Process for making surfactant capped metal oxide nanocrystals, and products produced by the process

This application claims priority under 35 U.S.C. 119, based an Provisional patent application Ser. No. 60/162,299, filed Oct. 28, 1999 is a continuation-in-part of and application Ser. No. 09/702,219 entitled, A Process For Making Surfatant Capped Nanocrystals, filed Oct. 30, 2000, now U.S. Pat. No. 6,440,213, both of which are incorporated by reference herein in their entirety.

The invention described herein arose in the course of, or under, contract No. DE-AC03-SF00098 between the United States Department of Energy and the Regents, University of California for the operation of the Ernest Orlando Lawrence Berkeley National Laboratory.

This invention relates to a process for making dispersable surfactant capped nanocrystals. More particularly, it relates to a process for making dispersable surfactant capped nanocrystals of metal oxides with non-hydroxylated particle surfaces using a non-hydrolytic single precursor approach, and to the nanocrystals made thereby. Still more particularly, it relates to a process for making dispersable surfactant capped nanocrystals of transition metal oxides.

The advent of new methods to prepare semiconductor and metal nanocrystals, specifically the injection of molecular precursors into hot organic surfactants, has yielded markedly improved samples with good size control, narrow size distributions, and good crystallinity of individual and dispersable nanocrystals (see references 1-3). It is of considerable interest to apply these methods to the synthesis of metal oxide nanocrystals, particularly transition metal oxide nanoparticles, which typically are prepared by methods involving water as solvent or reactant (4-9). Using nonhydrolytic preparations of metal oxide nanocrystals at high temperature in organic surfactants in the absence of air or water, one observes markedly different properties with respect to defect structure and surface composition. So far as applicants are aware, there has been only one example of the solution-based nonhydrolytic synthesis of individual TiO 2 nanocrystals (10). Metal oxide nanocrystals with nonhydroxylated surfaces are believed to have significant advantages for applications in catalysis, ceramics, energy storage, magnetic data storage, sensors, ferrofluids, etc.

The following references contain useful background information pertaining to this invention , and are incorporated by reference herein in their entirety.

1. Murray, C. B.; Norris, D. J.; Bawendi, M. G. J. Am. Chenm Soc. 1993, 115, 8706-8715.

2. Peng, X.; Wickham, J.; Alivisatos, A. P. J. Am. Chem. Soc. 1998, 120, 5343-5344.

3. Sun, S.; Murray, C. J. Appl. Phys. 1999, 85, 4325-4330.

4. Charles, S. W.; Popplewell, J. Ferromagnetic Materials; Northholland Publishing Co.: Amsterdam, New York, Oxford, 1982; Vol. 2.

5. Brinker, C. J.; Scherer, G. W. Sol-Gel Science; Academic Press: San Diego, 1990.

6. Ziolo, R. F.; Giannelis, E. P.; Weinstein, B. A.; O'Horo, M. P.; Ganguly, B. N.; Mehrotra, V.; Russell, M. W.; Hufftnan, D. R. Science 1992, 257, 219-223.

7. Matijevic, E. Chem. Mater. 1993, 5, 412.

8. Moumen, N.; Pileni, M. P. Chem. Mater. 1996, 8, 1128.

9. Ying, J. Y. Special Issue: Sol-Gel Derived Materials. In Cheim Mater. 1997, 9, 2247-2670.

10. Trentler, T. J.; Denler, T. E.; Bertone, J. F.; Agrawal, A.; Colvin, V. L. J. Am. Chem. Soc. 1999, 121, 1613-1614.

11. Comprehensive coordination chemistry: the synthesis, reactions, properties & applications of coordination compounds, 1st ed.; Pergamon Press: Oxford, England, 1987; Vol. 2, Chapter 15.9.2.

12. Helm, D. v. d.; Merritt, L. L.; Degeilh, R.; MacGillavry, C. H. Acta Crystallogr. 1965, 18, 355-362.

13. Elerman, Y.; Atakol, O.; Svoboda, I.; Geselle, M. Acta Crystallogr. C 1995, 51, 1520-1522.

14. Tamaki, K; Okabe, N. Acta Crystallogr. C 1996, 52, 1612-1614.

15. Keilner, R.; Prokopowski, P. Anal. Chim. Acta 1976, 86, 175-184.

16. Nolze, G.; Kraus, W. Powder Diffr. 1998, 13, 256-259.

17. Shmakov, A. N.; Kryukova, G. N.; Tsybulya, S. V.; Chuvilin, A. L.; Solovyeva, L. P. J. Appl. Crystallogr. 1995, 28,141.

18. Jarosch, D. Mineral. Petrol. 1987, 37, 15-23.

19. Neuburger, M. C. Z. Phys. 1930, 67, 845-850.

20. Cullity, B. D. Elements of X-ray Diffraction, 2nd ed.; Addison-Wesley: Reading, MA, 1978.

21. Bentzon, M. D.; Wonterghem, J. v.; Morup, S.; Tholen, A.; Koch, C. J. W. Philos. Mag. B 1989, 60, 169-178.

22. Murray, C. B.; Kagan, C. R.; Bawendi, M. G. Science 1995, 270, 1335-1338.

23. Chemseddine, A.; Weller, H. Ber. Bunsen-Ges. Phys. Chem. 1993, 97, 636-637.

24. Fleet, M. E. Acta Crystallogr. B 1981, 37, 917-920.

25. Cornell, R. M.; Schwertmann, U. The Iron Oxides-Structure, Properties, Reactions, Occurrence and Uses, 1st ed.; VCH Verlagsgesellschaft: Weinheim, Germany, 1996.

26 . Reagent Chemicals: American Chemical Society Specifications ; 8th ed.; American Chemical Society: Washington, D.C., 1993.

It is an object of this invention to provide a non-hydrolytic process for the preparation of metal oxide nanocrystals.

It is a further object of this invention to provide a non-hydrolytic process for the preparation of transition metal oxide nanocrystals having non-hydroxylated surfactant capped surfaces.

It is a further object of this invention to provide surfactant capped metal oxide nanocrystals with non-hydroxylated surfaces which are dispersable in organic solvents, and a process of preparing them

It is a still further object of this invention to provide new and novel metal oxide nanocrystals.

These, and other objects are achieved by (a) rnixing a solution of a metal cupferron precursor complex of the formula M X Cup X , wherein M X is a metal ion in the oxidation state X selected from the group consisting of elements in Group 2, Group 3-12 of the 4 th period, Group 3-6 of the 5 th and 6 th period, Group 10-12 of the 5 th period, Group 12 of the 6 th period, and Group 13 th to 15 th , and the Lanthanide and Actinide series of the periodic table, and Cup is a N-substituted N-Nitroso hydroxylamine, with a coordinating surfactant, and (b) heating the mixture at a temperature and for a sufficient period of time to cause thermal decomposition of the M X Cup X precursor and formation of the desired nanocrystals.

As used herein, the term “coordinating surfactant” means an organic molecule consisting of a polar headgroup and an apolar group providing stabilization against coagulation and precipitation of particles by binding to particle surfaces through the polar headgroup and allowing dispersion of particles in aliphatic, aromatic and halogenated hydrocarbons.

FIG. 1 shows the result of the thermal analysis of iron, manganese and copper cupferronate in a TGA/DTA apparatus under N 2 -flow (left). To the right, the xrd patterns of the respective decomposition products are displayed.

FIG. 2 and 3 present low-resolution TEM images as well as XRD patterns of γ-Fe 2 O 3 , Mn 3 O 4 , and Cu 2 O nanocrystals, respectively.

FIG. 4 presents low-resolution TEM images as well as derived particle size distributions of γ-Fe 2 O 3 nanocrystals prepared under varying synthesis conditions.

FIG. 5 shows particle size distributions derived from low-resolution TEM imaging as a function of refluxing time after the precursor injection.

FIG. 6 describes the increasing average particle size as the reaction precipitate is subsequently extracted with Toluene.

FIG. 7 , shows a low-resolution TEM image of a monolayer of individual γ-Fe 2 O 3 nanocrystals (10.0±1.5 nm) covering an area bigger than 2 μm 2 . Top left: High-resolution TEM image of one of the nanocrystals in this sample. The indicated lattice plane distances correspond to the (113) and (201) lattice planes of tetragonal γy-Fe 2 O 3 with ordered superlattice of the cation vacancies. Top right: FFT of the high-resolution TEM image looking down the [ 51-2 ] zone-axis.

FIG. 8 and 9 present XRD patterns as well as FF-IR spectra of the iron, manganese and copper cupferronates, respectively.

We have now discovered a new nonhydrolytic single-precursor approach to the synthesis of dispersable nanocrystals of metal oxides, particularly transition metal oxides.

The process comprises: (a) mixing a solution of a metal cupferron precursor complex of the formula M X Cup X , wherein M X is a metal ion in the oxidation state X selected from the group consisting of elements in Group 2, Group 3-12 of the 4 th period, Group 3-6 of the 5 th and 6 th period, Group 10-12 of the 5 th period, Group 12 of the 6 th period, and Group 13 th to 15 th , and the Lanthanide and Actinide series of the periodic table, and Cup is a N-substituted N-Nitroso hydroxylamine, with a coordinating surfactant , and (b) heating the mixture at a temperature and for a sufficient period of time to cause thermal decomposition of the M X Cup X precursor and formation of the desired nanocrystals.

While the process described below is focused primarily on the preparation of transition metal oxide nanocrystals, it is equally applicable to the formation of any metal oxide nanocrystal where the metal (M X ) is a metal ion in the oxidation state X selected from the group consisting of elements in Group 2, Group 3-12 of the 4 th period, Group 3-6 of the 5 th and 6 th period, Group 10-12 of the 5 th period, Group 12 of the 6 th period, and Group 13 th to 15 th , and the Lanthanide and Actinide series of the periodic table.

The value for X will vary, depending on which metal is used in the formation of the precursor, and the oxidation state of the metal. In general, however, X will range from 1 to 4.

Metal Cupferron complexes M X Cup X (M: metal ion; Cup: N-nitroso N-phenylhydroxylamine, C6H5N(NO)O—), with the metal ion coordinated via the oxygen atoms of the Cup ligand in a bidentate manner, proved to be promising as molecular precursors. Metal cupferronates are used in the precipitation or extraction of metal ions from aqueous solution, and are easily prepared for many metal elements (11). In the following, we demonstrate that injecting solutions of metal Cupferron complexes in octylamine into long-chain amines at 250-300° C. yields nanocrystals of iron oxide, manganese oxide,and copper oxide. These nanocrystals range from 4 to 10 nm in diameter, are crystalline, and are dispersable in organic solvents. As Cupferron complexes are available for almost all metal elements, the use of this precursor offers an interesting general approach to metal oxide nanocrystals.

The preparation of the metal Cupferron precursor is based on the precipitation of metal ions from aqueous solution at a specific pH with Cupferron, the ammonium salt of N-nitroso N-phenylhydroxylamine (details of the synthesis and characterization of the metalcupferronates M X Cu X (M: Fe, Cu, Mn) are provided infra). In all three cases the elemental analysis and the powder X-ray diffraction (XRD) patterns agreed well with single-crystal data (12-14). In addition, FT-IR spectroscopy excluded the presence of excess Cupferron, water, or solvent molecules, and agreed well with results given in the literature (15).

Dried powders of metal cupferronates show sharp decomposition temperatures (see FIG. 1 , left) of 180° C., 230° C., and 205° C. for FeCup 3 , MnCup 2 , and CuCup 2 , respective when heated in a DTA/TGA apparatus under nitrogen (ΔT=10° C. /min., 25° C.-500° C.). In FIG. 1 right, the XRDs of the respective decomposition products reveal that they consist of γ-Fe 2 O 3 , MnO, and Cu. The latter is a consequence of the reduction of CuO/Cu 2 O by the reducing atmosphere given by the organic decomposition products of the Cupferron complex. This result proves that metal Cupferron complexes can indeed act as molecular precursors for transition metal oxides in the absence of O 2 and H 2 O.

In the following, a typical synthesis yielding iron oxide nanocrystals with 6-7 nm particle diameters is described. To remove oxygen and water, 7 g of trioctylamine was heated to 100° C. for 1-1.5 h and repeatedly evacuated to 20 mTorr and purged with Ar. A solution of 0.3 M FeCup 3 in octylamine was treated the same way at 60° C. The reaction was initiated by the rapid injection of 4 mL of FeCup 3 stock solution into the trioctylamine at 300° C. under vigorous magnetic stirring and an Ar atmosphere. A color change of the liquid from colorless to dark-brown and the evolution of gas indicated the decomposition of the metal Cupferron complex. After the liquid was heated for 30 min at 225° C., the reaction was stopped and the liquid was allowed to cool. At room temperature the flask contained nanocrystals of iron oxide, in both a dark-brown, clear liquid supernatant and a precipitate. The latter results from the high concentration of nanocrystals and their limited solubility in trioctylamine at low temperature. Adding 1-2 mL of organic solvents such as toluene, hexane, CHCl 3 , etc. to this precipitate yielded clear, deep-brown dispersions of iron oxide nanocrystals which were stable for weeks at room temperature. Low-resolution TEM imaging (see FIG. 2 left) revealed particles with an average size of 6.7 nm and a standard deviation of 1.4 nr.

By addition of a 3-fold volume excess of methanol, the iron oxide nanocrystals could be reprecipitated as a brown powder. Adding methanol to the supernatant of the reaction led to a brown precipitate of iron oxide nanocrystals, which could also be dispersed and reprecipitated by suitable solvents. For both fractions, dispersion and reprecipitation a could be repeated several times.

By washing the precipitate excessively with methanol, the surfactant could be removed yielding typically about 50-60 mg of a brown insoluble powder after drying. This powder still contained about 10% of volatile compounds as indicated by therrnal gravimetric analysis between 300 K and 1100 K in an O 2 atmosphere. The overall yield of iron oxide is thereby estimated to 45 mg to 55 mg, which corresponds to about 50% of the iron injected as cupferron complex. Similar procedures were used in the synthesis of manganese oxide and copper oxide nanocrystals. In the case of manganese oxide, 4 ml of a 0.3 M solution of MnCup 2 in octylarmine were injected into 7 g trioctylamine at 360° C. and refluxed for 10 min. at 275° C. An orange precipitate was obtained after the reaction, which changed color to brown upon dispersion in toluene. Low-resolution TEM images (see FIG. 2 , middle) of this dispersion indicated the presence of 12.0±4.7 mrn Mn 3 O 4 particles.

In the case of CuCup 2 , hexadecylamine was used as a surfactant to disperse the precipitated nanocrystalline copper oxide in organic solvents like toluene, hexane, CHCl 3 etc. In addition, the reaction had to be stopped immediately after injection since the information of metallic Cu instead of the copper oxide is favored at high temperatures under the strongly reducing conditions of the reaction. In detail, 2 ml. of a 0.3 M solution fCuCu 2 in octylamine were injected into 5.5 g hexadecylamine at 250° C. The reaction was stopped by removing the heat as soon as the temperature reached again 230° C. The average particle size as determined from low-resolution TEM imaging is 6.9±2.5 nm (see FIG. 2 , right).

Powder XRD reveals the nanocrystalline nature of the described samples (FIG. 3 ). The diffraction patterns were fitted with the program PowderCell (16) using structure data for γ-Fe 2 O 3 , Mn 3 O 4 , and Cu 2 O as models (17-19). Using the Debye-Scherrer is equation (20), a crystallite size was determined for every reflection, and all were averaged to give the mean values reported in FIG. 3 . These averaged crystallite sizes are in good agreement with mean diameters determined from low-resolution TEM imaging (compare with results in FIG. 2 ).

Particles with average sizes down to 4 nrn were synthesized by lowering the injection temperature and/or lowering the injected precursor concentration. For instance, injecting the FeCup 3 precursor solution at 250° C. and refluxing at 200° C. for 30 mnin, instead of 300° C. and 225° C. as in the sample of FIG. 2 (see also FIG. 4 , left), yields γ-Fe 2 O 3 nanocrystals 5.2±1.5 nm in size (see FIG. 4 middle). However, the injection of an additional 2 mL of precursor solution after 5 min of refluxing at 200° C. results in particles with diameters of 6.1±1.8 nm (see FIG. 4 right).

The influence of the reflux time is shown in FIG. 5 for an injection of 4 ml 0.3 M FeCup 3 /Octylamine solution into 7 g trioctylamine at 250° C. and subsequent refluxing at 200° C. Apparently, the particle growth is essentially finished after 10 min. and only a small increase in the width of the size distribution is observed with longer refluxing tirne. This behavior is similar to the growth kinetics of semiconductor nanocrystals.(2) However, further studies are necessary to achieve a similar understanding of the growth kinetics of oxide nanocrystals.

In all cases it was observed that subsequent extractions of the reaction precipitate with small amounts of toluene yielded fractions containing continuously bigger particles. In the case of the preparation with an additional secondary injection, each subsequent extraction of the reaction precipitate resulted in approximately a 1 nm increase in the average particle diameter (seex FIG. 6 ). Consequently, the fifth fraction contained γ-Fe 2 O 3 nanocrystals which were 10.0±1.5 nm in diameter. A low-resolution TEM image of this fraction is shown in FIG. 7 . Apparently, the slow evaporation of a dispersion of nanocrystals in toluene on a TEM grid leads to the formation of an extended monolayer of particles. The covered area is larger than 2 μm 2 and extends beyond the part shown in FIG. 2 . Each particle is separated from neighboring particles by its shell of surfactant. Currently, the relatively broad size distribution of 15% prevents the formation of ordered superlattices as observed with other systems (3,21,22) However, we believe that by using size-selective precipitation (23), in addition to size-selective extraction during the isolation process, we will be able to achieve size distributions with less than 10% standard deviation.

The crystallinity of these particles is revealed by high-resolution TEM imaging as shown in FIG. 7 (top left). The observed lattice plane distances of 4.77 Å and 4.11 Å, as well as the angle of about 50° between the crossed fringes, indicate the presence of tetragonal γ-Fe 2 O 3 with an ordered superlattice of cation vacancies (17). The corresponding lattice planes are (113) and (201), respectively, and the FFFi (top right) of the high-resolution image indicates that the particle was imaged along its [51-2] zone axis. Neither cubic γ-Fe 2 O 3 with a statistical distribution of cation vacancies over the octahedral sites nor Fe 3 O 4 shows a similar agreement (17,24). In other high-resolution TEM images, we observe particles with lattice plane distances of 4.97 and 4.27 Å that also belong uniquely to the tetragonal superstructure phase of γ-Fe 2 O 3 (see following table 1 for comparison). Interestingly, it has been reported that this ordered superstructure phase occurs only in particles with diameters larger than 100 nm (25).

Prepaation of Metal Cupferron Precursors