摘要:

J O U R N A L O F C H E M I S T R Y Materials Feature Article Properties and substitutional chemistry of layered lead cuprate superconductors T. P. Beales MetalManufactures L td, High T emperature Superconductor Facility, Australian T echnology Park, Eveleigh NSW 1430, Australia This article reviews the relationship between structure and properties for a series of layered Pb-containing superconducting cuprates from the research viewpoint of the solid state chemist or materials physicist.The 0211, 1201, 1212, 1222 and 2213 structures containing (Pb,M)O layers and their derivatives are considered with the emphasis on the general physiochemical relationships observed between crystal structure, allovalent ion chemical substitution, materials processing and optimisation of superconductivity.It is now ten years since the discovery of high temperature coupling breaks down and the flux cylinders move uncoupled through the CuO5 planes. For practical applications, T * needs superconductivity in cuprates. If these materials are to enjoy widespread application in high and low power electrical engin- to be close to 77 K. To overcome the intrinsic problems of the known cuprate eering, then they will need to function in devices operating at, or just below, 77 K.Experience shows that it is near to 20 materials many groups across the world began to search for new materials that would keep the advantages of Y-123 i.e. years for a product engineered from a new material or process to enjoy commercial success from the point of innovation.If good flux pinning and simple chemical structure but would also overcome its disadvantages i.e. oxygen non-stoichiometry, high temperature superconductors follow this pattern, we are already half way into the cycle. Our present state of understand- orthorhombic–tetragonal distortion and chemical reactivity. Any new material would also need to incorporate the advan- ing of the materials in the liquid nitrogen regime is that each of the known cuprate superconductors has specific strengths tages of the Bi- and Tl-containing superconductors i.e.at least two CuMO layers containing the supercurrent (to ensure a TC and weaknesses. YBa2Cu3O7-d (Y-123) can be processed in bulk1 and thin above 77 K) and should overcome anisotropy problems by resembling Y-123 in having an eYcient non-superconducting film form2 to give high critical current densities (JC) at 77K in excess of 1×109A m-2.The drawbacks of Y-123 mainly arise dopant layer which was as atomically thin as possible (preferably a single non-superconducting oxide layer).12 from its crystal structure. The orthorhombic unit cell of Y-123 requires careful control during processing in both thin films Two likely candidate systems were discovered in the late 1980s: (Pb,Cu)-1212 with a general formula of (Pb1-yCuy)- and in bulk, as the biaxial alignment of Y-123 grains necessary for high JC values is diYcult to achieve.Labile oxygen on the Sr2(Y1-xCax)Cu2O7-d and Pb-2213, with a general formula of Pb2Sr2(Y1-xCax)Cu3O8+d which partly fitted the desired CuO2 chain site leads to Y-123 exhibiting an orthorhombic– tetragonal transition3,4 and the CuO2 chain site is also chemi- material requirements in several areas and were closely allied to Y-123 in crystal structure. This review attempts to chart the cally reactive.5 The corresponding bismuth-, thallium- and mercury-con- progress since then in the layered Pb-containing cuprates.Other ‘1212’ materials containing either Tl or Hg alone in the taining layered cuprates all have highly anisotropic layered crystal structures with macroscopic plate-like crystals.As a rocksalt structure and also other stoichiometries such as 2212 are not included in this review and are covered adequately consequence of this, they have been used to fabricate long conductors in the form of wires and tapes for high power elsewhere in the literature.13,14 applications.6 These materials have an advantage over Y-123 in that they have melting points below 960 °C, allowing them to be sinter-processed in an inert metallic silver sheath.The YSr2Cu3O7-d: the base unit for the layered Pb higher order Bi-, Tl- and Hg-containing layered cuprates (n>2) cuprates also have higher superconducting transition temperatures (TC) YSr2Cu3O7-d is derived from the substitution of Sr into than Y-123 (TC=92 K7) (TC=125 K for Tl-22238 and 110 K YBa2Cu3O7-d .The smaller ionic radius of SrII (cf. BaII) prevents for Bi-22239). The volatility of Hg and Tl combined with their a stable YSr2Cu3O7-d structure forming when synthesised toxicity has made their use in these materials unattractive.using normal solid state calcination procedures (maximum Sr Both the Bi- and Tl-containing superconductors are inferior substitution up to 60%)15–18 and requires static high pressure to Y-123 in regards to their magnetic flux pinning capabilities synthetic techniques.19 YSr2Cu3O7-d has a more tetragonal at 77 K, which is weak, and small magnetic fields applied structure than YBa2Cu3O7-d with a shorter a and c axis.19 along the crystallographic c axis severely limit their current- The smaller YSr2Cu3O7-d unit cell volume reflects a more carrying capability.10 compressed structure with a corresponding higher Cu oxi- The intrinsic flux pinning behaviour of layered high temperadation state.YSr2Cu3O7-d can be chemically stabilised by ture superconductors can be understood in terms of their cationic substitution as explored by Slater and Greaves,20 crystal structure.11 In these materials, the magnetic flux lines Sunshine et al.,18 Den and Kobayashi21 and others.These resemble a set of cylinders with radius equal to the coherence workers have shown that for two particular cases, four and length (l) and height equal to the width of the CuO5 planes six coordinate cations, the ‘123’ structure obeys general in the unit cell.Below a temperature T *, these cylinders are substitutional rules. Four coordinate substitution is favoured flux-pinned through the non-superconducting layers by a Josephson coupling interaction. Above T *, the Josephson in YSr2Cu3-xMxO7-d for small M, e.g. AlIII, GaIII or CrVI.J. Mater. Chem., 1998, 8(1), 1–12 1Six-fold coordinate substitution is favoured for larger M and for d values near full oxygenation and represents the incorporation of MO6 units into the CuO2 chain sites e.g. TiIV, MoVI or WVI. It is possible to obtain mixed coordination states in the same compound20 e.g. in YSr2Cu2FeO7.42, the formal Fe= 3.84 oxidation state implies a mix of FeIII/FeIV, most probably in four and six oxygen co-ordination respectively.It is possible to link the (Pb,M)-1212, Y-123 and Pb-2213 phases by comparing the unit cell structures and highlighting similar features. There is a common chemical block found in all three materials containing CuO5 square basal pyramid units able to transmit the supercurrent. This is the CuO5 plane or perovskite block, and is relatively stable to chemical attack.It also contains a group 2 ion astride the corner sharing CuO5 pyramids and a layer of lanthanide or (Ln/group 2 ion solid solution) which has no oxygen layer associated with it. In Y- 123, this layer contains Ba and Y ions respectively: in (Pb,M)- 1212, it contains Sr and Ln/Ca ions respectively and in Pb- 2213, it contains Sr/Ba and Y/Ca ions respectively. The perovskite unit is schematically shown in Fig. 1 and has the Fig. 3 Pictorial representation of the building block structure of the general chemical formula YSr2Cu2O6. ‘123’ phase The structural diVerence in each of the materials arises at the ends of the CuO5 units, where a new block consisting of ions able to act as charge reservoirs for doping the CuO5 orthorhombic–tetragonal transition.The Cu ions in the CuO2 planes is present. This block is unique to each structure. How site are also chemically diVerent from those in the CuO5 site successful this oxide layer is in accepting and accommodating and are more susceptible to chemical attack and substitution. electronic charge determines the overall degree of three dimen- The advantage of the CuO2 chains is that they endow on Ysionality of the material. 123 a higher dimensionality by allowing the flux cylinders in Fig. 2 shows the connecting block for the ‘123’ structure and adjacent perovskite blocks to couple through them resulting is composed of CuO2 chains. It is this block that eVectively in a high T *. Fig. 3 shows a representation of how the CuO2 determines the properties of Y-123: its labile oxygen produces units and YBa2Cu2O5 units in Y-123 are arranged.charge carriers from non-stoichiometry and its ability to undergo structural rearrangement is the root cause of the The (Pb,Cu)-1212 phase: (Pb1-yCuy)Sr2(Y1-xCax)Cu2O7-d In chemical terms, one can envisage the doping of Pb into the Sr-containing Y-123 phase in an idealised formula of YSr2(Cu3-yPby)O7-d.When this type of chemical substitution was tried, it was found that the Pb did preferentially substitute into the CuO2 site in the Y-123 structure as expected, but in doing so formed a new crystal structure. The Pb ions adopted a rocksalt arrangement rather than the perovskite structure favoured in the CuO2 chains. This new phase was called the ‘1212’ phase [usually written as (Pb1-yCuy)Sr2(Y1-xCax)- Cu2O7-d or (Pb,Cu)-1212].It can be thought of as a shear translation of the chain site oxygen in Y-123 by (c, 0, 0) to form the new layer shown schematically in Fig. 4. (The ‘123’ structure can really be thought of as a special case of the ‘1212’ structure i.e. CuO-YSr2Cu2O6). Fig. 5 shows how the rock salt and perovskite blocks fit together in the ‘1212’ structure.(Pb,Cu)-1212 was first synthesised with nominal stoichiometries of (Pb0.69Cu0.31)Sr2(Y0.85Ca0.15)Cu2O7-d22 and (Pb0.29Cu0.71)Sr2(Y0.73Ca0.27)Cu2O7-d23 and was originally thought not to be superconducting. A noticeable fact about the synthesis of (Pb,Cu)-1212 is that phase pure samples are diYcult to prepare, and (Pb,Cu)-1212 Fig. 1 Illustration of the YSr2Cu2O6 chemically stable perovskite block found in the ‘123’, ‘1212’ ‘1222’ and ‘2213’ structures Fig. 4 Illustration of the (Pb,M)O rocksalt structure block sited at Fig. 2 Illustration of the CuO2 chain block sited at the apical oxygen the apical oxygen of the YSr2Cu2O6 perovskite block in the ‘1212’ structure of the YSr2Cu2O6 perovskite block in the ‘123’ structure 2 J.Mater. Chem., 1998, 8(1), 1–12to d=0.1 for slow cooled samples for all values of x. In this system, superconductivity was observed between the narrow ranges 0.3x0.35 with TC onset values between 20 and 40 K. Fig. 6 shows a 3D projection of the observed literature values of TC onsets in (Pb1-yCuy)Sr2(Y1-xCax)Cu2O7-d plotted against the Pb and Ca contents determined after using diVerent processing conditions.The observed TC values show a strong correlation with the Ca content of (Pb,Cu)-1212 in the excess Pb content region with the highest TC onset values near to x=0.7. From the oxygen concentration measurements, it can be shown that increasing the Ca concentration alone is not solely responsible for increasing the hole concentration in (Pb,Cu)- 1212, but samples do become more metallic and show an increase in both the Hall and Seebeck coeYcients with increasing Ca concentration.32 In quenched samples with fixed Ca contents, those with d=0.0 show a lower resistivity and lower Hall and Seebeck coeYcients than for samples with d>0.0, indicating that quenching also increases the carrier density.These same samples also showed a positive magnetoconduc- Fig. 5 Pictorial representation of the building block structure of the tance and a ln(T ) behaviour at temperatures 4 K. This is ‘1212’ phase usually associated with the Kondo eVect and can be explained as arising from localised cationic distributions. Neutron diVraction data33 on (Pb0.65Cu0.35)Sr2(Y0.7Ca0.3)Cu2O7-d samples have been reported for the nominal compositions (Pb0.5Cu0.5)Sr2YCu2O7-d ,24 (Pb1-yCuy)Sr2(Y0.5Ca0.5)- showed that oxygen vacancies were statistically located on the (0, c, 0) site in the (Pb,Cu)O layer (i.e.equivalent to the O Cu2O7-d (0.2y0.5)25 and (Pb(1+x)/2Cu(1-x)/2)Sr2- (Y1-xCax)Cu2O7-d (x0.35)26 using conventional solid state chain site in Y-123). The evidence from neutron diVraction data implies that it is these oxygen vacancies that cause a synthesis techniques.The exact limits of the Ca solid solution range are diYcult to define from the literature XRD patterns random potential to occur in the (Pb,Cu)O layer which in turn leads to a decrease in carrier concentration. At present, due to an ambiguity in the phase purity. High Ca substitution levels (0.5) require both the increased substitution of Pb for it is not possible to diVerentiate whether this arises from Anderson localisation or from disruption of the PbIV–PbII Cu in the rocksalt layer and careful control of the overall oxygen stoichiometry (best achieved by quenching from high metavalence necessary to provide charge carriers in the selfdoping mechanism in these materials.temperatures).Using this procedure Naqvii and Boyd,27 synthesised reasonably phase pure XRD patterns for compositions of (Pb1-yCuy)Sr2(Y1-xCax)Cu2O7-d (0.25y0.35 and Substitutions for Cu: the x0.8). Phase pure samples of (Pb,Cu)-1212 have been syn- (Pb1-yMy)Sr2(Y1-xCax)Cu2O7-d series thesised using a laser ablation technique,28 which indicate that the Ca solid solubility limit is close to x=0.7.Both the solid The ‘1212’ lattice consists of stable perovskite YSrCu2O6 blocks state and thin film synthetic procedures agree that the TC value surrounding rocksalt (Pb,Cu)O layers. The thermodynamic is related to the amount of Ca present. A TC onset=82 K in stability of these YSrCu2O6 blocks means that chemical substi- (Pb0.75Cu0.25)Sr2(Y0.2Ca0.8)Cu2O7-d produced by the solid tution invariably takes place in the rocksalt layer [except in state route27 and a TC onset=86K in (Pb0.75Cu0.25)Sr2 the case where M=Tl which commonly cross substitutes for (Y0.3Ca0.7)Cu2O7-d produced by laser ablation28 are in Ca in the (Y1-xCax) layer].As will be described in more detail good agreement and suggest that this is probably the max- in this section, there is a strong correlation between the eVective imum TC value in this material.No evidence of supercon- ionic radius of the M ion and the maximum TC onset values ductivity has been reported in the Ca-free compound in the (Pb1-yMy)Sr2(Y1-xCax)Cu2O7-d series. Fig. 7 shows a (Pb0.5Cu0.5)Sr2YCu2O7-d . plot of the maximum observed TC values in selected (Pb,M)- In addition to the inducement of superconductivity by 1212 compositions versus the eVective ionic radius of the Mn+ doping the Y site with Ca, the properties of (Pb,Cu)-1212 show substituted into the (Pb,M)O layer.The synthesis and a marked sensitivity in its oxygen stoichiometry, which can be readily altered by processing in high pressure O2 gas, especially during post-synthesis annealing. For example, Tang et al.29 showed that the resultant oxygen stoichiometry in (Pb0.5Cu0.5)Sr2(Y0.6Ca0.4)Cu2O7-d was dependent on the sample cooling rate after calcination (at 940 °C).TC values below 10 K were observed for samples cooled at less than 10 K min-1 and TC values up to 67 K were observed for samples cooled at a rate above 10 K min-1. The low TC value for the slow-cooled samples was attributed to labile oxygen loss during cooling.A PbIV oxidation state was calculated from iodometric titration for these samples, the high TC samples exhibited a value of d=0.01 and the lower TC samples d>0.01. A value of d=0.01 in the above compound yields a net Cu valency= +2.2, which is close to the optimum value30 of+2.15 (assuming an overall valency of the rocksalt layer=+3) and can explain the higher TC value.Similar monitoring of the oxygen concentration by Kurusu et al.31 using coulometric titration in Fig. 6 3D plot of the eVect on the transition temperature observed in (Pb(1+x)/2Cu(1-x)/2)Sr2(Y1-xCax)Cu2O7-d (x0.35) showed (Pb,Cu)-1212 by altering the stoichiometry of Pb and Ca in (Pb1-yCuy)Sr2(Y1-xCax)Cu2O7-d (data taken from ref. 22–33) that samples quenched from 800 °C in air had d=0.0, compared J.Mater. Chem., 1998, 8(1), 1–12 3after the nominal XRD phase solid solution limit of Ca substitution is reached. This suggests that the mechanism for superconductivity in the group 2 1212 phases is due mainly to the doping eVect of CaII for YIII and not to an ability to selfdope as in (Pb,Cu)-1212. The mixed ion rocksalt layer induces a random oxygen stoichiometry which can be, in part, compensated for by a post synthesis high pressure O2 anneal.Oxygen vacancies have been directly observed in both (Pb,Ca)-1212 and (Pb,Sr)-1212 where values of d=0.3 were determined for stoichiometries of (Pb0.5M0.5)Sr2(Y0.5Ca0.5)Cu2O7-d (M=Sr or Ca),37,38 showing highest TC onset temperatures of 104 and 80 K, respectively.Single crystals of (Pb,Sr)-1212 have been grown from a self-flux, but only show a TC onset of 50 K.39 In low oxygen content (Pb,Sr)-1212 and (Pb,Ca)-1212, non-commensurate satellite reflections are observed in the electron diVraction data which are attributed to oxygen vacancies at the apical CuO5 rocksalt site. These features were seen to reduce in intensity after high pressure O2 annealing.(Pb,Sr)- 1212 also displays further reflections arising from an n×a Fig. 7 A plot of the observed TC onset value for a range of M ions superstructure along the a axis attributed to Pb and Sr cation substituted into the (Pb,M)-1212 structure versus the eVective ionic disordering.40 Similar cation disorder phenomena have been radius of Mn+ ions observed in the optical spectroscopy data from (Pb,Sr)-1212 single crystals.41 properties of each group of the Mn+ cations in the (Pb,M)- 1212 structure, according to their position in the Periodic Substitution by the d block elements Table, are discussed in the following section.Transition metals have been substituted into the (Pb,M)O layer to completely replace the Cu ions. Beales et al.42 syn- Substitution by group 2 metals (M=Mg, Ca and Sr) thesised a series of samples with the nominal stoichiometry (Pb0.5M0.5)Sr2(Y1-xCax)Cu2O7-d (M=Cd, Zn, Cu and The earliest substitution work in the 1212 system was the substitution of Cu by the group 2 metals.As in (Pb,Cu)-1212, 0.0x0.5). They found that pure ‘1212’ samples were formed at 750T /°C900 when x=0.0 but that only the phase purity is diYcult to achieve using conventional solid state synthetic techniques and secondary phases of SrCuO2, (Pb0.5Cd0.5)Sr2YCu2O7-d samples were superconducting as synthesised with TC onset=60 K. They concluded that the Sr2CuO3 and Sr5Pb3CuO12 regularly appear in the XRD patterns of published work.Maignan et al.34 used TEM and (Pb,Cd)O layers were able to spontaneously create charge carriers in the CuO5 layers, probably via a PbII–PbIV redox EDAX to carry out detailed analysis of 1212 phases.They found that cross-site cation substitution was common, along reaction. (Pb0.5Cu0.5)Sr2YCu2O7-d and (Pb0.5Zn0.5)- Sr2YCu2O7-d did not superconduct even with high pressure with the presence of secondary phases calculated to be in the order of 10% per sample.For the (Pb,Sr)-1212 phase O2 anneals, suggesting that interstitial oxygen is not generating charge carriers in these materials. As x was increased, the (Pb0.5Sr0.5)Sr2(Y0.5Ca0.5)Cu2O7-d, actual stoichiometries were found to be (Pb0.64Cu0.2Sr0.16)Sr2(Y0.6Ca0.32Sr0.19)Cu2O7-d phase purity of the samples began to fall and Sr2CuO3, SrCuO2 and Cd1-xCaxO impurity phases began to form.The highest and (Pb0.66Sr0.34)Sr2(Y0.6Ca0.21Sr0.19)Cu2O7-d, suggesting partial substitution of the Sr into the Y site and displacement TC=92 K in (Pb,Cd)-1212 was found for x=0.3.43 When the stoichiometry of the (Pb,Cd)O layer in (Pb1-yCdy)- some of the Ca. This is not entirely unexpected as similar cross-substitution is known in the bismuth cuprates where Pb, Sr2YCu2O7-d (0.25y0.75) was varied it was found that the phase purity and TC onset temperature of the samples began Sr and Bi form a solid solution.35 The cross substitutional eVect seems to be amplified when the M cation is particularly to decrease as y deviated from 0.5, with Sr2CuO3, SrCuO2, Sr5Pb3CuO12 present for y<0.5 and CdO identified for y>0.5.small, (as in MgII) where Maignan et al. found solid solutions of the nominal (Pb,Mg)-1212 phase34 (Pb0.5Mg0.5)- Beales et al.42 found that the change in both the a and c lattice parameters was proportional to the eVective ionic radius of Sr2(Y0.5Ca0.5)Cu2O7-d to contain actual stoichiometries from the EDAX analysis of (Pb0.6Mg0.17Cu0.21Sr0.02)- Mn+, providing strong indirect evidence that transition metal ions directly substitute into the (Pb,M)O layer.Liu et al.44 Sr2(Y0.56Ca0.44)Cu2O7-d and (Pb0.61Mg0.07Cu0.28Ca0.04)- (Sr1.92Ca0.04)(Y0.7Ca0.3)Cu2O7-d. This implies that Mg really also synthesised (Pb0.5Cd0.5)Sr2(Y0.5Ca0.5)Cu2O7-d and found TC=70 K. X-Ray emission spectrometry determined that the does not substitute at all and that a rather random cationic composition of the rocksalt layer occurs, and can explain the Cd was mainly present in the (Pb,Cd)O layer.Other workers have investigated substitution of Y for lanthanide ions.45,46 poor XRD phase purity shown by samples in this system. An independent report of the preparation of the (Pb,Mg)-1212 The (Pb,Cd)-1212 phase can be synthesised phase pure for Y1-xLnx for x1, for all lanthanide ions except for La and phase was given by Liu et al.,36 who used an additive of 10 mass% silver oxide and processed in 10 MPa of high Ce.Yu et al.47 investigated the properties of the solid solution series (Pb0.5Cd0.5) (Sr2-xBax) (Y0.5Ca0.5)Cu2O7-d for pressure O2 at 950 °C but only managed to see a very weak superconducting signal at 60 K. 0x0.5, and from XPS measurements concluded that the apical oxygen on the CuO5 pyramids plays an important part This latter point emphasises that aside from cation inhomogeneity, the group 2 substituted materials also suVer from in the superconductivity in this system.Table 1 summarises the findings of Yu et al.47 oxygen non-stoichiometry. Both (Pb,Sr)-1212 and (Pb,Ca)- 1212 can be made superconducting as-synthesised, but the Widder et al.48 reported on the substitution of VII in (Pb1-yVy)Sr2(Y1-xCax)Cu2O7-d (0.1y0.7 and 0x1).superconducting properties are improved by post-synthesis annealing in high pressure O2. The presence of Ca on the Y They found that none of the samples were superconducting as synthesised, but the composition (Pb0.7V0.3)Sr2(Y0.6Ca0.4)- site is essential for superconductivity in these materials.The complex cross-substitutional and phase equilibria of these Cu2O7-d showed a TC onset=40 K when annealed in Ar at 500 °C after synthesis. The behaviour of this compound again materials is reflected in the fact that TC is seen to increase even 4 J. Mater. Chem., 1998, 8(1), 1–12Table 1 Listing of experimentally determined lattice parameters and onset values to near 90 K by processing a composition of transition temperatures for some (Pb1-yCdy)Sr2(Y1-zBaz)Cu2O7-d (Pb0.7In0.3)Sr2(Y0.2Ca0.8)Cu2O7-d at 1150 °C in 6 GPa of O2.(Pb,Sb)-1212 has been reported by Widder et al.58 but these formula y z volume/A ° 3 samples were not superconducting and no single phase materials could be made. Given the very small size of SbIII and the (Pb1-yCdy)Sr2(Y1-zBaz)Cu2O7+d 0.5 0.0 173.5 (Pb1-yCdy)Sr2(Y1-zBaz)Cu2O7+d 0.5 0.1 173.4 impure XRD data, it is not clear if Sb enters this structure at (Pb1-yCdy)Sr2(Y1-zBaz)Cu2O7+d 0.5 0.2 174.2 all, especially as (Pb,Cu)-1212 and Pb-2213 impurity phases (Pb1-yCdy)Sr2(Y1-zBaz)Cu2O7+d 0.5 0.3 174.7 were also found to be present.(Pb1-yCdy)Sr2(Y1-zBaz)Cu2O7+d 0.5 0.4 175.7 It is possible however to synthesise (Pb,Bi)-1212.(Pb1-yCdy)Sr2(Y1-zBaz)Cu2O7+d 0.5 0.5 176.3 Bauernfeind et al.59 for instance, looked at the partial substitution of Bi for Pb in (Pb,Cu)-1212. For a nominal starting stoichiometry of (Cu1-(x+y)[PbxBiy])Sr2YCu2O7-d , near is evidence that the smaller metavalent transition metal ions phase pure samples were obtained for compositions where the in the (Pb,M)O layer lead to oxygen and/or cation disorder average valency in the rocksalt layer was +3.They found that and are deleterious for superconductivity. for samples containing Pb, only partial Bi substitution was At approximately the same time that the mercury cuprates possible in the range 0.25x0.5 and that all samples were were discovered,49 (Pb,Hg)-1212 phases were reported.Early semiconducting despite a range of annealing conditions at reports showed50 a TC onset=90 K for the composition diVerent temperatures and atmospheres. This was attributed (Pb0.5Cd0.5)Sr2(Y0.7Ca0.3)Cu2O7-d . Chmaissen and Sheng51 once again to a random oxygen stoichiometry caused by reported a TC=100 K for (Pb,Hg)-1212 and suggested that cation disorder in the rocksalt layer leading to localised there was evidence for partial cation disorder in the (Pb,Hg)O trapping of holes.Superconductivity in (Pb,Bi)-1212 layer and that, like Tl, Hg can cross-substitute onto the Ca has also been reported at 94 K by Bauer et al.60 in site. The determined stoichiometry for a nominal composition (Pb0.45Bi0.4)Sr1.9(Y0.7Ca0.4)Cu2.25O7-d using a melt process of (Hg0.7Pb0.3)Sr2(Y0.3Ca0.7)Cu2O7-d by Chmaissen et al.route followed by a rapid cool. It is likely that the composition using Rietveld refinement was actually (Hg0.651Pb0.3Cu0.049)- reported by these authors also has a mixed Pb–Cu–Bi rocksalt Sr2(Y0.264Ca0.616Hg0.12)Cu2O6.84. Table 2 summarises the layer but that the processing conditions has resulted in a more results from a series of transition metal doped (Pb,M)-1212 uniform oxygen distribution and that partial substitution of Y materials processed under diVerent synthetic conditions.by Ca has added additional charge carriers for superconductivity. Substitution by the p block elements (M=Sb, In, Bi The (Pb,Tl)-1212 materials have been extensively investiand Tl ) gated and are worthy of a review in themselves.In the context of this article, it is worthwhile to summarise the properties (Pb,In)-1212 was first reported by Liu et al.52 who substituted of these materials. Both Pb61 and Bi62 can be substituted into In for Tl in (TlyPb1-y)Sr2(Y1-xCax)Cu2O7-d. Bulk superconthe rocksalt layer along with TlIII. The Y site can also be ductivity53 in this system occurs at 60 K with TC onset at substituted with Ca and other lanthanides.63,64 The 80 K for the composition (Pb0.7In0.3)Sr2(Y0.2Ca0.8)Cu2O7-d.highest TC value of 108 K in this system is found in Superconductivity in the (Pb,In)-1212 phase was also con- (Pb0.5Tl0.5)Sr2(Y0.2Ca0.8)Cu2O7-d . The substantial diVerence firmed by other authors.54,55 However, the small size of InIII between the Tl-containing systems and the (Pb,M)O systems and its mixed valence causes problems with cation disorder is that in the former, complete Ca replacement for Y is possible and oxygen non-stoichiometry.Liu et al.56 found that Cu can and Tl-containing materials can be made to undergo metal– form a solid solution in the rocksalt layer and that (Pb,In)- semiconductor transitions concomitant with the level of Ca 1212 samples are not phase pure.For a nominal starting doping.65 composition of (Pb0.55In0.2)Sr2(Y0.5Ca0.5)Cu2.25O7-d (showing As a summary of the eVects of cation substitution in (Pb,M)- the highest TC onset), the phase composition by EDAX was 1212 materials Fig. 8 shows the correlation of a axis decrease found to be (Pb0.69In0.06Cu0.25)Sr1.99(Y0.53Ca0.47)Cu2.25O7-d. with increasing eVective ionic radius of the Mn+ ion and as an Evidence of corresponding oxygen non-stoichiometry comes from the enhancement of TC by Ohta et al.57 who increased TC alternative way of representation, Fig. 9 shows the correspond- Table 2 Listing of experimentally determined lattice parameters and transition temperatures for some (Pb1-yMy)Sr2(Ln1-xCax)Cu2O7-d series materials formula y x TC /K volume/A ° 3 ref.(Pb1-yCdy)Sr2(Y1-xCax)Cu2O7+d 0.25–0.75 0.0–0.5 92 173.4 42 (Pb1-yZny)Sr2(Y1-xCax)Cu2O7+d 0.5 0.0–0.5 50 173.6 42 (Pb1-yCuy)Sr2(Y1-xCax)Cu2O7+d 0.5 0.0–0.5 — 173.6 42 (Pb1-yCdy)Sr2(Y1-xCax)Cu2O7+d 0.5 0.5 70 174.4 44 (Pb1-yCdy)Sr2(Lu1-xCax)Cu2O7+d 0.5 0.5 — 173.8 45 (Pb1-yCdy)Sr2PrCu2O7+d 0.5 0.0 — 179.6 46 (Pb1-yCdy)Sr2NdCu2O7+d 0.5 0.0 — 179.0 46 (Pb1-yCdy)Sr2SmCu2O7+d 0.5 0.0 — 177.9 46 (Pb1-yCdy)Sr2EuCu2O7+d 0.5 0.0 — 177.1 46 (Pb1-yCdy)Sr2GdCu2O7+d 0.5 0.0 — 176.6 46 (Pb1-yCdy)Sr2TbCu2O7+d 0.5 0.0 — 174.6 46 (Pb1-yCdy)Sr2DyCu2O7+d 0.5 0.0 — 174.8 46 (Pb1-yCdy)Sr2YCu2O7+d 0.5 0.0 — 174.7 46 (Pb1-yCdy)Sr2HoCu2O7+d 0.5 0.0 — 174.1 46 (Pb1-yCdy)Sr2ErCu2O7+d 0.5 0.0 — 173.6 46 (Pb1-yCdy)Sr2TmCu2O7+d 0.5 0.0 — 172.9 46 (Pb1-yCdy)Sr2YbCu2O7+d 0.5 0.0 — 173.8 46 (Pb1-yHgy)Sr2(Y1-xCax)Cu2O7+d 0.5 0.7 90 174.3 50 (Pb1-yHgy)Sr2(Y1-xCax)Cu2O7+d 0.7 0.7 100 174.1 51 (Pb1-yVy)Sr2(Y1-xCax)Cu2O7+d 0.3 0.4 40 172.6 48 J.Mater. Chem., 1998, 8(1), 1–12 5and also by partially substituting Pb for Cu to form (Bi0.45Pb0.35)Sr2(Y0.5Ca0.5)Cu2.15O7-d . This material showed a TC onset of 110 K and Meissner volume fraction=50%. A neutron diVraction study by Gopalakrishnan et al.70 on (Bi0.5Cu0.5)Sr2Y0.8Cu2.2O7-d identified that it was via the removal of oxygen that (Bi,Cu)-1212 achieved superconductivity.The (Bi,Cu)O layer did act as an eVective charge reservoir (from the relatively elongated CuMO bond length along the bridging oxygen), but unusually this system suVered from having excess oxygen vacancies in both the rocksalt layer and the perovskite CuO5 layers.A similar conclusion was arrived at by Kambe et al.71 who measured the hole density in (Bi,Cu)-1212 samples from the Hall eVect and concluded that the observed figure of 3.6×1020 cm-3 was too low to induce superconductivity. Beales et al.72 reported on the synthesis and properties of a series of (Bi,M)-1212 materials with M=Cd, Zn and Cu, 0.0x0.5 and 0.5y0.75.Ca-free samples (x=0) could only be prepared for samples with y=0.7 suggesting that the Fig. 8 A plot of the observed shift in the a axis lattice parameter optimum composition for phase stability in the rocksalt layer versus the eVective ionic radius of Mn+ ions in the substituted (Pb,M)- is the (Bi0.3M0.7) composition. Virtually phase-pure (Bi,Cd)- 1212 structure 1212 phases were formed in air at calcination temperatures 850T /°C975, (Bi,Zn)-1212 formed at slightly higher calcination temperatures between 900T /°C975 and the (Bi,Cu)- 1212 phase required higher calcination temperatures around 975 °C.The a axis increased from Cd, Zn to Cu as 3.802, 3.814 and 3.814 A° and the c axis decreased from Cd, Zn to Cu as 11.96, 11.75 and 11.72 A ° .No superconductivity was observed in (Bi,M)-1212 samples as synthesised, but (Bi,Cd)-1212 annealed in high pressure O2 showed a TC onset of 40 K. No (Bi,M)-1212 samples with Y substituted by Ca (x>0) could be synthesised by Beales et al. (Bi,Cd)-1212 was also reported by Tang et al.73 and Qian et al.74 The former workers only found semiconducting behaviour, while the latter reported a TC of 26 K after processing in 10 MPa of O2 at 800 °C.Very few transport measurements have been performed on the (Bi,Cd)-1212 materials. Beales and Parberry75 performed thermoelectric power measurements on (Bi0.33Cd0.67)Sr2YCu2O7-d yielding a value of S290 K=50 mV K-1 and a hole concentration= 0.08 per planar Cu atom. This value of p suggests a theoretical maximum TC=85 K for the (Bi,Cd)-1212 system.As even high pressure O2 annealing only obtains a maximum Fig. 9 A plot of the observed shift in the c axis lattice parameter versus TC of 40 K, it implies that (Bi0.33Cd0.67)Sr2YCu2O7-d is highly the atomic number of the Mn+ ion in substituted (Pb,M)-1212 intrinsically underdoped. compounds Other (Bi,M)-1212 phases that have been reported are (Bi0.3Hg0.7)Sr2(PrxCa1-x)Cu2O7-d [(Bi,Hg)-1212] for ing increase in c axis length with increasing M atomic number. 0.25x1.0 which showed the highest TC onset at 91 K,76 Both these phenomena reflect the move towards a higher the highly non-stoichiometric (Bi,In)-1212 phase77 oxidation state of the Cu in the ‘1212’ lattice perovskite blocks (Bi0.21In0.33Pb0.25)Sr2(Gd0.89Ca0.11)Cu2.21O7-d with TC onset resulting in an increase in TC as highlighted in Fig. 7. at 70 K and the (Bi,Mo)-1212 phase78 in which the Cu and Mo are in solid solution in the rocksalt layer. The highest TC Alternatives to Pb: (A1-yMy)Sr2(Y1-xCax)Cu2O7-d in the series (Bi1-xMox)0.33Cu0.67)Sr2YCu2O7-d (0.0x1.0) was 37 K. Table 3 summarises the results obtained from several The (Bi1-yMy)Sr2(Y1-xCax)Cu2O7-d series: (Bi,M)-1212 authors on the crystal parameters and properties of the (Bi,M)- 1212 series. (Bi,Cu)-1212 was first reported by Ehmann et al.66 who found that using solid state calcination procedures a higher synthesis temperature [compared to that for (Pb,Cu)-1212] of at least The (Ce1-yMy)Sr2(Y1-xCax)Cu2O7±d series: (Ce,M)-1212 970 °C was required to form relatively clean XRD patterns.They found that the highest TC of 68 K occurred in a sample Phases with this composition were first synthesised by Beales et al.72 for M=Cd, Zn and Cu with (0.0x0.5 and of nominal composition (Bi0.5Cu0.5)Sr2Y0.8Cu2.2O6.95 including a post-calcination anneal in flowing O2 at 500 °C. TEM 0y0.5). For Ca-free samples (x=0), virtually phase-pure (Ce,Cd)-1212 was formed in air between 850T /°C975. on the samples67 did not show the presence of any intergrowths.Since the as-synthesised samples were not superconducting, a (Ce,Zn)-1212 also formed at slightly higher calcination temperatures between 900T /°C975 and (Ce,Cu)-1212 phase non-optimised (Bi,Cu)O layer was implied. Later samples with stoichiometry (Bi0.3Cu0.7)Sr2YCu2O7.125 were prepared by required higher calcination temperatures around 975 °C to form a reasonably phase pure sample.Phase purity decreased Wang et al.68 and although they were phase pure, the TC onset value of samples prepared in high pressure O2 was low at 18 K as y deviated from 0.5 suggesting that the (Ce0.5M0.5) composition was the optimum in the rocksalt layer for phase stability.with a small Meissner fraction. The TC onset value was found to increase significantly by Zoller and co-workers69 by using The a axis increased from Cd, Zn to Cu as 3.808, 3.824 and 3.823 A ° and the c axis decreased from Cd, Zn to Cu as 12.10, a post-calcination anneal in Ar gas rather than in O2 6 J. Mater. Chem., 1998, 8(1), 1–12Table 3 Listing of experimentally determined lattice parameters and transition temperatures for some of the (Bi1-yMy)Sr2(Y1-xCax)Cu2O7-d series materials formula a/A ° c/A ° TC/K ref. (Bi0.5Cu0.5)Sr2Y0.8Cu2.2O7+d 3.815 11.73 68 66 (Bi0.3Cu0.7)Sr2YCu2O7+d 3.816 11.71 18 68 (Bi0.45Pb0.35)Sr2(Y0.5Ca0.5)Cu2.15O7+d 3.819 11.81 102 69 (Bi0.33Cd0.67)Sr2YCu2O7+d 3.802 11.96 40 72 (Bi0.33Zn0.67)Sr2YCu2O7+d 3.814 11.75 — 72 (Bi0.33Cu0.67)Sr2YCu2O7+d 3.814 11.72 — 72 (Bi0.5Cd0.5)Sr2YCu2O7+d 3.812 11.91 — 73 (Bi0.5Hg0.5)Sr2(Pr0.5Ca0.5)Cu2O7+d 3.863 12.09 91 76 (Bi0.5In0.5)Sr2YCu2O7+d 3.838 11.84 70 77 11.92 and 11.88 A ° .Magnetisation measurements showed that and calcining at temperatures near to 900 °C for up to 90 h. Martin et al.95 reported the synthesis and structure of none of the (Ce,M)-1212 samples were found to exhibit superconductivity as synthesised, but after annealing in high pressure (Pb0.5Tl0.5)Sr2CuO5-d but did not detect superconductivity in as-synthesised or O2 annealed samples.However superconduc- O2 for 16 h at 900 °C, a weak superconducting transition with a TC onset=30 K was observed in (Ce,Cd)-1212. For samples tivity at TC (onset)=40 K, with approx. 2% Meissner fraction, was reported by Bourgault et al.96 in the structurally and containing Ca, the phase purity of (Ce,Cd)-1212 samples decreased with increasing x and the impurity phases Sr2CuO3, chemically related (Tl0.8Pr0.2) (Sr1.6Pr0.4)CuO5-d . The (Pb,M)-1201 structure is schematically shown in Fig. 10. SrCuO2 and Cd1-xCaxO appeared in the XRD data. Magnetisation measurements showed none of these samples The Cu occupies a typical perovskite site in an octahedral oxygen geometry, with the larger Sr and La ions located at were superconducting, even after annealing in high pressure O2.79 the apical oxygen ions of the CuO6 octahedra. In (Pb,M)- 1201, this is the equivalent of the stable perovskite YSrCu2O6 unit found in (Pb,M)-1212.As in the (Pb,M)-1212 structure, Other (A,M)-1212 series the rocksalt (Pb,Cu)O units simply sit on the outside of this Tl has long been a favourite ion to substitute for Pb in the base unit, attached at the apical oxygen of the CuO6 octahedra.(Pb,M)O layer, and the M ion in (Tl,M)-1212 materials has The phase stability of the (Pb,Cu)-1201 phase has been been subjected to substitution from all over the Periodic Table, determined97,98 for the series (Pb0.5Cu0.5) (La1-xSr1+x)CuO5-d with often haphazard logic.Liu et al.80 substituted a number (0.0x0.1). The (Pb,Cu)-1201 phase has many similarities of group 4 and 5 elements into the Tl layer of to the corresponding (Pb,Cu)-1212 phase. It is diYcult to TlSr2(Y1-xCax)Cu2O7-d and obtained a range of TC onset prepare phase pure and Beales et al.42 have pointed out the values from 40 to 100 K.Ce81 (TC onset values around 66 K), presence of impurity phases such as Sr5Pb3CuO12 in the XRD Mo82,83 (TC onset=70–100 K), Cr,83,84 Bi85 and Hg86 (TC patterns during synthesis. Also the physical properties are onset=92 K) have all similarly been substituted into the Tl strongly dependent on the processing: the TC onset value of layer.However, the tendency of Tl to cross-substitute onto the (Pb,Cu)-1201 can be increased to 36 K97 by quenching from Y site and also the tendency of Ca to cross-substitute onto the 900 °C or by annealing at 500 °C for 24 h in N2.99 In addition, Tl site means that the exact relationship between (Tl,M)O and a post synthesis high pressure anneal at 600 °C in 50 MPa O2, the optimisation of TC in these materials is nearly always was shown to both decrease TC and increase the normal state impossible to determine, as extra doping eVects will certainly resistivity.This suggests that the material is either in, or close be present from the cation disorder. The overall tendency of to, the overdoped state as synthesised. Very little work has these materials is to behave similar to the parent Tl-1212 phase.been done on determining the oxygen stoichiometry of the The most common other substituted element for Pb forming (Pb,Cu)-1201 phase, compared to that carried out on the a rocksalt layer (aside from thallium) is mercury. (Hg,M)-1212 (Pb,Cu)-1212 phase. Recent results determining the oxidastructures have been reported for M=Nb and Ta87 (TC onset= tion state of the Pb and Cu in the (Pb,Cu)O layer by Shida 80 K), Pr,88 Re,89 Cr,90 Ga91 (TC onset=80 K) and Mo92 (TC et al.100 in (Pb0.5Cu0.5) (Sr2-xLax)CuO5-d (1.0x1.2) and onset=98 K).However, indications are that these materials (Pb0.6Cu0.4) (Sr2-xLax)CuO5-d (0.9x1.2) prepared in owe little to the unique nature of the (Hg,M)O layer and act flowing O2 at 1010 °C, show that the TC onset value is related more like the parent HgO-layered phases (of which these can to d.As the structure loses oxygen, the TC onset value rises. be considered as sub-sections). The TC value in (Hg,M)-1212 However, the TC onset value was found to be insensitive to compounds seems dependent more on the processing method both substitution of SrII by LaIII and to increases in PbIV to yield TC values similar to the corresponding Hg-1212 phase rather than any particular unique chemical properties of the (Hg,M)O layer.Other layered cuprate materials containing (Pb,M) layers The (Pb1-yMx)(La2-xSrx)CuO5-d series: (Pb,M)-1201 It was Mochiku et al. who first reported the 1201 structure93 in 1989 for Tl(La,Sr)2CuO5-d and noted that for a series of compositions, the highest TC onset observed was close to 40 K.Shortly afterwards, a (Pb,Cu)O rocksalt layer in place of the Tl layer was reported by Adachi et al.94 in (Pb,Cu)-1201 for a nominal composition of (Pb0.6Cu0..5) (La,Sr)CuO5-d , with an observed TC onset near to 28 K. This material was easily synthesised using the conventional solid state method of mixing Fig. 10 Idealised crystal structure of the (Pb1-yMx)(La2-xSrx)- CuO5-d phase stoichiometric quantities of oxides and/or carbonates J. Mater. Chem., 1998, 8(1), 1–12 7concentration, suggesting that interstitial oxygen is mainly ) and d to the c axes, respectively. The anisotropy ratio (lab/lc) for all samples was found to be close to 1 reflecting a responsible for superconductivity in this phase.Shida et al. believe that the Cu ions in the rocksalt layer charge compensate more 3D nature than in the ‘1212’ phase. Resistivity and thermopower power measurements confirmed that the ‘1201’ for the electrochemical changes caused by the chemical substitution or changes in d. phase is intrinsically highly overdoped with high hole concentrations 0.2 as synthesised and is close to the theoretical Like for the corresponding ‘1212’ structure, the ‘1201’ phase preferentially substitutes on the rocksalt layer rather than into maximum TC value.the thermodynamically stable perovskite layer. Beales et al.42 reported the formation of the (Pb,Cd)-1201 and (Pb,Zn)-1201 (Pb1-zMz)(AE1-xLn¾x)2(Ln1-yLn¾y)2Cu2O9-d: (Pb,M)-1222 phases as well as (Pb,Cu)-1212 for (Pb0.5M0.5) (La2-xSrx)- CuO5-d (1.0x1.25).Between 750T /°C900, virtually Higher order members incorporating Pb-containing rocksalt layers have been synthesised. The (Pb,M)-1222 lattice contains phase-pure samples of (Pb,Cd)-1201 were formed for all values of x. (Pb,Zn)-1201 also formed in this temperature range but bridging layers of fluorite-structure lanthanide ions of similar configuration to those found in the T¾ [(Nd,Ce)CuO4] struc- only for compositions with x#1.(Pb,Cu)-1201 needed a higher calcination range of 850T /°C900 to form a reasonably ture.107 The (Pb,M)-1222 structure can be mentally constructed by splitting the (Pb,M)-1212 structure shown in Fig. 5 in half phase-pure material but again, only for compositions with x#1.The elemental substitution into the rocksalt layer was at the Y site, and then inserting a bridging fluorite layer as shown schematically in Fig. 11. reflected in the decreasing a axis lengths of 3.784, 3.776 and 3.757 A ° and increasing c axis lengths of 8.627, 8.659 and One of the first (Pb,M)-1222 phases to be reported was (Pb,Cu)(Sr,Pr)2Pr2Cu2O9-d or (Pb,Cu)-1222 by Adachi et al.108 8.891 A ° for M=Cu, Zn and Cd respectively.Beales et al. found that only the (Pb,Cd)-1201 phase could be made phase pure, (this phase is isostructural with TlSr2(Nd,Ce)2Cu2O9-d).109 As synthesised, (Pb,Cu)(Sr,Pr)2Pr2Cu2O9-d is non-supercon- with the (Pb,Zn)-1201 and (Pb,Cu)-1201 phases forming impurities of Sr5Pb3(Cu,Zn)O12 and Sr5Pb3CuO12 respectively. ducting. The idealised bridging fluorite layer found in (Pb,Cu)-1222 structure is shown in Fig. 12. The Sr site in (Pb,Cd)-1201 had a TC onset=40 K for the x=1 composition, which deteriorated with decreasing x. The corresponding (Pb,M)-1212 connects to the first fluorite layer (designated as an M¾ ion site) and can contain a mixture of either (Pb,Cu)-1201 and (Pb,Zn)-1201 phases with x=1 were found to have TC onsets of 28 K and 15 K, respectively.These results lanthanide or Sr/Ln ions. This sits astride a central fluorite layer of M lanthanide ions which also contains CuO5 square- confirm the beneficial eVect of Cd in the rocksalt layer as was also found in the ‘1212’ materials and suggests for the 1201 basal pyramids and rocksalt (Pb,Cu)O monolayer. Maeda et al.110 found that elemental substitution on the Ln sites could phase, the (Pb0.5Cu0.5) composition in the rocksalt layer is near optimised for superconductivity in the nominally undoped induce superconductivity in (Pb,Cu)-1222 with a TC onset of 25 K for the composition (Pb0.5Cu0.5) (Sr0.875Eu0.125)2- compositions (x=1.0).Several other (Pb,M)-1201 structures have been reported in (Eu0.75Ce0.25)2Cu2O9-d annealed in O2 at 400 °C for 10 h.Rietveld refinement of the XRD data by Maeda et al.111 found the literature, notably (Pb0.5Tl0.5) (Sr2-xLax)CuO5-d (TC=40 K101), (Pb0.5Hg0.5) (Sr2-xLax)CuO5-d (TC=32 K102), that the M¾ and the M sites are crystallographically distinct with the M¾ site=(Sr,Eu) and the M site=(Eu,Ce). The Eu (Hg0.5Bi0.5) (Sr2-xLax)CuO5-d,103 (Bi0.2Hg0.8)Ba2CuO5-d (TC onset=75 K104), and finally (Hg0.7V0.3) (Sr2-xLax)CuO5-d; could be replaced by either a single lanthanide e.g.Pr, Sm, Gd, Dy, Ho and Y or by a mixture of lanthanides e.g. (La,Eu), (TC=50 K105). Very little physical characterisation has been performed on these interesting materials. Recently, Beales et al. (Nd,Eu), (Nd,Dy), (Nd,Y), (Nd,Eu), etc. When a mixture of lanthanides was used, it was found that the larger ions (La, reported on the magnetisation data for (Pb0.5Cd0.5)- (La2-xSrx)CuO5-d (x=1.0±0.2).106 The strongest Meissner Sm and Nd) preferentially substituted onto the M¾ site and the smaller lanthanides onto the M site.Superconductivity signal occurred at the x=1.0 composition, with an intragrain JC calculated to be in the order of 1×109Am-2 at 4 K.The could only be induced in samples in which the average cation radius on the M site was close to 0.104 nm. Using the above, London penetration depth, lL, was found to be the smallest for the x=1.0 composition with lL=3.8 mm and 4.4 mm for H Maeda et al. were able to increase TC onset to 32 K. Further Fig. 11 Pictorial representation of the building block structure of the ‘1222’ phase by incorporating a fluorite block into the ‘1212’ structure 8 J.Mater. Chem., 1998, 8(1), 1–12The low TC onset values in the (Pb,Cu)-1222 series and lack of superconductivity in the (Pb,M)-1222/(M,Cu)-1222 series can be explained as being due to a combination of, or all of, the following: random cationic site substitution; oxygen disorder in the (Pb,Cu)O layers and non-stoichiometric intergrowths. In addition, Beales et al.42 pointed out that the fluorite layer favours n-type superconductivity while the (Pb,M)O layer favours p-type superconductivity which can lead to an electrochemical mismatch.However, one can point to potentially higher TC values in this system. The oxygen stoichiometry may well be optimised by processing in high pressure O2 and the work of Beales et al.42 showed that the a axis length in (Pb,Cd)- 1222 material was the shortest found in the (Pb,M)-1222 series, approaching that found in (Pb,Cu)-1212.This suggests that the (Pb,Cd) rocksalt mixture combined with annealing in high pressure O2 could provide a Cu valence nearer to that required for superconductivity and push TC onset temperatures to much higher values in the ‘1222’ materials. Other layered Pb-containing materials It is possible to incorporate an additional fluorite layer into the (Pb,Cu)-1222 materials, acting as another bridging unit to Fig. 12 Idealised crystal structure of the fluorite block found in that described in the previous section to form the (Pb,Cu)- (Pb1-zMz) (AE1-xLn¾x)2(Ln1-yLn¾y)2Cu2O9-d 1232 phase.Fig. 13 shows the idealised structure of the fluorite block in (Pb,Cu)-1232 materials. Wada et al.118 managed to synthesise (Pb,Cu)Sr2(Ln,Ce)3Cu2O11-d but this material was Rietveld refinement of the X-ray diVraction data by Maeda not found to be superconducting. Rietveld analysis of the et al.112 of (Pb0.5Cu0.5) (Sr0.875Nd0.125)2(Ho0.69Ce0.31)2Cu2O9-d neutron diVraction data119 of a (Pb,Cu)-1232 material of revealed that the (Pb,Cu)O layer was highly disordered [similar composition (Pb0.5Cu0.5)Sr2(Ho0.33Ce0.67)3Cu2O11-d again to the (Pb,Cu)-1212 materials] with 13% oxygen deficiency at showed a highly disordered (Pb,Cu)O monolayer similar to the apical oxygen site connecting the outer rocksalt (Pb,Cu)O those seen in (Pb,Cu)-1212 and (Pb,Cu)-1222.It is probable layer and the CuO5 square basal pyramids in the (Pb,M)-1212 that this is the prime cause of the lack of superconductivity in derived unit. The origin of this disorder was attributed to the these materials (a net Cu oxidation state of +2.12 in the CuO5 lattice mismatch between the (Pb,Cu)O and CuO5 layers and planes is calculated if one assumes cationic valencies of PbIV, that this in turn caused hole localisation and a corresponding SrII, HoIII and CeIV in the above composition).low TC value. Saskura et al.113 synthesised (Pb0.5Cu0.5)- The Pb-0201 structure, formed by doping lead into the non- (Sr1-xLnx)2(Sr1-yLay)2Cu2O9-d (Ln=Gd, Dy, Ho, Er; x= superconductor La2CuO4-d to form (La2-xPbx)CuO4-d , was 0.975, 0.9, 0.875, 0.875 and y=0.3, 0.45, 0.325, 0.3, respectively) synthesised by Maignan et al.120 for 0.5x0.5. This had a which showed similar results to those obtained by Maeda et al.K2NiF4-related structure with lattice parameters of a= on materials that contained tetravalent ions. Only the Gd- 5.3481(4), b=5.3367(4) and c=13.2186(6) A ° . A TC (onset) containing composition was superconducting showing a weak close to 15 K was achieved with a Meissner fraction of ca.magnetic TC onset at 20 K. 25% after a post-synthesis annealing at 450 °C in high pressure Beales et al.42 synthesised a series of samples with the O2. Similarly, the Pb-0212 structure, Pb0.5La1.4Sr1.1Cu2O6-d , nominal stoichiometry (Pb0.5M0.5) (Nd1+xCe1-x)Cu2O9-d has been reported by Seling et al.121 with a maximum TC onset (M=Cd, Zn, Cu and 0.0x0.5).They found that moderately of 75 K. Fig. 14 shows an idealised crystal structure for the phase pure samples of (Pb,Cd)-1222, (Pb,Zn)-1222 and Pb-0212 phase. This material contains eight co-ordinate Sr (Pb,Cu)-1222 phases were formed at 750 °C for x#0.3. The atoms sandwiched between the CuO5 planar units. These are samples were indexed with a tetragonal structure, space group then surrounded by the La ions and the Pb-0212 structure I4/mmm.The lattice parameters were observed to change with therefore contains no rocksalt layers. Doping the La sites with substitution into the rocksalt layer, the a axis increasing as Pb induces superconductivity in this system via a PbIV/PbII 3.856, 3.874 and 3.880 A ° and the c axis shortening from 29.7, charge transfer mechanism in contrast to the La2(Sr,Ca)Cu2O6 29.4 and 29.4 A ° forM=Cd, Zn and Cu, respectively.Compared system which requires processing under high pressure O2 to with their work on corresponding 1201 and 1212 phases, induce superconductivity.122 Beales et al. found that the 1222 phase was more diYcult to Mateev et al.123 managed to synthesise a (Ge,Cu)-1223 phase form and that calcination at higher temperatures induced phase decomposition inducing disorder along the c axis from the presence of intergrowths of thicker fluorite-type blocks of (Nd,Ce) ions and (Pb,M)-1212 layers.None of the (Pb,M)- 1222 phases synthesised by Beales et al. were found to be superconducting. A similar semiconducting (Pb,Cd)-1222 structure was reported by Min et al.114 with a composition of (Pb0.5Cd0.5) (Sr0.9Eu0.1)2(Eu0.7Ce0.3)2Cu2O9-d who found by Rietveld refinement that the (Pb,Cd)O layers are displaced from their ideal crystallographic sites.It is possible to substitute for Pb in the ‘1222’ structure and semiconducting compositions of (M,Cu)-1222 substituted materials have been reported by Wada and colleagues115,116 for (Fe0.75Cu0.25)Sr2(Y0..5Ce0.5)2Cu2O9-d , and for (Ce1-zCdz)- Sr2(Ln2-x-yCexSrz)Cu2O9-d (Ln=Nd, Y; 0.797z0.844; Fig. 13 Idealised crystal structure of the fluorite block found in (Pb,Cu)Sr2(Ln,Ce)3Cu2O11-d 0.676x0.743 and 0.076y0.101) by Shizhong et al.117 J. Mater. Chem., 1998, 8(1), 1–12 9Ca doping to synthesis Pb2Sr2(Y1-xCax)Cu3O8±d can be achieved by using an appropriate Y1-xCaxSr2Cu3O7 precursor. 127 It is also possible to synthesis Pb-2213 directly from a stoichiometric mixture of oxides and carbonates, but this requires some care to obtain phase-pure material. Kadowski et al.128 subjected the starting mixture to repeated firing and grinding at temperatures between 800 and 850 °C to prevent Pb loss. This mixture was then pelletised and heated between 900 and 930 °C for 1 to 2 h before being quenched. The pellet was made superconducting by annealing in N2 at 800 °C for up to 5 h, cooled to 700 °C, and held at that temperature for about a day.Pb-2213 is thermally unstable. At temperatures above 850 °C serious Pb volatilisation occurs (if the synthesis temperature is too low than scavenging phases such as SrPbO3 and unreacted binary oxides occur in the product).Partial decomposition of Pb-2213 to YSr2(Pb,Cu)3O7-d and other complex oxides occurs at temperatures above 900 °C and the material melts just above 1000 °C: the melt can be cooled to form a mixed reaction product of Pb-2213 crystalline plates and complex mixed oxides. Plate-like Pb-2213 crystals measuring up to 2×2×0.1 mm3 have been made using a PbO/PbF2 flux.129 The same group have also tried fluxes composed of CuO, PbO and PbF2 as well as pure PbO and PbF2.The flux solvent was mixed with Pb-2213 and heated in an Au crucible to 980 °C and held there for 20 to 50 h. The crucible was then cooled at 1–5 K h-1 to Fig. 14 The idealised Pb0.5La1.4Sr1.1Cu2O6-d, Pb-0212 crystal structure temperatures between 850 and 600 °C and furnace cooled in 1 to 2% O2 in N2 gas.Other groups have used Pb-2213 precursor mixtures rich in CuO, SrO and PbO, cooling the melt from using powders encased in Au capsules synthesised in a belt 1025 to 1050 °C in corundum crucibles held in a nitrogen apparatus at 6 GPa and 1300 °C. The composition was found atmosphere. Superconducting crystals of 1×1×0.5 mm3 were to be slightly Cu-rich at (Ge0.4Cu0.6)Sr2(YxCa1-x)Cu3O9.3 reported using this synthesis route.130 (0.1x0.3) and showed a TC onset=90 K.The 1223 phase A more novel route to synthesise Pb2Sr2(Y1-xCax)Cu3O8±d is similar in structure to the corresponding 1212 phase but has films using laser ablation has also been used. Naqvi et al.131 an extra (YxCa1-x)CuO2 plane in between the CuO5 layers. achieved this successfully on MgO substrates using stoichiometric targets without ex situ annealing, provided that a reduc- Pb2Sr2(Y1-xCax)Cu3O8-d: the Pb-2213 phase ing atmosphere in the growth chamber was present.They also Superconductivity in Pb2Sr2(Y1-xCax)Cu3O8±d was first grew multilayered Sr2(Y0.5Ca0.5)Cu3Od5PbO structures by reported in 1988 by Cava et al.124 and shortly after by other ablation using a Nd5YAG laser and then annealed these ex situ groups.125,126 This material is best treated as a sub-set of the at 864 °C in air for 3 h.Four point resistivity measurements layered Pb cuprates. It has the same perovskite block seen in showed the samples to have a TC onset of 83 K. Hughes et al.132 the ‘1212’ phase but instead of a rocksalt bridging unit, a also attempted to grow Pb-2213 films using laser ablation, but (Pb2Cu)O unit, shown in Fig. 15 is present. using a heavily Pb-compensated target. They reported preparing The preparation of Pb-2213 using conventional powder predominantly ‘1212’ films grown in situ on (100) LaAlO3 metallurgical routes requires rather more care than that needed substrates with a TC onset temperature of 90 K.for preparing Y-123. This requires a multi-step process involv- Pb-2213 can undergo substitution on the Pb site by Bi,133 ing mixed oxide precursors. Cava et al.124 found that the best on the Sr site by Ba134,135 and extensive substitution by method was to mix pellets of YSr2Cu3O7 and calcine at lanthanides on the Y site.136–139 However, the material still suVers from extensive charge localisation causing detrimental temperatures between 920 and 980 °C for 16 h (with a single eVects to any gains from cation doping140 and structural intermediate grinding stage) and then mix in PbO.The new similarities to the ‘1212’ and ‘123’ lattices often lead to powder mix was then re-calcined at temperatures between 860 intergrowths and stacking faults, especially in bulk prepared and 925 °C for 16 h in a reducing atmosphere of 1% O2 in N2.samples.141 The Pb-2213 phase has been the subject of several reviews, see for example Naqvi et al.142 Conclusions The layered curates incorporating a (Pb,M)O rocksalt layer, or its analogues, are a remarkably versatile chemical system to study the eVects of chemical substitution on superconductivity in oxides.Despite having large numbers of constituent cations, these systems are thermodynamically stable, and more often than not form the major phase during conventional solid state calcination synthesis. This method of fabricating such complex systems does have its drawbacks, the main ones being non-stoichiometric cationic site substitution (outside the stable perovskite blocks), anionic oxygen disordering, intergrowths (for the higher order members) and the tendency to form scavenging impurity phases usually composed of simpler Fig. 15 Illustration of the (Pb2Cu)O block sited at the apical oxygen of the YSr2Cu2O6 perovskite block in the ‘2213’ structure ternary and binary oxides of the cations. 10 J. Mater. Chem., 1998, 8(1), 1–1223 J. Y. Lee, J. S. Swinnea and H.Steinfink, J. Mater. Res., 1989, These can be overcome by using a more sophisticated fabri- 4, 763. cation technology such as laser ablation or other thin film 24 Y. Tokura, H. Takagi and S. Uchida, Nature (L ondon), 1989, growth techniques which should lead to pure single crystal 337, 345. quality samples being synthesised. These synthetic methods have 25 A. Maignan, T.Rouillon, D. Groult, J. Provost, M. Hervieu, not been widely employed at all at present on the layered Pb C. Michel, B. Raveau, R. S. Liu and P. P. Edwards, Physica C, 1991, 177, 461. cuprates with the exception of a few groups who have produced 26 T. Maeda, K. Sakuyama, S. Koriyama, H. Yamauchi and S. laser ablated films of Pb-2213 (and managed to produce some Tanaka, Phys. Rev.B, 1991, 43, 7866. (Pb,Cu)-1212 impurities as well ). Since the potential application 27 S. H. H. Naqvi and I. W. Boyd, Physica C, 1993, 213, 161. of these materials is clearly in the electronics industry, this 28 R. A. Hughes, Y. Lu, T. Timusk and J. S. Preston, Appl. Phys. fabrication route is compatible with that employed in present L ett., 1991, 59, 2597. day semiconductor manufacturing.High temperature supercon- 29 X. X. Tang, D. E. Morris and A. P. B. Sinha, Phys. Rev. B, 1991, 43, 7936. ducting electronics is in its infancy and we have only scratched 30 M. R. Presland, J. L. Tallon, R. G. Buckley, R. S. Liu and N. E. the surface. The lesson from semiconductors is that as devices Flower, Physica C, 1991, 176, 95. become smaller, materials issues come to the fore.Layered 31 K. Kurusu, H. Takami and K. Shintomi, Analyst (L ondon), 1989, cuprates may well prove useful in the future as alternatives to 114, 1341. Y-123 and may have more versatility in future devices. The 32 M. Kosuge, T. Maeda, K. Sakuyama, H. Yamauchi, non-superconducting forms may even provide ready-made lat- N. Koshikuza and S. Tanaka, Physica C, 1991, 182, 157. 33 T.Maeda, K. Sakuyama, F. Izumi, H. Yamauchi, H. Asano and tice-matched substrates. The crystal chemistry still needs much S. Tanaka, Physica C, 1991, 175, 393. investigation before economic exploitation can fully arise. 34 A. Maignan, D. Groult, R. S. Liu, T. Rouillon, P. Daniel, Unfortunately in today’s commercial climate, this is unlikely to C. Michel, M. Hervieu and B.Raveau, J. Solid State Chem., 1993, be undertaken by industry. However, this area can and should 102, 31. be researched by academia. The field is still wide open and the 35 B. Raveau, C. Michel, M. Hervieu, D. Groult and J. Provost, Adv. potential rewards very great indeed. Mater., 1990, 2, 299. 36 H. B. Liu, D. E. Morris and A. P. B. Sinha, Physica C, 1993, 204, 262. The author wishes to thank BICC Cables Ltd for financially 37 T.Rouillon, J. Provost, M. Hervieu, D. Groult, C. Michel and supporting this work and grateful thanks go to Ms. Carol B. Raveau, J. Solid State Chem., 1990, 84, 375. Chambers in the BICC Wrexham Technology Centre Library 38 T. Roullion, A. Maignan, M. Hervieu, C. Michel, D. Groult and for invaluable help in compiling the references.B. Raveau, Physica C, 1990, 171, 7. 39 E. L. Belokoneva, L. I. Leonyuk, A. G. Vetkin and N. I. Leonyuk, Supercond. Phys. Chem. T echnol., 1992, 5, 948. References 40 T. Rouillon, J. Provost, M. Hervieu, D. Groult, C. Michel and B. Raveau, Physica C, 1989, 159, 201. 1 M. Murukami, Supercond. Sci. T echnol., 1992, 5, 185. 41 G.-Yu. Babonas, R. Dagis, G. Pukinskas and L. I. Leonyuk, 2 M.Ye, J. Schroeder, M. Mehbod, R. Deltour, A. G. M. Jansen Supercond. Phys. Chem. T echnol., 1992, 5, 1022. and P. Wyder, in Applied Superconductivity 1995, IoP Conference 42 T. P. Beales, W. G..Freeman, S. R. Hall, M. R. Harrison and J. M. Series No 148, vol. 2, p. 987, ed. D Dew-Hughes, IoP, Bristol, Parberry, Physica C, 1993, 205, 383. 1995. 43 T. P. Beales, C. Dineen, W.G. Freeman, D. M. Jacobson, S. R. 3 P. K. Gallagher, H. M. O’Bryan, S. A. Sunshine and D. W. Hall, M. R. Harrison and S. Zammattio, Supercond. Sci. T echnol., Murphy, Mater. Res. Bull., 1987, 22, 995. 1992, 5, 47. 4 E. D. Specht, C. J. Sparks, A. G. Dhere, J. Brynestad, O. B. Cavin, 44 R. S. Liu, D. Groult, A. Maignan, S. F. Hu, D. A. JeVerson, D. M. Kroeger and H. A. Oye, Phys.Rev. B, 1988, 37, 7426. B. Raveau, C. Michel, M. Hervieu and P. P. Edwards, Physica C, 5 T. Den and T. Kobayashi, Physica C, 1992, 196, 141. 1992, 195, 35. 6 T. P. Beales, L. Le Lay, M. Mo� lgg, C. Ferrari and W. Segir, Nuovo 45 Y. Qian, H. Huang, K. Tang, G. Cao and Z. Chen, Chinese Phys. Cimento D, 1994, 16, 2067. L ett., 1992, 9, 552. 7 M. K. Wu, J. R. Ashburn, C. J. Torng, P.H. Hor, P. L. Meng, 46 G. C. Che, Y. C. Lan and Z. X. Zhao, J. Alloys Compd., 1995, L. Gao, Z. J. Huang, Y. Q.Wang and C. W. Chu, Phys. Rev. L ett., 228, 45. 1987, 58, 908. 47 W.-J. Yu, Z. Q. Mao, X. M. Liu, M. L. Tian, G. E. Zhou and Y. H. 8 Z. Z. Sheng and A. M. Hermann, Nature (L ondon), 1988, 332, 55. Zhang, Physica C, 1996, 261, 27 9 H. Maeda, Y. Tanaka, M. Fukutumi and T.Ansano, Jpn. J. Appl. 48 W. Widder, M. Franz, L. Bauernfeind and H. F Braun, Physica Phys., 1988, 27, L209. C, 1993, 217, 121. 10 T. Hikata, M. Ueyama, H. Mukai and K. Sato, Cryogenics, 1990, 49 S. N. Putilin, I. Bryntse and E. V. Antipov,Mater. Res. Bull., 1991, 30, 924. 26, 1299. 11 M. P. Maley, J. Appl. Phys., 1991, 70, 6189. 50 S. F. Hu, D. A. JeVerson, R. S. Liu and P. P. Edwards, J.Solid 12 R. S. Liu, D. N. Zheng, J. W. Loram, K. A. Mirza, A. M. Campbell State Chem, 1993, 103, 280 and P. P. Edwards, Appl. Phys. L ett., 1992, 60, 1019. 51 O. Chmaissen and Z. Z. Sheng, Physica C, 1995, 247, 125. 13 Z. Z. Sheng, Y. F. Li and D. O. Pederson, in Superconductivity 52 R. S. Liu, P. T. Wu, S. F. Wu, W. N. Wangh and P. Edwards, and its applications, 5th Annual Conference, BuValo, NY, Physica C, 1990, 165, 111. 24–26 September, 1991, AIP Conference Proceedings No. 251, 53 R. S. Liu, P. P. Edwards and P. T.Wu, Adv.Mater., 1990, 2, 369. 1992, pp. 477–488. 54 R. Suryanarayanan, H. Pankowska, L. Ouhammou, O. Gor- 14 B. Raveau, C. Michel, M. Hervieu and A. Maignan, J. Mater. ochov and A. Halli, J.Mater. Sci. L ett., 1990, 9, 33. Chem., 1995, 5, 803. 55 S. Claude, M. Hervieu and P. T. Moseley, J. Mater. Chem., 1992, 15 T. Wada, S. Adachi, T. Mihara and R. Inaba, Jpn. J. Appl. Phys., 2, 471. 1987, 26, L706. 56 R. S. Liu, S. F. Hu and P. P. Edwards, Appl. Phys. L ett., 1991, 16 A. Ono, T. Tanaka, H. Nozaki and Y. Ishikawa, Jpn. J. Appl. 59, 985. Phys., 1987, 26, L1687. 57 M. Ohta, M. Tsutsumi, J. Yoshimoto and B. Okai, Physica C, 17 B.W. Veal, W. K. Kwok, A. Umezawa, G. W. Crabtree, 1991, 185–189, 667. J. D. Jorgensen, J. W. Downey, L. J. Nowicki, A. W. Mitchell, 58 W. Widder, L. Bauernfeind, M. Franz and H. F. Braun, J. Alloys A. P. Paulikas and C. H. Sower, Appl. Phys. L ett., 1987, 51, 279. Compd., 1993, 195, 61. 18 S. A. Sunshine, L. F. Schneemeyer, T. Siegrist, D. C. Douglass, 59 L. Bauernfeind, W. Widder, W.HoVman and H. F. Braun, J. V. Waszczak, R. J. Cava, E. M. Gyorgy and D. W. Murphy, Physica C, 1994, 235–240, 947. Chem. Mater., 1989, 1, 331. 60 A. Bauer, P. Zoller, J. Glaser, A. Ehmann, W. Wischert and 19 B. Okai, Jpn. J. Appl. Phys., 1990, 29, L2180. S. Kemmler-Sack, Physica C, 1996, 256, 177. 20 P. R. Slater and C. Greaves, Physica C, 1991, 180, 299. 61 M. A. Subramanian, C.C. Torardi, J. Golpalakrishnan, P. L. Gai, 21 T. Den and T. Kobayashi, Physica C, 1992, 196, 141. J. C. Calabrese, T. R. Askew, R. B. Flippen and A. W. Sleight, 22 M. A. Subramanian, J. Golpalakrishnan, C. C. Torardi, P. Science, 1988, 242, 249. L. Boyes, T. R. Askew, RB. Flippen, W. E. Farneth and A. 62 M. A. Subramanian, P. L. Gai and A. W. Sleight, Mater. Res. Bull., 1990, 25, 101.W. Sleight, Physica C, 1989, 157, 124. J. Mater. Chem., 1998, 8(1), 1–12 1163 Z. Z. Sheng, L. Sheng, X. Fei and A. M. Hermann, Phys. Rev. B, 106 T. P. Beales, J. D. Johnson and J. M. Parberry, Supercond. Sci. T echnol., 1997, 10, 502. 1989, 39, 2918. 107 J. Akimitsu, S. Suzuki, M. Watanabe and H. Sawa, Jpn. J. Appl. 64 C. N. R. Rao, A. K. Ganguli and R. Vijayaraghavan, Phys.Rev. Phys., 1988, 27, L1859. B, 1989, 40, 2565. 108 S. Adachi, O. Inoue, S. Kawashima, H. Adachi, Y. Ichikawa, 65 R. S. Liu, P. P. Edwards, Y. T. Huang, S. F. Wu and P. T. Wu, K. Setsune and K. Wasa, Physica C, 1990, 168, 1. J. Solid State Chem., 1990, 86, 334. 109 C. Martin, D. Bourgault, M. Hervieu, C. Michel, J. Provost and 66 A. Ehmann, S. Kemmler-Sack, S. Lo� sch, M.Schlichenmaier, B. Raveau, Mod. Phys. L ett. B, 1989, 3, 993. W. Wischert, P. Zoller, T. Nissel and R. P. Huebener, Physica C, 110 T. Maeda, K. Sakuyama, S. Koriyama, A. Ichinose, H. Yamauchi 1992, 198, 1. and S. Tanaka, Physica C, 1990, 169, 133. 67 H.-U. Nissen, M. Cantoni, S. Kemmler-Sack, A. Ehmann, S. 111 T. Maeda, K. Sakuyama, N. Sakai, H. Yamauchi and S. Tanaka, Lo� sch, W.Wischert and P. Zoller, Physica C, 1992, 203, 436. Physica C, 1991, 177, 337. 68 J. Wang, S. Takano, M. Wakata, R. Usami, K. Hamada, A. 112 T. Maeda, N. Sakai, F. Izumi, T. Wada, H. Yamauchi, H. Asano Fukuoka and H. Yamauchi, Physica C, 1994, 219, 33. and S. Tanaka, Physica C, 1992, 193, 73. 69 P. Zoller, J. Glaser, B. Seling, A. Ehmann, W. Wischert and 113 H. Sasakura, H.Teraoka, K. Nakahigigashi, S. Minamigawa and S. Kemmler-Sack, Physica C, 1994, 235–240, 955. K. Yoshiara, Jpn. J. Appl. Phys., 1992, 31, L25. 70 I. K. Gopalakrishnan, H. Rajagopal, J. V. Yakhmi, A. Sequeira 114 J. Min, J. Liang, C. Wong, F. Wu, H. Chen, C. Dong and G. Rao, and R. M. Iyer, J. Supercond., 1994, 7, 857. J Phys.: Condens.Matter, 1995, 7, 5975. 71 S. Kambe, T. Akao, I.Shime, S. Ohshima and K. Okuyama, 115 A. Nara, T. Wada, A. Ichinose, H. Yamauchi and S. Tanaka, Mater. Sci. Eng. B, 1995, 32, 57. Physica C, 1991, 185–189, 599. 72 T. P. Beales, C. Dineen, S. R. Hall, M. R. Harrison and J. M. 116 T. Wada, A. Nara, A. Ichinose, H. Yamauchi and S. Tanaka, Parberry, Physica C, 1993, 207, 1. Physica C, 1992, 192, 181. 73 W. Tang, J. Liang, J. Yang and G.Rao, J. Alloys Compd., 1996, 117 W. Shizong, T. Kaibin, Q. Yitai, Y. Peidong, C. Zuyao, 240, L4. Y. Weichao and Z. Guien, Physica C, 1994, 226, 121. 74 Y. Qian, K. Tang, P. Yang, Z. Chen and R. Li, Physica C, 1993, 118 T. Wada, A. Ichinose, H. Yamauchi, and S. Tanaka, Physica C, 209, 516. 1990, 171, 344. 75 T. P. Beales and J. M. Parberry,Mater. Res. Bull., 1997, 32, 1293. 119 T.Wada, A. Ichinose, F. Izumi, A. Nara, H. Yamauchi, H. Asano 76 J. B. Shi, H. J. Tsao and M. F. Tai, Physica C, 1996, 261, 201. and S. Tanaka, Physica C, 1991, 179, 455. 77 A. Kharlanov and J. P. Attfield, Physica C, 1996, 265, 315. 120 A. Maignan, C. Michel, M. Hervieu, J. Provost, D. Groult and 78 S. Kambe,, T. Akao, S. Ohshima and K. Okuyama, Physica C, B. Raveau, Physica C, 1992, 191, 333. 1995, 255, 151. 121 B. Seling, A. Ehmann, S. Kemmler-Sack, J. Linhart and 79 T. P. Beales and J. M. Parberry, J. Alloys Compd., in press. W. Reimers, Phys. Status Solidi A, 1996, 156, 451. 80 R. S. Liu, W. Zhou and P. P. Edwards, Appl. Phys. L ett., 1990, 122 R. J. Cava, B. Batlogg, R. B. van Dover, J. J. Krajewskii, J. V. 57, 2492. Waszczak, R. M. Fleming, W. F. Peck Jr., L.W. Rupp, P. Marsh, 81 Z. Iqbal, B. L. Ramakrishna and J. C. Barry, Physica C, 1990, A. C. W. P. James and L. F. Schneemayer, Nature (L ondon), 1990, 169, 396. 345, 602. 82 Y. F. Li and Z. Z. Sheng,Mod. Phys. L ett. B, 1991, 5, 1661. 123 A. T. Matveev, J. Ramirez-Castellanos, Y. Matsui and E. 83 Y. F. Li, Z. Z. Sheng and D. O. Pederson, in Superconductivity Takayama-Muromachi, Physica C, 1996, 262, 279.and its applications, 5th Annual Conference, BuValo, NY, 124 R. J. Cava, B. Batlogg, J. J. Krajewski, L. W. Rupp, L. 24–26 September 1991, AIP Conference Proceedings No. 251, F. Schneemeyer, T. Siegrist, R. B. VanDover, P. Marsh, W. 1992, pp. 376–387. F. Peck Jr., P. K. Gallagher, S. H. Glarum, J. H. Marshall, R. C. 84 Y. F. Li, O. Chmaissen and Z. Z. Sheng, Physica C, 1995, 248, 42.Farrow, J. V. Waszczak, R. Hull and P. Trevor, Nature (L ondon), 85 V. Hardy, A. Ruyter, A.Wahl, A. Maignan, D. Groult, J. Provost, 1988, 336, 211. Ch. Simon and H. Noel, Physica C, 1996, 257, 16. 125 M. A. Subramanian, J. Gopalakrishnan, C. C. Torardi, P. L. Gai, 86 R. S. Liu, S. F. Hu, D. A. JeVerson, P. P. Edwards and P. D. E. D. Boyes, T. R. Askew, R.B. Flippen, W. E. Farneth and A. W. Hunneyball, Physica C, 1993, 205, 206. Sleight, Physica C, 1989, 157, 124. 87 K.-B. Tang, Y.-T. Qian, Z.-Y Chen, L. Yiang and Y.-H. Zhang, 126 O. Inoue, S. Adachi, Y. Takahashi, H. Hirano and S. Kawashima, Physica C, 1995, 248, 11. Jpn. J. Appl. Phys.,1989, 28, L60. 88 G. Van Tendeloo, M. Hervieu, X. F. Zhang and B. Raveau, 127 H. W. Zandbergen, K. Kadowski, M. J. V. Menken, A. A. J. Solid State Chem., 1995, 114, 369. Menovsky, G. Van Tendeloo and S. Amelinckx, Physica C, 1989, 89 Z. Hiroi, J. Shimoyama and K. Kishio, Physica C, 1996, 257, 210. 158, 155. 90 O. Chmaissen and Z. Z. Sheng, Physica C, 1995, 242, 23. 128 K. Kadowski, M. J. V. Menken and A. C. Moleman, Physica C, 91 O. Chmaissen and Z. Z. Sheng, Physica C, 1995, 242, 228. 1989, 159, 165. 92 K.-B. Tang, Y.-T. Qian, Z.-Y Chen, L. Yiang, L-B. Wang and Y.- 129 Z. Korczak, W. Korczak, S. Kolesnik, T. Skoskiewicz and H. Zhang, Physica C, 1995, 242, 216. J. Igalson, Supercond. Sci. T echnol., 1990, 3, 370. 93 T. Mochiku, T. Nagashima, M.Watahiki, Y. Fukai and H. Asano, 130 M. M. Lukina, V. N. Milov, V. V. Moschchalkov and A. V. Jpn. J. Appl. Phys., 1989, 28, L1926. Gavrilov, Supercond. Chem. Phys. T echnol., 1991, 3, 1962. 94 S. Adachi, K. Setsune and K. Wasa, Jpn. J. Appl. Phys., 1990, 131 S. H. H. Naqvi, F. Beech and I. W. Boyd, Appl. Surf. Sci., 1992, 29, L890. 54, 166. 95 C. Martin, D. Bourgault, C. Michel, J. Provost, M. Hervieu and 132 R. A. Hughes, Y. Lu, T. Timusk and J. S. Preston, Appl. Phys. B. Raveau, Eur. J. Solid State Inorg. Chem., 1989, 26, 1. L ett. B, 1991, 58, 762. 96 D. Bourgault, C. Martin, C. Michel, M. Hervieu, J. Provost and 133 R. Retoux, C. Michel, M. Hervieu and B. Raveau, Mod. Phys. B. Raveau, J. Solid State Chem., 1989. 78, 326. L ett B, 1989, 3, 591. 97 H. Sasakura, K. Nakahigashi, A. Hirose, H. Teraoke, S. 134 W. T. Fu, H. W. Zandbergen, W. G. Haije and L. J. DeJongh, Minamigawa, K. Inada, S. Noguchi and K. Okuda, Jpn. J. Appl. Physica C, 1989, 159, 210. Phys., 1990, 29, L1628. 135 X. X. Tang and D. E. Morris, Phys. Rev. B, 1991, 44, 4553. 98 S. Adachi, K. Setsune and K. Wasa, Jpn. J. Appl. Phys., 1990, 136 M. Reedyk, T. Timusk, J. S. Xue and J. E. Greedon, Phys. Rev. 29, L1799. B, 1992, 45, 7406. 99 S. Adachi, K. Setsune and K. Wasa, Jpn. J. Appl. Phys., 1990, 137 A. L. Kharlanov, E. V. Antipov, L. M. Kovba, L. G. Akselrud, 29, L2183. I. G. Muttik, A. A. Gippus and V. V. Moshchalkov, Physica C, 100 M. Shida, E. Ohshima, M. Kikuchi, M. Nagoshi and Y. Syono, 1990, 169, 469. Physica C, 1996, 268, 95. 138 T. Mochiku, M. Osawa and H. Asano, Jpn. J. Appl. Phys., 1990, 101 S. Adachi, O. Inoue, H. Hirano, Y. Takahashi and S. Kawashima, 29, L1406. Jpn. J. Appl. Phys., 1990, 28, L775. 139 S. Adachi, O. Inoue, S. Kawashima, H. Adachi, Y. Ichikawa, 102 R. S. Liu, D. S. Shy, S. F. Hu and D. A. JeVerson, Physica C, 1993, K. Setsune and K. Wasa, Physica C, 1990, 168, 1. 216, 237. 140 M. Marezio, A. Santoro, J. J. Caponi, E. A. Hewat, R. J. Cava 103 D. Pelloquin, C. Michel, M. Hervieu, F. Struder and B. Raveau, and F. Beech, Physica C, 1990, 169, 401. 141 P. Goodman, D. G. Jensen and T. J. White, Physica C, 1989, Physica C, 1996, 257, 195. 158, 173. 104 D. Pelloquin, A. Maignan, A. Guesdon, V. Hardy and B. Raveau, 142 S. H. H. Naqvi and I. W. Boyd, Mater. Sci. Eng. B, 1995, 33, 67. Physica C, 1996, 265, 5. 105 J. B. Mandel, B. Bandyopadhyay, F. Fauth, T. Chattopadhyay and B. Ghosh, Physica C96, 264, 145. Paper 7/03804F; Received 2nd June, 1997 12 J. Mater. Chem., 1998, 8(1), 1–12

ISSN:0959-9428    

DOI:10.1039/a703804f

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

J O U R N A L O F C H E M I S T R Y Materials Communication Polymer mesofibres Stacy A. Johnson,a Deepa Khushalani,b Neil Coombs,c Thomas E. Mallouk*a and GeoVrey A. Ozin*b aDepartment of Chemistry, T he Pennsylvania State University, University Park, Pennsylvania 16802, USA bMaterials Chemistry Research Group, L ash Miller Chemical L aboratories, University of T oronto, 80 St. George Street, T oronto, Ontario, Canada M5S 3H6 cImagetek: Analytical Imaging, 32Manning Avenue, T oronto, Ontario, Canada M6J 2K4 samples were then shaken in 48% hydrofluoric acid overnight and filtered to leave the organic replica materials. Ashing A novel process is described for synthesizing high aspect ratio, controlled diameter mesoscale poly(phenolformaldehyde) analysis of the products consistently gave compositions with fibres using the well defined channels of mesoporous silica as a <2% residual silicate.Characterization of the samples was mold and their extraction with the structure intact. accomplished using powder X-ray diVraction (PXRD), adsorption studies, elemental analysis, 13C CP MAS NMR, FTIR spectroscopy and HRTEM. PXRD analysis of the silica/polymer composite (before etching) displayed a pattern at low 2h values that is characteristic of the hexagonal form of mesoporous silica.The low Ever since the discovery of the MCM-41S family of mesoporous electron contrast of the carbon-based polymer compared to materials by Kresge et al.,1 a great deal of interest has been that of the silica host had little eVect on the intensity of the generated in the study of various aspects of these materials PXRD pattern of the composite relative to the empty host.At including varying pore sizes through a variety of techniques,2 a 2h value of ca. 25°, a broad peak was observed suggesting tuning the framework composition,3 and employing the the presence of disordered polymer in a glassy silica host internal channels for probing reactions in restricted matrix.Moreover, this peak was retained for the sample dimensions.4 obtained after etching, characterizing the presence of an Here we present a route to the formation of a polymer mold amorphous polymer. Elemental analysis on the silica/polymer of the channel structure of the hexagonal phase of MCM-41 composite sample gave 45% carbon content.For an ideal mesoporous silica. The technique involves creating a replica complete filling of the channels, the required carbon content by polymerizing formaldehyde and phenol inside the channels. needed to be 59%, and hence the channels can be assumed to This replica, through the novel idea of etching the host around be ca. 75% filled. This degree of mesopore filling is consistent it, is then able to be characterized ex situ by various physical with the results of 77 K N2 adsorption isotherms recorded for methods.High resolution transmission electron microscopy calcined mesoporous silica before and after polymer encapsul- (HRTEM) in particular allows the study of the structure and ation in the channels which show that the accessible pore morphology of the resultant polymer fibres after extraction. It volume of the mesoporous silica has diminished to about 85 should be noted that in the past there has also been some vol%. 13C CP MAS NMR spectra of the extracted polymer debate over the actual access of diVerent sized molecular displayed two broad signals at 40 and 130 ppm. These peaks adsorbates into the mesoporous silica. Channel restriction or can be assigned to the presence of the methylene and aromatic blockage was envisaged in certain cases.Also, it was debatable groups, respectively. Aside from some slight line narrowing in whether the reactions were occurring inside the mesoporous the aromatic region and small intensity changes in the methyl- structure or instead on the external surface of the particle.The ene region, the spectrum of the extracted polymer was more- technique presented here helps to clarify these issues and or-less the same as that of bulk poly(phenolformaldehyde). provides a way of directly visualizing the length of the channels Further evidence of the presence of the polymer was obtained within mesoporous silica materials. by FTIR spectroscopy. Characteristic bands for poly(phenol- Mesoporous silica was synthesized according to the literaformaldehyde) at ca. 3500 cm-1 were assigned to nOH stretch- ture procedure5 followed by calcination at 540 °C over 12 h. ing modes and several sharp bands at ca. 1500 cm-1 were The final surface area, pore size, and pore volume were attributed to CMO, CMC groups similar to those found for determined to be 969.2 m2 g-1, 38.7 A ° , and 1.40 ml g-1, the bulk poly(phenolformaldehyde). The combination of the respectively.In order to fill the pores with the phenolic resin, NMR and FTIR results indicated that the mesoscale size solid phenol was added to a flask containing the MCM-41 constraint aVorded by the channels did not cause extensive template and incubated at 65 °C under reduced pressure overdeviation in polymer connectivity and dynamics from that of night.The amount of phenol incorporated was determined the bulk. Nevertheless, NMR relaxation studies and the [1.57 g (g MCM-41)-1] by the pore volume of the dehydrated measurement of the elastic constants of the fibres by, for MCM-41 material. Excess solid paraformaldehyde [1.27 g (g example, AFM will be needed to explore whether there exists MCM-41)-1] was heated to 120 °C in order to liberate monany alteration in the dynamical and mechanical properties omeric formaldehyde, which was transferred as a vapour to between the mesofibre and bulk forms of the polymer.the phenol/MCM-41 composite. To initiate polymerization, The most interesting information was gained by HRTEM. anhydrous HCl vapor produced from a mixture of NaCl and It should be emphasized that because of the low electron H2SO4 was allowed to enter the reaction chamber.The contrast of the extracted sample, HRTEM studies of the resulting polymer/porous silica composite was cured in argon polymer necessitated the use of direct imaging on ultrathin at 125 °C for 3 h and then at 500 °C for 12 h in order to induce ‘holey’ carbon film (ca. 7 A° thick), and indirect imaging using cross-linking. A subsequent second loading of the monomer was performed to ensure complete filling of the channels. The heavy metal negative staining using a dilute solution of uranyl J. Mater. Chem., 1998, 8(1), 13–14 13Fig. 1 Transmission electron micrograph (using negative staining technique with uranyl acetate solution) of polymer mesofibres supported on a carbon film.Magnification bar=100 nm. Fig. 2 Transmission electron micrograph of control sample—polymer deposited on amorphous silica (Cab-O-Sil). Magnification bar= acetate on thin carbon-support film, Fig. 1. From the micro- 100 nm. graph, it is obvious that only fibre-like morphology is seen for the extracted poly(phenolformaldehyde).The width of the fibres, as determined through the micrographs, is found to be viously reported,4,6,7 no studies have been described on the extraction of these materials and their structural characteriz- ca. 20 A ° with an eVective resolution of 7 A ° (combination of uranyl acetate stain and other specimen parameters). These ation. Hence, the concept of synthesizing copies of the channels of mesoporous silica and extracting them provides access to a are comparable to the pore diameter of the host taking into consideration that only between 75 and 85% of the available diversity of mesofibres with a range of compositions and sizetunable properties.volume is occupied. Furthermore, depending on extraction, sonication and solvent protocols both single and bundles of polymer strands are observed.These bundles are discernible T.M. would like to acknowledge the National Institute of Health (GM 43844) and G.A.O. would like to acknowledge from single fibres by their size and texture. These display extensive regions of curvature thereby indicating appreciable the Natural Sciences and Engineering Research Council for support of this work.D.K. is grateful for a University of flexibility of the poly(phenolformaldehyde) mesofibres. The mesofibres were found to display extensive lengths which were Toronto Open Scholarship. comparable to the particle size of the mesoporous silica host, and the aspect ratios (ratio of the length to the diameter) of References the extracted polymer fibres were found to exceed 103.This suggests that for the mesoporous silicas, the channels run 1 C. T. Kresge, M. E. Leonowicz, W. J. Roth, J. C. Vartuli and J. S. Beck, Nature (L ondon), 1992, 359, 710. practically the entire length of the particles (whose sizes are in 2 D. Khushalani, A. Kuperman, G. Ozin, K. Tanaka, J. Garces, the range 1–10 mm, depending on the synthesis conditions). M. Olken and N. Coombs, Adv.Mater., 1995, 7, 842.Control TEM experiments have been performed for bulk 3 W. Zhang, M. Froba, J.Wang, P. Tanev, J.Wong and T. Pinnavaia, forms of poly(phenolformaldehyde), as well as samples synthe- J. Am. Chem. Soc., 1996, 118, 9164. sized on the surface of non-porous Cab-O-Sil silica that had 4 T. Maschmeyer, F. Rey, G. Sankar and J. M. Thomas, Nature (L ondon), 1995, 378, 159; J. Felipe Diaz and K. J. Balkus Jr., J. Mol. been pre-treated under identical conditions to those used for Catal. B, 1996, 2, 115. the mesoporous silica channel host material. The TEM images 5 D. Khushalani, A. Kuperman, N. Coombs and G. Ozin, Chem. of the polymer mesofibres show distinct morphologies, Fig. 1, Mater., 1996, 8, 2188. while only shapeless polymer particulates are extracted from 6 C.-G. Wu and T. Bein, Science, 1994, 264, 1757. Cab-O-Sil, Fig. 2, and only non-descript polymer agglomerates 7 T. Kyotani, T. Nagai, S. Inoue and A. Tomita, Chem. Mater., 1997, are obtained for the bulk form of the polymer. 9, 609. We wish to emphasize that while mesoporous silica encapsulated polymers, metals and semiconductors have been pre- Communication 7/06791G; Received 18th September, 1997 14 J. Mater. Chem., 1998, 8(1), 13–14

ISSN:0959-9428    

DOI:10.1039/a706791g

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

J O U R N A L O F C H E M I S T R Y Materials Communication Structural and electrical properties of (BEDTTTF) 2X(diiodoacetylene) (X=Cl, Br): the novel self-assembly of neutral Lewis-acidic molecules and halide anions in a molecular metal Hiroshi M. Yamamoto,*† Jun-Ichi Yamaura and Reizo Kato* T he Institute for the Solid State Physics, T he University of T okyo, Roppongi, Minato-ku, T okyo 106, Japan lar conductors.We report here the novel self-assembly of this iodine-substituted neutral molecule and halide anion in the A neutral Lewis-acidic molecule, diiodoacetylene, forms onedimensional chains with halide anions (Cl- and Br-) and crystals of metallic cation-radical salts. Dark brown plates of (BEDT-TTF)2X(DIA) (X=Cl, Br) aVects the donor (BEDT-TTF) arrangement in the cation radical salts, which exhibit metallic resistivity behaviour down were obtained by galvanostatic oxidation of a solution containing BEDT-TTF (ca. 8 mg), DIA (ca. 90 mg) and tetraphe- to 1.6 K. nylphosphonium chloride (ca. 30 mg: X=Cl) or tetra(nbutyl )ammonium bromide (ca. 30 mg: X=Br) in chlorobenzene (20 ml ) under argon atmosphere at 45 °C. A standard Hshaped cell and platinum electrodes (1 mm diameter) were used.A constant current (1.0 mA) was applied for 7 days. X- In the field of molecular conductors, three-component systems Ray structure analyses were performed on these cation-radical such as (BEDT-TTF)2[M(CF3)4](trihaloethane) [BEDTsalts.‡ The crystal structure of (BEDT-TTF)2Cl(DIA) (Cl salt) TTF=bis(ethylenedithio)tetrathiafulvalene; M=Cu, Ag, Au]1 is shown in Fig. 1. The Br salt is isomorphous with the Cl salt. have attracted much attention because the third (neutral) The two donor molecules in the unit cell are interrelated by component (trihaloethane in this example) aVects the crystal an inversion centre, and both Cl- and DIA are on inversion structure and thus tunes its physical properties. In many cases, centres.There is no positional disorder in the terminal ethylene- however, the third component comes from the solvent used dithio fragments of the donor molecule. The donor arrange- for the crystal growth and is incorporated incidentally. The ment resembles the b-type, but is not the same.3 solvent molecules tend to be packed loosely and are often The most striking feature of this crystal is the one-dimen- disordered.In order to produce a well-organized architecture sional (1D) chains constructed from DIA molecules and halide for the three-component system, a more rational method for anions, with the DIA molecules and Cl- anions alternating the introduction of neutral molecules into the molecular conalong the chains. The interatomic Cl···I distance [3.041(1) A ° ] ductors is required.is noticeably shorter than the sum [3.65 or 3.82 A ° ] of the Electron-deficient neutral molecules with halogen atoms can anion radius of Cl- (1.67 A° )4 and the van der Waals radius of form complexes with Lewis bases such as halide anions. iodine [1.98 (Bondi)5 or 2.15 A ° (Pauling)6]. This characteristi- Recently, Ghassemzadeh et al. revealed that diiodoacetylene cally short distance should originate from a strong acid–base (DIA) and halide anions generate two-dimensional networks interaction.In the Br salt, a similar characteristic structure by this type of acid–base interaction.2 This self-assembly should and interatomic distances are also observed [I···Br: 3.154(1); be applicable to the formation of new architectures for molecur( Br-)+r(I)=3.80 (Bondi) or 3.97 A ° (Pauling)]. The donor molecules fit into the channels formed by the 1D chains of DIA and Cl- along the ‘a–c’ direction (Fig. 2). The CMC bonds of the terminal ethylene groups of BEDT-TTF ‡ X-Ray diVraction data were collected on a MAC Science automatic four-circle diVractometer (MXC18) with graphite-monochromated Mo-Ka radiation up to 2h=55°. The intensities were corrected for Lorentz and polarization eVects.The structures were solved by direct methods and refined by full-matrix least-squares methods using reflections with I3s(I). Analytical absorption correction was carried out. Anisotropic thermal parameters were used for non-hydrogen atoms. All calculations were performed with use of ‘teXsan’ crystallographic software package of Molecular Structure Co.Crystal data for (BEDT-TTF)2Cl(DIA): C22S16H16I2Cl, MW= 1082.59, triclinic, space group P19, a=7.642(3), b=17.477(6), c= 6.728(3) A° , a=99.49(3), b=104.90(3), c=83.20(3)°, V=853.8(6) A° 3, Z=1, Dc=2.105 g cm-3, R=0.095, Rw=0.093, 4257 reflection measured, of which 3438 used [I>3s(I)]. Crystal data for (BEDT-TTF)2Br(DIA): C22S16H16I2Br, MW= 1127.04, triclinic, space group P19, a=7.743(1), b=17.651(2), c= 6.739(9) A ° , a=99.82(1), b=106.90(1), c=82.39(1)°, V=864.9(2) A ° 3, Z=1, Dc=2.164 g cm-3, R=0.079, Rw=0.078, 3966 reflection measured, of which 2935 used [I>3s(I)].Fig. 1 Crystal structure of (BEDT-TTF)2Cl(DIA) and distances of Full crystallographic details, excluding structure factors, have been short contacts deposited at the Cambridge Crystallographic Data Centre (CCDC). See Information for Authors, J.Mater. Chem., 1998, Issue 1. Any request to the CCDC for this material should quote the full literature citation and the reference number 1145/69. † E-mail: yhiroshi@kappa.issp.u-tokyo.ac.jp J. Mater. Chem., 1998, 8(1), 15–16 15dynamically.7 This stabilization eVect seems to align BEDTTTF molecules parallel to the 1D chain.It should be noted that in b-(BEDT-TTF)2X salts (X-=linear triatomic anion) the weak CMH···X contacts arrange the X- anion almost perpendicular to the donor plane.8 Overlap integrals of the HOMOs of the donor molecules are shown in Fig. 2. Two strong interactions (r and s) are observed along the direction parallel to the 1D chain. An interaction in this direction is also strong in the b-type salts.In the b-type salts, however, donor molecules along this direction repeat by a unit translation c. The Fermi surface calculated by the tight-binding method is rather one-dimensional (Fig. 3). On the other hand, the Fermi surface of the b- type consists of one- and two-dimensional components. Fig. 4 shows the temperature dependence of the d.c.resistivities (r) of the single crystals measured by a standard four-probe method. Both Br and Cl salts remain metallic down to 1.6 K. The r (room temp.)/r(4.2 K) ratio is fairly large for the Br salt Fig. 2 Crystal packing viewed along the b axis. Overlap integrals (ca. 1500). (S/10-3) are as followings: S(c)=2.45, S(p)=-1.56, S(q)=-1.45, It should be noted that conventional BEDT-TTF salts with S(r)=-8.85, S(s)=-8.94 for the Cl salt; S(c)=2.18, S(p)=-2.17, halide anions (Cl, Br) frequently crystallize from unrelated S(q)=-1.89, S(r)=-8.50, S(s)=-8.59 for the Br salt.electrolytes or solvents (e.g. CoCl4-, CH2 Cl2) and contain water molecules.8 The structural properties of all of these salts are diVerent from those of our system, and these salts exhibit a metal–insulator transition at low temperatures.In summary, we have prepared molecular metals with a novel polymeric configuration based on halide anions and neutral Lewis-acidic molecules. Among polymeric anion layers, this type of self-assembly possesses an advantage over conventional ones such as Cu+[N(CN)2-]Br-,9 because there is no need to adjust the charge in designing the polymeric framework.In addition, the physical and chemical properties of the neutral molecule (for example, electronic polarization of soft iodine atoms in DIA), which can be tuned by the choice of functional groups and molecular shape, should open new possibilities for molecular conductors. Fig. 3 Fermi surface of (BEDT-TTF)2Cl(DIA) References 1 J. A. Schlueter, J.M. Williams, U. Geiser, H. H. Wang, A. M. Kini, M. E. Kelly, J. D. Dudek, D. Naumann and T. Roy,Mol. Cryst. L iq. Cryst., 1996, 285, 43. 2 M. Ghassemzadeh, K. Harms and K. Dehnicke, Chem. Ber., 1996, 129, 259. 3 T. Mori, F. Sakai, G. Saito and H. Inokuchi, Chem. L ett., 1986, 1037. 4 R. D. Shannon, Acta Crystallogr., Sect. A, 1976, 32, 751. 5 A. Bondi, J. Phys. Chem., 1964, 68, 441. 6 L. Pauling, T he Nature of the Chemical Bond, Cornell University Press, Ithaca, N.Y., 1942. 7 H. H. Wang, J. R. Ferraro, K. D. Carlson, L. K. Montgomery, U. Geiser, J. M. Williams, J. R. Whitworth, J. A. Schulueter, S. Hill, M.-H. Whangbo, M. Evain and J. J. Novoa, Inorg. Chem., 1989, 28, 2267; J. J. Novoa, M.-H. Whangbo and J. M. Williams, Mol. Cryst. L iq. Cryst., 1990, 181, 25. 8 J. M. Williams, J. R. Ferraro, R. J. Thorn, K. D. Carlson, U. Geiser, H. H. Wang, A. M. Kini and M. H. Whangbo, Organic Fig. 4 Temperature dependence of resistivity along the c axis (the Superconductors (Including Fullerenes): Synthesis, Structure, most grown axis) for (BEDT-TTF)2X(DIA); (a) X=Cl and (b) X=Br Properties and T heory, Prentice Hall, Englewood CliVs, N.J., 1992 and references sited therein. 9 A. M. Kini, U. Geiser, H. H. Wang, K. K. Carlson, J. M. Williams, are almost parallel to the 1D chain. Several distances between W. K. Kwok, K. G. Vandervoort, J. E. Thompson, D. L. Stupka, hydrogen atoms and iodine atoms or chloride anion are short D. Jung and M. H. Whangbo, Inorg. Chem., 1990, 29, 2555. enough to be regarded as van der Waals contacts (Fig. 1). These short contacts stabilize the crystal structure thermo- Paper 7/07301A; Received 9th October, 1997 16 J. Mater. Chem., 1998, 8(1), 15–16

ISSN:0959-9428    

DOI:10.1039/a707301a

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

J O U R N A L O F C H E M I S T R Y Materials Communication EVect of the porosity and relative surface hydrophobicity of MCM-41 materials on the adsorption of [60]fullerene at the solid/toluene interface: comparison of Al- and Zr-doped silicates with the purely siliceous sample Ireneusz Piwonski, Jerzy Zajac,* Deborah J. Jones, Jacques Rozie`re and Stanislas Partyka L aboratoire des Agre�gats Mole�culaires et Mate�riaux Inorganiques, ESA 5072, Universite� Montpellier II, C.C. 015, Place E. Bataillon, 34095 Montpellier Cedex 5, France ratio Si/X where X=Al, Zr. C14 indicates the 14 carbon atoms of the surfactant tail used in the synthesis. The adsorption of [60]fullerene from toluene on three MCM-41 type materials: purely siliceous, aluminosilicate An automated Philips diVractometer with Cu-Ka radiation was used to acquire X-ray powder diVraction data.X-Ray and zirconiosilicate, is thought to occur primarily on the pore walls, and diVerences in the amount adsorbed between diVraction patterns of as-synthesised and calcined samples showed two or three peaks, which could be indexed in the the three samples can be ascribed to the diVerent size, volume and surface area of their pores rather than to the diVerent hexagonal system.The unit cell parameter was calculated using a0=2d100/Ó3.1 It is clear that the addition of aluminium and hydrophobic character of their surfaces. zirconium does not greatly aVect this parameter. 27Al MAS NMR spectroscopy of SiAl29C14 gave a signal at 53 ppm in both as-synthesised and calcined samples, indicating the presence of four-coordinate aluminium only.Nitrogen adsorption–desorption measurements were per- In spite of numerous theoretical and experimental studies on formed on all calcined samples using an automated volumetric the physical and chemical characteristics of MCM-41 type apparatus (Sorptomatic 1800) at 77 K. The isotherms are of mesoporous solids over the past five years, their interfacial type IV according to the IUPAC classification, do not exhibit properties have received much less attention.In view of their hysteresis loops, and have a well marked capillary condensation potential applications in selective catalysis and adsorption, it step, indicating that the pore size distribution is very homo- is necessary to describe the porous structure and hydrophilic– geneous in each case.The nitrogen adsorption isotherms served hydrophobic character of the MCM-41 materials prepared as a basis for determining some important parameters of the using diVerent synthesis routes and incorporating (or not) materials synthesised, including BET specific surface area heteroatoms in the silica matrix. In this paper the results of (SBET), pore surface area (Spore), external surface area (Sext), butanol adsorption from water and nitrogen gas adsorption– pore volume (Vpore), and the most probable pore diameter (d). desorption measurements on three MCM-41 samples are The methods applied have been described previously.4 These reported in order to describe the porosity and hydrophobic results are collected in Table 1.All the MCM-41 type materials character of their surfaces. These studies have been supprepared have high surface areas and the eVective diameter of plemented by X-ray diVraction and 27Al MAS NMR spectheir pores is in the mesoporous size range. troscopy. These properties are used to explain the adsorption Flow microcalorimetry was used to measure the enthalpy of C60 fullerene at the solid/toluene interface.of butanol adsorption from water. Details of the experiment The three MCM-41 samples were synthesized as follows. have been given elsewhere.5 Adsorption from solution is known (1) Purely siliceous SiC14, prepared using Beck’s method1 to have a competitive character. When adsorbed from water, using 45 g of deionised water; 14 g of sodium silicate solution, the butanol molecules will be retained on the surface because Na2O 7.5–8.5%, SiO2 25.5–28.5%, (Merck); 46.6 g (25 mass% their hydrophobic moieties are expelled from an aqueous solution) of n-tetradecyltrimethylammonium bromide environment and simultaneously attracted by the non-polar (Lancaster); 0.9 g of 96% sulfuric acid (Carlo Erba).(2) (hydrophobic) sites in the solid surface.The oncoming alcohol Aluminosilicate SiAl29C14, prepared according to Klinowski and co-workers2 using 50 g of deionised water; 5.9 g of sodium Table 1 Interfacial properties of the MCM-41 materials silicate solution (Merck); 4.5 g of fumed silica Cab-O-Sil M5; 35.9 g (25 mass% solution) of n-tetradecyltrimethylammonium SiC14 SiAl29C14 SiZr25C14 bromide (Lancaster); 10 g of tetramethylammonium hydroxide, 25 mass% solution (Aldrich); 96% sulfuric acid (Carlo Erba); SBET/m2 g-1 761 1029 979 Amberlite IRA-420C strongly basic ion exchanger (Sigma); Sext/m2 g-1 116 11 13 Spore/m2 g-1 583 999 939 0.42 g of aluminium sulfate.(3) Zirconiosilicate SiZr25C14, Vpore/cm3 g-1 0.59 0.83 0.78 synthesised using the method developed by Maireles-Torres d/nm 2.8 3.3 2.8 and co-workers3 using 18.7 g of tetraethoxysilane 98% a0/nm 4.0 4.1 4.1 (Aldrich); 1.7 g of zirconium(IV) propoxide, 70 mass% solution DdplH/mJ m-2 5.2 2.0 2.8 in propan-l-ol (Aldrich); 5 ml of tetraethylammonium hydrox- Cads/mmol g-1 2 11 3 ide 35 mass% solution in water (Aldrich); 15.1 g of n-tetradecyl- Cdes/mmol g-1 -2 -10 -3 trimethylammonium bromide (Lancaster); 26.1 ml of ethanol; DdplH=integral molar enthalpy (per unit surface area of the adsorbent) 6.7 ml of n-propanol.of adsorption of n-butanol from water using a 10 g l-1 solution at In syntheses (2) and (3) SiXmC14, m represents the atom 298 K; Cads=amount of C60 adsorbed on the adsorbent surface from 0.7 mmol l-1 toluene solution at 298 K; Cdes=amount of C60 desorbed from the adsorbent surface by the adsorbing toluene molecules at 298 K.* E-mail: zajac@univ-montp2.fr J. Mater. Chem., 1998, 8(1), 17–18 17molecules will displace interfacial water. The butanol concen- magnitude greater than those of the other two solids. If the tration used in the study corresponds to the formation of a adsorption had been limited only to the external surface, SiC14 monolayer on the hydrophobic part of the MCM-41 surface.would have adsorbed many more C60 molecules than the Therefore, the integral enthalpy of displacement per unit others. On the contrary, this sample gives the lowest level of surface area, i.e., DdplH, is proportional to the surface area of adsorption. Furthermore, this diVerence cannot be explained hydrophobic sites in the MCM-41 materials.The relative by the diVerent hydrophobic character of the MCM-41 surfaces hydrophobicity decreases in the order SiC14>SiZr25C14> because SiAl29C14 and SiZr25C14, which have almost the SiAl29C14. For SiC14, the relative hydrophobicity same relative hydrophobicity, have very diVerent adsorption is about twice that of the samples doped with Al or Zr. aYnities for C60.It is more reasonable to relate such behaviour The presence of Al or Zr species in the silicate structure is to diVerent accessibility of the MCM-41 porous structure to reported to induce the formation of Lewis acid sites.2,3 The C60 adsorption. The two materials having smaller pore sizes increased Lewis acidity of SiZr25C14 and SiAl29C14 compared adsorb fewer C60 molecules, inferring that adsorption seems to with SiC14 causes the percentage of polar surface sites in the occur mainly on the internal surface.Between these two first two adsorbents to increase and, consequently, the number samples, the amount adsorbed obtained with SiC14 is the of butanol molecules adsorbed from water to be greatly smallest because this support has the lowest pore volume and reduced.As a result, the relative surface hydrophobicity surface area of the samples studied. The nature of active sites decreases. in the MCM-41 surface relevant to the phenomenon is still The same flow microcalorimeter equipment was used to not clear and some additional studies are necessary. In particudetermine the adsorption of C60 fullerene from a toluene lar, C60 adsorption and desorption at diVerent bulk concensolution. Previously reports have described the inclusiotions are currently being determined and compared with [60]fullerene (C60) in siliceous MCM-41 by sublimation of the related enthalpy eVects.The results will be reported the former at 873 K under vacuum6 and, unusually, grinding subsequently. together of the two components.7 The possibility of loading aluminophosphates with C60 from the liquid and gas References phases has also been studied.8 Here, the amount of C60 adsorbed was determined using an on-line UV detector at l= 1 J.S. Beck, J. C. Vartuli, W. J. Roth, M. E. Leonowicz, C. T. Kresge, 404 nm. The results are presented in Table 1. Fullerene adsorp- K. D. Schmitt, C. T-W. Chu, D. H. Olson, E.W. Sheppart, tion seems to be reversible, but the amounts adsorbed are S. B. McCullen, J. B. Higgins and J. L. Schlenker, J. Am. Chem. Soc., generally very small. However, there are pronounced diVer- 1992, 114, 10 834. 2 Z. Luan, C. F. Cheng, W. Zhou and J. Klinowski, J. Phys. Chem., ences in the amount adsorbed between the three samples, 1995, 99, 1018. SiAl29C14 giving by far the greatest adsorption value.Using 3 D. J. Jones, J. Jime�nez-Jime�nez, A. Jime�nez-Lopez, P. Maireles- the BET specific surface area of each MCM-41 sample, the Torres, P. Olivera-Pastor, E. Rodriguez-Castellon and J. Rozie`re, average surface area available to each adsorbed C60 molecule Chem. Commun., 1997, 431. can be calculated. The following values have been obtained: 4 M.J. Meziani, J. Zajac, D. J. Jones, J. Rozie`re and S. Partyka, SiC14, 634 nm2; SiZr25C14, 580 nm2; and SiAl29C14, 156 nm2. L angmuir, 1997, 13, 5409. 5 J. Zajac and A. J. Groszek, Carbon, 1997, 35, 1053. These values are far from the 0.8 nm2 which was determined 6 F. Rachdi, L. Hajji, C. Goze, D. J. Jones, P. Maireles-Torres and from the projection of a C60 molecule in a hexagonal close- J. Rozie`re, Solid State Commun., 1996, 100, 237. packed arrangement onto a flat surface.5 Clearly, only a very 7 J. Chen, Q. Li, H. Ding, W. Pang and R. Xu, L angmuir, 1997, small proportion of the total MCM-41 surface is occupied by 13, 2050. the adsorbing C60 molecules under the dynamic conditions 8 A.Gu� gel, K. Mu� llen, H. Reichert, W. Schmidt, G. Scho� n, F. Schu� th, used. It is of interest to consider which part of the solid surface J. Spickermann, J. Titman and K. Unger, Angew. Chem., Int. Eng. Engl., 1993, 32, 556. is responsible for C60 adsorption and, here, the diVerent relative external and pore surface areas of the three samples are of use. The external surface area of SiC14 is about one order of Communication 7/07158B; Received 3rd October, 1997 18 J. Mater. Chem., 1998, 8(1

ISSN:0959-9428    

DOI:10.1039/a707158b

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

J O U R N A L O F C H E M I S T R Y Materials Communication Ultrafine La–Mo and Ce–Mo complex oxide particle catalysts for selective oxidation of toluene Wenxing Kuang,* Yining Fan, Jinghen Qiu and Yi Chen Department of Chemistry, Institute of Mesoscopic Solid State Chemistry, Nanjing University, Nanjing 210093, China gelation was completed, and then the gels thus prepared were dried at 393 K for 4 h and calcined in air at 773 K for the It has been found that by decreasing the particle size of La–Mo and Ce–Mo complex oxides to nanoscale, the La–Mo gel and 673 K for the Ce–Mo gel to aVord the oxide catalysts.reactivity of lattice oxygen ions and thus the selectivity for oxidation of toluene to benzaldehyde can be remarkably The morphology, particle size and structure of the oxide catalysts were characterized by using X-ray diVraction (XRD), improved. transmission electron microscopy (TEM) and laser Raman spectroscopy (LRS) techniques.The Brunauer–Emmett–Teller (BET) surface areas of the samples were measured by using a Micromeritics ASAP-2000 instrument (N2 adsorption at 77 K). The oxide catalysts were introduced into a U-type quartz fixed Selective oxidation of hydrocarbons to organic oxygenate bed microreactor and their catalytic properties for the selective compounds over metal oxide catalysts is one of the most oxidation of toluene to benzaldehyde were evaluated under important catalytic process in the chemical industry.It is well the following reaction conditions: 0.1 MPa, air/toluene=951 known that Mo-based and V-based oxide catalysts have been (vol/vol), F (flow rate of feed gas)/W (catalyst mass)=1900 ml widely used for selective oxidation of hydrocarbons to alde- (h g cat)-1. Reaction products were analyzed by on-line gas hydes.1 In the past decades, considerable attention has been chromatography.devoted to the design of highly selective oxidation catalysts, The image results show that the particle size are in the range and great eVorts have been made in improving the catalytic of 40–80 nm for La–Mo oxide (La/Mo=1.0) and 20–40 nm selectivity by adding some transition metal oxides to the for Ce–Mo oxide (Ce/Mo=1.0).This indicates that the La–Mo catalysts. It has been found that Bi, Sn, Fe and W oxides are and Ce–Mo oxide catalysts prepared by the sol–gel method eVective promoters of Mo-based and V-based oxide catalysts are ultrafine oxide particles (<100 nm). The BET surface areas for oxidation of toluene to benzaldehyde,2–7 and that the for La–Mo and Ce–Mo complex oxides are 21.3 and preparation conditions, oxide composition and catalyst struc- 19.0 m2 g-1, respectively.The XRD patterns of La–Mo and ture exert great influences on the catalytic properties.In recent Ce–Mo samples show that the rare earth oxide and molybyears, ultrafine metal oxide particles have attracted much denum oxide in both samples form complex oxides having the research interests in terms of materials science and heterostructures of La2Mo2O9 and Ce2(MoO4)3 with an excess of geneous catalysis.8,9 These new catalytic materials are expected CeO2, respectively.to have unique catalytic properties because of their nanoscale The toluene oxidation reaction was used as a probe to particle sizes. evaluate the catalytic properties of the ultrafine La–Mo and Recently, it has been found that some rare earth species Ce–Mo complex oxide particle catalysts. Under our reaction such as lanthanum and cerium exert an influence on the conditions the reaction products on the above catalysts were selective oxidation of toluene to benzaldehyde over V-based mainly CO, CO2, H2O and benzaldehyde.The specific activities oxide catalysts.10,11 For instance, Yan et al.10 reported that for selective oxidation of toluene to benzaldehyde (benzaladding cerium to vanadium oxide catalysts further increased dehyde yield per specific surface area of catalysts) over La–Mo selectivity for benzaldehyde.However, studies on lanthanum and Ce–Mo complex oxide catalysts with diVerent molar ratios or cerium used as promoters of Mo-based oxide catalysts to of metallic ions are shown in Fig. 1. Both rare earth oxides improve the selectivity for benzaldehyde in oxidation of toluene are rather limited.On the other hand, although Mo-based oxide catalysts have been extensively investigated, relatively little work has been done on the preparation of ultrafine Mobased oxide particle catalysts and their catalytic properties for selective oxidation of hydrocarbons to organic oxygenate compounds. In this work, we report our recent results on the preparation of ultrafine La–Mo and Ce–Mo complex oxide particles and their catalytic properties for the selective oxidation of toluene to benzaldehyde.Ultrafine La–Mo and Ce–Mo complex oxide particle catalysts with diVerent molar ratios of metallic ions (La/Mo or Ce/Mo) were prepared by the sol–gel method. Lanthanum nitrate, La(NO3)3·6H2O, or cerium nitrate, Ce(NO3)3·6H2O, ammonium molybdate, (NH4)6Mo7O24·6H2O (the total amount of metallic ions is 0.06 mol), and citric acid (0.02 mol) aqueous solutions were prepared separately and then mixed.The pH values of mixture solutions were adjusted to 1.0 for Fig. 1 The specific activity for selective oxidation of toluene to benzalthe La–Mo solution and 0.5 for the Ce–Mo solution. The dehyde over La–Mo oxide catalysts (a, at 753 K) and Ce–Mo oxide catalysts (b, at 673 K) above solutions were first kept in a water bath at 353 K until J.Mater. Chem., 1998, 8(1), 19–20 19Table 1 The structure and catalytic properties of Ce–Mo complex oxide particle catalysts (Ce/Mo=1.0) with diVerent particle sizes preparation phase structure particle conversion benzaldehyde MoNO method by XRD size/nm of toluene (%) selectivity (%) LRS band/cm-1 sol–gel Ce2(MoO4)3, CeO2 20–40 34.0 37.0 928 precipitation Ce2(MoO4)3, CeO2 >100 35.5 16.0 953 are inactive for the selective oxidation of toluene to benzal- the lattice oxygen ions have higher reactivity.The higher selectivity for oxidation of toluene to benzaldehyde on the dehyde. However, addition of lanthanum and cerium to molybultrafine particle catalyst may be correlated to the higher denum oxides leads to an improvement of the specific activities reactivity of their lattice oxygen ions.for the selective oxidation. It can be seen that with increasing The above results have shown that besides the composition rare earth/(rare earth+Mo) atomic ratios, the specific activities of oxides, the oxide particle size exerts a great influence upon are first increased and then decreased remarkably.Obviously, the catalytic selectivity of complex oxide catalysts. By decreas- the composition of the complex oxides has a great influence ing the particle size to nanoscale, the reactivity of lattice on the catalytic properties of the ultrafine oxide particle oxygen ions and thus the selectivity for oxidation of toluene catalysts. The maximum catalytic activities are reached in the to benzaldehyde can be remarkably improved.The ultrafine vicinity of La/(La+Mo)=0.2 and Ce/(Ce+Mo)=0.5 for complex oxide particles may be potentially new catalytic La–Mo and Ce–Mo oxide catalysts, respectively. materials for selective oxidation reactions. For comparison, the catalytic properties of the ultrafine Ce–Mo oxide particles prepared by the sol–gel method and The support of the National Natural Science Foundation of the corresponding larger oxide particles with the same com- China and SINOPEC is gratefully acknowledged. position prepared by a conventional precipitation method are listed in Table 1.It is interesting to note that the conversions of toluene on the ultrafine Ce–Mo oxide particles and the References corresponding larger particles are very similar, but the selec- 1 V.D. Sokolovskii, Catal. Rev. Sci. Eng., 1989, 32, 1. tivities to benzaldehyde are diVerent, which can not be 2 K. van der Wiele and P. J. van den Berg, J. Catal., 1975, 39, 437. accounted for by the diVerence in particle size or specific 3 J. Buiten, J. Catal., 1968, 21, 188. surface area. Apparently, the nature of the active species for 4 S.Tan, Y. Moro-oka and A. Ozaki, J. Catal., 1970, 17, 125. the selective reaction over the ultrafine and larger oxide 5 M. Ai and T. Ikawa, J. Catal., 1975, 40, 203. 6 J. E. Germain and R. Laugier, C. R. Acad. Sci., Ser. C, 1973, particles should be taken into consideration. Haber12 demon- 276, 1349. strated that the lattice oxygen species (MoNO) in molyb- 7 K.A. Reddy and L. K. Doraiswamy, Chem. Eng. Sci., 1969, 24, denum-based oxide catalysts are the main active species 1415. responsible for selective oxidation of aromatics. The state of 8 G. M. Pajonk, Appl. Catal., 1991, 72, 217. the lattice oxygen species in the above ultrafine and large 9 M. Schneider and A. Baiker, Catal. Rev. Sci. Eng., 1995, 37, 515. 10 Z. Yan, S. Lars and T. Andersson, J. Catal., 1991, 131, 350. Ce–Mo oxide particles were studied by using LRS. As pre- 11 P. J. van den Berg, K. van der Wiele, J. J. J. den Ridder, in Proc. sented in Table 1, the vibrational frequency of MoNO in the 8th Int. Congr. Catal., Berlin, 1984, Elsevier, Amsterdam, 1985, large Ce–Mo oxide particles is at 953 cm-1, while that of vol. 5, p. 393. MoNO in the ultrafine Ce–Mo oxide particles shifts to 12 J. Haber, in Proc. 8th Int. Congr. Catal., Berlin, 1984, Elsevier, 928 cm-1. This red shift indicates that chemical bonding Amsterdam, 1985, vol. 1, p. 85. between MoNO in the ultrafine Ce–Mo oxide particles is weaker than that in the corresponding large particles, so that Communication 7/06525F; Received 8th September, 1997 20 J. Mater. Chem., 1998, 8(1), 19–20

ISSN:0959-9428    

DOI:10.1039/a706525f

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

J O U R N A L O F C H E M I S T R Y Materials The use of novel cyclic monomers in hydrogel synthesis C. B. St. Pourcain,* A. W. P. Jarvie and B. J. Tighe Department of Chemical Engineering and Applied Chemistry, Aston University, Aston T riangle, Birmingham, UK B4 7ET A range of hydrogels was prepared by the copolymerisation of 2-hydroxyethyl methacrylate (HEMA) with various cyclic monomers (CM) derived from cis- and trans-1,2-dihydroxycyclohexa-3,5-diene (DHCD). The specific systems investigated were the cis-diacetate, -dimethyl carbonate and -dipivalate, and the trans-diacetate and -dimorpholinocarboxy.It was observed that the cis-derivatives polymerised more readily than the trans-derivatives. The poly(HEMA–CM) hydrogels prepared from cisderivatives showed superior mechanical properties compared with the corresponding systems containing styrene or methyl methacrylate.The hydrogels are a class of materials which have been used thetic hydrogels could also have a marked influence on water binding. The copolymerisation of DHCD derivatives with widely in biomedical applications. They have the particular property that they absorb and contain water but do not hydrophilic monomers oVers a potential route to high strength, high water content materials.The diacetate 1, dimethyl carbon- dissolve in it. The absorbed water aVects such characteristics as mechanical and surface properties,1,2 permeability and ate 2 and dipivalate 3 systems derived from cis-DHCD and the diacetate 4 and dimorpholinocarboxy 5 from trans-DHCD biocompatibility.The eVect of the contained water is mainly beneficial in that it acts as a plasticiser, a transport medium were copolymerised with HEMA to form a new group of hydrogels. Various properties of these materials were measured for dissolved species and as a link between the body fluids and the synthetic polymer. Its main disadvantage is that it is and the results of these investigations are discussed. HEMA was selected as the comonomer because of the predominance generally observed that the mechanical strength decreases as the water content increases.3 of poly(HEMA) hydrogels in many applications.A number of strategies have been developed for the preparation of hydrogel type materials which are mechanically strong and have high water contents.Included amongst these is the grafting of the hydrogel onto a polymer with superior mechanical properties so that the surface of the support exhibits greatly improved biocompatibility whilst the composite as a whole retains the mechanical strength. These materials are limited in use in particular applications, where permeability is an important factor.4,5 A more generally useful approach has been the development of a new family of OAc OAc O2COMe O2COMe O2CBut O2CBut OAc OAc O2C O2C N O N O 1 2 3 4 5 hydrogels based on interpenetrating polymer networks (IPNs).It is found that IPN formation produces materials with dramatically increased mechanical strength compared with hydro- Experimental gel copolymers of similar water content.6,7 Materials We describe here exploratory studies directed towards an alternative approach to the synthesis of hydrogel materials All solvents were dried and distilled prior to use.The cis-1,2- with high water content and good mechanical properties. If dimethoxycarboxy- 2 and cis-1,2-dipivaloyloxy-cyclohexa-3,5- conventional monomers such as styrene and methyl metha- diene 3 were gifts from ICI (courtesy of D.G. H. Ballard) and crylate are added to hydrogels, e.g. HEMA–N-vinylpyrrolidone the 2-hydroxyethyl methacrylate was supplied by Ubichem. (NVP)–crosslinker systems, in order to stiVen the polymer The cis-1,2-diacetoxycyclohexa-3,5-diene 1 was synthesised by backbone they tend to reduce the hydrophilicity of the material the method of Ballard and co-workers9 and the trans-1,2- and lower the water content of the swollen state.2,8 The diacetoxycyclohexa-3,5-diene 4 according to the procedure of monomer 1,2-dihydroxycyclohexa-3,5-diene (DHCD) and its Platt and Oesch.11 The IR and NMR spectra of these products derivatives could be an exception to this rule.9,10 These dienes were in accord with those observed previously.can be polymerised with acrylic monomers to produce in- The trans-1,2-bis(morpholinocarboxy)cyclohexa-3,5-diene 5 chain cyclohexene rings.These rings would be expected to was synthesised as outlined in Scheme 1. enhance the mechanical strength of the acrylic material. The in-chain ring structure which is absent in conventional syn- trans-1,2-Bis(morpholinocarboxy)-4,5-dibromocyclohexane 7. The trans-1,2-dihydroxy-4,5-dibromocyclohexane11 6 (13.7 g, Table 1 Composition of hydrogels 50 mmol) triethylamine (17.4 cm3, 0.125 mol), morpholinocarbonyl chloride (14.6 cm3, 0.125 mol) and 4-dimethylaminopyri- monomer monomer content of hydrogels, mol% dine (DMAP) (0.09 g, 0.74 mmol) were refluxed for 3 days in 1 9, 17, 24 dry THF (150 cm3).The THF was removed and dichloro- 2 9, 17, 24 methane (150 cm3) added.The solution was washed with water 3 9, 17, 24 (150 cm3), a saturated aqueous solution of sodium hydrogen 4 10, 20, 30 carbonate (3×50 cm3) and water (2×50 cm3), then dried over 5 5, 10, 15 magnesium sulfate. Solvent removal yielded a crude solid J. Mater. Chem., 1998, 8(1), 21–24 21APT (attached proton test) or DEPT (distortionless enhanced polarisation transfer) spectra.Elemental microanalyses were performed by Medac Ltd, Department of Chemistry, Brunel University, Uxbridge, Middlesex. Melting points were determined in capillary tubes with a Gallenkamp Melting Point Apparatus, Model No. ME-370, and are uncorrected. Measurement of the physical properties of hydrogels Determination of equilibrium water content (EWC). EWC determinations were carried out on five separate pieces of gel and the average value calculated.A No. 4 cork borer was used to cut out small discs of gel, which were then placed in a Br Br OH OH Br Br O2C O2C N O N O O2C O2C N O N O i ii 6 7 5 sample bottle of distilled water. For each determination the Scheme 1 Reagents and conditions: i, morpholinocarbonyl chloride, disc was blotted lightly with filter paper, to remove surface DMAP, THF; ii, Li2CO3, LiCl, HMPA water, and weighed.Dehydration of the gel was achieved by placing it in a microwave oven for 12 min, after which time the gel was re-weighed. The EWC was then calculated using product. Washing with diethyl ether gave 16.2 g (32 mmol, eqn. (1) 65%) of the desired material 7, as a pale beige amorphous solid, mp 129–130 °C; nmax/cm-1 (KBr) 1697 (CNO); dH EWC= mass of water in the gel mass of the swollen gel ×100 (1) (CDCl3) 2.35 (d, br, 2H), 2.57 (s, br, 2H), 3.40 (s, br, 8H), 3.59 (s, br, 8H), 4.47 (s, br, 2H), 5.18 (s br, 2H); dC (CDCl3) 34.42, Agreement between samples was at worst ±1% and typically 44.00, 44.31, 50.04, 66.53, 71.24, 154.2 (Found: C, 38.25; H, ±0.5%.The hydrogels were dried to constant mass in the 5.03; N, 5.42. C14H24Br2N2O6 requires C, 38.42; H, 4.84; microwave oven.Identical values were obtained for N, 5.60%). poly(HEMA) dehydrated by the microwave method and dried in a vacuum oven at 60 °C. trans-1,2-Bis(morpholinocarboxy)cyclohexa-3,5-diene 5. The trans compound 7 (5.0 g, 30 mmol) LiCl (3.6 g, 85 mmol) and Measurement of mechanical properties.All mechanical tests Li2CO3 (5.6 g, 76 mmol) in HMPA (100 cm3) were heated, were performed with a Hounsfield Tensometer (Model: HK under nitrogen, for 2.5 h at 90–100 °C. After cooling to room 10KN) fitted with a 10 N load cell, jaws specially designed for temperature dichloromethane (120 cm3) was added followed use with hydrogel samples and a test speed of 8 mm min-1. by the dropwise addition of a 7% v/v aqueous solution of HCl Five samples were tested for each gel and a minimum of three (100 cm3).The aqueous layer was then extracted with dichloroconsistent values used for further calculations. Dumbbell methane (3×60 cm3). The combined organic extracts were shaped samples were used for the cis-1,2-diacetate, -dimethyl washed with water (120 cm3) and a saturated aqueous solution carbonate and -dipivalate containing hydrogels.The samples of sodium hydrogen carbonate (100 cm3). The organic phase for the other gels were parallel sided. A gauge length of 10 mm was dried over magnesium sulfate and the solvent removed. and a width of 3 mm were used for both sets of samples.2 The Purification of the crude material by dry flash chromatography thickness of each sample was measured with a micrometer in (silica gel; ethyl acetate) gave 3.2 g (9.5 mmol, 32%) of the five diVerent places on the gel.An average value was then monomer 5 as a viscous yellow liquid, nmax/cm-1 (neat liquid) calculated. The average values were in the range 0.2–0.5 mm. 1708 (CNO); dH (CDCl3) 3.27 (m, br, 8H), 3.45 (s, br, 8H), 5.44 (s, 2H), 5.67–5.69 (m, 2H), 5.81–5.85 (m, 2H); dC (CDCl3) 43.74, 66.04, 73.17, 124.65, 125.76, 154.0 (Found: C, 56.40; H, Results and Discussion 6.63; N, 7.94. C14H22N2O6 requires C, 56.78; H, 6.55; N, 8.28%).The eVect of CM concentration on EWC for the poly(HEMA–CM) hydrogels is shown in Fig. 1. Hydrogel synthesis12 A mixture of HEMA, the comonomer, and 1% w/w ethylene glycol dimethacrylate were degassed by bubbling nitrogen through the solution for 15 minutes.Azoisobutyronitrile (0.5% w/w) was added and the solution injected into a mould, the needle was removed and the mould was placed in an oven at 60 °C for 3 days followed by 2 h postcure at 90 °C. The mould was then separated and the xerogel placed in 175 cm3 of distilled water to equilibrate for at least two weeks, the water being changed daily.The mould consisted of two polyethylene gaskets (10 cm×6 cm external; 6 cm×2.5 cm internal) sandwiched between two Melinex (polyethylene terephthalate) sheets which had been attached to two glass plates by spray mount adhesive (manufactured by 3M and purchased from BDH). The compositions mentioned in the discussion refer to the composition of the feed mixture. Analyses IR spectra were recorded on either a Nicolet 510 Fourier Transform Infrared Spectrometer or a Perkin Elmer 1710 Fourier Transform Infrared Spectrometer. Solid samples were prepared as KBr discs and liquids as thin films between sodium chloride plates.NMR spectra were recorded on a Bruker AC Fig. 1 EVect of CM concentration on EWC: (+) 1; (&) 2; (+) 3; (6) 4; (%) 5 300 spectrometer.The 13C spectra were recorded as either 22 J. Mater. Chem., 1998, 8(1), 21–24Fig. 3 Comparison of the mechanical properties of poly(HEMA–CM) Fig. 2 EVect of CM concentration on the mechanical properties of hydrogels with those of poly(methyl methacrylate) and polystyrene at poly(HEMA–CM) hydrogels: (+) 1; (&) 2; (+) 3; (6) 4; (%) 5 equivalent composition: (+) 1; (&) 2; (+) 3; (6) styrene; (%) methyl methacrylate Mechanical properties The hydrogels containing the pure cis-derivatives showed the expected decrease in EWC with decrease in HEMA content.The initial tensile strengths and initial moduli of the The decrease in EWC with decreasing HEMA content is poly(HEMA–CM) hydrogels are shown in Fig. 2. The hydrosubstantially less for the cis-diacetate 1 and -dimethyl carbon- gels obtained from the trans-derivatives exhibited comparable ate 2 derivatives than for the bulky dipivalate.The HEMA or lower values of tensile strength than poly(HEMA) itself, copolymers of the cis-diacetate and dimethyl carbonate mon- and variable values for initial modulus due to poor network omers produced clear hydrogels, when hydrated, in all pro- stucture.Hydrogels containing the cis-cyclic monomers all had portions studied, whereas the cis-dipivalate 3 gels were significantly higher tensile strengths and Young’s moduli than heterogeneous due to the limited miscibility of this monomer poly(HEMA). The gel containing the highest concentration of with HEMA. Hydrogels derived from the trans-diacetate 4 or dipivalate derivative was highly rigid and tended to fracture the trans-dimorpholinocarboxy 5 derivatives showed a slight when cut.It was practically impossible to obtain an intact increase in EWC with decrease in HEMA content. Whilst this sample of this material. The relatively low tensile strength of behaviour might be expected for the hydrogels containing the this sample was probably due to fracturing prior to testing.The tensile strengths and initial moduli of the hydrophilic trans-dimorpholinocarboxy derivative, the reverse poly(HEMA–CM) hydrogels containing cis-DHCD deriva- would be expected for the hydrophobic trans-diacetate derivatives were compared with those reported for the corresponding tive. The xerogels of the HEMA–CM hydrogels which con- HEMA–styrene and HEMA–methacrylate hydrogels.3 The tained derivatives of trans-DHCD were homogeneous; when dipivalate systems with the two highest dipivalate contents swollen they became heterogeneous and increased in fragility have been omitted from the comparisons of initial moduli with decrease in HEMA content.These observations are for clarity. consistent with a poor network stucture and increased oligomer For a given composition the poly(HEMA–CM) systems had formation resulting in a higher EWC for the hydrogel.It was higher values of tensile strength and initial moduli than did observed previously10 that the trans-diacetate derivative homothe the comparable methyl methacrylate systems, (Fig. 3). This polymerised poorly, giving low conversion of monomer to was also true for the styrene analogues at higher concentrations polymer and low molecular weight polymer.No explanation of HEMA, but the situation was reversed as the concentration has yet been advanced to explain the diVerence in the polymof HEMA decreased. This abrupt change in the properties of erisation behaviour of cis- and trans-DHCD derivatives.The the styrene system can been accounted for by the loss of availability of the p-electrons could be an important factor in plasticising freezing water, at higher concentrations of styrene.1 determining the polymerisation properties of the cis- and trans- Comparison of tensile strength and initial moduli with EWC monomers. It seems probable that the pendant groups shield show that the poly(HEMA–CM) systems possesed superior both faces of the trans-DHCD derivative whereas for the cismechanical properties to their styrene and methyl methacryal- DHCD derivatives only one face of the ring is shielded.In ate analogues (Fig. 4). It appears that the cis-DHCD deriva- consequence the cis-compounds undergo polymerisation tives are better comonomers for poly(HEMA) hydrogels than more readily.J. Mater. Chem., 1998, 8(1), 21–24 23polymer chains is probably responsible for the greater tensile strengths and initial moduli exhibited by these HEMA–CM hydrogels compared with their HEMA–methyl methacrylate and HEMA–styrene analogues. Conclusions Poly(HEMA–cis-DHCD) hydrogel copolymers showed higher tensile strengths and initial moduli than their styrene and methyl methacrylate analogues, at similar EWC and in most cases at similar composition. Thus cis-DHCD derivatives appear to be superior to styrene and methacrylate as comonomers in poly(HEMA) hydrogel systems.It was found that the trans-derivatives of DHCD polymerise with diYculty, giving poly(HEMA–trans-DHCD) copolymers with a poor network structure and lower values of tensile strength and initial modulus than poly(HEMA) itself.References 1 P. H. Corkhill, A. S. Trevett and B. J. Tighe, Proc. Instn. Mech. Engnrs., 1990, 204, 147. 2 A. Barnes, P. H. Corkhill and B. J. Tighe, Polymer, 1988, 29, 2191. 3 P. H. Corkhill, Novel Hydrogel Polymers, Ph.D. Thesis, Aston University, 1988, p. 103. 4 V. Kudela, Hydrogels, in Encyclopedia of Polymer Science and Engineering, ed.J. I. Kroschwitz, Wiley, New York, 1987, vol. 7, p. 783. 5 B. D. Ratner and A. S. HoVmann, Synthetic Hydrogels for Biomedical Applications, in Hydrogels for Medical and Related Applications, ACS Symp. Ser., ed. J. D. Andrade, ACS,Washington Fig. 4 Comparison of the mechanical properties of poly(HEMA–CM) DC, 1976, vol. 31, p. 1. hydrogels with those of poly(methyl methacrylate) and polystyrene at 6 P. H. Corkhill and B. J. Tighe, J.Mater. Chem., 1992, 2, 491. equivalent EWC: (+) 1; (&) 2; (+) 3; (6) styrene; (%) methyl 7 P. H. Corkhill and B. J. Tighe, Polymer, 1990, 31, 1526. methacrylate 8 M. B. Huglin and M. M. A.-M. Rehab, Polymer, 1987, 28, 2200. 9 D. G. H. Ballard, A. Courtis, I. M. Shirley and S. C. Taylor, Macromolecules, 1988, 21, 294. styrene or methyl methacrylate as they strengthen the polymer 10 D. R. McKean and J. K. Stille,Macromolecules, 1987, 20, 1787. but degrade the hydrophilicity to a lesser extent. 11 K. L. Platt and F. Oesch, Synthesis, 1977, 449. It would be expected that the poly(HEMA–DHCD) systems 12 P. H. Corkhill, A. M. Jolly, C. O. Ng and B. J. Tighe, Polymer, would show enhanced main chain stiVness due to the presence 1987, 28, 1758. both of the cyclohexene rings and interactions of the pendant Paper 7/03806B; Received 2nd June, 1997 groups on these rings. This decrease in the mobility of the 24 J. Mater. Chem., 1998, 8(1), 21–24

ISSN:0959-9428    

DOI:10.1039/a703806b

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

J O U R N A L O F C H E M I S T R Y Materials Synthesis and characterisation of telechelic regioregular head-to-tail poly(3-alkylthiophenes) Ahmed Iraqi* and George W. Barker School of Physics and Chemistry, University of L ancaster, L ancaster, UK L A1 4YA An investigation into the Stille cross-coupling reaction as an alternative synthetic route to regioregular head-to-tail poly(3-alkylthiophenes) is presented. 2-Iodo-3-alkyl-5-tri-n-butylstannylthiophene derivatives were made, isolated and characterised and subsequently used in Pd0 catalysed cross-coupling reactions in diVerent solvents. Short and long chain regioregular poly(3-hexylthiophenes) with over 96% head-to-tail coupling ratios between adjacent thiophene rings were obtained. The materials were analysed by gel permeation chromatography and NMR spectroscopy.Polymers with varying degrees of polymerisation were obtained depending on the conditions used. End group analysis has shown that polymers, functionalised with both iodo and tri-nbutyltin end groups can be obtained under certain reaction conditions. Preliminary homocoupling reactions were also conducted. There has been a great deal of interest in the study of organic materials (2,5-dibromo-3-alkylthiophene) since these should lead to random couplings of thiophene rings and coupling conjugated polymers over the last few years in view of their potential applications in a number of areas of molecular defects.In a recent communication,4 McCullough et al. reported the preparation of a regioregular oxazoline electronics.The structure of these rigid rod macromolecules plays an important role in determining their physical functionalised polythiophene using a modified Stille coupling reaction involving 2-bromo-3-[2-4,5-dihydro-(4,4-dimethylox- properties.1 Among these materials, polythiophenes have attracted sig- azol-2-yl )ethyl]-5-trimethylstannylthiophene as intermediate. nificant interest in view of their environmental stability in both neutral and oxidised states.The physical properties of poly(3- alkylthiophenes) are dramatically altered by changing the regioregularity of polymer chains. On increasing the ratio of head-to-tail coupling of adjacent monomers on poly(3-alkylthiophenes), a decrease of steric hindrance between alkyl S R Br X 1 X = MgBr 2 X = ZnBr substituents on adjacent rings is observed leading to an increase in the linearity of polymer chains and hence greater conjugation In this paper, we present our findings on the potential use of along the polymer backbone.Lower band gap polymers are 2-iodo-3-alkyl-5-tri-n-butylstannylthiophene derivatives as thus obtained. This was established by McCullough et al.2 and intermediates in Stille type cross-coupling reactions5 for the Riecke et al.3 who demonstrated the eVect of regiochemical preparation of telechelic regioregular head-to-tail poly(3- control on the properties of such materials.Enhanced conduc- alkylthiophenes). tivities (one to two orders of magnitude greater) are observed from doped head-to-tail regioregular poly(3-alkylthiophenes) Results and Discussion when compared with conductivity values of regioirregular polymers.The use of 2-iodo-3-alkyl-5-tri-n-butylstannylthiophene deriva- Synthetic strategies for the preparation of regioregular head- tives as intermediate monomers in palladium(0) catalysed to-tail poly(3-alkylthiophenes) so far reported in the literature cross-coupling reactions for the production of regioregular involve nickel catalysed Grignard cross-coupling reactions head-to-tail poly(3-alkylthiophenes) presents several advanusing 2-bromo-3-alkyl-5-bromomagnesiothiophene 12 and 2- tages.A major advantage in using these intermediates is their bromo-3-alkyl-5-bromozinciothiophene 23 derivatives. In such kinetic, air and moisture stability. This permits their separation reactions the intermediate species are made in situ and sub- and purification from undesired residual starting materials.sequently polymerised. Good head-to-tail ratios are obtained Another advantage in using these derivatives is the possibility from such syntheses providing that a strict kinetic control is of preparing telechelic polymer chain fragments that could be observed since at high temperatures in the presence of nickel involved in the synthesis of 3-functionalised polythiophene catalysts intermediates 1 and 2 can isomerise to 2-bromomag- block copolymers with diVerent substituents on respective nesio-3-alkyl-5-bromothiophene and 2-bromozincio-3-alkyl-5- blocks since both iodo and tri-n-butyltin groups can, in prinbromothiophene, respectively, by a series of transmetallation– ciple, be retained at the chain ends of polymers after reaction reductive eliminations followed by oxidative addition.work-up and polymer separation, as opposed to synthetic Reproducibility in molecular mass distributions of the resulting schemes involving the use of intermediates 1 or 2, and where polymers in the nickel catalysed cross-coupling polymerisation the Grignard ends of polymer chains are hydrolysed at the of intermediates 1 is linked to their freedom from starting work-up stage.materials (2-bromo-3-alkylthiophene). Trace amounts of unre- 2-Iodo-3-hexyl-5-tri-n-butylstannylthiophene 3 was targeted acted 2-bromo-3-alkylthiophene monomers still present at the as a model monomer in these reactions in view of the wealth polymerisation stage should contribute to chain termination.of data on poly(3-hexylthiophenes) available in the literature. Reproducibility in the percentages of head-to-tail coupling of It was made by selective iodination of 3-hexylthiophene at the the resulting polymers in the nickel catalysed cross-coupling 2-position using iodine with mercury(II) oxide in benzene, polymerisation of 2 is also linked to their freedom from starting followed by selective lithiation of the resulting 2-iodo-3-hexylthiophene at the 5-position with lithium diisopropylamide (LDA) and reaction with tri-n-butyltin chloride at low tempera- * E-mail: a.iraqi@lancaster.ac.uk J.Mater. Chem., 1998, 8(1), 25–29 25Table 2 Comparison of molecular mass distributions of poly(3-hexylthiophenes) after soxhlet extraction with methanol and then hexane.UV–VIS data and times of polymerisation in diVerent solvents. lmax/nm entry solvent MW MN polydispersity (CHCl3) t/h S R S R I S R I Bu3Sn i ii,iii 3 Scheme 1 Reagents and conditions: i, I2, HgO; ii, LDA; iii, Bu3SnCl 1 toluene 19 732 16 118 1.22 449 18 2 CH2 ClCH2Cl 16 625 11 517 1.44 449 18 3 CH2 ClCH2Cl 14 484 10 889 1.33 449 66 precipitation in methanol (lower molecular mass fractions).Upon Soxhlet extraction of the high molecular mass fractions with methanol for 24 h and then a further Soxhlet extraction S R I Bu3Sn S R Bu3Sn S S I R R Pd(PPh3)4 n R = C6H13 with hexane for 24 h, materials with higher average molecular Scheme 2 masses and lower polydispersities are obtained (Table 2). Experiments in 1,2-dichloroethane for longer periods of time tures.The resulting materials from these reactions are always (66 h) where the resulting polymers are subjected to the same found to contain trace amounts of the staring material (2- work up procedure are also presented in Table 2. iodo-3-hexylthiophene). These are removed by evaporation Results in Table 1 show in the first instance that the rate of in vacuo, to leave the desired product in quantitative yields these polymerisation reactions is slower than that of polymer- (see Scheme 1).isation reactions involving 2-bromo-3-alkyl-5-bromomag- Monomer 3 is kinetically stable and can be handled in an nesiothiophene or 2-bromo-3-alkyl-5-bromozinciothiophene air and water atmosphere without degradation.NMR analysis intermediates, where reactions occur at room temperature or of the product shows exclusive functionalisation at the 5- lower and where higher molecular mass materials are position on the thiophene ring. 1H NMR analysis also shows obtained.2,3 This is presumably due to a lower rate of halogen 3J coupling of the proton at the 4-position on thiophene rings exchange between tin and palladium as compared to that with naturally abundant tin isotopes with nuclear spins (117Sn between magnesium and nickel, zinc and nickel or zinc and and 119Sn) [Fig. 1(a)]. palladium. The new 2-iodo-3-hexyl-5-tri-n-butylstannylthiophene Results in Table 1 also reveal that higher molecular mass derivative is then used as monomer in palladium(0) crossmaterials are obtained from reactions carried out in toluene coupling catalysed reactions leading to highly regioregular and 1,2-dichloroethane, when compared to materials obtained poly(3-hexylthiophenes) with head-to-tail coupling ratios from reactions carried out in THF which are mainly composed exceeding 96%. (see Scheme 2).The resulting polymers are of long regioregular oligomeric chains. readily soluble in methylene chloride, chloroform and THF.Table 1 also shows results of polymerisations carried out with monomer 3 when contaminated with trace amounts of Solvents eVects on the reaction starting material (2-iodo-3-hexylthiophene) (entry 6). Lower Polymerisation reactions were carried out in diVerent solvents molecular mass polymers are obtained in lower yields as under the same conditions for set periods of time (18 h).Cross- compared to experiments run with purified monomer 3 under coupling reactions proceed in all three solvents used, tetra- the same conditions This is explained by chain termination hydrofuran, toluene and 1,2-dichloroethane at reflux tempera- reactions caused by the presence of 2-iodo-3-hexylthiophene. tures. It is found that in all solvents used, selective head-to- NMR analysis indicates that the head-to-tail regioregularity tail regiochemical coupling of monomers is observed. However, of these polymers is, however, still maintained.polymers with diVerent molecular masses are obtained in each In experiments run in toluene, the solubility of the growing case. GPC analysis results from these experiments are shown polymer chains plays an important role on the extent of in Table 1.polymerisation and could be a limiting factor on chain growth. The work up procedure of the polymers obtained from Indeed, in polymerisations carried out in toluene, precipitation reactions carried out in toluene and 1,2-dichloroethane, of part of the polymer is observed before the end of the presented in Table 1, involved precipitation of polymers in experiment. 1,2-Dichloroethane was chosen as an alternative methanol at the end of reactions, followed by washing with solvent in view of its enhanced solvation power as compared methanol several times at room temperature. The materials to toluene. However, as shown in Table 2, a maximal degree obtained are then extracted with hexane at room temperature of polymerisation of around 67 is not surpassed even after allowing separation of the polymer into two fractions; the extended reaction periods (66 h). hexane insoluble fractions (which have the higher molecular masses) and those separated from the hexane extracts by Spectroscopic studies NMR studies in chloroform on these materials explain a Table 1 Comparison of molecular mass distributions of poly(3-hexylnumber of these findings. 1H NMR spectra of monomer 3 and thiophenes), UV–VIS data and yields of polymerisation in diVerent solvents for reactions run for 18 h polymers with varying degrees of polymerisation from reactions in toluene are shown in Fig. 1. lmax/nm yield Fig. 1(d) shows the spectrum of the polymer with the highest entry solvent MW MN polydispersity (CHCl3) (%) average molecular mass in these experiments (DP=97, entry 1, Table 2).It reveals a single chemical environment for 1 toluene 9992 7366 1.36 447 50 the aromatic protons of the thiophene rings on the backbone 2 2659b 2239 1.19 430 20 3 CH2 ClCH2Cl 13 711 7977 1.72 448 54 of the polymer (dH 6.97) and essentially a single chemical 4 4198b 3538 1.48 434 17 environment for the methylene groups attached directly to the 5 THF 2315 1409 1.64 391 <10 thiophene rings (dH 2.80), with a small trace (<4%) of another 6 toluenea 7693 4220 1.82 439 40 triplet in a diVerent environment at dH 2.58, suggesting a homogeneous structure for the polymer and >96% head-to- aReaction carried out with monomer 3 contaminated with trace tail coupling along the backbone.It is worth noting at this amounts of 2-iodo-3-hexylthiophene. bFractions that are soluble in hexane at room temperature, separated by precipitation with methanol. stage that no signals from tri-n-butyltin end groups are 26 J. Mater. Chem., 1998, 8(1), 25–29Fig. 2 13C NMR spectra of (a) monomer 3, (b) poly(3) (degree of polymerisation ca. 13, entry 2, Table 1) and (c) poly(3) (degree of polymerisation ca. 44, entry 1, Table 1) Fig. 1 1H NMR spectra of (a) monomer 3, (b) poly(3) (degree of polymerisation ca. 13, entry 2, Table 1), (c) poly(3) (degree of polymerisation ca. 44, entry 1, Table 1) and (d) poly(3) (degree of polymerisation same reaction mixture. The signals most probably originate ca. 97, entry 1, Table 2) from methylene groups directly attached to thiophene rings close to the chain ends. 13C NMR spectra of monomer 3 as well as those of polymers observed, suggesting that they are hydrolysed in the work up to isolate the polymer. Fig. 1(c) shows the spectrum of the in entries 1 and 2 in Table 1 are shown in Fig. 2. Fig. 2(c) shows the spectrum of the polymer in entry 1, Table 1 (DP= parent polymer before Soxhlet extraction with methanol and hexane, which has a lower average molecular mass (DP=44, 44).It essentially reveals four carbon environments for the thiophene rings with additional background signals originating entry 1, Table 1). Similar features to those of the higher molecular mass polymer are observed. However, in addition from carbon on thiophene rings at the chain ends on shorter polymeric chains suggesting a homogeneous structure of the to peaks discussed above, there are additional background signals in the aromatic region originating from hydrogens on polymer and ca.>96% selective head-to-tail coupling along the backbone. Signals from butyl groups on tin residues are thiophene rings at the chain ends on shorter polymeric chains. It is also possible to distinguish signals from tri-n-butyltin end distinctly observed at dC 10.86, 13.54, 27.19 and 28.90.Fig. 2(b) shows the spectrum of a lower molecular mass polymer groups from the 1H spectra with signals occurring at dH 1.12 and 1.58. These are attributed respectively to hydrogens on a- (oligomer) (DP=13, entry 2, Table 1). The peak pattern in the aromatic region shows a multitude of peaks in view of the carbons and c-carbons on butyl groups of tin residues, while hydrogens from b-carbons are masked under six methylene shorter length of thiophene chains.Both intensity and multiplicity of signals attributed to tri-n-butyltin end groups (centred hydrogens from hexyl substituents at dH 1.38 and methyl hydrogens from the tin residues also masked under the methyl at the same chemical shifts) are increased, reflecting a wide distribution of oligomers of diVerent lengths.groups of hexyl substituents at dH 0.92. This shows that tri-nbutyltin end groups are preserved on polymer chains and are NMR analysis of polymers prepared in 1,2-dichloroethane also reveal that highly regioregular head-to-tail materials are only cleaved if the polymer is subjected to treatment with refluxing methanol as in the case of the previous polymer after formed.However, end group analysis of the polymers shows the absence of signals from tri-n-butyltin residues even prior Soxhlet extraction with hot methanol. Fig. 1(b) shows the spectrum of an even lower molecular mass polymer (oligomer) to Soxhlet extraction with methanol (entries 3 and 4, Table 1) suggesting their cleavage as reactions proceed.This also (DP=13, entry 2, Table 1). The intensity of the additional signals at the aromatic region beside the main peak at dH 6.97 explains the similarity in molecular mass distributions and polydispersities between materials obtained after reaction times is greater. Similarly, the intensity of signals attributed to tri-nbutyltin end groups is also greater.Polymers with low molecu- of 18 and 66 h, respectively (entries 2 and 3, Table 2). NMR analysis of materials obtained from reactions in THF, lar masses display, in addition to signals at dH 2.80, intense signals at dH 2.58. These additional signals do not correspond reveals a mixture of oligomers (DP=6, entry 5, Table 1). Signals from tri-n-butyltin groups are still apparent suggesting to coupling defects and regioirregular chains since these would be duplicated in longer polymeric chains isolated from the that the cross-coupling polymerisation proceeds at very low J.Mater. Chem., 1998, 8(1), 25–29 27Table 3 Homocoupling of telechelic polymers An important possible application of materials made in the present study, especially those functionalised with both iodo lmax b/nm and tri-n-butyltin end groups, is in the preparation of polythio- MW MN polydispersity (CHCl3) phene multi-block copolymers with diVerent substituents on alternating blocks.starting polymera 2999 2312 1.58 424 final polymer 8867 4765 1.86 439 Experimental section aPolymer obtained from reaction in THF after 66 h at reflux temperature. bUV–VIS spectra were recorded on samples of initial NMR spectra were recorded using a JEOL GSX400 specand final reaction mixtures and were found to be concordant with trometer (400 MHz, 1H and 100.5 MHz, 13C).J Values are spectra of isolated materials. in Hz. UV–VIS spectra were recorded using a Unicam 8700 spectrophotometer. IR spectra were obtained using a Nicolet rates in the solvent and pointing to a kinetic eVect rather than 205 FTIR.The gel permeation chromatography (GPC) system to the cleavage of end groups during the reaction. This is comprised a Waters 510 pump coupled via a Rheodyne injecsupported by data from reactions carried out in the same tion port to Polymer Laboratories columns. A Waters 410 solvent for longer periods of time (66 h) and where a degree diVerential refractometer was used to detect molecular mass of polymerisation of around 14 is achieved.NMR analysis fractions eluting from the columns. The data were manipulated also shows regioregular head-to-tail coupling of adjacent monusing Polymer Laboratories GPC software operating under omers in these materials, together with conservation of tin end Microsoft Windows 3.11. Calibration of the columns was groups on the polymers formed.carried out using polystyrene standards. All samples were run Absorption spectra on these polymers in chloroform, show in tetrahydrofuran (THF). The flow rate was set to 1 ml min-1 maximum absorptions at lmax between 424 nm for low molecuand toluene was added as a flow rate marker.lar mass polymers and 449 nm for higher molecular mass Toluene and benzene were distilled over sodium. THF and polymers (Tables 1, 2 and 3). The increase in lmax is shown to diethyl ether were distilled over sodium–benzophenone, 1,2- be proportional to the chain length of polymers and hence to dichloroethane was dried and distilled over CaH2. the extent of electronic delocalisation.Diisopropylamine was dried and distilled over sodium hydroxide. 3-hexylthiophene6 and Pd(PPh3)47 were made according Homocoupling of telechelic polymers to literature procedures. Tri-n-butyltin chloride was obtained Preliminary investigations on the potential use of short tele- commercially and used without further purification. chelic polymer chains in the preparation of block copolymers 2-Iodo-3-hexylthiophene were also conducted.Homocoupling of telechelic polymer chains was demonstrated upon treatment of polymers with a To a solution of 3-hexylthiophene (2 g, 11.88 mmol) in benzene low degree of polymerisation (DP~14, starting polymer, (20 cm3) at 0°C was added, in small portions, mercury(II) Table 3) with catalytic amounts of Pd(PPh3)4 in refluxing oxide (2.57 g, 11.88 mmol, yellow crystals) and iodine (3.02 g, toluene (Scheme 3). 11.90 mmol) over 40 min. The mixture was stirred at room Polymers with a higher degree of polymerisation (DP~28, temp. for 1 h, and the orange precipitate was filtered oV and final polymer, Table 3) and increased maximum absorptions washed with diethyl ether. The filtrate and washings were (lmax=439 nm) resulted.combined and washed with aqueous sodium thiosulfate and dried over magnesium sulfate. Solvent was removed by rotary Conclusions evaporation and the residue distilled in vacuo (88 °C at 0.5 mmHg) to give a colourless oil (Yield 2.00 g; 57%). dH An investigation into the use of 2-iodo-3-alkyl-5-tri-n- (CDCl3, SiMe4) 0.89 (t, J 7, 3H), 1.31 (m, 6H), 1.56 (m, 2H), butylstannylthiophene derivatives as intermediates in Pd0 Stille 2.54 (t, J 7, 2H), 6.74 (d, J 5, 1H), 7.36 (d, J 5, 1H).dC (CDCl3, cross-coupling polymerisation for the production of regio- SiMe4) 13.99 (1C), 22.49 (1C), 28.78 (1C), 29.89 (1C), 31.52 regular head-to-tail poly(3-alkylthiophenes) was undertaken (1C), 31.99 (1C), 73.86 (1C), 127.83 (1C), 130.16 (1C), 147.04 in three diVerent solvents. Reactions in THF aVorded low (1C).molecular mass materials with a degree of polymerisation up to 14 with no sign of cleavage of tin end groups. Reactions in 2-Iodo-3-hexyl-5-tri-n-butylstannylthiophene 1,2-dichloroethane led to highly regioregular (>96% head-totail ratio) polymers with a degree of polymerisation up to 70. To a solution of diisopropylamine (6.50 g, 64.21 mmol) in diethyl ether (100 cm3) was added n-butyllithium (32 cm3 of However, tri-n-butyltin end groups are cleaved during reactions in this solvent.Reactions in toluene also aVord highly regio- 1.6 M solution in hexanes, 51.2 mmol) at room temp. After 30 min stirring at room temp., the resulting LDA solution was regular head-to-tail poly(3-akylthiophenes) with >96% headto- tail coupling ratios with a degree of polymerisation up to cooled to -80 °C.A solution of 2-iodo-3-hexylthiophene (15.00 g, 51.00 mmol) in diethyl ether (100 cm3) was cooled to 97. Careful work-up of these materials in methanol and hexane at room temperature allows separation of polymers with -80 °C and added at the same temperature to the LDA solution and the temperature allowed to rise slowly to -40 °C, preservation of both iodo and tri-n-butyltin end groups.Treatment of these polymers with methanol at elevated whereupon it was stirred for 1 h. The solution was then cooled to -80 °C and tri-n-butyltin chloride (16.6 g, 51.00 mmol) temperatures leads to the cleavage of tin end groups. Preliminary homocoupling reactions of telechelic polymer added. The mixture was allowed to warm to room temp.overnight and was then poured onto ice, The products were chains were successful and yielded polymers with higher molecular masses. then extracted into diethyl ether (3×100 cm3) and washed S R Bu3Sn S S I R R n S R Bu3Sn S S I R R xn Pd(PPh3)4 toluene Scheme 3 28 J. Mater. Chem., 1998, 8(1), 25–29with water (2×100 cm3), saturated aqueous copper sulfate Homocoupling of telechelic polymers (100 cm3), saturated sodium hydrogen carbonate (100 cm3) To 0.22 g (9.25×10-2 mmol) of telechelic poly(3) (DP~14, and water again (200 cm3).The diethyl ether extracts were prepared as described above using THF as a solvent at reflux dried over magnesium sulfate and the solvent removed by for 66 h) in toluene (50 cm3) was added Pd(PPh3)4 (21 mg, rotary evaporation to leave an orange–red oil.NMR analysis 1.85×10-2 mmol) under inert atmosphere and the solution of the residual oil revealed trace amounts of starting material refluxed for 18 h with stirring. At the end of the reaction, the (2-iodo-3-hexylthiophene). These were removed on heating the solution was poured onto methanol (300 cm3). The polymer oil under high vacuum leaving the title product as a red oil was filtered oV and washed with methanol until the washings (Yield 25.58 g, 86%).dH (CDCl3, SiMe4) 1.15 (m, 12H), 1.34 were colourless and dried in vacuo. (Yield 0.20 g; 96%). (m, 6H), 1.58 (m, 12H), 1.82 (m, 8H), 2.82 (t, J 8, 2H), 7.04 (s, UV–VIS spectra were recorded on samples of initial and final 1H). dC (CDCl3, SiMe4) 10.79 (1C), 13.50 (1C), 13.95 (1C), reaction mixtures and were found concordant with spectra of 22.48 (1C), 27.09 (1C), 28.76 (1C), 28.84 (1C), 29.99 (1C), 31.52 isolated materials [starting polymer lmax (CHCl3)=424 nm, (1C), 31.58 (1C), 77.20 (1C), 136.10 (1C), 142.31 (1C), 147.86 final polymer lmax (CHCl3)=439 nm].(1C). Polymerisation procedure We would like to acknowledge the EPSRC for financial support of this work (studentship G.W.B) In a typical polymerisation reaction, monomer 3 (3.00 g, 5.14 mmol) was solubilised in the appropriate dry distilled solvent (50 cm3).The solution was then deoxygenated, the References catalyst (178 mg, 0.154 mmol, 3 mol%) added under inert atmosphere, and the solution refluxed for set periods of time 1 J. Roncali, Chem. Rev., 1997, 97, 173. with stirring. At the end of the reaction, the solution was 2 (a) R. D. McCullough, R. D. Lowe, M. Jayaraman and D. L. poured onto methanol (300 cm3). The polymer was filtered oV Anderson, J. Org. Chem., 1993, 58, 904; (b) R. D. McCullough and S. P. Williams, J. Am. Chem. Soc., 1993, 115, 11608 and washed with methanol until the washings were colourless. 3 (a) T.-A. Chen, X.Wu and R. D. Rieke, J. Am. Chem. Soc., 1995, 117, Extraction of the polymer with hexane at room temp. 233. (b) T.-A. Chen and R. D. Rieke, J. Am. Chem. Soc., 1992, allowed the separation of the polymer into two fractions; the 114, 10087. hexane insoluble fractions (which have the higher molecular 4 R. D. McCullough, P. C. Ewbank and R. S. Loewe, J. Am. Chem. masses) and those separated from the hexane extracts by Soc., 1997, 119, 633. precipitation in methanol. 5 J. K. Stille, Angew. Chem., Int. Ed. Engl., 1986, 25, 508. 6 C. Van Pham, H. B. Mark and H. Zimmer, Synth. Commun., 1986, For experiments reported in Table 2, separation of the 16, 689. polymers was accomplished through Soxhlet extraction with 7 D. R. Coulson, Inorg. Synth., 1990, 28, 107. methanol for 24 h, then hexane for 24 h, of either crude polymers or those with a high molecular mass described above, and aVorded higher molecular mass materials. Paper 7/06583C; Received 9th September, 1997 J. Mater. Chem., 1998, 8(1), 25–29 29

ISSN:0959-9428    

DOI:10.1039/a706583c

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

J O U R N A L O F C H E M I S T R Y Materials Covalent binding of redox active centres to preformed regioregular polythiophenes Ahmed Iraqi,† Joe A. Crayston and John C.Walton School of Chemistry, University of St. Andrews, St. Andrews, Fife, UK KY16 9ST The synthesis of regioregular head-to-tail poly[3-(6-bromohexyl)thiophene] is reported, together with its reaction with 2- carboxyanthraquinone (Anth) to give an example of a regioregular polythiophene containing pendant functional groups (87% loading).NMR data on the two soluble polymers are reported together with preliminary studies of some of their physical properties. Cyclic voltammetric studies of anthraquinone polymer coated electrodes show that the observed response is coverage dependent: thin films display four redox couples due to the Anth0/-/2- processes and the p- and n-doping of the conjugated thiophene backbone.Thick films are rectifying in the sense that reduction of the Anth groups is inhibited on the negative sweep. Spectroelectrochemical studies confirm the nature of the anodic p-doping process (the film turns red to nearly colourless) and show characteristic changes on reduction (red to black).There is considerable interest in the binding of redox active (6-bromohexyl)thiophene] (precursor polymer), together with its further reaction with 2-carboxyanthraquinone and prelimi- and mesogenic groups to organic conjugated polymers for their potential electrocatalytic, photochemical and electronic nary studies on the physical properties of the two polymers.We shall show that this functionalisation process is remarkably applications.1–3 Polythiophene is emerging as a popular choice for the conjugated polymer backbone compared to, say, poly- eYcient. We shall also compare our results with our previously reported regio-random polymer [poly(4)]8b based on the elec- pyrrole, not only due to the relative ease of synthesis of thiophene derivatives but also the relative stability of the trochemical random polymerisation of monomer 4.Since we completed this work we became aware of work by Bauerle polymer towards over-oxidation.4 Recent examples of redox active groups attached to polythiophene include ferrocenes,5,6 et al. which describes the post-functionalisation of an ester substituted polymer; in this case it is important to note that viologens,6 fullerene,7 quinones,8 nitroxide9 and nitro groups10 and tetrathiafulvalene (TTF).11 Other types of functional group the reactions were carried out with the polymer remaining on the electrode surface.23 have included metal complexes12 or complexing agents such as crown ethers13 or calixarenes.14 Cyano groups have a marked influence on the redox properties of thiophene,15 while photochromic,16 polymerisable,17 or molecular recognition18 sites confer additional properties or reactivity to the final polymer.Most of the reported redox functionalised conjugated polymers were formed by direct chemical or electrochemical polym- O O O O S 4 erisation of the corresponding functionalised monomers. The first major problem with the monomer functionalization approach used in most of the examples cited above is that it Results and Discussion often led to insoluble materials that were diYcult to characterise fully by NMR spectroscopy and GPC.Secondly, one of Poly[3-(6-bromohexyl)thiophene] was the novel key precurthe main factors aVecting the physical properties of conjugated sor polymer to be made using the regioselective synthetic polymers was shown to be the regioregularity and homogeneity strategy.The synthesis begins with selective bromination of 3- of their backbone structures. This was clearly demonstrated in (6-bromohexyl)thiophene 1 at the 2-position using N-bromothe case of alkyl and alkoxy substituted polythiophenes19 succinimide in DMF. The brominated monomer 2 was then where structurally homogeneous materials displayed lower selectively lithiated at the 5-position, followed by reaction with band gaps and higher intrinsic conductivities than the corre- magnesium dibromide–diethyl ether.Addition of the nickel sponding materials with random structures. catalyst aVorded poly(1) in 60% yield (see Scheme 1). We decided to investigate a new route to soluble and easily Poly(1) was isolated as a deep red powder readily soluble processible functionalised polymers relying on the post-func- in methylene chloride and chloroform.Gel permeation chromationalisation in solution of soluble regioregular preformed tography (GPC) in chloroform revealed an average molecular polymer precursors. Recently, we discovered that the Grignard mass of 12 300 with a polydispersity of 1.5.These data are coupling reaction leading to regioregular poly(3-alkylthio- comparable with those reported for regioregular poly(3-alkylphenes) could be extended to a number of substituted polythi- thiophenes) prepared by a similar route. NMR studies of ophenes.20,21 As a result of this work poly(3-bromo- poly(1) in chloroform revealed a single chemical environment alkylthiophenes) were chosen as attractive precursors for for the aromatic protons of the thiophene rings on the backfunctionalised polymers.The 3-(v-bromoalkyl)thiophene bone of the polymer (dH 6.98) and essentially a single chemical monomers were already known to be useful intermediates for environment for the methylene groups attached directly to the the synthesis of 3-functionalised thiophene monomers.22 In this thiophene rings (dH 2.80), with a small trace (6%) of another contribution we report the regioregular synthesis of poly[3- triplet in a diVerent environment at dH 2.60.The 13C NMR spectrum revealed only four carbon environments for the thiophene rings suggesting a homogeneous structure of the † Present address: School of Physics and Chemistry, University of Lancaster, Lancaster, UK LA1 4YA.polymer and ca. 94% selective head-to-tail coupling along the J. Mater. Chem., 1998, 8(1), 31–36 31directly attached to the backbone of the polymer (dH 2.80 signal, 3H, CH2-thienyl) to one on the anthraquinone (dH 8.84, 1H, aromatic proton adjacent to ester linkage). UV–VIS spectral analysis on polymer 3 in solution in CHCl3 showed an absorption peak at lmax 445 nm with a band edge (onset of absorption) at 576 nm.For films cast from CH2Cl2 an absorption at lmax 530 nm with a band edge at 700 nm (band gap 1.77 eV) was observed. This energy is comparable to that of poly(dodecylthiophene) in the solid state and suggests a lamellar structure in the film.19a There were, however, no long wavelength shoulders on this band (‘fine structure’), which, if present, are thought to be indicative of long range order in the sample.Electrochemical properties of polymer 3 Polymer-coated electrodes were prepared by droplet evaporation of a CH2Cl2 or CHCl3 solution of polymer 3. We first describe the cyclic voltammetric behaviour of thin polymer films with C10-8 mol cm-2 corresponding to an estimated thickness of 50 nm, assuming unit density of the film.These films showed not only the expected thiophene-like response at positive potentials (Fig. 1), but also the two cathodic reductions associated with the pendant anthraquinone groups. The thiophene oxidation wave was somewhat more symmetrical (Epa+0.716 vs. SCE, Epc+0.656 V), and the peak separation (DEp 60 mV) somewhat smaller than usual for thiophene films (the ideal surface wave should have zero peak separation). As expected, the oxidation of the polythiophene backbone occurs at lower potentials than non-regioregular polymers.4 But there was no evidence for two clearly separated oxidation processes as observed in previous work on regioregular polythiophenes. 4,19b The first reduction was signalled by an unsymmetrical set of peaks with a broad cathodic wave (Epc-1.028 V vs.SCE) and a much sharper anodic wave (Epa-0.82 V). This S Br S Br Br S Br Br BrMg S S S Br Br Br S S S O O O C C C Anth Anth Anth O O O n n 1 2 poly(1) poly(3) i ii–iv v vi Scheme 1 Reagents and conditions: i, NBS, DMF; ii, LDA, THF, -60 °C, 90 min; iii, MgBr2·OEt2, -60 °C, 20 min; iv, -60�-5 °C; v, 0.5 mol% [NiCl2(dppp)],-80�20 °C, 15 h; vi, 2-carboxhraquinone, DBU, THF, 80 °C, 15 h backbone.UV–VIS spectral analysis on the regioregular polymer in solution in CHCl3 showed an absorption band (lmax 445 nm) with a band edge (onset of absorption) at 560 nm. For films cast from CH2Cl2 an absorption band was observed at lmax 511 nm with a band edge at 670 nm (band gap 1.8 eV).This band gap is comparable with those reported for alkylsubstituted polythiophenes made by the same method.19 Poly(1) was further reacted with 2-carboxyanthraquinone in the presence of 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) to yield the anthraquinone functionalised polymer 3 as a deep red powder isolated in 56% yield. Polymer 3 was also readily soluble in methylene chloride and chloroform.NMR studies on polymer 3 in chloroform confirmed that 87±2% of the Fig. 1 Cyclic voltammetry of a poly(3) coated glassy carbon electrode total number of thiophene chains were grafted to an anthraqui- (area 0.071 cm2) in 0.1 mol dm-3 Bu4NPF6–acetonitrile. The coverage none group. The extent of anthraquinone incorporation was is approximately 6.9×10-9 mol cm-2 of Anth sites.Scan rates: (a) 100 and (b) 10 mVs-1. determined by comparing integral heights of the hydrogens 32 J. Mater. Chem., 1998, 8(1), 31–36peak was assigned to the formation of the semiquinone radical anion (AnthV-) from the anthraquinone (Anth): eqn. (1). Anth+e-�AnthV- (1) OMe O S n 5 This wave considerably sharpened at slower scan rates As we mentioned earlier, the grafting of the anthraquinone [Fig. 1(b), 10 mV s-1], behaviour typical of sudden changes onto the polymer occurs on 87% of possible sites. For these in the solvation of redox polymers such as polyvinylferrocences. thin films the electroactivity approached 100% based on the At slow scan rates the film will have had time to reach its amount of polymer applied to the electrode during droplet compact, non-solvated, neutral state at potentials between evaporation.We may then use the ratio of the observed peak -0.8 and +0.4 V. Reduction of the quinone sites (via the areas for quinone and thiophene redox processes to calculate electronic conducting backbone as there is always considerable the number of electrons removed per thiophene ring (equivalent residual electronic conductivity in polythiophene even in its to the ‘doping level’). This calculation yields a redox level of neutral state) demands rapid ingress of cations and associated 0.20 to 0.25 electrons per ring (1 electron per 4–5 thiophene solvent in order to achieve neutrality.rings). This result is entirely consistent with accepted doping The second reduction wave will not be subject to these levels for polythiophenes.4 eVects, and indeed was more symmetrical (Epc -1.408, Epa -1.344 V) with a peak area similar to the first cathodic peak Electrochemistry of thick polymer films area, consistent with formation of the dianion: eqn.(2). When polymer film coverages greater than about AnthV-+e-�Anth2- (2) 50 nmol cm-2 are used (ca. 50 nm thickness) a vastly diVerent We should point out that the regioregular polymer was redox response is observed (Fig. 3). In this case all cathodic found to be much more stable to redox cycling than the current is postponed until the onset of n-type doping near random polymer.8b Thus after 10 min of cycling at 20 mV s-1 -1.8 V. This means that before the first scan the film is not over both the thiophene backbone and the quinone waves, suYciently solvent swollen (and therefore permeable to ions) poly(4) had lost over 50% of the faradaic charge associated to allow charge transfer to the quinones.Subsequent scans with these redox processes. Under the same conditions polymer and lower scan rates [Fig. 3(b)] have the eVect of allowing the 3 had lost less than 2% of its charge.Moreover, polymer 3 polymer to equilibrate with the solution. Similar unusual eVects seemed to be more stable after the deliberate addition of have been reported for other polymers (often described as the oxygen or water (as indicated by the unchanged CV response). first scan or ‘break-in’ eVect). The break-in eVect is particularly In the case of water addition to the solvent, the quinone waves noticeable during the n-doping process; this is thought to be began to merge (at 50% v/v water–acetonitrile) into a single due to the diYculty of cation injection, particularly when the wave centred at about -0.80 V (Fig. 2). cation is a tetraalkylammonium ion. Less significant break-in We also noted in Fig. 1 a further reversible reduction at (Epc eVects were for previously described polythiophenes or poly- -1.824, Epa -1.792 V).This peak has a similar height to that pyrroles containing quinone groups.8 The reason for this may of the thiophene oxidation and has a comparable position to be that the regioregular polymer begins the cycle in its neutral, that of similar polythiophenes such as regioregular 5 which has Epc -1.95 V for n-doping.20 We therefore attribute this reduction to the reversible n-type doping of the polymer.Fig. 3 Cyclic voltammetry of a thick film of poly(3) deposited on a Fig. 2 (a) Cyclic voltammetry (100 mV s-1) of a poly(3) coated glassy glassy carbon electrode (area 0.071 cm2) in 0.1 mol dm-3 Bu4NPF6–acetonitrile. The coverage is approximately 3.5×10-8 carbon electrode (area 0.071 cm2; coverage ca. 6.9×10-9 mol cm-2 of Anth sites) in 0.1 mol dm-3 Bu4NPF6–acetonitrile (i) before and mol cm-2 of Anth sites. (a) 100 mVs-1; (i) 1st scan, (ii) 2nd scan. (b) 10 mVs-1. (ii) after 1% v/v water addition. (b) After 50% v/v water addition. J. Mater. Chem., 1998, 8(1), 31–36 33insulating state whereas the electropolymerised polymers still was observed, but at +0.8 V a sudden bleaching of colour occurred with a significant decrease in absorbance at this retain some residual conductivity as they are prepared in the fully conducting (p-doped) state.Our own studies of the wavelength. At the same time a new absorption peak at lmax 810 nm started to develop. The sudden approach to this state electropolymerisation of 4 to give regiorandom polythiophene with pendant anthraquinone groups8b showed both redox probably reflects the ‘break-in’ eVect mentioned earlier.Visually, the electrode colour faded from deep red at 0 V, to couples clearly, although the wave due to reaction (2) was diminished. However, the polymerisation was not very eYcient colourless at +0.8 V, finally reaching a light yellow shade at +1.0 to +1.3 V. These transformations are associated with and only rather low polymer film coverages were achievable.In summary, for the thick films the polymer backbone must the introduction of bipolaron sub-gap states into the polymer backbone on oxidation.24 The transition to a transparent undergo redox processes in order to cause the film to swell with ions and solvents suYciently to allow electron transfer to conducting material appears to be characteristic of these regioregular polymers as we ourselves have noted elsewhere.20 the quinone groups.The behaviour for thick films mirrors the fast scan behaviour for thin films. The insertion of ions and Other authors have noticed that further oxidation increases the amount of absorption at longer wavelengths24 and there is solvent into the film during the quionone redox processes is a much slower process than for the polymer processes because evidence for such absorption at >1000 nm in our spectra. The radical anion, AnthV-, is expected to be green and the the structure of the thiophene polymer film is more compact in its neutral state.doubly charged anion, Anth2-, is blue. On reduction, provided that the cell was thoroughly degassed, it was possible to see a reversible darkening of the red neutral polymer film into a Spectroelectrochemistry black film at potentials more negative than where the first The spectroelectrochemistry and electrochromic properties of quinone wave appears (ca.-1.0 V). The in situ spectroelectropolymer 3 were investigated after it was cast onto an chemistry revealed that the initial reduction (ca.-1.1 V) led indium–tin oxide (ITO) coated glass electrode from a solutionincreases in the bands at 375 and 1090 nm (Fig. 5). Further of the polymer in methylene chloride. It was possible to reduction at -1.5 V causes the appearance of a shoulder at measure the change in absorption as a function of the potential ca. 650 nm on the main polymer band and a simultaneous applied at the electrode (Fig. 4). At 0 V vs. Ag/Ag+ reference increase in absorption between 700 and 1000 nm due to broad electrode the polymer p–p* absorption dominates at lmax 530 nm. On increasing the potential to +0.6 V no colour change Fig. 5 UV–VIS spectroelectrochemistry of the reduction of anthraqui- Fig. 4 UV–VIS spectroelectrochemistry of the oxidation of anthraqui- none thiophene polymer 3 coated onto an ITO electrode (3.8 nmol cm2) in 0.1 M Bu4NPF6 in degassed acetonitrile.Successive none thiophene poly(3) coated onto an ITO electrode (3.8 nmol cm-2) in degassed 0.1 M Bu4NBF4–acetonitrile at the following potentials: spectra were recorded at (a) -1.1, (b) -1.3, (c) -1.4 and (d) -1.5 V vs. Ag/AgCl. (a) +0.6, (b) +0.8, (c) +1.0, (d) +1.1 and (e) +1.3 V vs. Ag/Ag+ 34 J.Mater. Chem., 1998, 8(1), 31–36bands centred at ca. 800 and 920 nm. These changes appear Gemini-200 spectrometer (200 MHz, 1H and 50 MHz, 13C). J Values are in Hz. UV–VIS spectra were recorded using a to occur near the potentials expected for AnthV- and Anth2- formation and indeed we would expect the spectra to indicate Perkin-Elmer Lambda 5 or Lambda 14 spectrophotometer. A digital temperature controller was included in the spectropho- the build up of AnthV-.The spectral changes do not seem to correspond precisely to previous studies such as the polypyr- tometer, allowing the temperature to be increased stepwise. Before each absorption spectrum was recorded, the sample role–anthraquinone case described by Audebert et al.(using tetraethylammonium tetrafluoroborate as the electrolyte),25 or was kept at the desired temperature for 5 min. Electrochromic measurements were performed on polymer films coated from the aqueous work on the spectroelectrochemistry of anthraquinones. 26 However, our results are comparable with those on a a CH2Cl2 solution (6.3 mg polymer cm-3; or 15 mmol dm-3 with respect to each monomeric unit) onto an indium–tin fully conjugated polyanthraquinone which is reduced first to the semiquinone (bands at 380 and 546 nm), and then to the oxide (ITO) glass electrode (9×50×2 mm, resistance 20 V/%).A known volume of solution (microlitre syringe) was placed dianion (412, 658, 743 and 885 nm).27 The bands at longer wavelength are not expected for monomeric quinones and in onto the ITO electrode, covering an area of 2 cm2.The solvent was allowed to evaporate in a solvent-saturated atmosphere line with these authors we assign them to polaronic or bipolaronic species in the polymer. Spectroelectrochemical at room temp., and then the film was removed and dried in an oven at 70 °C for 30 min. The film coated electrode was studies of polythiophene and polybithiophene have shown that n-doping leads to similar changes in the UV–VIS spectrum to then immersed in 0.1 M tetra-n-butylammonium tetrafluoroborate (TBAT, Fluka, puriss.) or hexafluorophosphate those occurring on p-doping, with the loss of the main visible band and the appearance of broad absorption bands at longer (TBAPF6, Fluka, puriss.)–acetonitrile solution in a quartz UV cell equipped with a platinum wire as a counter electrode and wavelengths.28 If the polymer-coated ITO was held at potentials -1.5 V for extended periods then the polymer became a silver wire as reference electrode.The cell was degassed with a fine stream of argon supplied from a balloon; it was partially detached from the electrode. Our spectra also display considerable amounts of ‘noise’.The reason for this is unclear, particularly important to degas the cell thoroughly on the reductive cycle. Controlled potentials across the electrochemi- but may reflect the fact that we need thick films for the optical measurements and we noticed that the diYculty of cation cal UV cell were generated using a Pine Instruments RDE4 potentiostat or an EG & G PAR 273A potentiostat (both insertion into these films on reduction (noted above) caused patchy changes in the film transmission.from EG & G, Wokingham, UK). Cyclic voltammetry was performed on films cast onto 3 mm diameter glass carbon electrodes (BAS, Stockport, UK). GPC measurements were Thermochromism kindly performed by Rapra Technology Ltd., Shrewsbury.The The polymer poly(3) changed colour when heated either in GPC results were calibrated against polystyrene standards and solution or in the solid state, although the changes were more the reported molecular masses are therefore expressed as prominent in solution. In CHCl2CHCl2 solution a stepwise polystyrene equivalent molecular masses. increase in the temperature was accompanied by a shift of lmax Tetrahydrofuran and diethyl ether were distilled over sodium to higher energies, i.e.from 445 nm at 10 °C to 426 nm at benzophenone; chloroform was dried and distilled over CaCl2. 90 °C. The changes were reversible and are thought to arise Dimethylformamide (DMF) was dried and distilled over CaH2. from the twisting of the backbone of the polymer due to Diisopropylamine was dried and distilled over sodium hydroxthermal agitation with increased movement of the substituents ide. 3-(6-Bromohexyl)thiophene22 1 and [NiCl2(dppp)]30 aVecting the planarity of the backbone and hence the extent [dppp=1,3-bis(diphenylphosphino)propane] were made of conjugation.29 These changes can be described as a continu- according to literature procedures. Light petroleum (bp ous modification of the polymer chain with a progressive 40–60 °C), N-bromosuccinimide and magnesium dibromide– increase of conformational defects with increased temperature diethyl ether were obtained commercially and used without leading to a monotonic blue shift of the absorption maximum.further purification. A less prominent blue shift was observed when the measurements were done on polymer 3 in the solid state (film) over 2-Bromo-3-(6-bromohexyl )thiophene (2) the temperature range (10–90 °C), although a pronounced A solution of N-bromosuccinimide (6.48 g; 36.41 mmol) in colour change from deep red to orange was observed on DMF (45 cm3) was added dropwise under nitrogen, in the heating to 200 °C.dark and at -20 °C, to a solution of 1 (9.0 g; 36.41 mmol) in DMF (70 cm3) over a period of 4 h.The solution was left Conclusions stirring at room temp. overnight and then poured over ice (300 g) and extracted with methylene chloride (3×200 cm3). The functionalised regioregular polymers described here dis- The organic extracts were washed with water (2×150 cm3), play some, but not all, of the typical physical, spectroscopic dried over magnesium sulfate and the solvent evaporated to and electrochemical properties familiar from the parent polydryness in vacuo to yield a yellow oil.Remaining traces of alkylthiophenes. The overriding advantage that they oVer is DMF were eliminated by heating the oil (50 °C for 1 h) in high their solubility and processing possibilities. However, from the vacuum (yield 11.5 g; 97%).dH (CDCl3, SiMe4) 1.43 (m, 4H), optical and electrochemical evidence these materials do not 1.60 (m, 2H), 1.85 (m, 2H), 2.62 (t, J 7, 2H), 3.44 (t, J 7, 2H), appear to have structures as ordered as their relatives, and 6.79 (d, J 6, 1H), 7.20 (d, J 6, 1H). dC (CDCl3, SiMe4) 28.50 their unusual electrochemical behaviour shows that the solvent (1C), 28.83 (1C), 29.78 (1C), 30.10 (1C), 33.27 (1C), 34.59 (1C), content of the neutral film is low.In a future publication we 109.47 (1C), 125.86 (1C), 128.72 (1C), 142.13 (1C). shall address the properties of poly(3) coated electrodes for the electrocatalysis of O2 reduction. We are also interested in Preparation of head to tail poly (1) the rates of charge transport within conducting polymers which may be very large due to the proximity of redox groups and To a solution of diisopropylamine (2.74 g; 27.11 mmol) in THF increased number of charge hopping pathways.(85 cm3) was added n-butyllithium (16.33 cm3 of 1.6 M solution in hexanes, 26.11 mmol) at room temp. After 15 min stirring at room temp., the resulting LDA solution was cooled to Experimental -80 °C. A solution of 2 (8.52 g; 26.13 mmol) in THF (100 cm3) was added at -80 °C and the temperature allowed to rise NMR spectra were recorded using a Bruker AM300 spectrometer (300 MHz, 1H and 75.5 MHz, 13C) and a Varian slowly to -40 °C where it was stirred for 40 min.The solution J. Mater. Chem., 1998, 8(1), 31–36 353 J. A. Crayston, A. Iraqi and J. C. Walton, Chem. Soc. Rev., 1994, was then cooled to -60 °C and MgBr2·Et2O (6.75 g; 23, 147. 26.14 mmol) added. The mixture was stirred at -60 °C for 4 J. Roncali, Chem. Rev., 1992, 92, 711. 20 min then at -40 °C for 20 min and allowed to rise to -5 °C 5 R. Back and R. B. Lennox, L angmuir, 1992, 8, 959; G. Zotti, slowly whereupon all the MgBr2·Et2O had reacted. The S. Zecchin, G. Schiavon, A. Berlin, G. Pagani and A. Canavesi, resulting solution was cooled down to -80 °C and Chem.Mater., 1995, 7, 2309; G. Zotti, G. Schiavon, S. Zecchin, A. Berlin, G. Pagani and A. Canavesi, Synth.Met., 1996, 76, 255. [NiCl2(dppp)] (100 mg; 0.18 mmol) was added. The mixture 6 P.Ba� uerle and K. U. Gaudl, Synth.Met., 1991, 43, 3037. was allowed to warm to room temp. overnight. The solution 7 A. Yassar, M. Hmyene, D. C. Loveday and J.P. Ferraris, Synth. was then evaporated under reduced pressure to half the volume Met., 1997, 84, 231. and was poured over methanol (800 cm3). The resulting red 8 (a) J. Grimshaw and S. D. Perera, J. Electroanal. Chem., 1990, 278, precipitate was filtered, washed with methanol, water and 287; (b) J. A. Crayston, A. Iraqi, P. Mallon and J. C.Walton, Synth. methanol again, and dried under high vacuum to aVord the Met., 1993, 55, 867. 9 A. Iraqi, J. A. Crayston, J. C. Walton and A. Harrison, J. Mater. polymer as a deep red powder (yield 4.0 g; 62%). Chem., 1995, 5, 1291. The polymer was readily soluble in methylene chloride, 10 G. Zotti, A. Berlin, G. Pagani, G. Schiavon and S. Zecchin, Adv. chloroform and tetrahydrofuran. NMR spectra of the polymer Mater., 1995, 7, 48. in chloroform were recorded and showed 94% head-to-tail 11 A.Charlton, A. E. Underhill, G. Williams, M. Kalaji, P. J. Murphy, coupling. dH (CDCl3, SiMe4) 1.48 (m, 4H), 1.70 (m, 2H), 1.90 K. M. A. Malik and M. B. Hursthouse, J. Org. Chem., 1997, 62, (m, 2H), 2.60 (br t, 0.12H), 2.80 (br t, 2H), 3.40 (t, J 7, 2H), 3098; C. Thobiegautier, A. Gorgues, M. Jubault and J.Roncali, Macromol., 1993, 26, 4094; M. R. Bryce, A. D. Chissel, J. Gopal, 6.98 (s, 1H). dC (CDCl3, SiMe4) 28.01 (1C), 28.66 (3C), 29.31 P. Kathirgamanathan and D. Parker, Synth.Met., 1991, 39, 397. (1C), 30.35 (1C), 32.76 (1C), 33.96 (1C), 128.81 (1C), 130.78 12 J. A. Crayston, A. Iraqi, J. J. Morrison and J. C. Walton, Synth. (1C), 133.87 (1C), 139.80 (1C). UV–VIS (CHCl3) lmax 445 nm, Met., 1997, 84, 441; S.S. Zhu and T. M. Swager, Adv.Mater., 1996, band edge 560 nm. (Film cast from CH2Cl2, lmax 511 nm, band 8, 497; D. D. Graf and K. R. Mann, Inorg. Chem., 1997, 36, 150; edge 670 nm (1.8 eV). Gel permeation chromatography (GPC) 141; G. King, S. J. Higgins and N. Price, Analyst, 1992, 117, 1243; in chloroform revealed an average molecular mass of 12 300 G.King, S. J. Higgins and A. Hopton, J. Chem. Soc., Dalton T rans., 1992, 3403; J. L. Reddinger and J. R. Reynolds, Synth. Met., 1997, with a polydispersity of 1.5. 84, 225; J. X. Wang and F. R. Keene, J. Electroanal. Chem., 1996, 405, 71; J. X. Wang, M. Pappalardo and F. R. Keene, Aust. Preparation of polymer 3 J. Chem., 1995, 48, 1425; M. O. Wolf and M. S. Wrighton, Chem. Mater., 1994, 6, 1526; S.Higgins and J. A. Crayston, Synth. Met., 1,8-Diazabicyclo[5.4.0]undec-7-ene (0.24 g; 1.57 mmol) was 1993, 55, 879. added to a solution of 2-carboxyanthraquinone (0.78 g; 13 P. Bauerle and St. Scheib, Acta Polym., 1995, 46, 124. 3.09 mmol) in THF (50 cm3). The solution was stirred for 1 h 14 M. J. Marsella, R. J. Newland, P. J. Carroll and T. M. Swager, after which a precipitate was formed and poly(1) (0.38 g; J.Am. Chem. Soc., 1995, 117, 9842. 1.55 mmol) was then added and the mixture placed in a sealed 15 P. Hapiot, F. Demanze, A. Yasser and F. Garnier, J. Phys. Chem., 1996, 100, 8397; F. Demanze, A. Yassar and F. Garnier, tube and heated at 100 °C for 18 h. The solution was evaporated Macromolecules, 1996, 29, 4267. to dryness in high vacuum.The residue obtained was dissolved 16 A. E. Yassar, C. Moustrou, H. K. Youssoufi, A. Samat, in methylene chloride (20 cm3), and the precipitate (ammonium R. Guglielmetti and F. Garnier, J. Chem. Soc., Chem. Commun., bromide) formed was filtered oV. Upon reducing the solution 1995, 471. to 10 cm3 and addition of light petroleum (bp 40–60 °C) 17 J. Lowe and S. Holdcroft,Macromolecules, 1995, 28, 4608. 18 K. Faid and M. Leclerc, Chem. Commun., 1996, 2761. (250 cm3), a red precipitate formed and was filtered and 19 (a) R. D. McCullough, R. D. Lowe, M. Jayaraman and isolated. It was then mixed with water (250 cm3) and stirred D. L. Anderson, J. Org. Chem., 1993, 58, 904; (b) R. D. McCullough for 30 min, filtered and washed with MeOH (250 cm3) then and S. P.Williams, J. Am. Chem. Soc., 1993, 115, 11608; (c) T.- diethyl ether (200 cm3) and dried under high vacuum to aVord A. Chen and R. D. Rieke, J. Am. Chem. Soc., 1992, 114, 10087; the polymer as a deep red powder. (Yield 0.47 g; 73%). The (d) T. A. Chen, X. M. Chu and R. D. Rieke, J. Am. Chem. Soc., polymer was readily soluble in methylene chloride and chloro- 1995, 117, 233. 20 A.Iraqi, J. A. Crayston and J. C. Walton, J. Mater. Chem., 1995, form.NMRspectra of the polymer in chloroform were recorded 5, 1831. and showed that 87% grafting of anthraquinone had occurred. 21 A. Iraqi, J. A. Crayston and J. C. Walton, UK Pat. Appl., 1996, The ratio of anthraquinone incorporation was determined by 9601634.0. comparing integral heights of the hydrogens directly attached 22 P. Ba�uerle, F. Wu� rthner and S. Heid, Angew. Chem., Int. Ed. Engl., to the backbone of the polymer to the ones from the anthraqui- 1990, 29, 419. none substituents. dH (CDCl3, SiMe4) 1.51 (m, 4H), 1.61 (m, 23 M. Hiller, C. Kranz, J. Huber, P. Bauerle and W. Schuhmann, Adv. Mater., 1996, 8, 214. 2H), 1.77 (m, 2H), 2.76 (br t, 2H), 4.37 (m, 2H), 6.90 (s, 1H), 24 (a) M. Trznadel, M. Zagorska, M. Lapkowski, G. Louarn, 7.75 (m, 2H), 8.28 (m, 4H), 8.84 (m, 1H). UV–VIS spectral S. Lefrant and A. Pron, J. Chem. Soc. Faraday T rans., 1996, 92, analysis on polymer 3 showed an absorption peak at lmax 1387; (b) G. Louarn, M. Trznadel, J. P. Buisson, J. Laska, A. Pron, 445 nm in solution in CHCl2CHCl2 with a band edge (onset M. Lapkowski and S. Lefrant, J. Phys. Chem., 1996, 100, 12532. of absorption) at 576 nm. For films cast from CH2Cl2 an 25 P. Audebert, G. Bidan and M. Lapkowski, J. Electroanal. Chem., absorption at lmax 530 nm with a band edge at 700 nm (band 1987, 219, 165. 26 R. S. K. A. Gamage, A. J. McQuillan and B. M. Peake, J. Chem. gap 1.77 eV) was observed. Soc., Faraday T rans., 1991, 87, 3653. 27 T. Yamamot and H. Etori, Macromolecules, 1995, 28, 3371. 28 A. R. Hillman and E. F. Mallen, J. Chem. Soc., 1991, 87, 2209. References 29 (a) C. Roux and M. Leclerc, Chem.Mater., 1994, 6, 620; (b) C. Roux and M. Leclerc, Macromolecules, 1992, 25, 2141. 1 Handbook of Conducting Polymers, Vol. 1 and 2, ed. T. J. Skotheim, 30 G. R. V. Hecke and W. D. Horrocks, Inorg. Chem., 1966, 5, 1968. Marcel Dekker, NY, 1986. 2 Molecular Electronics ed. G. J. Ashwell, Research Studies Press, Wiley, Chichester, 1992. Paper 7/03397D; Received 16th May 1997 36 J. Mater. Chem., 1998, 8(1), 31&nd

ISSN:0959-9428    

DOI:10.1039/a703397d

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

J O U R N A L O F C H E M I S T R Y Materials Attachment of human keratinocytes to plasma co-polymers of acrylic acid/octa-1,7-diene and allyl amine/octa-1,7-diene Richard M. France,a Robert D. Short,*a Rebecca A. Dawsonb and Sheila MacNeilb aL aboratory of Surface and Interface Analysis, Department of Engineering Materials, University of SheYeld, Mappin Street, SheYeld, UK S1 3JD bDepartment of Medicine, Clinical Sciences Centre, University of SheYeld, Northern General Hospital, SheYeld, UK S5 7AU Plasma co-polymers (PCPs) of acrylic acid/octa-1,7-diene and allyl amine/octa-1,7-diene have been prepared and characterised using X-ray photoelectron spectroscopy (XPS).The use of a hydrocarbon diluent in the monomer feed allowed the deposition of films with controlled concentrations of carboxylic acid and nitrogen-containing functional groups.Human keratinocytes were cultured on these PCP surfaces, tissue culture poly(styrene) (TCPS) and collagen I. The level of keratinocyte attachment over 24 h was measured. PCP surfaces containing low concentrations of carboxylic acid groups (2.3%) were found to promote keratinocyte attachment. The performance of these PCPs was similar to collagen I, a well established substratum for attachment. Nitrogencontaining PCP surfaces were found to promote attachment at higher functional group concentrations, although the attachment did not attain the level achieved on the acid functionalised PCP surfaces.Plasma polymers (PPs) are thin polymeric pinhole free films, teins such as fibronectin, laminin, vitronectin and collagen.6 The tripeptide sequence Arg–Gly–Asp (RGD) has been ident- that can be produced from continuous-wave radio-frequency ified in many cell adhesion proteins.7 Synthetic RGD has been induced non-equilibrium plasmas of volatile organic comcoated on surfaces8 or covalently bound9 to study the depen- pounds.The technique is attractive, as deposition is possible dence of cell adhesion on RGD surface concentration.onto almost all solid materials, irrespective of geometry, with The response of human keratinocytes to natural or synthetic minimal or no pre-treatment required. It allows exact control surfaces is of importance in wound care and healing. The over film thickness, and polymerisation takes place in a clean influence of extracellular matrix proteins on human kera- environment.Films deposited by plasma polymerisation are tinocyte attachment, proliferation and transfer to a dermal free of remnants of initiator or solvents. In a plasma, there is wound bed model has been studied.10 Matrigel, collagen I and considerable fragmentation of the original compound and a IV were found to enhance initial attachment; RGD, vitronectin, wide range of functional groups are incorporated into the fibronectin and irradiated 3T3 fibroblasts did not.Proliferation deposit. In this laboratory we have examined the relationship of cells was also found to be positively influenced (although to between the plasma deposition parameters (plasma power, P a lesser extent than initial attachment) on matrigel, collagen I and monomer flow rate, F) and functional group retention, in and IV and irradiated 3T3 fibroblasts.the deposit. We have shown that by employing low P/F it is The response of cells to specific surface functional groups possible, in many cases, to fabricate films with a high degree represents the most fundamental level of study. The basic of functional group retention.This has been demonstrated for requirement for this type of study is a surface of well defined various methacrylates,1 allyl alcohol2,3 and acrylic acid.4 chemistry. Cell–surface interactions have been investigated on Recently, plasma co-polymerisation has been used in this ion exchange materials,11 self assembled monolayers (SAMs)12 laboratory to control the amount of functional groups present and plasma polymers.13 in the plasma polymer.The introduction of a hydrocarbon This study investigates the attachment of human kera- diluent into the monomer feed has been used to prepare plasma tinocytes to PCP surfaces. The level of keratinocyte attachment co-polymers (PCPs) with controlled amounts of carboxylic to carboxylic acid and nitrogen functional groups, over a range acid and nitrogen-containing functional groups.5 The technique of functional group concentrations has been studied.Tissue of plasma co-polymerisation also allows the production of culture poly(styrene) (TCPS), a hydrocarbon plasma polymer films which are insoluble in aqueous media. and collagen I have been employed as reference surfaces. TCPS Cell–surface interactions influence or control many aspects is generally acknowledged to be a poor substrate for the of cell physiology, including adhesion, proliferation and diVerkeratinocyte, while collagen I has been shown to be a preferred entiation. Knowledge arising from the study of cell–surface substrate for the keratinocyte.10 interactions may be applied in the development of agents to enhance or inhibit cell–substrate interactions for tissue regeneration or biomaterials integration.In general cell–surface inter- Experimental actions are poorly understood. This is due to the complexity Plasma co-polymerisation of these interactions and the large number of parameters which control them. Surface chemistry is known to influence cellular Acrylic acid, allyl amine and octa-1,7-diene were obtained attachment, either directly, or through the protein layer which from Aldrich (UK).All monomers were used as received, after rapidly absorbs to surfaces in contact with serum containing several freeze–pump–thaw cycles. Polymerisation was carried culture media. To examine cell–surface interactions, many out in a cylindrical reactor vessel (of 8 cm diameter and 50 cm simplified models have been used.Biologically active molecules in length), evacuated by a two-stage rotary pump. The plasma have been coated or immobilised to various substrata to was sustained by a radio-frequency (13.56 MHz) signal generproduce bioactive surfaces. Molecules that have been investi- ator and amplifier inductively coupled to the reactor vessel. The base pressure in the reactor was 3×10-3 mbar.gated include glycosaminogylcan matrix components and pro- J. Mater. Chem., 1998, 8(1), 37–42 37Acrylic acid and octa-1,7-diene were co-polymerised at a nitrogen functionalised PCPs, an MTT–ESTA assay17 was used. This estimates the viable cell number, the assay having plasma power of 2 W and a total flow rate of 2.0 sccm [sscm= cm3(STP) min-1].Allyl amine and octa-1,7-diene were co- been previously shown to parallel increases in cell number for human keratinocytes.18 Cells were incubated with 0.5 mg ml-1 polymerised at a plasma power of 3 W and a total flow rate of 2.0 sccm. Plasma co-polymers were deposited onto tissue of MTT in PBS for 40 min. The stain was then eluted with acidified isopropyl alcohol.An optical density measurement culture wells and clean aluminium foil (for XPS analysis). The pressure during co-polymerisation was typically was then made at 540 nm with a protein reference wavelength of 630 nm which was subtracted. 4.0×10-2 mbar. For all co-polymerisations, a deposition time of 20 min was used. The monomer mixtures were allowed to flow for a further Results 20 min after the plasma was switched oV.This was done in an attempt to minimise the uptake of atmospheric oxygen by the Characterisation of PCPs deposits on exposure to the laboratory atmosphere. XP survey scan spectra of PCPs prepared from acrylic acid and octa-1,7-diene revealed only carbon and oxygen in the X-Ray photoelectron spectroscopy deposits. The O/C ratios were measured and are shown in Table 1.The O/C ratio was observed to increase as the molar XP spectra were obtained on a VG CLAM 2 photoelectron fraction of acrylic acid in the monomer feed increased. If ideal spectrometer employing Mg-Ka X-rays. Survey scan spectra gas behaviour is assumed then the molar fraction of acrylic (0–1100 eV) and narrow spectra were acquired for each sample acid is equal to the flow rate ratio Faa/Ftot, where Ftot= using analyser pass energies of 50 and 20 eV respectively.Faa+Foct. The C 1s core level spectra of the PCPs were peak Spectra were acquired using Spectra 6.0 software (R. Unwin fitted for various oxygen-containing functionalities.15 First, Software, Cheshire, UK). Subsequent processing was carried spectra were corrected for sample charging, setting the hydro- out with Scienta data processing software (Scienta Instruments, carbon signal to 285 eV.The following functionalities were Uppsala, Sweden). The spectrometer was calibrated using the then fitted: alcohol/ether (CmOH/R) at a shift of +1.5 eV; Au 4f 7/2 peak position at 84.00 eV, and the separation between carbonyl (CnO) at +3.0 eV; carboxylic acid/ester (COOH/R) the C 1s and F 1s peak positions in a sample of PTFE at +4.0 eV; and a b-shifted carbon bonded to carboxylate measured at 397.2 eV, which compares well with the value of (CmCOOH/R) at +0.7 eV.The results of peak fitting are 397.19 eV reported by Beamson and Briggs.14 shown in Table 1 and an example peak fit (Faa/Ftot=1) is shown in Fig. 1(a). In the peak fit the FWHM of component Cell culture peaks were kept equal and were in the range 1.4–1.6 eV. The Normal human dermal keratinocytes (obtained from breast Gaussian to Lorentzian ratios (G/L) of the component peaks reductions and adominoplasties) were isolated from the dermal/ were also kept constant and were in the range 0.8–0.9. While epidermal junction as previously described.15 Cells were cul- XPS cannot distinguish between carboxylic acid and ester tured in complete Green’s media, which included cholera toxin groups, grazing angle IR spectroscopy of plasma polymerised (0.1 nM), hydrocortisone (0.4 mg ml-1), EGF (10 ng ml-1), acrylic acid has shown, that at the low powers employed in adenine (1.8×10-4 M), triiodo-L-thyronine (2×10-7 M), insulin this study, the carboxylate peak in the XP spectra can be (5 mg ml-1), transferrin (5 mg ml-1), glutamine (2×10-3 M), assigned to carboxylic acid rather than ester.4 There is a direct fungizone (0.625 mg ml-1), penicillin (1000 IU ml-1), strepto- relationship between the proportion of acrylic acid in the mycin (1000 mg ml-1) and 10% fetal calf serum. Cells were monomer feed and the concentration of acid groups in the cultured at 37 °C, in a 5% CO2 atmosphere.PCPs. This can be seen in the data reported in Table 1. This Experiments were carried out using both freshly isolated relationship is described in more detail elsewhere.5 Other and subcultured (passaged) cells. For the latter, cells were carbon–oxygen functionalities present in the PCPs (besides cultured (in collagen I coated flasks) for four days, being re- carboxylic acid) include carbonyl and alcohol/ether.These fed once with fresh media. The cells were then trypsinised arise as a result of fragmentation of the monomer in the (0.5% trypsin in EDTA), prior to seeding on the experimental plasma. Reaction between the deposit and water desorbed substrates. TCPS was obtained from Corning Glass Co. (USA).from the walls of the plasma vessel (during polymerisation) Collagen coated tissue culture plates were prepared by air and atmospheric oxygen and water (after polymerisation) also drying a solution of collagen I (32 mg cm-2) in 0.1 M acetic contribute. The CmOH/R peak is thought to be predominantly acid (200 mg ml-1) in a laminar flow cabinet overnight. ether. In the pure hydrocarbon PCP (Faa/Ftot=0), the O/C ratio determined from the C 1s and O 1s peak areas is 0.01.Cell attachment assay This is considerably lower than the O/C calculated from the peak fit (0.03). In the calculation of the O/C ratio from the Cells were seeded at densities of between 1.6 and 7.0×105 cells ml-1 onto six well (3.5 cm diameter) tissue culture plates. The wells were either uncoated, PCP coated or collagen I Table 1 Summary of XPS results for PCPs prepared from acrylic acid coated. 1 ml was added to each well. Cells were seeded on and octa-1,7-dienea PCPs within 48 h of preparation. Keratinocytes were allowed % of functionality in the C 1s core level to attach for 24 h; unattached cells were removed by a gentle wash with 1 ml of phosphate buVer solution (PBS).Faa/Ftot O/C ratio CmC, CmH CmOR CnO COOH/R For all substrates, the level of cell attachment was determined, by an estimation of total DNA content using the 0 0.01 97.2 3.4 — — Hoechst stain, as detailed in ref. 16. Cells were incubated in a 0.52 0.10 90.3 5.4 0.1 2.3 saline sodium citrate (SSC) + urea + sodium dodecyl sulfate 0.64 0.14 84.0 6.3 0.5 4.7 0.80 0.28 70.7 7.4 1.3 10.4 (SDS) digestion buVer for 1 h.Following digestion, cells were 0.90 0.43 58.3 6.8 2.2 16.4 stained with Hoechst stain (in an SSC buVer at 1 mg ml-1). 1.00 0.53 47.0 7.5 3.5 21.0 Fluorescence was measured using excitation and emission wavelengths of 355 and 460 nm, respectively. DNA content aA b shift (at +0.7 eV from the hydrocarbon) of equal magnitude to was then estimated from a standard curve of known DNA the carboxylate has been added to the peak fit.Conditions for concentrations. polymerisation were Ftot=2.0 sccm, power=2W, deposition time= 20 min. To determine keratinocyte attachment and viability on the 38 J. Mater. Chem., 1998, 8(1), 37–42vessel walls. The C 1s core level spectra of the PCPs were peak fitted for nitrogen-containing functionalities, using chemical shift values reported in the literature.14,19 The functionalities fitted were: amine (CmNR2) at +0.9 eV; imine (CnN) at +1.7 eV; and amide (CNO) at +3.0 eV.The results of these peak fits are shown in Table 2 and an example peak fit (Faa/Ftot=1) is displayed in Fig. 1( b). The FWHM used were in the range 1.5–1.7 eV and the G/L ratios in the range 0.8–0.9. It is obvious from the peak fit results that there has been lower selectivity towards the amine functionality in allyl amine/octa- 1,7-diene PCPs than might have been anticipated, particularly when compared with the acrylic acid/octa-1,7-diene PCPs.Imine and amide functionalities are present in the allyl amine/ octa-1,7-diene PCPs. In addition to the low selectivity, the nature of the amine, i.e.primary, secondary or tertiary, is unknown. Keratinocyte culture on PCPs containing carboxylic acid groups The eVect of carboxylic acid group concentration on keratinocyte attachment was investigated. Six well tissue culture plates were coated with PCPs. Freshly isolated keratinocytes were seeded at 6.2×105 cells ml-1 on these PCP surfaces. Three wells were used on each plate.Carboxylic acid functional group concentration was found to have a significant eVect on the level of cell attachment. The results are shown in Fig. 2(a). On PCP surfaces containing no carboxylic acid groups (PP of octa-1,7-diene), the level of attachment was found to be poor; comparable to that on TCPS. On PCP surfaces containing 2.3% carboxylic acid Fig. 1 Peak fitted C 1s core levels of PPs of (a) acrylic acid and (b) allyl amine.A: CmC, CmH, B: CmCOOH/R, C: CmOR/H, D: CnO, E: COOH/R, F: CmNR2, G: CnN, H: CNO. peak fit, the total number of carbons bonded to oxygen is divided by 100 carbon atoms to yield an O/C ratio (for Faa/Ftot=0, we calculate 3.4 oxygen atoms per 100 carbons). In the peak fit, the ether functionality will be counted twice, since two carbons experience the same shift brought about by one (shared) oxygen atom.In the PCPs the number of CmOH/R environments is constant; between 5 and 7 per 100 carbons. In the hydrocarbon PP of octa-1,7-diene the number is a little lower (3.4 per 100 carbons). XP survey scans were made of PCPs prepared from allyl amine and octa-1,7-diene. These deposits contained carbon, nitrogen and oxygen.The XPS data were quantified and the results are shown in Table 2. Small amounts of oxygen were detected in these PCPs. This oxygen was most likely incorporated on exposure to the laboratory atmosphere. However, oxygen may also have been incorporated within the plasma. Water is known to be constantly desorbing from the reactor Table 2 Summary of XPS results for PCPs prepared from allyl amine and octa-1,7-dienea % of functionality in the C 1s core level CmC, CmNR2 Faa/Ftot N/C ratio O/C ratio CmH (CmOR) CnN CNO 0 0 0.01 97.5 0 (3.5) 0 0 Fig. 2 Attachment of keratinocytes to PCP surfaces containing car- 0.48 0.06 0.03 82.9 12.3 4.8 0.4 0.74 0.15 0.08 71.2 13.7 11.6 3.9 boxylic acid groups.Attachment was estimated from the total DNA of cells attached to surface.(a) Freshly isolated cells (seeded at 1.00 0.37 0.06 59.5 20.2 17.6 3.0 6.2×105 cells ml-1). (b) First passage keratinocytes (from another donor, seeded at 1.6×105 cells ml-1). Results shown are the aXPS analysis of sample prepared at Faa/Ftot=0.74 was carried out 11 days after preparation. Conditions for polymerisation were Ftot= means±s.e.m. of triplicate wells of cells (s.e.m.=standard error of the mean). 2.0 sccm, power=2 W, deposition time=20 min. J. Mater. Chem., 1998, 8(1), 37–42 39groups very high levels of cell attachment were measured. The significant influence on the level of cell attachment. Cell attachment was estimated by both DNA and MTT–ESTA level of attachment was comparable to that measured on collagen I.On PCPs containing.2.3% carboxylic acid groups, assays (three wells on the same tissue culture plate being used for each experiment). The results are shown in Fig. 4(a) and the level of attachment was observed to be reduced. However, the amount of attachment measured was still in excess of that (b), respectively. In Fig. 4(a), the level of cell attachment is shown to increase with the number of nitrogen-containing measured on the PCP surfaces with no carboxylic acid chemistry.functionalities. However, the overall level of attachment is lower than that observed upon TCPS. It should be noted here Freshly isolated cells were examined by light microscopy, prior to washing with PBS. Optical micrographs are shown in that we will subsequently argue that TCPS is not a good reference surface, because of possible variability in surface Fig. 3. Cells on PCP surfaces of low concentrations of acid groups and on collagen I, Fig. 3( b) and (e) respectively, display chemistry between samples. In our first experiment (with acid functional groups) there was a similar level of cell attachment morphological features consistent with good attachment and spreading.In contrast, cells seeded on the PCP surfaces to both TCPS and to the pure hydrocarbon plasma polymer surface. Also, in that experiment a marked diVerence was seen containing no acid groups and on TCPS, Fig. 3(a) and (d), respectively, show poor attachment: the cells are rounded and between TCPS and collagen. The pure hydrocarbon plasma polymer surface used in this experiment was identical to the lighter in appearance.Cells seeded on the PCP surfaces of higher acid group concentrations, Fig. 3(c), show a mixed one used before. Therefore, this surface should be considered as the negative control in both experiments. response: some have attached well and spread, whilst others are poorly attached. The level of attachment as estimated by the MTT–ESTA assay, Fig. 4(b), mirrors the data shown by the DNA assay, The attachment of passaged cells (from a diVerent donor, first passage, seeded at 1.6×105 cells ml-1) to PCP surfaces Fig. 4(a). Enhanced cell attachment is observed at higher nitrogen functional group concentrations. However, even at was also investigated, using the remaining three wells of the six well tissue culture plates. The results from the DNA assays the highest concentration, the level of attachment does not approach that achieved on collagen I.are shown in Fig. 2(b). Significant enhancement in the level of attachment (over that observed upon PCP surfaces containing Keratinocytes were examined by light microscopy prior to washing with PBS. Optical micrographs are shown in Fig. 5. no carboxylic acid groups) was again seen on the PCP surfaces containing a low concentration (2.3%) of carboxylic acid Cells on the hydrocarbon PCP, Fig. 5(a), show poor attachment: the cells are round and light in appearance. The darker groups. Enhanced attachment was also seen on surfaces containing higher acid group concentrations, although the level of cells present on this micrograph are diVerentiated keratinocytes. Cells seeded upon both TCPS and collagen I, attachment to these surfaces failed to reach that measured on collagen I.Keratinocyte culture on PCPs containing nitrogen functional groups The attachment of freshly isolated keratinocytes (seeded at 7.0×105 cells ml-1) to PCP surfaces with diVerent concentrations of nitrogen-containing functionalities was investigated.Nitrogen-containing functional groups were found to have a Fig. 3 Freshly isolated keratinocytes on PCPs containing carboxylic Fig. 4 Attachment of keratinocytes to PCPs containing amine groups acid groups and on control substrates. Micrographs were taken 24 h post cell addition, prior to washing with phosphate buVer solution. (a) and to control surfaces. The level of attachment was estimated from (a) the total DNA content of cells and (b) the MTT–ESTA assay Hydrocarbon PCP (no carboxylic acid groups).(b) PCP surface containing 2.3% carboxylic acid groups. (c) PCP surface containing (which assesses cell viability). All cells were freshly isolated (seeded at 7.0×105 cells ml-1). Results shown are the means±s.e.m. of triplicate 21.0% carboxylic acid groups.(d) Tissue culture poly(styrene). (e) Collagen 1. wells of cells. 40 J. Mater. Chem., 1998, 8(1), 37–42increase in cell attachment may result from an increase in the concentration of one specific nitrogen functionality, e.g. amine. Alternatively, cell attachment may have increased as a result of the increase in the total amount of nitrogen surface functionality. Significantly, the level of attachment on theses PCP surfaces was lower than on collagen I.Again, the hydrocarbon plasma polymer surface did not promote attachment. A serious problem with one of the controls, TCPS, was identified. In the experiment with acid functionalised PCP surfaces, the level of attachment to the TCPS was comparable to the hydrocarbon plasma polymer surface. In the experiment with nitrogen functionalised PCP surfaces the TCPS performed significantly better than the hydrocarbon plasma polymer surface.Given that cells from diVerent donors were used in these two experiments, it is important to have confidence in the controls. In these two experiments, collagen I and hydrocarbon plasma polymer surfaces were also used. In both experiments, there were approximately 20 times more cells attaching to the collagen I than to the hydrocarbon plasma polymer surface.The hydrocarbon plasma polymer surface can thus be considered the negative control, which does not promote attachment, and collagen I the positive control. We can use the results on these two surfaces to ‘normalise’ our data. The variable performance of the TCPS must still be explained.Keratinocytes from diVerent donors do display considerable variation in culture as we have previously noted.10 Fig. 5 Freshly isolated keratinocytes on PCPs containing amine groups and on control substrates. Micrographs were taken 24 h post However we suspect that in this case the cause of this variation cell addition, prior to washing with phosphate buVer solution.(a) is the TCPS. TCPS is known to receive a propriety treatment, Hydrocarbon PCP (no amine groups). (b) PCP surface containing probably corona or plasma discharge. The surface of the TCPS 13.7% amine groups. (c) PCP surface containing 20.2% amine groups. was examined by XPS and it was found to contain oxygen. (d) Tissue culture poly(styrene). (e) Collagen I. The surface O/C ratio was approximately 0.25.Peak fitting revealed a range of diVerent oxygen-containing functional groups. In a separate study we have investigated the surface Fig. 5(d) and (e), respectively, display morphological features indicative of good attachment and spreading. In these micro- oxidation of PS.20 We have shown that above an O/C ratio of 0.18, PS surfaces are not ‘stable’ to aqueous solutions.That is, graphs, more attached and well spread cells (those darker in appearance) are observed upon collagen I, cf. TCPS, consistent low molecular mass material can be extracted. It is also uncertain whether diVerent batches of TCPS would have with the cell attachment assay results. The cells seeded upon PCPs containing nitrogen functionalities, Fig. 5( b) and (c), received exactly the same level of surface treatment, or whether in any one batch, the surface treatment is stable to ageing.show a mixture of good attachment/spreading and poor attachment. Based on this, we suggest that TCPS surfaces do not make for good controls. Other methods have been used to prepare surfaces of well Discussion defined chemistry, on which cellular attachment has then been measured.Recent reports have appeared describing the prep- Human keratinocytes have been successfully cultured on PCP surfaces containing carboxylic acid and nitrogen functional aration of such surfaces using self-assembly monolayers (SAMs). SAMs are typically formed by the adsorption of groups. The use of co-polymerisation has allowed the eVect of functional group concentration to be studied.alkylthiols onto gold supports21 or by the chemisorption of alkylsilanes onto silica surfaces.22 Recent papers describe the Keratinocytes show high levels of attachment to PCP surfaces containing low levels of carboxylic acid functionalities culturing of cells on SAMs with a variety of diVerent surface chemistries.23–25 3T3 murine fibroblasts have been successfully (2.3%).The number of cells attached to these surfaces is comparable to the number of cells attaching to collagen I, a cultured on SAMs formed by the adsorption of carboxylic acid- and methyl-functionalised alkylthiols on gold surfaces.12 preferred substratum material for the culturing of keratinocytes. Optical microscopy of cells cultured on these two A strong preference was exhibited for the COOH terminated monolayer.On this surface there was a greater number of surfaces reveals morphologies characteristic of both attachment and spreading. Enhanced attachment was also observed at attached cells, and these cells exhibited a more well spread morphology. On the methyl-terminated surface, cells were higher acid group concentrations. The same trend was found for both freshly isolated cells and passaged cells.Hydrocarbon more rounded and clumped together. Human osteoblasts have also demonstrated a preference for COOH terminated sur- plasma polymer surfaces did not promote attachment. The attachment of keratinocytes to PCPs of allyl amine and faces.25 Endothelial cells have shown a strong preference for COOH surfaces, and also for surfaces containing mixtures of octa-1,7-diene was greatest at high functional group concentrations.Optical microscopy revealed a mixed response on OH and COOH.26 The use of SAMs, as opposed to plasma polymers, for this these surfaces: some cells were attached and well spread, others were poorly attached. These PCP surfaces do not contain one type of study oVers real advantages.First, there is a greater level of control over substratum chemistry. Even in the car- unique nitrogen functionality. The original objective of this study was to prepare amine-containing surfaces, however, the boxylic acid PCPs some other oxygen-containing functional groups have been introduced. This problem does not arise selectivity towards amine functionalities in these PCPs was lower than anticipated.The concentration of other nitrogen with SAMs. Further, structural order can be introduced into SAMs. functionalities (imine and amide) increased disproportionally as the amount of allyl amine in the monomer feed increased. Plasma polymerisation also oVers advantages, particularly when it comes to device fabrication. PPs can be deposited This confuses the interpretation of our data.The observed J. Mater. Chem., 1998, 8(1), 37–42 41directly onto most surfaces. Objects of complicated geometry revealed that on these low acid content surfaces, and on collagen I, keratinocytes were well attached and spread. On can be coated. In PPs, the distribution of functional groups, because of the nature of the process, will be uniform across PCP surfaces of higher acid content, the level of attachment was still significantly above that seen on the negative control the surface. The surface concentration of functional groups can be controlled and surfaces containing more than one (a hydrocarbon PP), but well below the optimum (collagen I).Plasma co-polymerisation of allyl amine and octa-1,7-diene functional group should, in principle, be readily prepared.PPs can be engineered with specific hydration properties, through produced surfaces containing a mixture of nitrogen functional groups, predominantly amine and imine. The attachment of the control of cross-links. These two approaches should be considered complementary. keratinocytes to PCP surfaces containing nitrogen functional groups increased with the nitrogen content of these surfaces, The preference exhibited here by keratinocytes for PCPs containing the carboxylic acid functional groups (over TCPS although cells never attained the level of attachment seen on collagen I.or hydrocarbon surfaces) confirms other findings with fibroblasts,12 osteoblasts25 and endothelial cells.26 Copolymerisation has allowed control over the surface carboxylic References acid concentration.The data show keratinocyte preference for 1 A. P. Ameen, L. O’Toole, R. D. Short and F. R. Jones, J. Chem. surfaces with low amounts of this functionality. The depen- Soc., Faraday T rans., 1995, 91, 1363. dence of cell attachment on functional group concentration 2 L. O’Toole and R. D. Short, J. Chem. Soc., Faraday T rans., 1997, has yet to be properly explored.SAMs have been prepared 93, 1141. from mixtures of COOH and OH terminated alkanethiols.26 3 A. P. Ameen, R. D. Short and R. J.Ward, Polymer, 1994, 35, 4382. The level of attachment to binary mixtures of COOH and OH 4 L. O’Toole, A. J. Beck and R. D. Short, Macromolecules, 1996, 29, 5172. is greater than to single component COOH terminated SAMs. 5 A. J. Beck, F. R. Jones and R. D. Short, Polymer, 1996, 37, 5537. Enhanced attachment of cells to surfaces containing mixtures 6 P. D. Drumheller, J. A. Hubell, T he Biomedical Engineering of functionalities, has also been observed on fluorinated sur- Handbook, ed. J. D. Bronzino, CRC Press, Inc., 1995. faces.27 In ref. 27, PPs (containing both CF2 and CF3 groups) 7 E.Ruoslahti and M. D. Pierschbacher, Science, 1987, 238, 491. were observed to support the growth of bovine aortic endo- 8 P. A. Underwood and F. A. Benett, J. Cell Sci., 1989, 93, 641. thelial cells. Single component SAMs (terminated with either 9 I. I. Singer, D. W. Kawa and S. Scott, J. Cell Biol., 1987, 104, 573. 10 R. A. Dawson, N. J. Goberdhan, E. Freedlander and S. MacNeil, CF2 or CF3 groups) did not support cell growth.Burns, 1996, 22, 93. There is only a limited literature pertaining to cellular 11 R. M. Shelton and J. E. Davies, T he Bone-Biomaterial Interface, behaviour on surfaces containing nitrogen functional groups. ed. J. E. Davies, Toronto University Press, 1991. The attachment and growth of human endothelial cells and 12 E. Cooper, R.Wiggs, D. A. Hutt, L. Parker, G. J. Leggett and fibroblasts on PPs containing amine and amide functional T. L. Parker, J. Mater. Chem., 1997, 7, 435. groups have been previously explored.28 In this work, cell 13 H. J. Griesser, R. C. Chatelier, T. R. Gengenbach, G. Johnson and J. G. Steele, J. Biomater. Sci., Polym. Ed., 1994, 5, 531. attachment followed nitrogen content within the films.It was 14 G. Beamson and D. Briggs, High Resolution XPS of Organic suggested that the amide functionality was the main promoter Polymers: T he Scienta ESCA300 Handbook, John Wiley and Sons, of cell attachment. These results are certainly in agreement Chichester, 1992. with those reported here for keratinocytes. 15 N. J. Goberdhan, M. Edgecombe, E. Freedlander and S. MacNeil, The study of cell–surface interactions, whether using SAMs Burns, 1997, 23, 122.or PPs, is very much in its infancy. Yet, general trends are 16 J. Rao and W. R. Otto, Anal. Biochem., 1992, 207, 186. 17 P. A. Ealey, M. E. Yateman, S. J. Holt and N. J. Marshall, J. Mol. already emerging. However much work is still to be Endocrinol., 1988, 1, 1. undertaken. 18 S. MacNeil, R. A.Dawson and G. Crocker, Br. J. Dermatol., 1993, The link between protein adsorption (to the substrate) and 128, 143. cellular attachment is extremely important, as surface func- 19 F. Fally, C. Doneux, J. Riga and J. J. Verbist, J. Appl. Polym. Sci., tional groups influence the nature and conformation of the 1995, 56, 597. 20 R. M. France and R. D. Short, J. Chem. Soc., Faraday T rans., 1997, adsorbed protein layer, and may only indirectly influence 93, 3173.cellular attachment. In particular, vitronectin and fibronectin 21 C. D. Bain, E. B. Troughton, Y-T. Tao, J. Evall, G. M. Whitesides have been implicated in cellular attachment.29 Understanding and R. G. Nuzzo, J. Am. Chem. Soc., 1989, 111, 321. of the nature of the adsorbed protein layer will be critical in 22 S.R. Wasserman, Y-T. Tao and G. M. Whitesides, L angmuir, 1989, understanding cellular interactions with substrates and will 5, 1074. form the basis of future work from this laboratory. 23 G. P. Lopez, M. W. Albers, S. L. Schreiber, R. Carroll, E. Peralta and G. M. Whitesides, J. Am. Chem. Soc., 1993, 115, 5877. 24 P. A. DiMilla, J. P. Folkers, H. A. Biebuyck, R. Harter, G. P. Lopez and G. M. Whitesides, J. Am. Chem. Soc., 1994, 116, 2225. Conclusions 25 C. A. Scotchford, M. Barley, K. Hughes, E. Cooper, G. J. Leggett and S. Downes, T ransactions of the Soc. For Biomater., 23rd Annual Plasma co-polymerisation was used to prepare surfaces on Meeting, 1997, vol. XX, p. 202. which keratinocyte attachment was investigated. Enhanced 26 C. D. Tidwell, A. M. Belu, B. D. Ratner, B. Tarasevich, S. Atre and attachment was observed to surfaces containing low concen- D. L. Allara, T ransactions of the Soc. For Biomater., 23rd Annual trations of carboxylic acid groups and to surfaces with high Meeting, 1997, vol. XX, p. 203. concentrations of nitrogen functional groups. 27 D. G. Castner, C. D. Tidwell, B. D. Ratner, M-W. Tsao and J. F. Raboult, T ransactions of the Soc. For Biomater., 23rd Annual Plasma co-polymerisation of acrylic acid and octa-1,7-diene Meeting, 1997, vol. XX, p. 206. gave surfaces containing predominantly (but not exclusively) 28 H. J. Griesser, R. C. Chatelier, T. R. Gengenbach, G. Johnson and carboxylic acid functional groups. The attachment of kera- J. G. Steele, J. Biomater. Sci., Polym. Ed., 1994, 5, 531. tinocytes to PCP surfaces containing concentrations of 2.3% 29 J. G. Steele, B. A. Dalton, G. Johnson and P. A. Underwood, acid groups was found to be comparable to the level of J. Biomed.Mater. Res., 1993, 27, 927. attachment observed to collagen I, the latter being a preferred surface for the culturing of keratinocytes. Optical microscopy Paper 7/05098D; Received 16th July, 1997 42 J. Mater. Chem., 1998, 8(1), 37–42

ISSN:0959-9428    

DOI:10.1039/a705098d

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

J O U R N A L O F C H E M I S T R Y Materials Studies of equid hoof horn material by EPR spectroscopy Barry C. Cope,a Lyn Hopegood,a Roger J. Latham,*a Roger G. Linford,a John D. Reilly,b Martyn C. R. Symonsa and Fatai A. Taiwoa aSchool of Applied Sciences, DeMontfort University, T he Gateway, L eicester, UK L E1 9BH bRoyal Army Veterinary Corps, Defence Animal Centre,Melton Mowbray, UK EPR spectroscopy has been used to show that stable free radicals are formed when hoof horn material from equids is cut.The form of the EPR signal suggests that these are stable sulfur-centred radicals. The detection of melanin and thus the distinction between pigmentation levels of hoof horn is noted. Possible implications of these results are discussed. Hoof horn material is remarkable in that it accommodates Bruker 6/1 EMX EPR spectrometer operating at a modulation major stresses, both chemical and physical, to which the hoof frequency of 100 MHz.Details of instrumental parameters are capsule is exposed. Historically, hoof management has been as stated for the respective spectra. regarded as a skill rather than a science and consequently ‘Uncut’ samples refers to long slivers (usually about there is a lack of information concerning structure–property 12×2×2 mm) obtained from original hoof clippings supplied relationships in both the scientific and veterinary literature.It by farriers. ‘Cut’ samples were obtained by cutting the sliver has been suggested recently,1 however, that multidisciplinary along its length, into chips (approximately 2×2×2 mm). scientific studies of hoof horn involving, for example, biochem- Samples from both pigmented and non-pigmented hoof horn ists, materials scientists and veterinarians may provide the best were examined.way of understanding the structure of this important natural material. The structure of hoof horn is highly complex,1–3 but Results and Discussion it can be described as a composite material based on tubules and intertubular material, whose biomechanical properties are Initial experiments have been carried out on samples from the very dependent on its structure.A major structural component inner and outer walls of both donkey and, for comparison, of the material is the protein a-keratin, which exists as a triple horse hoof horn. A typical EPR spectrum for a sample of the helix, strengthened greatly by cross-links comprised of sulfur– inner wall from a donkey hoof is shown in Fig. 1(a). Peaks are sulfur bonds. observed at g values 4.3, 2.003 ( line width 9.3 G) and 2.380 EPR (electron paramagnetic resonance) spectroscopy has ( line width 290 G). The spectrum for a sample from the outer been used to study aspects of other natural materials such as wall of the hoof is similar but the peak at g=2.380 is absent, stratum corneum from rats4 and human fingernail.5,6 It was as shown in Fig. 1(b). When compared, the spectra for the reported that when human fingernails are cut at room tempera- inner and outer walls of horse hoof horn are similar in ture, remarkably high yields of trapped free radicals were appearance, but they diVer from those of the donkey hoof formed that were characterised by EPR spectroscopy.5 The horn samples in that the peaks are at g values of 4.3 and 2.2 signals were analysed in terms of a species having highly (broad, line width 1050 G).The EPR parameters of the main characteristic features associated with a specific sulfur-centred features of the spectra of hoof horn are summarised in Table 1.radical. The structure of these radicals, which is controversial, The signal at g=4.3 is due to FeIII which has the configuris discussed below. These radicals are remarkably stable at ation d5 in the high spin spectroscopic state and high symmetry. room temperature, showing no detectable decay over periods This is attributed to haemichrome formation as a result of of several weeks.It is possible therefore that EPR could degradation of haemoproteins, predominantly from blood, provide information on the basic properties of hoof horn whereas no signal is expected from haemoglobin in fresh blood together with an understanding of neoplastic hoof horn con- containing FeII with its d6 configuration (low spin S=0, high ditions such as keratoma.spin S=2 but high transition energy and hence no EPR signal ). Its immediate oxidative product FeIII is expected to give a Experimental Samples Suitable clippings of hoof horn were obtained by farriers during regular hoof maintenance and sharp hoof cutters were used in order to prevent tearing of the sample. The samples were wrapped immediately in three overlapping layers of Parafilm (Parafilm ‘‘M’’ Laboratory Film, American National Can, CT 06836, USA) to make an airtight seal which moulded to the shape of the sample.The labelled, wrapped samples were then stored at 4 °C prior to examination. The portions removed for testing were from the midline dead centre site (MDC).7 Fig. 1 First derivative X-band spectrum of donkey hoof at 77 K; (a) sample obtained from the inner wall section and (b) sample obtained from the outer wall section (instrument settings: centre field 2880 G, EPR spectra sweep width 5000 G, microwave frequency 9.480 GHz, microwave First derivative EPR spectra were recorded at room tempera- power 2 mW, modulation amplitude 4 G, time constant 40 ms, sweep time 84 s) ture (ca. 298 K) and 77 K (in liquid nitrogen) on an X-band J.Mater. Chem., 1998, 8(1), 43–45 43Table 1 EPR parameters of the main features in the spectra of hoof deliberately cut by the farrier, and may undergo microfracture7 horn during exercise, on impact with sharp, stony material, for example. EPR spectroscopy also may be useful in the characterg value line width ( lw) or A value/G source isation of neoplastic hoof conditions such as keratoma. 2.382 lw 290 FeIIIMhigh spin, 4.3 high symmetry Nature of the free radicals gx=2.005 lw 5.1 (room temp.) melanina In terms of chemical expectation, bond-homolysis is likely, gy=2.0047 and (pigmentation gz=2.0025 lw 7.5 (77 K) element) such that protein strands are severed. It can be argued that these strands are so long that movement of the sharp blades giso=2.005 Aiso=94.21 MnII d5–low spin between the ends of these protein strands is extremely unlikely (non-protein) gx=2.061 aHx=8.0 RMS• and thus the only alternative process is bond-homolysis to gy=2.026 aHy=8.0 (on hoof matrix fracture, give radical ends to the strands although some strands may gz=1.998 aHz=5.0 this work) be forced aside by the cutting blade without breaking.In the gx=2.061 aHx,y,z=8.5 RMS• cutting process it is necessary, however, to break the crossgy= 2.025 (human finger nails, links and the main protein backbone to sever the strands.The gz=2.000 ref. 6) SMS bonds are expected to break, as they are relatively weak, generating RS• radicals which were not detected in these aThe melanin signal gives diVerent line widths at room temperature experiments. However, an equal number of carbon-centred and 77 K.radicals, R• are also expected, and are also not detected in hoof horn. This is in contrast with the results for EPR spectra of crushed bone at 77 K.9,10 Because of the high concentration of RS•S(R)R and RS•S(R¾)R Symons11 has made a strong case that a stable radical centre, frequently encountered in organosulfur chemistry, has the general structure RS•S(R¾)R.However, at the time this structure was first proposed,12 Gordy and co-workers13 proposed that these radicals have the structure RSS•. The former structure is far more probable for the radicals formed herein, mainly because it is diYcult to envisage any mechanism for the formation of RSS• centres and this assignment is accepted herein.RS•SR2 radicals are one of a range of ‘three-electron’ radicals Fig. 2 First derivative X-band spectrum of white (unpigmented) often described as ‘s* radicals’ because the unpaired electron donkey hoof at 77 K; (a) uncut sample and (b) cut sample (instrument is in the SMS antibonding orbital. These bonds readily break settings: centre field 3381 G, sweep width 200 G, microwave frequency 9.495 GHz, microwave power 2 mW, modulation amplitude 1 G, time to give RS• radicals and R2S molecules.However, if this occurs constant 40 ms, sweep time 84 s) within the rigid hoof matrix, rapid return to give the s* adduct again is to be expected. The results from these experiments indicate that there are significant diVerences between the EPR signal at g=6 (high spin, low symmetry).The absence of this spectra of samples from the hooves of donkeys and horses, signal, and the appearance of its decay product at g=4.3, and also between pigmented and white hoof. The pigment, would suggest a history of bleeding into the hoof from the melanin, contains a relatively high concentration of occluded blood supply. Another possible explanation would be semiquinone-type radicals, but it is likely that the sulfur the presence in the hoof material of iron species which appear radicals are still formed on cutting, even though the EPR as uptake from soil.This, however, would most likely be spectrum is not detected directly. These radical centres are present in the form of oxides which are low spin FeIII species, well separated and cannot react together in the solid hoof leading to broad lines on the spectrum corresponding to the horn material.It is not possible to distinguish any spatial existence of clusters. Trace amounts of MnII were also observed. diVerences in the concentration of these radicals. All regions This is attributed to adsorption from the soil as the EPR are rich in keratin so this is not surprising.parameters (see Table 1) are more inorganically derived. As stressed above, deliberate cutting or microfracture occurs Symons et al.5 demonstrated that there is a signal due to the extensively, so there is probably always a low concentration formation of sulfur centred radicals when human fingernail is of these radical centres present. If they can be detected without cut and it is particularly interesting that there are no apparent deliberate cutting, the signals could be of use in providing a diVerences between the EPR spectra of cut and uncut pigmeasure of the extent of the matrix fracture.It is unlikely that mented donkey hoof.† On the other hand, cut non-pigmented these potentially active radicals present any problem to the samples of donkey hoof do give rise to EPR spectra showing hoof in themselves, because they are so firmly occluded within the presence of sulfur centred radicals in Fig. 2. The signals the material. were well defined for white hoof horn material from non- Although the microstructure of the hoof horn is not yet pigmented hooves, but their presence was concealed for clearly understood, at the molecular level the predominance strongly pigmented samples because of overlap by very intense of cysteine would suggest a three-dimensional ordered network signals from the melanin radicals.8 However, spectral subtracof –SMS– units such that fracture in the matrix of the fibres tion showed that the sulfur-centred species were still formed by cutting or breaking in any direction would always lead to after cutting.bond homolysis and formation of an RS•. Depending on the There is no obvious explanation for the diVerence in signals locus of such fractures, RS• formation may be important in from samples of pigmented and non-pigmented hoof. The role the understanding of processes within the hoof horn material. of melanin may be relevant, but the presence of free radicals Formation of free radicals initiates a host of biological reac- in hoof material can be detected.These results may prove to tions, often resulting in lipid peroxidation and carcinogenesis.14 be of more than academic interest as hoof horn material is Formation of RS• may be at, or close to, the coronary corium/ papillae or the laminar corium, all of which are richly supplied † Although Symons et al.did not refer to pigmentation phenotypes in by blood and nerve endings. In these tissues the RS•, though the sources of fingernail clippings, the element of pigmentation and its intensity would be expected to account for spectral diVerences. occluded in the rigid fibres, may involve formation of more 44 J. Mater. Chem., 1998, 8(1), 43–45reactive species like O2•- and RO2•, generally referred to as The Donkey Sanctuary, Sidmouth, Devon is thanked for the funding of a studentship for L.H. reactive oxygen species (ROS), which are initiators of lipid peroxidation and carcinogenesis, and may be involved in keratoma. Such an incident is a possibility during shoeing References when nails are driven into the hoof. A misplacement across such sensitive zones in the laminae would pose potential 1 J.D. Reilly, Equine Vet. J., 1995, 27, 166. dangers. 2 J. D. Reilly and S. A. Kempson, in Proc. of 7th Int. Symp. on Procedures of hoof resection and other forms of general Diseases of T he Ruminant Digit, ed. K. Mortensen, Denmark, 1992. hoof surgery make significant cuts in the horn and it would 3 C. C. Pollitt, Equine Vet.Ed., 1992, 4, 219. 4 A. Alonso, N. C. Meirelles, V. E. Yushmanov and M. Tarbak, be of great importance to note the formation of RS• in the horn. J. Invest. Dermatol. 1996, 106, 1058. The line width of the melanin signal varies with temperature. 5 H. Chandra and M. C. R. Symons, Nature, 1987, 328, 833. The increase in line width with temperature suggests an 6 M. C. R. Symons, H.Chandra and J. LWyatt, Radiation Protection increase in the rate of tumbling of the melanin centres in a Dosimetry, 1995, 58, 11. matrix which undergoes a phase transition. As the matrix 7 J. D. Reilly, D. F. Cottrell, R. J. Martin and D. Cuddeford, becomes more fluid the mobility of the melanin increases and Biomimetics, 1996, 4, 23 8 B. Collins, T. O. Poehler and W. A. Bryden, Photochem.Photobiol., it tumbles faster, leading to a narrower line. Similar obser- 1995, 62, 557. vations have been made for lipid–protein interactions in which 9 R. Partridge, M. C. R. Symons and J. Wyatt, J. Chem. Soc., the otherwise rigid lipid membranes become fluid with increas- Faraday T rans., 1993, 89, 1285. ing temperature, aVording increased lipid–protein interaction 10 M. C. R. Symons, Free Radicals Biol. Med., 1996, 20, 831. as reported in spin labelling studies15 and hydration induced 11 D. J. Nelson, R. L. Petersen and M. C. R. Symons J. Chem. Soc., fluidity in stratum corneum,4 for example. Perkin T rans. 2, 1977, 225. 12 M. C. R. Symons, J. Chem. Soc., Perkin T rans. 2, 1974, 168. 13 J. H. Hadley and W. Gordy, Proc. Natl. Acad. Sci. USA, 1974, 72, 3106. Conclusions 14 B. Halliwell and J. M. C. Gutteridge, Mol. Aspects Med., 1985, EPR spectroscopy can provide useful information on a complex 8, 89. 15 F. Severcan and S. Cannistraro, Chem. Phys. L ipids, 1990, 53, 17. natural material such as hoof horn. Cutting or fracture of hoof horn may produce free radicals which are similar to those detected when human fingernail is cut. Paper 7/06199D; Received 26th August, 1997 J. Mater. Chem., 1998, 8(1), 43–45 45

ISSN:0959-9428    

DOI:10.1039/a706199d

出版商:RSC    

年代:1998

数据来源: RSC