Carbon coating objects for SEM/EDS

Suzanne Davis <davissl [at] umich__edu> writes

>The geologists usually coat their samples for SEM with carbon. As I
>understand it, this helps prevent stray X-rays from nearby areas
>from interfering with analysis of the particular spot you wish to
>analyze. We are placing the entire seal (they are quite small) in
>the SEM chamber and our analyses would be more reliable if we could
>coat the seals themselves with the carbon. Does anyone have
>experience doing this?
>...
>... We coated a glass slide this way, and I find that the
>graphite is very easy to remove with ethanol:de-ionized water in a
>1:1 ratio. I feel fairly confident that the coating can be applied
>and removed without difficulty, but I would be interested to hear
>from the conservation-science community any opinions/experiences in
>using a coating on small objects for SEM.

Unless you don't mind carbon stuck in the nooks and crannies of your
stone seals, I would recommend not carbon coating stone objects.

The particles of carbon that come off the rod will enter into each
and every fracture and pore space on the surface of your seals!  If
your stone seals do not have the absolutely smooth surface that
glass does, then removing all of the carbon will be next to
impossible.  Even if your seals are obsidian, they may have
crystalline inclusions or vesicles that can get carbon stuck in
them.  It is possible to see the carbon in a low powered light
microscope, and this remaining small amount of carbon can change the
apparent color of the object.

It is possible to quantitatively analyze un-coated samples in a
standard SEM. You can place the seals onto a carbon base and secure
them with carbon putty or carbon tape. There are many carbon-based
products available for use in SEM studies and your geologist
colleagues should have information on where to obtain these.   The
trick is to use enough carbon adhesive to ensure that a charge does
not build up from the electron beam.  If a charge does build up,
then the electron beam will be deflected away from the area you want
to analyze.

An article by Ian Freestone and Heike Bronk describes SEM/EDS
analysis of manufactured glass without carbon coating the samples:

    Ian Freestone and Heike Bronk.
    "A Quasi Non-destructive Microsampling Technique for the
    Analysis of Intact Glass Objects By Sem/edxa"
    Archaeometry, Nov. 2001, 43 (4), pp. 517-527

    Abstract: A novel sampling technique for the analysis of glass
    is described, which involves the removal of-minute particles
    from an object with a diamond-coated file, followed by
    energy-dispersive X-ray analysis in the scanning electron
    microscope (SEM/EDXA). The particles are fixed to adhesive
    carbon discs and carbon coated without grinding or polishing.
    Mean compositions are determined for 10 arbitrarily selected
    particles above a-minimum grain size of 150-mm and normalized to
    totals of 100%. Tests were carried out on two standard
    soda-lime-silica glasses of well-characterized composition,
    using two files of different grade. The analyses showed good
    agreement with the accepted values of all elements. Although the
    precision is somewhat reduced, this highly portable and quasi
    non-destructive microsampling procedure provides almost the same
    information as that gained from samples embedded and polished in
    the normal way. Its application is thought to be especially
    useful for the investigation of intact glass objects and ceramic
    glazes.

Good luck,

Colleen Stapleton
Geologist
Archaeological Scientist
Asst. Prof.
Mercer University
3001 Mercer University Dr.
Atlanta, Georgia 30341


                                  ***
                  Conservation DistList Instance 17:53
                Distributed: Wednesday, February 4, 2004
                       Message Id: cdl-17-53-003
                                  ***