Muhammad S. Khan*a, Muna R. A. Al-Mandharya, Mohammed K. Al-Sutia, Timothy C. Corcorana, Yaqoub Al-Mahrooqia, J. Paul Attfieldb, Neil Feederb, William I. F. Davidc, Kenneth Shanklandc, Richard H. Friendd, Anna Köhlerd, Elisabeth A. Marseglia*d, Emilio Tedescod, Chiu C. Tange, Paul R. Raithby*f, Jonathan C. Collingsg, Karl P. Roscoeg, Andrei S. Batsanovg, Lorna M. Stimsong and Todd B. Marder*g
aDepartment of Chemistry, Sultan Qaboos University, P.O. Box 36, Al Khod 123, Sultanate of Oman. E-mail: msk@squ.edu.om
bDepartment of Chemistry, University of Cambridge, Lensfield Road, Cambridge, UK CB2 1EW
cISIS Facility, Rutherford Appleton Laboratory, Chilton, Didcot, Oxon, UK OX11 0QX
dCavendish Laboratory, University of Cambridge, Madingley Road, Cambridge, UK CB3 0HE
eCLRC Daresbury Laboratory, Daresbury, Warrington, UK WA4 4AD
fDepartment of Chemistry, University of Bath, Claverton Down, Bath, UK BA2 7AY. E-mail: p.r.raithby@bath.ac.uk
gDepartment of Chemistry, University of Durham, South Road, Durham, UK DH1 3LE. E-mail: Todd.Marder@durham.ac.uk
First published on 25th November 2002
A series of 1,4-diethynylbenzene (1) derivatives, H–C
C–R–C
C–H with R
=
C6H3NH2
(2), C6H3F (3), C6H2F2-2,5 (4), C6F4
(5), C6H2(OCH3)2-2,5 (6) and C6H2(OnC8H17)2-2,5 (7) has been synthesised and their crystal structures determined by single crystal (2–5) or powder (6, 7) X-ray diffraction. The C
CH⋯πC
C hydrogen bonds dominating structure 1 are gradually replaced by C
C–H⋯F ones with the increase of fluorination (3
→
5), or completely replaced by C
CH⋯N and NH⋯πC
C bonds in 2, and C
CH⋯O in 6 and 7. The related platinum-based polymers, trans-[–Pt(PnBu3)2–C
C–R–C
C–]n
(R
=
as above and C6H4,) have been prepared and characterised by spectroscopic methods and thermogravimetry, which show that the amino- and methoxy-derivatives have lowest thermal stability while the fluorinated ones exhibit increasing thermal stability with increasing fluorination. Optical spectroscopic measurements reveal that substituents on the aromatic spacer group do not create strong donor–acceptor interactions along the rigid backbone of the organometallic polymers.
Recently, attention has also been directed toward oligomers, taken not only as model compounds, but also as potential materials for device applications.6 They are more crystalline than the corresponding polymers, allowing for a detailed structural analysis and, thus, an assessment of the structure–electronic property relationship.7 There is some evidence that the solid state arrangement of these molecules represents an important factor in determining their optical behaviour.5–8 While the π–π stacking interaction between adjacent oligomer molecules or polymer chains has been identified as a key ‘interchain’ interaction in solids,9 it is possible that hydrogen bonding can also play an important rôle in this context.
Hydrocarbon groups are generally poor donors of hydrogen bonds, but the ethynyl moiety with its acidic H atom is an exception.10 Furthermore, the concentration of electron density in the triple bond can act as an acceptor, albeit a rather weak one, of an H-bond.11 Thus ethynyl groups can form co-operative H-bonds,12 similarly to OH groups.13 Such a cooperative effect is evident in the crystal structure14 of 1,4-diethynylbenzene 1. In order to investigate the effect of various substituents on the crystalline architecture of these organic poly-yne precursors, particularly the stability of the C
CH⋯πC
C‡ H-bonding in competition15 with other intermolecular interactions, e.g. X–H⋯πC
C
(X
=
N, C) and C–H⋯Y (Y
=
N, O, F16), we synthesised a series of amino-, fluoro- and alkoxy-derivatives of 1,4-diethynylbenzene (2–7, see Scheme 1) and structurally characterised them by single crystal and powder X-ray diffraction. The structures of 4 and 5 were determined at variable temperature, to investigate the supposed effects of the latter on weak hydrogen bonds.17
![]() | ||
Scheme 1 Compounds HC C–R–C CH. | ||
In addition, these substituted diethynylbenzene ligands have been incorporated into rigid-rod platinum(II)-containing polymers and their thermal and opto-electronic properties investigated systematically.
![]() | ||
Scheme 2 Reagents: (i) Pd(OAc)2, PPh3, CuI, iPr2NH/CH2Cl2; (ii) KOH, MeOH/THF, or Bu4NF, THF; (iii)
trans-[PtCl2(PnBu3)2], CuI, iPr2NH/CH2Cl2. R = C6H41, 8; C6H3NH22, 9; C6H3F 3, 10; C6H2F2-2,5 4, 11; C6F45, 12; C6H2(OMe)2-2,5 6, 13; C6H2(OnC8H17)2-2,5 7, 14. | ||
Compared to the parent 1, the amino, fluoro and the alkoxy derivatives are relatively stable at low temperature in the absence of light and air; if exposed to either, the initially colourless solids slowly turn off-white, yellow, light brown or light purple. Long storage times at ambient temperature and under aerobic conditions led to the formation of some insoluble material (<5%), which was presumed to be a polymerisation product.
The platinum-containing poly-yne polymers were obtained by the organometallic polycondensation reaction of the diethynylbenzene derivatives with one equivalent of trans-[Pt(PnBu3)2Cl2] in CH2Cl2/iPr2NH solution in the presence of catalytic quantities of CuI (Scheme 2). The resultant polymers were purified by passing through a short alumina column, and isolated as off-white or yellow solids. The materials were characterised by spectroscopic techniques. The IR spectra showed a single ν(C
C) stretching frequency at 2095 cm−1 consistent with the trans-arrangement of the ethynyl groups. The 31P NMR spectrum of each polymer displayed the expected singlet signal at approximately δ
−138 ppm with JPt-P coupling confirming the trans-configuration of the phosphine ligands. The 1H NMR spectra of the polymers show the expected resonances corresponding to the substituted 1,4-diethynylbenzene derivatives. The weight average molecular weights (Mw) of the polymers indicate a high degree of polymerisation. The number average molecular weight (Mn) values are in the range of 74
000 to 95
000 g mol−1. These values correspond to between 95 and 128 repeat units in the polymer chain. The molecular weights should be viewed with caution in view of the difficulties associated with utilizing GPC for rigid-rod polymers. GPC is not a direct measure of molecular weight but a measure of the hydrodynamic volume. Rod-like polymers in solution possess very different hydrodynamic properties from flexible polymers. Thus, by GPC using randomly coiled polystyrene standards, the observed average molecular weights of rigid rod poly-ynes are likely to be inflated to some extent relative to the actual molecular weights. However, the lack of discernable resonances that could be attributed to end groups in the NMR spectra provides support for the view that there is a high degree of polymerisation in these poly-ynes.
C bonds) H-bonding sites are not only congruous (moderately polar) but also equal in numbers. This favours the formation of a two-dimensional net of co-operative H-bonds, linking molecules into a puckered layer (Fig. 1, Table 1). In both recent studies this layer was described as parallel to the (1 0 0), i.e. bc, plane.14b,c In fact, as can be seen from Fig. 2, it is parallel to the (1 0 −2) plane. It is noteworthy that the same mistake was repeated in the description of three crystal structures of p-XC6H4C
CH (X
=
Cl, Br or I), pseudo-isomorphous with 1.19 There are infinite π–π stacks between molecules of adjacent layers (related by the a translation), at normal interplanar separations (d in Table 1).![]() | ||
| Fig. 1 Crystal structures of 1 (a), 3 (b) and 4 (c); projections on the (1 0 −2) plane. | ||
| Compound | 114b | 114c | 3 | 4(A) | 4(B) | 5(A) | 5(B) | 5(C) |
|---|---|---|---|---|---|---|---|---|
| a See notation in Figs. 1 and 4.b Mean interplanar separation in the π–π stack.c Probably erratum in the original, cannot be verified as atomic coordinates are unavailable. | ||||||||
| T/K | 295 | 125 | 180 | 180 | 120 | 273 | 180 | 150 |
| (i)a H⋯π/Å | 2.68 | 2.60 | 2.75 | 2.87 | 2.85 | 2.94 | 2.93 | 2.89 |
| (i) C–H⋯π/° | 176 | 175 | 168 | 161 | 160 | 129 | 127 | 127 |
| (ii) H⋯F/Å | 2.66 | 2.60 | 2.58 | |||||
| (ii) C–H⋯F/° | 121 | 122 | 122 | |||||
| (iii) H⋯F/Å | 2.86 | 2.94 | 2.88 | 2.49 | 2.41 | 2.41 | ||
| (iii) C–H⋯F/° | 90 | 116 | 117 | 132 | 136 | 134 | ||
| db/Å | 3.71c | 3.53 | 3.47 | 3.43 | 3.41 | 3.63 | 3.60 | 3.59 |
![]() | ||
| Fig. 2 Crystal structure of 1, viewed down the y axis. | ||
Introduction of a polar (amino) group in 2 contributes two additional “active” H atoms, but only one potential acceptor site (the lone pair of N). The prominent feature of this structure (Fig. 3) is a ring system of co-operative H-bonds, involving four molecules. The ring (having crystallographic Ci symmetry) includes two relatively strong N–H⋯πC
C bonds (H⋯π
=
2.51 Å, N–H⋯π
=
145°)§ and two C
CH⋯N bonds (H⋯N
=
2.44 Å, C–H⋯N
=
159°), directed towards the lone pair of the substantially pyramidalised N atom (the valent bond angles at N average 116°). As each molecule participates in two such tetrameric systems, this generates a molecular layer, parallel to the crystallographic (1 0 −2) plane. The other amino-H atom is involved in a bifurcated interaction with two (crystallographically non-equivalent) C
C bonds of two different molecules (H⋯π
=
2.89 and 2.87 Å, N–H⋯π
=
128 and 140°, respectively), belonging to the next layer. The ethynyl group not involved in the co-operative H-bonding points almost perpendicularly towards the benzene ring of one of these molecules (C–H vector/ring plane angle of 81°), with the shortest contacts H⋯C(4) 3.04 and H⋯C(5) 2.98 Å typical for van der Waals interactions.20 The outcome is a peculiar orientation of molecular planes: normal to each other but all normal to the layer plane. Molecules belonging to different layers (as defined by cooperative H-bonds) are nevertheless engaged in π–π stacking with each other (interplanar separations of 3.56 Å), albeit with considerable offset. Unlike all other compounds reported herein, molecule 2 has no crystallographic symmetry and is slightly ‘bent’, the directions of the two bonds forming an angle of 170.8(2)°.
![]() | ||
Fig. 3 (a) Crystal structure of 2, projected on the (1 0 −2) plane; (b) H-bonds and intermolecular contacts. Symmetry transformations: (i) 1 − x, ½ + y, ½ − z; (ii) 2 − x, 1 − y, 1 − z; (iii)
x + 1, ½ − y, ½ + z; (iv)
x − 1, y, z; (v)
x, ½ − y, ½ + z; (vi) 2 − x, y − ½, ½ − z. | ||
Crystal structures of 3 and 4
(Fig. 1) are isomorphous with that of 1. In each case, the molecule lies at a crystallographic inversion centre, which means that in 3 the fluorine atom is disordered, being distributed equally between two positions, related by this centre. Hydrogen bonding in 1, 3 and 4 is compared in Table 1. The ‘organic’
(i.e. bonded to carbon) fluorine atom, notwithstanding its high formal electronegativity, is known to be very poor H-bond acceptor, and is able to participate in them only in the absence of competition from stronger ones, e.g. oxygen.21 However, structures 3, 4 and 5 satisfy the conditions for the formation of C–H⋯F hydrogen bonds, as pointed out by Desiraju et al. in a study on fluorobenzenes:22 they contain only C, H and F atoms, and the H atoms are bonded to sp2 or sp-hybridised C atoms. Indeed, in 3 and 4 the C
CH⋯πC
C H-bond of 1 is weakened, or rather replaced by a bifurcated H-bond to both the C
C group and the fluorine atom of the same molecule, the C
CH⋯F interaction being the stronger of the two. Thus, the H⋯π distance increases by ca. 0.2 Å from that in 1 to that in 4. There is also a much longer, but probably electrostatically attractive, contact (iii) with another F atom. Comparison of the structure of 4 at different temperatures (180 and 120 K) shows similar (ca. 0.02 Å) thermal expansions of the H-bonds (i) and (ii) and the interplanar separations (d) in the π–π stack, but a larger extension of contact (iii). The layers in all three structures are similarly puckered: the dihedral angles between H-bonded molecules are 48°
(1), 52°
(3) and 54°
(4).
The increased degree of fluorination in compound 5 gives rise to completely different crystal packing (Fig. 4). The molecule also possesses crystallographic Ci symmetry, but the principal synthon in structure 5 is the pair of inversion-related, parallel and nearly coplanar molecules, linked by two C
CH⋯F hydrogen bonds (iii), shorter than in 3 and 4
(see Table 1). Thus molecules are linked into an infinite ‘ribbon’, parallel to the (1 0 −1) direction. There are no continuous π–π stacks of molecules, although each molecule has one neighbour contacting it in a π–π fashion. These molecules, related by an inversion centre, have a longitudinal (parallel to the C
C bond) shift of ca. 4.3 Å, so that a C
C bond of one molecule overlaps with the benzene ring of another. The interplanar separation contracts from 3.63 Å at 273 K to 3.59 Å at 150 K.
![]() | ||
| Fig. 4 Crystal structure of 5. | ||
This double ribbon is surrounded on all sides by four other ribbons, propagating in the same direction but with perpendicular molecular planes. Each ethynyl H atom participates in a rather long inter-ribbon C
CH⋯πC
C contact (i)
(Table 1); the interaction thus can be interpreted as a very asymmetric bifurcated H-bond. Structure determinations at 273, 180 and 150 K did not reveal any substantial change, although thermal expansion in the direction of the strongest H-bond (iii) is somewhat greater than in other directions.
The crystal packing of di-alkoxy derivatives 6 and 7 is shown in Figs. 5 and 6. Both molecules have crystallographic Ci symmetry and adopt planar (6) or nearly planar (7) conformations. Structure 6 comprises puckered layers, parallel to the (1 0 −2) plane and broadly similar to those observed in structures 3 and 4, in which the molecules are linked by C
CH⋯O hydrogen bonds (H⋯O
=
2.34 Å, C–H⋯O
=
146°), while the C
CH⋯πC
C interaction between the same molecules is much weaker than in structures 3 and 4
(H⋯π
=
3.07 Å, C–H⋯π
=
137°). Unlike 1–4, structure 6 contains no π–π stacks, molecules of adjacent layers contacting in a herring-bone fashion.
![]() | ||
| Fig. 5 Crystal structure of 6. | ||
![]() | ||
| Fig. 6 Crystal structure of 7. | ||
Structure 7 contains nearly-flat layers, in which a synthon, broadly similar to that of structure 5, can be identified: a centrosymmetric pair of molecules, linked by two C
CH⋯O hydrogen bonds. These, however, are much weaker than in 6
(H⋯O
=
2.80 Å, C–H⋯O
=
167°). Probably, the crystal structure of 7 is dominated by the close packing of (all-trans) n-alkyl chains, to which the H-bonding pattern has to adjust. This structure is in good agreement with the structural model proposed recently for the corresponding polymer.3b
°C) from thermal analysis
![]() | ||
| Fig. 7 Simultaneous thermogravimetric (TG) curve (above) and differential thermal analysis (DTA) curve (below) for 10. The onset and peak decomposition temperatures, as defined in the text, are marked. | ||
![]() | ||
| Fig. 8 Absorption and luminescence spectra of the platinum-containing poly-yne polymers 8, 12 and 13 taken from thin films at room temperature with excitation at 3.4 eV. | ||
The emission spectra consist of two bands. The band just below the onset of absorption at 3 eV is emission from the S1 state (fluorescence). The band below 2.5 eV has been well characterised for the polymer 8 and other aromatic/hetero-aromatic spacers by time-resolved luminescence measurements and is assigned to emission from the triplet excited state T1 (phosphorescence).8,24 Substitution of the aromatic ring shifts the energy of the T1 emission by about the same amount as the S1 emission, so that the S1–T1 energy difference remains constant. This can be used to accurately fine-tune the energy of the T1 state for energy harvesting purposes25 particularly in analogous organic polymers, where T1 emission (phosphorescence) is spin-forbidden and cannot be observed. While such a constant S1–T1 energy gap has been observed for these organometallic and analogous organic polymers with a variety of spacers its cause is not yet fully understood.23,25
The emission spectra normalised to the fluorescence peak for the polymers with one, two and four fluorine substituents (polymers 10, 11 and 12, respectively) are shown in Fig. 9. The relative intensity of phosphorescence increases strongly with the fluorine content.
![]() | ||
| Fig. 9 Room temperature thin-film luminescence spectra of platinum-containing poly-yne polymers 10, 11 and 12 normalised to the peak of the S1 emission and shown on a logarithmic scale. | ||
The packing analysis of the structures of the polymer precursors has also thrown light on the co-presence of weak hydrogen bonding in model compounds of polymers that are of technological interest. In summary the ethynylenic moiety provides a good hydrogen bond donor and acceptor, which is able to compete in terms of intermolecular interactions and as structure defining synthons, with more electronegative atoms, such as nitrogen. C
CH⋯πC
C hydrogen bonds dominate the crystal structure of 1, and are gradually replaced by C
C–H⋯F ones with the increase of fluorination (3
→
5), as has been noted previously,19 or completely replaced by C
CH⋯N and NH⋯πC
C bonds in 2, and C
CH⋯O in 6 and 7. However, it is hard to judge, based solely on geometrical measurements, the relative strength of the C–H⋯F interactions in these derivatives. A C–H⋯πC
C bond in 1 has been estimated being of ca. 4 kJ mol−1.26 When alkoxy substituents are present, C–H⋯O interactions occur, but the presence of long alkyl chains in the structure can impede the formation of these short contacts.
The substituted 1,4-diethynyl benzene ligands have also been incorporated into rigid-rod Pt(II) polymers and their thermal and opto-electronic properties investigated. The substituents on the aromatic spacer have a major effect on the thermal stability of the polymeric complexes but their electronic properties are affected only to a minor extent.
NMR spectra were recorded on Bruker WM-250 or AM-400 spectrometers in CDCl3 and the 1H and 13C{1H} were referenced to solvent resonances. Infrared spectra were recorded as CH2Cl2 solutions, in a NaCl cell, on a Perkin-Elmer 1710 FT-IR spectrometer, mass spectra on a Kratos MS 890 spectrometer by the electron impact (EI) technique. Microanalyses were performed in the Department of Chemistry, University of Cambridge. Column chromatography was performed in Kieselgel 60 (240–400 mesh) silica gel. GC/MS analyses were performed on a Hewlett-Packard 5890 Series II/5971A MSD instrument equipped with an HP 7673A autosampler and a fused silica column (30 m
×
0.25 mm
×
0.25 μm, cross-linked 5% phenylmethyl silicone). The following operating conditions were used: injector, 260
°C; detector, 280
°C; oven temperature was ramped from 70° to 260
°C at the rate of 20
°C min−1; helium (UHP grade) was used as the carrier gas; toluene (BDH, 99.7%) was used as an internal integration standard.
°C for 20 h when TLC, GC analysis and IR spectroscopy indicated that all the starting material had been consumed and the coupling reaction was completed. After being cooled to room temperature, the mixture was filtered to eliminate the ammonium salt and the solvent mixture was removed in vacuo.
C–), 3300 (C
CH), 3428 (–NH–). 1H NMR (250 MHz, CDCl3): δ 3.12 (s, 1H, C
C–H), 3.50 (s, 1H, C
C–H), 4.70 (br s, 2H, NH2), 7.21 (d, 2H, ar), 7.73 (dd, 1H, ar). 13C NMR (100.6 MHz, CDCl3): δ 78.2, 80.6, 83.1, 84.0 (C
C), 107.2 (C1), 117.5 (C3), 121.4 (C4), 123.0 (C6), 132.2 (C5), 148.0 (C2). EI mass spectrum: m/z 141 (M+). Found: C, 85.03; H, 4.94; N, 9.90; Calc. for C10H7N: C, 85.08; H. 5.00; N, 9.92%.
C–) and 3300 (–C
C–H). 1H NMR (250 MHz, CDCl3): δ 3.21 (s, 1H, C
C–H), 3.38 (s, 1H, C
C–H), 7.22 (d, 2H, ar) and 7.46 (d, 1H, ar). 13C NMR (100.6 MHz, CDCl3): δ 77.1, 80.2, 81.8, 84.1 (C
C), 111.51 (C1), 119.02 (C3), 124.36 (C4), 128.10 (C6), 134.20 (C5) and 162.65 (C2). EI mass spectrum: m/z 144 (M+). Found: C, 83.29; H, 3.58. Calc. for C10H5F: C, 83.32; H, 3.50%.
C–) and 3299 (–C
C–H). 1H NMR (250 MHz, CDCl3): δ 3.41 (s, 2H, C
CH), 7.24 (d, 2H, ar), 13C NMR (100.6 MHz, CDCl3): δ 75.7, 85.5 (C
C), 112.8 (C1,4), 120.2 (C3,6), 158.9, (C2,5). EI mass spectrum: m/z 162 (M+). Found C, 74.11; H, 2.45; Calc. for C10H4F2: C, 74.08; H. 2.49%.
C–) and 3297 (–C
C–H). 1H NMR (250 MHz, CDCl3): δ 3.72 (s, 2H, C
CH). 13C NMR (100.6 MHz, CDCl3): δ 68.4, 91.5 (C
C), 104.5 (C1,4), 147.5 (C2,3,5,6). EI mass spectrum: m/z 198 (M+). Anal. Found: C, 60.71; H, 1.04; Calc. for C10H2F4: C, 60.63; H, 1.02%.
C–) and 3299 (–C
C–H). 1H NMR (250 MHz CDCl3): δ 3.33 (s, 2H, C
C–H), 3.92 (s, 6H, OCH3), 7.19 (s, 2H, ar). 13C NMR (100.6 MHz, CDCl3): δ 69.89 (OCH3), 79.96, 82.57 (C
C), 113.41 (C1,4), 117.89 (C3,6), 154.14 (C2,5). EI mass spectrum: m/z 186 (M+). Found: C, 77.46; H, 5.43; Calcd for C12H10O2: C, 77.40; H, 5.41%.
C–) and 3299 (–C
C–H). 1H NMR (250 MHz CDCl3): δ 0.86 (t, 6H, CH3), 1.26, 1.40 (both m, 20H, CH2), 1.78 (m, 4H, CH2), 3.32 (s, 2H, C
C–H), 3.95 (t, 4H, OCH2), 6.93 (s, 2H, ar). 13C NMR (100.6 MHz, CDCl3): δ 14.29 (CH3), 22.85, 26.09, 29.31, 29.42, 29.48, 31.99 (all CH2), 69.84 (OCH2), 79.98, 82.60 (C
C), 113.45 (C1,4), 117.92 (C3,6), 154.18 (C2,5). EI mass spectrum: m/z 382 (M+). Found: C, 81.52; H, 10.04; Calc. for C26H38O2: C, 81.63; H, 10.01%.
C–C6H3(NH2)–C
C)–]n9. CuI (10 mg) was added to a mixture of trans-[Pt(PnBu3)2Cl2]
(0.670 g, 1.0 mmol) and 1,4-diethynyl-2-amimobenzene (0.141 g, 1.0 mmol) in CH2Cl2/iPr2NH (50 cm3, 1∶1 v/v). The yellow solution was stirred at room temperature over a period of 15 h, after which all volatile components were removed under reduced pressure. The residue was dissolved in dichloromethane and filtered through a short alumina column. After removal of solvent by rotary evaporator, an off-white solid of polymer 9 was obtained in 85% yield (0.750 g). Further purification can be accomplished by precipitating the polymer from dichloromethane solution in methanol. IR (CH2Cl2): ν/cm−1 2094 (–C
C–). 1H NMR (250 MHz, CDCl3): δ 0.87 (t, 18H, CH3), 1.40 ((sex, 12H, CH2), 1.65 (brs, 12H, CH2), 2.25 (m, 12H, PCH2), 4.18 (s, 2H, NH2), 6.58 (d, 2H, ar), 7.12 (d, IH, ar). 31P{1H} NMR (101.3 MHz, CDCl3): δ
−138.08, JPt-P
=
2330 Hz. Found: C, 55.17; H, 8.14; Calc. for (C34H59NP2Pt)n: C, 55.26; H. 8.05%. GPC (THF): Mn
=
94
700 g mol−1
(n
=
128), Mw
=
151
500 g mol−1, polydispersity
=
1.6.
C–C6H3F–C
C)–]n10. Off-white solid (82% yield). IR (CH2Cl2): ν/cm−1 2095 (–C
C). 1H NMR (250 MHz, CDCl3): δ 0.82 (t, 18H, CH3), 1.38 (sex, 12H, CH2), 1.58 (brs, 12H, CH2), 2.29 (m, 12H, PCH2), 7.38 (d, 2H, ar), 7.28 (d, IH, ar). 31P{1H} NMR (101.3 MHz, CDCl3): δ
−138.50, JPt-P
=
2330 Hz. Found: C, 55.15; H, 7.74; Calc. for (C34H57P2FPt)n: C, 55.05; H. 7.76%. GPC (THF): Mn
=
94
650 g mol−1
(n
=
128), Mw
=
160
900 g mol−1, polydispersity
=
1.7.
C–C6H2F2–C
C)–]n11. Off-white solid (78% yield). IR (CH2Cl2): ν/cm−1 2094 (–C
C–). 1H NMR (250 MHz, CDCl3): δ 0.77 (t, 18H, CH3), 1.34 (sex, 12H, CH2), 1.52 (brs, 12H, CH2), 2.20 (m, 12H, PCH2), 7.28 (d, 2H, ar). 31P{1H} NMR (101.3 MHz, CDCl3): δ
−138.50, JPt-P
=
2330 Hz. Found: C, 53.65; H, 7.44; Calc. for (C34H56P2F2Pt)n: C, 53.74; H. 7.43%. GPC (THF): Mn
=
88
230 g mol−1
(n
=
116), Mw
=
158
810 g mol−1, polydispersity
=
1.8.
C–C6F4–C
C–]n12. Off-white solid (72% yield). IR (CH2Cl2): ν/cm−1 2095 (–C
C–). 1H NMR (250 MHz, CDCl3): δ 0.75 (t, 18H, CH3), 1.38 (sex, 12H, CH2), 1.58 (brs, 12H, CH2), 2.20 (m, 12H, PCH2). 31P{1H} NMR (101.3 MHz, CDCl3): δ
−138.50, JPt-P
=
2330 Hz. Found: C, 51.36; H, 6.79; Calc. for (C34H54P2F4Pt)n: C, 51.31; H. 6.84%. GPC (THF): Mn
=
82
500 g mol−1
(n
=
103). Mw
=
156
770 g mol−1, polydispersity
=
1.9.
C–C6H4(OCH3)2–C
C–]n13. Light yellow solid (85% yield). IR (CH2Cl2): ν/cm−1 2095 (–C
C–). 1H NMR (250 MHz, CDCl3): δ 0.86 (t, 18H, CH3), 1.38 (sex, 12H, CH2), 1.58 (brs, 12H, CH2), 2.20 (m, 12H, PCH2), 3.70 (s, 6H, OCH3), 7.10 (s, 2H, ar). 31P{1H} NMR (101.3 MHz, CDCl3): δ
−138.10, JPt-P
=
2330 Hz. Found: C, 55.22; H, 8.03; Calc. for (C36H62P2O2Pt)n: C, 55.16; H. 7.97%. GPC (THF): Mn
=
74
100 g mol−1
(n
=
95), Mw
=
111
200 g mol−1, polydispersity
=
1.5.
C–C6H4(OC8H17)2–C
C–]n14. Light yellow solid (90% yield). IR (CH2Cl2): ν/cm−1 2095 (–C
C–). 1H NMR (250 MHz, CDCl3): δ 0.84 (t, 24H, CH3), 1.22–1.38 (m, 32H, CH2), 1.62 (m, 16H, CH2), 2.20 (m, 12H, PCH2), 3.75 (t, 4H, OCH2), 6.82 (s, 2H, ar). 31P{1H} NMR (101.3 MHz, CDCl3): δ
−138.10, JPt-P
=
2330 Hz. Found: C, 61.36; H, 9.29; Calc. for (C50H90P2O2Pt)n: C, 61.26; H. 9.25%. GPC (THF): Mn
=
94
850 (n
=
97) g mol−1, Mw
=
151
750 g mol−1, polydispersity
=
1.6.
°C running in THF at 1 cm3 min−1 with a Roth Mocel 200 high precision pump. A DAWN DSP (Wyatt Technology) Multi-Angle Laser Light Scattering (MALLS) apparatus with 18 detectors and auxiliary Viscotek model 200 differential refractometer/viscometer detectors was used to calculate the molecular weights (referred to GPC LS).![[small lambda, Greek, macron]](https://www.rsc.org/images/entities/i_char_e0cc.gif)
=
1.54178 Å); for 4(B) and 5(C) on a SMART 3-circle diffractometer with a 1K CCD area detector, using graphite-monochromated Mo-Kα radiation (![[small lambda, Greek, macron]](https://www.rsc.org/images/entities/i_char_e0cc.gif)
=
0.71073 Å). In all cases, crystals were cooled with Cryostream (Oxford Cryosystems) open-flow N2 gas cryostat. Absorption corrections were performed for 3 and 4
(A and B) by a semi-empirical method, based on ψ-scans. The structures were solved by direct methods using the SHELXS-86 program31
(or SIR9232 for 5) and refined by full-matrix least squares against F2, using SHELXL-97 software.33CCDC reference numbers 196099–196102. See http://www.rsc.org/suppdata/nj/b2/b206946f/ for crystallographic data in CIF or other electronic format.
| 2 | 3 | 4(A) | 4(B) | 5(A) | 5(B) | 5(C) | 6 | 7 | |
|---|---|---|---|---|---|---|---|---|---|
| a Total number of inequivalent reflections in the powder pattern;b Rwp for the Rietveld fits;c Rp for the Rietveld fits. | |||||||||
| Formula | C10H7N | C10H5F | C10H4F2 | C10H2F4 | C12H10O2 | C26H38O2 | |||
| Mw/g mol−1 | 141.17 | 144.14 | 162.13 | 198.12 | 186.20 | 380.55 | |||
| T/K | 180 | 180 | 180 | 120 | 273 | 180 | 150 | 290 | 290 |
| λ/Å | 1.54178 | 1.54178 | 1.54178 | 0.71073 | 1.54178 | 1.54178 | 0.71073 | 1.2996 | 1.54178 |
| Crystal system | monoclinic | monoclinic | monoclinic | monoclinic | monoclinic | triclinic | |||
| Space group | P21/c | P21/c | P21/c | P21/c | P21/c | P![]() | |||
| a/Å | 5.919(2) | 3.878(1) | 3.8535(7) | 3.828(1) | 6.1510(6) | 6.112(1) | 6.109(2) | 9.1718(1) | 7.6769(8) |
| b/Å | 12.030(4) | 5.947(1) | 5.906(1) | 5.887(1) | 11.524(1) | 11.463(2) | 11.456(4) | 6.04823(7) | 11.682(2) |
| c/Å | 11.026(3) | 15.800(5) | 16.221(5) | 16.204(4) | 5.8204(7) | 5.764(1) | 5.751(2) | 9.54469(1) | 6.926(2) |
| α/° | 90 | 90 | 90 | 90 | 90 | 90 | 90 | 90 | 104.29(1) |
| β/° | 104.17(2) | 93.46(3) | 95.23(3) | 95.21(1) | 98.71(1) | 99.27(2) | 99.35(2) | 107.428(1) | 96.637(8) |
| γ/° | 90 | 90 | 90 | 90 | 90 | 90 | 90 | 90 | 99.671(9) |
| V/Å3 | 761.2(4) | 363.7(2) | 367.6(2) | 363.7(1) | 397.1(2) | 398.6(1) | 397.1(2) | 505.17(1) | 586.0(2) |
| Z | 4 | 2 | 2 | 2 | 2 | 2 | 2 | 2 | 1 |
| μ/mm−1 | 0.57 | 0.77 | 1.03 | 0.12 | 1.42 | 1.45 | 0.16 | 0.67 | 0.51 |
| Refls. total | 1043 | 920 | 510 | 4764 | 753 | 510 | 4179 | ||
| Refls. unique | 959 | 460 | 466 | 966 | 691 | 506 | 907 | 243a | 70a |
| Rint | 0.031 | 0.026 | 0.026 | 0.020 | 0.040 | 0.059 | 0.023 | — | — |
Refls., I > 2σ(I) | 777 | 369 | 431 | 902 | 601 | 333 | 807 | ||
| wR(F2) | 0.136 | 0.122 | 0.073 | 0.106 | 0.129 | 0.072 | 0.085 | 0.070b | 0.031b |
R[I > 2σ(I)] | 0.050 | 0.045 | 0.028 | 0.034 | 0.043 | 0.035 | 0.031 | 0.045c | 0.024c |
=
1.2996 Å) using a flat plate geometry, that for 7 on a Stoe powder X-ray diffractometer equipped with a linear PSD, using Cu-Kα radiation (![[small lambda, Greek, macron]](https://www.rsc.org/images/entities/i_char_e0cc.gif)
=
1.5406 Å) from a sealed-tube source. Both diffraction patterns were indexed using the DICVOL91 program34 and each structure solved using the DASH suite of programs.35 The structure solution details are summarised in Table 4.
| 6 | 7 | |
|---|---|---|
| Data range used/degrees | 5–40 | 7–34 |
| Number of intensities extracted | 74 | 66 |
| χ2 for Pawley profile fit | 7.67 | 2.94 |
| Number of atoms in model, excluding dummy | 12 | 33 |
| Internal degrees of freedom | 1 | 8 |
| Initial simulation temperature (χ2 units) | 300 | 100 |
| Final simulation temperature (χ2 units) | 280 | 4.5 |
| Initial χ2 for intensities | 3740 | 1825.6 |
| Final χ2 for intensities | 63.1 | 24.4 |
χ2 for Rietveld profile fit of SA model (refine scale + ITF only) | 18.17 | 5.61 |
It was clear from the space groups and lattice parameters that the molecules in both structures lie at crystallographic inversion centres and have no translational degree of freedom. The trial structures were subjected to an optimisation in which torsion angles were the only internal degrees of freedom, and limits on the external degrees of freedom were derived from the Euclidean normalisers of the relevant space groups,36 while the molecule was anchored on a dummy atom lying at the origin of the unit cell. Structure solutions were deemed to have been obtained when the χ2 figure of merit fell below a predetermined value.
Finally, the fractional coordinates obtained at the end of the simulated annealing runs were verified by Rietveld refinements using the GSAS program.37 In both cases, the scale factor, cell constants, and parameters describing a linear interpolated background function and the diffraction peak shape were varied. The highly-resolved synchrotron powder pattern of 6 contained sufficient data for the atomic coordinates and isotropic U-factors of all non-hydrogen atoms to be refined independently while the H atom positions were restrained to idealised positions, giving an excellent fit with the final χ2
=
5.32. The powder diffraction pattern of 7 contained an amorphous component at 2θ
>
34°, therefore only the data below this limit were used in the solution and subsequent Rietveld refinement. The refinement of atomic coordinates being unstable, these were fixed and an overall isotropic U-factor was refined, giving a good fit with the final χ2
=
2.24.
SHELXTL software38 was used for analysing the geometry and for graphical presentation of all structures (1–7).
°C min−1 to 465
°C. The thermocouple readings were calibrated using a series of DTA standard materials: KNO3, In, Sn, Ag2SO4, and K2SO4 as well as Pb and Al as secondary standards, using the same heating rates as the samples.
613 CAS; J. D. Dunitz and R. Taylor, Chem. Eur. J., 1997, 3, 89 CrossRef CAS.Footnotes |
| † Electronic supplementary information (ESI) available: atomic cooordinates for 6 and 7. See http://www.rsc.org/suppdata/nj/b2/b206946f/ |
‡ πC C is the midpoint of the C C bond. |
| § All hydrogen bond parameters are calculated for idealized bond lengths C–H 1.08 Å and N–H 1.01 Å. |
| This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2003 |