The IEEE website places cookies on your device in order to provide you with the best user experience. By using our website, you agree to the placement of these cookies. To learn more, please read our privacy policy.

The first general-purpose microprocessor Intel 8080 was released in 1974 and can execute approximately 500,000 instructions per second. At the time, this seemed very lively.

Today, the most advanced descendants of the 8080 run at 100,000 times faster. This amazing advancement is a direct result of the semiconductor industry’s ability to reduce the size of the basic building blocks of microprocessors—many of its metal oxide semiconductor field effect transistors (MOSFETs) act as micro switches. Through the magic of photolithography, billions of them are usually built collectively on the surface of silicon wafers.

Over the years, as these transistors have become smaller and smaller, more transistors can be accommodated on the chip without increasing its overall cost. They also gained the ability to open and close at faster and faster speeds, allowing the microprocessor to buzz at higher speeds.

But shrinking the MOSFET to far beyond its current size (tens of nanometers) will be a daunting challenge. In fact, at some point in the next few years, it may become impossible to make them smaller due to basic physics rather than specific engineering. So people like me are always looking for other ways to increase their speed. In particular, we have been working hard to use compound semiconductors such as gallium arsenide to make them, which will enable such transistors to switch much faster than their silicon counterparts.

This strategy is by no means new. In fact, since the invention of the silicon MOSFET in 1960, engineers have been trying to come up with a version of gallium arsenide suitable for large-scale integrated circuits. No one has succeeded yet. These repeated failures have led to the oldest joke in Silicon Valley: gallium arsenide-it is the technology of the future, and it will always be.

But this long-standing skepticism may be about to disappear. My colleagues at the Berke Nanotechnology Center at Purdue University in West Lafayette, Indiana, and other researchers in industry and academia have recently made some progress and may soon allow the use of gallium arsenide or related Compound transistors are used in large-scale digital ICs. This ability will greatly help bring us microprocessors that can run at three or even four times the best speed today. Achieving this goal will undoubtedly require other improvements in semiconductor technology at the same time, but gallium arsenide or something close to it may be the key. No wonder some of us are unwilling to give up this extraordinary material.

The two main components of gallium arsenide come from the third column (gallium) and the fifth column (arsenic) on the right side of the periodic table, which is why the industry refers to it as a III-V semiconductor. There are more than a dozen such compounds, including gallium nitride and indium phosphide, but gallium arsenide is the most common example and is therefore the best studied. It currently accounts for approximately 2% of the semiconductor market.

The cost of gallium arsenide devices is much more expensive than silicon-made devices-the price of raw materials is about 10 times that of silicon devices-but they are very suitable for some special applications, including high-efficiency solar cells, laser diodes and a very special field ——Effect transistor: High electron mobility transistor or HEMT, used in mobile phones, communication systems and radars.

HEMTs are extraordinary devices because they overcome a fundamental problem of solid-state physics. Semiconductors, as the name suggests, usually do not conduct electricity very well. Usually, they must be doped with other kinds of atoms to conduct electricity. But these impurities often interfere with the movement of electrons through the semiconductor lattice, thereby limiting the conductivity that can be obtained.

In HEMT, electrons are introduced into III-V semiconductors not by doping but by contacting the material with another doped III-V compound. Essentially, the electrons fall a short distance into the undoped material, making one of its thin layers—the channel—conduct very well when the transistor is turned on.

HEMT can be used alone or in integrated circuits, for example, 100 or even 1000 clusters together, but they cannot be used in microprocessors yet. The problem is that too much electrons should flow from the source electrode of the transistor through the channel to its drain, but seep out from the control input electrode (gate), thereby generating heat. With millions of leaky transistors squeezed on the same chip, things will quickly become hot enough to melt.

In silicon MOSFETs, an intermediate insulating layer (traditionally silicon dioxide) prevents electrons from sliding into the gate from the channel. In HEMT, the channel is separated from the gate by a semiconductor, which, as you might expect, has a certain degree of conductivity. Of course, what is needed here is an insulator, but for decades, there has been no good gate insulator for gallium arsenide. Over the years, researchers seem to have discovered a promising material from time to time, but it was only recently realized.

It is easy to understand, at least in the general sense, why it is easier to find a gate insulator for silicon than for gallium arsenide. Silica is the natural oxide of silicon-a naturally occurring coating that grows when silicon is exposed to oxygen. Fortunately, silicon dioxide forms an excellent chemical bond with the silicon it covers: only one of the 100,000 silicon atoms on the interface fails to bond with the adjacent silicon dioxide, leaving so-called dangling bonds. These defects can disrupt the flow of electrons in the channel, but they are very rare and do not significantly degrade the performance of the transistor.

Gallium arsenide is another matter. When it oxidizes, it forms a complex mixture of Ga2O3, As2O3 and As2O5. Since the 1960s, some researchers have tried to use these natural oxides for gate insulators, but it turns out that this strategy is worse than useless, because natural oxides will produce various defects at the interface with gallium arsenide. Thereby destroying the conductivity of adjacent channels. Obviously, if there is any hope to make gallium arsenide MOSFETs for digital ICs, better materials need to be found.

Researchers have continued decades of exploration, testing candidates such as silicon dioxide, silicon nitride, silicon oxynitride, and aluminum oxide. They also tried to add a third material between the substrate and the insulator, such as sulfur, silicon or germanium, to eliminate the harmful effects of dangling bonds. However, the results were always disappointing, and by the early 1990s, most investigators simply gave up. Two notable exceptions are Minghwei Hong and Matthias Passlack of Bell Laboratories, who developed a method of depositing gallium oxide-gadolinium oxide composite insulators on III-V family substrates. Passlack was later a company that Motorola and Motorola peeled off. Improved this strategy in 2004, Freescale Semiconductor.

At that time, engineers who made silicon MOSFETs also began to encounter problems with gate insulators. As the size of these transistors shrinks, the silicon dioxide of the insulated gate no longer functions. In fact, it has become so thin that electrons can pass through it like a sieve. Huge efforts have been made to find alternative materials with higher dielectric constants that can be made thicker without affecting the electrical function of the transistor. In the end, a suitable compound was found. For example, Intel is now using hafnium-based gate insulators on some of its most advanced microprocessors [see "High-k solutions", IEEE Spectrum, October 2007].

To control the thickness of these "high-k" dielectrics (the name refers to the symbol used for the material's dielectric constant, the Greek letter kappa), manufacturers apply them to silicon substrates using a technique called atomic layer deposition. This is really clever, really. The trick is that you use a chemical carrier molecule that adheres to the target surface but not itself. Therefore, this chemical deposits a molecularly thick veneer. A second treatment with another carrier molecule removes the first carrier, leaving a diatomic thick layer of the desired material. The repeated application of two carrier gases, one alternating with the other, allows chip manufacturers to deposit various high-k gate insulators on silicon with atomic precision.

In 2001, my colleague Glen Wilk at Bell Labs and I decided to try atomic layer deposition to place a high-k gate insulator (in this case aluminum oxide (Al2O3)) on top of gallium arsenide. All the rage. After 2003, our team continued to study this method at Agere Systems in Allentown, Pennsylvania, a spin-off company of Bell Labs and Lucent Technologies, and our team moved there. This exercise was more successful than we thought.

This is not to say that we can directly use gallium arsenide to make a functionally perfect MOSFET. On the contrary, what shocked us early on was that atomic layer deposition allowed us to apply Al2O3, even though we did not take any measures to remove troublesome natural oxides from gallium arsenide. As researchers at the University of Texas at Dallas recently detailed, the reason is that the first carrier, a molecule called trimethylaluminum, will attack the natural oxide of gallium arsenide, even though all reasonable As a precaution, these oxides tend to cover the substrate. It is the atomic equivalent of the mold on the floor of the old porch. As any homeowner knows, if you want to repaint these boards, it is best to scrape off the trash first.

Using trimethyl aluminum is like having a one-piece product that can be stripped, primed and painted at the same time. If you want an additional coating—that is, a thicker Al2O3 film—just repeat the use of trimethylaluminum and second carrier water vapor in alternate steps.

Once you have grown an appropriately thick aluminum oxide layer on gallium arsenide in this way, you can use traditional photolithography techniques to build the drain, source, gate, and other components of the MOSFET. No special process is required. The problem is that the transistor you end up with will be a waste product: it will not pass more current through its path than some of the failed designs of the past few decades.

When I came to Purdue three and a half years ago, Yi Xuan, a postdoctoral researcher in the research group, took over the current problem of poor ability. Around that time, Intel announced that its engineers were seriously considering the use of III-V semiconductors in its future chips. IBM has also expressed interest in this technology. The pursuit of speed seems to be driving research on how to make III-V semiconductors for digital applications. However, despite the attention given by some big players in the industry, no one knows how to achieve sufficient current-carrying capacity for III-V MOSFETs. The challenge is greatest for those working in enhanced mode, which means that only when a voltage is applied to the gate will electrons flow from the source to the drain, as is the case with silicon MOSFETs in digital ICs.

Based on the published work done at Bell Labs nearly ten years ago and my own research on depletion MOSFETs, which turn off when a voltage is applied to the gate, I realized that the related III-V semiconductor-indium arsenide Gallium-will serve the channel better. In this compound, indium atoms replace gallium to an arbitrarily adjustable degree. In other words, you can mix mainly indium atoms, mainly gallium atoms, or mix these two kinds of atoms with arsenic atoms in a ratio of 50:50.

Modifying the indium content allows us to design the electronic properties of the substrate as needed, rather than trying to solve a given problem for a particular material. After a lot of experiments, we determined a composition with a 65:35 ratio of indium to gallium. With it, we can build a MOSFET that can carry more than 1 ampere per millimeter of channel width-this is the highest current density produced in 40 years of operation of a gallium arsenide MOSFET. In fact, it is so big that it initially made our semiconductor parameter analyzer beyond scale!

A well-known difficulty of this method is that the mechanical properties of indium gallium arsenide are so poor that using it to make wafers will be problematic, if not impossible. Pure gallium arsenide is more robust. Our wafer supplier, a British company called IQE, was able to overcome this obstacle by growing a thin layer of indium gallium arsenide on a thick indium phosphide substrate. These two compounds have crystal lattices of similar sizes, so they combine fairly well. Although the mechanical properties of indium phosphide are not ideal, they have proven to be sufficient for us to construct various test transistors.

Passlack and his colleagues at Freescale Semiconductor and the University of Glasgow have also been experimenting with indium gallium arsenide, using gallium oxide-gadolinium oxide insulators for the past few years. Hong, now at the National Tsing Hua University in Taiwan, also continues to study this combination. Although such MOSFETs have shown considerable current-carrying capabilities, they will be difficult to manufacture. The problem is that they require two applications of high-vacuum deposition techniques called molecular beam epitaxy: one is to deposit indium gallium arsenide, and then to coat it with gate oxide. Performing molecular beam epitaxy twice in the laboratory while maintaining a high vacuum is possible in the laboratory, but this will be a challenge for industrial-scale production.

Research teams from the National University of Singapore and IBM are looking for another design that has shown promise recently, a design that uses a chemical method to add a layer of amorphous silicon between the semiconductor and the gate insulator. This approach is similar to the strategy tried 20 years ago, and it is also similar to the work being done at the University of Texas at Austin and the State University of New York at Albany.

More importantly, some researchers are seeking to build a very different III-V field effect transistor suitable for digital applications, which can work without gate oxide at all. These devices work in a similar way to HEMTs because the semiconductor provides a barrier between the gate and the highly conductive, undoped channel. Intel and QinetiQ, headquartered in the United Kingdom, have achieved impressive performance over the past few years using transistors made in this way using indium-antimony channels. Jesús A. del Alamo and his colleagues at the Massachusetts Institute of Technology are also studying how to make this HEMT-like transistor smaller and less prone to gate leakage, so that it can one day be used in digital applications.

In fact, a lot of work is being done all over the world to introduce III-V semiconductors into the only field of silicon that has long been. In addition to the efforts of the industry, the academic team has also established research centers at the University of California, Santa Barbara, the University of Glasgow, and the University of Tokyo to conduct these investigations.

Before any of these new field-effect transistors can replace the slower silicon counterparts in microprocessors, memory chips, and other digital ICs, considerable progress must be made. In particular, in addition to current-carrying capacity and gate leakage current, device engineers must also optimize many parameters. For example, they also hope that these transistors can work at low voltages to reduce another troublesome heating source: the power consumed when the transistors switch states. (Indium-rich indium gallium arsenide holds great promise in this regard.) Designers also want to ensure that the current that flows when the transistor is nominally "off" is very small, so that no power is consumed—nor Generate heat-useless. Doing so, while making these transistors as small as today’s silicon miracles, will be no small feat.

In addition, manufacturers may have to find ways to place III-V semiconductors on silicon wafers. In other words, the goal of chip manufacturers must be to use compound semiconductors only where they are needed, rather than trying to completely replace silicon, although getting all these materials to work properly is a tricky task and the subject of a lot of research.

One of the reasons to keep as much silicon as possible is that it has better physical properties and can be used to make large wafers used in semiconductor manufacturing. In addition, silicon is cheap and environmentally friendly, while gallium arsenide is expensive, and because it contains arsenic, it can be toxic.

Another reason for not expecting an all-gallium arsenide microprocessor to appear soon is that III-V semiconductors can only accelerate half of the transistors in CMOS chips: n-channel transistors, which transport current in the form of negative charges-electrons. CMOS integrated circuits require a combination of n-channel and p-channel MOSFETs, which consume power together only when they switch states, such as when the n-channel transistor is turned on and the p-channel transistor connected in series with it is turned on. When not switching between states, this complementary pair does not consume power, which is why CMOS chips are so energy-efficient.

Although gallium arsenide allows electrons to pass through it particularly easily, it does not have any advantage over silicon in terms of positive charge carriers-"holes", which are locations in the semiconductor lattice that lack shell electrons. Therefore, it will be very difficult to manufacture high-performance p-channel MOSFETs using gallium arsenide or other III-V compounds. The current consensus is that the semiconductor industry may use germanium in these transistors. For example, the Duallogic Academic Industry Alliance in Europe is working to combine germanium and III-V semiconductors in this way.

The III-V devices that my Purdue colleagues and I recently built represent a huge leap because these MOSFETs are both easy to manufacture and capable of carrying record currents. Competing designs also provide some attractive features. Nevertheless, there are still many obstacles preventing their widespread use. In particular, chip manufacturers must learn to mix and match some very different semiconductors on a single wafer. Perhaps chipmakers will have to piece together indium gallium arsenide and germanium on a silicon bed, or it will become more complicated. But if there is any lesson to be learned from the dizzying advances in computing power over the past four years, it is that the industry is thriving amidst challenges.

PEIDE D. YE is an associate professor of electrical engineering at Purdue University, where he is called Peter. He adopted this nickname at Bell Labs, where he and many of his semiconductor research colleagues were required to wear special blue suits and headscarves in clean rooms, which made it difficult to determine who was who. In this place they call the "Blue Zoo", having a name that Westerners can easily recognize proved to be an absolute advantage.

Details of the author's work in this field can be found in "High-performance inverted enhancement mode InGaAs MOSFET with maximum drain current exceeding 1 A/mm", authors Y. Xuan, YQ Wu and PD Ye, IEEE Electron Device Letters 29: 294, April 2008.

To learn more about the use of gallium oxide-gadolinium oxide insulators on III-V family substrates, see "High Mobility III" by M. Passlack, R. Droopad, K. Rajagopalan, J. Abrokwah, P. -V MOSFET technology". Zurcher, R. Hill, D. Moran, X. Li, H. Zhou, D. Macintyre, S. Thoms and I. Thayne at http://www.gaasmantech.org/Digests/2007/2007 Papers/12c.pdf .

Does more than 100 mobile robots indicate that everyday robots are inevitable?

Last week, Google, Alphabet or X or whatever company you want to call it announced that its Everyday Robots team has grown enough and made enough progress. Now it’s time to make it its own thing. Now you guessed it. You guessed it, "Everyday Robots." There is a problem with the design of a new website, and there are a lot of fluffy descriptions about the content of Everyday Robots. But fortunately, there are some new videos and enough details about the engineering and team approach, it’s worth spending a little time trekking through the chaos to see what Everyday Robots has done in the past few years and what plans they have done. For the recent.

The place near the arm does not seem to be suitable for placing an emergency stop, right?

Our headline may sound a bit acrimonious, but the headline of Alphabet’s own announcement blog post is "Every day robots leave the laboratory (slowly)." It is not so much a sarcasm, but rather an admission that the mobile robot is in a semi-structure. Effective operation in a chemical environment has been and will continue to be a huge challenge. We will go into details later, but the high-level news here is that Alphabet seems to have invested a lot of resources behind this effort, has a long time span, and its investment is beginning to pay off. Considering the random state of Google Robotics over the years (at least from the appearance), this is a nice surprise.

According to Astro Teller, who is in charge of Alphabet's moon landing program, the goal of Everyday Robots is to create "a universal learning robot", which I think sounds enough. To be fair, they deployed a lot of hardware, Hans Peter Brøndmo of Everyday Robots said:

This is a lot of robots, which is great, but I have to question the actual meaning of "autonomy" and the actual meaning of "a series of useful tasks". For us (or anyone?), there is really not enough public information to evaluate what Everyday Robots does with its 100 prototype fleets, how much robot support is needed, the constraints of their operation, and whether to call their work." "Useful" is appropriate.

If you don't want to browse Everyday Robots's weirdly over-designed website, we have extracted good things (mainly videos) and reposted them here, along with some comments below each.

0:01 — Is it just me, or does the gearing behind these actions sound a bit, eh, unhealthy?

0:25-I think the claim of winning the Nobel Prize by picking up the cup from the table is a bit exaggerated. Robots are very good at sensing and grasping cups on the table, because this is a very common task. Like, I understand, but I just think there are better examples to illustrate the current problems of humans and robots.

1:13 — It’s not necessarily useful to make an analogy between a computer and a smartphone and compare them to a robot, because certain physical realities (such as motors and operating requirements) prevent the kind of zoom that the narrator refers to.

1:35 — This is a red flag for me, because we have heard about "this is a platform" many times before, but it has never been successful. But in any case, people continue to try. It may be effective when limited to the research environment, but fundamentally, “platform” usually means “making it do (commercially?) useful things is someone else’s problem. I’m not sure if this has ever been A successful model of a robot.

2:10-Yes, okay. This robot sounds much more normal than the one at the beginning of the video; what's the matter?

2:30 — I am a big fan of Moravec's Paradox, and I hope that when people talk about robots with the public, it will be mentioned by more people.

0:18-I like the door example because you can easily imagine how many different ways it can be disastrous for most robots: different levers or knobs, different glass positions, variable The weight and resistance, then, of course, the threshold and other similar annoying things.

1:03-Yes. I can't emphasize enough, especially in this case, computers (and robots) are really bad at understanding things. Know things, yes. Know them, not so much.

1:40 — People really like to cast a shadow on Boston Dynamics, don't they? But this seems unfair to me, especially for companies that Google once owned. What Boston Dynamics is doing is very difficult, very impressive, come on, very exciting. You can admit that when you are dealing with different difficult and exciting problems yourself, others are dealing with difficult and exciting problems, and don’t feel a little annoyed by what you do, for example, not so flashy Or whatever.

0:26 — It doesn't make sense to say that the robot is low cost, without telling us how much it costs. Seriously: the "low cost" of a mobile manipulator like this is easy (and almost certainly is) at least tens of thousands of dollars.

1:10 — I like to include things that don’t work. Everyone should do this when presenting a new robotics project. Even if your budget is unlimited, no one can always do everything well, and we all know that others are as flawed as we are, and we will all feel better.

1:35 — When talking about robots trained using reinforcement learning techniques, I personally avoid using words such as "intelligence" because most people associate "intelligence" with the kind of basic world understanding that robots don't really have .

1:20 — As a research task, I think this is a useful project, but it is important to point out that this is a bad way of automatically sorting recyclables from trash. Since all the garbage and recyclables have been collected and (probably) taken to several centralized locations, in fact you only need to have your system there, where the robots can stand still and have some control over their environment And do better and work more efficiently.

1:15-Hope they will talk more about this later, but when considering this montage, it is important to ask in the real world what tasks do you actually want the mobile manipulator to perform, and which do you only want to perform Tasks are automated in some way because they are very different things.

0:19 — It may be a bit premature to talk about ethics at this point, but on the other hand, there is a reasonable argument that it is not too early to consider the ethical significance of robotics research. To be honest, the latter may be a better point of view, and I am happy that they are thinking about it in a serious and positive way.

1:28 — A robot like this will not take your job away. I promise.

2:18 — Robots like this are not the robots he is talking about here, but the point he puts forward is very good, because in the short to medium term, robots will become the most valuable role, and they can increase what humans can do by themselves. Things are not to completely replace human beings to increase human productivity.

3:16 — Again, the idea of ​​that platform...blarg. The whole "someone wrote these applications" thing, uh, who the hell is it? Why are they doing this? The difference between a smartphone (which has a lucrative application ecosystem) and a robot (which does not) is that there are no third-party applications at all, and the smartphone has enough useful core functions to justify its cost. It will take a long time for robots to reach that point, and if software applications are always someone else’s problem, they will never get there.

I am a little upset about this whole thing. A fleet of 100 mobile manipulators is amazing. It is also great to invest money and manpower to solve robotics problems. I'm just not sure whether the vision of the "daily robot" we are asked to buy must be realistic.

The impression I got from watching all these videos and browsing the website is that Everyday Robot wants us to believe that it is actually working hard to bring the universal mobile robot into the daily environment in the way people (outside of the Google campus) can benefit from it. Maybe the company is working towards that exact goal, but is it a practical goal? Does it make sense?

The ongoing basic research seems very solid; these are definitely difficult problems, and solving these problems will help promote the development of this field. (If these technologies and results are released or shared with the community in other ways, then these advancements may be particularly important.) If the reason for this work in a robotic platform is to help stimulate this research, that would be great, I There is no objection to this.

But I really hesitate to accept the vision of this universal home mobile robot to perform useful tasks autonomously. This approach may be of great help to anyone who actually watches the Everyday Robotics video. Perhaps this is the whole point of the vision of the moon landing-try hard to do something that will not be rewarded for a long time. Again, I have no problems with this. However, if this is the case, Everyday Robots should pay attention to how it puts its efforts (and even its success) in context and portrays it, why it works on a specific set of things, and how external observers should set it Our expectations. Time and time again, the company's commitment to useful and affordable robots is too high, but delivery is insufficient. I hope that Everyday Robots will not make the same mistake.

Here are ways to encourage daughters to pursue STEM careers

In my 2016 article "Fathers' Views on Daughters and Engineering", I shared my disappointment at the lack of role models and cultural information that made my two smart daughters — and many of their female friends — -To pursue an engineering career.

After the article was published, I received an email from Michelle Travis, she was writing a book about fathers and daughters. She wanted to know my thoughts on creating stronger channels for girls to pursue careers in science, technology, engineering, or mathematics (STEM), and what can be done to change the narrative of engineering to highlight their public service role.

Travis is a professor at the University of San Francisco Law School, where she co-directs her work law and judicial projects. She researches and writes articles on employment discrimination laws, gender stereotypes, and work/family integration. She is also a founding member of the Work and Family Researchers Network and serves on the board of non-profit organizations.

Her latest book, "Dads for Daughters" (Dads for Daughters) is a guide for male allies to support gender equality. (I am one of the fathers who appear in this book.) She wrote that my mother who won the prize has two jobs, which is a children's picture book celebrating working mothers.

Over the years, we have kept in touch, followed each other's work, and looked for other ways of cooperation.

In the past few months, I have been frustrated by the news that girls from certain countries are either not allowed to go to school or risk safety even if they are officially allowed to go to school. This is one reason why I feel I need to talk to Travis and learn from her what else can be done to change fathers and men's perceptions of women's abilities and women's success in almost all fields (including engineering).

Last month, I asked her a few questions about her book and what her father can do to better support women. In the next interview, she gave a sneak peek and listed some resources for engineering dads who wish to encourage their daughters to pursue STEM careers.

QA: As a lawyer, why did you decide to research and write articles about fathers and daughters? Is it personal?

MT: My interest in making my daughter's father an advocate for gender equality is both professional and personal. As a lawyer and law professor, I have been using legal tools for years to promote equality for women in the workplace-seeking stronger employment discrimination laws, equal pay for equal work, and family leave policies. Over time, I realized that the law has limits on what it can accomplish. I also realized that we are asking women to do too much heavy work to break down barriers and break the glass ceiling. Most importantly, I realized that to make progress, you need the commitment of a powerful male leader.

I started to ask myself how women can get more men to participate in gender equality work. At the same time, I noticed the powerful influence of my two daughters on my husband. He has always regarded the equality of women as an important goal, but it was not until he started to think about the world his daughter entered that he completely internalized his personal responsibility and influence. With a daughter, he is eager to take action. He wants to be an outspoken advocate for girls and women, not just bystanders.

"The father of an engineer has a unique position and can be an ally in expanding opportunities for girls and women."

Watching this transition prompted me to study the father-daughter relationship. I found that my husband's experience is not unique. Researchers found that having a daughter tends to increase men’s support for anti-discrimination laws, equal pay policies, and reproductive rights, which tends to reduce men’s support for traditional gender roles. This has a significant impact in the workplace. For example, fathers of daughters are more likely to support gender diversity than other male leaders. Compared with CEOs run by non-father men, CEOs who are daughter fathers tend to have a smaller gender pay gap in the company.

Of course, many men without daughters are allies of women, and not all fathers with daughters are advocates of gender equality. We have even heard some men-including famous politicians-citing their "daughter's father" in a dishonest way.

But fathers of most daughters are genuinely interested in promoting equal opportunities for girls and women. This makes the father-daughter relationship an excellent entry point for inviting men to form partnerships to build a fairer world.

QA: Why do people want to read your book?

MT: Today's fathers are training self-confident and capable daughters who believe that they can achieve anything. But the world is still unequal, the workplace is run by men, there is a gender pay gap, and deep-rooted gender stereotypes. My book celebrates the role father can play in creating a better world for the next generation of girls.

Inspired by their daughters, fathers are fully capable of becoming powerful allies for girls and women. But in the post #MeToo world, it may be difficult for men to step in and speak out. This is where the father of the daughter can help. It provides fathers with the data they need to advocate for gender equality. It also provides specific strategies to illustrate how they play a role in various fields, from sports fields to science laboratories, from conference rooms to ballot boxes.

In addition to serving as a guide, it also shares the stories of fathers who have joined the battle. All the men who emphasized praised their daughters for inspiring them to pay more attention to gender equality. These include a CEO who invests in female entrepreneurs to manage parts of his company's supply chain, and a lawyer who sets up a part-time position in his company-which allows women to maintain partnerships. Another head coach hired the first female assistant coach in the NBA. Another governor broke the partisan line and signed a bill to expand the rights of sexual assault victims. An engineer provides computer skills training to support girls who have become victims of sex trafficking in India. In addition, there is a teacher, a U.S. Army colonel, a plumber, a firefighter, and a construction contractor who have joined forces to fight for equality in the girls’ high school sports program.

All these fathers, and many others, are inspired by their daughters to support gender equality. Their stories can inspire other dads to participate. Fathers who are committed to seeing their daughters realize their dreams have the opportunity to improve the world their daughters will enter, and fathers born for their daughters will support them in this journey.

QA: What do you think is the difference between engineer fathers and other fathers, and why?

MT: Being an engineer's father has a unique advantage and can be an ally in expanding opportunities for girls and women. We all know that there is a huge gender imbalance in the STEM field. This leads to a large loss of talents. Daughters’ fathers can take small but influential steps in their homes, communities, and workplaces, welcoming more girls and women into engineering careers.

At home, the father can fill the home with books, toys and activities, so that the girl can imagine that she is the engineer of the future. The father of engineering has created some great resources for this. For example, Greg Helmstetter found that his daughter lacked an engineering role model, so he created the STEAMTeam 5 series of books, which shared the adventures of five girls using STEM skills to meet challenges. Inspired by his daughter, Anthony Onesto created the Ella the Engineer comic book series, which depicts a superhero girl who uses her engineering knowledge to solve problems and save the world.

Other excellent children's books include Rosie Revere by Andrea Beaty, engineer, and Tanya Lee Stone's "Who Says Women Can't Be Computer Programmers?" With Mike Adamick's father's Book of Awesome Science Experiments. Daughter’s dad can also follow Ken Denmead’s GeekDad blog, check the Go Science Girls website, and purchase one of Debbie Sterling’s GoldieBlox engineering suites for his daughter’s next birthday.

As an engineer, dads can have a broader impact in their communities by volunteering with girl technology organizations such as EngineerGirl, TechGirlz, Girls Who Code, Girl Develop It, or CoolTechGirls. These organizations are always looking for engineers to share their expertise and passion for STEM careers with talented young girls.

Engineer fathers can also become gender equality leaders in their workplaces. Hiring, mentoring, and funding women is a key step in expanding women’s representation in the engineering field. Dads can further support women by joining programs such as the Million Women Mentoring Program or cooperating with IEEE Women in Engineering or the Society of Women Engineers. The empathy that fathers gain from their daughters can also enable them to create a safer workplace culture by fighting hostile work environments and fighting gender prejudice.

QA: From the perspective of an adult daughter, what makes a father different from a husband or friend?

MT: In a recent survey, fathers listed strength and independence as the primary qualities they wish to instill in their daughters-this is different from the characteristics that men value their wives most. From the perspective of the daughter, this can make the father a particularly effective ally on their behalf.

When the father is involved in the daughter's life, this relationship can have a profound impact. Participating fathers will produce women who are more confident, self-esteem, and mentally healthy. Girls supported by their fathers have stronger cognitive abilities and are more likely to go to school and achieve greater financial success. The fathers involved also helped their daughters establish healthier adult relationships with other men.

For fathers, daughter relationships are a powerful way to build men’s empathy and raise men’s awareness of gender discrimination and gender inequality. For example, men generally understand the challenges of work/family integration better when they look at their adult daughters taking care of the needs of their careers and their mothers.

The failure mode, impact and diagnostic analysis (FMEDA) sets the safety and reliability calculation standard of the automatic protection system to IEC61508. However, FMEDA results are only as good as the failure rate data used to create them. The new component reliability database (CRD) overcomes limitations and improves accuracy.