ntroduction
Organic and printed electronics (OPE) is a new emerging technology, which uses conventional graphic arts manufacturing printing platforms, such as screen printing, ink-jet printing, gravure and offset lithography, to fabricate electronic devices. The printing process has advantages over conventional print and etches processes as it has lower processing cost and higher production throughput. Its sequential build-up process generates less waste, which makes it more environmentally friendly.

As a result of its low-cost and large area printing characteristic, OPE opens up a new market for cheap disposable electronics, such as RFID, diagnostic devices and sensors, where mass-scale low cost, lightweight, and flexible electronics devices are required. The flexibility and thinness nature of OPE makes it suitable for new products such as flexible e-paper and displays. Other applications of OPE include memories, rollable photovoltaics (PV) and batteries, which forecasts a multibillion market in the near future.

According to recent IDTechEx studies, there are about 2,250 organizations who are working on printed electronics, which is a twofold increase from 2007. Since 2008, nearly USD 2 billion has been invested by U.S. and European companies.

Experimental study
The global interest in printed and organic electronics is growing, however, OPE is still very much in its infancy. Materials, equipment, processing technology and new applications of OPE still require much further development. Nevertheless, this paper will focus only on using two conventional graphical printing techniques. For the gravure printing, technology to print continuous conductors instead of discrete pigment dots will be presented. For the ink-jet printing, larger part of the investigation will be focused on how to achieve low temperature curing of printed features, which involved in the development of an ink for ink-jet printing and the corresponding low temperature curing method.

Gravure printing
The gravure printer used in this experiment was a Printing Proofer. The gravure plate used in this experiment was 1mm thick stainless steel, and the pattern image was engraved onto it by a laser system. As conductors were the target features for this printing experiment, the traditional engraving method to produce the pattern image on the gravure plate, where the pattern image was formed by a number of individual raster cells with different spacing, was no longer applicable. Instead, grooves were routed on the gravure plate to make the line continuity for electrical conduction as shown in Figure 2. After laser engraving, the surface was mechanically polished to remove any steel burr surrounding the edges of the grooves. Otherwise the protruded burr would mask ink and affect printing quality as well as damaging the doctor blade.

The gravure ink used for the printing was Parelec silver ink. It is a mixture of organic solvent and silver particles which acts as conducting media after curing. Silver ink is selected as it has higher conductivity and would not prone to oxidation. The substrate used was Dupont 50 m Kapton HN polyimide (PI). The printed conductors are shown in Figure 3 and the minimum width achieved is about 50 m. The thicknesses of the printed conductors were found to be directly proportional to the groove width. The resistance of the conductors was measured by Agilent 4338B milliohmmeter after curing of the printed conductors and the resistivity was calculated accordingly. The resistivity was found to be inversely proportional to the groove width. The resistivity of 150 m conductors was found to be 9.01E-08 m, which was very close to the resistivity of the bulk silver, 1.59E-8 m [3]. Theoretically, the resistivity should be identical for each line width as the same ink was used. However, due to (i) better packaging of mic on-sized silver particles for wider and thicker line widths on substrates, and (ii) wider grooves on the gravure plate made silver particles embed more easily, this resulted in more silver particles being transferred to the substrates in subsequent printing. These two factors helped to decrease the resistivity of wider conductors.

Ink-jet printing
The silver ink used for gravure printing, where the typical particle size of silver of the order of a few microns to over ten microns is not suitable for inkjet printer as it will clog the nozzle. In addition, nanoparticles are preferred to achieve low temperature curing due to their lower melting point. Therefore, in this experiment, silver nanoparticles were used to fabricate silver ink for ink-jet printing.

Preparation of silver ink was divided into two parts; the first part involved the fabrication of silver nanoparticles, and the second part was the formulation of the silver ink. Chemical reduction method was adopted to fabricatesilver nanoparticles. First, the silver ion was reduced to silver nanoparticles, and they were then protected by a stabilizer. The size of the silver nanoparticle was found to be a function of the following factors:1.Concentration of silver ion;2.Type and concentration of stabilizer;3.Reaction environment (neutral, or alkalisolution);4.Sequence of reaction.

A multimodal silver nanoparticle mixture was observed on the fabricated silver nanoparticles. The size of most of larger particles was found to be around 20nm, surrounded by smaller silver nanoparticles (less than 10nm). These small silver nanoparticles are critical as they allowed the printed conductors to be cured at lower temperature due to their low melting point. The melting of these nanoparticles helped to fuse comparatively larger silver nanoparticles together, and to become electrically conductive. The multimodal particle mixtures also allowed more eff icient particle packing for electrical conductivity than monosized nanoparticles, as the smaller silver nanoparticles can f ill-up the spaces between the larger silver nanoparticles.

15% of silver nanoparticles by weight was left in the solution. 0.25% by weight of the adhesion promoter was then added to the solution to f inalize the fabrication process of the ink-jet silver ink.

The ink-jet equipment used in the experiment was a Microdrop MD-K-140 dispenser head with MD-E-201-H drive electronics. Photos of a printed conductor are shown in Figure 4. The actual thickness of printed conductors was determined by an atomic force microscope (AFM), fell in the range of 400-500nm.

Ink-jet printed conductors were cured by baking and flashing methods respectively. The time required for 100% curing of ink-jet printed conductors using baking method is summarized in Table 1.Flashing was conducted using a camera flashlight, Nikon Speed light SB-22. The maximum temperature of the flashing was recorded to be about 109oC by an infrared (IR) camera (NECTH9100PWV).

The resistivity of ink-jet printed conductors on polyimide substrates was found to be 3.60E-6 m and 7.05E-6 m by baking and flash curing respectively. As there is no signif icant difference of resistivity of these two curing methods, flash curing is preferred due to its high throughput that makes it compatible to reel-to-reel production.

Conclusion
The gravure printer was found to have the capability of printing conductors which could be further evaluated for other materials such as insulatingmaterials and semiconductors. Modification of the gravure plate is required for conductor printing in contrast to conventional printing. Grooves instead of individual cells are required.

A multimodal silver nanoparticle mixture (20nm and <6nm) was produced for ink-jet printing by the chemical reduction method using silver nitrate as the silver source. Factors affecting the size of the silver nanoparticles were determined. Two types of curing methods were studied and compared. Flash curing was found to be more suitable for curing printed conductors due to its high throughput. The resistivity of the cured printed conductors was determined.