Printed electronics, when combined with displays, can facilitate the integration of electronic intelligence to a virtually limitless number of products.
by Klaus Ludwig, Jürgen Ficker, Wolfgang Clemens, and Wolfgang Mildner
IN just about every futuristic science-fiction movie ever made, no matter what direction the plot takes, the ubiquity of electronics is a common thread. Film makers love to portray the future as a world where electronics saturates the landscape, thus creating the true information society.
Such a scenario is no longer the stuff that dreams are made of. Thanks to large strides made in the field of polymeric semiconductors, printable electronics can pave the way for a world in which nearly every product will incorporate electronics, with displays serving as the primary user interface.
By utilizing the advantages of polymer materials, such as the ability to combine soluble electronic polymer materials with high-volume printing processes, low-cost printed electronics can become a reality. When this day finally arrives, the ramifications of the development of new displays, along with myriad other components, are virtually unlimited.
But first, the developers of these displays need to understand the special properties of the new printed electronics because the electronic platform is the linkage between each component, serving as a center of basic logic functions that can power components such as displays, power sources, keyboards, etc.
This article will provide a brief overview of the technologies associated with the attempt to combine printed electronics and soluble organic semiconductors and a progress report on the advances that have been made during the past several years. Possible fields of application will also be discussed along with the general requirements for the integration of displays, derived from the constraints of printed electronics.
The printing methods used to create electronic circuits requires the use of fundamentally new materials that must be handled in a completely different manner. New organic conductor and semiconductor materials will be used instead of silicon.
The adaptation of new technologies into high-volume printing processes can lead to low-cost electronic products such as radio-frequency identification (RFID) tags and displays. In this article, the required materials, devices, and production technology will be identified (a more detailed description can be found in Ref. 1).
Organic semiconductor and conductor materials have enormous advantages over silicon because many of theses materials can be processed in solution and do not have to be available as high-purity crystals. As a result, organic electronics can be produced by a printing process, presenting opportunities for high-volume low-cost applications.
However, because of their considerably lower charge-carrier mobilities, the electronic devices themselves are slower than traditional silicon devices. A comparison of the conductive and semiconductive properties of organic materials with other materials is shown in Fig. 1. The conductivity of intrinsically conductive, solution-processable polymers, such as polyaniline-emeraldin salt (PANI) or poly-(ethylenedioxy-thiophene)/poly(styrene sulfonic) acid (PEDOT/PSS), covers a wide range, from very low values up to about 200 S/cm [Fig. 1(a)]. This is sufficient for simple logic devices but not for high-power electronics because it is orders of magnitudes below the conductivity of metals such as copper.
In Fig. 1(b), the charge-carrier mobility of organic materials is compared to that of amorphous silicon (a-Si), polycrystalline silicon (poly-Si), and the crystalline states of silicon. Mobility is one of the most important parameters of a semiconductor because it describes the speed at which the charge carriers, i.e., electrons or holes, can be transported within a device. This figure indicates that polymers cover a mobility of up to 0.2 cm2/V-sec,2 reaching that of a-Si, while for small molecules values upwards of 1.5 cm2/V-sec are possible. In Fig. 1(c), the chemical structure as well as the calculated geometrical structure of poly(3-alkylthiophene) (P3AT) – the most widely used semiconductive hole-conducting polymer – is shown.
Because of the availability of conductive, semiconductive, and insulating polymers, active electronics consisting only of polymers can be realized.3 A transistor, defined as a current switch driven by an applied electrical voltage, is the fundamental electronic device in microelectronics. All microelectronics chips are based on the logical combination of transistors.
An organic field-effect transistor (OFET) essentially consists of four different components: electrically conducting layers, an insulator layer, a semiconducting material, and a carrier substrate. A cross section of a polymer field-effect transistor (PFET) as made in the lab by a rapid prototyping process is shown in Fig. 2. First, a gold layer is sputtered on a thin substrate of polymer film, from which the source–drain electrode is patterned photolithographically. The semiconductor is dissolved typically in chloroform and spin-coated on the source–drain layer, resulting in a homogeneous 30–100-nm-thick layer. In a similar manner, the insulator layer, in the form of a dissolved polymer, is spin-coated onto the semiconductor layer. The solutions are processed to form homogeneous layers 200–1000-nm thick. The last step of the process is the sputtering and patterning of the gate electrode, which again is made of gold. In this setup, the electrodes are still metal, but the work function of gold corresponds very well with that of organic conductors such as polyaniline or PEDOT.
The basic function of organic transistors is very simple and comparable to that of conventional silicon thin-film transistors (TFTs). Based on the structure of a TFT, the electrodes at the lower layer of a transistor are referred to as source–drain electrodes and that at the upper layer gate electrode (Fig. 2). This gate electrode provides electrons that migrate between source and drain, and without applying a gate voltage, current cannot flow between the source and drain electrodes because the semiconducting layer is not doped and instead acts as an insulator. If a voltage is applied to the gate, a conductive channel at the interface between the semiconductor and insulator results from the accumulation of charge carriers, and, as a consequence, a current will flow from the source to the drain electrode. The extent of the current depends on the gate voltage, which determines the number of charge carriers.
Fig. 1: (a) Comparison of the electric conductivity of organic materials with filled metal paste (e.g., Ag) and copper as a typical bulk metal. (b) Comparison of the charge-carrier mobility of organic semiconductors with amorphous silicon (a-Si), polycrystalline silicon (poly-Si), and crystalline silicon (Si-crystal). (c) Regioregular poly-3alkylthiophene (P3AT) is the most widely used semiconductive hole-conducting polymer.
Fig. 2: Schematic cross section of a top-gate polymer field-effect transistor (PFET) placed on top of a polyester foil. Typical thicknesses: electrodes, 40–200 nm; semiconductor, 30–100 nm, and insulator, 200–1000 nm.
Electrical measurements of our PFETs show good saturation of the drain–source current and a high on/off ratio [Fig. 3(a)]. Low off-currents are important in digital circuits. From the transmission characteristics, i.e., the drain–source current versus the gate voltage [Fig. 3(b)], a threshold voltage Uth of about -5 V is determined for this transistor. Above this voltage, the transistor is off, and below this value the transistor is on. High on-currents are required, for example, in display and amplifier circuits. Because a PFET has the ability to drive voltages from a few volts up to 100 V, these printed electronic devices offer compatibility with high-voltage applications such as display drivers.
Integrated Polymer Circuits
In reality, individual transistors are only the building blocks of an integrated circuit. The connection of many transistors to a logic unit is called an integrated polymer circuit (IPC) or a "polymer chip." The successful utilization of ring oscillators fabricated by using printed electronics is proof that complicated integrated circuits can be made. A ring oscillator is constructed by connecting an odd number of inverters in such a manner that the output of every inverter is linked to the input of the next one. The output of the last inverter is linked to the input of the first inverter. Because of the odd number of inverters, the device begins to oscillate, but only if the supply voltage applied to the ring oscillator is sufficiently high.
Devices such as a seven-stage poly-(3-hexylthiophene) (P3HT) based ring oscillator with an operating frequency of 106 kHz at a supply voltage of –80 V was demonstrated as well.4 Polymer circuits operating at a frequency of more than 192 kHz at a supply voltage of -60 V have been reported.5 Even at lower supply voltages, the device still works, but at lower frequencies. The highest frequency obtained thus far for a five-stage ring-oscillator based on P3HT is 0.6 MHz.
As previously mentioned, ring oscillators are the basic circuits whose construction using printed electronics have indicated that this technology is suitable for higher-order circuits. Figure 4 shows a 4-bit transponder chip6 consisting of 183 single transistors on an area of 10 x 7 mm. Such a device, combined with an antenna, is a major step towards the development of a transponder tag, referred to as a radio-frequency identification label.
The above description of the structure of IPCs is an indication of how simple PFET ICs can be produced in comparison to silicon-based electronics, even when using photolithographic processes. But actual low-cost applications can only be realized with continuous manufacturing processes, such as printing techniques.
Fig. 3: (a) Transistor output characteristics: drain current versus drain voltage using the gate voltage as a parameter (P3HT, L = 10 μW = 10,000 μm, dIns = 300 nm, εr = 3.2). (b) Transmission characteristics (left y axis): drain current versus gate voltage. The charge-carrier mobility can then be evaluated (right y axis).
Compared to conventional printing processes, the printing of polymer chips involves completely different preconditions. First, continuous line structures are required by the drain–source electrodes with resolutions in the micron range. The semiconductor and insulator layers have to be stacked on each other as very thin, homogeneous, defect-free layers, and the gate structure must be applied as accurately as possible on the drain–source layer. Second, the rheological values of the inks, such as those of viscosity, adhesion, and coating, used in printing IPCs differ from typical printing inks by several orders of magnitude. The conventional solution to this problem would be to combine the inks with additives to adapt these parameters to ones more suitable for printing. Unfortunately, the electrical performance depends critically on the purity of the solvated polymers and, thus, the functionality would be severely reduced. Finally, the overall electrical performance of the materials and the entire setup must be sufficient to meet the requirements of microelectronics. The demands placed on the machines as well as on the materials used are enormous.
Nevertheless, promising results in the printing of polymer-electronic components were demonstrated.7-9 By using current printing methods, seven-stage ring oscillators consisting of several integrated PFETs have been fabricated.10 Figure 5 shows such a device which has a printed channel length of approximately 50 μm. The ring oscillator consists of 15 integrated transistors connected to seven inverter stages. In the figure, the output characteristics of a PFET and an inverter stage are shown. The maximum clock frequency of this ring oscillator is 0.6 Hz at a supply voltage of –100 V, as measured from the output signal at the outcoupling transistor of the device. This indicates that the construction of a circuit, within the kHz range and operating at a sufficient supply voltage, will be possible using the adapted materials, processes, and layouts.
These results will have to be applied to continuous roll-to-roll processes that lead to low-cost production, creating new mass markets. Besides low production costs and high volumes, these printing processes also have the advantage that the electronics can be changed within a short cycle time by simply changing the printing plates.
Polymer-Chip Structures on Flexible Film
One additional important aspect is the chemical stability of the organic materials. Further encapsulation of the circuits will not be cost effective, especially for low-cost high-volume applications. After extensive research, the devices demonstrated a very-high shelf life – more than 2 years without the use of any further encapsulation – and operational lifetime.11
Printed electronics based on IPCs enable an unlimited field of applications for low-cost high-volume products, for which printed electronics perform sufficiently.
Radio-frequency identification (RFID) tags enable the transmission of information without line of sight from a tag mounted on a product to a reader/receiver. By using polymer electronics, these tags can be mounted onto nearly every product or package and can replace the widely used optical barcode in the areas of brand protection, anti-theft stickers, electronic tickets, logistics, track and trace, electronic product code, and many more.
Printed electronics will also enable a large variety of applications for "intelligent" products. These so-called "smart objects" can be combined with displays as well as other components, such as sensors, batteries, photovoltaic cells, memories, etc., for use in smart cards, games, and marketing products. Printed electronics can also be used as driving circuits in flexible, low-cost passive- or active-matrix displays.
Displays for Printed Electronics
The technology that PolyIC has developed allows for the production of roll-to-roll printed polymer electronics for use in RFID tags, for example. Although the manufacture of RFID tags has been given top priority at PolyIC, smart objects are also of significant interest. The company, however, does not plan to manufacture displays just for themselfs, but intends to integrate displays and combine them with smart objects.
Although many applications can be realized by using printed electronics, many also require a low-cost display or indicator in order to communicate with the user. These displays also must be compatible with the properties of printed electronics. Among the display technologies that can be integrated with printed electronics, the most typical are electro-phoretic, liquid crystal, electrochromic, and organic light-emitting diode (OLED). Some of the basic requirements include
• Pattern Layout. The display design strongly depends on the application, but the envisioned simple smart objects can be realized with segmented displays or indicators.
• Flexiblity. Printed electronics is flexible, and the devices will be produced and stored on rolls. If printed electronics is combined with display functionality, the resulting product will also be distributed on rolls. Therefore, the displays must also be flexible, and as an added advantage, these devices can be placed on products with curved surfaces.
• Thinness. Thin displays are preferable to thick displays during production and for storage and integration and are easier to incorporate into packages and posters.
• Low Current. As previously mentioned, printed transistors are limited in drive current. The maximum drive current for printed transistors depends upon the design parameters, such as the amount of surface area. The maximum current is typically in the range of about 100 μA.
• Low Cost. Because printed electronics will be introduced to mass markets at low cost, displays integrated with electronics must be low in cost too.
• Integration. The integration of printed electronics with displays and other components must be possible under mass-production conditions.
But, in addition to the basic compatibility requirements described above, there are more specific requirements that vary from application to application.
• Bistability. Many flexible-display technologies retain their optical information after turning the power off, which is preferable for applications such as e-books. But there are other applications where a self-erasing display is advantageous, mostly due to the simplicity of the design of the required driving electronics. The manufacture of blinking indicators, for example, would be much simpler if self-erasing displays were used instead of bistable displays.
• Image-Switching Speed. In many applications, the image-switching speed in the display can be slower than the detection speed of the human eye. A refresh rate of approximately 1 sec is usually acceptable for a single-patterned indicator. The more display segments switched on or off in sequence, the faster the switching speed usually has to be.
• Shelf Life. The period of time between production and a smart object's first use can be in the range of years. Therefore, all components of the smart-object system – the electronics, display, interconnections, etc. – must also last as long as the application requires, a minimum of a year.
• Operational Lifetime. Many smart objects are used only occasionally, sometimes only once or twice during product life. Therefore, the operational lifetime is, in many applications, much shorter than the storage lifetime.
• Switching Voltage. For printed electronics, a very large range of driving voltages is possible. The maximum voltage provided by printed electronics for a display is more or less dependent on the available energy source.
• Foldability. Foldable displays enable myriad new applications, such as smart objects attached to paper.
• Contrast Ratio. The displays must be easily readable under both ambient and sunlight conditions. Self-emitting displays are favorable, but not necessary.
There are many other specifications for displays in smart objects, which change from application to application. The aforementioned items are intended to present a brief overview of the most important concerns faced when dealing with a specific application. Table 1 lists the display requirements and includes a range of values that cover most of the applications previously discussed, but they are not universally valid.
Printed electronics is not intended to replace standard electronics based on silicon, but rather to enable the realization of bringing electronic intelligence to products where there are no electronics today. Printed electronics, especially in combination with simple displays, will enable a large variety of new applications. The displays integrated with printed electronics can have both low resolution and speed, but they also have to be compatible with the technical and economic requirements of printed electronics for mass production.
But electronics and displays are not the only components needed to engineer a smart object. There must be an integrated power source that can provide energy to the complete system. If these conditions are met, the opening of new markets will follow closely behind.
Fig. 6: Photograph of polymer-chip structures on flexible foil.
Table 1: Overview of the specifications for displays used in printed-electronics applications
1W. Clemens, W. Fix, J. Ficker, A. Knobloch, and A. Ullmann, "From polymer transistors toward printed electronics," J. Mater. Res.19(7), 1963-1973 (2004).
2A. Ullmann, J. Ficker, W. Fix, H. Rost, W. Clemens, I. McCulloch, and M. Giles, "High-Performance Organic Field-Effect Transistors and Integrated Inverters," Mater. Res. Soc. Symp. Proc. 665, C7.5 (2001).
3H. Rost, A. Bernds, W. Clemens, W. Fix, J. Ficker, A. Ullmann, S. R. Moreno, and I. McCulloch, "All-Polymer Organic Field-Effect Transistors," Proc. Mater. Week, 2001 (2002).
4W. Fix, A. Ullmann, J. Ficker, and W. Clemens, "Fast polymer integrated circuits," Appl. Phys. Lett. 81(9), 1735-1737 (2002).
5W. Fix, J. Ficker, A. Ullmann, D. Zipperer, and W. Clemens, Polymer Electronics, presented at SPIE's 48th Annual Meeting, 5217-01 (2003).
6J. Krumm, E. Eckert, W. H. Glauert, A. Ullmann, W. Fix, and W. Clemens, "A Polymer Transistor Circuit Using PDHTT," IEEE Electron. Dev. Lett. 25(6), 399-401 (2004).
7A. Knobloch, A. Bernds, and W. Clemens, "Printed Polymer Transistors," Proc. Polytronic 84, 84-90 (2001).
8A. Manuelli, A. Knobloch, A. Bernds, and W. Clemens, "Applicability of Coating Techniques for the Production of Organic Field-Effect Transistors," Proc. Polytronic, 201-204 (2002).
9A. Knobloch, A. Bernds, and W. Clemens, "An Approach towards the Printing of Polymer Circuits," Mater. Res. Soc. Symp. Proc.736, D6.3.1-D6.3.5 (2003).
10A. Knobloch, A. Manuelli, A. Bernds, and W. Clemens, "Fully printed integrated circuits from solution-processable polymers," J. Appl. Phys. 96(4), 2286-2291 (2004).
11J. Ficker, A. Ullmann, W. Fix, H. Rost, and W. Clemens, "Stability of polythiophene-based transistors and circuits," J. Appl. Phys.94(4), 2638-2641 (2003). •