Aerosol assisted chemical vapour deposition of Cu

Download PDF
4 downloads 0 Views 447KB Size Report
Comment
function of temperature, EDX for composition, SEM for surface morphology and XRD for crystalline phases. ..... data and structure refinement parameters are listed in Table 1. ..... perhaps related to the adjacency of Cu and Zn in the Periodic.
PAPER

www.rsc.org/dalton | Dalton Transactions

Aerosol assisted chemical vapour deposition of Cu–ZnO composite from single source precursors† Muhammad Shahid,a Muhammad Mazhar,*a Mazhar Hamid,a Paul O’Brien,*b Mohammad A. Malik,b Madeleine Helliwellb and James Rafteryb Received 18th February 2009, Accepted 17th April 2009 First published as an Advance Article on the web 28th May 2009 DOI: 10.1039/b903406d Two heterobimetallic precursors [Zn(TFA)3 (m-OH)Cu3 (dmae)3 Cl]·THF (1) and [Zn(TFA)4 Cu3 (dmae)4 ] (2) [dmae = N,N-dimethylaminoethanolate and TFA = trifluoroacetate], have been synthesized and characterized by their melting points, elemental analysis, FT-IR spectroscopy, mass spectrometry, TGA and single crystal X-ray diffraction methods. Both complexes were used to deposit thin films of Cu–ZnO composite on glass substrates by aerosol assisted chemical vapor deposition (AACVD) method. The films were characterized by “scotch tape” test for adhesion, thickness measurement as a function of temperature, EDX for composition, SEM for surface morphology and XRD for crystalline phases. Thin film deposition studies at 250, 325, 400, 475 ◦ C indicated the increase in thickness with temperature reaching a maximum at 400 ◦ C and then decreasing. EDX and PXRD results showed the uniform distribution of cubic metallic copper and hexagonal zinc oxide phases which make them useful for nanocatalysis on structured surfaces.

Introduction The catalytic synthesis of methanol from CO/CO2 /H2 and oxidation of CO to CO2 are important well-known industrial and technological areas of current interest.1,2 The conversion of carbon monoxide to carbon dioxide at ambient conditions is important for respiratory protections, mining industries, deep sea diving and space exploration missions.3,4 The mixed metal/metal-oxide composite, Cu/ZnO, is well reported as the most widely used catalyst for the large scale industrial production of methanol5 and oxidation of carbon monoxide.6,7 The activity of the catalyst is mainly accounted for by high surface area and homogenous dispersion of copper on the zinc oxide phase for methanol synthesis, water-gas shift reactions and various hydrogenations.8–10 Further, the catalytic efficiency of the Cu/ZnO nanocomposite having the desired application area varies with surface structures and morphologies i.e. thin films,11 nanowires,12 nanoparticles etc.13 The industrial catalysts are generally prepared by empirically well-developed aqueous co-precipitation techniques. Metal carbonates are precipitated out by the addition of alkali metal carbonates in respective metal salt solutions (nitrates, acetates) resulting in the formation of malachite Cu2 (CO3 )(OH)2 , aurichalcite (Cu,Zn)5 (CO3 )2 (OH)6 or hydrozincite Zn5 (CO3 )2 (OH)6 as well as other related phases having various ratios of copper and zinc.14 The product obtained after drying, aging, and calcinations can

a Department of Chemistry, Quaid-I-Azam University, Islamabad, 45320, Pakistan. E-mail: [email protected]; Fax: +92(0)5190642241; Tel: +92(0)5190642106 b The School of Chemistry and School of Materials, The University of Manchester, Oxford Road, Manchester, UK M13 9PL. E-mail: [email protected] † CCDC reference numbers 633129 and 662553 for compounds (1) and (2). For crystallographic data in CIF or other electronic format see DOI: 10.1039/b903406d

This journal is © The Royal Society of Chemistry 2009

be adjusted for its particle size and crystallinity by the pH, concentration, stirring rate, duration of precipitation and aging, and the calcination conditions.15 However, these inorganic solidstate precursors are neither soluble nor volatile. This limits their application as precursors for the impregnation of highly porous supports based on alumina or silica. Further, in conventional syntheses, a relatively higher temperature is usually required to achieve a sufficiently fast diffusion. It is reported16,17 that the composition as well as the crystal structure of the bimetallic composites can be fine-tuned by the thermolysis conditions such as time, temperature and the atmosphere. The controlled thermolysis of the precursor can be carried out at comparatively low temperature if suitable organic leaving groups such as carboxylates and amino alcohols are present. Such systems prove to be suitable precursors for the preparation of bimetallic nanocomposites that incorporate the initially blended metals. The synthesis and the utility of new heterometallic coordination complexes as single molecule precursors (SMPs) for the synthesis of mixed-metal oxides have been continuously investigated over the years.18–20 This is particularly adapted for sol–gel processing21 as well as for the growth of metal oxide thin film materials by chemical vapor deposition techniques.22,23 However, initially particular emphasis remained around the single-phased metal oxide materials. Further progress in the design of single source precursors to produce single step multiphased composite materials opened the gateway for the synthesis of particular precursors.24 In heterogeneous catalysis, a finely dispersed nanoscale metallic phase bound to a metal oxide support is found to possess superior properties25 and tailored single molecular precursors for thin films may prove to be an interesting alternative approach for catalyst preparation. Therefore, we were attracted to investigate the soluble and volatile aerosol assisted chemical vapor deposition (AACVD) single molecular precursors for the deposition of Cu–ZnO Dalton Trans., 2009, 5487–5494 | 5487

composite thin films at different temperatures for their possible use as successful catalyst systems.

Experimental All manipulations were carried out under an atmosphere of dry argon using Schlenk tube and glove box techniques. The solvents and reagents were purchased from Aldrich. Solvents were rigorously dried on sodium benzophenoate, while N,N-dimethylaminoethanol (dmaeH) was dried by refluxing over K2 CO3 for 10 h and distilled immediately before use. Cu(dmae)2 and [Cu(OCH3 )Cl]4 were prepared by literature procedures.26,27 Elemental analyses were performed using a CHN Analyzer LECO model CHNS-932. Melting points were determined in a capillary tube using an electrothermal melting point apparatus, model MP.D Mitamura Riken Kogyo (Japan) and are uncorrected. FT-IR spectra were recorded on a single reflectance ATR instrument (4000–400 cm-1 , resolution 4 cm-1 ). Mass spectra were recorded on a Kratos concept IS instrument. Single crystal X-ray data were collected on a Bruker Smart Apex CCD diffractometer at 100(2) K using monochromatic MoKa radiation. Data were reduced using SAINTPLUS28 and the structures solved and refined with the SHELXTL29 suite. All and only non-hydrogen atoms were modeled with anisotropic ADPs. Hydrogens were placed in calculated positions and refined with 1.5 (methyl) or 1.2 (methylene) Ueq of the parent atom, with the exception of the hydroxyl hydrogen in (1) which was located from the difference map. TGA measurements were carried out by using a Seiko SSC/S200 thermal analyzer at a heating rate of 10 ◦ C/min under N2 gas flow. Powder X-ray diffraction (PXRD) measurements were carried out by means of a Bruker AXS D8 diffractometer using monochromatic Cu-Ka radiation. The morphology of films was determined by a FEG-SEM Philips XL30 electron microscope. Samples were carbon coated before observation and EDAX-DX4 was used to calculate the composition (metallic ratio) of films. The thickness of films was measured with a Dektak 8 Stylus profilometer. The current–voltage characteristics of thin films were measured with Jandel Voltmeter, Model RM3, by the four-probe method. Synthesis of [Zn(TFA)3 (l-OH)Cu3 (dmae)3 Cl]·THF (1) 0.6 g (2.06 mmol) of zinc trifluoroacetate [Zn(TFA)2 ·xH2 O] was added to a solution of 0.56 g (2.99 mmol) of [Cu(dmae)Cl]4 in 10 ml of toluene at room temperature. After stirring for 3 h, unreacted Zn(TFA)2 ·xH2 O and precipitated ZnCl2 were eliminated by cannula filtration. The reaction mixture was evaporated to dryness under vacuum to give a dry powder of (1) which was crystallised from THF to give dark blue block-like crystals with 75% yield after 10 days at -10 ◦ C, m.p. 160 ◦ C. Anal. Calc. For C22 H39 ClCu3 F9 N3 O11 Zn: C, 26.50; H, 3.84; N, 4.21. Found: C, 26.23; H, 3.75; N, 4.05%. FT-IR (KBr, cm-1 ): 3406 br, 2989 w, 2864 m, 1678 vs, 1463 m, 1435 m, 1383 w, 1204 vs, 1185 s, 1150 s, 1066 s, 1014 m, 951 m, 895 m, 840 m, 795 m, 725 s, 637 w, 588 w, 522 m, 486 w, 430 w. MALDIMS (positive mode) m/z: 901 [Cu3 (dmae)3 (TFA)(THF)]+ , 822 [Cu3 (dmae)2 (O)(TFA)3 ZnCl]+ , 783 [M - (dmae)(TFA)]+ , 759 [Cu2 (dmae)2 (OH)(TFA)3 ZnCl]+ , 681 [Cu(dmae)2 (TFA)3 ZnCl]+ , 5488 | Dalton Trans., 2009, 5487–5494

593 [Cu(dmae)(TFA)3 ZnCl]+ , 550 [Cu2 (O)Zn(TFA)3 ]+ , 523 [Cu2 (dmae)(O)Zn(TFA)2 ]+ , 478 [Cu2 (dmae)(TFA)2 Cl]+ , 442 [Cu2 (dmae)(TFA)2 ]+ , 389 [Cu2 (TFA)2 Cl]+ , 381 [Zn(dmae)3 (O)Cl]+ . TGA: 25–85 ◦ C (3.70 wt% loss), 85–183 ◦ C (7.45 wt% loss), 183– 393 ◦ C (26.9% residue mass).

Synthesis of [Zn(TFA)4 Cu3 (dmae)4 ] (2) A solution of 1.20 g (5.01 mmol) Cu(dmae)2 in 15 ml toluene was transferred to a suspension of 1.00 g (3.43 mmol) Zn(TFA)2 ·xH2 O in 10 ml warm toluene. The reaction mixture was stirred at about 50 ◦ C for 3 h and allowed to cool. The excess of Zn(TFA)2 ·xH2 O was removed by filtration through a cannula, followed by vacuum evaporation to dryness. The solid was redissolved in THF to crystallise. Blue cubic crystals suitable for single crystal X-ray analysis were harvested from solution at -10 ◦ C with a 75% yield after two weeks, m.p. 125 ◦ C Anal. Calc. For C24 H40 Cu3 F12 N4 O12 Zn: C, 27.18; H, 3.80; N, 5.28; Zn, 6.16 Found: C, 26.89; H, 3.65; N, 5.04; Zn, 6.35%. FT-IR(KBr, cm-1 ): 2977 w, 2886 m, 1694 vs, 1465 m, 1204 vs, 1183 s, 1140 s, 1067 m, 1015 m, 950 m, 897 w, 843 m, 796 m, 726 s, 643 w, 522 w, 489 w, 433 w. APCI-MS (positive scan) m/z: 1059.4 [M]+ , 946.4 [Cu3 (dmae)4 Zn(TFA)3 ]+ , 885.2 [Cu2 (dmae)4 Zn(TFA)3 ]+ , 682.8 [Cu2 (dmae)3 Zn(TFA)2 ]+ , 551.4 [Cu2 (dmae)(TFA)3 ]+ , 459.8 [Cu2 (dmae)3 Zn]+ , 417 [Cu2 (dmae)2 (TFA)]+ , 303.5 [Cu2 (dmae)2 ]+ , 241 [Zn(dmae)2 ]+ , 238.4 [Cu(dmae)2 ]+ , 149.5 [Cu(dmae)]+ . TGA: 26–86 ◦ C (3.70 wt% loss), 86–182 ◦ C (7.45 wt% loss), 182–393 ◦ C (residue of 26.9%).

Deposition of thin films Thin films were deposited on soda glass substrates in a hot walled reactor by gas phase reactions of the precursors in an argon environment using a self-designed AACVD assembly described elsewhere.30 In a typical experiment, 0.1 g of precursor dissolved in 15 ml THF in a two neck round bottomed flask was connected via rubber tubing to a quartz reactor loaded with 2.5 ¥ 1 cm glass substrates inside a carbolite tube furnace and a flow of argon gas was regulated using a Platon flow gauge on the other neck. The flask was fitted on an ultrasonic humidifier equipped with a piezoelectric modulator for atomization of the precursor solution to tiny droplets of aerosol that were ultimately transferred by the carrier gas into the reactor chamber. The substrates were washed with concentrated nitric acid, followed by washings with deionized water several times and then oven dried at 100 ◦ C. Deposition was carried out at four different temperatures i.e., 250, 325, 400 and 475 ◦ C at a constant argon flow rate of 130 ml/min on substrates for 2.5 h.

Results and discussion Zn(TFA)2 ·xH2 O reacts with [Cu(dmae)Cl]4 and Cu(dmae)2 in toluene to give blue coloured heterometallic tetranuclear complexes, Zn(TFA)3 (m-OH)Cu3 (dmae)3 Cl·THF (1) and Zn(TFA)4 Cu3 (dmae) (2), respectively (eqn (1) and eqn (2)). The progress of the reaction was monitored through dissolution of Zn(TFA)2 ·xH2 O in the solvent mixture. This journal is © The Royal Society of Chemistry 2009

Toluene 2Zn(TFA)2 ◊x H 2O + 3/4[Cu(dmae)Cl]4 æCr æææ Æ Zn(TFA) 3 yst THF

( µ -OH)Cu 3 (dmae ) 3 Cl◊THF + ZnCl 2 + CF3COOH (1)

2Zn(TFA)2 ◊ x H 2O + 3Cu(dmae)2 æToluene æææ Æ Zn(TFA)4Cu 3 (dmae )4 + Zn (dmae)2 (2) Both complexes were characterized by melting point, elemental analysis, FT-IR, mass spectrometry and single-crystal X-ray analysis. The FT-IR spectra of (1) and (2) account for the presence of trifluoroacetate ligands in the complexes. A very strong absorption band at 1678 cm-1 corresponding to n asy (CO2 ) suggests the presence of bridging TFA ligands for (1), while that at 1694 cm-1 for (2) indicates monodentate ones. In the case of complex (1), the broad band at 3406 cm-1 is due to absorption by the O–H group which is absent in the spectrum for (2) due to the absence of any O–H group in the complex. The spectra of both complexes also exhibits three strong C–F and C–O absorption bands in the range 1120– 1210 cm-1 corresponding to TFA ligands.31 In the MALDI mass spectrum of (1), the molecular ion peak is absent whereas the APCI-MS (positive scan) of (2) exhibits a molecular ion peak at m/z 1058.4. The incomplete cube-like core M4 (m-O)4 of complex (1) breaks down under MS conditions and gives ionic fragments with different m/z ratios. Most dominant fragments include m/z (901) [Cu3 (dmae)3 (TFA)3 Cl(THF)]+ and the base peak at m/z (442) corresponding to [Cu2 (dmae)(TFA)2 ]+ . In the case of (2), the tetranuclear complex was further confirmed by the presence of various other ions derived from the fragmentation of the complex, with the base peak at m/z (417) [Cu2 (dmae)2 (TFA)]+ . Structural analysis of [Zn(TFA)3 (l-OH)Cu3 (dmae)3 Cl]·THF (1) The X-ray structure of (1) is shown in Fig. 1 and selected bond distances and angles are given in Table 2, while the crystal

Fig. 1 ORTEP drawing showing the molecular structure of [Zn(TFA)3 (m-OH)Cu3 (dmae)3 Cl]·THF (1). THF molecule and weak Cu–Cu interactions are omitted for clarity. Ellipsoids are drawn at 40% probability.

data and structure refinement parameters are listed in Table 1. The complex crystallizes in the monoclinic space group C2/c. ˚ , X-ray Considering weak copper–oxygen interactions up to 2.8 A crystallographic analysis reveals the asymmetric Cu3 ZnO4 core representing a structure that can be described as an incomplete cube, missing one bond, with metal and oxygen atoms on alternate positions. The crystallographic study of this complex demonstrates that its tetranuclear skeleton has no crystallographically imposed symmetry. Each discrete molecule in the unit cell contains one noncoordinating THF solvate. The m-dmae, m-TFA and triply bridged hydroxyl groups link the metal centers resulting in copper–copper ˚ and and copper–zinc separations in the range 3.171(1)–3.266(1) A ˚ , respectively. All three TFA as well as dmae 3.061(1)–3.147(1) A

Table 1 Crystal data and refinement parameters for the complex [Zn(TFA)3 (m-OH)Cu3 (dmae)3 Cl]·THF (1) and [Zn(TFA)4 Cu3 (dmae)4 ] (2) Empirical formula Formula weight Temperature (T) ˚) Wavelength (A Crystal system Space group ˚) Unit cell dimensions (A (◦ ) ˚ 3) Volume (A Z Density (calculated) (Mg/m3 ) Absorption coefficient (mm-1 ) F(000) Crystal size (mm) q range for data collection (◦ ) Index ranges Reflections collected Independent reflections Refinement method Goodness-of-fit on F 2 Final R indices [I>2s (I)] R indices (all data) ˚ -3 ) Largest diff. peak and hole (e ¥ A

C22 H39 ClCu3 F9 N3 O11 Zn 984.00 100(2) 0.71073 Monoclinic C2/c a = 16.423(5), b = 12.205(4), c = 35.531(11) a = g = 90, b = 94.755(6) 7097(4) 8 1.842 2.619 3960 0.50¥ 0.40 ¥ 0.40 2.08 to 26.39 -12 ≤ h ≤20, -14 ≤ k ≤ 15, -44 ≤ l ≤ 41 19510 7223 [R(int) = 0.0993] Full-matrix least-squares on F 2 1.046 R1 = 0.0403, wR2 = 0.1003 R1 = 0.0536, wR2 = 0.1044 1.061 and -0.633

This journal is © The Royal Society of Chemistry 2009

C24 H40 Cu3 F12 N4 O12 Zn 1060.59 100(2) 0.71073 Monoclinic P21 /c a = 13.9527(14), b = 20.757(2), c = 14.0478(14) a = g = 90, b = 104.623(2) 3936.7(7) 4 1.789 2.315 2132 0.35 ¥ 0.30 ¥ 0.24 1.79 to 28.31 -18 ≤ h ≤ 18, -27 ≤ k ≤ 27, -18 ≤ l ≤ 18 33326 9308 [R(int) = 0.0883] Full-matrix least-squares on F 2 1.174 R1 = 0.0603, wR2 = 0.1237 R1 = 0.0757, wR2 = 0.1285 1.403 and -0.673

Dalton Trans., 2009, 5487–5494 | 5489

Table 2 Selected metrical data for complex (1)

Structural analysis of [Zn(TFA)4 Cu3 (dmae)4 ] (2)

˚) Bond distances (A

Molecular structure with atomic labeling of the complex (2) is displayed in Fig. 2. Crystal data and refinement parameters are tabulated in Table 1. One dmae ligand was disordered over two positions; the major conformation (65%) will be used in discussing the structure. Selected bond lengths and angles are listed in Table 3. The complex crystallizes in the monoclinic space group P21 /c. Counting Cu–O weak interactions, (2) exhibits a cubelike central core [M4 O4 ] with metal and oxygen atoms occupying alternate corners of the cube (Fig. 2). As for complex (1), all the trifluoroacetates behave in a similar way but coordinate in a monodentate manner with weak interactions between noncoordinating TFA oxygen and neighbouring copper atoms. In addition, the four dmae ligands coordinate in a similar fashion as that found for complex (1) with their oxygen atoms bridging two metal centers. Again, m2 -oxygen atoms of dmae ligands are loosely bound by the third nearest copper candidate.

Cu(1)–O(7) Cu(1)–O(2) Cu(1)–O(10) Cu(1)–Cu(2) Cu(1)–Cu(3) Cu(2)–O(8)

1.968(2) 2.380(2) 2.711(3) 3.266(1) 3.171(1) 2.779(3)

Cu(2)–O(4) Zn(1)–O(10) Zn(1)–O(8) Zn(1)–O(9) Zn(1)–Cu(1) Zn(1)–Cu(3)

171.90(11) 86.15(10) 84.53(9) 101.58(10) 100.84(11) 160.35(10)

O(3)–Cu(2)–O(4) O(9)–Cu(2)–O(4) O(8)–Zn(1)–O(9) O(10)–Zn(1)–Cl(1) O(8)–Zn(1)–Cl(1) O(9)–Zn(1)–Cl(1)

2.315(3) 1.967(2) 1.968(2) 1.969(2) 3.061(1) 3.147(1)

Bond angles (◦ ) O(1)–Cu(1)–N(1) O(8)–Cu(1)–O(2) O(1)–Cu(1)–O(2) O(7)–Cu(1)–O(2) N(1)–Cu(1)–O(2) O(3)–Cu(2)–O(9)

103.03(10) 96.33(10) 97.35(10) 115.89(7) 121.26(7) 120.63(7)

ligands behave in a similar way. The TFA groups are bridging two copper metal atoms, while the dmae ligands behave as chelating– bridging with m2 -oxygen atoms connecting zinc with copper metal atoms. The three copper atoms are also almost identical: each has a CuO3 N coordination sphere comprising one O,N-chelating dmae ligand, two bridging trifluoroacetate groups and one triply bridged hydroxyl group. So, the geometry of Cu(1) can be described as slightly distorted square pyramidal, with normal Cu–O/N bonds in the basal plane. The bond distance of copper to the ˚ ] is much greater in value than apical oxygen O(2) [2.380(2) A ˚ but comparable to the the sum of the ionic radii of 1.92 A 32 values found in similar compounds. All the bond angles between two consecutive coordinating atoms in the basal plane of square pyramidal geometry around Cu(1) are close to 90◦ , while those with the capping oxygen, O(2), lie in range 84.53(9)–101.58(10)◦ . In addition, Cu(1) is weakly bonded to the oxygen atom O(10) at ˚ that gives a distorted octahedral coordination sphere 2.711(3) A for Cu(1). The geometries about Cu(2) and Cu(3) are also distorted square pyramidal, with angles in the basal plane which are fairly close to 90 and 180◦ . The capping oxygen atoms, O(4) and O(6), ˚ , respectively, to Cu(2) are at distances of 2.315(3) and 2.282(2) A ˚ and Cu(3) and the atoms O(8) and O(9) are 2.779(3) and 2.823(3) A from Cu(2) and Cu(3), giving a pseudo distorted octahedral environment. The angles from the atoms in the basal planes to the capping oxygens range from 85.50(9) to 103.0(1)◦ and 87.7(1) to 98.4(1)◦ for Cu(2) and Cu(3), respectively. The Zn(1) centre is coordinated in a distorted tetrahedral fashion by three oxygen and one chloride atoms. The core unit ZnO3 Cl consists of O(8), O(9) and O(10) from three bridging dmae ligands and a terminal chloride. The bond lengths Zn(1)– O(8), Zn(1)–O(9) and Zn(1)–O(10) with 1.968(2), 1.969(2) and ˚ , respectively are normal and lie in the range observed 1.967(2) A for similar compounds.33,34 The bond angles about Zn(1) that range from 97.35(10) to 121.26(7)◦ deviate from the ideal 109.5◦ but are in accordance with the literature.26 Intermolecular hydrogen bonding [O(1)–H(1O) ◊ ◊ ◊ O(11) (symmetry equivalent x - 1/2, y - 1/2, z)] is present between the hydrogen atom [H(1O)] of the bridged hydroxyl group in complex (1) with oxygen atom [O(11) (symmetry equivalent x - 1/2, y - 1/2, z)] of THF solvate. 5490 | Dalton Trans., 2009, 5487–5494

Fig. 2 ORTEP drawing showing the molecular structure of [Zn(TFA)4 Cu3 (dmae)4 ] (2). The weak Cu–O interactions are omitted for clarity. All ellipsoids are drawn on 40% probability level. (b) Core unit of the complex with the coordination arrangement around the metallic centers; atoms are drawn at 30% probability level.

This journal is © The Royal Society of Chemistry 2009

Table 3 Selected metrical data for complex (2) ˚) Bond distances (A Cu(1)Zn(1)–O(9) Cu(1)Zn(1)–N(1) Cu(1)Zn(1)–Cu(2)Zn(2) Cu(1)Zn(1)–Cu(4)Zn(4) Cu(1)Zn(1)–O(1) Cu(1)Zn(1)–O(11)

1.952(2) 2.046(2) 3.130(1) 3.225(1) 2.560(2) 2.756(3)

Cu(2)Zn(2)–O(2) Cu(2)Zn(2)–N(2) Cu(3)Zn(3)–O(8) Cu(3)Zn(3)–N(3) Cu(4)Zn(4)–O(5) Cu(4)Zn(4)–N(4)

1.946(2) 2.046(2) 1.948(2) 2.042(3) 1.957(2) 2.053(2)

Bond angles (◦ ) O(7)–Cu(1)Zn(1)–O(12) 90.44(7) O(7)–Cu(2)Zn(2)–O(2) 94.73(8) O(7)–Cu(1)Zn(1)–O(9) 169.27(8) O(11)–Cu(3)Zn(3)–O(4) 169.35(11) O(12)–Cu(1)Zn(1)–O(9) 95.94(8) O(11)–Cu(3)Zn(3)–O(8) 89.66(9) O(7)–Cu(1)Zn(1)–N(1) 85.57(8) O(4)–Cu(3)Zn(3)–O(8) 96.21(8) O(12)–Cu(1)Zn(1)–N(1) 174.01(8) O(11)–Cu(3)Zn(3)–N(3) 88.08(12) O(9)–Cu(1)Zn(1)–N(1) 88.68(8) O(11)–Cu(4)Zn(4)–O(5) 95.42(11) O(8)–Cu(2)Zn(2)–O(2) 172.64(9) O(11)–Cu(4)Zn(4)–N(4) 173.30(10)

All the four metal sites in the complex are partially occupied by copper and partially by zinc, with an approximate occupancy ratio of 3:1 (Cu:Zn). Occupancy refinement proved to be unstable, perhaps related to the adjacency of Cu and Zn in the Periodic Table, and as a consequence all the metal sites were assigned the 3:1 Cu:Zn ratio consistent with the elemental analysis and consonant with the very similar coordination environment of the metal sites. Eliminating weak interactions, the overall geometry around each metal centre can best be described as slightly distorted square planar. So the Cu1 Zn1 metal site is bound by three oxygens O(7), O(9), O(12) of which O(7) connects Cu1 Zn1 to Cu2 Zn2 with ˚ and O(12) acts as a bridge between a separation of 3.130(1) A ˚ and the Cu1 Zn1 and Cu4 Zn4 with a separation of 3.226(1) A nitrogen atom N(1). The Cu1 Zn1 –O/N bond distances are similar to Cu–O/N lengths observed in earlier published compounds.26 Also the bond angles between cis coordinating donor atoms, falling in the range 85.57(8)–95.94(8)◦ with an average of 90.16◦ , are unexceptionable. The coordination environment of Cu1 Zn1 is augmented by two weak Cu1 Zn1 –O interactions [with O(1) ˚ ] and can be described as a 2.562(3) and O(11) 2.789(3) A tetragonally elongated octahedron.

Fig. 3 Thermogram of the complex (1) and complex (2) showing the weight loss with increase in temperature.

Further temperature increase does not affect the stability of the proposed composite, Cu3 ZnO. Complex (2) is pyrolysed with only one mentionable decomposition step at 216 ◦ C which ceased at 260 ◦ C. This behavior indicates the instantaneous decomposition of the coordination complex to a stable oxidic species with the removal of all the organic part. The residual amount of 25.31% is obtained which is slightly less than but in good agreement with the calculated value (25.6%) of the accumulative sum of the proposed composite Cu3 ZnO, as in the case of precursor (1). Thin film study A thin film deposition study was performed at different temperatures i.e., 250, 325, 400, and 475 ◦ C utilizing both the precursors (1) and (2) to investigate the effect of temperature on film thickness, crystallite sizes, conductivity and phase stability. In both cases, the quality of the deposited films is good with smooth, compact and uniform surface morphology and this behavior is observed to be more dominant with an increase in temperature. They qualify for the “scotch tape” test for surface adhesion. In the case of film deposited from precursor (1), electron micrographs (Fig. 4) show almost spherical sized particles with low

Thermogravimetric studies Thermal decomposition behavior of the complexes was studied using a Seiko SSC/S200 thermal analyzer at a heating rate of 10 ◦ C/min, in a N2 atmosphere. Thermolysis patterns of complex (1) (Fig. 3) indicate three steps weight losses with two minor and one major downfall in the thermogram. During the first step at 152 ◦ C and second step at 190 ◦ C, a relatively small fraction of weight loss indicates the most probable removal of smaller fragments. The major degradation dip in the curve is observed at 232 ◦ C for the complete removal of the organic part. These laddered thermolysis patterns point towards the stepwise breakdown of the complex rather than simultaneous decomposition. The weight loss ceased at about 287 ◦ C resulting in the formation of a stable residue of 27.76% that is in good agreement with the calculated amount of 27.7% for Cu3 ZnO based upon the EDX metal ratio and PXRD phase identification results. This journal is © The Royal Society of Chemistry 2009

Fig. 4 SEM image of thin films deposited from precursor (1) at reactor temperatures of (a) 250 ◦ C, (b) 325 ◦ C, (c) 400 ◦ C, and (d) 475 ◦ C.

Dalton Trans., 2009, 5487–5494 | 5491

crystallinity at 250 ◦ C. The particles seem to grow with the increase of temperature without any change in shape or morphology, and the crystallinity of the film is improved greatly as shown by powder X-ray diffractograms. However, the particle size distribution seems to be narrow for each individual sample. At lower temperature i.e., 250 ◦ C, particles of size 0.2 mm are found to be deposited that grow to 0.57, 1.15 and 1.4 mm at 325, 400 and 475 ◦ C respective temperatures. In the case of films from precursor (2), the electron micrographic study (Fig. 5) shows that film deposition and particle growth patterns are similar to that in the first case, with minor morphological differences. Initially at 250 ◦ C (Fig. 5a), particle growth is very little, without clear boundaries and with a poor crystalline nature. At that stage, particles size seem to be roughly 0.06–0.09 mm in diameter. At the relatively higher temperature i.e., 325 ◦ C, particles grow to sizes 0.26 mm with clear individual boundaries, in the form of clusters of grains. Further, with the increase of film deposition temperatures to 400 and 475 ◦ C (c & d), grains grow to 0.47–0.55 mm and seem to be more clear and segregated without any appreciable difference at these two temperatures. It is observed from SEM, Fig. 6(b), that some large particles seem to fragment and shatter into daughter grains that grow further at higher temperature i.e., 400 ◦ C. The overall EDX analysis indicated that the metallic ratio of Zn:Cu in the films deposited from both precursors is close to 1:3, which also confirms the retention of the same metallic ratio as that in the complexes.

Fig. 6 Powder X-ray patterns of the films deposited from (1) at temperatures (a) 250 ◦ C, (b) 325 ◦ C, (c) 400 ◦ C, and (d) 475 ◦ C. x = ZnO [01–076–0704], and y = Cu [01–085–1326].

Fig. 7 Powder X-ray patterns of the films deposited from (2) at temperatures (a) 250◦ C, (b) 325◦ C, (c) 400◦ C, and (d) 475◦ C. x = ZnO [00–079–0206], and y = Cu [01–085–1326].

Zn(TFA) 3 (µ -OH)Cu 3 (dmae) 3Cl ◊ THF æAACVD æææ Æ 3Cu + ZnO + Volatiles Zn(TFA)4Cu 3 (dmae)4 æAACVD æææ Æ 3Cu + ZnO + Volatiles Fig. 5 SEM image of thin films deposited from precursor (2) at reactor temperatures of (a) 250 ◦ C, (b) 325 ◦ C, (c) 400 ◦ C, and (d) 475 ◦ C.

X-ray diffraction studies of the films deposited at four different temperatures were performed to identify the stable crystalline phases under CVD experimental conditions. It was observed in both cases (Fig. 6 & Fig. 7) that at lower temperature (250 ◦ C), the film is almost amorphous or poorly crystalline with undeveloped copper phase peaks at characteristic 2q values of 43.32◦ , 50.45◦ , 74.13◦ and 89.94◦ .35 At the augmented temperature, the height of the peaks increases and width decreases indicating the better crystallinity with increased crystallite size. The formation of two stable well-developed crystalline hexagonal ZnO36,37 and cubic Cu35 phases at 450 ◦ C is shown in eqn (3) and eqn (4) for both the precursors (1) and (2), respectively. 5492 | Dalton Trans., 2009, 5487–5494

(3)

(4)

XRD data of complex (2) (Fig. 7) also show that the ZnO phase is more dominant at 325 ◦ C and with the increase of temperature, the Cu phase seems to be more crystalline as compared to ZnO. The results concluded from the intensity and peak broadening at 475 ◦ C are evident of the crystallites growth, which are in good agreement with SEM results. The thickness of the films deposited at various temperatures was measured and found to be increased with a maximum at 400 ◦ C in both cases (Fig. 8). Initially at 250 ◦ C, surface decomposition reactions of the precursors take place, which lead to nucleation and growth of crystallites at higher temperatures i.e., 325 and 400 ◦ C. As the microstructure is dependent upon the film thickness, observations recorded here are in accordance with SEM and XRD results.38 Further, with the increase in temperature beyond the 400 ◦ C, film thickness decreases due to limited exposure of the substrate to the precursor and the increased desorption rate. This journal is © The Royal Society of Chemistry 2009

Table 4 Numerical data for the thickness and electrical properties of film deposited at different temperatures for both the precursors Deposition Film thickness temp. (◦ C) (mm)

Fig. 8 Plot showing the variation of film thickness with substrate temperature for (a) precursor (1) and (b) precursor (2).

Electrical characterization of the films deposited from precursors (1) and (2) were studied from the voltage–current curves obtained for the films deposited at 400 and 475 ◦ C and are presented in Fig. 9 and Fig. 10, respectively. Linear conduction behavior shown by both the films observed at these temperatures indicated their ohmic conduction nature. Such conduction behaviour was

Fig. 9 V –I curves of Cu–ZnO films deposited from precursor (1) at 400 and 475 ◦ C.

Resistance (X) Resistivity (X cm)

Precursor (1) 250 325 400 475

0.90 3.41 5.65 3.11

— — 15 ¥ 10-3 10 ¥ 10-3

— — 3.8 ¥ 10-5 1.4 ¥ 10-5

Precursor (2) 250 325 400 475

1.23 3.95 7.38 3.11

— — 12.5 ¥ 10-3 7.5 ¥ 10-3

— — 4.2 ¥ 10-5 1.1 ¥ 10-5

not shown by the films deposited at lower temperatures. From the slope of the curves (V /I) and thickness (d) of films, resistivity was calculated using the relationship as r = 4.532 (V /I) ¥ d, where 4.532 is an instrumental constant and the results are summarized in Table 4.

Conclusion Two new heterobimetallic complexes [Zn(TFA)3 (m-OH) Cu3 (dmae)3 Cl]·THF (1), [Zn(TFA)4 Cu3 (dmae)4 ] (2) were synthesized by a simple one step reaction and were fully characterized including single X-ray crystallography. Both complexes deposited crystalline thin films of Cu–ZnO composite by the AACVD method. The morphology and the thickness of the films including the shape and size of the crystallites changed with change in the deposition temperature. The crystallinity and the particle sizes are found to increase with increasing temperature, while film thickness increases to a certain maxima and then decreases at the saturation point. Complex (2) is comparatively a better precursor for deposition of Cu–ZnO composite due to its relatively low decomposition temperature and better crystallinity with small particle sizes. The films deposited are also found to be ohmic conductors as indicated by V –I measurements. The detailed characterization of the films provides useful information for their possible applications in the field of catalysis and electrocatalysis on nanostructured surfaces.

Acknowledgements M.S. and M.M. would like to thank “Higher Education Commission Islamabad, Pakistan for financial support through the “Indigenous 5000 and open merit 200 Ph.D. Scholarship Scheme” and Pakistan Science Foundation through project PSF/RES/ C-QU/CHEM(408).

Notes and references

Fig. 10 V –I curves of Cu–ZnO films deposited from precursor (2) at 400 and 475 ◦ C.

This journal is © The Royal Society of Chemistry 2009

1 J. E. Baker, R. Burch and S. E. Golunski, Appl. Catal., 1989, 53, 279 and references therein. 2 S. Angelov, D. Mehandjiev, B. Piperov, V. Zarkov, A. Terlecki-Baricevic, D. Jovanovic and Z. Jovanovic, Appl. Catal., 1985, 16, 431. 3 S. H. Taylor, G. J. Hutchingsa and A. A. Mirzaeib, Chem. Commun., 1999, 1373. 4 J. W. Couves, J. M. Thomas, D. Waller, R. H. Jones, A. J. Dent, G. E. Derbyshire and G. N. Greaves, Nature, 1991, 354, 465. 5 A. Cybulski, Catal. Rev. Sci. Eng., 1994, 36, 557. 6 A. A. Mirzaei, H. R. Shaterian, S. H. Taylor and G. J. Hutchings, Catal. Lett., 2003, 87, 103.

Dalton Trans., 2009, 5487–5494 | 5493

7 G. C. Chinchen and M. S. Spencer, Catal. Today, 1991, 10, 293. 8 A. M. Pollard, M. S. Spencer, R. G. Thomas, P. A. Williams, J. Holt and J. R. Jennings, Appl. Catal., A, 1992, 85, 1. 9 Y. Choi, K. Futagami, F. Fijitani and J. Nakamura, Appl. Catal., A, 2001, 208, 163. 10 Y. Guo, R. Weiss and M. Epple, Eur. J. Inorg. Chem., 2005, 3072. 11 O. V. Cuchillo, U. Pal and C. V. Lopez, Modern Phys. Lett. B, 2001, 15, 675. 12 S. M. Zhou, X. H. Zhang, X. M. Meng, K. Zou, X. Fan, S. K. Wu and S. T. Lee, Nanotechnology, 2004, 15, 1152. 13 L. Guo and S. Yang, Chem. Mater., 2000, 12, 2268. 14 M. S. Spencer, Catal. Lett., 2000, 66, 255. 15 M. Kurtz, H. Wilmer, T. Genger, O. Hinrichsen and M. Muhler, Catal. Lett., 2003, 86, 77. 16 M. Rehbein, M. Epple and R. D. Fischer, Solid State Sci., 2000, 2, 473. 17 Y. Guo, R. Weiss and M. Epple, J. Mater. Chem., 2005, 15, 424. 18 M. Veith, J. Chem. Soc., Dalton Trans., 2002, 2405. 19 V. G. Kessler, Chem. Commun., 2003, 1213. 20 S. Polarz, A. V. Orlov, M. W. E. van den Berg and M. Driess, Angew. Chem., Int. Ed., 2005, 44, 7892. 21 L. G. Hubert-Pfalzgraf, Coord. Chem. Rev., 1998, 180, 967. 22 W. A. Herrmann, N. W. Huber and O. Runte, Angew. Chem., 1995, 107, 2371; W. A. Herrmann, N. W. Huber and O. Runte, Angew. Chem., Int. Ed. Engl., 1995, 34, 2187. 23 R. C. Mehrotra, A. Singh and S. Sogani, Chem. Rev., 1994, 94, 1643. 24 M. Veith, S. Faber, R. Hempelmann, S. Janssen, J. Prewo and H. Eckerlebe, J. Mater. Sci., 1996, 31, 2009.

5494 | Dalton Trans., 2009, 5487–5494

¨ 25 G. Ertl, H. Knozinger, and J. Weitkamp, Handbook of Heterogeneous Catalysis, Wiley-VCH, Weinheim, 1997. 26 B. F. G. Johnson, M. C. Klunduk, T. J. O’Connell, C. McIntosh and J. J. Ridland, J. Chem. Soc., Dalton Trans., 2001, 10, 1553. 27 M. Hamid, A. A. Tahir, M. Mazhar, M. Zeller and A. D. Hunter, Inorg. Chem., 2006, 45, 10457. 28 Bruker (2003). SMART (Version 5.630) and SAINT-Plus (Version 6.45A). Bruker AXS Inc., Madison, Wisconsin, USA. 29 Bruker (2001) SHELXTL (Version 6.12). Bruker AXS Inc., Madison, Wisconsin, USA. 30 M. Shahid, M. Mazhar, P. O’ Brien, M. A. Malik and J. Raftery, Polyhedron, 2009, 28, 807–811. 31 J. Zhang, L. G. Hubert-Pfalzgraf and D. Luneau, Inorg. Chem. Commun., 2004, 7, 979. 32 M. Mazhar, S. M. Hussain, F. Rabbani, G. Kociok-Kohn and K. C. Molloy, Bull. Korean Chem. Soc., 2006, 27, 1572. 33 R. Becker, J. WeiB, M. Winter, K. Merz and R. A. Fischer, J. Organomet. Chem., 2001, 630, 253 and references therein. 34 M. Hamid, A. A. Tahir, M. Mazhar, M. Zeller and A. D. Hunter, Inorg. Chem., 2007, 46, 4120. 35 H. E. Swanson and E. Tatge, Natl. Bur. Stand. US Circ. 539, 1953, 359, 1. 36 H. Schulz and K. H. Thiemann, Solid State Commun., 1979, 32, 783. 37 J. Albertsson, S. C. Abrahams and A. Kvick, Acta Crystallogr., Sect. B: Structural Science, 1989, 45, 34. 38 H. O. Pierson, Handbook of Chemical Vapor Deposition, Principles, Technology, and Applications, 2nd Edn, Noyes Publications, Park Ridge, New Jersey, USA, 1992, p. 63.

This journal is © The Royal Society of Chemistry 2009