Thursday, August 30, 2007

prelude to the pascaline part 2

Building and simulating the octet truss has caused me to get over my fear of working with larger parts. I am thinking it might be time to revisit a mechanical calculator in earnest, like actually design one before I start building anything...starting tomorrow. Here is something I am pretty sure I will need: 10 spoke spur gears

So far I like this design because they are robust, and I can trim off spokes to create a "ones" gear. Plus I should be able to attach diamondoid shafts or cnts to the sides. This is a shot in the dark, but does anyone have any engineered drawings of a pascaline?

Wednesday, August 29, 2007

carbon nanotube octet truss simulation -or- it's a ROCK

The simulation was at 300K for 1500 frames at 20 femtoseconds per frame. It has been my experience that if stuff comes apart, it happens before then. Looks pretty solid, as it should I guess.

Sorry, no QuteMol. I tried, but I'll have to wait until I have a day to devote to it. I think the first plug-in for NE1 I write will automate the QuteMol animation process considerably. I could do it now with a Windows app that accepts some parameters and then recreates my very repeatable keyboard and mouse movements (advance frame, save, advance frame, save...), but that's not the point or elegant. If only I had access to some people that knew the ins and outs of the NE1 source code! Wait a minute, I'm getting an idea...

*EDIT: that's 2 femtoseconds per frame not 20

an explanation on how you're seeing the dna

Maybe you are thinking that the DNA strands in the last few pictures doesn't really look like anything but beads on a chain? This is because the DNA strands used to build these cross motifs is being represented in PAM5. This is a convention that uses five groups to represent the major structural components of the DNA strand. I think, and correct me if I'm wrong, but there is a group (bead) to represent a phosphate, each of the two sugars, the nucleotides and the hydrogen bonding on the nucleotides. That's five. That's what I'm going with tonight. It is a hell of a lot easier to work with these groups than the atoms they represent. However I must admit the atomistic representation of the DNA chains is more visually appealing:

The octet sim is done in half the time I thought. It is no wonder I am usually early for appointments. I'll post it later tonight or tomorrow.


Tuesday, August 28, 2007

CNT - DNA hybrid array in technicolor

The title says it all.

The different colors on the cnts is because they are not b0nded together, merely placed next to each other. I know. I'll bond them. At least this gives you an idea of the scale of things. I am very interested in combining DNA with cnts, so this could turn into something.

The octet truss sim is crunching away on the old sim-computer. I am expecting it to take around 100 hours. It will not be a terribly long simulation.


Sunday, August 26, 2007

a nanofactory graphic

This was obviously inspired by the nanofactory animation, but I only spent an hour or so on it. I would like to switch out the bar shafts for hexagons and increase the tool tip number. We'll see where this thread goes.

We lost power here, so I had to restart the octet truss simulation. I had to choose between a backup power supply and a backup storage device. I went with the hard drive; what can you do?
I may need to restart it again to save time. Right now I am running it on 1.6 GHz Celeron processor, not my normal simulation computer. After 4 hours it is 0% done. I'll try to swap some things around so we can see how it turns out this year.

Saturday, August 25, 2007

a short DNA primer for the future: sticky ends

This is probably the first of a series of my personal tutorials on building stuff with DNA. I am sure as things get more complicated I will cop out and just link to some fine resources, but for now you're getting my take on things. This may be review for some, but not for others, and I would like to build a one-stop-shop for the information I may use in the future. I did do a couple science fair projects involving DNA in junior high. One included building a model out of dowel rod and window blinds. It rotated. I got an A++ (brag!). The other was an ill fated attempt to mutate Oak trees which I may still revisit one day. Unfortunately due to time constraints, I must assume you did a couple similar projects in school too, so you would know all the DNA- "blue print of life" stuff. Please let me know of any errors.

Let's just dive in...

For building DNA structures we first need to focus on the structure of DNA itself and how strands of DNA join (or won't join) together.

The frame work, or backbone, of a DNA strand is composed of alternating sugar and phosphate groups. I am going to represent these by S and P respectively.

So we have a chain of:

These sugars are the deoxyribose in deoxyribose nucleic acid, and they come in two kinds: 3' and 5' (pronounced 3-prime and 5-prime). The 3 and the 5 represent how many carbon atoms the sugar contains.

A single strand of DNA will have alternating sugar groups and alternating ends:

But doesn't most DNA come double stranded?
YES, so the groups on a double strand of DNA would look like this:

These strands are connected together by hydrogen bonding of the nucleotides, A, T, C, G

I have developed a somewhat "blue" method for remembering which nucleotides join up. I was inspired by a pin-up art calendar a friend bought me for Christmas years ago: Just remember T and A always go together.

The nucleotides join up at the sugars:
A---T--- C---G
The dashes between nucleotides are just for formatting purposes

Also it is probably important to point out that hydrogen bonds are weak bonds, allowing the DNA to seperate.

If we have a strand of DNA, it is possible to cut it apart using an enzyme called a restriciton endonuclease. These enzymes can cut apart a strand of DNA in a manner that leaves a "sticky end", meaning an end that would happily join to a sticky end of another DNA strand:



Here the overhanging nucleotides form the sticky end. I used NE1 to create sticky ends on the DNA chains in the cross structure I built last night.

So that was the first DNA tutorial.


Friday, August 24, 2007

exploring the DNA features of NE1

First for the cnt data storage system, I am trying to produce an image of the electrostatic potential (ESP) of the N and F atoms, just to see what I can see really. However I have not been able to yet, still figuring out how to do it with NanoHive. <- that's a link to the NanoHive gallery. There is an example there of an ESP simulation of a carbon nanotube. This evening I was trying to use NanoHive to make the ESP image. I spent a half hour on it and then fell asleep. I had an amazing dream that I totally figured it out, everything was clear as day; I was even using the program to forecast hurricanes, something I am pretty sure it can't do. Alas it was just a dream... Also I haven't forgotten about the octet truss. I am going to anchor the ends and run a simulation to see if I can get an idea of how rigid it is. Depending on what happens there I may expand the simulation ambitions.

Other than just playing around, I have not done anything with the DNA modules of NE1, until tonight. Here is a DNA cross motif I built following a tutorial...more or less:
This image was rendered in QuteMol and is of DNA chains connected together to form a cross.

These arrays have actually been synthesized, so we are more into science fact than fiction. I'm just getting into myself, but I understand DNA nanostructures have big time potential.

Definitely MTF

Wednesday, August 22, 2007

first simulation of the carbon nanotube data storage system

Here is the first simulation. It is short and not that interesting, other than everything staying put, no nitrogen atoms shooting off or anything. That's always nice to see.

I plan on doing some more with this.


Monday, August 20, 2007

Tool Tip inspired....

I am trying to build something that resembles the tool tips featured in Nanorex's Molecular Manufacturing Gallery and this Nanofactory animated film:

Here is what 30 minutes has produced:

Again this was built to look like it might work. I have no idea how it would function in the role of a tool tip. I just want a place holder for my own animations, and this will work until I reverse engineer the real ones.

stills from the NASA data storage system

Still working/waiting on the simulation for these, but here are some still shots recreating NASA's CNT data storage system:

The yellow atoms are fluorine and the white ones are hydrogen. In the original diagram fluorine is being used to represent a binary one, and hydrogen a zero. I am looking into how this is all supposed to work. Right now I think it must have something to do with moving electrons away from the blue nitrogen atom to the fluorines, just based on the electron affinities of these atoms.
Anyone with anything else to add?

Also, in the interest of disclosure, today was my first day as a full-time Nanorex employee.

Sunday, August 19, 2007

the assembled DDM

My work here is done. It was fun. I would like to thank Damian, of course, Nanorex for providing the tools, the Coca Cola company for their fine refreshments, Johnson & Johnson for cooking up an extra strong batch of Visine, Gino's Pizza Rolls, and my computer for spontaneously turning off only once.

On to the next project...

Saturday, August 18, 2007

last part of the DDM

Looks like my deposition mechanism is also a fan. In my last post I called this the base. That really is a misnomer. I have thought of it as the "base" in my mind because I started calling the bearing part the "top". This "base" is really a section of a surface to show what the deposition mechanism does: deposit atoms. Mine thinks I am #1.

So now I think all that's left is putting everything together.


close ups of the DDM base and serendipity

Here is a picture of the DDM base. I have not built mine yet, but I don't anticipate any problems:

The red dots are where I deleted the green atoms, fluorine. The white dots are hydrogen as usual. After I wrap up the DDM (tomorrow?), I am going to reuse this base and its fluorine/hydrogen surface to make this:

This image is from NASA, and I believe it is therefore in the public domain. The one I build will look way better than that, and it will move. Maybe I should worry about NASA stealing mine? Just kidding. If they want it, I would be honored to let them have it.

Thursday, August 16, 2007

elbow simulation

Here are the DDM arms not doing too much of anything, which is good. This simulation was at 300K.

ready to simulate the elbows

I connected the arm parts with an elbow joint. Again, my elbow is one the left, Damian's on the right. The black blobs on the bottom are a mass of anchor jigs. While my elbow joint may be an exact copy, it might not. That is to say I believe it is, but instead of trying to match the original atom for atom, I am going to simulate both at 300K and if mine stays rigid, then I will call it good. Of course this is assuming Damian's does too. Stay tuned!


Wednesday, August 15, 2007

positioning the arms on the DDM

Not much tonight, just getting ready to build the elbow that connects the top and bottom arm segments. Here is the original:

And here are my segments positioned to be connected:

And a little movie of the DDM top performing as a bearing:

Tuesday, August 14, 2007

continuing the DDM

First thing I did tonight was extrude the ring I built yesterday to make the first arm part of the DDM:

Next I decided to build the deposition tip. For this all I did was isolate Damian's and approached it like building a big o'l moiety. I probably put a half mile on my mouse with all the view changes and rotations. However I found a better way to build things like this: use the ball and stick view. Switching to this display helped a lot because it is easier to focus on just the atoms without the bonds. Here are the tips in CPK for comparison:

You can probably tell that the slightly irregular one on the left is mine:

It still needs some fine tuning and some shifting, and I just check the properties of both these chunks when I grab that picture; mine is missing a couple carbons and some hydrogen. I guess I'll go chase those down.


Monday, August 13, 2007

taking on the DDM arm

Here's a picture of the DDM arm and deposition tip. My next step is to rebuild the arm portion. Notice that I am starting with the parts that are similar to what I've built before?

Same steps here, just delete the deposition tip and see what I can see:

This is the arm displayed in the tubes view. Right away I picked out six pieces, thinking that a single piece could be used as a moiety. This piece to be exact:

And if you extrude this, and manually bond the bottom carbons, you get the ring back:

Mine still looks a little loose because it has not had the energy minimized yet. But I'm on the right track.


Sunday, August 12, 2007

I was wrong...this was the easy part

There is an inner ring that sits inside the outer one I just built. This creates a bushing mechanism.

Here is a pic of the original nested in the outer ring:

See it?

Here is just the inner ring displayed in tubes view:

This diamondoid ring looks familiar. I know because I have already spent an hour taking it apart to build Damian's carbon nanotube bearing. This is a good lesson: Don't feel like you have to reinvent the wheel. If you get something that works, keep using it. A slightly less talked about, but equally valuable, lesson is: Never throw anything away.

I just happen to have a copy of the ring moiety laying around here:

Extruding this one gives a ring like the original:

Mine is on the left; or is it on the right? Hmmm. Just kidding. What? Have you heard that one before?

Putting them together:

Beautiful. However I might need to go back and extrude the outer ring to 25 instead of 24. It looks a little tight. I guess I could actually measure Damian's ring's diameter. We'll see.

I am now really sure that was the easy part.


building the outer ring on the DDM

As my good friend Dr. G used to say before beginning a new lecture, " a modicum of review":

I was in the midsts of rebuilding Damian's DM, starting with the outer ring on the deposition arm, and trying to piece out a moiety I could use to build a new ring.

I settled on this one:

This would not produce an exact copy of the original ring because of the differing locations of the oxygen atoms around the circumference. However I now believe this one will work.

To arrive at this conclusion I built another copy of this moiety from scratch and bonded it to the one above in the plane that goes into the screen you are reading this on. You can imagine that the moiety in the above picture is the keystone of an arch with the rest of the arch branching out and downwards to the left and right. These arches would connect to form the ring. I have doubled the thickness of this moiety.

(Big Time Review: for the purposes of this site we are calling the smallest repeating pattern of a part a moiety)

I got this:

This would be rotated 90 degrees from the above picture, so we are looking into the cross section of the ring.

Extruding this 24 times as a ring produces something pretty close to the original:

Mine is on the left; or is it on the right? Hmmm. Just kidding. You can see by the placement of the oxygen that the new ring is the one on the left.

Alright, that was probably the easy part.


Saturday, August 11, 2007

a little more messing with the DDM

Here I am carving up Damian's ring to try to find a potential moiety.

This could be one. Let's see what happens if this moiety is extruded 18 times in a ring formation:

The resulting ring is one on the right and Damian's original is on the left. It's not exact, but that was expected. I don't know yet if I can use it to make a final ring for the DDM. If I can, then I will have to go back and build the moiety from scratch one atom at a time to reform the ring, just for my own sense of satisfaction.


messing around with Damian's deposition mechanism (DDM)

I planned on using the same basic approach to recreate this one as I have with others, one piece at a time, and small pieces at that. I decided to start with the outer ring, so I needed to isolate it. The entire part was sent as a protein databank file, so it is all one chunk. I need to delete everything not part of the ring:

Here I am removing everything else. Next I try to find a symmetry or some basic pattern I can start with:

Orientating the ring this way it appears that maybe there are two similar rings joined together. The middle seems to resemble a type of seam to me. I figure I can cut the ring in half, reverse-engineer one half and join two together to recreate this part. I deleted all the atoms left of the seam, leaving the red oxygens:

Here is one half of the ring with the oxygens displayed in CPK view. If you look closely you'll see there is not a perfect pattern for the locations of the oxygen atoms. Damian said this was built before NE1 and its extrude function. This means I cannot piece-out a single moiety and extrude it to get this exact ring. Unfortunately I don't have the patience (and probably skill!) to build a ring like this without using the extrude function. <- That's sort of the way I do these: piece out the moiety and extrude it, just like with carbon nanotube bearing. I am going to have to deviate from the original design, hopefully the changes will be immaterial.

Thursday, August 9, 2007

general update

Here is the current status of things at Machine Phase:

  • I am still thinking of what original project I want to tackle next and open to suggestions.
  • The C60 pump isn't dead yet, just mostly dead, so partially alive.
  • My sim-computer is now free, but I am hesitant to start a crazy long simulation on the octet truss until I am sure I won't need it for the pump. Also I am still considering the best way to simulate it.
  • This might be my favorite structure in Damian's gallery, not least because there is a lot to it I don't really know how to build. Plus I just think it looks great. I would say it is the Grand National of the gallery. <- a vague comparison but a compliment none the less.
And look what I got in my inbox yesterday:

For now I think I am going to spend a few hours taking this apart and putting it back together again and showing you how too. Just in case you are wondering, I got Damian's blessing for this series of posts.

Addendum: OR maybe it is time to look at some of the stuff NE1 can do with DNA?
I'll sleep on it...

Tuesday, August 7, 2007

still here, some bad news on the C60 pump, and copying Damian...again

Well after around 400 hours all we got was a wobbly tube. It looks like the cnt paddle might be too flimsy for this design as it just slips past the fullerenes. The good news is everything held together, so I could probably build a SiC bar rotor and really turn up the juice (slang for increasing the rotor speed and torque).

This is the first long simulation I have done in the Alpha 9 release. A while ago Damian filled me in on some of the improvements the simulator underwent between Alpha 8 and 9. There were five areas addressed, but I believe the one most relevant to the pump and nanotube systems used here were extending the parameters used to calculate graphic systems. These were made more rigorous. This may explain why the C60 survived the crash tests so well, but kept breaking apart in the pump.

Also remember when I tried calculating the modulus of rigidity and Young's modulus of a carbon nanotube and they didn't come out right? Well, those were done in A8 with old parameters, so I should probably redo them in A9.

Other than that I have been benchmarking Damian again. What is it they say about good and great artists, something about copying and stealing? I have been trying to recreate his carbon nanotube crimped junction. So far mine is different because the tubes are bonded to the rings. From what I can tell from Damian's site, his tubes are not. I have built a ring that is not connected to the tubes by placing nitrogen around the inner circumference, but this is not like Damian's either. I am sure he would probably just send me his to figure out, but this is kind of fun. In NE1 if you have a scroll wheel on your mouse, you can press it and freely rotate the view of a part. I don't know how many times I will be squinting at some picture in Damian's gallery and push the scroll button down to try and rotate the view.

Here is mine thus far:

And my ring that's not right:

I am still wrapping up my other project. I will punch back in full-time soon.