
Osmium #
Orbitals and Energies #
Note – these are listed in BINDING ENERGY
Cu 2p ≈ 933 eV
Cu 2s ≈ 1096 eV
Cu 3s ≈ 120 eV
Cu 3p ≈ 74 eV
Theory and Background #
-
Asymmetry in metallic lineshapes
-
Like Pt and Ir, metallic Os exhibits a strong Doniach–Šunjić asymmetry in its 4f peaks.
-
This makes fitting tricky — naïve Gaussian–Lorentzian fits tend to misassign oxidation states.
-
-
Multiple oxidation states overlapping
-
Os is “multi-valent” in oxides; you can have Os(IV), Os(VI), Os(VIII) present simultaneously.
-
Their 4f signals overlap closely (often <1 eV apart). Deconvolution is nontrivial and sometimes ambiguous.
-
-
Close overlap with Auger features
-
In certain cases, Os Auger lines can fall close to 4f binding energies, complicating background subtraction.
-
-
Surface oxidation / instability
-
Os is notorious for forming OsO₄ under even mild oxidative conditions.
-
This changes the XPS spectrum dramatically, and OsO₄ is volatile (and extremely toxic). Samples can “age” under air and show higher BE shoulders.
-
-
Peak width differences
-
Higher oxidation states of Os tend to broaden the 4f peaks compared to metallic Os.
-
This can look like unresolved multiplets, but it’s mostly due to chemical disorder and shake-up contributions.
-
Experimental Advice #
-
Minimise air exposure. Prepare samples in glovebox or under inert gas if possible; transfer under vacuum or with a sealed capsule.
-
Avoid heating in air. Thermal treatments can produce OsO₄ and change surface composition.
-
Surface cleanliness: Os is reactive — adventitious species (C/O) and thin oxide layers form quickly. Document handling history.
-
If you must sputter-clean, be careful: Common Ar⁺ sputtering tends to reduce Os oxides and can change oxidation states. Prefer low-energy sputtering or cluster-ion sputtering (gas-cluster, if available) and validate effects on reference samples first.
Data Analysis Guidance #
-
Always use asymmetric line shapes (e.g. Doniach–Šunjić) for metallic Os.
-
Constrain the 4f spin–orbit splitting (~2.9 eV) and area ratio (4f₇/₂ : 4f₅/₂ ≈ 4:3).
-
Be cautious when assigning oxidation states — Os(IV) vs Os(VI) differences can be <1 eV.
-
Cross-check with O 1s region (lattice vs oxide oxygen) and possibly XAS if available.


