Rm BSIHydrogenLike, Add BSI with charge Z #2013

n01r · 2017-05-03T15:00:25Z

The BSIHydrogenLike model was much too narrow since it was only applicable
to hydrogenlike states, i.e. states that have charge (Z-1)+.
The following replacement was done in

F_BSI = E_i^2 / (4*Z)

Z (nuclear charge) --> Z_i (ion residual charge)

That means that in a naive way the inner electrons shield the nuclear charge
from the electron that is to be ionized.

The `BSIHydrogenLike` model was much too narrow since it was only applicable to hydrogenlike states, i.e. states that have charge (Z-1)+. The following replacement was done in F_BSI = E_i^2 / (4*Z) Z (nuclear charge) --> Z_i (ion residual charge) That means that in a naive way the inner electrons shield the nuclear charge from the electron that is to be ionized. --- Also the documentation has been refined to better describe the assumptions.

PrometheusPi

Just some minor changes in the code but please update the comment and used schemes in all speciedDefinition.param e.g. here

PrometheusPi · 2017-05-04T12:52:49Z

src/picongpu/include/particles/ionization/byField/BSI/AlgorithmBSI.hpp

                /* ionization potential in atomic units */
                const float_X iEnergy = GetIonizationEnergies<ParticleType>::type()[cs];
+                /* the charge that attracts the electron that is to be ionized:
+ +               * equals `protonNumber - #allInnerElectrons`


OPTIONAL:
perhaps just write number instead of #

or no. for number of

that would be fine too

PrometheusPi · 2017-05-04T12:53:42Z

src/picongpu/include/particles/ionization/byField/BSI/BSI.def

-     *   of ionization energies according to strong external electric fields
+     * - This model accounts for naive ion charge shielding by inner electrons
+     *   as it assumes that the charge the electron 'feels' is equal to
+     *   `Z - #NumberOfBoundInnerShellElectrons`.


REQUIRED:
you are using both # and number - please just use number

PrometheusPi · 2017-05-04T14:13:55Z

@n01r What happens now to BSIEffectiveZ? Where is the difference between the new BSI and BSIEffectiveZ?

n01r · 2017-05-04T14:46:03Z

@PrometheusPi The difference lies in how the residual ion charge is determined.
In the simple BSI case one would naively say: The charge the electron sees comes from the proton number reduced by the number of other bound electrons. This reduces the residual ion to a single point and neglects all geometric effects. So when I introduced BSIEffectiveZ I tried to account for that by using effective charges that were calculated from shielding and considering orbital structure.

Just for illustration (this will also find its way into the documentation).

You see here nuclear potentials influenced by a homogenous external electric field. The potentials are created just by varying the charge Z. We compare Hydrogen with Carbon-1+ (often denoted as C-II). The external electric field is the same in all three cases and is just exactly the BSI field strength that you would calculate with BSIEffectiveZ. It becomes apparent that even though the outermost electron in C-II lies energetically lower than the one in Hydrogen both will become classically ionized in the BSIEffectiveZ model. The simple BSI model sets the barrier to a larger value and thus does not allow ionization at this point. That does not make it better, though. Other effects are neglected as well in both models, e.g. Stark upshift of ionization energies which would deepen the potential well and increase the necessary field strength for ionization. Both of these models should just be used as a simple continuation from the tunneling regime to the barrier suppression regime when one increases external field strength. There are other empirical models for this purpose as well, though.

ax3l

Looks good to me besides Richard comments about updating all input files in examples, too.

I think you can move BSIEffectiveZ up to the "normal" models, too.

PrometheusPi · 2017-05-08T07:14:12Z

@n01r Could you please add the changes.

n01r · 2017-05-08T07:44:14Z

Sorry for the delay.

Changes have been implemented. Please check if you're okay with them, @ax3l and @PrometheusPi .

PrometheusPi · 2017-05-08T07:46:59Z

@n01r thank you for the fast reply, I will have a look at the code today

ax3l · 2017-05-08T07:47:35Z

@PrometheusPi looks good to me, if you are ok I can review and merge it

PrometheusPi · 2017-05-08T08:40:43Z

I am fine with that. Go ahead.

all comments addressed

ax3l

does not compile, see comments inline

ax3l · 2017-05-08T09:12:13Z

src/picongpu/include/particles/ionization/byField/BSI/AlgorithmBSI.hpp

+ +              float_X effectiveCharge = chargeState + float_X(1.0);
                /* critical field strength in atomic units */
-                float_X critField = iEnergy*iEnergy / (float_X(4.0) * protonNumber);
+                float_X critField = iEnergy*iEnergy / (float_X(4.0) * effectiveCharge);


this should be a float_X const

there is a + in front of this line

oO where the hell did that come from?

copied from a diff?

mmh, must be it 🤕

ax3l · 2017-05-08T09:12:50Z

src/picongpu/include/particles/ionization/byField/BSI/AlgorithmBSI.hpp

+                /* the charge that attracts the electron that is to be ionized:
+ +               * equals `protonNumber - no. allInnerElectrons`
+                 */
+ +              float_X effectiveCharge = chargeState + float_X(1.0);


this should be a float_X const

there is a + in front of this line

ax3l · 2017-05-08T09:13:02Z

src/picongpu/include/particles/ionization/byField/BSI/AlgorithmBSI.hpp

                /* ionization potential in atomic units */
                const float_X iEnergy = GetIonizationEnergies<ParticleType>::type()[cs];
+                /* the charge that attracts the electron that is to be ionized:
+ +               * equals `protonNumber - no. allInnerElectrons`


there is a + in front if this line

ax3l · 2017-05-08T09:14:16Z

src/picongpu/include/particles/ionization/byField/BSI/AlgorithmBSI.hpp

            /* verify that ion is not completely ionized */
            if (chargeState < protonNumber)
            {
                uint32_t cs = math::float2int_rd(chargeState);


should be uint32_t const cs = ...

Move the `BSIEffectiveZ` model up from the R&D section to the usable models. Also: implement changes to comments requested in PR 2013.

n01r · 2017-05-08T11:22:59Z

fixed the 4 lines and amended the commit

PrometheusPi

looks good to me

Adds the following changes to the `release-0.3.0` branch: - Multiple ionizers per species: `ionizer< ... >` became a sequence of `ionizers< ... >` ComputationalRadiationPhysics#1999 - Thomas-Fermi: - fixes in domain decomposition ComputationalRadiationPhysics#2007 - fixes compile issue (missing include) ComputationalRadiationPhysics#2003 - BSI models restructured ComputationalRadiationPhysics#2013 - ADK: fix effective principal quantum number `nEff` ComputationalRadiationPhysics#2011 - multiple ionization algorithms can be applied per species, e.g. cut-off barrier suppression ionization (BSI), probabilistic field ionization (ADK) and collisional ionization ComputationalRadiationPhysics#1999 - FieldTmp: Gather support to fill GUARD ComputationalRadiationPhysics#2009 - Fix undefined device variable in tilted laser pulse ComputationalRadiationPhysics#2017 - CompileTime Accessor: Type (::type) ComputationalRadiationPhysics#1998 - TBG Macros: Outdated Comment ComputationalRadiationPhysics#2004 Increases the state of the `0.3.0` release candidate to `rc3`.

n01r added the refactoring code change to improve performance or to unify a concept but does not change public API label May 3, 2017

n01r added this to the 0.3.0: Release Candidate milestone May 3, 2017

n01r requested a review from ax3l May 3, 2017 15:00

n01r force-pushed the topic-replaceBSIHydrogenLike branch from bdab947 to d7a9774 Compare May 3, 2017 15:01

n01r force-pushed the topic-replaceBSIHydrogenLike branch from d7a9774 to 4d41698 Compare May 3, 2017 15:08

PrometheusPi previously requested changes May 4, 2017

View reviewed changes

PrometheusPi self-assigned this May 4, 2017

ax3l approved these changes May 4, 2017

View reviewed changes

ax3l changed the title ~~Rename BSIHydrogenLike to BSI and fix charge Z~~ Rm BSIHydrogenLike, Add BSI with charge Z May 8, 2017

ax3l requested changes May 8, 2017

View reviewed changes

ax3l self-assigned this May 8, 2017

Move BSIEffectiveZ up to usable models

d7eb57c

Move the `BSIEffectiveZ` model up from the R&D section to the usable models. Also: implement changes to comments requested in PR 2013.

n01r force-pushed the topic-replaceBSIHydrogenLike branch from a4ec604 to d7eb57c Compare May 8, 2017 11:20

ax3l approved these changes May 8, 2017

View reviewed changes

PrometheusPi approved these changes May 8, 2017

View reviewed changes

PrometheusPi merged commit 32f679d into ComputationalRadiationPhysics:dev May 9, 2017

ax3l mentioned this pull request May 11, 2017

Fixes & Ionizers for 0.3.0-rc3 #2035

Merged

PrometheusPi mentioned this pull request May 17, 2017

BSI documentation: move BSIEffectiveZ to usable #2019

Closed

Rm BSIHydrogenLike, Add BSI with charge Z #2013

Rm BSIHydrogenLike, Add BSI with charge Z #2013

Uh oh!

Conversation

n01r commented May 3, 2017

Uh oh!

PrometheusPi left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

ax3l May 8, 2017 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

PrometheusPi commented May 4, 2017

Uh oh!

n01r commented May 4, 2017

Uh oh!

ax3l left a comment

Choose a reason for hiding this comment

Uh oh!

PrometheusPi commented May 8, 2017

Uh oh!

n01r commented May 8, 2017

Uh oh!

PrometheusPi commented May 8, 2017

Uh oh!

ax3l commented May 8, 2017 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

PrometheusPi commented May 8, 2017

Uh oh!

ax3l left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

n01r commented May 8, 2017

Uh oh!

PrometheusPi left a comment

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

ax3l May 8, 2017 •

edited

Loading

ax3l commented May 8, 2017 •

edited

Loading