Debye length when ionic strength = 0

[rwxayheee]

Dear APBS experts,

I have a question on how to properly run an APBS calculation without ions. In my input file, I set the ion concentration to be 0. But from the APBS output, I found that the Debye length = 0 A. I'm wondering whether that means the screening is turned off, or maybe I have made some mistake in the input which leads to a Debye length of 0..

Below is my input:

read
mol pqr protein.pqr
end
elec
mg-auto
dime 513 513 513

cglen 1000 1000 1000
fglen 256 256 256

cgcent mol 1
fgcent mol 1

mol 1
lpbe
bcfl sdh

ion 1 0.00 1.62 # Na+ Born radius
ion -1 0.00 2.26 # Cl- Born radius

pdie 4
sdie 78
srfm smol

chgm spl2

sdens 10.0
srad 1.40
swin 0.30
temp 298
calcenergy total

calcforce no
write pot dx protein

end
quit

Also, I was wondering how diluted the solvent needs to be for the application of the screened coulomb potential. I was planning to calculate binding energy at different ionic strengths from 0 to 1M, but I'm worried whether 1M is valid or too high for the "mg-auto" type of calculation.

Any comments/suggestions would be really appreciated. Thank you for your time and kind advice in advance!

Thanks,
Amy

[sobolevnrm]

Hi @rwxayheee --

The Debye length isn't defined at zero ionic strength so I just had the output report "0". This should probably be "infinite" or "undefined" instead.

In general, Poisson-Boltzmann theory does OK treating non-specific ion behavior at relatively low ionic strengths. I get skeptical of PB results when any of the following conditions are met:

• There is a known specific-ion effect (e.g., fluoride behaves differently than chloride, potassium behaves differently than sodium, etc.) -- this is actually fairly common.
• There is a significant concentration of divalent (or higher valency) ions; e.g., greater than 50 mM
• The total ionic strength is greater than 400 mM

I don't have strong justifications for all of these conditions -- they're just rules of thumb that I use when applying PB methods to systems. Caveat emptor.

Thanks,

Nathan

[rwxayheee]

Hi Nathan,
Thank you so much for your kind reply. That is very helpful!!