JEOL 8900 Electron Microprobe Capabilities

The electron microprobe at Binghamton University is fully computer automated and has the following major specifications:
  • 4 wavelength-dispersive spectrometers (WDS): crystals = STE, LDE1, TAPJ, PET, LIF
  • 1 energy-dispersive spectrometer (EDS): detector = Si(Li), Be window
  • Electron beam imaging: Secondary-electron imaging
  • Back-scattered imaging: Compositional (mean Z) and Topographic modes
  • Light microscopy: reflected and transmitted (polarized) light
  • Line analyses: (beam scan, X-Y stage scan):
  • Multi-element area maps (beam scan, X-Y stage scan)
  • Image analysis: Minimum pixel size for line and stage scans is 0.02 micron. Grey level or false-color digital enhancement
  • Data manipulation:
    • Analyses: Matrix corrections using ZAF, CITZAF, and Bence-Albee schemes.
    • Data and Image output: Standard photographic media, magnetic media, color thermowax printer, or direct internet transfer to your e-mail address.
    • Sample size: 100 x 100 x 50 (H) mm (max.), analyzable area of 90 x 90 mm (max.)
    • Best designed to handle conventional petrographic thin sections and 1-inch round disks.

Is the Electron Microprobe the right instrument for your project?

What the JEOL 8900 Electron Microprobe Can Do:

  • Quantitative analysis of elements from Fluorine to Uranium
  • Semi-quantitative analysis of elements from Boron to Oxygen
  • Digital imaging using Back-scattered and Secondary Electrons – images generally good for particles down to 0.1 µm in width
  • Digital X-ray mapping
  • Four-spectrometer WDS system, Be-window EDS system
  • Accept standard petrographic thin sections, 1-inch diameter, and 1.25-inch diameter metallurgical mounts

What the JEOL 8900 Electron Microprobe Cannot Do:

  • Cannot routinely analyze particles < 5 µm in width or depth
  • Cannot detect concentrations of elements below about 500 ppm (0.05 wt%)
  • Cannot analyze H, He, Li, or Be because H and He produce no characteristic X-rays and because Li and Be have extremely low X-ray fluorescent yields
  • Cannot do “thin film” analyses
  • Cannot image features at the “nano” scale, i.e. 1-100 nm (10-1000 Å)
  • Cannot analyze highly volatile or vacuum-sensitive samples
  • Cannot analyze large samples (greater than 3/4-inch cube)

Optimal sample conditions (solids) for microprobe analyses:

  • Hard (able to be polished)
  • Flat (polish surface to ~0.5 µm grit diameter)
  • Thick (>30 µm thick, thicker than a normal rock thin section)
  • Conducting – but we routinely carbon-coat insulating materials

For instrument use and availability:

David Collins

E-mail address:

David M. Jenkins
Email address:

Telephone: (607) 777-7623


Typical Microprobe Mineral Analyses*

Oxide (wt %)
Kakanui Hornblende Observed
(USNM 14365)
Durango Apatite Observed**
(USNM 104021)Reported**
SiO2 40.77 40.37 0.33 0.34
Al2O3 14.28 14.90 0.0 0.07
TiO2 4.73 4.72 nd nd
MgO 12.64 12.80 0.03 0.01
FeO 11.09 10.92 0.07 0.05
MnO 0.09 0.09 nd nd
CaO 9.86 10.30 53.42 54.02
Na2O 2.68 2.60 nd 0.23
K2O 2.09 2.05 nd 0.01
P2O5 nd 0.0 41.62 40.78
F nd nd 3.23 3.53
Cl nd nd 0.27 0.41
-O=F,Cl     -1.42 -1.58
Total 97.54 98.75 97.54 97.88

* nd = not determined

Standards used are: Si=SiO2, Al, K=orthoclase, Ti=TiO2, Mg=MgO, Fe=hematite, Mn=spessartine, Ca=diopside, Na=albite (Amelia), P=apatite (Wilberforce), F=fluorite, Cl=Kcl.

** Does not include 2.3 wt % total of REE, Na, S, etc.

Other standards available: (1) oxide/metal standards from C. M. Taylor Corp., (2) selected mineral standards from the Smithsonian Institution, and (3) selected glass standards from the Carnegie Institution of Washington, Geophysical Laboratory.

Last Updated: 1/6/15