In this work, we report on a new method of SrCu₂O₂ synthesis that is more economical, potentially environmentally friendly, and more conducive to rapid exploration of process and compositional phase space. We report the non-vacuum synthesis of SrCu₂O₂ from liquid precursors. The precursors are synthesized inexpensively and are amenable to ready compositional substitution.

The liquid precursor approach was chosen with the intention that they could readily be incorporated into inkjet printing in the future. Inkjet printing allows for control of the deposition space as well as a relatively inexpensive method of deposition. Greater control of deposition space allows scientists to deposit materials only where needed, thus reducing the amount of material used for certain applications. Multiple inkjet head printing may allow for elemental substitution gradients similar to the way color inkjet printers can create a gradient of colors. For example, a Ca liquid precursor may be simultaneously printed with the SrCu₂O₂ precursor just as yellow is simultaneously printed with cyan in conventional inkjet printing. This method could create gradients of cation substitution, altering the p-type characteristics of the TCO throughout the sample.

The first step toward inkjet printing of thin film SrCu₂O₂ is the synthesis of the correct phase using the non-vacuum processing and the liquid precursors. We report the synthesis and substitution of bulk phase SrCu₂O₂ through spray deposition of liquid precursors. Spin casting allows similar precursors to be spun into thin films. The viability of inkjet printing liquid precursors has also been explored.

Multiple liquid precursors in various solvents were successfully spin cast onto fused silica substrates. The 4 cm² substrates were mounted onto a spin caster (photoresist spinner, Headway Research Inc.). Liquid precursor was filtered through a 0.2 µm syringe filter and 2 drops were deposited onto the substrate. Samples were spun at 2000 rpm for 30 seconds at room temperature. The samples were warmed to 100° C for 30 minutes and then heated to 400° C for 30 minutes. In some cases, multiple layers were deposited by repeating the same process. The same annealing conditions are used as for the spayed case above.

Several liquid precursors in multiple solvents were also inkjet printed onto fused-silica substrates. A 50-micron inkjet print head (50 µm MicroJet™, MicroFab Technologies Inc.) was used to print the liquid precursors.

A second batch of samples was prepared using the exact same method; however, these were annealed in the vacuum chamber at 775° C in 2.0 x 10^-5 Torr O₂, a decrease from the original pressure.

Samples from both batches were characterized using X-ray diffraction (XRD, Scintag Model X1, Cu Kα) and inductively coupled plasma emission spectroscopy (ICP, Varian Liberty 150).

Figure 1. XRD θ/2θ spectra for bulk phase SrCuO2. Liquid precursor spray was deposited on fused silica substrate and annealed at 775° C under argon flow for 20 hours. Bottom panel: expected powder pattern intensities (JCPDS 38-1179).

Figure 2. XRD θ/2θ spectra of bulk phase (a) SrCu2O2 and (b) Ca substituted SrCu2O2. Samples were annealed at 775° C and 2.0 x 10-5 Torr for 4 hours. Bottom panel: expected powder pattern intensities (JCPDS 38-1178).

Table 1. Precursor* and Annealed Powder† Ratios *Ratio calculated according to preparation of liquid precursor †Ratio measured by ICP emission spectroscopy

The spin-cast films were similarly characterized. The precursor is a 0.5 M solution of copper formate and strontium acetate in a 70:30 mix of water and 2-propanol. This spectra was taken from a film deposited by spinning two layers of 0.5 M precursor. ICP results for the films are shown in Table 2.

Table 2. Thin Film Annealed Samples Showing Sr Deficiency

The viability of inkjet printing of the liquid precursors was also explored. The water-based precursor could be consistently printed under conditions given in Table 3.

Table 3. Stable Inkjet Printer Conditions with Water-Based Liquid Precursor

Discussion and Conclusion

Figure 3. Sr-Cu-O phase space for Cu/Sr = 2.

The XRD pattern for samples annealed at 775° C in 2.0 x 10^-5 Torr O₂ show that they appear to be highly crystalline and phase pure. From Table 1 it is clear that the Ca source added to the liquid precursor has survived the annealing process. This suggests that elemental substitution may be achieved using liquid precursors. Facile elemental substitution will allow for greater control of p-type conduction.

XRD patterns and ICP results for the spin-cast films indicate that the sample is less pure and less crystalline than the spray deposited counterparts (Table 2). Note the presence of Cu₂O in addition to SrCu₂O₂. This suggests that, after annealing, the thin film samples are Sr-deficient. The ICP results given in Table 2 show further evidence of the incorrect stoichiometry after annealing of several thin film samples. The loss of Sr from the precursor during anneal was unexpected and demonstrates a potential problem for the thin film samples. Mixed phase thin films will exhibit physical properties unlike those of the desired phase of pure TCO. Additional Sr may need to be added to the precursor or Sr vapor pressure may need to be controlled during anneal.

The inkjet printing of liquid precursors has developed as a viable option for the deposition of SrCu¬2O₂. The time and cost of preparing these samples appear to be less than other popular methods such as sputtering. Additionally, as the technique develops, composition gradients and control of deposition space may further prove inkjet printing to be a valuable to tool for solid-state chemists.

We have demonstrated the synthesis of phase-pure bulk SrCu₂O₂ through the use of liquid precursors for the first time. Ca has been substituted through the use a liquid precursor and been shown to endure the annealing process. Spin-cast thin films have been created using water based liquid precursors. Annealing of the thin films results in a mixed phase of SrCu₂O₂ and CuO. The ability to inkjet print the precursors to SrCu₂O₂ has also been demonstrated. This work suggests that p-type TCOs may be derived from liquid precursors and easily substituted to allow greater control of p-type conductivity. These methods may open doors to new and better solar cells, flat panel displays and various other opto-electronic devices.

References

Coutts, T.J. et al. (2000) Characterization of Transparent Conducting Oxides. Materials Research Society Symposium Proceedings 623:199.

Ginley, D.S. and C. Bright (2000). Transparent Conducting Oxides. Materials Research Society Bulletin 25:15

Kawazoe, H. et al. (1997) P-type Electrical Conduction in Transparent Thin Films of CuAlO2. Nature 389:939

Kudo, A. et al. (1998a) SrCu2O2: A p-type conductive oxide with wide band gap Applied Physics Letters 73:220.

Kudo, A. et al. (1998b) A New p-type Conductive Transparent Oxide: Cu₂SrO₂. Materials Research Society Symposium Proceedings 526:299

Kudo, A. et al. (1999) Fabrication of transparent p–n heterojunction thin film diodes based entirely on oxide semiconductors. Applied Physics Letters 75:2851

Suzuki, R.O. et al. (1992) Thermodynamics and Phase-Equilibria in the SR-CU-O System. Journal of the American Ceramics Society 75:2833.

Yanagi, H. et al. (2000) Chemical Design and Thin Film Preparation of p-Type Conductive Transparent Oxides. Journal of Electroceramics 4:407.