|
|
| (12 intermediate revisions by 2 users not shown) |
| Line 2: |
Line 2: |
| | | | |
| | == Current Projects == | | == Current Projects == |
| − | === Molecular Dynamics of Block-copolymers (MDBCP) === | + | === [[MDBCP:Main Page|Molecular Dynamics of Block-copolymers (MDBCP)]] === |
| | + | * [[MDBCP:Main Page|Main Page]] |
| | + | * [[Running]] |
| | | | |
| − | ====Scaling up BCP Runs on DESE computers==== | + | === [[X9|X9 Beamline (X9)]] === |
| − | Division of Electrical and Software Engineering (DESE) in the Center for Devices and Radiological Health at the FDA is providing high performance computer resources and support for large-scale molecular dynamics simulations. Initial efforts will be on scaling up the moderate size runs (<100,000 beads for ~10 millions iterations) to larger size runs, meaning systems upwards of 1,000,000 particles. Simulations of films with an air interface, although scientifically of great interest, are inherently processor-inefficient and then thus difficult to maintain running without complaints from IT staff. Therefore, we will be working with the less scientifically important case of a bulk polymer material, ie one with no air interface. Two initial system sizes will be run for fairly short periods of time:
| + | * [[X9|Main Page]] |
| − | Warewulf Runs:
| + | * [[X9:User|User's page]] |
| − | *[[2X Runs]] (105600 atoms)
| + | * [[X9:Procedures|Beamline procedures]] |
| | + | * [[X9:Usage Log|Usage log]] |
| | + | * [[X9:Hints|Hints and notes]] |
| | + | * Software: [[X9:View.gtk|view.gtk]] and [[X9:pyXSanalysis|pyXSanalysis]] |
| | | | |
| − | *[[16X Runs]] | + | === [[ScatterSim:Main Page|Scattering Simulation Software (ScatterSim)]] === |
| | + | * [[ScatterSim:Main Page|Main Page]] |
| | + | * [[X-ray modeling|Traditional X-ray modeling]] |
| | + | * [[X-ray analysis|Traditional X-ray analysis]] |
| | + | * [[ScatterSim:MultiComponentModel|Nano-object on lattice modeling (Multi Component Model)]] |
| | | | |
| − | ====Making HiDef Movies of Simulations==== | + | === [http://yager-research.ca/softmatter/index.php/Main_Page Proposals for NSLS-II Beamlines] === |
| − | Goal: 720p movies of simulation trajectories.
| + | * [http://yager-research.ca/softmatter/index.php/Main_Page Main Page] |
| − | Procedure:
| |
| − | * set povray width/height flags to 1280x720
| |
| − | * collect consecutive snapshot images w/ name format as follows: image%04d.png
| |
| − | * using ffmpeg ( http://ffmpeg.org/ ), create a playable movie file as follows:
| |
| − | ffmpeg -b 500000 -i %04d.png -r 12 -vcodec msmpeg4v2 -s 800x600 nice.asf
| |
| − | The flags may be optimized to improve movie quality in the future.
| |
| − | The msmpeg4v2 codec was used to create .asf formatted movies, as surprisingly, this was the only combination that would provide playable movies for powerpoint on a windows machine, specifically my FDA laptop.
| |
| − | | |
| − | ====Manuscript Drafts====
| |
| − | * [[Letter_Draft]]
| |
| − | ===== Figures =====
| |
| − | | |
| − | Working on perl script to call 'configuration.cpp' and perform calculations.
| |
| − | | |
| − | To be run in directory containing dump data, param_set.txt and config_dumpfiles.in. The script creates a subdirectory, EXT-FIFO, if it does not exist. The directory contains all necessary input files to do an extension. The dump output is written at a very high rate to a named pipe, which is then fed into the configuration.cpp program. Thus, no dumpfile accumulates and the lammps output only exists long enough to do configuration.cpp calculations on it.
| |
| − | | |
| − | usage: mkext-fifo-2-24-10.pl [ jobs ] where [ jobs ] are the command line arguments taken by configuration.cpp (a.out).
| |
| − | | |
| − | /home/cforrey/CODE/datedCode/mkext-fifo-2-24-10.pl
| |
| − | | |
| − | ======simulation procedure======
| |
| − | * [[Spin Casting Procedure]]
| |
| − | * [[Spincast "Selective Etch"]]
| |
| − | * [[Temp vs time]]
| |
| − | | |
| − | ======polymer characterization======
| |
| − | * [[Surface Tension]]
| |
| − | * [[χ Determination]]
| |
| − | | |
| − | ======morphology vs substrate/temp======
| |
| − | * [[Fine-grained Temp / Morphology]]
| |
| − | | |
| − | ======lamellar characterization======
| |
| − | * [[Density(n)]]
| |
| − | * [[Projected Chain Density]]
| |
| − | * [[Inter-tether Pair Correlation]]
| |
| − | * [[Block Persistence]]
| |
| − | | |
| − | * [[End-to-End Distance]]
| |
| − | * [[Fourier Transform]]
| |
| − | * [[Lo(T)]]
| |
| − | * [[φ(z)]]
| |
| − | * [[θ(time)]] (end-to-end vector w.r.t. z-axis)
| |
| − | | |
| − | ======Next paper:======
| |
| − | | |
| − | * [[Energy vs Horizontal/Vertical Lamellae]]
| |
| − | * [[Pressure Profile]]
| |
| − | * [future][[Partial Layer - Multiple Scans of Commens.]]
| |
| − | | |
| − | | |
| − | * [[Project Planning]]
| |
| − | * [[Paper Weaknesses]]
| |
| − | | |
| − | ====Recent Results====
| |
| − | * [[Bulk Sims]]
| |
| − | * [[Escaping Energy Minimum]]
| |
| − | * [[Chain_Configuration_Analysis]]
| |
| − | * [[Propogation of Alignment]]
| |
| − | * [[Fine-grained Temperature - Effect on Surface Features]]
| |
| − | * [[Partial_layer_runs]]
| |
| − | * [[Morphology_vs_ChainComp]]
| |
| − | * [[33x45_kbend2_moprhology]]
| |
| − | * [[pxx_vs_z]]: Also includes an analysis of Pzz(z)
| |
| − | * [[MDBCP:Fourier_transform]]: The FT data is coming along; might be useful...
| |
| − | * [[MDBCP:Log_2009_Nov_17]]: kbend 2, T1.4, g0.2 run for '''7 million''' steps. The 'weird' morphology has equilibrated into something much more reasonable (vertical LAM with no tilt). This suggests that kbend 2 is working fine; just needs to be run for longer.
| |
| − | * [[MDBCP:Log_2009_Nov_13]]: kbend 2 data.
| |
| − | * [[MDBCP:Log_2009_Nov_12]]: Larger system size simulation for LAM of higher bending energy (kbend=2), neutral surface (g=0.5), and intermediate temperature (T=1.4).
| |
| − | * [[MDBCP:Log_2009_Nov_12-cylinders]]: Small box size (7x16x2), for CYL (Na=3 Nb=7), for various temperatures (1.1 to 1.6) and surface energy (g = 0.2, 0.5, 0.8).
| |
| − | * [[MDBCP:Log_2009_Nov_10]]: gamma 0.2, 22x30 box; T data from 1.357 to 1.54 (including film pressure data). Islands/hole density changes with T. | |
| − | * [[MDBCP:Log_2009_Oct_20]]: early results of 'medium' box sims, showing 3 temperatures and 3 gammas.
| |
| − | | |
| − | ====Progress====
| |
| − | * '''Done:'''
| |
| − | ** See [[Done]]
| |
| − | | |
| − | * '''Running:'''
| |
| − | ** See [[Running]]
| |
| − | | |
| − | * '''To be done:'''
| |
| − | ** Composition and surface energy phase diagram (at a particular T, like 1.4?)
| |
| − | | |
| − | * '''Resources:'''
| |
| − | ** See [[MDBCP:Scripts]] for the source code of existing scripts.
| |
| − | | |
| − | ====Plan for Paper====
| |
| − | # Main points of paper:
| |
| − | ## We have a valid strategy for simulating BCP thin films.
| |
| − | ## We can reproduce many expected results for BCP thin films (islands and/or holes, L0 vs. T, etc.). Surface and substrate wetting play a crucial role, as already amply demonstrated in the experimental and theoretical literature.
| |
| − | ## We have identified internal film pressures/tensions/stresses as having a significant role in controlling film behavior (morphology orientation, L0, islands and holes... and also defect density, ordering kinetics, etc.):
| |
| − | ### A "happy" (stable, low-energy) film has a low internal tension. The formation of an island or hole is one place to "pay the energy penalty" of having incommensurate thickness. Internal film stresses is another.
| |
| − | #### ''Currently speculative:'' The tradeoff between forming islands-and-holes versus building up internal stress depends on the bending energy (chain stiffness).
| |
| − | #### ''Currently speculative:'' Tension builds up to a critical level before being released ("pop") with the formation of an island or hole (e.g. as T is changed and thus L0 changes).
| |
| − | ### ''Currently speculative:'' The vertical state is low-energy than horizontal on neutral surfaces, because a tension mismatch between the two blocks (slightly different cohesive energy) creates an energy penalty in horizontal configuration, which can be relieved (slightly) in the vertical configuration via surface "puckering". (We have experimental data showing the puckering.)
| |
| − | #### ''Currently speculative:'' A perfectly vertical state (rather than vertical with tilt) should be preferred because this can minimize chain perturbation at the film surface (and also minimizes surface area?). Thus a perfectly vertical state should appear when the bending energy is increased (and the surface tensions increase?).
| |
| − | # Figures: (tentative, of course)
| |
| − | ## A graph of T vs. time to explain our general simulation protocol. Sub-panels show generic image of a spin-cast starting state, and a final state (with an expected morphology; e.g. commensurate horizontal on a strongly wetting substrate). This justifies our methodology.
| |
| − | ## A few "boring" results (e.g. morphology vs. composition?), mostly to justify that our protocol is "getting the right answer".
| |
| − | ## Phase diagram (images) of gamma and T for LAM. This will show many expected trends (horizontal vs. vertical), and we will point out some interesting features (islands and holes, tilt angles, etc.)
| |
| − | ## L0 vs. T (and corresponding pressure vs. T?); trend is in expected direction (L0 decreases as T increases for the usual reasons) but exact scaling is informative.
| |
| − | ## Phase diagram of tensions for the gamma vs. T images. We can show the "residual final tension" or the "tension in z direction" or the "tension along LAM normal" or whatever best makes our point: that internal stress and tension mismatch between the blocks drives behavior.
| |
| − | ## Phase diagram (images) of composition and gamma for T=1.4 (or something). This will show the interplay between composition and film-confinement effects. There are plenty of papers showing, e.g. conversion from cylinder to sphere morphology in thin films. We can reproduce this behavior and (hopefully) show the tension/stress patterns that exist in the film (and, we argue, drive morphology changing...)
| |
| − | ## And some more...
| |
| − | | |
| − | ====Future Directions/Projects====
| |
| − | * Explore role of substrate topography (roughness, channels) on controlling orientation.
| |
| − | * Explore role of chemical patterning on BCP order. In particular look at BCP pattern fidelity (e.g. how are errors or LER in the pattern translated into the BCP LER), defect tolerance, persistence of templating in the z-direction, and so on. In other words, all the effects of note for using BCP chemical patterning for litho. Be aware of significant work already done: [http://dx.doi.org/10.1021/ma702514v DOI: 10.1021/ma702514v], [http://dx.doi.org/10.1103/PhysRevLett.102.197801 DOI: 10.1103/PhysRevLett.102.197801], [http://dx.doi.org/10.1039/b902283j DOI: 10.1039/b902283j] and others by Nealey.
| |
| − | * Combine MD simulations with scattering modeling, to compare with experimental data. (Possibly fit experimental data by adding an energetic bias in LAMMPS based on fit quality?)
| |
| | | | |
| | == Getting started with a Wiki == | | == Getting started with a Wiki == |
| Line 127: |
Line 28: |
| | * [http://www.mediawiki.org/wiki/Manual:FAQ MediaWiki FAQ] | | * [http://www.mediawiki.org/wiki/Manual:FAQ MediaWiki FAQ] |
| | * [https://lists.wikimedia.org/mailman/listinfo/mediawiki-announce MediaWiki release mailing list] | | * [https://lists.wikimedia.org/mailman/listinfo/mediawiki-announce MediaWiki release mailing list] |
| | + | * [http://meta.wikimedia.org/wiki/Help:Displaying_a_formula Working with equations] |
| | Some useful links: | | Some useful links: |
| | * [[Special:Upload|Upload a file]]. | | * [[Special:Upload|Upload a file]]. |
