Preface

We people working in semiconductor IC packaging business should be proud of ourselves. When the first cellular phone was demonstrated in 1973, it was the size of a brick.[1] To put it more precisely, it weighed 2.5 pounds, was 9 inches long, 5 inches deep, and 1.75 inches wide.[2] Now, a cellular phone can fit comfortably in the pocket; some of them are even smaller than the size of a palm. It’s unlikely it weighs more than 10 ounces.[2] People travel with a personal mobile device, a smartphone, or an iPad, and make a phone call right through either a cellular based network or through a voice over IP (VoIP) access point nearby. People are able to watch a movie using their own portable devices while in an inter‐continental flight. The achievement of the profession is enormous. This book details miniaturization technology that has helped to realize this newly found human experience—mobility, an experience that may be as significant as the freedom discovered by homo erectus.

Mobility is a result of concentrated efforts by the best global brains who contributed to creation of the huge communications networks; wired and wireless, hardware and software. As mentioned, this book concentrates on hardware technology that led to today’s success in miniaturization. The hardware technology includes the manufacturing craft from two distinct groups: more Moore (MM referring to the front end of a wafer foundry) and more than Moore (MTM referring to technology of packaging a die). Later in the preface, we will introduce how these two technologies took shape.

Without a doubt, mobility can be symbolized by a smartphone, or any other mobile phones. Considering the hypes every August‐September time frame in recent years when the largest smartphone equipment providers introduce their new phones. We know exactly what we are tuned in for in mobility: the speed of the multi‐core ARM processor, the multi‐GB capacity in the main mobile memory; the mixed signal capability of RF transceiver, and the multi‐mode, multi‐band capabilities of the RF front end (FE).

In The Dawn, below, we introduce how MM and MTM started, along with a historical event related to mobility.

P.1 The Dawn—Background of MM, MTM, and Mobility

Key historical events occurred before the digital revolution are discussed, which include:

1. Logic‐centric components in a computer, the CPU and its main memory, had created the need of and fueled advances in MM; without MM, there is no mobility.
2. MTM, like MM is another mass production technology; it is capable of manufacturing many, in the scale of several hundreds of million. MTM enhances miniaturization and affordability of mobile devices.
3. AT&T break‐up and the demonstration of the first analog handheld phone by Motorola helped usher in a new era of mobility.

P.1.1 The Computer, the Electrical Brain (DianNao 電腦)

One of very important features of the modern computing machine is the binary coding scheme, or the digits. The concept of digits and the system of carrying‐to‐the‐next‐significant‐digit is very similar to an ancient Chinese counting tool, the Chinese abacus (SuanPan, 算盤),[3], which was used traditionally in daily business transactions.

The modern concept of the computing machine originated from Alan Turing, through the concept of algorithm or artificial intelligence (AI). During war time from 1939 to 1945, Alan Turing, a crypt‐analyst, built a computing machine to crack the German Enigma codes, and is considered as the father of artificial intelligence. The Turing machine is a computing machine, based on mathematical and statistical models, to crack encrypted codes. One important consequence of the algorithm or AI is that it can be programmable, or reprogrammed. Interestingly, in movies, such as The Imitation Game, the “Electrical brain” (DianNao, or 電腦, which is literally “electrical brain” in English) was uttered out by Turing to describe his computing machine.[4]

The world’s first digital, electronic, general‐purpose computer, ENIAC, was built by the Moore School of Electrical Engineering, at the University of Pennsylvania in 1943 to 1946. It was Turing‐complete, digital, and capable of being reprogrammed to solve a full range of computing problems. ENIAC's design and construction was financed by the U.S. Army. This computer employed vacuum tubes as its digital elements. The service year of the first electronic computer was less than a year (February 14, 1946 to November 9, 1946), due mainly to the bulky, and thermally unsustainable digital elements.

At the time, John von Neumann worked closely with the Moore school faculty. In the second ENICA proposal, he began the concept of memory, where instructions can be stored. The basic architecture of the computer known as the “von Neumann Architecture,”[5] remains in use today.

John von Neumann didn’t invent the computer, however. The distinction rightly belongs to two men at the University of Pennsylvania, Professors Presper Eckert and John Mauchly.[6]

With these basic traits: 1) digital (similar to the beads and carry‐up system in the Chinese abacus), 2) algorithm, artificial intelligence, or programmable, attributed to Alan Turing, and 3) memory as a part of the computer architecture, attributed to John von Neumann, where the programs can be stored, these computers were still far from resembling the modern computing machines today.

Today’s modern computing machines can be either in a huge room, full of servers, handling complicated AI problems, such as speech recognition, or inside mobile devices. This difference in scale has to do with two things: one related to solid state transistors, the digital elemental devices; the other related to te monolithic principle of building both active devices and passives on the same platform, not only shrinking the sizes of electronics components, but also reducing the number of interconnections.

Note that vacuum tubes were used in ENIAC, the first digital, electronic, general‐purpose computer that was in operation from 1943 to 1946. John Bardeen and Walter Brattain at AT&T’s Bell Labs in the United States, during the so‐called “Miracle Month” of November 17 to December 23, 1947,[7] performed experiments and observed that when two gold point contacts were applied to a crystal of germanium, a signal was produced with the output power greater than the input. Solid State Physics Group leader William Shockley saw the potential in this, and over the next few months worked to greatly expand the knowledge of semiconductors. Bardeen, Shockley, and Brattain were presented with the Nobel Prize in Physics in 1956.[7]

Of equal importance, Jack Kilby (1923–2005)[8] came upon the idea he called the monolithic principle, in which he tried to build all the different parts of an electronic circuit on a silicon chip, when he was working at Texas Instruments. A U.S. Patent (3,138,743) "Miniaturized Electronic Circuits," for the first integrated circuit (IC), was filed on February 6, 1959.

Robert Noyce (1927–1990)[8] used a series of photographic and chemical techniques known as the planar process (which had just been developed by a colleague, Jean Hoerni) to produce the first, practical, integrated circuit.

Robert Noyce formed a company called Fairchild Semiconductor in 1957. In 1968, Intel was created by the famous former Fairchild employees, including Gordon Moore and Andrew Grove. This actually started the so‐called Moore’s Law claimed by Gordon Moore in an Electronics article in 1965.[9] According to Moore’s Law, the number of devices on a microchip IC doubles roughly every 2 years. This is a powerful law, from which shrinking the device’s feature size became the one and only focused goal of many wafer fab manufacturers. Interestingly, the development of modern computers (along with memory, i.e., logic‐centric circuits) has orchestrated 50 years of Moore’s revolution beginning in 1965.[9] More Moore (MM) refers to technologies employed to prolong the life of Moore’s Law.

In Chapter 1, we will discuss the most recent advances in MM technologies, such as immersion technology and TriGate (or FinFET). MTM, discussed briefly below, will be presented in Chapters 2 to 4.

P.1.2 Technology that Manufactures Hundreds of Million Many

The basis of MTM was started in early 1980s by IBM’s C4 (controlled collapse chip connection) and TCM (thermal conduction module) technologies (Figure P.1) using HTCC (high‐temperature co‐fired ceramic), and later replaced HTCC by LTCC (low‐temperature co‐fired ceramic) multi‐layer ceramic with thin‐film redistribution technology. In the 1990s, Motorola licensed IBM’s C4 technology, and developed eutectic flip‐chip (FC) and ball grid array (BGA) using organic boards, which made MTM mass production possible.

In around 1965, there were only wirebond packages. In early 1980, IBM showed the world their thermal conduction modules (TCMs) and controlled collapse chip carrier (C4) technology. Many people would probably agree that IBM then just started a new profession called IC packaging engineers. Since IC packaging is working intimately with the IC itself, and IC is a product from Moore’s Law, we sometimes call IC packaging...