source: TTI MarketEye article
Dennis M. Zogbi’s article on thin film and integrated passive components.
Introduction and Premise
There is a clear increase in global interest and investment in new processes and materials for the production of advanced next generation capacitors, resistors and inductors. This trend includes spending on capital equipment and expanding in-house knowledge based on a larger and more diverse collection of advanced electronic material that have been field proven to generate nanofarads, nanohms and nanohenries. Two key market drivers pulling electronic components into the future are component miniaturization trends and new product introductions at higher operational frequencies.
Keeping Up with the Roadmaps
Roadmaps have finally pushed the limits of volumetric efficiency to new levels of existence with the 008004 EIA (Inches or 0201 in mm) case size chip, now available in ceramic chip capacitors and ceramic chip inductors (but not noticeably available yet in chip resistors, thick or thin film, and this absence is being felt in the supply chain). Such a small product becomes increasingly difficult to handle to accurately “pick and place,” but like any good technology, the learning curve is soon mastered. In passives, the enabling technology in ceramic capacitors and ceramic inductors to create such a small electronic component was the movement to thin film technology.
Expect the Dinosaurs to Fight Back
First let me start by saying that it is possible to extend thick film technology much further than the design engineers thought possible 20 years ago. As with any disruptive technology, it is only as good as its allowed to be by the massive establishment of pick and place machines and people with amber waves of signatures on paper work- (otherwise known as the dinosaurs); however, like any good disruptive technology it takes about 20 years of pushing the technology; the bric-a-brac of the trade show booth, the never-ending transition of the overhead slide, to finally bring it into the popular mindset, where its “design-in” is as easy a process as any of the primary dielectrics. One of the successful technologies that have gradually clawed its way into the mainstream has been the thin film process used to produce passive components.
The Guys Who Take up the Most Room Try and Make Things Smaller and Integrated
Figure 1.0 is a graphic illustration of how integrated passive electronic components have developed. It shows the three primary segments of the passive components industry- capacitors, resistors and inductors; with further breakouts for capacitors into the ceramic, tantalum, aluminum and DC film segments. From fixed capacitors, we see how ceramics were employed for use with thick film R/C networks. From the resistor side of the equation, where the majority of interest lies in R/C networks, we see how networks were an obvious trend from the planar resistor segment which started out with thick film resistor chips, then thick film networks, which then combined with ceramic capacitors to form R/C networks. The chart also shows how multichip arrays were developed from resistor chips, and subsequently included in network statistics. Moreover, the chart shows how IPECs evolved out of the semiconductor industry to compete for the same markets as the thick film R/C networks.
Thin Film Equipment Requirements for Passive Component Generation
The general maxim in the industry is that thin film passive components are a step up from the historical thick film methods of producing passive components (i.e. multilayer stacking, winding, screen printing, etc.,), and a step down from the production of semiconductor components. The key differentiation is that semiconductor equipment is capital intensive, while thick film component production process is raw material intensive, and this has always been a barrier to market entry. However, with the advent of new, higher frequency product lines AND the continued downward march toward greater volumetric efficiency of individual components, the volumes of consumption for thin film chips are now in the billions of pieces are warranting large investment from the larger passive component manufacturers (i.e. TDK and Murata).
To manufacture thin film chips, the production equipment required is minimal, and largely associated to the physical vapor deposition equipment. However, over time, additional equipment may be required to produce fully integrated passive devices as follows.
A PVD machine is required for the physical deposition of the tantalum nitride resistive layer. It as an absolute requirement for standard production of thin film passive components. Used PVD equipment can be used here because the requirements for thin film passives are not as stringent as that of state of the art semiconductors. It is generally believed that at least one additional “back-up” PVD machine is required in a standard operation, just in case the initial machine breaks down. A plasma etcher and mask stepper are required for each layer of screening (i.e. once for resistance, once for capacitance and once for the passivation. An ion implanter is required to manipulate the silicon to form silicon dioxide or silicon nitride capacitors; and ESD protection diodes, such as zeners, silicon avalanche diodes, or Schotzky diodes. An ion implanter may also be used to form the crystalline ferroelectric layer for inductance. A final passivation layer on the thin film integrated die is required. This is generally accomplished with a Chemical Vapor Deposition (CVD) machine. And while capacitance, resistance and inductance layers can be applied with PVD equipment; the fourth passivation layer is generally applied with a CVD machine. Clean room costs are very high because they require triple containment; and must be impervious to organic materials and gases. Vent hoods must put out air that is cleaner than when it went in. Large AC units are required for filtration. It is generally agreed that the more laser trimming machines the better. Laser trimming devices are needed to set the resistance values on the dies. Packaging is very expensive and may be as high as 75% of the costs in some instances because the process is labor intensive. None of the American companies use domestic sources for packaging due to the high labor costs unless, of course it is prototype work. Packaging is generally accomplished in the Asian countries outside of Japan. Tape and reel equipment will be required to satisfy 90-95% of orders. Most high volume operations require electronic components to be packaged in a tape and reel format for easy insertion into pick and place machines. However, some end-use markets, such as those in the industrial market segment for motor controllers require waffle packs instead of tape and reel. Test equipment is a requirement and is viewed as a bottleneck to the integrated passive device production. Component tests for military/aerospace applications are very stringent and time consuming. And must be taken into consideration when doing cost planning for thin film processing because the defense and aerospace sector is a large part of consumption, especially for defense communications, guidance and munitions.
When Markets Grind, Profits Spark
One area of the market where profits fly like sparks as two edges of great markets grind past each other is in the continuing practice of making components smaller. This process creates a conflict between the scientific maxim that states that capacitance and resistance is limited by the physical size of the finished component or the available surface area within that component. But the process dictates that the vendor provide what the customer would like more of. The logic of the mosaic is to satisfy the consumer by making smartphones extremely thin, slick and stylish; and to create a seemingly infinite source of applications; a device that is ubiquitous and encompassing; and destined to continue to require advances in hardware and software to benefit from the churn that is coming when phone penetration rates by global population reach parity. But the research and development dollars are substantial, and the need for technology to be adopted now to keep up with the requirements of a changing market are well vested.
Integration Solutions for Passive Components = “Disruptive Technology”
Passive Component Integration has been a painfully slow process because the technology is disruptive and not necessarily in the best interests of those who produce thick film chips, nor even necessary until their customer base insist that thin film is important to continue the movement to smaller case sizes and to operate at higher frequencies. There is also a massive investment in pick and place equipment globally and this is a true roadblock to getting new thin film technology adopted. However, as our friends in Japan point out, the resulting products will be so revolutionary but require a new investment in tooling- a big up front cost. This is evident in the thin film production of the 008004 MLCC and Chip Inductor, which are considered enabling sizes for manufacturers of ultra-small communications modules. It is important for the reader to understand that in the chart below (Figure 2.0) there are specific groups of vendors who work within each discipline shown, and yet the overlap between the markets is limited. Therefore, the company that expands its knowledge base on how specific materials are manipulated to produce nanofarads, nanohenries and nanohms.
Figure 2: Integration Solutions of Passive Components FY 2016-2021
How C,R,L is being Achieved in Networks and Modules
75% of module solutions use re-packaged discrete passive components and 25% use manipulated materials (target sputtering). A list of materials is included below (Figure 3.0) and it is important for the reader to understand that in the future that thick film manufacturers will have to adopt thin film platforms in order to continue to satisfy the requirements of their largest and most important customers in smartphones. In the future we foresee the requirements for passive component manufacturers to have engineers with an increasingly broader understanding of materials and their functions in both thick and thin film as is shown below.
Figure 3: How Capacitance, Resistance and Inductance and Being Achieved in Networks and Arrays to Achieve
Historical Methods of Achieving Capacitance, Resistance, Inductance and Protection in Electronic Circuits
These materials and their sourcing, engineering, application and processing will be key to moving technology forward over the next 100 years with respect to volumetric efficiency.
A combination of thick and thin film will continue to command the global market, but over time thin film will gradually win more slots in high frequency applications and in power and communications markets where volumetric efficiency is a design parameter that moves constantly. The following materials list includes exotic materials that have been known to produce C, R and L over time.
Figure 4: Historical Methods of Achieving Capacitance, Resistance, Inductance and Circuit Protection in Electronic Circuits
Synergistic Technology Platforms in Thin Film Passive Component Production
Certain technology platforms have synergies, and certain materials lend themselves to these processes and have tech data support and 10 to 20 year real world application data. A major brand name customer noted that they needed to begin to branch out and learn more about varying materials, to expand their comfort zone beyond the usual boxes on the period table. Figure 5 below shows how each discipline has specific process methods that lend themselves to the engineering process for the respective feed stocks. Design engineers of the future will require a greater knowledge of numerous materials and process steps to create components that are “practically invisible, precise in their operation from lot-to-lot, and integrated with other components.”
Figure 5: Synergistic Technology Platforms in Passive Component Production: FY 2016
What Future Passive Component Integration Means to Intellectual Property Development Today
(1) Intellectual property and (2) market opportunities abound in circuit protection and inductance because of the “mystery factor solutions in search of a problem” coupled with the lack of technical support data on both product lines and the general limitation of global knowledge for such large markets. Support data on capacitors and linear resistors is much greater globally. It has always been this way for 30 years; noise control, ESD, overcurrent and overvoltage protection all have premiums because they are an after thought of the design engineer and do not facilitate the function of the final product. However, in terms of creating new intellectual property for the integration of passive components in the interests of improving overall module or chipset size, the focal point should be inductors because of the limit of points of reference in the literature on the subject when compared to capacitors and resistors.
Figure 6: What Future Passive Component Integration Means to Intellectual Property Development Today