With respect the the Samsung: Autofocus: good in bright light, passable in low light, it really is better than the old canons (5Dii) in mid-low light. Nowhere near the newest AF SLRs. There is no good way to control focus in C-AF mode. Weight: Very nice! Even with the 16-50 f2-2.8. Durability: Appears to be at or above the 5Diii level. Not quite as overbuilt as an D810. It feels basically like a 7D. Usability: Complete garbage! Nice button placement but boneheaded firmware. In video mode, we let it slide because the video is so awesome. Support: 2 Firmware updates already, new ones on the way (sadly these ones are video focused). Weather Sealing: S lenses are gasketed, body appears robust. We won't know till someone tears one down though. Lenses: 24-70 eq available and quite good. 70-230 eq reportedly of similar quality sharp and fast. Neither are landscape lenses. 85mm f1.4 (135mm eq) check, very sharp very corrected a bit high on CA but its a system of lenses that are supposed to be software corrected. 60mm macro check. 30mm pancake check. UWA zoom check minus. Viewfinder Quality: A new standard as far as latency, good sharpness as compared to A7. Lightroom support out of the box: It includes a copy of lightroom! Sadly no Capture One support other than through an included DNG converter. At the moment it really is an astounding 4k video camera. Stills are good enough for events, there are reports of sports working fine if you can fill the frame. Its a PITA for slow fine arts stuff because of wide open focusing and a nerfed AF-S mode.
I'm pretty sure the interesting metric is the fracture toughness. Interestingly enough there is a NIST page on it: http://www.ceramics.nist.gov/s...
which ranges from 1.89 to 4.45 MPam1/2
and a nice paper on anealed borosilicate glass fracture toughness: http://dx.doi.org/10.1016/j.ph... which ranges from 1.5 to 1.7 MPam1/2 depending on loading. Of course Gorilla glass might have slightly higher values.
I think the OP is refering to plastic as in the adjective plastic 1632, "capable of shaping or molding," from L. plasticus, from Gk. plastikos "able to be molded, pertaining to molding," from plastos "molded," from plassein "to mold" (see plasma). Surgical sense of "remedying a deficiency of structure" is first recorded 1839. The noun
Lol, shouldn't it be an Otto-Bahn after the inventor/discoverator of the diesel cycle? Also I laugh a little as the Autobahn (car track?!) here in Austria seems to be a toll road. I'm not sure since I don't own the cars I drive but we have to have a special sticker that costs a couple hundred euro a year to use them.
NO! I think he should be rewarded by those who can appreciate the mastery and wish to buy a signed print. Do prints of mainstream photography that have made it onto postcards deminish in value? They don't. The artificial scarcity is interacting with creator and recieving an individual product.
Free loaders are fine, people will find a way to show how bad-ass they are that they can afford to advocate art, not only generic art but art that fits their perception of good.
maybe in a perfect world... but check out some of honeywell's patents on parasitic power for thermostats that have been reissued without significant changes multiple times. The patent office is broken which is part of the problem.
spoken like someone who has never tried to make a direct copy (even of software) that functions to the same extend as the original. Copying takes work and takes developement. Why do you think that science isn't dominated by the Chinese? They have the grad student populations to put a smart student to work trying to reproduce every interesting scientific paper (where they even give you the recipe most of the time), it doesn't really work out because by the time the student has "discovered" how to do the cutting edge stuff from 2-3 yrs ago; the entire field has moved on to something different.
Another fun example: there is a small korean motorcycle company that literally copies the suzuki SV650 as in the engine castings look nearly identical. They can therefore sell the bike for 20-30% less than the japanese do, however, the copy only makes 70% of the power of the original and has attrocious suspension and brakes.
Copying is not easy with tangible goods or technology
I have to agree! I really want to teach a digital logic course to high-school students. My professor was crazy (in a bad way) but it was the most influential course for me too.
It also works in most American universities if you avoid the engineering prerequisite courses, just dropping your head in and asking to sit in on a course (as long as you are a student) works suprisingly well. Otherwise there is the option to sign up for many more courses than you plan to take and drop the ones you don't want before the drop date. I did that rather often in grad-school with math courses. The university won't tell you to do it but it is a legitimate use of the rules they design. It is really the only advantage to the all you can eat pricing that you get/have to pay as a full time student.
So why is this a game changer? Well its not yet but it could be... So basically we have to make small chips, this is because the parasitics decrease with size so we get more efficient as we get smaller (up to a point), we also get cheaper for simply geometric reasons (wafer/exposure area pretty much fixed cost) so yay more goodies from a chip.
Except... Tooling up is expensive and HARD, dude what do you mean a.032m lithography mask costs $500,000+ and it probably won't work perfectly for at least a couple of revisions??! That sucks!
Why is this placement technique cool? One of the neat uses for big chips today are graphics chips. They are huge but relatively simple, they have a bunch of repeated processing units and a buttload of cache and some neat front end or back end. The top generation GPUs are usually made about as big as one can physically make a chip (we normally expose wafers in chunks, and precision and field of view are competitive design goals when thinking about the optics for these systems), because they problem is embarrassingly parallel we can use all that performance and it even scales reasonably well across multiple cards/nodes. So imagine we could make a functional dice that contains a very small number of processing units, some cache and some glue logic that makes it easy to connect to some magical buss, now we can make a GPU as big as we want. If we are a GPU company we like this since we now have something like linear scaling for processing costs to go to MUCH bigger chips, which means we can sell a new super computer every few years without having to compromise our consumer line.
Other cool things: Lab on a Chip, you can throw down MEMS and processing components on a small device, imagine a full digital assay built into the pricking needle. Harry Potter Newspapers, if you can do alignment and finishing after the fact you can print your e-ink display and driver in whatever size you need. All thats needed is good yield on the sub-displays. Disposable-lazy-electronics, we are getting here anyway with RFID, but how cool would it be to carry a roll of stickers that would act as line of sight GPS extenders but were solar or wifi powered. Add some sensors and you have a crowd-sourcable metro monitor that knows where every train is and what stops are crowded.
To summarize, this invention has to capability to possibly decrease the cost of sticking widgets together, thus we can have many little widgets instead of monolithic widgets.
except for they aren't refering dies as in what you use to smash semi-soft metal into new shapes but actually dice, as in what you get after you dice (or cut up) a wafer.
TLDR: you are correct with tool dies, he/she is correct with wafer dice.
define blowtorch, in colloquial american english it seems to be a Propane or Oxy-Propane/MAPP torch. If thats what the parent is refering to I've used all 3 within the last year. My first date with my gf was introducing her to glass-working using the lab's Oxy-H torch and she started with fused quartz. I would argue if acess is available and some sort of guide or manual is near many people will pick up any old tool if they think it will help with the job.
I've been in an out of the piracy game since irc times but there was one approach that really worked for me. The author of Lux (a java based Risk game) had a nice system for detering privacy: 1st: The game was free to play for 10-20 times and then it required registration (simple key code) 2nd: The author had set up a website so when you searched google: lux warez, serialz, serial, keygen, his website was the first site you got to where he asked crackers to respect his tiny cottage industry (I think it was 5-15$ for a lifetime key), and at the same time pointed out to users that by stealing his software they were poisoning his part of the ecosystem.
It seemed to work. I never found keys to the software (this was 6-7 years ago), and we didn't pirate that piece of software. I stopped looking for keys after I'd read his page and that was the important part.
On the other hand I have very little problem pirating professional software to play around with 3DStudio and Photoshop, however once I got into photography (and had spent much more than the cost of software on gear) I've had to change my approach. I pay for my Raw software (Capture One Pro) and I use gimp or open source tools instead of PS. Sometimes I want to dick around with CAE software and I have no problem pirating that since I'm interested in demo-ing it and not using it as a tool in my business. I think reminding users what they would be paying for (its your time not the tool) is the best approach.
except that E-beam lithography is in effect lithography, the following steps are harder and require lots of infrastructure. Here is a typical process for getting a single layer into a chip. Step 1: Clean the substrate of any organics. Step 2: Apply resist (usually using a spin on process) Step 3: Expose resist (E-beam -- Photolithography it doesn't matter). The hard part here is exposing in the correct places. Step 4: Develop resist, usually wet chemistry which will remove or leave only the areas exposed in the previous step. Step 5: Use the patterns made with the resist (Deposition, Etching, or Implantation) Step 6: Remove the exposed resist, usually a different wet chemistry.
Then remember that you are going to do this entire process numerous times ( A simple P-MOS needs 4+ cycles without considering metalization). It also HAS to be done in a clean room if you want ANY flexibility as you have to switch the substrate between different machines for each step.
If we decide to go at it like 3D printers where one machine does every step (implantation is still kinda difficult but it could be done with a FIB) then we need to be able to predict exactly what the beam will do which we can't do yet. We are working on learning how to do that but we are not there yet. If we had all the knowledge to be able to build chips like a 3D printer I'm still not so sure we would as the general case since the batch process mentioned above is very cheap (per device). Then again I would not have expected the 3D printer movement to have taken off so quickly, so it could happen.
The costs are already coming down quickly, there are desktop SEMs that cost less than an expensive SUV, the gas injection systems are nothing more than capilarry tubes and solenoids (neglecting that most of the deposition gasses are wicked toxic and may explode if they contact air) so I would argue that the tools are already very much on thier way to being cheaper. The problem is we really don't have any systematic approach to using them to do what I think you are suggesting.
Sorry, I went all internet tough guy back there... I should clarify what I meant. 1st: E-beam lithography as I know it; with an E-beam resist is pretty much the creme of the crop if you want ultra high resolution. It is also a very old technique IE they were looking at it to replace photo-lithography as far back as the '80s but there are difficulties with making a bright electron beam to do the lithography in a parallel manner. Therefore its been used serially with a beam rastering the resist to make the desired patterns. With this techniques you can make very small features.
2nd: I am un-aware (doesn't mean it doesn't exist, just that its outside of my research area) of any analogous ion beam processes; in that we are talking about using a polymer resist activated by an ion beam. There are however very interesting nano patterning methods that use implanted ions either in a sacrificial layer or in the substrate itself, followed by selective etching that could arguably be thought of as ion-beam lithography.
3rd: Focused Ion Beams (FIB) is a rather mature technique for circuit repair and editing because it acts as both and additive and a subtractive process. With the FIB we can make deep holes using gas assisted etching, and then deposit with gas deposition both conductors and insulators. The real advantage of this technique is that we can see what we are doing!! Imaging can be done either with the ion beam or a separate electron beam allowing us to see the structures we are working on with the same or better resolution than we can write or etch with. Normally however FIBs use Gallium (Ga) ions as they are a convenient ion source (the melting point is low and the vapor pressure is also low) these ions are rather heavy and cause damage to the substrate (this can be mitigated through careful selection of the beam energy and angle), Ga also acts as a dopant in silicon.
4th: There was a company that tried to deal with the serial nature of focused ion beam milling. This company developed a 1024 beam array where each beam could be individually steered or turned on or off using a selector plate made with standard Si manufacturing techniques. This device used Argon (Ar) ions to avoid doping. Sadly it seems this company has stopped developing this device. They might be entering with a similar setup for electron beams in the future. My understanding is that the ion beam device worked best for gas-assisted processes where the deposition or etching gas is activated by secondary electrons freed when the ion hits the target. Seeing as an electron beam also free secondary electrons I think they changed directions to an electron only technique but these are only rumors I've heard around work.
Both Ion beam and electron beam techniques are more difficult than they appear as the yield (either sputtering or secondary electron) is dependent upon the incidence angle between the beam and the surface. It therefore becomes much more difficult to predict the interactions once the surface is no longer planar.
My comment about the 30nm not being all that sexy was with respect to TFA, I saw this on FEI's facebook page a couple of weeks ago and thought the same thing. Yes its neat that they can make shapes at this size with good control (heaven knows we can't do it yet with electron deposition or fib deposition [we can make cute test cases but we are far from arbitrary shapes even though we can do overhangs already]), but for me the real limitation is that they seem quite limited with respect to the materials that they can make things out of. I'm sure this is a great thing and we will see some neat tricks in the future with people either using these printed structures at templates for some nano imprint lithography, or as high tech resist with some neat deposition into the voids. My real problem with TFA is that they are using polymers and I don't like polymers.
So anyway, yes we can make feature sizes less than 30nm with both electron beams and ion beams; however we are still a long way away from being able
depends on what you mean by smaller features. With 30keV Ga ions on Si the effective range is on the order of 27nm which basically limits your z resolution to something around 30nm, You can do a bit better with lateral resolution, FEI claims something on the range to sub 10s of nm, but I'm really having difficulty with the choice of the term lithography.
Lithography usually refers to some sort of masking procedure but the real advantage of ion beam is that you can do deposition and milling. You can do similar things with electron beams but its usually referred to as electron beam gas deposition or etching. E-beam lithography usually refers to using photons generated when the electron beam hits the resist to induce a chemical change in the resist which is then developed similarly to standard photo-lithography.
So while you could use a FIB to activate your resist... I don't really see why you would as the resolution is crap compared to a good electron beam. If you aren't using a resist in your so called ion beam lithography then I need some more explanation as to what you mean by ion beam lithography.
A small disclaimer: my PhD research is in the simulation of FIB milling.
Well that and they are serial and thus slow. (Yes I know about the parralell methods for both E-beam and Ion beam [also ion-beam litho, not direct write maybe for making nano-imprint-masks]) So the reason they are expensive (they aren't: E-beam is way cheap for the resolution, its just you'd never want to wait for even a single layer of a real device with E-beam litho on a production scale) is that you need lots of them to get anywhere near the throughput you get with photo-lithography.
Sure this technique might be a neat way to make nano-imprint masks, but then again 30nm isn't all that sexy.
Of course the European contract-less plans sometimes get even better. Here in Austria I pay €10/month for 500min +500txt +5GB. Voice and txt coverage is good data is good everywhere I have to wait (trains and platforms), roaming can be a bitch but since its a Chinese smartphone it has two sim cards so picking up another shortterm number when traveling is rather trivial.
That and even very efficient (nearly perfect these days) desalinization takes a bunch of energy. I was being a bit tongue in cheek, though we do harvest certain elements from sea water using electrolysis.
Or we could ya know... USE the SALTS. Those are valuable chemicals! Think of all the magnesium and sodium and bromide we could get if we honestly had cheap enough power to be worried about the ecological impact of desalination.
That's not quite true,
very often people with money are smart enough to not re-invent the wheel if they don't need to and would much rather pay you a consulting fee or a service contract to tailor the new software to the job they need done. That produces a positive incentive innovate as you now receive feedback as to the parts of your innovations that have marketable value.
The trend of clean room reimplementations came about BECAUSE of overly strong copyright, if you can purchase the time of the creator for a reasonable fee why would you ever pay more to steal it??!
Slashdot Mirror
With respect the the Samsung:
Autofocus: good in bright light, passable in low light, it really is better than the old canons (5Dii) in mid-low light. Nowhere near the newest AF SLRs. There is no good way to control focus in C-AF mode.
Weight: Very nice! Even with the 16-50 f2-2.8.
Durability: Appears to be at or above the 5Diii level. Not quite as overbuilt as an D810. It feels basically like a 7D.
Usability: Complete garbage! Nice button placement but boneheaded firmware. In video mode, we let it slide because the video is so awesome.
Support: 2 Firmware updates already, new ones on the way (sadly these ones are video focused).
Weather Sealing: S lenses are gasketed, body appears robust. We won't know till someone tears one down though.
Lenses: 24-70 eq available and quite good. 70-230 eq reportedly of similar quality sharp and fast. Neither are landscape lenses. 85mm f1.4 (135mm eq) check, very sharp very corrected a bit high on CA but its a system of lenses that are supposed to be software corrected. 60mm macro check. 30mm pancake check. UWA zoom check minus.
Viewfinder Quality: A new standard as far as latency, good sharpness as compared to A7.
Lightroom support out of the box: It includes a copy of lightroom! Sadly no Capture One support other than through an included DNG converter.
At the moment it really is an astounding 4k video camera.
Stills are good enough for events, there are reports of sports working fine if you can fill the frame. Its a PITA for slow fine arts stuff because of wide open focusing and a nerfed AF-S mode.
chopped up or whole? I'm thinking either way with a standard sedan you've just managed to end up on the less protected list.
I'm pretty sure the interesting metric is the fracture toughness.
Interestingly enough there is a NIST page on it:
http://www.ceramics.nist.gov/s...
which ranges from 1.89 to 4.45 MPam1/2
and a nice paper on anealed borosilicate glass fracture toughness:
http://dx.doi.org/10.1016/j.ph...
which ranges from 1.5 to 1.7 MPam1/2 depending on loading.
Of course Gorilla glass might have slightly higher values.
I think the OP is refering to plastic as in the adjective
plastic
1632, "capable of shaping or molding," from L. plasticus, from Gk. plastikos "able to be molded, pertaining to molding," from plastos "molded," from plassein "to mold" (see plasma). Surgical sense of "remedying a deficiency of structure" is first recorded 1839. The noun
Lol,
shouldn't it be an Otto-Bahn after the inventor/discoverator of the diesel cycle?
Also I laugh a little as the Autobahn (car track?!) here in Austria seems to be a toll road. I'm not sure since I don't own the cars I drive but we have to have a special sticker that costs a couple hundred euro a year to use them.
Yeah, but the cheap solar is poly silicon, which is slightly more labor intensive than aluminium.
NO! I think he should be rewarded by those who can appreciate the mastery and wish to buy a signed print. Do prints of mainstream photography that have made it onto postcards deminish in value? They don't. The artificial scarcity is interacting with creator and recieving an individual product.
Free loaders are fine, people will find a way to show how bad-ass they are that they can afford to advocate art, not only generic art but art that fits their perception of good.
maybe in a perfect world... but check out some of honeywell's patents on parasitic power for thermostats that have been reissued without significant changes multiple times. The patent office is broken which is part of the problem.
spoken like someone who has never tried to make a direct copy (even of software) that functions to the same extend as the original. Copying takes work and takes developement. Why do you think that science isn't dominated by the Chinese? They have the grad student populations to put a smart student to work trying to reproduce every interesting scientific paper (where they even give you the recipe most of the time), it doesn't really work out because by the time the student has "discovered" how to do the cutting edge stuff from 2-3 yrs ago; the entire field has moved on to something different.
Another fun example: there is a small korean motorcycle company that literally copies the suzuki SV650 as in the engine castings look nearly identical. They can therefore sell the bike for 20-30% less than the japanese do, however, the copy only makes 70% of the power of the original and has attrocious suspension and brakes.
Copying is not easy with tangible goods or technology
I have to agree! I really want to teach a digital logic course to high-school students. My professor was crazy (in a bad way) but it was the most influential course for me too.
It also works in most American universities if you avoid the engineering prerequisite courses, just dropping your head in and asking to sit in on a course (as long as you are a student) works suprisingly well. Otherwise there is the option to sign up for many more courses than you plan to take and drop the ones you don't want before the drop date. I did that rather often in grad-school with math courses. The university won't tell you to do it but it is a legitimate use of the rules they design. It is really the only advantage to the all you can eat pricing that you get/have to pay as a full time student.
So why is this a game changer?
Well its not yet but it could be...
So basically we have to make small chips, this is because the parasitics decrease with size so we get more efficient as we get smaller (up to a point), we also get cheaper for simply geometric reasons (wafer/exposure area pretty much fixed cost) so yay more goodies from a chip.
Except... .032m lithography mask costs $500,000+ and it probably won't work perfectly for at least a couple of revisions??! That sucks!
Tooling up is expensive and HARD, dude what do you mean a
Why is this placement technique cool?
One of the neat uses for big chips today are graphics chips. They are huge but relatively simple, they have a bunch of repeated processing units and a buttload of cache and some neat front end or back end. The top generation GPUs are usually made about as big as one can physically make a chip (we normally expose wafers in chunks, and precision and field of view are competitive design goals when thinking about the optics for these systems), because they problem is embarrassingly parallel we can use all that performance and it even scales reasonably well across multiple cards/nodes. So imagine we could make a functional dice that contains a very small number of processing units, some cache and some glue logic that makes it easy to connect to some magical buss, now we can make a GPU as big as we want. If we are a GPU company we like this since we now have something like linear scaling for processing costs to go to MUCH bigger chips, which means we can sell a new super computer every few years without having to compromise our consumer line.
Other cool things:
Lab on a Chip, you can throw down MEMS and processing components on a small device, imagine a full digital assay built into the pricking needle.
Harry Potter Newspapers, if you can do alignment and finishing after the fact you can print your e-ink display and driver in whatever size you need. All thats needed is good yield on the sub-displays.
Disposable-lazy-electronics, we are getting here anyway with RFID, but how cool would it be to carry a roll of stickers that would act as line of sight GPS extenders but were solar or wifi powered. Add some sensors and you have a crowd-sourcable metro monitor that knows where every train is and what stops are crowded.
To summarize, this invention has to capability to possibly decrease the cost of sticking widgets together, thus we can have many little widgets instead of monolithic widgets.
except for they aren't refering dies as in what you use to smash semi-soft metal into new shapes but actually dice, as in what you get after you dice (or cut up) a wafer.
TLDR: you are correct with tool dies, he/she is correct with wafer dice.
define blowtorch, in colloquial american english it seems to be a Propane or Oxy-Propane/MAPP torch. If thats what the parent is refering to I've used all 3 within the last year. My first date with my gf was introducing her to glass-working using the lab's Oxy-H torch and she started with fused quartz. I would argue if acess is available and some sort of guide or manual is near many people will pick up any old tool if they think it will help with the job.
I've been in an out of the piracy game since irc times but there was one approach that really worked for me.
The author of Lux (a java based Risk game) had a nice system for detering privacy:
1st: The game was free to play for 10-20 times and then it required registration (simple key code)
2nd: The author had set up a website so when you searched google: lux warez, serialz, serial, keygen, his website was the first site you got to where he asked crackers to respect his tiny cottage industry (I think it was 5-15$ for a lifetime key), and at the same time pointed out to users that by stealing his software they were poisoning his part of the ecosystem.
It seemed to work. I never found keys to the software (this was 6-7 years ago), and we didn't pirate that piece of software. I stopped looking for keys after I'd read his page and that was the important part.
On the other hand I have very little problem pirating professional software to play around with 3DStudio and Photoshop, however once I got into photography (and had spent much more than the cost of software on gear) I've had to change my approach. I pay for my Raw software (Capture One Pro) and I use gimp or open source tools instead of PS. Sometimes I want to dick around with CAE software and I have no problem pirating that since I'm interested in demo-ing it and not using it as a tool in my business. I think reminding users what they would be paying for (its your time not the tool) is the best approach.
except that E-beam lithography is in effect lithography, the following steps are harder and require lots of infrastructure.
Here is a typical process for getting a single layer into a chip.
Step 1: Clean the substrate of any organics.
Step 2: Apply resist (usually using a spin on process)
Step 3: Expose resist (E-beam -- Photolithography it doesn't matter). The hard part here is exposing in the correct places.
Step 4: Develop resist, usually wet chemistry which will remove or leave only the areas exposed in the previous step.
Step 5: Use the patterns made with the resist (Deposition, Etching, or Implantation)
Step 6: Remove the exposed resist, usually a different wet chemistry.
Then remember that you are going to do this entire process numerous times ( A simple P-MOS needs 4+ cycles without considering metalization). It also HAS to be done in a clean room if you want ANY flexibility as you have to switch the substrate between different machines for each step.
If we decide to go at it like 3D printers where one machine does every step (implantation is still kinda difficult but it could be done with a FIB) then we need to be able to predict exactly what the beam will do which we can't do yet. We are working on learning how to do that but we are not there yet. If we had all the knowledge to be able to build chips like a 3D printer I'm still not so sure we would as the general case since the batch process mentioned above is very cheap (per device). Then again I would not have expected the 3D printer movement to have taken off so quickly, so it could happen.
The costs are already coming down quickly, there are desktop SEMs that cost less than an expensive SUV, the gas injection systems are nothing more than capilarry tubes and solenoids (neglecting that most of the deposition gasses are wicked toxic and may explode if they contact air) so I would argue that the tools are already very much on thier way to being cheaper. The problem is we really don't have any systematic approach to using them to do what I think you are suggesting.
Sorry, I went all internet tough guy back there...
I should clarify what I meant.
1st: E-beam lithography as I know it; with an E-beam resist is pretty much the creme of the crop if you want ultra high resolution. It is also a very old technique IE they were looking at it to replace photo-lithography as far back as the '80s but there are difficulties with making a bright electron beam to do the lithography in a parallel manner. Therefore its been used serially with a beam rastering the resist to make the desired patterns. With this techniques you can make very small features.
2nd: I am un-aware (doesn't mean it doesn't exist, just that its outside of my research area) of any analogous ion beam processes; in that we are talking about using a polymer resist activated by an ion beam. There are however very interesting nano patterning methods that use implanted ions either in a sacrificial layer or in the substrate itself, followed by selective etching that could arguably be thought of as ion-beam lithography.
3rd: Focused Ion Beams (FIB) is a rather mature technique for circuit repair and editing because it acts as both and additive and a subtractive process. With the FIB we can make deep holes using gas assisted etching, and then deposit with gas deposition both conductors and insulators. The real advantage of this technique is that we can see what we are doing!! Imaging can be done either with the ion beam or a separate electron beam allowing us to see the structures we are working on with the same or better resolution than we can write or etch with. Normally however FIBs use Gallium (Ga) ions as they are a convenient ion source (the melting point is low and the vapor pressure is also low) these ions are rather heavy and cause damage to the substrate (this can be mitigated through careful selection of the beam energy and angle), Ga also acts as a dopant in silicon.
4th: There was a company that tried to deal with the serial nature of focused ion beam milling. This company developed a 1024 beam array where each beam could be individually steered or turned on or off using a selector plate made with standard Si manufacturing techniques. This device used Argon (Ar) ions to avoid doping. Sadly it seems this company has stopped developing this device. They might be entering with a similar setup for electron beams in the future. My understanding is that the ion beam device worked best for gas-assisted processes where the deposition or etching gas is activated by secondary electrons freed when the ion hits the target. Seeing as an electron beam also free secondary electrons I think they changed directions to an electron only technique but these are only rumors I've heard around work.
Both Ion beam and electron beam techniques are more difficult than they appear as the yield (either sputtering or secondary electron) is dependent upon the incidence angle between the beam and the surface. It therefore becomes much more difficult to predict the interactions once the surface is no longer planar.
My comment about the 30nm not being all that sexy was with respect to TFA, I saw this on FEI's facebook page a couple of weeks ago and thought the same thing. Yes its neat that they can make shapes at this size with good control (heaven knows we can't do it yet with electron deposition or fib deposition [we can make cute test cases but we are far from arbitrary shapes even though we can do overhangs already]), but for me the real limitation is that they seem quite limited with respect to the materials that they can make things out of. I'm sure this is a great thing and we will see some neat tricks in the future with people either using these printed structures at templates for some nano imprint lithography, or as high tech resist with some neat deposition into the voids. My real problem with TFA is that they are using polymers and I don't like polymers.
So anyway, yes we can make feature sizes less than 30nm with both electron beams and ion beams; however we are still a long way away from being able
depends on what you mean by smaller features. With 30keV Ga ions on Si the effective range is on the order of 27nm which basically limits your z resolution to something around 30nm, You can do a bit better with lateral resolution, FEI claims something on the range to sub 10s of nm, but I'm really having difficulty with the choice of the term lithography.
Lithography usually refers to some sort of masking procedure but the real advantage of ion beam is that you can do deposition and milling. You can do similar things with electron beams but its usually referred to as electron beam gas deposition or etching. E-beam lithography usually refers to using photons generated when the electron beam hits the resist to induce a chemical change in the resist which is then developed similarly to standard photo-lithography.
So while you could use a FIB to activate your resist... I don't really see why you would as the resolution is crap compared to a good electron beam. If you aren't using a resist in your so called ion beam lithography then I need some more explanation as to what you mean by ion beam lithography.
A small disclaimer: my PhD research is in the simulation of FIB milling.
Well that and they are serial and thus slow. (Yes I know about the parralell methods for both E-beam and Ion beam [also ion-beam litho, not direct write maybe for making nano-imprint-masks]) So the reason they are expensive (they aren't: E-beam is way cheap for the resolution, its just you'd never want to wait for even a single layer of a real device with E-beam litho on a production scale) is that you need lots of them to get anywhere near the throughput you get with photo-lithography.
Sure this technique might be a neat way to make nano-imprint masks, but then again 30nm isn't all that sexy.
Darn,
no more people coming back and thinking science is sexy.
Of course the European contract-less plans sometimes get even better. Here in Austria I pay €10/month for 500min +500txt +5GB. Voice and txt coverage is good data is good everywhere I have to wait (trains and platforms), roaming can be a bitch but since its a Chinese smartphone it has two sim cards so picking up another shortterm number when traveling is rather trivial.
I'm an ex-pat dating an Austrian girl I met in China.
Being in an alien land does a lot to force one to revise their preconceptions. I really recommend the experience to everyone!
That and even very efficient (nearly perfect these days) desalinization takes a bunch of energy. I was being a bit tongue in cheek, though we do harvest certain elements from sea water using electrolysis.
Or we could ya know... USE the SALTS. Those are valuable chemicals! Think of all the magnesium and sodium and bromide we could get if we honestly had cheap enough power to be worried about the ecological impact of desalination.
That's not quite true, very often people with money are smart enough to not re-invent the wheel if they don't need to and would much rather pay you a consulting fee or a service contract to tailor the new software to the job they need done. That produces a positive incentive innovate as you now receive feedback as to the parts of your innovations that have marketable value. The trend of clean room reimplementations came about BECAUSE of overly strong copyright, if you can purchase the time of the creator for a reasonable fee why would you ever pay more to steal it??!