Add plugin: XPS #518

superstar54 · 2023-10-20T12:14:00Z

This PR adds a plugin for calculating X-ray Photoelectron Spectroscopy (XPS) spectra using the XpsWorkChain of aiida-quantumespresso package.

Setting Panel

The Setting panel allows users to

select the peak (element and orbital) to be calculated. The availability of peak in the panel depends on the corresponding pseudopotentials.
adjust the parameters based on the input structure.

Results Panel

The Result panel displays the final XPS spectrum.

broadened spectrum using a Voight profile (combined Lorenzian and Gaussian)
the chemical shift values (differences in total energy relative to the lowest value)
absolute binding energy
the Offset Energies (δ)

Pseudopotentials
XPS calculations in QE require core-hole pseudopotentials. We supply a set of pseudopotentials specific to XPS calculations in the XPS Pseudopotential Repository. This repository provides the core-hole pseudopotentials as an AiiDA archive file. There are two kind of pseudopotentials available:

pbe
pbesol

WorkChain

To make the setting as simple as possible for user. The following default setting are used.

core_hole_treatment.
- full for molecule
- xch_smear for metal
- xch_fixed for insulator
supercell_min_parameter updated based on protocol.

codecov · 2023-10-20T12:24:00Z

Codecov Report

Attention: 131 lines in your changes are missing coverage. Please review.

Comparison is base (54bc0cf) 77.48% compared to head (8f8fd4a) 75.82%.

Files	Patch %	Lines
src/aiidalab_qe/plugins/xps/result.py	10.57%	93 Missing ⚠️
src/aiidalab_qe/plugins/xps/workchain.py	18.42%	31 Missing ⚠️
src/aiidalab_qe/plugins/xps/setting.py	92.39%	7 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #518      +/-   ##
==========================================
- Coverage   77.48%   75.82%   -1.67%     
==========================================
  Files          55       60       +5     
  Lines        4011     4277     +266     
==========================================
+ Hits         3108     3243     +135     
- Misses        903     1034     +131

Flag	Coverage Δ
python-3.10	`75.82% <50.75%> (-1.67%)`	⬇️
python-3.8	`75.85% <50.75%> (-1.67%)`	⬇️
python-3.9	`75.85% <50.75%> (-1.67%)`	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

superstar54 · 2024-01-29T11:12:31Z

@PNOGillespie, Please review this PR. Thanks!

superstar54 · 2024-01-30T13:34:47Z

It seems that the plugins still use the default pseudo family from pw workchain.

ValueError: required pseudo family `SSSP/1.2/PBEsol/efficiency` is not installed. Please use `aiida-pseudo install` toinstall it.

for more information, see https://pre-commit.ci

PNOGillespie

Hi @superstar54, I've had a look at the code and, generally speaking, it looks good. I've left some suggested changes here and there to improve readability and clean up commented-out blocks of code.

The only important change IMO is a selection widget for the element to display in the Result panel - see the relevant comment for discussion.

PNOGillespie · 2024-02-01T14:45:28Z

src/aiidalab_qe/plugins/xps/result.py

+        voigt_profile_help = ipw.HTML(
+            """<div style="line-height: 140%; padding-top: 10px; padding-bottom: 10px">
+        The Voigt function:
+        </div>"""
+        )


Was there meant to be a description of the Voigt function here? I would suggest adding something for the user's benefit, at least just to tell them what the two slider values are.

I added a description for the parameters and also a link to Wikipedia.

PNOGillespie · 2024-02-01T14:55:56Z

src/aiidalab_qe/plugins/xps/result.py

+        for element, data in spectra.items():
+            # print(element, data)
+            for site, d in data.items():
+                g.add_scatter(x=d[0], y=d[1], fill="tozeroy", name=f"{element}_{site}")


Would it be possible to add a dropdown/selection widget to change between different elements? (e.g. like what the XAS plugin has)

On the one hand, the user should know that XPS spectra from different elements aren't comparable and to look at the different elements separately. On the other hand, it would tidy up the UI a lot - particularly when considering a structure with many inequivalent atoms (e.g. a complex molecule)

Thanks! I added the selection dropdown for different elements, like in the XAS plugin.

PNOGillespie · 2024-02-01T15:00:51Z

src/aiidalab_qe/plugins/xps/setting.py

+
+    pseudo_title = ipw.HTML(
+        """<div style="padding-top: 0px; padding-bottom: 0px">
+        <h4>Pseudopotential</h4></div>"""


Just to make things easier to read:

Suggested change

<h4>Pseudopotential</h4></div>"""

<h4>Core-Hole pseudopotential group</h4></div>"""

I changed as you suggested.

PNOGillespie · 2024-02-01T15:06:50Z

src/aiidalab_qe/plugins/xps/setting.py

+    peak_title = ipw.HTML(
+        """<div style="padding-top: 0px; padding-bottom: 0px">
+        <h4>Select peak</h4></div>"""
+    )
+    peak_help = ipw.HTML(
+        """<div style="line-height: 140%; padding-top: 6px; padding-bottom: 0px">
+        The list of peaks to be considered for analysis.
+        </div>"""
+    )


Would something like this make sense for readability? "Core-Hole" or "Core-Level" seems like it would be more intuitive, but if "Peak" is the correct technical term in this case then don't worry.

Suggested change

peak_title = ipw.HTML(

"""<div style="padding-top: 0px; padding-bottom: 0px">

<h4>Select peak</h4></div>"""

)

peak_help = ipw.HTML(

"""<div style="line-height: 140%; padding-top: 6px; padding-bottom: 0px">

The list of peaks to be considered for analysis.

</div>"""

)

peak_title = ipw.HTML(

"""<div style="padding-top: 0px; padding-bottom: 0px">

<h4>Select core-level</h4></div>"""

)

peak_help = ipw.HTML(

"""<div style="line-height: 140%; padding-top: 6px; padding-bottom: 0px">

The list of core-levels to be considered for analysis.

</div>"""

)

Thanks! I change it to core-level

PNOGillespie · 2024-02-01T15:14:57Z

src/aiidalab_qe/plugins/xps/setting.py

+            # self.core_hole_treatment_title,
+            # self.core_hole_treatment_help,
+            # self.core_hole_treatment,
+            self.pseudo_title,
+            self.pseudo_help,
+            self.pseudo_group,
+            self.peak_title,
+            self.peak_help,
+            ipw.HBox(
+                [self.peak_list],
+            ),
+            # self.supercell_title,
+            # self.supercell_help,
+            # ipw.HBox(
+            # [self.supercell_min_parameter],
+            # ),


There are quite a few commented-out sections here which should be cleaned up if possible:

Suggested change

# self.core_hole_treatment_title,

# self.core_hole_treatment_help,

# self.core_hole_treatment,

self.pseudo_title,

self.pseudo_help,

self.pseudo_group,

self.peak_title,

self.peak_help,

ipw.HBox(

[self.peak_list],

),

# self.supercell_title,

# self.supercell_help,

# ipw.HBox(

# [self.supercell_min_parameter],

# ),

self.pseudo_title,

self.pseudo_help,

self.pseudo_group,

self.peak_title,

self.peak_help,

ipw.HBox(

[self.peak_list],

),

Does this also mean (based on the fact that supercell_min_parameter is not used) that the supercell size used in calculation is defined entirely by the protocol?

Yes. This would be more user friendly

…alab-qe into feature/plugin/xps

superstar54 · 2024-02-14T12:07:10Z

@PNOGillespie Thanks for your review. I made the changes as you suggested. I added the dropdown button to select the spectrum for different elements. I also added a table to show the offset energies, which are used to calibrate the calculated binding energy.

PNOGillespie

Hi @superstar54. Code looks good for the most part. I've added a few final changes, after which I think this PR is complete.

src/aiidalab_qe/plugins/xps/result.py

PNOGillespie · 2024-02-14T15:33:10Z

src/aiidalab_qe/plugins/xps/result.py

+                        g.add_scatter(
+                            x=d["x"], y=d["y"], fill=fill.value, name=d["site"]
+                        )


This section here causes a crash because you're calling the value from the UI element rather than converting it to a type of fill method (e.g. "tozeroy"):

Suggested change

g.add_scatter(

x=d["x"], y=d["y"], fill=fill.value, name=d["site"]

)

if fill.value:

fill_type = "tozeroy"

else:

fill_type = None

g.add_scatter(

x=d["x"], y=d["y"], fill=fill_type, name=d["site"]

)

thanks! I changed it.

PNOGillespie · 2024-02-14T15:40:03Z

src/aiidalab_qe/plugins/xps/result.py

+        correction_energies_table = ipw.HTML(
+            """<h4>Offset Energies (δ)</h4>
+            <div>The binding energies should be corrected by these constant offsets. These offsets mainly depend on chemical element and core level, and are determined by comparing the calculated core electron binding energy to the experimental one.</div>
+            <table><tr><th>Core-level</th><th>Value (eV)</th></tr>"""
+        )


The first sentence should remind the user how the correction energies are used in practice:

Suggested change

correction_energies_table = ipw.HTML(

"""<h4>Offset Energies (δ)</h4>

<div>The binding energies should be corrected by these constant offsets. These offsets mainly depend on chemical element and core level, and are determined by comparing the calculated core electron binding energy to the experimental one.</div>

<table><tr><th>Core-level</th><th>Value (eV)</th></tr>"""

)

correction_energies_table = ipw.HTML(

"""<h4>Offset Energies (δ)</h4>

<div>When comparing binding energies reported by AiiDALab-QE to experimental data, these should be corrected by the given offset listed below</div>

<table><tr><th>Core-level</th><th>Value (eV)</th></tr>"""

)

I would also recommend moving the second sentence (or maybe the relevant concept or description) to the documentation since this is not really needed here.

thanks! I made the changes.

PNOGillespie · 2024-02-14T15:52:17Z

src/aiidalab_qe/plugins/xps/result.py

+        for core_level, value in correction_energies.items():
+            element = core_level.split("_")[0]
+            if element not in self.spectrum_select_options:
+                continue
+            exp = value["exp"]
+            correction_energies_table.value += (
+                f"<tr><td>{core_level}</td><td>{exp}</td></tr>"
+            )
+        return correction_energies_table


Just a little addition to make things clearer:

Suggested change

for core_level, value in correction_energies.items():

element = core_level.split("_")[0]

if element not in self.spectrum_select_options:

continue

exp = value["exp"]

correction_energies_table.value += (

f"<tr><td>{core_level}</td><td>{exp}</td></tr>"

)

return correction_energies_table

for core_level, value in correction_energies.items():

element = core_level.split("_")[0]

if element not in self.spectrum_select_options:

continue

exp = value["exp"]

if exp >= 0.0:

sign = "+"

elif exp < 0:

sign = ""

correction_energies_table.value += (

f"<tr><td>{core_level}</td><td>{sign}{exp}</td></tr>"

)

return correction_energies_table

If the values can be positive or negative, then it's usually a good idea to give the sign explicitly even if the number is positive.

The string for the core level name should probably also have the underscore removed, since this doesn't format so well in non-monospace text. I understand this is also the way that the labels for the core levels are written in the Setting panel, but I wouldn't worry about changing those at this stage.

thanks! The sign is indeed very important. I just noticed reverse the sign for all values.

Co-authored-by: Peter Gillespie <[email protected]>

superstar54 · 2024-02-14T17:47:14Z

@PNOGillespie, thanks for your review and very useful suggestions! I made all the changes as suggested.

PNOGillespie

Looks good to me, nice work @superstar54!

superstar54 added 5 commits October 9, 2023 12:17

use the latest plugin format

71d706a

fix workchain and add test

e0b29d4

fix molecule

803a05c

Merge branch 'main' into feature/plugin/xps

8f96fc5

Merge branch 'main' into feature/plugin/xps

7c22c5f

superstar54 marked this pull request as draft October 20, 2023 12:14

superstar54 added 4 commits October 22, 2023 07:51

Merge branch 'main' into feature/plugin/xps

46c8bb2

Merge branch 'main' into feature/plugin/xps

436b4a8

Merge branch 'main' into feature/plugin/xps

c6efec4

add workchain_labels

ff3cd2e

superstar54 marked this pull request as ready for review January 29, 2024 10:13

superstar54 requested a review from unkcpz January 29, 2024 10:59

update test for setting panel

251399f

superstar54 force-pushed the feature/plugin/xps branch from ae6fab8 to 251399f Compare January 29, 2024 16:09

superstar54 added 2 commits January 30, 2024 08:09

Merge branch 'main' into feature/plugin/xps

493b2c3

Merge branch 'main' into feature/plugin/xps

6edff70

unkcpz removed their request for review January 30, 2024 12:51

superstar54 and others added 3 commits January 30, 2024 14:38

update overrides to fix pseudo family problem

de5d290

udpate description for pseudo group

ea21cd8

[pre-commit.ci] auto fixes from pre-commit.com hooks

9944e2a

for more information, see https://pre-commit.ci

PNOGillespie suggested changes Feb 1, 2024

View reviewed changes

support select element for plot

85e777b

superstar54 force-pushed the feature/plugin/xps branch from bdcca63 to 85e777b Compare February 7, 2024 11:15

superstar54 added 2 commits February 7, 2024 12:20

load correction energies

9f71bce

load correction energies

9268914

superstar54 force-pushed the feature/plugin/xps branch from 9f71bce to 9268914 Compare February 9, 2024 09:13

Merge branch 'feature/plugin/xps' of https://github.com/aiidalab/aiid…

7953d87

…alab-qe into feature/plugin/xps

superstar54 force-pushed the feature/plugin/xps branch from 55016fd to 7953d87 Compare February 14, 2024 11:41

superstar54 added 2 commits February 14, 2024 12:43

Merge branch 'main' into feature/plugin/xps

49e00a2

replace peak by core-level

1ceec1b

superstar54 force-pushed the feature/plugin/xps branch from b0c5f89 to 1ceec1b Compare February 14, 2024 12:02

superstar54 requested a review from PNOGillespie February 14, 2024 12:05

PNOGillespie suggested changes Feb 14, 2024

View reviewed changes

superstar54 and others added 2 commits February 14, 2024 17:23

Update src/aiidalab_qe/plugins/xps/result.py

dfe2ab4

Co-authored-by: Peter Gillespie <[email protected]>

fix fill_type and sign of the offset energy

9ce6b92

superstar54 force-pushed the feature/plugin/xps branch from deae7cd to 9ce6b92 Compare February 14, 2024 17:26

Merge branch 'main' into feature/plugin/xps

8f8fd4a

superstar54 force-pushed the feature/plugin/xps branch from a6aae7b to 8f8fd4a Compare February 14, 2024 17:32

superstar54 requested a review from PNOGillespie February 14, 2024 17:46

PNOGillespie approved these changes Feb 14, 2024

View reviewed changes

superstar54 merged commit f951964 into main Feb 15, 2024
16 of 17 checks passed

superstar54 deleted the feature/plugin/xps branch February 15, 2024 07:35

This was referenced Mar 20, 2024

XPS plugin #595

Closed

Example structure SiO2 has extra tags #617

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Add plugin: XPS #518

Add plugin: XPS #518

superstar54 commented Oct 20, 2023 •

edited

Loading

codecov bot commented Oct 20, 2023 •

edited

Loading

superstar54 commented Jan 29, 2024

superstar54 commented Jan 30, 2024

PNOGillespie left a comment •

edited

Loading

PNOGillespie Feb 1, 2024

superstar54 Feb 14, 2024

PNOGillespie Feb 1, 2024

superstar54 Feb 14, 2024

PNOGillespie Feb 1, 2024

superstar54 Feb 14, 2024

PNOGillespie Feb 1, 2024

superstar54 Feb 14, 2024

PNOGillespie Feb 1, 2024

superstar54 Feb 14, 2024

superstar54 commented Feb 14, 2024

PNOGillespie left a comment

PNOGillespie Feb 14, 2024

superstar54 Feb 14, 2024

PNOGillespie Feb 14, 2024

superstar54 Feb 14, 2024

PNOGillespie Feb 14, 2024

superstar54 Feb 14, 2024 •

edited

Loading

superstar54 commented Feb 14, 2024

PNOGillespie left a comment

	<h4>Pseudopotential</h4></div>"""
	<h4>Core-Hole pseudopotential group</h4></div>"""

-                        g.add_scatter(
-                            x=d["x"], y=d["y"], fill=fill.value, name=d["site"]
-                        )
+                        if fill.value:
+                            fill_type = "tozeroy"
+                        else:
+                            fill_type = None
+                        g.add_scatter(
+                            x=d["x"], y=d["y"], fill=fill_type, name=d["site"]
+                        )

Add plugin: XPS #518

Add plugin: XPS #518

Conversation

superstar54 commented Oct 20, 2023 • edited Loading

Setting Panel

Results Panel

WorkChain

codecov bot commented Oct 20, 2023 • edited Loading

Codecov Report

superstar54 commented Jan 29, 2024

superstar54 commented Jan 30, 2024

PNOGillespie left a comment • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

superstar54 commented Feb 14, 2024

PNOGillespie left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

superstar54 Feb 14, 2024 • edited Loading

Choose a reason for hiding this comment

superstar54 commented Feb 14, 2024

PNOGillespie left a comment

Choose a reason for hiding this comment

superstar54 commented Oct 20, 2023 •

edited

Loading

codecov bot commented Oct 20, 2023 •

edited

Loading

PNOGillespie left a comment •

edited

Loading

superstar54 Feb 14, 2024 •

edited

Loading