31P NMR data of these compounds may be reversed.

Page 7: Phosphorus–nitrogen compounds: Part 26. Syntheses, spectroscopic and structural investigations, biological and cytotoxic activities, and DNA interactions of mono and bisferrocenylspirocyclotriphosphazenes

Fig. 2. HPLC profile of compound 3a a chiral column (250–4.6 mm) and a solvent system of 99% n-hexane–1% THF at a flow-rate of 1.0 mL/min.

Table 2Chromatographic conditions for TLC and HPLC resolution of racem ic cyclotriphos- phazene derivative (3a).

Compound HPLC parameters a TLCb

t1c t2

d k01e k02

e af RSg Rf

h

3a 25.04 26.26 7.24 7.64 0.95 2.13 0.43

a The mobile phase was a 99:1 (v/v) mixture n-hexane–THF and flow-rate was set at 1.0 ml/min for separation on a Whelk-01 column.

b The mobile phase was a 5:2 (v/v) mixture toluene–THF for separation on an F254 silica gel plate.

c First elution time (min).d Second elution time (min).e Capacity factor, k = ti � t0/t0.f Separation factor, a = k02/k01.g Resolution factor, RS = 1.18(t2 � t1)/(w0.5)1 + (w0.5)2.h Retention factor, Rf = distance traveled by the compound/distance travelled by

the solvent front.

256 N. Asmafiliz et al. / Inorganica Chimica Acta 400 (2013) 250–261

The microanalys es, FTIR, APIES-MS spectra and NMR data are consistent with the proposed structures of the compounds. While the mass spectra of 1a, 1b, 2a, 2b, 3a, 4a, 2c, 5d, and 3b, 5a, 5b,1d, 2d, 4d show the molecular (M+) and protonated molecular ion (MH+) peaks, respectivel y, in the spectra of 1c, 3c, 3d, 4b and4c the M+ peaks do not appear but important fragments are ob- served at m/z 837 [M �(CH2OH)]+ for 1c, at 1033 [M �OH]+ for 3c,at 959 [M �FcCH2]+ for 3d, at 895 [M �(Cl + C2H4)]+ for 4b, and at 1047 [M �OH]+ for 4c.

3.2. NMR and IR spectrosco py

The 1H-decoup led 31P NMR spectral data of the new phospha- zene derivatives are listed in Table 3. The spin systems are assigned as AB 2, AX 2, ABC, AMX, and ABX. The mono- (1a–5a), geminal- (1b–5b), and tri-DASD-su bstituted (1c–4c) ferrocenylphosp hazene derivatives show a 12-line resonance pattern consisting of a dou- blet of doublets for all of the P-atoms. Whilst the different substi- tution patterns of the phospho rus atoms of tetra-DASD -substituted ferrocenylphos phazenes (1d and 3d) give rise to one triplet and one doublet in the 1H-decouple d 31P NMR spectra. The spectra of the other tetra-DASD- substituted ferrocenylphos phazenes (2d, 4dand 5d) exhibit a total of eight signals for AB 2 spin systems (2JPP/Dm = 1.5 for 2d, 1.9 for 4d and 1.5 for 5d) [31]. The geminal struc- tures of 1b–5b are confirmed by 1H-coupled 31P NMR spectral data, indicating that the expected non-geminal products (cis or trans)

could not have been isolated. On the other hand, the mono- DASD-su bstituted ferroceny lphosphazenes (1a and 2a) which have two stereogenic phosphorus centers, exhibit a pair of signals for all the phosphorus atoms, indicating that 1a and 1a0, and 2a and 2a0

are diastereoiso mers (Fig. 1).Taking into account the basis of the chemical shifts, multiplici -

ties, and coupling constant s, the assignment s of the 1H and 13CNMR signals of the DASD-su bstituted ferrocenylphos phazene derivatives are unambiguou sly made (Tables S1 and S2 ). All the DASD-su bstituted ferrocenylphos phazenes give complex 1H NMR spectra (Table S1 ). The spectra of mono- (1a–5a) and tri-DASD- substitut ed (1c–4c) mono and bisferroceny lphosphazenes , and gem-DASD -substituted monoferr ocenylphospha zenes (1b and 2b)are especially highly complex, because all the aliphatic protons are diastereotopi c. The H2, H3, and H4 protons of some of the phosphaz ene derivatives are separated from each other and can be easily distingui shed using HSQC spectra. The HSQC spectrum of 1d is depicted in Fig. S1 , as an example. The FcCH 2N protons of fully-substitut ed ferrocenylphosp hazenes (1d–5d) give rise to doublets due to the vicinal coupling with the 31P nucleus, indicat- ing that both geminal protons are equivalent. The 3JPH values of FcCH2N protons are between 4.8 and 8.6 Hz.

The expected signals of carbons are observed in the 13C NMR spectra of the compound s and all of them are determined (Table S2 ). The Fc CH2, NCH2 (spiro) and NCH2 (DASD) carbon sig- nals are clearly interpreted by the HSQC experiment (Fig. S1 ).The two groups of carbon peaks for OCO, NCH2, OCH2 and NCH 2CH2

(DASD) are easily distinguishable for geminal- (1b and 2b) and tet- rasubstitute d (1d and 2d) monoferroceny l, and trisubstituted mono and bisferrocenyl phosphazenes (1c–4c). Moreover, in the mono-DASD -substituted ferrocenylphos phazenes (1a and 2a)which have two stereogenic phosphorus centers, all the carbon atoms are observed as the pair of signals, indicating that 1a and1a0, and 2a and 2a0 are diastereoi somers. As mentioned before, this situation is clearly distinquisha ble in the 31P and 1H NMR spectra of 1a and 2a.

The compounds 2b and 2c exhibit triplets in the case of dou- blets (Fig. S2 ) for three-bond coupling s of NCH 2CH2 (DASD) with phospho rus atoms. This may be attributed to the second-order ef- fects and it is observed previously [20,21]. The coupling constant s, 3JPC, are calculated using the external transitions of the triplets [32]. The average values of the 3JPC between the ipso-C atoms of the Fc (C1) and P atoms are in the order of 12.1, 7.9, and 4.5 Hz with the compounds six-, five-, and seven-m embered spirorings, respectivel y. The 2JPC values between the NCH2 spiro-carbon atoms and P atoms with the compounds containing five-membered spir- orings (3a, 3b and 3d) are larger than the 2JPC values of the other ones.

The FTIR spectra of all the DASD-substitu ted phosphazenes ex- hibit intense bands between 1247 and 1151 cm �1 related to mPN

bonds of the phosphaz ene rings [6,20,21]. The most characterist ic bands of all the DASD-su bstituted phosphaz enes are assigned for mCOC in the range of 1122–1156 cm �1. As expected, the asymmetr ic and symmetric vibrations of mPCl2 emerge for 1a–5a and 1b–5b, at 576–538 and 548–514 cm �1, respectively . In addition, in tetrasub- stituted cyclotriphosph azenes, the mPCl2 bands are disappeared.

3.3. X-ray structures of 4a, 2b, 5b and 1d

The molecula r and solid-state structure s of 4a, 2b, 5b and 1dalong with the atom-num bering schemes , are illustrated in Figs. 3–6, respectively . The crystallo graphic data are given in Table 4, and selected bond lengths and angles are listed in Table 5. The com- pound 4a contains half of the cyclohexane solvent molecule in the asymmetric unit. The phosphazene rings of 4a and 5b are in twisted-for ms (Fig. S3a ; u2 = �138.4(9)�, h2 = 101.3(9)�, Fig. S4a ;

Page 8: Phosphorus–nitrogen compounds: Part 26. Syntheses, spectroscopic and structural investigations, biological and cytotoxic activities, and DNA interactions of mono and bisferrocenylspirocyclotriphosphazenes

Table 331P NMR (decoupled) spectral data of the compounds. Chemical shifts (d) repor ted in ppm and J values in Hz. a

Compound Spin system d (ppm) 2Jpp

PCl2 P(NR)2 (spiro) P(NR)2 P(NR)(Cl)

1a* AMX 22.04 (dd) 12.57 (dd) – 25.54 (dd) 45.3; 45.4; 56.3 1a0* AMX 22.39 (dd) 14.42 (dd) – 25.93 (dd) 46.4; 46.5; 59.6 2a* ABX 22.83 (dd) 15.52 (dd) – 23.96 (dd) 43.8; 47.2; 62.7 2a0* ABX 22.54 (dd) 15.74 (dd) – 24.25 (dd) 42.0; 46.5; 62.7 3a ABX 25.06 (dd) 18.98 (dd) – 26.48 (dd) 36.0; 43.5; 64.7 4a AMX 22.37 (dd) 14.41 (dd) – 25.93 (dd) 46.3; 46.6; 59.6 5a ABX 21.93 (dd) 16.27 (dd) – 23.75 (dd) 42.1; 43.8; 60.8 1b ABX 23.44 (dd) 22.03 (dd) 16.81 (dd) – 30.5; 47.1; 64.6 2b ABX 23.74 (dd) 17.32 (dd) 17.20 (dd) – 43.3; 44.8; 47.2 3b AMX 24.94 (dd) 20.62 (dd) 18.24 (dd) – 38.5; 46.0; 54.2 4b AMX 18.92 (dd) 11.82 (dd) 2.18 (dd) – 32.4; 44.5; 46.0 5b ABX 22.78 (dd) 18.32 (dd) 17.44 (dd) – 45.3; 45.9; 49.7 1c ABX – 19.87 (dd) 21.86 (dd) 22.65 (dd) 40.0; 45.3; 47.2 2c ABC – 21.47 22.15 22.51 41.3; 47.5; 52.4 3cb ABX – 17.99 2.15 16.80 4c AMX – 16.63 (dd) 1.55 (dd) 14.06 (dd) 36.5; 42.0; 43.2 1d AB2 – 19.38 (t) 20.93 (d) – 39.2 2d AB2 – 21.27 21.97 – 31.3 3d AX2 – 26.38 (t) 22.40 (d) – 41.5 4d AB2 – 22.09 20.63 – 38.1 5d AB2 – 22.04 21.6 – 33.5

a 31 P NMR measurements in CDCl 3 solutions at 293 K. b 2JPP values could not have been calculated due to the overlapping signals.

* Compounds 1a and 1a0 , and 2a and 2a0 are diastereoisomers and 31P NMR data of these compounds may be reversed.

N. Asmafiliz et al. / Inorganica Chimica Acta 400 (2013) 250–261 257

u2 = 161.2(6)�, h2 = 134.5(4)�], while that of 2b and 1d are in boat conformations [Fig. S5a ; u2 = �67.8(3)�, h2 = 76.4(2)�, Fig. S6a ;u2 = 133.7(2)�, h2 = 99.1(2)�] having total puckering amplitudes QT of 0.127(2) Å (for 4a), 0.240(1) Å (for 2b), 0.205(1) Å (for 5b),and 0.268(1) Å (for 1d) [33]. In the gem-DASD- substituted mono and bisferroceny lspirocyclotriphos phazenes (2b and 5b), the se- ven-member ed rings (P3/N4/N5/C1–C4) are in twisted-forms [Fig. S5b ; QT = 0.833(2) Å, u2 = -66.6(2)�, h2 = 33.7(1)�, Fig. S4b ;QT = 0.884(2) Å, u2 = 109.9(2)�, h2 = 32.3(2)�], and in 4a and 1d,the six-membered rings, (P3/N4/N5/C1–C3), are in chair conforma- tions [Fig. S3b ; QT = 0.700(3) Å, u2 = 26.9(2)�, h2 = 90.4(3)� andFig. S6b ; QT = 0.732(2) Å, u2 = �150.8(1)�, h2 = 93.2(1)�].

In the phosphazene rings of 4a, 2b, 5b and 1d, the endocyclic P–N bond lengths are in the ranges of 1.554(2)–1.626(2) Å (for 4 a),1.560(1)–1.628(1) Å (for 2b), 1.558(2)–1.622(2) Å (for 5b), and 1.586(1)–1.603(1) Å (for 1d) and there are regular variations with the distances from P2: P2–N1 � P2–N2 < P1–N3 � P3–N3 < P1–N1� P3–N2 in gem-DASD -substituted ferroceny lphosphazenes (2band 5b). Besides, the average endocyclic P–N bond lengths in phos- phazene rings are 1.586(2), 1.593(1), 1.590(2), and 1.598(1) Å,which are shorter than the exocyclic P–N bonds of 4a, 2b, 5b and1d, respectively (Table 5).

As can be seen from Table 5, in 4a, 2b and 5b the endocylic N2–P3–N3 (a) angles are fairly narrow, whereas in 1d the correspond- ing a angle is almost the same with the ‘‘standar d’’ compound, N3P3Cl6. In addition, the exocyclic N4–P3–N5 (a0) angles are slightly changed. In gem-DASD-sub stituted ferroceny lphosphaz- enes (2b and 5b), the N1–P1–N3 (c) angles are highly narrowed,

whilst the P1–N3–P3 (b) angles are slightly expanded. In N3P3Cl6,the a, a0, b and c angles are 118.3(2)�, 101.2(1)�, 118.3(2)� and121.4(1)�, respectively [34].

On the other hand, in phosphazenes , the PN single and double bonds bonds are generally in the ranges of 1.628–1.691 and 1.571–1.604 Å, respectivel y [35], and they are among the most intriguing bonds in chemistry. It is suggested that, two proposal shave been made for the formatio n of the PN double bonds in phosphaz enes; namely ‘‘island’’ [36] and negative hyperconjug a- tion [37] models. Both of the models have been discussed in the lit- erature [38–41]. In recent years, particular ly, the effects of steric hindranc es, electronic factors, and negative hypercon jugation to the bond lenghts and angles have been discussed in our papers [6,7,20,21]. In this study, the variations of the bond lenghts and an- gles in mono-, gem-, and tetra-DASD- substituted ferroceny lphosp- hazenes are presented above, and these findings are in good agreement with the our previous results. Consequently, it is mentioned that the shortening, narrowing and expansion of the endo- and exocyclic P–N bonds and angles considerably depend on the electronic factors, steric hindrances and negative hypercon jugation.

Furthermore, compound s 2b and 5b have intramolecu lar C–H� � �N hydrogen bonds between Fc-CH 2 and nitrogen atoms of the phosphaz ene rings, indicating that the CH 2 protons of these com- pounds have acidic propertie s (Figs. 4 and 5). In compounds 4a,2b, and 5b, there are intramolecular C–H� � �N and intermolecu lar C–H� � �N and C–H� � �O hydrogen bonds (Figs. 3–5). The intermolecu- lar hydrogen bonds link the molecule s into a two-dimensional net-

Page 9: Phosphorus–nitrogen compounds: Part 26. Syntheses, spectroscopic and structural investigations, biological and cytotoxic activities, and DNA interactions of mono and bisferrocenylspirocyclotriphosphazenes

Fig. 3. ORTEP-3 [42] drawing of 4a with the atom-numbering scheme. Displace- ment ellipsoids are drawn at the 50% probability level. Hydrogen bonds are shown as dashed lines. Hydrogen atoms not involved in hydrogen bonding have been omitted for clarity.

Fig. 4. ORTEP-3 [42] drawing of 2b with the atom-numbering scheme. Displace- ment ellipsoids are drawn at the 50% probability level. Hydrogen bonds are shown as dashed lines. Hydrogen atoms not involved in hydrogen bonding have been omitted for clarity.

Fig. 5. ORTEP-3 [42] drawing of 5b with the atom-numbering scheme. Displace- ment ellipsoids are drawn at the 50% probability level. Hydrogen bonds are shown as dashed lines. Hydrogen atoms not involved in hydrogen bonding have been omitted for clarity.

Fig. 6. ORTEP-3 [42] drawing of 1d with the atom-numbering scheme. Displace- ment ellipsoids are drawn at the 50% probability level. Hydrogen atoms have been omitted for clarity.

258 N. Asmafiliz et al. / Inorganica Chimica Acta 400 (2013) 250–261

work (Figs. S3c and S5c ). In addition, compound 1d has three kinds of intramolecu lar hydrogen bonds; C–H� � �N, C–H� � �O and N–H� � �O(Table S3 ). The p� � �p contacts between the ferrocene rings, Cg2–Cg3 (for 1d), Cg2–Cg3 (for 2b), Cg2–Cg3 and Cg4–Cg5 (for 4a) and Cg2–Cg3 and Cg4–Cg5 (for 5b) [where Cg2, Cg3, Cg4 and Cg5 are the centroids of the rings (C5–C9), (C10–C14), (C16–C20) and (C21–C25) (for 1d and 4a) and (C6–C10), (C11–C15), (C17–C21)and (C22–C26) (for 2b and 5b)] with the centroid–centroid distances of 3.285(1) Å (for 1d), 3.291(1) Å (for 2b), 3.301(2) and 3.290(2) Å(for 4a) and 3.307(2) and 3.293(2) Å (for 5b) may be effective in the stabilizati on of the crystal structure.

3.4. Antimicrobia l activity

The DASD-substitu ted ferrocenylphosp hazene derivatives (3a–5a and 3c) exhibit weak antibacterial activity against B. subtilis ATCC 6633 (G+) and S. aureus ATCC 25923 (G+). The compounds

3a and 4a display moderate antibacterial activity against B. cereus NRRL-B-3 711 (G+) (Section S1).

3.5. Interactions of DNA with the compoun ds 3a–5a, 2b, 4b, 3c, 4c and1d–5d

Interactions of supercoiled pBR322 DNA with the compound sare investigated using the agarose gel electroph oresis. It is ob- served that these compound s are able to cleave the DNA (SectionS2). The presence of the linear form III in the DNA-compo und mix- tures indicates the conformationa l changes of DNA. Moreover, restrictio n analyses indicate that phosphazene derivatives bind to A/A of the DNA.

3.6. Cell viability assay

A large amount of chemical synthesized are characteri zed chemical ly and biologica lly [43,44]. Biological activities of the compound s that can be used as drugs has been focused on their cytotoxic , antimicrobial, antioxidant activities, enzyme inhibition and DNA interaction and many others. DNA is a main target of some anticance r agents. Some drug action depends on DNA cleavage s.

Page 10: Phosphorus–nitrogen compounds: Part 26. Syntheses, spectroscopic and structural investigations, biological and cytotoxic activities, and DNA interactions of mono and bisferrocenylspirocyclotriphosphazenes

Table 4Crystallographic data for 4a, 2b, 5b, and 1d.

4a 2b 5b 1d

Empirical formula C35H46Fe2N6O2P3Cl3 C29H44FeN7O4P3Cl2 C40H54Fe2N7O4P3Cl2 C42H66FeN9O8P3

Fw 893.76 774.37 972.41 973.80 Crystal system triclinic triclinic orthorhombic monoclinic Space group P�1 P�1 Pbca P21/na (Å) 10.7743(3) 11.4651(2) 11.4518(3) 12.1077(3)b (Å) 14.1603(4) 11.6595(2) 19.8455(5) 20.4979(5)c (Å) 14.2453(3) 15.1867(3) 36.5374(8) 18.7391(5)a (�) 60.921(2) 68.810(2) 90.00 90.00 b (�) 79.641(3) 86.808(3) 90.00 104.8400(10)c (�) 82.612(3) 63.050(2) 90.00 90.00 V (Å3) 1866.31(9) 1672.95(5) 8303.7(4) 4495.6(2)Z 2 2 8 4l (cm�1) 1.164 (Mo Ka) 0.801 (Mo Ka) 0.995 (Mo Ka) 0.505 (Mo Ka)q (Calc.) (g cm �1) 1.590 1.537 1.556 1.439 Number of reflections total 33 775 29 229 142 568 42 492 Number of reflections unique 9380 8301 10 689 11 234 Rint 0.0279 0.0210 0.0699 0.0300 2hmax (�) 57.16 56.86 57.40 56.84 Tmin/Tmax 0.7441/0.8448 0.6996/0.8187 0.6630/0.8258 0.8429/0.9057 Number of parameters 478 419 523 572 R [F2 > 2r(F2)] 0.0392 0.0273 0.0356 0.0348 wR 0.0960 0.0700 0.0755 0.0886

Table 5Selected bond lengths (Å) and angles (�) for 4a, 2b, 4b, and 1d.

4a 2b 4b 1d

P1–N1 1.591(2) 1.623(1) 1.622(2) 1.603(1)P1–N3 1.574(2) 1.592(1) 1.593(2) 1.594(1)P2–N1 1.576(2) 1.560(1) 1.566(2) 1.603(1)P2–N2 1.554(2) 1.560(1) 1.603(3) 1.602(1)P3–N2 1.626(2) 1.628(1) 1.558(2) 1.586(1)P3–N3 1.592(2) 1.592(1) 1.600(2) 1.600(1)P3–N4 1.658(2) 1.634(1) 1.637(2) 1.678(1)P3–N5 1.652(2) 1.637(1) 1.644(2) 1.663(1)N1–P1–N3 119.95(12) 114.78(6) 115.40(9) 115.80(7)N1–P2–N2 119.80(13) 122.85(6) 121.54(9) 117.23(7)N2–P3–N3 114.42(13) 114.62(6) 113.26(9) 117.67(17)N5–P3–N4 102.85(11) 104.68(6) 101.45(9) 103.02(7)P1–N1–P2 118.62(15) 119.43(7) 118.57(11) 121.80(8)P3–N2–P2 123.31(14) 119.21(8) 122.47(11) 120.85(8)P1–N3–P3 122.61(15) 124.10(8) 123.90(11) 120.89(8)

N. Asmafiliz et al. / Inorganica Chimica Acta 400 (2013) 250–261 259

In this study, after antimicrobial analysis and DNA cleavage as- say, HeLa cancer cell lines have been used to determine the cyto- toxic effects of 2b, 1d, 2d and 4d on mammalian cells. Paclitaxel and cis-platin are used as positive control. The cell viability is eval- uated following exposure of HeLa cervical cancer cells treatment of the phosphaz enes by 4,5-dimethy lthiazol-2-yl-2,5- diphenyl

Fig. 7. Relative cell viability (%) of HeLa cells following exposure of various concentrationuntreated control cell for 24 h.

tetrazoliu m bromide (MTT) assay [45]. The cells are adjusted to 10,000 cells/well and exposed to 2b, 1d, 2d and 4d at different con- centrations (5 lg/mL, 10 lg/mL, 25 lg/mL, 50 lg/mL, 75 lg/mL,100 lg/mL and 200 lg/mL) for 24 h. Following exposure of cells with MTT dye for 4 h, formazan products are dissolved in DMSO and then the absorbance of the solution is measured by spectroscopy at a wavelength of 540 nm. The results exhibit that 25 lg/mL 2b decreases cell viability about 60%, 50 lg/mL 2d 75%,5 lg/mL 4d 50% and 25 lg/mL cisplatin 57%, 25 lM paclitaxel 58% (Fig. 7). The results show that the anti-prol iferative activity of compound 4d is higher against HeLa cancer cell lines than the other compounds , cisplatin and paclitaxel. In addition, compond 1d (5 lg/mL, 10 lg/mL, 25 lg/mL and 50 lg/mL) has very high anti-prol iferative activity on HeLa cervical cancer cells. Thus, three different compounds (1d, 2d and 4d) which have different cyto- toxic effects on HeLa cervical cells are selected. To test the poten- tial effect of these compounds, the cell survival within 96 h is basically scrutinized (Fig. 8). The most cytotoxic compound 4d de-creases cell count in the first 24 h and 1d decrease s drastically after 48 h. Therefore, 4d and 1d may be candidat e for potential drugs.

Recently , it has been observed that some of the cyclopho spha- zene derivatives and poly(2-dimethylamino ethylamino )phospha-zene were very active against A549, HCV29T, HL-60 and Neuro 2A cells [46,47]. On the other hand, in contrast to our results it is

s of compounds 2b, 1d, 2d, 4d, cisplatin and paclitaxel between 5 and 50 lg/mL and

Page 11: Phosphorus–nitrogen compounds: Part 26. Syntheses, spectroscopic and structural investigations, biological and cytotoxic activities, and DNA interactions of mono and bisferrocenylspirocyclotriphosphazenes

Fig. 8. Trypan blue exclusion assay is preceded in HeLa cervical cancer cells to determine cell survival in the presence of 1d, 2d and 4d. The cell survival is determined following exposure of HeLa cells to 1d, 2d and 4d (10 lg/mL 1d, 5 lg/mL 4d) for 96 h. The each line graph represents the average of two replicates.

260 N. Asmafiliz et al. / Inorganica Chimica Acta 400 (2013) 250–261

reported that cyclophosphaz ene hydrazone derivatives are not suf- ficiently effective against HeLa cell lines and they conclude that adding more biologically compatible spacer groups may help the activity of compounds [48]. In this study, it can be thought that pendant ferrocenyl and four DASD groups may increase the anti- proliferative effects of cyclotriphos phazene rings against HeLa cell lines.

4. Conclusion

The reactions of tetrachlor o mono (1 and 2) and bisferrocenyl- spirocyclotri phosphazenes (3–5) with DASD have led to the formation of mono- (1a–5a), gem- (1b–5b), tri- (1c–4c), and tet- ra-DASD-subs tituted (1d–5d) ferrocenylcyclotr iphosphazen es via the gradually chloride replacemen t reactions. When the reactions have been carried out with 1 equiv of 1–5 and 2 equiv of DASD, the geminal products (1b–5b) are isolated. The tri-DASD-subs ti- tuted products are likely to be the modal compounds containing a single replaceable Cl atom because they can be easily converted to oxygen bridge phosphaz ene derivatives. The fully-DASD- substi- tuted ferrocenylcyclotr iphosphazenes can be thought of the good candidates as possible ligating agents for transition metal cations. Besides, mono- (1a–5a), gem- (1b and 2b), and tri-DASD-su bsti- tuted (1c–4c) mono and bisferroceny lphosphazenes have stereo- genic P-center(s), and the chirality of 3a is determined using chiral HPLC column, as an example. The solid-state structures of 4a, 2b, 5b and 1d reveal the intra- and/or intermolecu lar hydrogen bondings. Furthermor e, the compounds 3a–5a and 3c have poten- tial antibacterial activities against Gram (+) bacteria, except Enterococcus faecalis ATCC 292112. Interactions between the com- pounds 3a–5a, 2b, 4b, 3c, 4c and 1d–5d and pBR322 plasmid DNA exhibit that the compounds are effective in changing the mobility of the DNA. The restriction analyses disclose that some phospha- zene derivatives bind to A/A of the DNA. Compound 4d has been proven to be stronger anti-proliferative activity than the cisplatin and paclitaxel against HeLa cancer cell lines. Thus, compound 4dmay be considered as the agent with potential antitumor activity and seems to be a good potential candidat e for further studies.

Acknowled gements

The authors acknowledge the ‘‘Ankara University, ScientificResearchs Unit’’ Grant No. 12B4240004 and also thank ‘‘Medicinal Plants and Medicine Research Center of Anadolu University, Eskis�ehir’’ for using the X-ray facility. T.H. is indebted to ‘‘Hacet- tepe University, Scientific Research Unit’’ (Grant No. 02 02 602 002) for their financial support. L.A. thanks ‘‘State Planing Organizatio n’’ (Grant No. 1998K1214 80).

Appendi x A. Supplementar y material

Supplement ary data associated with this article can be found, in the online version, at http://dx.doi .org/10.1016/j. ica.2013.03.001 .

References

[1] R.A. Shaw, Phosphorus, Sulfur 45 (1989) 103. [2] Y. Pareek, M. Ravikanth, Chem. Eur. J. 18 (2012) 8835. [3] D. Kumar, N. Singh, K. Keshav, A.J. Elias, Inorg. Chem. 50 (2011) 250. [4] M.R. Rao, G. Gayatri, A. Kumar, G.N. Sastry, M. Ravikanth, Chem. Eur. J. 15

(2009) 3488. [5] E.E. _Ilter, N. Çaylak, M. Is �ıklan, N. Asmafiliz, Z. Kılıç, T. Hökelek, J. Mol. Struct.

697 (2004) 119. [6] M. Is �ıklan, N. Asmafiliz, E.E. Özalp, E.E. _Ilter, Z. Kılıç, B. Ços�ut, S. Yes �ilot, A. Kılıç,

A. Öztürk, T. Hökelek, L.Y. Koç Bilir, L. Açık, E. Akyüz, Inorg. Chem. 49 (2010)7057.

[7] A. Okumus �, Z. Kılıç, T. Hökelek, H. Dal, L. Açık, Y. Öner, L.Y. Koç, Polyhedron 30 (2011) 2896.

[8] N.N. Bhuvan Kumar, K.C. Kumara Swamy, Chirality 20 (2008) 781. [9] S. Bilge, S�. Demiriz, A. Okumus �, Z. Kılıç, B. Tercan, T. Hökelek, O. Büyükgüngör,

Inorg. Chem. 45 (2006) 8755. [10] G. Yenilmez Çiftçi, H. Dal, E. Tanrıverdi Eçik, T. Hökelek, A. Kılıç, E. S�enkuytu,

Polyhedron 29 (2010) 1209. [11] N. Satish Kumar, K.C. Kumara Swamy, Polyhedron 23 (2004) 979. [12] C.W. Allen, Chem. Rev. 91 (1991) 119. [13] V. Chandrasekhar, S. Karthikeyan, S.S. Krishnamurthy, M. Woods, Indian J.

Chem. 24A (1985) 379. [14] H.R. Allcock, U. Diefenbach, S.R. Pucher, Inorg. Chem. 33 (1994) 3091. [15] V. Chandrasekhar, V. Krishnan, Adv. Inorg. Chem. 53 (2002) 159. [16] K.C. Kumara Swamy, M.D. Poojary, S.S. Krishnamurthy, H. Manohar, J. Chem.

Soc., Dalton Trans. (1985) 1881. [17] A.A. Van der Huizen, J.C. van de Grampel, J.W. Rusch, T. Wilting, F. Bolhius, A.

Meetsma, J. Chem. Soc., Dalton Trans. (1986) 1317. [18] Z. Kılıç, A. Okumus �, S�. Demiriz, S. Bilge, A. Öztürk, T. Hökelek, J. Incl. Phenom.

Macrocycl. Chem. 65 (2009) 269. [19] N. Asmafiliz, E.E. _Ilter, Z. Kılıç, T. Hökelek, E. S�ahin, J. Chem. Sci. 120 (2008) 363. [20] E.E. _Ilter, N. Asmafiliz, Z. Kılıç, L. Açık, M. Yavuz, E.B. Bali, A.O. Solak, F.

Büyükkaya, H. Dal, T. Hökelek, Polyhedron 29 (2010) 2933. [21] N. Asmafiliz, Z. Kılıç, A. Öztürk, T. Hökelek, L.Y. Koç, L. Açık, Ö. Kısa, A. Albay, Z.

Üstündag, A.O. Solak, Inorg. Chem. 48 (2009) 10102. [22] H. _Ibis�og ˘lu, G. Yenilmez Çiftçi, A. Kılıç, E. Tanrıverdi, _I. Ün, H. Dal, T. Hökelek, J.

Chem. Sci. 121 (2009) 125. [23] S. Kumaraswamy, M. Vijjulatha, C. Muthiah, K.C. Kumara Swamy, U.

Engelhardt, J. Chem. Soc., Dalton Trans. (1999) 891. [24] C.M. Myer, C.W. Allen, Inorg. Chem. 41 (2002) 60. [25] S. Sengupta, Tetrahedron Lett. 44 (2003) 7281. [26] C. Nataro, C.N. Myer, W.M. Cleaver, C.W. Allen, J. Organomet. Chem. 637–639

(2001) 284. [27] K. Muralidharan, A.J. Elias, Inorg. Chem. 42 (2003) 7535. [28] BRUK ER program 1D WIN-NMR (release 6.0) and 2D WIN-NMR (release 6.1).[29] Bruker, SADABS, Bruker AXS Inc., Madison, Wisconsin, USA, 2005. [30] G.M. Sheldrick, SHELXS-97; SHELXL-97, University of Göttingen, Göttingen,

Germany, 1997. [31] F.A. Bovey, Nuclear Magnetic Resonance Spectroscopy, Academic Press, San

Diego, CA, 1988. p. 159. [32] V. Vicente, A. Fruchier, H.J. Cristau, Magn. Reson. Chem. 41 (2003) 183. [33] D. Cremer, J.A. Pople, J. Am. Chem. Soc. 97 (1975) 1354. [34] G.J. Bullen, J. Chem. Soc. A (1971) 1450. [35] F.H. Allen, O. Kennard, D.G. Watson, L. Brammer, G. Orpen, R.J. Taylor, J. Chem.

Soc., Perkin Trans. 2 (1987) 1. [36] M.J.S. Dewar, E.A.C. Lucken, M.A. Whitehead, J. Chem. Soc. (1960) 2423.

Page 12: Phosphorus–nitrogen compounds: Part 26. Syntheses, spectroscopic and structural investigations, biological and cytotoxic activities, and DNA interactions of mono and bisferrocenylspirocyclotriphosphazenes

N. Asmafiliz et al. / Inorganica Chimica Acta 400 (2013) 250–261 261

[37] A.B. Chaplin, J.A. Harrison, P.J. Dyson, Inorg. Chem. 44 (2005) 8407. [38] S.S. Krishnamurthy, Sulfur Silicon Relat. Elem. 87 (1994) 101. [39] E.W. Ainscough, A.M. Brodie, A.B. Chaplin, A. Derwahl, J.A. Harrison, C.A. Otter,

Inorg. Chem. 46 (2007) 2575. [40] R.J. Davidson, E.W. Ainscough, A.M. Brodie, J.A. Harrison, M.R. Waterland, Eur.

J. Inorg. Chem. (2010) 1619. [41] D.G. Gilheany, Chem. Rev. 94 (1994) 1339. [42] L.J. Farrugia, J. Appl. Crystallogr. 30 (1997) 565. [43] M. Siwy, D. S�ek, B. Kaczmarczyk, I. Jaroszewicz, A. Nasulewicz, J. Med. Chem.

49 (2006) 806.

[44] Y.S. Sohn, Y.J. Jun, Polyphosphazenes for Biomedical Applications, John Wiley &Sons, Inc., Hoboken, New Jersey, 2009. p. 249 (Chapter 14).

[45] S. Patel, N. Gheewala, A. Suthar, A. Shah, Int. J. Pharm. Pharm. Sci. 1 (2009) 38. [46] M. Siwy, D. Sek, B. Kaczmarczyk, J. Wietrzyk, A. Nasulewicz, A. Opolski,

Anticancer Res. 27 (2007) 1553. [47] H.K. de Wolf, M. de Raad, C. Snel, M.J. van Steenbergen, M.H.A.M. Fens, G.

Storm, W.E. Hennink, Pharm. Res. 24 (2007) 1572. [48] B.R. Patil, S.S. Machakanur, R.S. Hunoor, D.S. Badiger, K.B. Gudasi, S.W. Annie

Blighb, Pharm. Chem. 3 (4) (2011) 377.


Recommended