Session C12 6166 Disclaimer —



Download 55.8 Kb.
Date20.10.2016
Size55.8 Kb.
#6879

Session C12

6166


Disclaimer — This paper partially fulfills a writing requirement for first year (freshman) engineering students at the University of Pittsburgh Swanson School of Engineering. This paper is a student, not a professional, paper. This paper is based on publicly available information and may not be provide complete analyses of all relevant data. If this paper is used for any purpose other than these authors’ partial fulfillment of a writing requirement for first year (freshman) engineering students at the University of Pittsburgh Swanson School of Engineering, the user does so at his or her own risk.
INTEGRATED CARBON NANOTUBE COMPUTER CHIP TECHNOLOGY ALLOWS FOR NEW SKYSCRAPER CHIP DESIGN
Tyler Murin, tdm41@pitt.edu, Bursic 2:00, Kevin McGoogan, kcm54@pitt.edu, Bursic 2:00



Abstract—Due to society’s growing dependence on computing systems, innovative advancements to improve efficiency are necessary in order to keep stride with the increased implementation of computers. One possible breakthrough is apparent in the overall design of a computer chip and consists of using carbon nanotubes along with the existing silicon-based microprocessor to create a never-before-seen “skyscraper” designed chip.

The process of embedding a silicon medium with carbon nanotubes is the first step in constructing the chip and has two main mechanisms, chemical vapor deposition and soft lithography. Chemical vapor deposition deposits the carbon nanotubes on a silicon wafer, and soft lithography is then used to pattern and replicate the existing nanotubes, which serve as transistors.

Since carbon nanotube transistors generate much less heat in an electric circuit, these new processors can be stacked on top of each other like floors of an apartment building. The layered chip and its numerous data pathways would allow information to be transferred vertically through metaphorical elevators, and thus outperform conventional approaches by a wide margin. The increased computing efficiency will in turn lead to long term sustainability in the computer industry.

Upgraded computer performance would enhance supercomputers with advanced programming applications, specifically molecular modeling and simulation. Machines with elite computing power can now be created with less cost and bulk, and data-intensive programs used in research and development will run more effectively, which could lead to future scientific breakthroughs.
Key Words—Carbon Nanotube Transistors, Clock Speed, Microprocessor, Moore’s Law, Nanotechnology, Skyscraper Computer Chip
COMPUTING THE FUTURE

The past quarter century can be characterized by rapid technological expansion, specifically in the computer industry, which seemingly controls not only the present, but the foreseeable future as well. Computers are continuously being upgraded and improved to support the increased technological demands of society, and as a result, the processes used in their manufacturing must be constantly updated as well. However, breakthroughs in the form of new innovative devices available on the consumer market are occurring at a much faster rate than advancements to the core components that allow computers to function. Current methods of fabricating these components, namely microprocessors, will suffice for the next decade, but physical limitations will soon be reached, which could prove devastating to the industry and halt computing innovation all together. In order to address the problem at hand, a new never-before-seen microprocessor has been prototyped and has the potential to revolutionize computer speed and capabilities. Stanford researchers have named the solution “The Skyscraper Chip”, as it is comprised of vertically stacked components rather than the traditional flat array seen in current devices. The new orientation in conjunction with carbon nanotubes, the key material needed for manufacturing, will provide limitless opportunities for future computing applications.



THE MICROPROCESSOR:

A BRIEF HISTORY
Setting the Stage
Before being able to comprehend the skyscraper designed chip, one must first understand the purpose of a microprocessor and the developments that lead to this innovation. The evolution of the computer from mechanical systems that ran on binary algorithms to the powerful modern day digital machines would have never been made possible without the creation of the microprocessor. Although small in size, the microprocessor’s effect on the computer industry was revolutionary, as it consolidated various computer components into one compact place. In the decades leading up to Intel’s release of the device, other breakthroughs in the industry allowed for the ingenious design to be thought of as feasible. The first and most important of these discoveries was the transistor, which replaced the bulky and heat generating vacuum tubes used in circuits [1]. As a direct result of the small size and increased energy efficiency offered by these devices, the notion of an integrated circuit was brought to life. The integrated circuit used the same semiconducting material for all components of an electric circuit, which in turn greatly reduced the size of circuits, as individual components no longer need to be manufactured manually [1]. In addition to the transistor and ensuing integrated circuit, the last of the major breakthroughs came in the form of the programming language FORTRAN. The language was the first ever to make use of a computer compiler, enabling programmers to code at much faster speeds and unlocking new methods of problem solving and data analysis [2]. All of these contributions to the computer industry set the stage for Intel, which successfully designed and manufactured the first single chip commercial microprocessor, or modern day “computer chip”.
Intel 4004
After initially being tasked with designing a working printing calculator, Intel engineers devised an innovative solution to the problem at hand in 1971, namely Intel 4004. The design consisted of four small chips, all oriented on one parent chip which allowed the individual components to function together. Intel 4004 was the first computer chip to implement integrated circuit technology, and used circuit lines slimmer than a human hair [3]. The first and most notable of the four chips was the central processing unit (CPU), which was responsible for interpreting instructions and performing arithmetic calculations. In addition to the CPU, random access memory (RAM) and read only memory (ROM) chips were introduced and were responsible for processing memory and creating custom programs, respectively [3]. Lastly, the shift register chip was to allow for different input/output devices to be connected to the module, allowing user input and computer output [3]. Not only was this model instrumental in changing the computer interface, but it also served as the foundation for microprocessor architecture. Microprocessors used over four decades later in recent years still have the same basic schematic/layout, consisting of memory modules and a central processing unit all on one chip.
Evolution of the Processor
To better understand modern processor technology, it is important to discuss the numerous advancements made to microprocessors since their mainstream introduction. One glaring observation would be the 2,300 transistors on the original processor versus the 820,000,000 transistors of current models [3]. Nanotechnology breakthroughs have allowed for transistor size to be greatly reduced, allowing computers to run at higher speeds with more efficiency. Along with transistor numbers, multi-core chips, which are often associated with 64-bit operating systems, have been designed and contain multiple processor cores on one single chip. The operating system perceives each core as an individual processor, which allows data manipulation onto multiple threads, optimizing the multitasking capabilities of computers and allowing for more advanced calculations and simulation models [4]. To express the advancements quantitatively, clock speed, a factor used to assess computer speed, has increased from 108 KHz to over 3 GHz, a factor well over 100,000 [3]. The evolution of computers started with Intel 4004, but has since been guided and powered by breakthroughs in nanotechnology and microprocessor architecture. Every improvement to the physical, chemical, or electrical aspect of the design in the past century has contributed to the sustained increase in computer upgrades and advanced computing power evident in modern society.

SMALL ISSUE, BIG PROBLEM

Although computers and their components have evolved a great deal since their initial introduction to the market, problems and inefficiency still exist. As briefly mentioned earlier, the number of transistors has rapidly increased with better microprocessor architecture and design. This correlation, which is often referred to as Moore’s Law when discussed in industry, is used as a prediction model for computer chip manufacturing. The law is often represented as a linear plot and can be simplified to mean the number of transistors per square inch on an integrated circuit will double each year [5].



https://www.indybay.org/uploads/2006/05/18/moore__sl_small.jpglsjprm.jpg
FIGURE 1 [6]

Moore’s Law graph of number of transistors over time

As seen in Figure 1, the bottom dotted line, showing the doubling of the number of transistors on chips every two years, is an accurate line of best fit for the data points. However, the linear trend in recent years has flattened, as seen by the highlighted curved line on the top right. The reason for this deviation is the physical limitations of transistors in the nanotechnology industry is quickly being approached. Current techniques can produce chips with a precision of 14 nanometers, which has recently been made possible by advancements in nano-manufacturing [3]. Even if new methods or techniques allow for smaller chip production, current transistors cannot function at smaller sizes. Electric current will not flow because the transistors dissipate heat at too slow of a rate, resulting in overheating circuits and dysfunctional electric signals.



In addition to the physical size of transistors, current circuitry has other flaws that up to this point seemed unavoidable. Data signals are currently sent on paths that are longer than ideal due to the flat architecture of the chip [7]. This problem is magnified when multiple chips are used in conjunction, as data pathways become stretched over multiple chips leading to slower data transfer and processing. Furthermore, bottlenecking, in which too many signals try to travel the same path, is another problem evident in modern circuits that restricts data flow and is amplified when more than one chip is involved [7]. Although computers of this age appear powerful and sometimes flawless, many problems still exist and prevent them from reaching their ultimate potential.
THE SKYSCRAPER
The Design
As a solution to all of previously mentioned problems with circuits and physical limits, a newly designed chip that can be stacked vertically rather than positioned horizontally has been created. The skyscraper chip is comprised of several standard silicon computer chips coated with carbon nanotubes, which themselves act as transistors. Carbon nanotubes are essential to the design as they generate little heat, conduct electricity extremely well, and can hold microprocessors together like glue when placed on top of one another [7]. Implementing nanotubes allows for central processing units and memory modules to be placed on top of each other, unlocking shorter and many more data pathways for signals to travel. The chip addresses all the main problems with currently used microprocessors, as it will reduce bottlenecking and data path length with more pathways and less distance between chips, while at the same time using carbon nanotubes as an alternative to existing transistors.

Previous attempts at constructing such a chip have proved unsuccessful because of heat generation involved with chip fabrication. In order to vertically orient silicon chips, the chips must be manufactured in one whole unit rather than individual stacked layers. To create and etch one silicon chip on the stack, the silicon must be heated to roughly 1800 degrees Fahrenheit, which subjects existing microprocessors to intense heat [7]. Even though some successful processors can be produced on the column using this method, fabrication of nearby processors will torch the recently created ones, rendering them incapable of functioning in a computer. In addition to fabrication aspect, silicon chips at the top of a processor-stacked column generate too much heat for the ones near the bottom, which causes the lower microprocessors to overheat [7]. Nanotubes solve both problems because they can function as transistors at much smaller sizes and produce less heat in a circuit. Also, carbon nanotube growth can occur at much lower temperatures, which prevents processors from being burned during the construction phase. Manufacturers will have the ability to fit more transistors onto existing computer chips, while simultaneously eliminating inefficiency, paving the way for ongoing computing innovation.

Innovation, as previously mentioned, is essential in the computer world, as the industry faces a looming issue regarding technological progress. The extent of the effect the skyscraper design will have on computer industry must be further analyzed and discussed. The breakthrough can be thought of as a security blanket, protecting the computer industry for at least several upcoming decades. Although an exponential increase in computer technology would be ideal, every innovation propels technology forward until reaching a barrier, whether it be physical or scientific. If no other groundbreaking innovations are made, technology will eventually reach its limits, as seen in the current situation. In reality, the skyscraper orientation will greatly assist in advancing capabilities of computers, but specific aspects in the manufacturing and design phases of the chip must be optimized and improved to truly sustain long term growth and advancement for the industry.
Apartment Building
In order to better understand the functionality of the chip, it may be helpful to think of the microprocessor in terms of an apartment building. Current chips connect components on a flat board, comparable to single storied structures in a suburban neighborhood [8]. The chip is based on the idea that being oriented horizontally requires much more space and order than being stacked vertically. The skyscraper chip can be thought as a consolidated suburban neighborhood stacked into one vertical building. In a suburb, traffic jams seen on crowded public roads metaphorically represent bottlenecking in electric circuits on standard flat chips, while the commute from home to work symbolizes non-ideal data pathways. Looking from a similar perspective, elevator rides in an apartment serve as the commute and directly take people to their desired destination [8]. Numerous elevators in an apartment can be thought of as the numerous pathways in the chip, as it is simple and easy to get from one floor to the next. In addition, cooling systems on each floor of the stacked chips mimic air conditioning units within an apartment building. All of the parallelism evident in the apartment metaphor illustrates the ease at which data can be transferred from floor to floor in stacked orientation versus street to street in the horizontal position.

              



HOLDING THE COMPUTER INDUSTRY TOGETHER: CARBON NANOTUBES

Physical Properties and Advantages
To truly understand the skyscraper chip’s uniqueness and potential, the specifics of carbon nanotubes must be examined in depth. The nanotubes themselves are comprised of a special form of carbon, namely graphite, whose molecular structure consists of sheets of hexagonal carbon rings stacked upon one another. In its natural state, graphite is very weak and brittle, as seen in the tips of pencils where graphite atoms easily move past one another and break apart [9]. However, nanotubes are the hexagonal graphite atoms rolled up into dense cylindrical tubes, a transformation that results in drastic changes in physical properties. To put this into perspective, carbon nanotubes are five times stronger than steel and can resist deformation up to 63 times better [9]. Another essential physical aspect of the nanotubes is their size, as their average diameter is roughly one nanometer, which is fourteen times smaller than width of current transistors [9]. Unlike silicon transistors, graphite is composed of a weak non-lattice structure, which allows for a smooth flow of electrons and a high rate of heat dissipation [9]. Electricity is able to flow due to the rounded tips of the tubes whose high electron density has the ability to produce a powerful electric field and eject electrons through the circuit, thus allowing for electric current [9]. Chemically, nanotubes are inert, meaning they do not react well with very many other substances, which makes them a valuable resource that is compatible with other materials and useable in a wide range of applications. The myriad of favorable properties nanotubes possess make them an appealing material on which to base the entire design of the skyscraper chip. In addition, the carbon nanotubes themselves will promote sustainability, as they will provide unmatched opportunities and open new doors for innovation. New three dimensional circuit arrangements will replace traditional linear circuitry, allowing new methods of data transfer to be investigated. The current perception of linear circuits will be revolutionized, and the future capabilities are not yet known, potentially allowing for sustained large scale computing advancements.
The Manufacturing Process
Carbon nanotubes, the primary reason for the skyscraper chip’s promising future, need to be carefully and strategically placed on the surface of silicon chips to ensure proper functionality within the electric circuits. In order to grow these microscopic fiber-like tubes on a silicon medium with a high degree of precision, engineers must use two main chemical processes: soft lithography and thermal chemical vapor deposition.
Soft Lithography
The first of the two major phases in nanotube production is soft lithography, which refers to a family of techniques used for the replica molding of various nanomaterials. One member of the family, photolithography, is essential for carbon nanotube growth, as it facilitates the uniform distribution of carbon nanotubes on a silicon wafer in an ordered rectangular array [10]. If carbon nanotube transistors are grown without proper patterning, placement on the silicon is sporadic and tubes often agglomerate, or stick together in large clumps.

Before beginning the patterning process, the silicon substrate, or base material, must be cleaned and prepped. Various chemical solutions, depending on silicon sample, will be applied to the wafer to rid the compound of unwanted organic, ionic, and metallic impurities [11]. The ensuing photolithography process can be broken down into four main phases, all of which are outlined in Figure 2 below.

FIGURE 2 [11]

Four major steps of photolithography process
After the sample is cleaned, a layer of silicon dioxide (SiO2) will be applied to the substrate, serving as a barrier for the next step in the process. A layer of photoresist, a material that is extremely sensitive to light, must then be applied the silicon wafer using a common technique referred to as spin coating [11]. The photoresist is placed onto the center of the silicon, which is spun at high speeds by a spin coater, resulting in the uniform dispersion of the photoresist on the wafer [11]. The force form spinning at high speeds, often referred to as the centrifugal force, evenly pushes the photoresist out onto to the edges of the silicon sample. Continuing with the process, the next step is to soft bake the substrate, or apply a constant supply of heat ranging from 90-100 degrees Celsius for 20 minutes to remove the coating solvent [12]. Now the photoresist must be exposed to ultraviolet light using a glass plate with metal patterned droplets, which causes the photoresist to dissolve in the silicon base [13]. After exposure, the silicon is subjected to hard baking, in which the silicon is heated once again but at higher temperatures to strengthen adhesion between the photoresist and wafer. The final step in the process is to remove the photoresist, which currently lies upon the silicon dioxide layer. When all steps are complete, the end product is a patterned silicon wafer that dictates where carbon nanotubes can and cannot be grown.
Chemical Vapor Deposition
The second and equally important process is thermal chemical vapor deposition, which is used to grow the carbon nanotubes on the silicon substrate. Before starting nanotube growth and serving as a continuation of the soft lithography process, a non-catalytic metal must be used to cover portions of the silicon wafer where no nanotubes are to be grown [10]. In order to attach the metal to the surface of the silicon, a technique known as electron beam evaporation must be employed. The metal, most commonly copper, is vaporized by a heated tungsten filament which ejects electrons guided by magnets onto a small target area all contained in a vacuum [13]. The high energy density from the electrons vaporizes the metal which condenses of the surface of the silicon, creating an even coat on the surface.

After protecting the spots on the silicon chip that are not to be covered with nanotubes, a film of a catalytic metal must be coated on the substrate. The purpose of the catalytic metal to enhance and speed up the growth of carbon nanotubes on the desired surface [10]. A specific form of deposition, sputter deposition, is used to fuse the metal piece to the silicon and create a thin film. The apparatus in which the process takes place is known as a sputter coater, and it uses several magnetic fields to maintain electrons in a toroidal or donut shaped path around the substrate [14]. When these electrons lose enough energy and fall to the surface, they fuse into the substrate and form a circular ring on the material [14]. Once the catalyst is in place, ammonia gas is then released into the container to “etch” the new layer of film and create nanoparticles [10]. The ammonia gas acts as a supervisor to the nanoparticles sites and facilitates their creation. Argon gas, an inert substance, is also present during this reaction to ensure any unwanted byproducts are removed from the solution. After nanoparticle sites are established by the ammonia gas, a hydrocarbon gas, such as acetylene or camphor, is released into the chamber and reacts with the surface of the substrate. After 60-90 minutes, depending on desired nanotube length and thickness, the chamber is cooled to room temperature to allow for the freshly grown nanotubes to harden.

As evident in Figure 3, the freshly grown nanotubes are assembled and aligned in a very specific pattern. If looking from the top view in the two left images, one can see the nanotubes are arranged uniformly in a rectangular array with even spacing. The earlier soft lithography phase allows for this to occur without clumping and disorder. In the right images, one can observe another property of nanotubes, their flexibility. The flexible nature of carbon nanotubes allows them to act as effective adhesives, as they are naturally attracted to each other.

http://www.tandfonline.com/na101/home/literatum/publisher/tandf/journals/content/uhte20/2016/uhte20.v037.i09/01457632.2015.1080570/20151229/images/medium/uhte_a_1080570_f0002_b.gif
FIGURE 3 [10]

High resolution images of carbon nanotubes after growth

Overall, nanotubes are a promising material with unmatched potential and a difficult fabrication process. All steps must be followed carefully and accurately to ensure proper nanotube growth and function. The main benefit of this current technique is that manufacturers will potentially be able to mass produce nanotubes on a large scale, thus making nanotubes a more plausible and sustainable solution for the skyscraper chip. In addition, the process can be further upgraded if a technique is found that can reduce the amount of conducting tubes after deposition growth. The process yields a majority of quality, functioning nanotubes with optimal capabilities, but some resulting nanotubes are conducting, in which current cannot travel as fast. In summary, current technology appears promising, but it is imperative all processes involved in chip production be continuously evaluated and improved to create the best possible skyscraper chip with maximum computing potential.
INTEGRATING THE SKYSCRAPER INTO MAINSTREAM TECHNOLOGY
Carbon nanotube based computers can yield extraordinary results due to their flexibility in design. Future computers with carbon-nanotube-transistor based microprocessors may even surpass the multitasking capabilities of modern super computers, responsible for important tasks in the research and development fields, such as data mining. Data mining, in particular, is the process of analyzing large databases through complex algorithms in order to find useful patterns. Common techniques include cluster detection, anomaly detection, and regression, all of which prove important to society [15]. Cluster detection searches for similarities within the data, while anomaly detection searches for abnormalities [15]. Both of the two processes categorize data in ways that allow for trends and patterns within any field to be visually represented through graphs and charts and analyzed. The third type of data mining, which will be most affected by the skyscraper chip, is regression, which allows computers to generate advanced prediction models for nearly any scenario. One field specifically that prediction modeling has a profound effect on is chemistry.

Chemical modeling, which often takes advantage of regression techniques, is used to better understand the mechanisms of reactions and orientations of molecules. Chemical modeling can be summarized as a computational study of various hypothetical reactions to determine what reactions happen and how often they occur. By running millions of prediction scenarios on these computers, the percentage of successful reactions and optimal conditions for that specific reaction can be calculated [16]. In addition, the best catalysts for a specific reaction can be determined and the resulting increased rate of reaction can be predicted [16]. Manufacturers in all industries will have access to this software and can use advanced multitasking algorithms to make better predictions on future sales to optimize yield and profit. Furthermore, not only will new more complicated programs be created, but existing modeling and prediction software can be instantly upgraded in terms of computer speed. Existing useful programs will be more efficient and the processor will enhance these programs, promoting long term sustainability due to its compatibility with current technology.

In addition to modeling software, servers are another important aspect of today’s culture. They are responsible for powering and supporting one of the world’s most important tools, the internet. A server is a computer designated to communicate with other computers wirelessly in order to relay data over a network connection. This often requires multiple servers to accomplish, even on the scale of a whole warehouse, called a server farm. These farms take up an enormous amount of space and require a vast amount of power to operate. Consequently, server farms have detrimental effects from both a financial and environmental standpoint. By utilizing the full capabilities of carbon nanotubes, not only will wireless data transfer improve drastically, but companies will be more profitable and the environment will be cleaner. Internet speed will improve, and downloading along with streaming will be upgraded as well. These chips can function with incredible results at a fraction of the size and power consumption of traditional chips. As a result, server farms will dwindle down to server gardens considering much less energy and maintenance will be needed for upkeep.

Another application of this technology is artificial intelligence. The term, often associated with a sci-fi theme, is a form of data mining that may become a reality in the very near future. A computer will absorb an input from its surroundings and analyze a large data base of actions to take; it will then select best response possible based off computation algorithms and predictions. Its purpose will be to aid the user in any way it can in the most humanly way possible. Unfortunately, an ideal artificial intelligence program must use portable hardware that is convenient for the user. The fastest cell phones on the market today can accomplish speeds of a subpar netbook, nowhere near data mining capabilities. Carbon nanotubes could break that barrier by enabling the creation of compact and efficient hardware capable of fishing through gigantic quantities of data while executing complex algorithms based on minimal user input.

The possibilities resulting from the implementation of the skyscraper chip are in no ways limited. Computers are an integral part of society, and the creation of a revolutionary processor will give computer more capabilities.
WHAT COULD GO WRONG?
Artificial Intelligence
A common ethical concern with rapid development of technology is the fear that computers will soon surpass the critical thinking skills of humans. As artificial intelligence continues to evolve, this fear may become a reality. However, in order for this technology to become integrated into society, computing power must increase. With the possible looming drought of computing speed, artificial intelligence initiatives will slow down significantly. The skyscraper chip is one step towards rapid computer evolution, possibly resulting in computers writing their own code.

Although current computer programs are extremely helpful by enhancing almost every aspect of society, computers will never reach its full potential if artificial intelligence is not present. After all, the programs are written by humans, and therefore they will only be as powerful as the author who wrote the code. If a computer were given the ability to write its own programs, the possibilities would be limitless. Computers would continuously search for algorithms to solve problems plaguing the world and code programs as solutions. This may be decades away from happening, but as computer power increases, so do the capabilities.

By giving computers its own state of mind without any way of supplying distinctive human qualities such as emotions and common sense, there is no guarantee a computer will make the right decision for the good of humans [17]. This question leads us to the fact that artificial intelligence systems would need to recognize the moral status of anything it encounters. For example, an inanimate object with no moral status can be subjected to all types of treatment, but a human must be treated not only as a means but also as an end. Therefore an AI system must be able to take into account the basics of moral restraints against humans [17]. Each case of these moral issues is extremely unique making it almost impossible to recreate artificially. Furthermore, if carbon nanotubes live up to their expectations, artificial intelligence will inevitably become smarter while morals remain the same. Although artificial intelligence seems it will play a prominent role in the future of society, it still must take significant steps even to become mainstream. The chip will provide a much needed boost to computing power, but in order to sustain the necessary technology other advancement must be made. Multithreaded applications involving multiple processors must be created allowing for computer generated response to actual problems. The reoccurring theme of continuous innovation along with the major breakthrough in the form of the chip is essential.
Cyber Security
Given the qualities of carbon nanotubes and the effect they would have on the computer industry, the software industry would also follow in that same path. Although great innovation would be put forth to better modeling fields such as chemistry and engineering, the same drive for innovation would backfire into the hands of hackers and identify thieves. With the computing power demonstrated by carbon nanotubes, algorithms could be written to improve the hijacking of an innocent user’s identity. As a result, in the early stages of the transition from silicon to carbon nanotubes, people with traditional silicon computers would be at a tremendous disadvantage to hackers using carbon nanotube based processors. Therefore many standard users not interested in upgrading will be extremely vulnerable to identify theft and credit card fraud as typical antivirus software will become useless due to its lack in computing power. This may seem like a major issue initially, but the creators of antivirus software also have the ability to improve their product. As hackers write new codes to cause havoc, antivirus creators will enhance their product to combat the surge in online crime. Conclusively, during the transition of computing eras, the problem of vulnerability to hacking must be addressed and accounted for in order to safely and successfully implement this new form of computing.
GETTING TO THE TOP FLOOR
Traditional flat silicon computer chips are reaching the brink of their physical limitations, as the minimum size of a standard transistor is being approached. As a result, the maximum attainable clock speed by this design is seemingly at its peak. Consequently, an alternative design must be found in order to keep pace with society’s high demand for technological improvement. Despite some ethical issues involving artificial intelligence and online security, carbon nanotube computing has much potential to fuel this demand. If properly implemented, this technology has potential to be one of the most significant breakthroughs of the past century. Therefore, it is imperative that heavy emphasis is put on investing time and resources into this new method of computing. The physical attributes of carbon nanotubes are far superior to those of silicon, most notably in the way they allow for circuits to be designed. The unique properties allow previously impossible arrangements of circuitry to become a reality, as seen in the skyscraper chip design. The design enables chips to be stacked as opposed to laid out flat, causing the efficiency in communication between computer components to increase drastically. This design in conjunction with the physical attributes of carbon nanotubes permits this technology to be a viable successor to current silicon chips. Once implemented into society, applications of this device are endless and the average person may have access to supercomputing processing speeds. Data mining capabilities will benefit primarily, as large sums of data can be analyzed at record speeds, enabling the creation of advanced prediction models and recognition of societal patterns. Unfortunately, increased computer power will raise questions regarding online security, but any advancement is accompanied with inevitable negative drawbacks. If the correct precautions and safety measures are taken, the software can be used to combat more advanced hacking, nullifying its negative effect. Therefore, in order to ensure the successful future of carbon nanotube computing, we must pursue further research in all aspects of its development and advancement.

REFERENCES
[1] “The History of the Integrated Circuit”. Nobel. (Online Article). http://www.nobelprize.org/educational/physics/integrated_circuit/history/

[2] (1999). “The FORTRAN Programming Language”. University of Michigan. (Online Article). http://groups.engin.umd.umich.edu/CIS/course.des/cis400/fortran/fortran.html

[3] “The Story of the Intel 4004”. Intel. (Online Article). http://www.intel.com/content/www/us/en/history/museum-story-of-intel-4004.html

[4] J. Barbic. (2007). “Multi-core Architectures”. Carnegie Mellon University. (Online Article). http://www.cs.cmu.edu/~fp/courses/15213-s07/lectures/27-multicore.pdf

[5] (2014). “Moore’s Law and Computer Processing Power”. University of California, Berkeley. (Online Blog). https://datascience.berkeley.edu/moores-law-processing-power/

[6](2006). “Moore’s Law Ending” IndyBay. (Online Image). https://www.indybay.org/uploads/2006/05/18/moore__sl_small.jpglsjprm.jpg

[7] J. Bennett. “‘Skyscraper’ Chips Could Make Computers 1,000 Times Faster”. Popular Mechanics. (Online Article). http://www.popularmechanics.com/technology/a18493/stanford-3d-computer-chip-improves-performance/.

[8] D. Poeter. (2016). “Skyscraper Chips Promise Powerful Computing Boost”. PC Magazine. (Online Article). ISSN: 2373-2830

[9] T. Adams. (2000). “Physical Properties of Carbon Nanotubes” Michigan State University. (Online Article). http://www.cs.cmu.edu/~fp/courses/15213-s07/lectures/27-multicore.pdf

[10] P. Asinari , M.I. Shahzad, M. Giorcelli, L. Ventola, D. Perrone, N. Shahzad, E. Chiavazzo, M. Cocuzza, A. Tagliaferro. (2016). “Convective Heat Transfer Enhancement for Electronic Device Applications Using Patterned MWCNTs Structures”. Heat Transfer Engineering. (Online Article). DOI: 10.1080/01457632.2015.1080570

[11] “Photolithography”. Georgia Tech University. (Online Article). http://www2.ece.gatech.edu/research/labs%20/vc/theory/%20photolith.html

[12] R. Darling. “Photolithography”. University of Wake Forest. (Online Article). http://users.wfu.edu/ucerkb/Nan242/L15 Photolithography.pdf

[13] (2016). “Electron Beam Evaporation”. Midwest Tungsten Service. (Online Article). http://www.tungsten.com/tips/electron-beam-evaporation/

[14] “Sputter Deposition”. Oxford Vacuum Science. (Online Article). http://www.oxford-vacuum.com/background/thin_film/sputtering.htm

[15] (2012). “The Benefits of Data Mining”. Concepts. (Online Article). http://charc-concepts.org/the-benefits-of-data-mining/

[16] “IBM Research- Almaden Science Colloquium Series”. International Business Machines (IBM). (Online Article). http://researcher.watson.ibm.com/researcher/view_group.php?id=4284



[17] N. Bostrom , E. Yudkowski. (2011). “The Ethics of Artificial Intelligence”. (Online Article). http://www.nickbostrom.com/ethics/artificial-intelligence.pdf
ACKNOWLEDGEMENTS
We would like to thank our writing co-chair, Chase Barilar, for helping us organize our paper and ideas. We would also like to thank our writing instructor, Keely Bowers, for providing helpful feedback on prior assignments that we submitted.





University of Pittsburgh Swanson School of Engineering

2016/03/04


Download 55.8 Kb.

Share with your friends:




The database is protected by copyright ©ininet.org 2024
send message

    Main page