Silicon photonics: will light replace electricity?
All-semiconductor CW laser solves previously insurmountable problem of two-photon absorption
Microelectronics already faces physical limitations (at the atomic level) in transmitting electrical signals between microcircuits. Possible Solution This problem may be the development of non-traditional technologies, in particular, silicon photonics.
Intel has already created many of the structures needed to make signaling between chips using light as easy as electrons now do. The main problem for this was the lack of a suitable light source. Recently, Intel announced a new breakthrough in this area, the first all-semiconductor continuous wave laser using a physical phenomenon called the Raman effect (in quantum mechanics, the Raman effect is described as the exchange of energy between scattering molecules and incident light), and built using standard commercial CMOS -crystals.
Using the power of semiconductors, Intel researchers were able to realize the functionality of a traditional, bulky Raman laser that uses glass and is typically the size of a suitcase by shrinking it down to the thickness of a single track on a silicon wafer.
This breakthrough in silicon photonics will lead to practical and affordable solutions for communications and computing, to the creation of new medical equipment and sensors, and the tunable semiconductor laser can replace its predecessors that cost hundreds and thousands of dollars. This achievement may also lead to the acceleration of the creation of new optical interconnects between microcircuits and external devices, because thin optical fibers take up less space than electrical cables and will provide Better conditions cooling of computers and servers.
The semiconductor laser demo wafer was manufactured using standard CMOS technology on an existing production line. This means that for these new technologies, the path from laboratory to production may not be long and complicated, as is the case for some non-traditional technologies, but rather direct and fast.
Photonic logic will not yet replace semiconductor logic, but it can already be used for data transmission. Both between devices and between processor cores.
Looking at the recent announcement of Apple's hardware innovations, one would like to say that new technologies are like tropical greenery: yesterday there was a small stunted shoot, and today it is already a powerful vine that has deeply taken root and firmly embraced the market trunk of computer technology with its shoots.
The appearance of the first "poppies" with the Thunderbolt interface was received with curiosity, but nothing more. Also at one time, the market looked at the outlandish FireWire port in Apple laptops PowerBook 3G.
The subsequent inclusion of Thunderbolt, combined with Display Port, in almost all Apple computing equipment made peripheral manufacturers seriously think about supporting this technology. Benefit a new controller designed by Intel, supports both Thunderclap and the USB 3.0 specification at the same time. And if everything is clear with the latest interface, then Thunderbolt is full of mysteries. What?
Well, for example, from the series "What's in my name to you?". After all, Thunderbolt is the market name for Intel's Light Peak Research Technology, where keyword is light - light. The ten gigabits per second that Thunderbolt now offers to the consumer, transmitting data over copper wires over a distance of up to three meters, is truly flowers in comparison with the fifty gigabits per second that Light Peak provides over a hundred meters of optical cable.
The appearance of an optical version of Thunderbolt is a matter of the near future. The future, in which, along with microelectronics familiar to us, the "queen of light" - photonics - will begin to help process data.
You can read about how Intel uses photonics in Silicon Photonics Link high-speed data exchange technology in the article "Download in a second: advances in silicon photonics".
Intel's Silicon Photonics Solutions Will Deliver Fifty Gigabits Per Second of Computer-to-Peripheral Interface Bandwidth
It's time to look at the components of systems based on silicon photonics in more detail. Systems, because Intel solutions are far from the only ones. And, most importantly, today it is no longer just laboratory exercises. Silicon photonics has acquired all the necessary capabilities and is quite ready to fruitfully cooperate with existing microelectronic solutions.
An example of such cooperation is the hero of this material - an IBM project with the apt name SNIPER (Silicon Nano-Scale Integrated Photonic and Electronic Transceiver).
Photonics. Building blocks of technology
Can photonics completely replace electronics in microcircuitry? Probably not. The propagation of light is based on the laws of optics, which imposes significant restrictions on the design of such basic components as transistors, capacitors and diodes. No, attempts to develop optical analogues of the transistor have been made for a long time, and even today they do not stop. Only now they cannot compete with the proven CMOS technology.
The photonic transistor circuit was proposed back in the eighties of the last century.
Where photonics really excels is in realizing high-speed links between components. digital circuits. That is, in those places where the electronics begin to slip more and more actively. An increase in the degree of integration of microcircuit components affects the dimensions of the metal conductors connecting them. With the transition to twenty-two nanometer technological process CMOS engineers were faced with the problem of transients in miniature copper bars. These phenomena can easily lead to errors in the operation of a complex computer system densely packed into a silicon chip.
The use of photonics as a communication medium for microcircuits allows technologists to simultaneously rid new chips of the influence of transient processes in copper conductors and significantly reduce the heating of the microcircuit. Unlike electrons that unproductively convert their energy into heat, photons, moving along an optical conductor, do not dissipate heat at all.
So, a compromise solution is a combination of electronics and photonics. Electronics remains the basis of digital circuitry, while photonics takes on the role of a universal conducting medium.
What is needed for such an environment? First of all, the source of photons is a laser. Next - a conducting medium through which photons can propagate inside the microcircuits - waveguides. In order for the zeros and ones formed by electronic components to turn into a luminous flux, and for the reverse conversion, modulators and demodulators will be required, but, of course, not simple ones, but optical ones.
Well, in order to achieve the high bandwidth required by the channels of current integrated circuits, multiplexers and demultiplexers (also, of course, optical ones) will be required. Moreover, all these components must be implemented on the same silicon base, which is used for CMOS technology.
The development of these "bricks" is the way that silicon photonics has been going for the last twenty years. During this time, a lot of unique solutions were proposed, which were the very "sum of technologies" that allow photonics to move to a qualitatively new level. The level of integrated optical electronic circuits.
Silicon lasers
Actually, the phrase "silicon laser" is an oxymoron. Being a so-called indirect-gap semiconductor, silicon is completely incapable of emitting light. This is why fiber optic telecommunications uses solutions based on other (direct-gap) semiconductors, such as gallium arsenide. At the same time, silicon is excellent for creating waveguides and detecting optical signals in electrical ones.
So what's the problem? You can use a laser external to the silicon circuit or develop a hybrid circuit based on silicon and, for example, the same gallium arsenide. But neither solution can be considered effective. In the case of using an external laser (and this is done in modern macro-level fiber optic systems), it is practically impossible to accurately calibrate the beam with respect to a nanometer-sized waveguide at the micro level. The inclusion of gallium arsenide in the technological process for the production of CMOS chips failed. Too different production conditions are needed for these two semiconductors.
So what, a silicon laser will never see (more precisely, not emit) light? Of course not. Silicon can be made to shine by applying various tricks. For example, dope it with a material that will emit photons for silicon. Or change the structure of silicon itself in such a way that it will be forced to light up. The third way is to apply Raman scattering of light (it is also called Raman scattering), which temporarily turns silicon into an almost direct-gap semiconductor.
One way to make silicon glow is to create a porous silicon structure.
Scheme and micrograph of a laser based on Raman scattering
Currently, scientists have achieved the greatest success in the field of silicon doping technologies. The best-known implementation of a continuous-wave silicon laser based on them is a laser developed by Intel in collaboration with the University of California, Santa Barbara. The scientists managed to “glue” a direct-gap semiconductor indium phosphide to a silicon waveguide using oxide. The thickness of the "glue" in this case is only 25 atoms. By creating a potential difference between silicon and indium phosphide (this is called "electric pumping"), they achieved the formation of photons that penetrate through the "glue" into the silicon waveguide.
Schematic diagram of a CW hybrid silicon laser
On the basis of such a scheme, versions of a hybrid silicon laser with different wavelengths (in the infrared range, transparent to silicon) are created, which makes it possible to implement a multichannel communication system.
Silicon modulators
The photon flux emitted by a silicon laser can be represented as a carrier frequency that needs to be modulated with a binary signal.
Optical modulators were considered impossible until scientists decided to use the phenomenon of light interference. In general, a modulated optical signal can be obtained by the interference of a reference beam of light and a beam that has passed through a material that changes the refractive index under the influence of an electric current (the so-called electro-optical effect). Unfortunately, silicon let us down here too - its symmetrical crystal lattice does not allow realizing the electro-optical effect. Doping again came to the rescue.
Scientists bifurcated a silicon waveguide and built up a layer of silicon nitride on one of its arms, which stretched the silicon crystal lattice. Applying a voltage to this section leads to the refraction of light in this arm of the waveguide. At the same time, in the other arm, the same flow propagates without distortion.
Photomicrograph of a section of the light refraction arm in a Mach-Zehnder modulator
Implementation of the entire Mach-Zehnder modulator and its variants.
Combining these streams at the output leads to their interference, while the output stream will be modulated by applying a voltage to the arm of the silicon nitride waveguide. Scientists didn't have to reinvent the wheel. A similar effect is widely used in Mach-Zehnder interferometers. Therefore, silicon modulators and demodulators were named exactly the same.
Silicon multiplexers
A plurality of modulated light fluxes from a plurality of lasers with different wavelengths can significantly increase throughput communication channel due to parallelization of data transmission. But how can this set of threads be combined into one? Yes, and in such a way that at the output the resulting total flow can again be divided. This is where multiplexers come in handy. Optical, of course.
The idea of an optical multiplexer based on an array of waveguides (AWG)
Micrograph of an AWG Multiplexer
Optical multiplexer based on a cascade of Mach-Zehnder modulators
At present, the technology of microminiature multiplexing of light by means of its spectral multiplexing (WDM - Wavelengths Division Multiplexing) has been proposed. Most often, for its implementation, a diffractive structure based on an array of waveguides and mirrors (AWG - Arrayed Waveguide Grating) is used, in which each beam of light moves along its own waveguide, curved in accordance with its wavelength. Closing, these waveguides give the resulting spectrally densified flow. Another common solution is to use a cascade of Mach-Zehnder modulators already known to us.
IBM SNIPER. Silicon terabit
Silicon Photonics Solutions Offered by Intel to Promote Photonic Technology in Interfaces peripherals. The closest commercial prospect is a fifty-gigabit optical version of the Thunderbolt interface (perhaps it will be called differently by the time of industrial implementation). In the longer term, Intel is considering increasing throughput to two hundred gigabits per second. To say that it is fast is to say nothing: for example, the content DVD disc at this rate, it can be transmitted in one second.
Exactly the same goal was set by IBM Research. Set and got it! True, IBM plans to use its terabit not in communication interfaces, but in high-speed buses connecting the cores of a multi-core processor.
Internuclear communication based on silicon photonics
The idea of the SNIPER project from IBM Research (the photonic part of the circuit is shown in blue)
The SNIPER project is a practical implementation of the idea of nanophotonics, which uses the "building blocks" discussed above to create a photonic communication network. This photonic network is integrated on top of a multi-layer system-on-a-chip "pie" that includes a multiprocessor module and a module random access memory. Having outputs to the outside, such a network connects this system on a chip to a high-speed optical data bus connecting the processor with peripherals. The internal waveguide wiring provides data routing between the cores of the processor module.
Six-channel photonic module of the SNIPER project
The SNIPER project currently boasts the implementation of a six-channel photonic transceiver module using hybrid silicon lasers, Mach-Zehnder modulators, and a waveguide array multiplexer. The bandwidth of each channel of this transceiver is twenty gigabits per second. Fifty such channels are implemented on a 25 square millimeter substrate, which provides the same terabit of bandwidth.
Project SNIPER photonic chip delivering terabit throughput
Most importantly, SNIPER is no longer a research project. Libraries of all elements of photonics for silicon lithography have been worked out for the production cycle. As well as the method of their integration with the CMOS logic of the system on a chip.
Where will this solution be applied first? Of course, in supercomputer systems and data centers cloud computing. Where the computing power of electronic circuits is most needed is to communicate at the speed of light.
However, one can be sure that the expansion of silicon photonics into consumer computing is not far off. Everything will start with interfaces for connecting peripherals, and then, you see, the buses for multi-core solutions will catch up, turning boring silicon inside our processors into a magic crystal sparkling with all the colors of the spectrum.
The past 2007 was a very successful year for the development of many Intel technologies, including those in the field of silicon photonics. The latest breakthrough achievements of Intel in this area were compared by the MIT Technology Review magazine with a triple win at the races - this is how the reviewers of the leading publication assessed a series of official announcements by the corporation.
As Justin Rattner, chief technology officer and head of Intel's Corporate Technology Group, said, “We have demonstrated empirically that manufacturing technologies compatible with silicon CMOS technology enable the creation of semiconductor optical devices. The proof of this fact was a huge achievement, but no less significant steps are needed for the further development of this technological direction. Now we need to learn how to integrate silicon photonics devices into standard computer components; we are not yet able to do this. But at the same time, we continue to actively work with our product development teams to provide manufacturers with models for using semiconductor photonics in Intel solutions.”
Researchers at Intel have developed the world's first semiconductor chip capable of producing a high-quality continuous laser beam. Eight lasers are integrated into one silicon chip.
Silicon photonics as a means to eliminate bottlenecks on the way to the era of tera computing
Silicon photonics is the most important component long-term development strategy of the Corporate Technology Group, aimed at accelerating the transition to tera computing. The point is that as development multi-core processors with enormous computing power, engineers face new challenges. For example, the need for data transfer speed between the memory and the processor will soon exceed the physical limits imposed by copper conductors, and the transfer rate of electrical signals will become less than the speed of the processor. Even now, the performance of powerful computing systems is often limited by the speed of data exchange between the processor and memory. Today's data transfer technologies are designed for much lower bandwidth than photonics, and as the distance over which data is transferred increases, the transfer rate becomes even lower.
“It is necessary to bring the speed of data transfer between the components of the computing platform in line with the speed of the processors. This is indeed a very important task. We see silicon photonics as a solution to this problem, and therefore we are pursuing a research program that puts us at the forefront of this field,” said Kevin Kahn, Distinguished Research Engineer at Intel Corporation.
Tests of a prototype optical memory module have shown that light, rather than electricity, can be used to access the server's memory.
A team led by Intel Lead Optics Researcher Drew Alduino is building an optical communication system between processor and memory for Intel platforms. A test platform has already been created based on a fully buffered FB-DIMM memory, on which it boots and runs Microsoft Windows. The working prototype is proof of the possibility of connecting memory to the processor using optical communication lines without compromising system performance.
Creating a commercial version of such a solution has huge benefits for users. Optical communication systems will eliminate the bottleneck associated with the difference in memory bandwidth and processor speed, and improve the overall performance of the computing platform.
From research to implementation
The Photonics Technology Lab, run by Distinguished Intel Research Engineer Mario Paniccia, has proven that all components for optical communications—laser, modulator, and demodulator—can be fabricated from semiconductors using existing manufacturing techniques. The PTL has already demonstrated key silicon photonics components operating at record performance, including modulators and demodulators capable of data rates up to 40 Gbps.
Semiconductor photonics technology requires six main components:
- a laser that emits photons;
- a modulator for converting the photon stream into an information stream for transmission between elements of the computing platform;
- waveguides that play the role of "transmission lines" for delivering photons to their destinations, and multiplexers for combining or separating light signals;
- case, especially necessary for the creation of assembly technologies and low-cost solutions that can be used in the mass production of PCs;
- a demodulator for receiving photon streams carrying information and converting them back into an electron stream available for processing by a computer;
- electronic circuits to control these components.
%%%
The implementation of all these components of optical communication based on semiconductor technologies is widely recognized as the most important research problem, the solution of which will lead to a huge technical breakthrough. PTL has already set a number of world records by developing high performance devices, modulators, amplifiers and demodulators that deliver data rates up to 40 Gbps. Over the next five years, Intel will look for ways to integrate these components into actual products.
One of the key components of silicon photonics is a modulator that provides transmission rates up to 40 Gbps.
In the field of semiconductor photonics, Intel has already reached the finish line. Research in the field of integration of optical elements has already moved from the stage of scientific or technological development to the stage of creating commercial products. The research team is now in the process of identifying the capabilities and specifications for designing innovative products based on this revolutionary technology. Ultimately, Intel creates prototypes and works closely with product development teams to accelerate adoption. new technology.
In addition to its own activities, Intel Corporation is funding some of the most promising research in this direction outside of CTG - in particular, it is collaborating with the University of California at Santa Barbara, which is developing a hybrid semiconductor laser. Talented graduates from various universities from other countries are also trained in the PTL laboratory.
Intel Lead Optics Researcher Richard Jones said, “We have two major challenges ahead of us in our hybrid semiconductor laser project. First, we must move the experimental production of hybrid lasers from the University of California to the Intel plant. Secondly, we have to combine a hybrid laser, a high-speed semiconductor modulator, and a multiplexer to prove that we can create a single optical transmitter based on CMOS-compatible manufacturing technology.”
The introduction of silicon photonics technologies will include the development of new manufacturing processes for manufacturing lasers on a large scale. The success of Intel Corporation in the field of photonics will allow it to significantly outperform potential competitors. PTL has already registered about 150 patents. The most prestigious publications, such as Nature, have noted the unprecedented achievements of Intel specialists. In addition, in 2007, Intel was awarded the EE Times ACE Award for Most Promising New Technology.
In pursuit of photons
Unlike the existing well-established and decades-old processes for manufacturing transistors, the technology for creating elements for semiconductor photonics is completely new. Certain problems stand in the way of its implementation: optimization of devices, increasing the reliability of the design, testing methodology, ensuring energy efficiency, and developing subminiature devices.
40 Gigabit Silicon Laser Modulator Test Bench
One of the most important problems is optimization, because the PTL laboratory develops optical devices for mass computing. While there are no other similar products, standards, or other benchmarks, it is up to engineers developing a new process to find solutions that best meet the needs of computer applications.
Currently, the PTL research team, relatively small by the standards of photoelectronics, is gradually switching to the commercialization of semiconductor photonics solutions and expects that the mass adoption of this incredible technology can begin as early as 2010. A group of optics specialists from the Digital Enterprise Group (DEG) under under the leadership of Victor Krutul, is developing applications that will provide the basis for the emergence of new technology. "We believe that Intel's products will continue to conform to Moore's Law through the development of optical communications," says Krutal.
When to transfer information between components of the same computing platform and between different systems not electrons, but photons will be used, the next computer revolution will be accomplished. Leading electronics manufacturers around the world are already joining this race to gain a competitive edge. The significance of the new technology can be compared to the invention of integrated circuits. Intel is leading the way in this research and in the development of semiconductor photonics components.
IBM announced a breakthrough in silicon photonics with the creation of the first fully integrated multiplexed chip. The new device will allow individual chips to communicate with each other using optical rather than electromagnetic waves, which will increase throughput to 100 GB per second and beyond. This chip resides on a single silicon chip and is critical to the long-term adoption of microscale optical technologies. But why did powerful companies like IBM and Intel spend decades studying silicon photonics?
In theory, with the help of silicon photonics, many of the serious problems associated with the further use of copper connectors can be solved. One of the main problems with copper wire is that it cannot be scaled as aggressively as other vital parts. modern processor. Beyond a certain point, it is physically impossible to shrink copper wire any further without compromising its performance and/or shelf life. In theory, optical connections can transfer data much faster while consuming less power. In addition, many companies believe that silicon photonics is necessary to create supercomputers with a computing power of about one exaflops (exascale computing).
Unfortunately, silicon is a poor medium for optical devices, since the scale of production is so varied (optical waveguides and other components are much larger than silicon CMOS) that there are no engineering solutions that can effectively and inexpensively integrate optical elements into existing CMOS using silicon rather than expensive alternative materials such as gallium arsenide. Now the company has been able to place silicon photonics chips directly on the processor module.
The graph from Intel's presentation on silicon photonics also illustrates the power consumption that manufacturers are trying to achieve. Long-term plans for silicon photonics offer the same bandwidth and energy per bit of information that is not available to copper connections.
After decades of work, silicon photonics may seem like just another crazy idea that looks good on paper, but is completely inapplicable in practice, but progress does not stand still, and although cutting-edge companies such as IBM, Intel or HP may not release the technology on commercial level in the near future, it will certainly find applications in scientific laboratories, supercomputers and data centers.
So spring has come ... And with it the time has come for the next Intel Developer Forum (IDF), held twice a year in sunny California and regularly visiting other cities of the world (more recently - in Russia). Moreover, spring this case came not just for a red word - in San Francisco, where the IDF once again takes place from March 1 to March 3 at the huge Moscone West convention center,
it’s really warm now, trees and bushes are blooming, pouring spring aromas, and locals walk the streets in shirts or light jackets if it’s not raining. Against this cheerful background, having arrived from snowy Moscow, it would not be so easy to sit all day in conference halls and press rooms, jostling among several thousand visitors and IDF organizers at show cases and on the sidelines. If not for that sometimes unique and exciting information that falls on you in huge portions, leaving not a moment of peace. Even I, a regular visitor to the central Intel Forums (as well as many other exhibitions and conferences of similar subjects), who, it would seem, is fed up with such events and perceives them almost as another Hollywood blockbuster, solidly molded according to long-known clichés, I often have to be surprised at that flow novelties prepared for IDF participants by its organizers. Surprise and even admire places ...
Our regular readers, probably, no longer need to explain what the Intel Developer Forum is and "what it is eaten with." This event, regularly held for many years by Intel Corporation and its closest friends in the IT shop, has its own individual characteristics that distinguish it from various computer exhibitions (like CeBIT, Computex, Comdex or CES, where hundreds and thousands of IT product manufacturers brag about their achievements in order to sell them more profitably), and from major world scientific and technical conferences (like the Material Research Society Meeting, IEEE and others, where hundreds of the world's leading institutions and research laboratories report on the latest scientific discoveries, inventions and technologies, the implementation of which is still to be dealt with for many more years). In my opinion, the IDF is closer to the latter than to the former. Since Intel, which spends more than $ 4 billion annually on Research & Development, is trying to demonstrate at IDF not so much current and ready-to-market products (microprocessors, platforms, etc.),
how much to tell the industry the vector in which it will develop over the next few years. To make public those current and future technologies that the corporation is implementing together with its partners and other IT developers, to attract new researchers and engineers (that is, “developers”, as the name of the Forum) to its side, and possibly discuss the feasibility of certain steps within the entire IT community. And although, of course, the “exhibition and sale” canvas is also present at IDF to some extent, the most valuable and interesting, in my opinion, is precisely its research and technological part.
So the "zero" day of the current IDF, which took place on February 28 for the leading press and analysts from around the world, presented several surprises, which I will try to tell about in this report, which precedes the story about the Forum itself.
Silicon Nanotechnology: Looking 20 Years Ahead
The first report of the zero day discussed how silicon technology for the production of computing devices can and will develop in the coming decades. Briefly and primitively, this could be called “the justification of Moore's law for 20 years ahead”, if such a banal at first glance message was not backed up by the breathtaking details of scientific research in the field of nanotechnology and their implementation in practice in industrial-scale technologies. The report was presented by Paulo Gargini (Paulo Gargini, pictured), director of Intel Technology Strategy and Intel Nanotechnology Research.
More than an hour of presentation was held at a very fast pace, not allowing for a second to come to his senses and calmly reflect on this or that slide. Her detailed retelling, apparently, would be useful to some of our thoughtful readers. But it would take up an unreasonable amount of space (that's about a hundred "serious" slides, each of which still needs to add a lot of comments). Therefore, I will note only some of the most interesting, in my opinion, points, especially since some of the details present in it, I and my colleagues have already described in my articles based on the results of previous IDFs and recent "technological breakthroughs" of Intel. I will elaborate on this material in more detail, perhaps another time.
For the past 40 years, the number of elements on silicon chips has steadily continued to double every two years, and the cost of a single transistor on a chip has declined at the same rate.
About 10 years ago, scientists predicted big problems in the transition to 100-nm devices, but, fortunately, this did not happen, and now industry leaders have well-studied prospects for the development of traditional silicon technology with planar CMOS transistors for another 10 years ahead (see. slide).
The need for fundamentally new electronic devices will arise only by the year 2013, when the possibilities of miniaturization of current devices will actually be exhausted.
Among the new silicon devices considered are multigate (for example, tri-gate) nanotransistors, devices based on silicon nanotubes completely surrounded by a gate, as well as devices with quasi-ballistic transport.
In a more distant perspective, carbon nanotubes with a diameter of a few nanometers are also considered, which, depending on the structure, can act as a metal or a semiconductor. Of interest for nanoelectronics are devices based on InSb heterostructures (with uniquely high mobility), see slide.
And what will happen after 2020, when CMOS technology exhausts the possibilities of miniaturization, reaching the atomic limit?
Then, perhaps, spintronics will be used - operating with the magnetic moments of elementary particles:
Some people also talk about quantum computers. In the meantime, CMOS technology is alive and Moore's law will be valid for at least another 15-20 years.
Silicon photonics: a new breakthrough
Another interesting event of the zero day of this IDF was the report on, created on a silicon chip at Intel. Strictly speaking, the news about this went around the world a few days before IDF (on February 17, a corresponding article was published in Nature and a press release from the corporation), but here the main developers of the new device publicly shared many hitherto unknown details and demonstrated to the audience numerous crystals with such lasers. For example, in this photo (author's photo), the crystal contains 8 such lasers at once.
Without going into details, we note that in order to create such a silicon laser, Intel scientists had to solve an important problem - the so-called "two-photon absorption", which previously prevented the creation of a continuous silicon laser.
The use of silicon as a material for creating a laser and for multiple amplification of infrared radiation (thanks to the giant, about 20,000 times the Raman effect),
Previously, it was problematic, since the Raman amplification saturated under high-power pumping, and the power obtained during saturation was not enough to create a cw laser.
The fact is that the energy of one infrared photon (quantum of light) is not enough to knock out (liberate) an electron from it when it collides with an atom of the silicon crystal lattice. However, if two photons collide with an atom at once (which often occurs when the laser is intensely pumped by external radiation), then ionization of the atom becomes possible, and free electrons in silicon begin to absorb photons themselves, thereby preventing further Raman amplification. The problem was solved by creating a so-called p-i-n structure along the optical channel (silicon regions with hole and electron conductivity, respectively, on the sides of an undoped optical channel in silicon, see figure).
By applying an electrical bias between the p and n regions of silicon, "two-photon" free electrons can be effectively removed from the region of the optical channel, thereby significantly increasing the Raman gain in silicon and creating a cw laser.
Based on this solution, it is possible to create two important optical devices directly on a single silicon crystal - an amplifier and a signal modulator.
And also with the help of cascades of mirrors (located directly on silicon) to make multiwave optical communication channels and compact lasers for various applications.
In the hands of Mario Paniccia, director of the Intel Photonic Technology Lab, a crystal of a new cw silicon laser (right) and a traditional expensive Raman optical amplifier (left):
This achievement of Intel employees opens up new horizons for the development of silicon photonics and its further implementation in traditional microelectronics.