[AMBER] Slab of water

From: M Conte <maxwell.conte.gmail.com>
Date: Mon, 5 Mar 2018 17:28:18 -0600

 Hi Amber users,
I'm trying to create a slab of water surrounded by blank space similar to
this:


(J. Dweik, M. Srour, K. Karaky, M. Kobeissi, W. Joumaa and K. Abou-Saleh,
"Molecular Simulation of Ion Transport at the Water/Vapor Interface," *Open
Journal of Physical Chemistry*, Vol. 2 No. 3, 2012, pp. 147-155. doi:
10.4236/ojpc.2012.23020 <http://dx.doi.org/10.4236/ojpc.2012.23020>.)


by first creating and equilibrating a cube of tip4p water (400 molecules,
22.90Å on all sides, 300K), saving the last frame of the production run as
a .pdb file, increasing the z-axis to 50Å in the final restart file of the
production run, and equilibrating the expanded box one more. I'm using the
following parameters:


imin=0,
ntb=2, ntp=3, iwrap=1, barostat=1,
irest=1,
ntx=5,
cut=8.0,
ntf=2,ntc=2,
csurften=3, ninterface=2, gamma_ten=80.0
tempi=300.0, temp0=300.0,
nstlim=100000,
dt=0.001,
ntpr=50, ntwx=50
/

After equilibration for 100ps, I get an initially expanded box, but over
time, the water molecules expand apart, as shown here
<https://drive.google.com/open?id=1Unla61s-G4C_H2t4FglFsyGTOSO_ZH3l>. I had
the idea to try to manipulate surface tension, but no
noticeable improvements occurred. I'm also unsure how to expand the box in
the negative z-direction as well; is there a way to do so? I think
expansion in the -z direction would lessen potential interactions with a
water molecule at the bottom of the slab crossing over the boundary and
entering in at the top of the box causing waters at the top of the slab to
gravitate towards it. Another idea I had was to limit the periodic boundary
conditions of the xy plane altogether instead of having PBC's for all three
planes. However, I'm not too sure if there's a function to do that.

Thank you very much for your suggestions and ideas,
Maxwell


-- 
*Maxwell Conte*
maxwell.conte190.topper.wku.edu
maxwell.conte.gmail.com



_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber

image.png
(image/png attachment: image.png)

Received on Mon Mar 05 2018 - 15:30:01 PST
Custom Search