SciELO - Scientific Electronic Library Online

vol.5 issue4Thermal, Structural and Morphological Characterisation of Freeze-dried Copper(II) Acetate Monohydrate and its Solid Decomposition ProductsSilica Derived from Burned Rice Hulls author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Materials Research

Print version ISSN 1516-1439On-line version ISSN 1980-5373

Mat. Res. vol.5 no.4 São Carlos Oct./ Dec. 2002 

Synthesis of Group IVB Metals Oxicarbides by Carboreduction Reactions

A.D. Mazzonia*, M.S. Conconib

CETMIC (Centro de Tecnología de Recursos Minerales y Cerámica) Cno. Centenario y 506. C.C. 49 (B1897ZCA) M.B.Gonnet - Buenos Aires, Argentina



Received: December 21, 2001; Revised: July 30, 2002


The metals of the group IV B (Ti, Zr, Hf) present a series of carbides and oxicarbides with scientific and technological interest. Many of these compounds belong to the subsystem "MO - MC" of the pseudoternary "MO - MN - MC" system (where M = Ti, Zr or Hf).

In this work carboreduction reactions of TiO2 and ZrO2 were performed in argon atmosphere, using temperatures from 1250° to 1650° and reaction times of 120 min. The oxicarbides obtained were in the range TiC0.16O0.84 to TiC0.73O0.27 and ZrC0.46O0.54 to ZrC0.90O0.10. respectively.

The reaction products were characterized by X-ray diffraction (XRD), with the calculation of their cell constants by means of the Rietveld method. Scanning Electron Microscopy (SEM) was used in the characterization of powdered materials. Additionally, the carborreduction reaction was followed by weight loss.

Keywords: carboreduction, anatase, badeleyite, oxicarbides, carbides

1. Introduction

The group IV B metals, as titanium (Ti) and zirconium (Zr), present pseudoternary systems of the type "MO - MN - MC" (where M = metal). Both systems show substitutional solid solutions which correspond to extensive monophasic zones. The solutions, named M(C,N,O), have the same cubic crystalline structure (S.G.:Fm3m) that MN, MC and MO phases in the corners of diagrams. The unit-cell parameter (ao) of M(C,N,O) phase depends on the MC, MO and MN content.

Figure 1 shows the pseudoternary phase diagram "TiO-TiN-TiC" proposed by Neumann et al.1. The monophasic zone covers the whole diagram at temperatures higher than 1100 °C. This zone corresponds to the Ti(C,N,O) phase. The cell parameter ao varies between 4.12 Å and 4.32 Å1.



The pseudoternary system "ZrO- ZrN- ZrC" was studied by Constant et al.2. These authors observed a considerable solubility of "ZrO" (added as an equimolar mix of ZrO2 + Zr) in ZrC, ZrN and Zr(C,N). Figure 2 shows the limit of monophasic zone at 1600 °C. The ao parameter of the phase Zr(C,O,N) changes between 4.570 Å and 4.692 Å.



The monophasic zones cover in part the MO-MC axes in both systems. Thus, the subsystems "MO-MC" present oxycarbides M(C,O). These oxycarbides, Ti(C,O) and Zr(C,O), have the same cubic structure of M(C,N,O) with ao depending on MC and MO content. The values of ao increase with the MC content , the highest value corresponds to MC pure phase.

The synthesis of these compounds is usually obtained by TiO, TiC or ZrO2, Zr, ZrC reactions in argon atmosphere or vacuum3,4. Carborreduction reactions of the MO2 (TiO2-anatasa, ZrO2- badeleyite) oxides constitute an alternative method to obtain these phases. This method is within the present tendency to use abundant and cheap raw materials to obtain ceramic powders (oxides as well as carbides). Ceramics are manufactured starting from well characterized raw materials and then purified in order to control the degree of impurities.

The Rietveld method allows to properly characterize crystalline phases by x-ray diffraction (XRD). This method was developed by Hugo Rietveld in 19695, in order to refine crystalline structures using neutron diffraction data. At the present time, it is also used to perform analysis of structure and crystalline defects, reticular parameter measurement and quantitative analysis in X-ray diffraction.

In this work, the carborreduction reactions of anatasa and badeleyite were studied by XRD. Reaction products were characterized by XRD (Rietveld) and scanning electron microscopy (SEM). The weight loss was also measured.

2. Materials And Methods

Raw materials used were badeleyite Riedel-de Haën from Germany with more than 99% (ZrO2 + HfO2) content, anatase Fluka from Switzerland with more than 99% TiO2 content and carbon black (carbon 97%, ash 1%, volatiles 2% and a specific surface area BET = 45 m2/g). All powders have an average particle size lower than 5 mm. The argon used contains less than 5 ppm of O2 and H2O.

Samples were prepared by wet mixing of the calculated amounts of regents, and then they were dried and pressed at 39 MPa into cylinders of 2 mm in height and 10 mm in diameter. Samples were called by a letter and the C/MO2 relation. Letter A was used for samples prepared from anatase and letter B for samples from badeleyite. Thus, we obtain the samples A200 (C/TiO2 = 0.200), A280, A360, A430 and A500 and the samples B180 (C/ZrO2 = 0.180), B210, B240, B300 and B450.

Reactions were performed in a horizontal alumina reactor with argon flowing through it. The pressure used was 0.05 Mpa above the atmospheric pressure, and the flow was 1 l/min. Experiments were carried out within the range of 1250 to 1650 °C with reaction times of 120 min and the temperature slopes were 10 °C/min upward as well as downward. The Ar flow was kept during cooling up to 200 °C.

Crystalline phases were characterized by X-ray diffraction with a Philips 3020 Goniometer with a PW 3710 controller, Cu-Ka, Ni filter, 40 kV-20 mA. The scanning was made between 10° and 75° with step size of 0.02° and a step counting time of 2 sec. Some samples were scanned with 40 kV-30 mA and a step counting time of 4 sec. The set of divergence, receiving and scattering slits were 1°, 0.2°, 1° and no monochromator was used. The unit cell parameters were refined using the Rietveld refinement FULLPROF 6 program. The starting crystallographic data for each phase were taken from literature7-11. The refining sequence began with the adjustment of the shift in 2q due to vertical sample displacement and the background. To do this, elemental Si was added to the samples as internal standard. Then, for each phase, the scale factors, the cell constants, the parameters for calculation of the full-width-at-half-maximum (FWHM), the profile (pseudo Voight) function coefficients, the preferred orientation parameter for the March function, were sequentially refined. Quantitative phase analysis was also performed by FULLPROF program based on the scale factors of each phase.

The Zr(C,O) and Ti(C,O) crystallite sizes were determined by the full-width-at-half-maximum of peak (200) for obtained after refinement, using the Scherrer equation and the PC-APD (PW 1877) Analytical Powder Diffraction Software, Version 3.6. Weight losses were measured by weighing the samples before and after the reaction process.

The reaction products were also studied by scanning electron microscopy (SEM) using a Philips 505 equipment.

3. Results And Discussion

The following theoretical equations describe the reactions carried out in these systems:

where a, b, c and d are stoichiometric coefficients (0 £ a £ 1, 1 £ b £ 2, 0.40 £ c £ 1 and 1.4 £ d £ 2). The phases {TiCaO2-b}and {ZrCcO2-d} are metal oxicarbides belonging to the MC-MO edge of the MC-MO-MN system. These phases are called Ti(C,O) and Zr(C,O) respectively.

Table 1 shows the results obtained in the TiO2 carborreduction. Estimated standard deviations of weight percents were derived from the estimated standard deviations on individual scale factors for the respective phases, and other error contributions were not included.

The Ti (C,O) content increase with the temperature and carbon content being the unique product obtained at high temperatures and high C content.

Table 2 shows the values of ao obtained by the Rietveld method for the Ti(C,O) phases. Ranges of values appear in some cases due to the necessity of using two or more Ti(C,O) phases in order to obtain a good fitting of peak widths. Figure 3a shows the final Rietveld plot for sample A360 treated at 1450 and Fig. 3b is part of the same graphic amplified to show the overlap of Ti(C,O) peaks belonging to different phases. In Fig. 3b, individual calculated profiles for both Ti(C,O) phases are also shown.



Within the temperature range used in these experiments, the Ti (C, O) cubic phase is throughout the TiO-TiC axis (Fig. 4) as well as the Ti (C, N, O) cubic phases cover the triangular diagram completely (Fig. 1). The carbon content of samples A would allow to obtain the complete composition range (100% TiO to 100%TiC). In Table 2, ao values of Ti(C,O) are between 4.234 Å and 4.308 Å. These values correspond to a composition range between TiC0.16O0.84 and TiC0.73O0.27 (between 16 and 73 mol% of TiC). The composition range was determined using the calibration curve ao vs. composition of Ti(C,O) (Fig. 4). This curve, from Neuman et al.1, was determined from samples prepared by reaction of pure TiO and TiC.



The values of ao tend to be higher with the increase of reaction temperature. It is not possible to determine a defined tendency of ao values changes with carbon content. This little ao variation is due to the low slope in Fig. 4 where big changes in composition produce little shift of the unit cell parameter.

Some titanium sub-oxides were also detected. These phases are Ti3O5, g-Ti3O5, Ti2O3, and Magnéli phases (TinOn+2) similar to those observed by other authors12. These oxides become more important at low temperature and in samples with low carbon content. In Table 3 the refined unit-cell parameters of these phases are in good agreement with published data8-11.



The weight losses showed in Table 4 allow us to see that the reaction progresses with the increase of carbon content and temperature, being the obtained tendency consistent with the detected crystalline phases, within the margin of experimental error.



Badeleyite carborreduction results (reaction B) for the 5 studied samples are shown in Tables 5 and 6.




All samples have enough carbon to obtain an oxicarbide as unique phase. Theoretically oxicarbide compositions would be between ZrC0.45 O0.55 for B180 and ZrC for B300.

The zirconium phases obtained were: Zr(C,O) phase, monoclinic ZrO2 (badeleyite) and tetragonal ZrO2 in lower proportions. Figure 5a shows the final Rietveld plot for sample B450 treated at 1600 °C and Fig. 5b is part of the same graphic amplified to show the overlap of Zr(C,O) peaks belonging to different phases. In Fig. 5b, individual calculated profiles for all Zr(C,O) phases are also shown.

Badeleyite is the principal phase at temperatures lower than 1600°. The amount of this phase decreases with the increase of temperature and carbon content, being the minority phase at higher temperatures. Mean unit-cell parameters refined for badeleyite were: a = 5.151(3) Å, b = 5.210(5) Å, c = 5.318(5) Å and b = 99.27(4)° Å.

Tetragonal zirconia content was between 1 and 3 wt.% in all samples containing ZrO2 presenting similar behavior to the badeleyite.

Zr(C,O) increases with the increase of temperature and carbon content. It is the principal phase at a temperature of 1600° or higher.

The obtained values of ao (4.640 Å - 4.683Å) show that the formed oxicarbides were within the range ZrC0.46O0.54-ZrC0.9 O0.1 (between 54 and 10 mol%.of ZrO). The composition range was determined using the calibration curve ao vs. Zr(C,O) composition (Fig. 6). This curve, from Kosolopova et al.4, was obtained from samples prepared by reaction of pure Zr, ZrO2 and ZrC.



The Rietveld analysis shows that the wide peaks can be fitted using two or more phases M(C,O) indicating that the wide peaks are actually due to overlap of several narrow peaks. In almost all the samples, the use of two phases allowed to have RBragg values near 2.85 for both M(C,O). The Rexp are between 4.63 and 5.97 and the Rwp are between 9.20 and 15.56. This phenomenon can be observed in other systems where there are phases having changeable compositions and unit-cell parameters associated13-15.

The crystal sizes of the Zr(C,O) and Ti(C,O) were calculated using the width of peaks (200) determined from the fitting results. Both phases presented similar behavior, with crystal size values between 560 Å and 3000 Å without a definite tendency.

Table 6 shows a weight loss increase with the increase of temperature in all samples. The carbon content has influence at 1600 °C or higher temperatures when the Zr(C,O) is the principal phase. High weight loss values obtained principally for sample B180 may be due to ZrO(g) loss (reaction C).

But in samples with high carbon content, the ZrO(g) loss my be limited by reaction D.

Finally, reagents and reaction products were characterized by scanning electron microscopy (SEM). Morphology variations caused by reactions were observed (Fig. 7).

When comparing the TiO2 reagent with the reaction products an important increase of the particle size is observed, indicating their sintering. This phenomenon can also be observed in low temperature samples. Besides, the TiO2 grain morphology and that of the products are different.

Nevertheless, B samples seem to preserve the morphology and size of the raw material that originates them (the badeleyite).

4. Conclusions

1. Carborreduction reactions of TiO2 and ZrO2 are a suitable method to prepare oxycarbides of TiO-TiC and ZrO-ZrC systems respectively.

2. All oxycarbides obtained belong to MO-MC axes of MO-MC-MN system.

3. The carbon content of oxycarbides increased with reaction temperature.

4. Carboreduction of TiO2 produced Ti(C,O) as unique product at temperatures higher than 1450 °C and carbon/mineral ratios higher than 0.360. In the other samples some titanium oxides and suboxides were obtained but non remainder TiO2 was detected.

5. Carboreduction of ZrO2 produced Zr(C,O) as unique reaction product.

6. The Rietveld method allowed to characterize titanium and zirconium oxycarbides obtained by carboreduction and also to refine unit cell parameters of minority phases.

7. The observation by SEM permitted to see the TiO2 and ZrO2 different behavior with regard to their morphology modification during these reactions.


1. Neumann, G.; Kieffer, R.; Ettmayer, P. Über das System TiC-TiN-TiO, Monatshefte für Chemie, v. 103, p. 1130-1137, 1972.        [ Links ]

2. Constant, K.; Kieffer, R.; Ettmayer,P. Über das pseudoternare system ZrO-ZrN-ZrC, Monatshefte fur Chemie, v. 106, p. 823, 1975.        [ Links ]

3. Krainer, H.; Konopicky, K. Untersuchungen von sinterhartmetallen. Berg-und Huttenm. Mh. v. 92, p. 166, 1950.        [ Links ]

4. Kosolapova, T.Y.; Fedorus, V.B.; Panasyuk, A.D.; Kotzina, G.K. Über das System ZrC-ZrO. Poroschkowaya metall. n.1, p. 109, 1972.        [ Links ]

5. Young R.A, The Rietveld Method. IUCr, Oxford University Press, NY, USA,1993.        [ Links ]

6. Rodriguez-Caravajal, Abstracts of the Satellite Meeting on Powder Difraction of XV Congres of the IUCr, Toulouse, France, p.127, 1990.        [ Links ]

7. Howard, C.J.; Hill, R.J.; Reichert, B.E. Structures of ZrO2 Polymorphs at Room Temperature by High-Resolution Neutron Powder Diffraction. Acta Crystallographica v. 44, p. 116-120, 1988.        [ Links ]

8. Wyckoff, R.W.G. Crystal Structures, Interscience, New York, 1963.        [ Links ]

9. Le Page,Y.; Strobel, P. Structural Chemistry of Magnéli Phases TinO2n-1(4 £ n £ 9). I Cell and Structure Comparisons, Journal of Solid State Chemistry, v. 43, p. 314-319, 1982.        [ Links ]

10. Åsbrink S.; Magnéli A. Crystal Structure Studies on Trititanium Pentoxide, Ti3O5, Acta Crystallographica, v. 12, p. 575-581, 1959.        [ Links ]

11. Hong, S.H.; Åsbrink, S. The structure of g-Ti3O5, Acta Crystallographica, v. B38, p. 2570-2576, 1982.        [ Links ]

12. Afir, A.; Achour, M.; Saoula, N. X - ray Diffraction study of Ti-O-C system at high temperature and in continuous vacumm, Journal of Alloys and Compounds, v. 288, p. 124-140, 1999.        [ Links ]

13. Mazzoni, A.D.; Aglietti, E.F.; Pereira, E. Carbonitriding of clay: Relation between the weight loss and crystalline phases during reaction, Journal of American Ceramic Society, v. 76, n. 9, p. 2337-2340, 1993.        [ Links ]

14. Mazzoni, A.D. Conconi, M.S.; Aglietti, E.F., Synthesis of Zr-Si-O-N Phases by Carbonitriding Reaction, Characterization of Crystalline Phases Using the Rietveld Method, Materials Research, v. 4, n. 2, p. 107-111, 2001.        [ Links ]

15. Mazzoni, A.D.; Conconi, M.S. Carbonitruración de ZrO2. Caracterización de productos mediante el método de Rietveld, Anales del V Congreso de Cerámica del Mercosur, Buenos Aires, Argentina, p. 32, 2000.        [ Links ]

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License