In general, biological samples, like insect specimen have a low absorption coefficient due to its high water content, but the skeleton has substantially higher absorption. These two phases, "soft" and "hard" can usually be separated well. Separation of "soft" phases is more difficult.

First of all, the target should be selected to have a spectrum in the similar range, that is a material with low atomic number, like Molybdenum.

For the parameters: Choose a low X-ray voltage, and a high current for good image quality. We used values like 50kV and 300µA with our system (GE Nanotom) for imaging flies or bees.

In general, dried insects with well defined chitinous exoskeletons will have greater contrast for the skeletal structures. This is due to the low absorptivity of air relative to the skeleton or tissue. The problem with dried specimens is the distortion of normal anatomy. X-ray CT have little native soft tissue contrast and discerning soft tissue structures without any contrast producing additives is difficult. Low energy scans can give better internal structure contrast (20 - 40 kV) but at the cost of very long scan times. Remember that kV sets the contrast level for the amount of tissue and the integral of the I, current, and Exposure,s, sets the image SNR level.

If possible is can be helpful to introduce an external contrast agent to produce additional absorption in image structures of interest. We have used a variety of materials to do this. This can be fed to live insects in the form of lead sugar, Lead(II) acetate. Which concentrates in metabolically active tissues. Post-morteum labeling agents like phosphotungstic acid will label fibrin, collagen and connective tissues. Other agents such as osmium tetroxide will stain lipids.

The use of agents allows a more complete approach to discerning soft tissue structures in specimens such as insects.

For a better contrast of fresh samples try to maintain the following conditions: humidified environment, long scan time, sample mechanical fixation and a long distance between the source, detector and the specimen (about 10cm). Under these conditions no fixation or staining is necessary for soft tissues imaging.

At the Manchester X-ray Imaging Facility we have found that staining is certainly the easiest way of increasing contrast in biological specimens and allowing very short scans. However the goal in XCT is usually to avoid staining and image in-situ. In this case I would recommend using a low kV as suggested by H. Morris above, although there are a number of ways of decreasing scan times depending on your system. In the case of Nikon instruments find the correct X-ray target for the specified energy this will ensure a better photon count and signal to noise ratio. It may also be possible to adjust the filament settings – this is usually hidden in the advanced settings and will effect filament lifetime. Be careful of increasing the current since this can adversely affect the resolution limit assuming you are using no optical magnification (as in the case of Xradia instruments). Lastly I have found that the contrast will be superior on a 16bit detector than a 14 bit detector assuming the detector is not damaged.

Thank you very much for answers.

I read also something about iodine or iodized solution. Do you know how to proceed, for exemple with dry insects.

Does it exist any text book or synthesis about this topic ?

Romain - You will find some pointers from the following two links -

First Brian D. Metscher has a nice review of different staining techniques in his paper available from BMC Central - "MicroCT for comparative morphology: simple staining methods allow high-contrast 3D imaging of diverse non-mineralized animal tissues" - http://www.biomedcentral.com/1472-6793/9/11 .

Second, some specific details about sample preparation are available from the Abstracts of the SkyScan Annual Users meeting. This meeting is quite a find for MicroCT techniques in a variety of fields. You may consider attending. The following link is from the Zoology and Marine presentations from a recent conference - http://www.skyscan.be/company/um_abstracts/zoology&marine.pdf . Faulwetter et al. discuss iodine staining methods for soft marine tissues (100% ethanol, 1% iodine). Alba-Tercedor & Caparros discuss X-ray CT of coleoptera beetles using a variety of techniques and describe in detail the drying and mounting techniques used.

These two links (the Skyscan link has many proceeding papers in it on different topics) will get you started well on your scanning of your specimens.

Let us hear how your experiments turned out. If you have any neat image results share them.

Weakly absorbing materials are difficult to form high contrast images. At the same time, these samples produce significant phase shifts in the X-ray beam. The phase variation is caused by variation of refractive index for the X-ray in the object. Therefore the internal structure of the object can be mapped as a distribution of refractive index for X-ray. In Additional, phase retrieval algorithms can be implemented in order to retrieve the phase distribution of the sample and improve significantly the CNR.