Yingbin Ge's Blog

Numerical Hessian for a plannar molecule

2011-12-21T10:08:00.000-08:00

4 atoms:(x1, y1, 0),(x2, y2, 0),(x3, y3, 0),(x4, y4, 0)

145 energy calculations need to be performed:

1: at equilibrium geometry.
4: for 1z-1z 2z-2z,3z-3z,4z-4z.
6*2: for 1z-2z, 1z-3z, 1z-4z, 2z-3z, 2z-4z, 3z-4z.
2*8^2: for the second derivatives with respect to x(s) and y(s)

1 + 4 + 12 + 128 = 145

Numeric Hessian for H2NO with a Cs Point Group

2011-12-19T16:02:00.000-08:00

$force temp(1) = 10.00, 20.00, 30.00, 40.00, 50.00,
60.00, 70.00, 80.00, 90.00, 100.00
method=fullnum projct=.t.
$end
$contrl maxit=200 mult=2 scftyp=rohf runtyp=hessian coord=unique $end
$contrl ispher=+1 $end
$contrl cctyp=cr-ccl $end
$basis gbasis=cct $end
$system timlim=1000000 mwords=400 memddi=0 $end
$scf soscf=.t. dirscf=.t. $end
$statpt nstep=100 opttol=0.00001 $end
$data
H2NO
Cs

N 7.0 -0.1357158331 0.0890286127 0.0000000000
H 1.0 -0.3822861363 -0.3680309994 0.8687014208
O 8.0 0.0165139267 1.3587083668 0.0000000000
$end

The following 2nd derivatives are evaluated:

Between atoms 1 and 1 9 (2x-x 4x-y 2y-y 1z-z)
Between atoms 1 and 2 (not calculated because 1-2 = 1-3)
Between atoms 1 and 3 (4*3*3=36)
Between atoms 2 and 2 (not calculated because 2-2 = 3-3)
Between atoms 2 and 3 21
Between atoms 3 and 3 18
Between atoms 4 and 1 18
Between atoms 4 and 2 (not calculated because 4-2 = 4-3)
Between atoms 4 and 3 (4*3*3=36)
Between atoms 4 and 4 9

Total = 147 + 1 = 148

Note that due to the Cs symmetry, E(z+dz) = E(z-dz) at z=0

E"(z=0) with respect to z = [E(z+dz) + E(z-dz) - 2E(z)]/dz^2

E"(z=0) = 2[E(z=dz) - E(z=0)]/dz^2

Also, at z=0, E(x+dx,z+dz) = E(x+dx,z-dz);
E(x-dx,z-dz) = E(x-dx,z+dz) due to the Cs symmetry.

Therefore,
E"(z=0) with respect to x and z
= [E(x+dx,z+dz) + E(x-dx,z-dz) - E(x-dx, z+dz) - E(x+dx, z-dz)]/4dxdz = 0

Numerical Hessian

2011-12-19T15:06:00.000-08:00

The total number of energy calculations for a N-Cartesian-coordinate system is 2N^2+1.

The single point energy of the geometry is evaluated first.

Then the coordinates are disturbed in the following ways:

For each diagonal element, disturb only 1 coordinate (x) at a time. Evaluate E(x+dx) and E(x-dx). The 2nd derivative can be derived in the following ways:

E'(x+dx/2)=[E(x+dx)-E(x)]/dx

E'(x-dx/2)=[E(x)-E(x-dx)]/dx

E" = [E'(x+dx/2)-E'(x-dx/2)]/dx

E" = [E(x+dx) + E(x-dx) - 2E(x)]/dx^2

For each off-diagonal elements, disturb 2 coordinates (x and y) at the same time. Evaluate E(x-dx,y-dy), E(x-dx,y+dy), E(x+dx,y-dy), and E(x+dx,y+dy).

The 2nd derivative can be derived in the following way:

E'(x+dx,y) with respect to y = [E(x+dx,y+dy)-E(x+dx,E(y-dy))]/2dy

E'(x-dx,y) with respect to y = [E(x-dx,y+dy)-E(x-dx,E(y-dy))]/2dy

E"(x,y) = {[E(x+dx,y+dy)-E(x+dx,E(y-dy))]/2dy - [E(x-dx,y+dy)-E(x-dx,E(y-dy))]/2dy}/2dx

E"(x,y) = {[E(x+dx,y+dy)-E(x+dx,E(y-dy))] - [E(x-dx,y+dy)-E(x-dx,E(y-dy))]}/4dxdy

E" = [E(x+dx,y+dy) + E(x-dx,y-dy) - E(x-dx, y+dy) - E(x+dx, y-dy)]/4dxdy

Note that E(x,y) is not used in the above off-diagonal element calculation.

In summary, 2N+1 energy calculations need to be done to obtain the N diagonal elements of the energy second derivative matrix (Hessian); 4[N(N-1)/2] energy calculations need to be done to obtain the N(N-1)/2 off-diagonal elements of the Hessian matrix.

2N+1+4[N(N-1)/2]=2N^2+1

For example, a 4-particle system has 12 Cartesian coordinates. 2(12^2)+1=289 energy calculations need to be carried out to obtain the Hessian matrix. A 5-particle system has 15 Cartesian coordinates. 2(15^2)+1=451 energy calculations need to be carried out to obtain the Hessian matrix.

The use of the internal coordinates reduces the number of coordinates by 5 or 6. A linear system has 3n-5 internal coordinates while a non-linear system has 3n-6 internal coordinates, where n is the number of particles.

The number of coordinates can be further reduced if the studied system has a symmetry element.

First snow on 11/11/2011

2011-11-16T12:36:00.000-08:00

2011/11/11

2010/11/9

2009/11/13

2008/11/1-23

IRC on one of multiple imaginary frequencies

2011-07-25T16:40:00.000-07:00

FREQ = The magnitude of the imaginary frequency, given
in cm**-1.
CMODE = An array of the components of the normal mode
whose frequency is imaginary, in Cartesian
coordinates. Be careful with the signs!

You must give FREQ and CMODE if you don't give a $HESS
group, when SADDLE=.TRUE. The option of giving these
two variables instead of a $HESS does not apply to the
GS2 method, which must have a hessian input, even for
restarts.

RK4 may work.

Install Fedora, configure network, and set up GAMESS

2011-06-23T16:40:00.000-07:00

Install Fedora-9-x86_64 on Poweredge1950/2950

4xPoweredge 1950: Xeon L5420 16GB memory 450GB HD
1xPoweredge 2950: Xeon L5240 16GB memory 450GB HD + RAID1

Details on Xeon
http://en.wikipedia.org/wiki/Xeon

Download Fedora-9-x86_64 (ISO file for DVD) from anl.gov mirror site (dead) of Fedora
Where to get Fedora 7-10 at 300-350kB/sec
http://mirror.anl.gov/pub/fedora/linux/releases/
I downloaded Fedora 7, 8, 9 at the same time, each at the speed of 300-350kB/s
http://archives.fedoraproject.org/pub/archive/fedora/linux/releases/9/Fedora/x86_64/iso/

Linux 2.6.25 kernel
Restart computer with DVD in the drive
Install or upgrade an existing system: graphics mode is default
"Loading vmlinuz... "
"Anaconda 11.4.0.82 installer"
Media test if the DVD is used the first time
Installation language: English
Keyboard: English
eth0,1,2,3 active on boot; automatically via DHCP (can be revised later)
Time: US Los Angeles CA; uncheck UTC
Enter root password
Remove all linux partitions (default)
Format disk
Select "software development"; uncheck "Office..."; "customize later"(default)
A total of 1110 or 1111 packages will be installed in ~20 minutes
Note the gnome is the default desktop
gcc with gfortran will be installed
Installation done and DVD will be rejected
Reboot
Welcome --> License --> Create user account --> Current time --> Do not send profile
Done with localhost.localdomain

Log in as root (for convenience)

(the following need to be done only 1 time)
(Open firefox --> log in CWU network --> shutdown --> wait 5 minutes --> reboot)

Installation of softwares:
yum install ncompress
yum install csh
yum install tcsh
The above 2 commands install the same tcsh-6.15-4.fc9.x86_64 package

yum provides libstdc++.so.5 (this is for installation of Intel Fortran Compiler Professional)

yum install compat-libstdc++-33
yum install compat-libstdc++-33.i386
Both of the above packages need to be installed for Intel Fortran compiler

Now increase shmmax and shmall
vi /etc/sysctl.conf

# added by Yingbin Ge for 16 GB of SysV memory
kernel.shmmax = 17179869184
kernel.shmall = 4194304

Linux allows this parameter to be set on the fly, by
/sbin/sysctl -w kernel.shmmax=17179869184
/sbin/sysctl -w kernel.shmall=4194304
to avoid a reboot.

Temporarily change selinux=enforcing to selinux=permissive by
vi /etc/sysconfig/selinux
Change selinux=enforcing after the installation of Intel Fortran Compiler (ifort)

reboot; log in as root

Google "Intel Fortran Compiler" and download l_cprof_p_11.0.74.intel64.tgz from Intel website. It includes ifort, debugger and MKL (math kernel lib)
It is free for non-commercial use. Do not close the page with License number.
download l_cprof_p_11.0.74.intel64.tgz to /root/download

gunzip *gz and tar -xvf *tar
cd l_cprof_p_11.0.74.intel64
./install.sh
follow the instructions to install ifort
set path=(. $path /opt/intel/Compiler/11.0/074/bin/intel64)

log out root and log in as regular user
double check network
csh
make a .cshrc file same as /root/.cshrc
source ~/.cshrc

Obtain /u1/guest/gamess.tar.Z and tinker.tar.Z from si.msg
uncompress *.Z
tar xvf gamess.tar
No need to install tinker unless SIMOMM calculations need to be done
cd gamess
cd ddi
vi compddi
make changes
./compddi
mv ddikick.x ..

cd ../tools
cp actvte.code actvte.f
vi actvte.f
:%s/\*UNX/4SPC/g
:wq
ifort actvte.f -o actvte.x
cd ..

vi comp compall lked
set platform = linux-ia64 to use ifort
change paths
compall >& compall.log
lked

vi rungms runall
change paths to /home/yingbin
mkdir ~/scr
su
mkdir /scr
mkdir /scr/yingbin
chown yingbin !$
chgrp yingbin !$
exit

vi /etc/hosts
make changes to combine ip and hostname
127.0.0.1 hostname will do

rungms exam23 failed in the middle due to convergence problem; all others from 1-43 succeeded.

Configure network:
http://www.yolinux.com/TUTORIALS/LinuxTutorialNetworking.html

Log in as root and run .ssh-keygen on each slave node; the produced id_rsa.pub should be appended to the authorized_keys in the .ssh directory of the head node. This is to allow ssh without passwd as a superuser.
NFS on Fedora 9 (PowerEdge 2950 + 1950)

ON THE HEAD NODE (192.168.2.101 or 192.168.2.102)

1. vi /etc/exports
TYPE the following compute nodes in /etc/exports.

# Allow rw access from compute nodes to server in a synchronized mode.
# Since no_root_squash is put in the parentheses,
# the remote root user will not be treated as a root but as a default nfs user.
/home 192.168.2.111(rw,sync,no_root_squash)
/home 192.168.2.112(rw,sync,no_root_squash)
/home 192.168.2.121(rw,sync,no_root_squash)
/home 192.168.2.122(rw,sync,no_root_squash)
/home 192.168.2.131(rw,sync,no_root_squash)
/home 192.168.2.132(rw,sync,no_root_squash)
/home 192.168.2.141(rw,sync,no_root_squash)
/home 192.168.2.142(rw,sync,no_root_squash)
/home 192.168.2.151(rw,sync,no_root_squash)
/home 192.168.2.152(rw,sync,no_root_squash)

2. /usr/sbin/exportfs -a
This command is to export file system as described in /etc/exports.

3. service nfs restart; service nfslock restart
These commands are to restart nfs and nfslock.

4. chkconfig nfs on; chkconfig nfslock on
These commands are to auto-run nfs and nfslock after reboot.

5. vi /etc/hosts.allow
ALL: .cwu.edu
ALL: 127.0.0.1
This /etc/hosts.allow file is to allow all access from *.cwu.edu
The second line is important to allow loopback to start NFS server.

6. vi /etc/hosts.deny
ALL: ALL
This /etc/hosts.deny file is to deny all access from all IP except from *.cwu.edu

ON EVERY COMPUTE NODE

7. mount -t nfs ge1:/home /home
This command is to mount the ge1:/home on server to a local directory on the compute node.
Then add this following line in /etc/fstab to let the compute node mount master automatically after reboot.
ge1:/home /home nfs hard,intr,rw 0 0

/etc/fstab:
< partition >< mount point >< file system >< mount options >< dump >< fsck order >

/etc/mtab handles the mounted devices and is automatically updated by the mount command. And it looks a bit similar to fstab but not the 100% same ( notice rw and ro for read/write and read only ) And it does only lists the mounted devices !

8. /sbin/chkconfig sendmail off (for security reasons)
/sbin/chkconfig network on (for network on reboot)

9. Set selinux=enforcing on the head node and selinux=disabled on the compute nodes
On the head node: chkconfig iptables on (to turn on the firewall) remember to allow "ssh" using the admin tools on the head node.
On the compute node: chkconfig iptables off

10. to reboot all computers
ssh ge1
su
reboot
ssh ge1
su
ssh na1
reboot
ssh mg1
reboot
ssh al1
reboot
ssh si1
reboot
ssh p1
reboot

How to create a user account (yyyy) to run GAMESS

su
useradd yyyy
passwd yyyy

chsh -s /bin/csh yyyy
chown yyyy /home/yyyy
chgrp yingbin /home/yyyy
mkdir /scr/yyyy
chown yyyy /scr/yyyy
chgrp yingbin /scr/yyyy

ssh yyyy@ge1
cp ~yingbin/.cshrc .
vi .cshrc
ssh-keygen
cd .ssh
cp id_rsa.pub authorized_keys
ssh yyyy@ge1
ln -s ~yingbin/gamess ~yyyy/gamess

ssh yyyy@p1
mkdir /scr/yyyy
chown yyyy /scr/yyyy
chgrp yingbin /scr/yyyy

CFLAG for compddi on TRU64 (OSF1)

2010-05-24T12:07:00.001-07:00

"-D_POSIX_PII_SOCKET -pthread -verbose"

Tru64 UNIX is a 64-bit UNIX operating system for the Alpha instruction set architecture (ISA), currently owned by Hewlett-Packard (HP). Previously, Tru64 UNIX was a product of Compaq, and before that, Digital Equipment Corporation (DEC), where it was known as Digital UNIX (formerly DEC OSF/1 AXP).

As its original name suggests, Tru64 UNIX is based on the OSF/1 operating system. DEC's previous UNIX product was known as Ultrix and was based on BSD UNIX.

-- from WIKI

MACBOOK STUCK ON STARTUP SCREEN

2010-05-03T18:41:00.000-07:00

From: http://support.apple.com/kb/ht1343

Key or key combination What it does
Option Display all bootable volumes (Startup Manager)
Shift Perform Safe Boot (start up in Safe Mode)
C Start from a bootable disc
T Start in FireWire target disk mode
N Start from NetBoot server
X Force Mac OS X startup (if non-Mac OS X startup volumes are present)
Command-V Start in Verbose Mode
Command-S Start in Single User Mode

From http://discussions.apple.com/thread.jspa?threadID=2009503&tstart=0
written by BDAqua

...
Tough without the Tiger Disk problems, but try fsck...

To use fsck, you must run it from the command line. Unlike using your mouse to open an application to do something, you'll need to type a text command at the prompt (#) to tell fsck what to do. The Terminal application (/Applications/Utilities) and single-user mode are two examples of command-line interfaces in which you can type such commands. To use fsck:

1. Start up your computer in single-user mode to reach the command line. Hold CMD+s keys down at bootup. Note: If necessary, perform a forced restart as described in the Emergency Troubleshooting Handbook that came with your computer. On desktop computers, you can do this by pressing the reset/interrupt button (if there is one) or holding down the power button for several seconds. On portable computers, simultaneously press the Command-Control-power keys. If your portable computer doesn't restart with this method, you may need to reset the Power Manager.

2. At the command-line prompt, type /sbin/fsck -fy

(SPACE between fsck AND -fy important)

3. Press Return. fsck will go through five "phases" and then return information about your disk's use and fragmentation. Once it finishes, it'll display this message if no issue is found: 
The volume (name_of_volume) appears to be OK  If fsck found issues and has altered, repaired, or fixed anything, it will display this message: 

*
o
+
#
* FILE SYSTEM WAS MODIFIED *****   Important: If this message appears, repeat the fsck command you typed in step 2 until fsck tells you that your volume appears to be OK (first-pass repairs may uncover additional issues, so this is a normal thing to do).

4. When fsck reports that your volume is OK, type reboot at the prompt and then press Return.

http://docs.info.apple.com/article.html?artnum=106214

Then Safe Boot, (holding Shift key down at bootup), & use Disk Utility from there to Repair Permissions, reboot once more.

MP2 calculation goes wrong without direct SCF

2010-04-23T15:21:00.000-07:00

Linux *.*.*.* 2.6.25-14.fc9.x86_64 #1 SMP Thu May 1 06:06:21 EDT 2008 x86_64 x86_64 x86_64 GNU/Linux
ifort v11
mkl v10
$GMS_COMM = sockets

The following MP2 calculation went wrong if dirscf is turned off.
It does not always happen though.

$contrl maxit=200 mult=1 scftyp=rhf runtyp=optimize coord=unique $end
$contrl mplevl=2 $end
$basis gbasis=n31 ngauss=6 ndfunc=1 $end
$system timlim=1000000 mwords=20 memddi=800 $end
$scf dirscf=.t. $end
$data
no comments
CNV 2

O 8.0 0.0000000000 0.0000000000 0.1360314251
H 1.0 0.7641065796 0.0000000000 0.7319842875
$end

diff test4.inp test5.inp
5c5
< $scf dirscf=.t. $end
---
> $scf dirscf=.f. $end

test4.log: E(MP2)= -76.1968470946
test4.log: MAXIMUM GRADIENT = 0.0004494 RMS GRADIENT = 0.0002158
test4.log: E(MP2)= -76.1968468130
test4.log: MAXIMUM GRADIENT = 0.0006107 RMS GRADIENT = 0.0002556
test4.log: E(MP2)= -76.1968477564
test4.log: MAXIMUM GRADIENT = 0.0001911 RMS GRADIENT = 0.0000941
test4.log: E(MP2)= -76.1968477898
test4.log: MAXIMUM GRADIENT = 0.0000469 RMS GRADIENT = 0.0000289
test5.log: E(MP2)= -76.1968470945
test5.log: MAXIMUM GRADIENT = 0.0004494 RMS GRADIENT = 0.0002158
test5.log: E(MP2)= -126.2224701040
test5.log: MAXIMUM GRADIENT = 1.3400855 RMS GRADIENT = 0.6443406
test5.log: E(MP2)= -126.2208660397
test5.log: MAXIMUM GRADIENT = 1.4001881 RMS GRADIENT = 0.6625852
test5.log: E(MP2)= -126.2703390929
test5.log: MAXIMUM GRADIENT = 1.5046911 RMS GRADIENT = 0.7201698
test5.log: E(MP2)= -126.3443479269
test5.log: MAXIMUM GRADIENT = 1.6611566 RMS GRADIENT = 0.8095194
test5.log: E(MP2)= -126.4231825510
test5.log: MAXIMUM GRADIENT = 1.8277026 RMS GRADIENT = 0.9085332
test5.log: E(MP2)= -126.5069942193
test5.log: MAXIMUM GRADIENT = 1.9991214 RMS GRADIENT = 1.0159385
test5.log: E(MP2)= -126.5957592381
test5.log: MAXIMUM GRADIENT = 2.1666212 RMS GRADIENT = 1.1290437
test5.log: E(MP2)= -2190.7996648700
test5.log: MAXIMUM GRADIENT = ********** RMS GRADIENT =**********

Occam's Razor and the Existence of God

2010-02-13T14:40:00.000-08:00

Occam's Razor: "The simplest explanation is the best." Or
"The simplest explanation that covers all the facts is the best."

One (Schmitt 2005) may argue that the concept of God does not help to explain the universe and thus God is irrelevant and should be cut away.

However, one may also argue that all the universe can be explained as being designed by God and thus Science is irrelevant and should be cut away.

What do you think?

An interesting Nature article: Leading scientists still reject God.

2010-02-03T17:57:00.000-08:00

NATURE| VOL 394 | PAGE 313 | 23 JULY 1998

Leading scientists still reject God
by Edward J. Larson and Larry Witham

A survey was sent in 1998 to all 517 National Academy of Sciences (NAS) members in the core disciplines of Class (i) Physical and Mathematical Sciences and Class (ii) Biological Sciences.
Slightly over 50% responded.

* 7.0% do believe in personal God.
********** 72.2% do not believe in personal God.
*** 20.8% have doubts or adopt agnosticism.

Notes:

Class (I): Physical and Mathematical Sciences
Section 11: Mathematics
Section 12: Astronomy
Section 13: Physics
Section 14: Chemistry
Section 15: Geology
Section 16: Geophysics

Class (II): Biological Sciences
Section 21: Biochemistry
Section 22: Cellular and Developmental Biology
Section 23: Physiology and Pharmacology
Section 24: Cellular and Molecular Neuroscience
Section 25: Plant Biology
Section 26: Genetics
Section 27: Evolutionary Biology
Section 28: Systems Neuroscience
Section 29: Biophysics and Computational Biology

when modem does not recognize router/pc

2010-01-09T12:02:00.000-08:00

connect modem with pc
unplug cable to wall
ipconfig /release *Con*
ipconfig /renew
ipconfig /all
plug cable
ipconfig /renew
ipconfig /all
=========================

USAGE:
ipconfig [/? | /all | /renew [adapter] | /release [adapter] |
/flushdns | /displaydns | /registerdns |
/showclassid adapter |
/setclassid adapter [classid] ]

where
adapter Connection name
(wildcard characters * and ? allowed, see examples)

Options:
/? Display this help message
/all Display full configuration information.
/release Release the IP address for the specified adapter.
/renew Renew the IP address for the specified adapter.
/flushdns Purges the DNS Resolver cache.
/registerdns Refreshes all DHCP leases and re-registers DNS names
/displaydns Display the contents of the DNS Resolver Cache.
/showclassid Displays all the dhcp class IDs allowed for adapter.
/setclassid Modifies the dhcp class id.

The default is to display only the IP address, subnet mask and
default gateway for each adapter bound to TCP/IP.

For Release and Renew, if no adapter name is specified, then the IP address
leases for all adapters bound to TCP/IP will be released or renewed.

For Setclassid, if no ClassId is specified, then the ClassId is removed.

Examples:
> ipconfig ... Show information.
> ipconfig /all ... Show detailed information
> ipconfig /renew ... renew all adapters
> ipconfig /renew EL* ... renew any connection that has its
name starting with EL
> ipconfig /release *Con* ... release all matching connections,
eg. "Local Area Connection 1" or
"Local Area Connection 2"

How to change E-23 to 10^(-23) in word

2009-07-24T11:36:00.000-07:00

Find and Replace

Find: -[0-9][0-9] (select wild card)
Replace with: (leave here blank and selece superscript format)

Replace all.

DFT convergence problem

2009-07-21T12:27:00.001-07:00

may be solved by changing the switch threshold:

$dft swoff=2.0e-2 switch=2.0e-5 $end
$scf soscf=.t. $end

How to clean up semaphores.

2009-04-22T22:39:00.000-07:00

ipcs -s | grep yingbin | cat > rm_sem
sed "/0x00000000/s//ipcrm -s/" rm_sem > rm_sem.csh
chmod 755 rm_sem.csh
rm_sem.csh
/bin/rm rm_sem.csh rm_sem

Quantum Chemistry Videos, Applets, and Images

2009-01-09T17:54:00.000-08:00

1/9/2009
http://www.youtube.com/watch?v=qzZDlXji0e0
(Quantum Mechanics for Dummies - Electrons Are Weird)
http://www.youtube.com/watch?v=DfPeprQ7oGc
(Dr Quantum - Double Slit Experiment)
http://www.youtube.com/watch?v=EN-jCuV7BoU
(Schrodinger's Cat for real)

1/12/2009
http://www.lon-capa.org/~mmp/kap28/PhotoEffect/photo.htm
http://lectureonline.cl.msu.edu/~mmp/kap28/PhotoEffect/photo.htm
(Applet: Photo Effect)

http://www.physics.uiowa.edu/~umallik/adventure/quantumwave.html
(Quantum Waves)

1/14/2009
http://www.stanford.edu/~ccecka/research/QuantumApplet.html
Applet (Time Evolution of the Schrodinger Equation)

required packages for loadleveler

2008-12-30T12:28:00.000-08:00

[yingbin@ge ~]$ yum info autoconf >> loadleveler.required
[yingbin@ge ~]$ yum info automake >> loadleveler.required
[yingbin@ge ~]$ yum info make >> loadleveler.required
[yingbin@ge ~]$ yum info gcc >> loadleveler.required
[yingbin@ge ~]$ yum info libtool >> loadleveler.required
[yingbin@ge ~]$ yum info rpm >> loadleveler.required
[yingbin@ge ~]$ cat !$
cat loadleveler.required
Loaded plugins: refresh-packagekit
Installed Packages
Name : autoconf
Arch : noarch
Version : 2.61
Release : 10.fc9
Size : 2.1 M
Repo : installed
Summary : A GNU tool for automatically configuring source code
URL : http://www.gnu.org/software/autoconf/
License : GPLv2+ and GFDL
Description: GNU's Autoconf is a tool for configuring source code and Makefiles.
: Using Autoconf, programmers can create portable and configurable
: packages, since the person building the package is allowed to
: specify various configuration options. You should install Autoconf
: if you are developing software and would like to create shell
: scripts that configure your source code packages. If you are
: installing Autoconf, you will also need to install the GNU m4
: package. Note that the Autoconf package is not required for the
: end-user who may be configuring software with an Autoconf-generated
: script; Autoconf is only required for the generation of the
: scripts, not their use.

Loaded plugins: refresh-packagekit
Installed Packages
Name : automake
Arch : noarch
Version : 1.10.1
Release : 2
Size : 1.3 M
Repo : installed
Summary : A GNU tool for automatically creating Makefiles
URL : http://sources.redhat.com/automake
License : GPLv2+ and GFDL and MIT
Description: Automake is a tool for automatically generating `Makefile.in' files
: compliant with the GNU Coding Standards. You should install
: Automake if you are developing software and would like to use its
: ability to automatically generate GNU standard Makefiles. If you
: install Automake, you will also need to install GNU's Autoconf
: package.

Loaded plugins: refresh-packagekit
Installed Packages
Name : make
Arch : x86_64
Epoch : 1
Version : 3.81
Release : 12.fc9
Size : 1.0 M
Repo : installed
Summary : A GNU tool which simplifies the build process for users
URL : http://www.gnu.org/software/make/
License : GPLv2+
Description: A GNU tool for controlling the generation of executables and other
: non-source files of a program from the program's source files. Make
: allows users to build and install packages without any significant
: knowledge about the details of the build process. The details about
: how the program should be built are provided for make in the
: program's makefile.

Loaded plugins: refresh-packagekit
Installed Packages
Name : gcc
Arch : x86_64
Version : 4.3.0
Release : 8
Size : 17 M
Repo : installed
Summary : Various compilers (C, C++, Objective-C, Java, ...)
URL : http://gcc.gnu.org
License : GPLv3+ and GPLv2+ with exceptions
Description: The gcc package contains the GNU Compiler Collection version 4.3.
: You'll need this package in order to compile C code.

Loaded plugins: refresh-packagekit
Installed Packages
Name : libtool
Arch : x86_64
Version : 1.5.24
Release : 6.fc9
Size : 2.5 M
Repo : installed
Summary : The GNU Portable Library Tool
URL : http://www.gnu.org/software/libtool/
License : GPLv2+ and LGPLv2+ and GFDL
Description: GNU Libtool is a set of shell scripts which automatically configure
: UNIX and UNIX-like systems to generically build shared libraries.
: Libtool provides a consistent, portable interface which simplifies
: the process of using shared libraries. If you are developing
: programs which will use shared libraries, but do not use the rest
: of the GNU Autotools (such as GNU Autoconf and GNU Automake), you
: should install the libtool package. The libtool package also
: includes all files needed to integrate the GNU Portable Library
: Tool (libtool) and the GNU Libtool Dynamic Module Loader (ltdl)
: into a package built using the GNU Autotools (including GNU
: Autoconf and GNU Automake). This package includes a modification
: from the original GNU Libtool to allow support for multi-
: architecture systems, such as the AMD64 Opteron and the Intel
: 64-bit Xeon.

Loaded plugins: refresh-packagekit
Installed Packages
Name : rpm
Arch : x86_64
Version : 4.4.2.3
Release : 2.fc9
Size : 3.6 M
Repo : installed
Summary : The RPM package management system
URL : http://www.rpm.org/
License : GPLv2+
Description: The RPM Package Manager (RPM) is a powerful command line driven
: package management system capable of installing, uninstalling,
: verifying, querying, and updating software packages. Each software
: package consists of an archive of files along with information
: about the package like its version, a description, etc.

antivirus for fedora 9 64bit

2008-12-30T11:53:00.001-08:00

clamav.x86_64 0:0.93.3-2.fc9
yum install clamav
for KDE yum install klamav

burnaware free CD/DVD burner

2008-12-18T10:19:00.000-08:00

http://www.burnaware.com/burnaware_free.html

it works fine: 5'15" for 3.9G Fedora-10-x86_64 DVD ISO file without verification

/boot/grub/menu.lst

2008-12-15T11:09:00.000-08:00

/boot/grub/menu.lst

# grub.conf generated by anaconda
#
# Note that you do not have to rerun grub after making changes to this file
# NOTICE: You have a /boot partition. This means that
# all kernel and initrd paths are relative to /boot/, eg.
# root (hd0,2)
# kernel /vmlinuz-version ro root=/dev/VolGroup00/LogVol00
# initrd /initrd-version.img
#boot=/dev/mapper/isw_dhdjhahafi_ARRAY
default=0
timeout=5
splashimage=(hd0,2)/grub/splash.xpm.gz
hiddenmenu
title CentOS (2.6.18-92.el5PAE)
root (hd0,2)
kernel /vmlinuz-2.6.18-92.el5PAE ro root=/dev/VolGroup00/LogVol00 rhgb quiet
initrd /initrd-2.6.18-92.el5PAE.img
title Other
rootnoverify (hd0,1)
chainloader +1
~

obtain gcc g77 and gfortran for CentOS

2008-11-04T10:27:00.000-08:00

yum groupinstall "Development Tools"
*be sure to include the Quotes.

Also from the Add/Remove Software menu you can Search for g77 and the click on the
compat-gcc-34-g77 - 3.4.6-4.i386 - Fortran 77 support for compatibility compiler

Installing MPICH on CentOS

2008-11-04T10:16:00.000-08:00

http://www-unix.mcs.anl.gov/mpi/mpich1/
http://www-unix.mcs.anl.gov/mpi/index.htm

I downloaded mpich.tar.gz from http://www-unix.mcs.anl.gov/mpi/mpich1/download.html to my home directory.

gunzip mpich.tar.gz
tar xvf mpich.tar
cd mpich-1.2.7p1
configure
make

so far so good.

But then I go to examples/test and do
make testing
I got

$ make testing
if [ "" != yes ] ; then \
(cd pt2pt && ./runtests -check ) ; fi
/home/yingbin/Desktop/mpich-1.2.7p1/bin/mpicc -DUSE_SOCKLEN_T -DUSE_U_INT_FOR_XDR -DHAVE_MPICHCONF_H -DHAVE_STDLIB_H=1 -DHAVE_UNISTD_H=1 -DHAVE_STRING_H=1 -DUSE_STDARG=1 -DHAVE_LONG_DOUBLE=1 -DHAVE_LONG_LONG_INT=1 -DHAVE_PROTOTYPES=1 -DHAVE_SIGNAL_H=1 -DHAVE_SIGACTION=1 -DHAVE_SLEEP=1 -DHAVE_SYSCONF=1 -c third.c
/home/yingbin/Desktop/mpich-1.2.7p1/bin/mpicc -DUSE_SOCKLEN_T -DUSE_U_INT_FOR_XDR -DHAVE_MPICHCONF_H -DHAVE_STDLIB_H=1 -DHAVE_UNISTD_H=1 -DHAVE_STRING_H=1 -DUSE_STDARG=1 -DHAVE_LONG_DOUBLE=1 -DHAVE_LONG_LONG_INT=1 -DHAVE_PROTOTYPES=1 -DHAVE_SIGNAL_H=1 -DHAVE_SIGACTION=1 -DHAVE_SLEEP=1 -DHAVE_SYSCONF=1 -c test.c
/home/yingbin/Desktop/mpich-1.2.7p1/bin/mpicc -o third third.o test.o
Failed to run simple program!
Output from run attempt was
*** Testing Unexpected messages ***
p0_28712: p4_error: Could not gethostbyname for host xxxxx; may be invalid name
: 61
*** Testing Unexpected messages ***
mpirun program was /home/yingbin/Desktop/mpich-1.2.7p1/bin/mpirun
mpirun command was
/home/yingbin/Desktop/mpich-1.2.7p1/bin/mpirun -mvhome -np 2 ./third >third.out 2>&1
make: *** [runtest] Error 1

The possible reason is that my host name pc##### is not associated with an IP?
So I added one line in /etc/hosts
127.0.0.1 pc##### pc#####

And then I make sure that no password is need for pc##### connecting itself thru ssh

The following is copied from http://linuxproblem.org/art_9.html in case it's deleted:

First log in on A as user a and generate a pair of authentication keys. Do not enter a passphrase:

a@A:~> ssh-keygen -t rsa
Generating public/private rsa key pair.
Enter file in which to save the key (/home/a/.ssh/id_rsa):
Created directory '/home/a/.ssh'.
Enter passphrase (empty for no passphrase):
Enter same passphrase again:
Your identification has been saved in /home/a/.ssh/id_rsa.
Your public key has been saved in /home/a/.ssh/id_rsa.pub.
The key fingerprint is:
3e:4f:05:79:3a:9f:96:7c:3b:ad:e9:58:37:bc:37:e4 a@A

Now use ssh to create a directory ~/.ssh as user b on B. (The directory may already exist, which is fine):

a@A:~> ssh b@B mkdir -p .ssh
b@B's password:

Finally append a's new public key to b@B:.ssh/authorized_keys and enter b's password one last time:

a@A:~> cat .ssh/id_rsa.pub | ssh b@B 'cat >> .ssh/authorized_keys'
b@B's password:

From now on you can log into B as b from A as a without password:

a@A:~> ssh b@B hostname
B

After doing the above,
cd examples/test
make testing

every single MPI function will be tested and no error is found!
Now we can use the 4 cores to run parallel calculations.

how to remove grub loader

2008-11-02T14:33:00.000-08:00

C:\WINDOWS
CD ..
FIXBOOT C:
FIXMBR
BOOTCFG /rebuild

How to add Linux (CentOS 5.2) to Windows PC

2008-10-29T14:58:00.000-07:00

1. BACKUP YOUR FILE!

2. Download EASEUS Partition Manager Home Edition (free) from http://www.partition-tool.com/

3. Install EASEUS

4. Use EASEUS to partition the hard drive. I spitted my hard drive in to halves: ~115G for XP, ~115 become unpartitioned.

5. You will be prompted to restart the computer after applying the new partition. You'll see the new disk sizes.

6. You might be asked to restart again. Do that.

7. Insert CentOS (5.2) CD/DVD and restart the computer.

8. Select the option "Use free space on selected drives and create default layout". It depends on where u like to put the CentOS. I have only 1 disk anyway.

9. eth0 or eth1 should be selected as" active on boot" ( it depends on which NIC is connected to the network, it can be changed later anyway -- so no worries)
DHCP for IPv4 and Auto for IPv6.
Automatically via DHCP (it works for my office computer and laptop, both of which use DHCP)

10. Select Gnome desktop or KDE desktop (I have no preference) and select ssh only for communication (for security reasons).

11. when restart computer and see "press enter to show a menu of OS), hit "enter" to get a menu of "CentOS and other OS". The other OS is Windows XP for me (you can change "other" to "XP" in the /boot/grub/menu.lst file.

12. You may change the default OS by changing the default value from 0 to 1 or whatever number that is proper in the /boot/grub/grub.conf file.

13. In case you get into the grub command line during restart, and you'd like to have graphical interface, the easiest way to do is to type "boot" and hit "enter" and then "esc". Then you will see a list of OS.

14. More tedious way is to type the commands under grub. To get windows, you need to type 3 commands like:
rootnoverify (hd0,1)
chainloader +1
boot

to get Linux (in my case CentOS), you need something like:
root (hd0,2)
kernel /vmlinuz-2.6.18-92.el5PAE ro root=/dev/VolGroup00/LogVol00 rhgb quiet
initrd /initrd-2.6.18-92.el5PAE.img
boot

connect a macbook (wireless) to cwu network

2008-08-20T20:31:00.000-07:00

http://vuksan.com/linux/dot1x/wpa-client-config.html#macosx

The following is from the above website in case this webpage is deleted.

Mac OS X

This is how you would configure a Mac OS X 10.3.x client to use TTLS+PAP to authenticate against WPA.

1. Click on the wireless icon.

2. Select "Open Internet Connect".

3. Click File then "New 802.1X Connection"

4. Click on 802.1X icon. Enter your username/password and the Wireless Network you want to connect to.

4. Click on Configurations then Edit Configurations. You will be prompted to save the configuration. Name it ie. CS 802.11. You will then get a window as follows

5. Check off everything other than TTLS.

6. Click on TTLS and click on Configure

7. Select PAP as TTLS Inner Authentication ie.

8. Click OK then OK again. Try to connect.

9. During authentication you will likely be prompted to accept certificate for the server.

10. Click Accept All

11. That should be it.

Windows XP SecureW2 setup
Get the Windows Version of SecureW2 tool from the url given below.
http://www.securew2.com/uk/download/index.htm
Its a zip file. Unzip and run the installation file. You will be prompted to reboot. After reboot do following.
Right Click on the Wireless Connection Icon in the taskbar and select “View available Wireless Networks”
Click on “Change Advanced Settings” in the Left Side bar.

Click on Wireless Networks tab and Select your WPA network ie. CSWIRELESS-WPA and click on Properties.
Go to Authentication tab
Under Eap type select “SecureW2”
Then Click on Properties
SecureW2 window will show up.

Click on 'Configure' for DEFAULT profile. Go to Certificates Tab and check off "Verify server certificate"

Next click on Authentication tab and make sure that PAP is the select authentication method ie.

Next select "User account" tab and fill out the appropriate info or check off "Prompt user for credentials".

Click OK and then try to connect to

network name "cwu" use ttls + pap

How to install and run pcgamess (binary download)

2008-08-20T20:22:00.000-07:00

1. http://classic.chem.msu.su/gran/gamess/index.html to download a copy

2. send an email to GAMESS@AMESLAB to get a password to extract a zip file

3. to install, send another email to Granovsky@MSU for another password for installation

4. All files will be in the C:\WINDOWS\system32 (for a WINDOWS XP 32 bit system)

5. To run pcgamess.exe, use My Computer -> Advanced -> Environment variables to set path, or use setx in command line

6. write a script file to delete PUNCH before running GAMESS and use a for-do loop to run all exam input files and benchmarks

4500 km, from Ames IA to Ellensburg WA

2008-08-03T08:09:00.000-07:00

We actually drove ~4500 km (~2800 miles) from Ames IA to Ellensburg WA

The following is schedule by using Google Map (not very accurate):

Total Mileage: 2277 Miles

Ames, IA
--306--> Sioux Falls, SD (passing MN)
--400--> Rapid city, SD (356 + 22*2: passing Badlands and Mount Rushmore)
--525--> Yellowstone, WY
--381--> Glacier, MT
--204--> Kellogg, ID
--241--> Ellensburg, WA
--110--> Seattle, WA --110--> Ellensburg, WA

Pass by: Iowa, Minnesota, South Dakota, Wyoming, Montana, Idaho, Washington

Total Drive time: 33'31" (2011")
4'32" + [5'31" + 30"x2] + 9'05" + 6'19" + 3'27" + 3'37"

7/26(Sat) 16:00 Take off from Ames, IA

306 Miles, Drive 4'32"

7/26(Sat) 20:45 Arrive Sioux Falls, SD
Country Inn ($82 + tax with AAA)

7/27(Sun) 08:30 Take off from Sioux Falls, SD

356+22*2 Miles, Drive 5'31" (pass Badlands) to Rapid City + 30"*2 to
and from Mount Rushmore

Stay at Super 8 ($72 + tax with AAA)

7/28(Mon) 08:30 Take off from Rapid City, SD

525 Miles, Drive 9'05"

7/28(Mon) 19:30 Arrive Yellowstone National Park

7/29-30(Tue-Weds) Fun at Yellowstone

Stay at City Center Motel ($71.1 + tax with AAA)

7/31(Thur) 11:00 Take off from Yellowstone

381 Miles, Drive 6'19"

7/31(Thur) 20:30 Arrive Hungry Horse, MT, close to Glacier National Park, MT

8/01 (Fri) Fun at Glacier National Park, MT

8/01 (Fri) 18:00 Take off from Glacier National Park, MT (after dinner)

204 Miles, Drive 3'27"

8/01 (Fri) 21:45 Arrive Kellogg, ID

8/01 (Fri) 22:00 Stay in Silverhorn Motel Inn ($87 + tax with AAA)

8/02 (Sat) 08:30 Take off from Kellogg, ID

241 Miles, Drive 3'37"

8/02 (Sat) 12:30 Arrive Ellensburg, WA

8/02 (Sat) 14:30 To and from Seattle, WA

8/02 (Sat) 21:30 Arrive Ellensburg, WA (final stop of the trip)

MP2/cc-pVDZ reaction list (change)

2007-11-29T15:40:00.000-08:00

Original (41 + 73TS)

CH3SiCl3 -> CH3SiCl2Cl removed
CH3SiCl2Cl -> CH3SiCl + Cl2 removed
CH3SiCl3 -> CH3SiCl + Cl2 kept with a TS of CH3SiCl2Cl -> CH3SiCl + Cl2
(40 + 72TS)

C2H2 + H2 -> C2H3 + H has a TS
C2H4 + H2 -> C2H5 + H has a TS
SiHCl + HCl -> SiH2Cl + Cl has a TS
SiCl2 + HCl -> SiHCl2 + Cl has a TS
C2H4 + C2H4 -> C2H3 + C2H5 has a TS
(35 + 77TS)

MTS->CHSiCl3+H2; MTS->CH2+SiHCl3

2007-11-20T15:07:00.000-08:00

1--A'
2--A''

MTS->CHSiCl3+H2 (Cs forbidden)
H2: 0 core 2 val e-
1
CHSiCl3: 21 core 30 val e-
1211
11212
11212
1
CH3SiCl3: 21 core 32 val e-
1121
12111
22112
12

MTS->CH2+SiHCl3 (Cs allowed)
CH2: 1 core 6 val e-
121
SiHCl3: 20 core 26 val e-
12111
21122
112
CH3SiCl3: 21 core 32 val e-
1121
12111
22112
12

Useful but hard-to-remember tips

2007-05-25T07:54:00.000-07:00

1. To keep the ssh connection alive:
more ~/.ssh/config
Host si.fi.ameslab.gov
ServerAliveInterval 240

2. To remove the ^M characters at the end of all lines in vi, use:
:%s/^V^M//g

3. CCL calc:Larger LMICRO (8 instead of the default 6) sometimes help convergence of solving the left eigenstate in ROHF-referenced CR-CCL calculations.

4. Full CI calculations with core electrons dropped often has problem with keeping core electrons out of active space at long bond distance when dealing with bond-breaking reactions.

5. CASSCF PARALLEL FULLNR RUN MUST HAVE VIRTUAL ORBITALS.
THE JACOBI CONVERGER DOES NOT RUN IN PARALLEL
-FCORE,MOFRZ,NOROT- MUST BE USED WITH
ONLY THE -FULLNR- OR -JACOBI- CONVERGERS.

6. g94: basis set reading error on HfTa but not on Ti

6.ipcs | grep [username] | awk '{printf("ipcrm -s %s\n", $2);}' to remove semaphores

7.:1,100s/^/blah/ can be used to add "blah" to the beginning of the first 100 lines.

Defects in SiC

2007-05-23T09:40:00.000-07:00

Voids: missing Si, C, Si-C
Impurity: Si crystal, C diamond, PyroC
Ring strain: 4,5,7,8,9,10,11,12 member-rings, individual Si-Si or C-C bonds
Dislocation: a displacement of part of a crystal lattice structure

Affecting rate constants:
Types of defects.
Distance: on site or 1,2,3,4 bonds away

Chemical Vapor Deposition Modeling

2007-05-16T10:29:00.000-07:00

A few thoughts:

1). Many CVD modeling oversimplify the chemistry.
E. g. Lamborn et al. J. Crystal Growth 299, p358 (2007) paper studied CVD of boron.
They have 2 gas phase reactions and 2 surface reactions.
S-1: BH3(g) + * -> BH2* + 1/2 H2(g)
S-2: BH2(*) -> B(s) + * + H2(g)

2). Sticking coefficients are widely and improperly used!
E. g. CH3 deposition on amorphous C. The rate-determining step could be production of active sites or association between CH3 and the active site. In the former case, the "sticking" coefficient corresponds to the rate constant of the H abstraction reation.

3). As to SiC CVD, we need to find a way to model surface species and reactions:
E. g.

(s)Si* + SiCl3 -> Si(Si)) + (s)SiCl3
(s)Si* + CH3 -> Si(SiC) + (s)CH3
(s)Si* + SiCl2 -> Si(Si)) + (s)Si*
(s)Si* + CH2 -> Si(SiC) + (s)C*

(s)C* + SiCl3 -> C(SiC) + (s)SiCl3
(s)C* + CH3 -> C(C) + (s)CH3
(s)C* + SiCl2 -> C(SiC) + (s)Si*
(s)C* + CH2 -> C(C) + (s)C*

(s)SiCl3 + H -> (s)Si* + HCl
(s)CH3 + H -> (s)C* + H2

4. Compare rate constants of gas phase and surface reactions.
M verse myu?

CH3SiCl2+Cl.s vib. freq. evolving from real to imaginary

2007-03-20T09:29:00.000-07:00

CH3SiCl2+Cl.s.390 real 97.28 cm-1
CH3SiCl2+Cl.s.395 real 54.47 cm-1
CH3SiCl2+Cl.s.396 real 38.83 cm-1
CH3SiCl2+Cl.s.397 imag 22.20i cm -1
CH3SiCl2+Cl.s.398 imag 48.42i cm -1
CH3SiCl2+Cl.s.399 imag 65.65i cm -1
CH3SiCl2+Cl.s.400 imag 81.76i cm -1
CH3SiCl2+Cl.s.405 imag 138.24i cm -1

Methyl rotation

Writing a Scientific Paper Copyright 2006, Michael K. gilson

2007-03-19T14:51:00.000-07:00

http://gilsonlab.umbi.umd.edu/writing2a.htm

Writing a Scientific Paper
Copyright 2006, Michael K. gilson

kinetic Monte Carlo simulation of SiC CVD

2007-03-01T07:16:00.000-08:00

a). KMC has been used on far less complicated system than SiC CVD. There are too many possible events that need to be considered in SiC CVD. It may take years to develop the code for modeling all significant events in SiC CVD, and the results may still not agree well experiments. The reasons are:
b). A slight orientation change may dramatically affect growth rate and morphology of the surface. Thus not only (111) orientation but others need to be studied.
c). Steps are important: surface reactions prefer to take place on steps than on terraces. However the density of steps need to be determined by experiments -- hard to predict theoretically.
d). Defects such as dislocation and voids need to be modeled. Surface relaxation and reconstruction need to be considered.
e). Diffusion would not be as important in SiC CVD as in metallic systems that has been studied. The Si-C, C-H, Si-Cl covalent bonds have much higher bond energy than metallic bonds, hence the diffusion reactions on the SiC surface would be slow due to the hardness to break a covalent bond.
f). Even missing one significant event or predicting one rate constant inaccurately may lead to wrong simulation results.

In a review paper by Battaile from Sandia Lab and Srolovitz from Princeton Univ, they described their kMC for the diamond CVD. Their results showed that the growth rates predicted by simulations agreed with experiments when the temperature is less than 1200 K, however they differ dramatically at higher temperatures. They also mentioned aforementioed drawbacks and limitations .

However, I think, although I was often wrong, that kMC is the most suitable method for studying the CVD of SiC because of its efficiency (compared to MD) and its ability of studying morphology as well as growth rates of the SiC surface.

Tinker's updates in SIMOMM

2007-02-26T18:42:00.000-08:00

1. There are static dimensions for MM parameters. Eg the maximum number of torsion5 parameters is 300 originally. However, user provided mm3.prm may easily go over the upper bound. Since tinker does not check the actual number of parameters, it will keep reading all parameters in mm3.prm by allowing extra parameters to overwrite out of bound, which causes unpredictable results. In GAMESS version after Nov 15, 2007, dimension check for parameters are performed in the Libtpr.f and the program dies if too many parameters are provided. Also maxnt5 for torsion5 parameters was changed to 600.

2. Tinkers has maxval=4 in sizes.i, which does not allow 5 or more bonded atoms in input file.
Easy fix is to change maxval=5 or 6 in sizes.i. However the string length of the MM coordinates will exceed the maximum string length. Therefore, to run a input file with 5-bonded atoms, the number of blanks in the line of the 5-bonded atoms need to be minimized to make sure READXYZ actually read in the parameters. Or you could increase the upper bound of string length. Unfortunately, the SIMOMM interface use GAMESS maxcol=80, hence it's simpler to shrink the MM coordinate lines than to change MAXCOL in both GAMESS and Tinker.
In Libtpr.f, line 3141 need to be changed accordingly so that the Tinker coordinates in an output file can be used directly (by copy&paste) in an input file.
-----------------------
c 30 format (i6,2x,a3,3f12.6,9i6)
30 format (1x,i4,2x,a3,3f12.6,9(1x,i4))
-----------------------
The reason is GAMESS only read in the coordinates within the first 80 columns, while an atom with 5 bonds will produce a coordinate line with more than 80 characters. The above change is to shrink the Tinker coordinate line so that the whole line will be read properly.

12/30/2006 - 1/1/2007 in Chicago

2007-01-02T07:34:00.000-08:00

12/31/2007: Art Institute of Chicago (> 6 hours)

1/1/2007:
Tut-ankh-amun (2.5 hours): http://en.wikipedia.org/wiki/Tutankhamun
ankh: life
amun: Amun (also spelled Amon, Amoun, Amen, and rarely Imen, Greek Αμμον Ammon, and Άμμον Hammon, Egyptian a-m-n) was the name of a deity, in Egyptian mythology, who gradually rose to become one of the most important deities, before fading into obscurity.
http://en.wikipedia.org/wiki/Amun

Field Museum (3 hours)

Driving:
A). Ames - Factory Store at Iowa City (117 miles)
B). Ames - Factory Store at Aurora, IL (330 miles)
C). Many rest area on I80, few on I88
D). Asian Bistro /(Ya Xuan): A Sze Chwan style restaurant at Golf Road/Higgins Road (Des Plaines, IL)

k_B T / h , k_B T / p (cm^3/particle) and (k_BT)^2/hp

2006-11-28T13:46:00.000-08:00

Temp k_BT/h (/s)
0298.15 6.212443e+12
0400.00 8.334655e+12
0600.00 1.250198e+13
0800.00 1.666931e+13
1000.00 2.083664e+13
1200.00 2.500396e+13
1400.00 2.917129e+13
1600.00 3.333862e+13
1800.00 3.750595e+13
2000.00 4.167327e+13

Temp k_BT/p(cm^3/particle)
0298.15 4.062579e-20
0400.00 5.450382e-20
0600.00 8.175574e-20
0800.00 1.090076e-19
1000.00 1.362596e-19
1200.00 1.635115e-19
1400.00 1.907634e-19
1600.00 2.180153e-19
1800.00 2.452672e-19
2000.00 2.725191e-19

Temp (k_BT)^2/hp (cm3/particle s)
0298.15 2.523854e-07
0400.00 4.542706e-07
0600.00 1.022109e-06
0800.00 1.817082e-06
1000.00 2.839191e-06
1200.00 4.088435e-06
1400.00 5.564814e-06
1600.00 7.268329e-06
1800.00 9.198979e-06
2000.00 1.135676e-05

#define R 8.314472
#define CAL2J 4.184
#define BOLTZMANN 1.380650e-23
#define PLANCK 6.6260693e-34
#define ATM 101325.0

CH3SiCl3 -> CH2 + SiHCl3

2006-07-25T08:32:00.000-07:00

A'' count: CH3SiCl3 (13) -> CH2(1) + SiHCl3(12) MATCH: Cs symmetry allowed!

CH3SiCl3 (74 e-): 12111 21112 11212 12121 11121 11212 11212 12
1/4 (1s of Si and Cl)
7/21 (1s of C, 2s and 2p of Si and Cl)
13/37 (valence)

CH2 (8 e-): 1121
0/0
0/1 (1s of C)
1/4 (valence)

SiHCl3 (66 e-): 21112 11121 12121 21121 11211 21212 112
1/4(1s of Si and Cl)
7/20 (1s of C, 2s and 2p of Si and Cl)
12/33 (valence)

An interesting blog

2006-04-09T18:22:00.000-07:00

http://spaces.msn.com/stevenwang1978/

Why does pyroC shrink?

2006-04-04T08:24:00.000-07:00

Possible reasons:

1. Nuclear energy enables the pyro C (density 1.8-1.9 g/cm^3) to overcome energy barrier to reduce defects to form more perfect graphite which has higher density (2.2 g/cm^3).

2. Fission products may react with neighbor graphenes to link them together

Some thoughts on modeling graphite or pyrolytic C

2006-03-28T17:42:00.000-08:00

1. QM/MM such as SIMOMM uses MM for the bulk. However does it describe the dispersion force between sheets correctly?

2. The C-C of C6H6 (benzene) is about 1.5 bond order while the C-C of a graphene is about 4/3 bond order, which suggests that QM region must be large enough, such as C24H12 or C54H18 to make C-C nearly 4/3 order.

3. Due to 2, a rather small (at most medium-sized) basis set should be used.
Eg. For C24H12, 6-31G -> 240 basis functions; 6-31G* -> 360 functions; 6-31G** or cc-pVDZ -> 396 functions.
For C54H18 even 6-31G -> 522 basis functions; 6-31G* -> 792 functions; 6-31G** or cc-pVDZ -> 846 functions.
NO WAY TO MODEL 2-3 LAYERS FOR DESCRIBING DISPERSION FORCE !!!

4. Due to 2, DFT methods should be considered. How about B3LYP, or BLYP or BPW91?
BPW91 better describes benzene-benzene interaction (1-2 kcal/mol) than B3LYP or BLYP.

5. How about triple-level ONIOM?

Notes on ccsd(T) in Gaussian

2006-03-24T16:52:00.000-08:00

1. uccsd(t) and ccsd(t) keywords are same for open shell system in Gaussian 94
2. e4t only gives an extra MP4(SDTQ) energy.
uccsd(t)/aug-cc-pvdz and uccsd(t,e4t)/aug-cc-pvdz give same uccsd(t) energy. uccsd(t)/aug-cc-pvdz does not give MP4(SDTQ) energy but uccsd(t,e4t)/aug-cc-pvdz does.

Fillipino and Thai

2006-03-09T17:52:00.000-08:00

magandang umaga
magandang hapon
magandang gabi
hindi ako marunong mag-Tagalog (I don's speak Filipino)

sawadi
put-thai-mai-dai
Chan-put-Thai-mai-dai (I don's speak Thai)
Khop Khun (Cap Coon) thank you
Khun/Chan mai roo a-rai (You/I don't know anything)
Roo mai? (understand?)

SIMOMM for graphene sheets?

2006-03-07T16:50:00.000-08:00

1. Each C-C bond has a 4/3 bond order on a graphene sheet. (3 sigma and 1 pi distributed to 3 C-C bonds)

2. If H are used as cap atoms then we are trying to use C-H single bond to model C-C 4/3 bond.

3 Possible solutions:

a. H as cap atoms
b. Benzene as caps
c. FMMM
d. buffer between QM and MM treated with frozen LMO

MOLPRO

2006-02-09T09:38:00.000-08:00

1. MOLPRO use standard (old) aug-cc-pvdz, which is consistent with the PNNL basis set database.

2. MOLPRO does not treat open shell system right by giving too high UHF energies

package comparison:

CH2.s rhf/uhf mp2/ump2 ccsd(t)
GAMESS -38.8841291 -38.9999600 -39.0315069
Gaussian -38.8841291 -38.9999600 -39.0315093
Molpro(ext b) -38.8841291 -38.9999600 -39.0315069
Molpro(std b) -38.8841291 -38.9999600 -39.0315069
Aces(dropmo) -38.8841291 -38.9999600 -39.0315069
Aces(core) -38.8841291 -39.0021547 -39.0335302
GAMESS(core) -38.8841291 -39.0021547 -39.0335302

CH2.t uhf ump2 ccsd(t)
GAMESS -38.9281162 -39.0252705 N.A.
Gaussian -38.9281162 -39.0252705 -39.0482933
Molpro(ext b) -38.8668580 -38.9812586 -39.0126656
Molpro(std b) -38.8668580 -38.9812586 -39.0126656
Aces(dropmo) -38.9281162 -39.0252705 -39.0482933
Aces(core) -38.9281162 -39.0274627 -39.0503947
GAMESS(core) -38.9281162 -39.0274627 N.A.
Gaussian(core) -38.9281162 -39.0274627 -39.0503947

SiCl2.s uhf ump2 ccsd(t)
GAMESS -1208.0149322 -1208.3875167 -1208.4342190
Gaussian -1208.0149322 -1208.3875167 -1208.4342190
Molpro(ext b) -1208.0149322 -1208.3875168 -1208.4342191
Molpro(std b) -1208.0149322 -1208.3875168 -1208.4342191
Aces(dropmo) -1208.0149322 -1208.3875167 -1208.4342190
Aces(core) -1208.0149322 -1208.4096195 -1208.4546794
GAMESS(core) -1208.0149322 -1208.4096195 -1208.4546794

SiCl2.t uhf ump2 ccsd(t)
GAMESS -1207.9580760 -1208.3069062 N.A.
Gaussian -1207.9580760 -1208.3069062 -1208.3496318
Molpro(ext b) -1207.7724952 N.A. N.A.
Molpro(std b) -1207.7724952 N.A. N.A.
Aces(dropmo) -1207.9580760 -1208.3069062 -1208.3496319
Aces(core) -1207.9580760 -1208.3286703 -1208.3699382
GAMESS(core) -1207.9580760 -1208.3286703 N.A.

CH3.d uhf ump2 ccsd(t)
GAMESS -39.5656619 -39.6982657 N.A.
Gaussian -39.5656619 -39.6982657 -39.7246525
Molpro(ext b) -39.3668603 -39.4847095 -39.5128610
Molpro(std b) -39.3668603 -39.4847095 -39.5128610
Aces(dropmo) -39.5656619 -39.6982657 -39.7246525
Aces(core) -39.5656619 -39.7009446 -39.7272145
GAMESS(core) -39.5656619 -39.7009446 N.A.

Translational Partition Functions

2005-11-03T09:46:00.000-08:00

k_B/h = 2.08366455 × 1010 s-1 K-1
k_B/p = 1.362595904 × 10-28 m3 K-1
k_B/h × k_B/p = 2.839192782 × 10-18 m3 s-1 K-2

Rate constant (uni) = 2.08366455 × 1010 s-1 K-1 * T * exp(delta G/RT)

Rate constant (bi) = 2.839192782 × 10-18 m3 s-1 K-2 * T * T * exp(delta G/RT)

R = 8.314 47215 J Mol-1 K-1 = 1.9872065 Cal Mol-1 K-1 (assuming 4.184 J/Cal)

When the translational energy levels are closed spaced, the delta E is much smaller than the k_B*T.
Hence we can use integral instead of sum to get Q(trans)

Q(trans) = intergral of e^(-n^2 h^2/8m l^2 k_B T) from n = 0 to infinity

=> Q(trans) = [(2 Pi m k_B T) ^ (1/2)] * l / h (in 1 dimensions)

=> Q(trans) = [(2 Pi m k_B T) ^ (3/2)] * [l^3] / [h^3] = [(2 Pi m k_B T / h^2) ^ (3/2)] * V (in 3 dimensions)

Note that this Q(trans) is dimensionless and V represent the volume per particle.

Now, let talk about the confusing part:

when calculating the rate constant k without the transmission coefficient (anyway tunneling is the large part of the transmission coefficient and is close to 1 at high temperature), there are 2 ways:

1) k = (k_B T/ h) * [e^(- delta G / RT)] (This is NOT the k we need for calculating the rate constant for not-uni-molecular reactions) The unit of the above k is always second^(-1)

2) k = (k_B T/ h) * [e^(- delta G / RT)] * [V ^ (Nr-1)] where Nr is the number of reactants and V is the volume per particle at the reaction temperature and pressure.
This k has the unit of [second ^ (-1)] * [m ^ 3] / particle and is the k we need for the rate constant calculations.

Why?
Because the real k we need for rate constant calculations does not only depend on partition functions and electronic energies but also on the volume per particle.

k = (k_B T/ h) (Q_ts/Q_react)_elec (Q_ts/Q_react)_vib (Q_ts/Q_react)_rot [(Q_ts/V)/(Q_react/V ^ (Nr))]_trans * e^(-E_0/RT)

=> k = [V ^ (Nr -1)] * (k_B T/ h) * [e^(- delta G /RT)]

Now, problem is solved.
For unimolecular reactions: k = (k_B T/ h) * [e^(- delta G /RT)]
For bimolecular reactions: k = V * (k_B T/ h) * [e^(- delta G /RT)] = (k_B T/p) (k_B T/ h) * [e^(- delta G /RT)]
(* Note that V is volume per particle, so V = k_B T /p )
which leads to a simplified term k = [(k_B T)^2 / (ph)] * [e^(- delta G /RT)]
The unit of k will be [kg m^2 s^(-2)]^2 / [kg m s^(-2) m^(-2) kg m^2 s^(-1)] = m^3 s^(-1) per particle

Properties of MTS (Methyltrichlorosilane)

2005-10-31T08:46:00.000-08:00

Detailed Reaction Kinetics for CFD Modeling of Nuclear Fuel Pellet Coating for High-Temperature Gas-Cooled Reactors

Precursor Compound: Methyltrichlorosilane (MTS) CH3SiCl3
Carrier Gas: H2 (+ He/Ar)

Experimental Values:

Bond Energy & Bond Length from
http://chemviz.ncsa.uiuc.edu/content/doc-resources-bond.html or http://chemviz.ncsa.uiuc.edu/content/doc-resources-bond.pdf

Bond Bond Energy (kJ/mol) Bond Length (pm)
H-H 432 74
H-C 411 109
H-Cl 428 127
H-Si 318 148
C-C 346 154
C-Cl 327 177
C-Si 318 185
Cl-Cl 240 199
Cl-Si 381 202
Si-Si 222 233

Heat of Formation (kcal/mol) from references cited by
Allendorf and Melius (JPC, 1993, 97, 720-728)
Species Lit. Values
CH3SiCl3 -136(3), -126.4
(CH3)2SiCl2 -111(2)
CH3SiH2Cl -50.2(2)
CH3SiHCl2 -93.9(2)
(CH3)2SiHCl -67.4(2)
CH3SiCl2 -58.1
(CH3)2SiCl -33.1
CH3SiHCl -17.0
CH3SiCl -13.6

Computational Methods: CCSD(T)//MP2 + empirical corrections

Numbers of Basis Functions using Various Basis Set using (5d,7f) and (6d,10f)

Aug-cc-pVDZ, aug-cc-pVTZ, and 6-31+G(d,p)

#Basis Functions Aug-cc-pVDZ
Aug-cc-pVTZ 6-31+G(d,p)
H 9(9) 23(25) 5
C 23(25) 46(55) 18(19)
Cl 27(29) 50(59) 22(23)
Si 27(29) 50(59) 22(23)
CH3SiCl3 158(168) 315(366) 121(126)
1st row 3s2p 4s3p2d 2s1p
2nd row 4s3p2d 5s4p3d2f 4s3p1d
3rd row 5s4p2d 6s5p3d2f 5s4p1d

ACCD (a.k.a. aug-cc-pV(D+d)Z) and ACCT (a.k.a. aug-cc-pV(T+d)Z)

#Basis Functions ACCD ACCT
H 9 23(25)
C 23(25) 46(55)
Cl 32(35) 55(65)
Si 32(35) 55(65)
CH3SiCl3 178(192) 335(390)
1st row 3s2p 4s3p2d
2nd row 4s3p2d 5s4p3d2f
3rd row 5s4p3d 6s5p4d2f

C2v or Cs: C2H4 <-> C2H2 + H2

2005-10-29T08:33:00.000-07:00

1. C2v constrained reaction is woodward-hoffman forbidden (A1 -> B1)

C2H2 + H2: A1 B1 A1 B1 A1 A1 B2 A1 || B1 A1
C2H4 : A1 B1 A1 B1 A1 A1 B1 B2 || A1 B1

2. Cs constrained reaction is allowed since A1 and B1 becomes A'. 1 stands for A' and 2 stands for A''
C2H2 + H2: 1 1 1 1 1 1 1 2 || 1 1
C2H4 : 1 1 1 1 1 1 1 2 || 1 1

3. C2 constrained reaction is frobidden
C2H2 + H2: A B A B A A B A || A B
C2H4 : A B A B A A B B || A B

4. Conopt does not work since often many instead of only 1 coupled vibration moded lead to the transition state.
If only 1 mode is selected, then conopt will faithfully follow that vib mode which usually leads to a weird structure with the same symmetry as that vib mode.