Sro SrCO3 SrF2 SrSO4  SrTiO3 SrZrO3 SrB4O7 SrSi2 BiSrCaCuO 

Basic XPS Information Section

The Basic XPS Information Section provides fundamental XPS spectra, BE values, FWHM values, BE tables, overlays of key spectra, histograms and a table of XPS parameters.
The Advanced XPS Information Section is a collection of overlays of significant key spectra, peak-fitting advice, data collection guidance, material info,
common contaminants, degradation during analysis, auto-oxidation, gas capture study, valence band spectra, Auger spectra, and more.
Published literature references, and website links are summarized at the end of the advanced section.
 Periodic Table – HomePage                     XPS Database of Polymers                  → Six (6) BE Tables


 

Strontium (Sr)

 

Tausonite – SrTiO3 Strontium Metal – Sro Celestite – SrSO4

 

  Page Index
  • Expert Knowledge & Explanations


Strontium (Sro) Metal

Peak-fits, BEs, FWHMs, and Peak Labels


  .
Strontium (Sro) Metal
Sr (3d) Spectrum – raw spectrum
Strontium (Sro) Metal
Peak-fit of Sr (3d) Spectrum (w/o asymm)

 Periodic Table – HomePage  
Strontium (Sro) Metal
Sr (3d) Spectrum – extended range 
Strontium (Sro) Metal
Peak-fit of Sr (3d) Spectrum (w asymm)
   

 

Survey Spectrum of Strontium (Sro) Metal
with Peaks Integrated, Assigned and Labelled


 Periodic Table 

XPS Signals for Strontium, (Sro) Metal

Spin-Orbit Term,  BE (eV) Value, and Scofield σ for Aluminum Kα X-rays (1486 eV, 8.33 Ang)

Overlaps Spin-Orbit Term BE (eV) Value Scofield σ from 1486 eV X-rays IMFP (TPP-2M) in Å
Sr (3s) 359 1.86 42.5
C (1s) overlaps Sr (3p1/2) 280 2.25 45.2
Cl (2s) overlaps Sr (3p3/2) 270 4.37 45.2
Pb (4f7) overlaps Sr (3d3/2) 135.9 2.06 49.1
P (2p) overlaps Sr (3d5/2) 134.2 2.99 49.1
Sr (4s) 39 0.291 51.9
Sr (4p) 21 0.775 52.5

σ:  abbreviation for the term Scofield Photoionization Cross-Section which is used with IMFP and TF to produce RSFs and atom% quantitation

Plasmon Peaks

Energy Loss Peaks

Auger Peaks

Energy Loss    Intrinsic Plasmon Peak:  ~xx eV above peak max
Expected Bandgap for SrF2:  6-7 eV
Work Function for Sr:  xx eV

*Scofield Cross-Section (σ) for C (1s) = 1.0

 Periodic Table 


 

Plasmon Peaks from Strontium, Sro Metal
 Fresh exposed bulk produced by extensive Ar+ ion etching

Sr (3d) – Extended Range Spectrum Sr (3d) – Extended Range Spectrum – Vertically Zoomed
 Periodic Table 

 

Sr (LMM) Auger Peaks from Sro Metal
 Fresh exposed bulk produced by extensive Ar+ ion etching

Sro Metal – high BE Auger peaks Sro Metal – full Auger range
 

Features Observed

  • xx
  • xx
  • xx

 Periodic Table 


 

Side-by-Side Comparison of
Strontium Sulfate and Strontium Fluoride
Peak-fits, BEs, FWHMs, and Peak Labels

SrSO4 SrF2

Sr (3d) from SrSO4
Flood Gun ON
Charge Referenced to C (1s) at 285.0 eV

Sr (3d) from SrF2
Flood Gun ON
Charge Referenced to C (1s) at 285.0 eV
 


 
SrSO4 SrF2
C (1s) from SrSO4
Flood Gun ON
Charge Referenced to C (1s) at 285.0 eV

C (1s) from SrF2
Flood Gun ON
Charge Referenced to C (1s) at 285.0 eV

 Periodic Table 

 
SrSO4 SrF2
O (1s) from SrSO4
Flood Gun ON
Charge Referenced to C (1s) at 285.0 eV

F (1s) from SrF2
Flood Gun ON
Charge Referenced to C (1s) at 285.0 eV



.
SrSO4
S (2p) from SrSO4
Flood Gun ON
Charge Referenced to C (1s) at 285.0 eV
 Periodic Table

 


 

Survey Spectrum of Strontium Sulfate, SrSO4
with Peaks Integrated, Assigned and Labelled

 

 Periodic Table 


 

 

Survey Spectrum of Strontium Fluoride, SrF2
with Peaks Integrated, Assigned and Labelled

 Periodic Table  


Overlays of Sr (3d) Spectra for
Strontium (Sro) metal, SrSO4 and SrF2

 

 Overlay of Sro metal and SrSO4     Sr (3d)
Flood Gun ON
Charge Referenced to C (1s) at 285.0 eV
Chemical Shift:  +0.8 eV

 Overlay of Sro metal and SrF2    Sr (3d)
Flood Gun ON
Charge Referenced to C (1s) at 285.0 eV
Chemical Shift:  -0.4  eV

 Periodic Table  Copyright ©:  The XPS Library 

 

Overlay of Sr (3d)
Sro Metal, SrSO4, & SrF2

Features Observed

  • xx
  • xx
  • xx

 Periodic Table 


 

Valence Band Spectra
SrSO4, SrF2

SrSO4
Flood gun is ON,  Charge referenced so C (1s) = 285.0 eV
SrF2
Flood gun is ON,  Charge referenced so C (1s) = 285.0 eV


Overlay of Valence Band Spectra
for SrSO4 and SrF2

Features Observed

  • xx
  • xx
  • xx

 Periodic Table 


 

 

Overlay of Sr (3d) Spectra from
Strontium Carbonate (SrCO
3) and Strontium Borate (SrBO3)
Strontium Carbonate, SrCO3
Peak-fit of Sr (3d) from SrCO3, powder

pressed onto stage
Charge Referenced to 285.0 eV
Strontium Borate, SrBO3
Peak-fit of Sr (3d) from SrBO3, single crystal

freshly cleaved
Charge Referenced to 285.0 eV


Overlay of Sr (3d) Spectra from
Sr metal, SrCO
3 and SrBO3

Features Observed

  • xx
  • xx
  • xx

Strontium Minerals, Gemstones, and Chemical Compounds

 

Goyazite – SrAl3(PO4)(PO3OH)(OH)6 Haradaite – SrVSi2O7 Tunellite – SrB6O9(OH)2-3H2O Ancylite – CeSr(CO3)2(OH) · H2O

 Periodic Table 



 

Six (6) Chemical State Tables of Sr (3d5/2) BEs

 

  • The XPS Library Spectra-Base
  • PHI Handbook
  • Thermo-Scientific Website
  • XPSfitting Website
  • Techdb Website
  • NIST Website

 Periodic Table 



 

Notes of Caution when using Published BEs and BE Tables from Insulators and Conductors:

  • Accuracy of Published BEs
    • The accuracy depends on the calibration BEs used to calibrate the energy scale of the instrument.  Cu (2p3/2) BE can vary from 932.2 to 932.8 eV for old publications 
    • Different authors use different BEs for the C (1s) BE of the hydrocarbons found in adventitious carbon that appears on all materials and samples.  From 284.2 to 285.3 eV
    • The accuracy depends on when the authors last checked or adjusted their energy scale to produce the expected calibration BEs
  • Worldwide Differences in Energy Scale Calibrations
    • For various reasons authors still use older energy scale calibrations 
    • Some authors still adjust their energy scale so Cu (2p3/2) appears at 932.2 eV or 932.8 eV because this is what the maker taught them
    • This range causes BEs in the higher BE end to be larger than expected 
    • This variation increases significantly above 600 eV BE
  • Charge Compensation
    • Samples that behave as true insulators normally require the use of a charge neutralizer (electron flood gun with or without Ar+ ions) so that the measured chemical state spectra can be produced without peak-shape distortions or sloping tails on the low BE side of the peak envelop. 
    • Floating all samples (conductive, semi-conductive, and non-conductive) and always using the electron flood gun is considered to produce more reliable BEs and is recommended.
  • Charge Referencing Methods for Insulators
    • Charge referencing is a common method, but it can produce results that are less reliable.
    • When an electron flood gun is used, the BE scale will usually shift to lower BE values by 0.01 to 5.0 eV depending on your voltage setting. Normally, to correct for this flood gun induced shift, the BE of the hydrocarbon C (1s) peak maximum from adventitious carbon is used to correct for the charge induced shift.
    • The hydrocarbon peak is normally the largest peak at the lowest BE. 
    • Depending on your preference or training, the C (1s) BE assigned to this hydrocarbon peak varies from 284.8 to 285.0 eV.  Other BEs can be as low as 284.2 eV or as high as 285.3 eV
    • Native oxides that still show the pure metal can suffer differential charging that causes the C (1s) and the O (1s) and the Metal Oxide BE to be larger
    • When using the electron flood gun, the instrument operator should adjust the voltage and the XY position of the electron flood gun to produce peaks from a strong XPS signal (eg O (1s) or C (1s) having the most narrow FWHM and the lowest experimentally measured BE. 

 Periodic Table 


Table #1

Sr (3d5/2) Chemical State BEs from:  “The XPS Library Spectra-Base”

C (1s) BE = 285.0 eV for TXL BEs
and C (1s) BE = 284.8 eV for NIST BEs

Element Atomic # Compound As-Measured by TXL or NIST Average BE Largest BE Hydrocarbon C (1s) BE  Source
Sr 38 BiSrCaCuOx 132.2 eV 285.0 eV The XPS Library
Sr 38 Sr-TiO3 (N*1) 132.7 eV 284.8 eV Avg BE – NIST
Sr 38 SrTiO3 132.7 eV 285.0 eV The XPS Library
Sr 38 Sr-(OH)2 (N*1) 133.0 eV 284.8 eV Avg BE – NIST
Sr 38 Sr-CO3 132.9 eV 285.0 eV The XPS Library
Sr 38 Sr-F2 (N*2) 133.7 eV 134.0 eV 284.8 eV Avg BE – NIST
Sr 38 Sr-BO3 133.8 eV 285.0 eV The XPS Library
Sr 38 Sr-F2 133.9 eV 285.0 eV The XPS Library
Sr 38 Sr-SO4 (N*2) 134.0 eV 134.3 eV 284.8 eV Avg BE – NIST
Sr 38 Sr(NO3)2 (N*1) 134.2 eV 284.8 eV Avg BE – NIST
Sr 38 Sr – element 134.3 eV 285.0 eV The XPS Library
Sr 38 Sr-SO4 134.9 eV 285.0 eV The XPS Library
Sr 38 Sr-I2 (N*1) 135 eV 284.8 eV Avg BE – NIST
Sr 38 Sr-O  (N*1) 135.3 eV 284.8 eV Avg BE – NIST

Charge Referencing Notes

  • (N*number) identifies the number of NIST BEs that were averaged to produce the BE in the middle column.
  • The XPS Library uses Binding Energy Scale Calibration with Cu (2p3/2) BE = 932.62 eV and Au (3d7/2) BE = 83.98 eV.  BE (eV) Uncertainty Range:  +/- 0.2 eV
  • Charge Referencing of insulators is defined such that the Adventitious Hydrocarbon C (1s) BE (eV) = 285.0 eV.  NIST uses C (1s) BE = 284.8 eV 
  • Note:   Ion etching removes adventitious carbon, implants Ar (+), changes conductivity of surface, and degrades chemistry of various chemical states.
  • Note:  Ion Etching changes BE of C (1s) hydrocarbon peak.
  • TXL – abbreviation for: “The XPS Library” (https://xpslibrary.com).  NIST:  National Institute for Science and Technology (in USA)

 Periodic Table 


Table #2

Sr (3d5/2) Chemical State BEs from:  “PHI Handbook”

C (1s) BE = 284.8 eV

 Periodic Table 

Copyright ©:  Ulvac-PHI


Table #3

Sr (3d5/2) Chemical State BEs from:  “Thermo-Scientific” Website

C (1s) BE = 284.8 eV

Chemical state Binding energy (eV), Sr (3d5/2)
SrO 132.9
SrCO3 133.4
SrTiO3 133.1

 Periodic Table 

Copyright ©:  Thermo Scientific 


Table #4

Sr (3d5/2) Chemical State BEs from:  “XPSfitting” Website

Chemical State BE Table derived by Averaging BEs in the NIST XPS database of BEs
C (1s) BE = 284.8 eV

 

 Periodic Table 

Copyright ©:  Mark Beisinger


Table #5

Sr (3d5/2) Chemical State BEs from:  “Techdb.podzone.net” Website

 

XPS Spectra – Chemical Shift | Binding Energy
C (1s) BE = 284.6 eV

XPS(X線光電子分光法)スペクトル 化学状態 化学シフト ケミカルシフト

Element Level Compound B.E.(eV) min max
Sr 3d5/2 SrRh2O4 133.0 ±0.3 132.7 133.3
Sr 3d5/2 SrCO3 133.3 ±0.2 133.1 133.5
Sr 3d5/2 SrMoO4 133.5 ±0.2 133.3 133.7
Sr 3d5/2 SrF2 133.8 ±0.3 133.5 134.0
Sr 3d5/2 SrSO4 134.4 ±0.3 134.1 134.6
Sr 3d5/2 Sr 134.5 ±0.3 134.2 134.7
Sr 3d5/2 Sr(NO3)2 134.8 ±0.3 134.5 135.0
Sr 3d5/2 SrO 135.4 ±0.3 135.1 135.6

 Periodic Table 



 

Histograms of NIST BEs for Sr (3d5/2) BEs

Important Note:  NIST Database defines Adventitious Hydrocarbon C (1s) BE = 284.8 eV for all insulators.

Histogram indicates:  134.3 eV for Sro based on 2 literature BEs Histogram indicates:  134.1  eV for SrO based on 2 literature BEs

Table #6


NIST Database of Sr (3d5/2) Binding
Energies

NIST Standard Reference Database 20, Version 4.1

Data compiled and evaluated
by
Alexander V. Naumkin, Anna Kraut-Vass, Stephen W. Gaarenstroom, and Cedric J. Powell
©2012 copyright by the U.S. Secretary of Commerce on behalf of the United States of America. All rights reserved.

Important Note:  NIST Database defines Adventitious Hydrocarbon C (1s) BE = 284.8 eV for all insulators.

 

Element Spectral Line Formula Energy (eV) Reference
Sr 3d5/2 Bi2Sr2Ca2Cu3Ox 131.10  Click
Sr 3d5/2 SrFeO3 131.40  Click
Sr 3d5/2 Bi1.99Sr2.00Ca2Cu3Ox 131.40  Click
Sr 3d5/2 Bi2.00Sr2.00Ca2Cu3Ox 131.40  Click
Sr 3d5/2 Bi2.01Sr2.00Ca2Cu3Ox 131.40  Click
Sr 3d5/2 Bi2.01Sr2.00Ca2Cu3Ox 131.40  Click
Sr 3d5/2 Bi2.01Sr2Ca2Cu3Ox 131.40  Click
Sr 3d5/2 Bi1.6Pb0.4Sr2Ca2Cu3Ox 131.40  Click
Sr 3d5/2 Bi2Sr2CaCu2O8+x 131.60  Click
Sr 3d5/2 Bi1.7Pb0.4Sr2Ca2Cu3O10+x 131.60  Click
Sr 3d5/2 Bi1.55Pb0.6Sr2Ca2Cu3.5O10+x 131.60  Click
Sr 3d5/2 BiPbSr2CaCu2O8+x 131.60  Click
Sr 3d5/2 Bi2Sr2CaCu2O8 131.70  Click
Sr 3d5/2 Bi2Sr1.4CaCu2Ox 131.80  Click
Sr 3d5/2 Bi2Ca1+xSr2-xCu2O8+y 131.80  Click
Sr 3d5/2 Bi2Ca1+xSr2-xCu2O8+y 131.80  Click
Sr 3d5/2 Bi2Sr2Ca2Cu2O8+x 131.80  Click
Sr 3d5/2 Bi2Sr2CaCu2O8+x 131.80  Click
Sr 3d5/2 Bi1.6Pb0.4Sr2CaCu2O8+x 131.80  Click
Sr 3d5/2 Bi2Sr2CaCu2Ox 131.90  Click
Sr 3d5/2 Bi2Sr2CaCu2O8+x 131.90  Click
Sr 3d5/2 Bi2CaSr2Cu2Ox 132.00  Click
Sr 3d5/2 Bi2CaSr2Ni0.2Cu1.8Ox 132.00  Click
Sr 3d5/2 Sr 132.10  Click
Sr 3d5/2 Bi2Sr2CaCu2O8+x 132.20  Click
Sr 3d5/2 Bi2Sr2CuO6 132.30  Click
Sr 3d5/2 Bi2Sr2Ca2Cu3Ox 132.30  Click
Sr 3d5/2 SrO 132.50  Click
Sr 3d5/2 Bi2Sr2YCu2Ox 132.50  Click
Sr 3d5/2 Bi2Sr2CaCu2Ox 132.50  Click
Sr 3d5/2 SrTiO3 132.70  Click
Sr 3d5/2 SrTiO3 132.70  Click
Sr 3d5/2 Sr/Si 132.70  Click
Sr 3d5/2 Bi2Sr2Ca2Cu2O8+x 132.70  Click
Sr 3d5/2 Bi2Sr2CaCu2O8+x 132.70  Click
Sr 3d5/2 SrTiO3 132.70  Click
Sr 3d5/2 SrO 132.80  Click
Sr 3d5/2 SrO 132.80  Click
Sr 3d5/2 SrCO3 132.90  Click
Sr 3d5/2 SrCO3 132.90  Click
Sr 3d5/2 SrCO3 132.90  Click
Sr 3d5/2 SrS 132.90  Click
Sr 3d5/2 SrS 132.90  Click
Sr 3d5/2 Bi2Sr2CaCu2O8+x 132.90  Click
Sr 3d5/2 SrTiO3 132.90  Click
Sr 3d5/2 Bi1.7Pb0.4Sr2Ca2Cu3O10+x 132.90  Click
Sr 3d5/2 Bi1.55Pb0.6Sr2Ca2Cu3.5O10+x 132.90  Click
Sr 3d5/2 BiPbSr2CaCu2O8+x 132.90  Click
Sr 3d5/2 SrCO3 132.90  Click
Sr 3d5/2 La0.8Sr0.2MnO3 132.90  Click
Sr 3d5/2 Sr2CuO2F2.6 132.90  Click
Sr 3d5/2 SrRh2O4 133.00  Click
Sr 3d5/2 Sr(OH)2.8H2O 133.00  Click
Sr 3d5/2 Sr(OH)2.8H2O 133.00  Click
Sr 3d5/2 Bi2Sr1.4CaCu2Ox 133.00  Click
Sr 3d5/2 Bi1.99Sr2.00Ca2Cu3Ox 133.00  Click
Sr 3d5/2 Bi2.00Sr2.00Ca2Cu3Ox 133.00  Click
Sr 3d5/2 Bi2.01Sr2.00Ca2Cu3Ox 133.00  Click
Sr 3d5/2 Bi2.01Sr2.00Ca2Cu3Ox 133.00  Click
Sr 3d5/2 Bi2.01Sr2Ca2Cu3Ox 133.00  Click
Sr 3d5/2 Bi2Sr2CaCu2Ox 133.10  Click
Sr 3d5/2 Bi2Ca1+xSr2-xCu2O8+y 133.15  Click
Sr 3d SrCO3 133.20  Click
Sr 3d5/2 Sr3[Mn(OH)6]2 133.20  Click
Sr 3d5/2 Bi2Sr2Ca0.2Y0.8Cu2Ox 133.30  Click
Sr 3d5/2 Bi2Sr2CaCu2Ox 133.30  Click
Sr 3d5/2 Bi2Sr2Ca0.5Y0.5Cu2Ox 133.30  Click
Sr 3d5/2 Bi1.6Pb0.4Sr2CaCu2O8+x 133.30  Click
Sr 3d5/2 Bi2Sr2CaCu2O8+x 133.40  Click
Sr 3d5/2 Sr2CuO3 133.40  Click
Sr 3d5/2 SrCO3 133.50  Click
Sr 3d5/2 SrMoO4 133.50  Click
Sr 3d5/2 SrF2 133.75  Click
Sr 3d5/2 SrF2 133.75  Click
Sr 3d5/2 La0.972Sr0.212NiO3-x 133.80  Click
Sr 3d5/2 La1.067Sr0.220NiO3-x 134.00  Click
Sr 3d5/2 La0.8Sr0.2CrO3 134.00  Click
Sr 3d5/2 La0.8Sr0.2FeO3 134.00  Click
Sr 3d5/2 SrF2 134.00  Click
Sr 3d5/2 SrF2 134.05  Click
Sr 3d5/2 SrF2 134.05  Click
Sr 3d5/2 SrSO4 134.05  Click
Sr 3d5/2 SrSO4 134.05  Click
Sr 3d5/2 SrF2 134.05  Click
Sr 3d5/2 SrSO4 134.06  Click
Sr 3d5/2 Sr(NO3)2 134.20  Click
Sr 3d5/2 Sr 134.20  Click
Sr 3d5/2 Sr(NO3)2 134.20  Click
Sr 3d5/2 Sr(NO3)2 134.20  Click
Sr 3d5/2 La0.8Sr0.2CoO3 134.20  Click
Sr 3d SrSO4 134.30  Click
Sr 3d5/2 La0.8Sr0.2YO3 134.30  Click
Sr 3d5/2 Sr 134.40  Click
Sr 3d5/2 Sr/Si 134.40  Click
Sr 3d Sr(NO3)2 134.70  Click
Sr 3d5/2 SrCl2 134.70  Click
Sr 3d5/2 SrCl2 134.70  Click
Sr 3d5/2 SrBr2 134.70  Click
Sr 3d5/2 SrBr2 134.70  Click
Sr 3d5/2 La0.996Sr0.159NiO3-x 134.70  Click
Sr 3d5/2 SrBr2 134.70  Click
Sr 3d5/2 SrCl2 134.70  Click
Sr 3d5/2 SrI2 135.00  Click
Sr 3d5/2 SrI2 135.00  Click
Sr 3d5/2 SrI2 135.00  Click
Sr 3d5/2 SrO 135.30  Click

 

 

Statistical Analysis of Binding Energies in NIST XPS Database of BEs

 

 

 Periodic Table 


 

Advanced XPS Information Section

 

Expert Knowledge, Spectra, Features, Guidance and Cautions
for XPS Research Studies on Strontium Materials

 

 


 

Expert Knowledge Explanations

 

 Periodic Table 


 

Strontium Chemical Compounds


Peak-fits and Overlays of Chemical State Spectra

Pure Strontium, Sro:  Sr (3d)
Cu (2p3/2) BE = 932.6 eV
SrCO3:  Sr (3d)
C (1s) BE = 285.0 eV
SrF2:  Sr (3d)
C (1s) BE = 285.0 eV

Features Observed

  • xx
  • xx
  • xx

 Periodic Table 


 

Overlay of Sr (3d) Spectra shown Above

C (1s) BE = 285.0 eV

 

Chemical Shift between Sr and SrCO3:  -0.3 eV
 Chemical Shift between Sr and SrF2:  -1.3 eV

 

 Periodic Table 


 

Strontium Carbonate (SrCO3)
pressed powder

Survey Spectrum from SrCO3
Flood gun is ON, C (1s) BE = 285.0 eV
Sr (3d) Chemical State Spectrum from SrCO3
Flood gun is ON, C (1s) BE = 285.0 eV

 
O (1s) Chemical State Spectrum from SrCO3
Flood gun is ON, C (1s) BE = 285.0 eV
C (1s) Chemical State Spectrum from SrCO3
Flood gun is ON, C (1s) BE = 285.0 eV

 
Valence Band Spectrum from SrCO3
Flood gun is ON, C (1s) BE = 285.0 eV

 

 

Features Observed

  • xx
  • xx
  • xx

 Periodic Table 

 


 

 

Strontium Chemical Compounds

 

Strontium Titanate, SrTiO3

Survey Spectrum Sr (3d) Spectrum


.
C (1s) Spectrum O (1s) Spectrum


.
Valence Band Spectrum Ti (2p) Spectrum

 Periodic Table 



 

Quantitation Details and Information

Quantitation by XPS is often incorrectly done, in many laboratories, by integrating only the main peak, ignoring the Electron Loss peak, and the satellites that appear as much as 30 eV above the main peak.  By ignoring the electron loss peak and the satellites, the accuracy of the atom% quantitation is in error.

When using theoretically calculated Scofield cross-section values, the data must be corrected for the transmission function effect, use the calculated TPP-2M IMFP values, the pass energy effect on the transmission function, and the peak area used for calculation must include the electron loss peak area, shake-up peak area, multiplet-splitting peak area, and satellites that occur within 30 eV of the main peak.

 

Quantitation from Pure, Homogeneous Binary Compound
composed of Strontium – SrSO4

This section is focused on measuring and reporting the atom % quantitation that results by using:

  • Scofield cross-sections,
  • Spectra corrected to be free from Transmission Function effects
  • A Pass Energy that does not saturate the detector system in the low KE range (BE = 1000-1400 eV)
  • A focused beam of X-ray smaller than the field of view of the lens
  • An angle between the lens and the source that is ~55 deg that negates the effects of beta-asymmetry
  • TPP-2M inelastic mean free path values, and
  • Either a linear background or an iterated Shirley (Sherwood-Proctor) background to define peak areas

The results show here are examples of a method being developed that is expected to improve the “accuracy” or “reliability” of the atom % values produced by XPS.

 Periodic Table 


 

 

Strontium Alloys

   
XxCu XxCu
 Periodic Table   
XxCu XxCu

 

Copyright ©:  The XPS Library 

 



 

XPS Facts, Guidance & Information

 Periodic Table 

    Element Strontium (Sr)
 
    Primary XPS peak used for Peak-fitting: Sr (3d5/2)  
    Spin-Orbit (S-O) splitting for Primary Peak: Spin-Orbit splitting for “d” Orbital, ΔBE = 1.8 eV
 
    Binding Energy (BE) of Primary XPS Signal: 134 eV
 
    Scofield Cross-Section (σ) Value: Sr (3d5/2) = 2.99.     Sr (3d3/2) = 2.06
 
    Conductivity: Sr resistivity =  
Native Oxide suffers Differential Charing
 
    Range of Sr (3d5/2) Chemical State BEs: 132-136 eV range   (Sro to SrF2)  
Signals from other elements that overlap
Sr (3d5/2) Primary Peak:
  P (2p)
Bulk Plasmons:   ~xx eV above peak max for pure
Shake-up Peaks: xx
Multiplet Splitting Peaks:   xx

 

 

General Information about
XXX Compounds:
  xx  
Cautions – Chemical Poison Warning

xx 

Copyright ©:  The XPS Library 

 Periodic Table 



 

Information Useful for Peak-fitting Sr (3d5/2)

  • FWHM (eV) of Sr (3d5/2) for Pure Sro ~0.8 eV using 25 eV Pass Energy after ion etching:
  • FWHM (eV) of Sr (3d5/2) for SrCO3 ~1.5 eV using 50 eV Pass Energy  (before ion etching)
  • Binding Energy (BE) of Primary Signal used for Measuring Chemical State Spectra:  134.3 eV for Sr (3d5/2) with +/- 0.2 uncertainty
  • List of XPS Peaks that can Overlap Peak-fit results for Sr (3d5/2):  P (2p)

 Periodic Table 


 

General Guidelines for Peak-fitting XPS Signals

  • Typical Energy Resolution for Pass Energy (PE) setting used to measure Chemical State Spectra on Various XPS Instruments
    • Ag (3d5/2) FWHM (eV) = ~0.90 eV for PE 50 on Thermo K-Alpha
    • Ag (3d5/2) FWHM (eV) = ~1.00 eV for PE 80 on Kratos Nova
    • Ag (3d5/2) FWHM (eV) = ~0.95 eV for PE 45 on PHI VersaProbe
    • Ag (3d5/2) FWHM (eV) = ~0.85 eV for PE 50 on SSI S-Probe
  • FWHM (eV) of Pure Elements: Ranges from 0.4 to 1.0 eV across the periodic table
  • FWHM of Chemical State Peaks in any Chemical Compound:  Ranges from 1.1 to 1.6 eV  (in rare cases FWHM can be 1.8 to 2.0 eV)
  • FWHM of Pure Element versus FWHM of Oxide:  Pure element FWHM << Oxide FWHM  (e.g. 0.8 vs 1.5 eV, roughly 2x)
  • If FWHM Greater than 1.6 eV:  When a peak FWHM is larger than 1.6 eV, it is best to add another peak to the peak-fit envelop.
  • BE (eV) Difference in Chemical States: The difference in chemical state BEs is typically 1.0-1.3 eV apart.  In rare cases, <0.8 eV.
  • Number of Peaks to Use:  Use minimum. Do not use peaks with FWHM < 1.0 eV unless it is a or a conductive compound.
  • Typical Peak-Shape:  80% G: 20% L,   or Voigt : 1.4 eV Gaussian and 0.5 eV Lorentzian
  • Spin-Orbit Splitting of Two Peaks (due to Coupling):  The ratio of the two (2) peak areas must be constrained.
  • Constraints used on Peak-fitting: typically constrain the peak area ratios based on the Scofield cross-section values
  • Asymmetry for Conductive materials:  20-30% with increased Lorentzian %
  • Peak-fitting “2s” or “3s” Peaks:  Often need to use 50-60% Lorentzian peak-shape

Notes:

  • Other Oxidation States can appear as small peaks when peak-fitting
  • Pure element signals normally have asymmetric tails that should be included in the peak-fit.
  • Gaseous state materials often display asymmetric tails due to vibrational broadening.
  • Peak-fits of C (1s) in polymers include an asymmetric tail when the energy resolution is very high.
  • Binding energy shifts of some compounds are negative due to unusual electron polarization.

 Periodic Table 


 

Contaminants Specific to Strontium

  • Strontium develops a thick native oxide due to the reactive nature of clean Strontium .
  • The native oxide of Sr Ox is 8-9 nm thick.
  • Strontium thin films often have a low level of iron (Fe) in the bulk as a contaminant or to strengthen the thin film
  • Strontium forms a low level of Carbide when the surface is ion etched inside the analysis chamber

 

Commonplace Contaminants

  • Carbon and Oxygen are common contaminants that appear on nearly all surfaces. The amount of Carbon usually depends on handling.
  • Carbon is usually the major contaminant.  The amount of carbon ranges from 5-50 atom%.
  • Carbon contamination is attributed to air-borne organic gases that become trapped by the surface, oils transferred to the surface from packaging containers, static electricity, or handling of the product in the production environment.
  • Carbon contamination is normally a mixture of different chemical states of carbon (hydrocarbon, alcohol or ether, and ester or acid).
  • Hydrocarbon is the dominant form of carbon contamination. It is normally 2-4x larger than the other chemical states of carbon.
  • Carbonate peaks, if they appear, normally appear ~4.5 eV above the hydrocarbon C (1s) peak max BE.
  • Low levels of carbonate is common on many s that readily oxidize in the air.
  • High levels of carbonate appear on reactive oxides and various hydroxides.  This is due to reaction between the oxide and CO2 in the air.
  • Hydroxide contamination peak is due to the reaction with residual water in the lab air or the vacuum.
  • The O (1s) BE of the hydroxide (water) contamination normally appears 0.5 to 1.0 eV above the oxide peak
  • Sodium (Na), Potassium (K), Sulfur (S) and Chlorine (Cl) are common trace to low level contaminants
  • To find low level contaminants it is very useful to vertically expand the 0-600 eV region of the survey spectrum by 5-10X
  • A tiny peak that has 3 or more adjacent data-points above the average noise of the background is considerate to be a real peak
  • Carbides can appear after ion etching various reactive s.  Carbides form due to the residual CO and CH4 in the vacuum.
  • Ion etching can produce low oxidation states of the material being analyzed.  These are newly formed contaminants.
  • Ion etching polymers by using standard Ar+ ion guns will destroy the polymer, converting it into a graphitic type of carbon

 Periodic Table 


 

Data Collection Guidance

  • Chemical state differentiation can be difficult
  • Collect principal Sr (3d) peak
  • Long time exposures (high dose) to X-rays can degrade various polymers, catalysts, high oxidation state compounds
  • During XPS analysis, water or solvents can be lost due to high vacuum or irradiation with X-rays or Electron flood gun
  • Auger signals can sometimes be used to discern chemical state shifts when XPS shifts are very small

 Periodic Table 


 

Data Collection Settings for Strontium (Sr)

  • Conductivity:  Strontium readily develops a native oxide that is sensitive to Flood Gun – Differential Charging Possible – float sample recommended
  • Primary Peak (XPS Signal) used to measure Chemical State Spectra:  Sr (3d5/2) at 134eV
  • Recommended Pass Energy for Measuring Chemical State Spectrum: 40-50 eV    (Produces Ag (3d5/2) FWHM ~0.7 eV)
  • Recommended # of Scans for Measuring Chemical State Spectrum:  4-5 scans normally   (Use 10-25 scans to improve S/N)
  • Dwell Time:  50 msec/point
  • Step Size:  0.1 eV/point   (0.1 eV/step or 0.1 eV/channel)
  • Standard BE Range for Measuring Chemical State Spectrum:  125 – 145 eV
  • Recommended Extended BE Range for Measuring Chemical State Spectrum:  120 – 220 eV
  • Recommended BE Range for Survey Spectrum:  -10 to 1,100 eV   (above 1,100 eV there are no useful XPS signals, except for Ge, As and Ga, above 1100 is waste of time)
  • Typical Time for Survey Spectrum:  3-5 minutes for newer instruments, 5-10 minutes for older instruments
  • Typical Time for a single Chemical State Spectrum with high S/N:  5-10 minutes for newer instruments, 10-15 minutes for older instruments 

 Periodic Table 


 

Effects of Argon Ion Etching

  • Carbides appear after ion etching Sr and various reactive surfaces.  Carbides form due to the presence of residual CO and CH4 in the vacuum.
  • Ion etching can produce low oxidation states of the material being analyzed.  These are newly formed contaminants.
  • Ion etching polymers by using standard Ar+ ion guns will destroy the polymer, converting it into a graphitic type of carbon

 

 Periodic Table 

Copyright ©:  The XPS Library 


Gas Phase XPS or UPS Spectra


 

Chemical State Spectra from Literature
from Thermo website

Interpretation of XPS spectra

  • Sr3d region has well resolved spin-orbit components (Δ=1.76eV, ratio=0.690).
    • When multiple chemical states are present, apparent resolution of these two spin-orbit components is reduced. (Compare Sr zirconate with SrO/SrCO3 spectra below.)



End of File