ARTICLE IN PRESS

Applied Radiation and Isotopes 66 (2008) 1083–1090

www.elsevier.com/locate/apradiso

Laboratory-based X-ray phase-contrast imaging technique for material

and medical science applications
Yogesh S. Kashyap, P.S. Yadav, Tushar Roy, P.S. Sarkar, M. Shukla, Amar Sinha
Laser and Neutron Physics Section, Bhabha Atomic Research Centre, Mumbai 400085, India
Received 1 August 2007; received in revised form 13 December 2007; accepted 13 December 2007

Abstract

In-line X-ray phase-contrast imaging technique is an emerging method for the study of materials such as carbon fibers, carbon
composite materials, polymers, etc. Similarly this technique is also well suited for the imaging of soft materials such as tissues,
distinguishing between tumor and normal tissue. These represent the class of materials for which X-ray attenuation cross-section is very
small. Thus this method promises a far better contrast for low X-ray absorbing substances than the conventional radiography method.
We have set up an experimental facility using a combination of X-ray CCD detector and a microfocus X-ray source. This facility is
dedicated to micro-imaging experiments such as microtomography and high-resolution phase-contrast experiments. In this paper, the
results of X-ray phase-contrast imaging experiments are described.
r 2007 Elsevier Ltd. All rights reserved.

Keywords: X-ray phase contrast; CCD detector; Microfocus source

1. Introduction improved by the use of coherent X-ray sources. The use of

The use of X-rays in the visualization of intricate details
within the object has reached a very high stage of maturity.
A large variety of sophisticated X-ray sources and
detectors are routinely used for different imaging applica-
tions, be it medical or industrial. The growth in the use of
X-ray imaging techniques can be attributed to the high
penetrating power of these radiations, which provides
information about internal details of the object. The
contrast in conventional absorption-based X-ray imaging
techniques depends upon the object thickness and its
atomic number. Hence, the contrast of the absorption-
based X-ray images decreases when either atomic number
of the object or object thickness decreases. In fact, this is
one of the major drawbacks of the conventional absorp-
tion-based X-ray imaging techniques. However, recent
investigations (Davis et al., 1995; Snigirev et al., 1995) have
proved that the contrast in the imaging process can be
Corresponding author.
E-mail address: yogesh78@barc.gov.in (Y.S. Kashyap).
coherent X-rays gives the opportunity to utilize both the
amplitude and phase information. This in turn contributes
to the improved contrast in the imaging process. These new
methods are based on the fact that just like any other wave,
propagation of X-rays through any material medium
introduces a change of phase across X-ray wavefront. This
change of phase across the wavefront depends upon the
spatial frequency of the feature, electron density of the
material, etc. Hence, the phase variation across the
wavefront carries detailed information about the different
features of the object. These phase gradients in the
wavefront can be made to manifest themselves as redis-
tribution of X-ray intensity in the imaging plane using
suitable techniques. This results in contrast enhancement
across the edges or boundaries for weakly absorbing
materials. As the contrast does not directly depend upon
the energy deposited in the object, the dose delivered to the
sample can be reduced. In this sense, these new imaging
technique can be truly considered as ‘‘non-destructive
imaging’’ techniques. These new imaging modalities are
collectively called ‘‘phase-sensitive imaging techniques’’.

0969-8043/$ - see front matter r 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.apradiso.2007.12.008
 ARTICLE IN PRESS
1084 Y.S. Kashyap et al. / Applied Radiation and Isotopes 66 (2008) 1083–1090
Fig. 1. Photograph of experimental set-up.
Fig. 2. Schematic representation of X-ray CCD used for phase-contrast
experiments.
They have been implemented in following three modes
depending upon the source characteristics:
1. Interferometric technique,
2. Diffraction enhanced imaging,
3. In-line phase contrast imaging.
Diffraction enhanced and interferometry-based imaging
methods have stringent requirement in terms of high
coherence of source, use of silicon crystals, low-vibration
of table, etc. for implementing these techniques (Chapman
et al., 1997; Momose et al., 1996). Since only synchrotron
sources satisfy the demanding requirement of the above
techniques, they have been implemented only in selected
facilities on the synchrotron sources. In contrast with the
above two methods, in-line phase-contrast technique is
most suitable for laboratory-based experiments as the only
stringent parameter is the spatial coherence of the source
(Wilkins et al., 1996). Hence among all the above
techniques, in-line phase-contrast imaging techniques
appear to be most suitable for clinical and industrial
applications. Moreover, it is easy to set up this facility for
specific imaging applications due to its low cost and ease of
implementation (Donelly et al., 2006). Depending upon the
source size, detector resolution and feature size, one has to
optimize different parameters to get good contrast images.
The effect of the polychromaticity of the source has been
investigated earlier, and was shown that (Pogany et al.,
1997; Yadav et al., 2006; Olivo and Speller, 2006) it results
in smearing of the higher-order interference fringes, while
the information about primary maxima can be easily
located. Hence, this technique can be easily implemented
using microfocus X-ray source without any need of
monochromatizing X-ray spectrum using an optical
element. In this paper we would like to discuss the results
of our recent investigation using ‘‘in-line phase contrast’’
technique.
2. Basics of coherent X-ray imaging
The refractive index of the materials in the X-ray region
can be expressed as n(l) ¼ 1d(l)ib(l), where l is the
X-ray wavelength (Xizeng Wu and Hong Liu, 2003;
Peterzol et al., 2005; Cowley, 1975). The image contrast
in conventional absorption radiography is obtained from
variations in the X-ray absorption (imaginary term b(l)) of
the object. Although this approach works well in distin-
guishing between hard (high Z) and soft (low Z) materials,
it fails in imaging soft materials such as polymers,
carbon–fiber composites and soft tissues or its distinguish-
ing between different soft materials. The contrast produced
in X-ray phase-contrast imaging depends not only on
differences in absorption, but also on differences in the real
part (d(l)) of the refractive index. The real part of the
 ARTICLE IN PRESS
Y.S. Kashyap et al. / Applied Radiation and Isotopes 66 (2008) 1083–1090 1085

refractive index is 1000 times greater than absorption
contrast in 10–100 keV range for many of the low Z
materials. This leads to better sensitivity and contrast as
compared to the conventional absorption radiography.
Hence, the objects with low-X-ray absorption coefficient can
also be studied using phase-contrast imaging techniques.
The phase contrast has been exploited in the visible regions
long ago. However, lack of availability of highly coherent
X-ray source, was a major hurdle for implementing this
technique in the X-ray region. Synchrotron sources provide
X-ray beam with high degree of coherence and therefore
they are the source of choice for implementation of these
techniques (Arfelli et al., 1998; Cloetens et al., 1997).
However, the use of these facilities is not only expensive, but
also not suited for day-to-day experimentation. With the
availability of microfocus-X-ray sources (focal spot-
o10 mm), we can also implement in-line phase-contrast
technique on a laboratory scale. This technique has potential
applications in the medical field as in early detection of
tumor, animal studies and also in material science field as in
the study of carbon composite materials, study of polymer-
based materials and so forth. Hence, this technique has a
great potential in the study of new class of materials
belonging to the industrial and medical science fields.

3. Experimental setup

Fig. 3. Image of carbon-composite fiber showing the internal layered

Fig. 1 shows the photograph of the experimental setup.
structure of the fiber. We have used microfocus X-ray source (160 kV, 1 mA) and

Fig. 4. (a) X-ray phase radiograph of a carbon composite sample used in aircraft industry and (b) phase-contrast image of the red-circled portion revealing
the structures of the fibers.

Fig. 5. (a) Images of polished, and (b) unpolished diamond showing phase enhancement at the edges.
 ARTICLE IN PRESS
1086 Y.S. Kashyap et al. / Applied Radiation and Isotopes 66 (2008) 1083–1090
fiber optically bonded CCD camera in all our studies.
Fig. 2 shows schematic layout of X-ray detector system.
The X-ray detectors consist of Gd2O2S:Tb phosphor layer
deposited on an entrance window of 4:1 glass fiber-optic
Fig. 6. Alumina microspheres coated with pyrocarbon of 40 mm thickness.
Fig. 7. Alumina coated with pyrocarbon of 40 mm thickness with tube
defocused.
Fig. 8. Alumina coated with pyrocarbon of 40 mm thickness with 2  2
binning of detector
taper. The taper is coupled to a front-illuminated CCD.
This CCD has active pixel pitch of 13.5 mm  13.5 mm and
1024  1024 pixels. The thick fiber-optic taper increases the
detector area and protects the CCD chip from direct
impact of X-rays. The microfocus X-ray source has focal
spot as low as 5 mm, when target power is less than 4 W.
The tube is demountable type so that target can be replaced
with some other suitable material target. This is transmis-
sion type of tube with emission in full 1801 cone, which can
Fig. 9. Zirconia microspheres of 500 mm diameter coated with pyrocar-
bon. This figure shows the non-uniformity of the coated pyrocarbon.
Fig. 10. Zirconia microspheres of 500 mm diameter with uniform
pyrocarbon coating of thickness 40–50 mm.
 ARTICLE IN PRESS
Y.S. Kashyap et al. / Applied Radiation and Isotopes 66 (2008) 1083–1090 1087

be delimited by the use of suitable collimator for reducing
unwanted scattered X-ray radiation.
4. Experimental results
4.1. Phase contrast images of carbon fiber
We have carried out X-ray radiography of carbon
composite fiber. The layered structure of fiber has been
brought out very clearly in Fig. 3. The strength and
stiffness of the fiber may change due the presence of micro-
defects or porosity in the bulk of the material. This
technique when used in combination with tomography
technique can give detailed information about the distribu-
tion of the micro-defects or porosity. Thus, this technique
can be a valuable tool in characterization of these
composite fibers. The fiber material, which is used in
design of the aircrafts was also imaged. Fig. 4 shows the
internal structure of the fiber in phase radiography mode.
4.2. Phase image of polished and raw diamond
We have collected images of raw and polished diamonds.
The polished diamond was taken from a DAC cell and
appears to be defect, free while the raw diamond (before
polishing) has some defects or scratches on its surface. Fig.
5(a,b) clearly shows the potential of phase imaging for
detecting flaws with high contrast. The process of getting
defect-free diamond from raw diamond is very laborious
and difficult. X-ray phase-contrast imaging can be easily
combined with tomography to find out the distribution of
flaws within the raw diamond. This data can be very
valuable for the automization of the diamond-cutting
process so as to optimize the cutting process.
4.3. Phase images of alumina microspheres coated with
pyrocarbon
We have done some studies on alumina sample coated
with pyrocarbon. The estimated coating thickness of
pyrocarbon on alumina microspheres was 40–60 mm. The
phase-enhancement effects can be seen at both the
alumina–air and alumina–carbon interfaces in Fig. 6. This
technique can be a valuable tool in detecting the interfaces
in multi-layerd or multi-coated structures. Fig. 7 shows
that as the focal spot of the tube increases the contrast
degrades. This clearly brings out the importance of high
spatial coherence for these kinds of experiments. Finally,

Buffer High dense

(Low dense PyC) PyC

ZrO2

Fig. 11. Phase-contrast X-ray image of pyrocarbon-coated zirconia micro-spheres with diameter of 500 mm showing the zirconia (ZrO2) low-dense PyC
and high-dense PyC.

I-PyC

Buffer

ZrO2

ZrO2

SiC

O-PyC

Fig. 12. Phase-contrast X-ray image of zirconia microspere showing the kernel (ZrO2), the buffer, the inner PyC, the SiC and the outer PyC.
 ARTICLE IN PRESS
1088 Y.S. Kashyap et al. / Applied Radiation and Isotopes 66 (2008) 1083–1090

Fig. 13. (a) Image of the mouse taken in phase-radiography mode; (b) image of the tumor growth just below the skin of the same mouse.

we have collected data by rebinning the CCD detector in

2  2 mode i.e. by decreasing its resolution. Fig. 8 shows
that as the detector-pitch increases the contrast decreases.
The contrast at alumina–carbon interface has degraded
more than the contrast at carbon–air interface. Thus a
combination of high-resolution detector and low-focal spot
source is a must for detecting the phase-enhancement
effects.

4.4. Phase imaging of zirconia microspheres coated with

pyrocarbon

We have carried out set of experiments for estimating Fig. 14. Image of a bee taken in phase radiography mode.
pryrocarbon (PyC) coating thickness on carbon-coated
zirconia microspheres. These kind of coated materials are
potential candidates for future generation of high-tem-
perature nuclear reactor. Zirconia microspheres of ap-
proximately 500 mm diameter were deposited with 40–50-
mm-thick layer of pyrocarbon in an electrically heated
graphite vessel under different conditions. These samples
were imaged at various magnifications to study the carbon
coating on zirconia microspheres. The non-uniformity of
the carbon coating can be easily seen in Fig. 9. Fig. 10
shows the possibility of obtaining uniform coating of
40–50 mm on surrogate zirconia microspheres. The varia-
tion in coating thickness is attributed to changes in control
parameters during the deposition process. This technique,
thus, becomes a valuable tool for estimating the right
parameter for correct thickness deposition. Fig. 11 shows
the images of zirconia microspheres having two different
layers of carbon coating. The average coating thickness of
low density (1.1 g/cm3) PyC and high density (1.9 g/cm3) on
ZrO2 microspheres was determined to be 132 and 66 mm,
respectively. Fig. 12 shows the phase image of the zirconia
(ZrO2) coated by a low density (1.1 g/cm3) PyC layer
(buffer) followed by three successive isotropic layers: an
inner dense (1.9 g/cm3) PyC layer (I–PyC), a silicon carbide Fig. 15. Radiographic image of the internal structure of a dead
layer (SiC) having density (3.18 g/cm3) and an outer dense cockroach.
 ARTICLE IN PRESS
Y.S. Kashyap et al. / Applied Radiation and Isotopes 66 (2008) 1083–1090
Fig. 16. (a) and (b) Images of dry leaves of two different trees. The internal tubular structure is very clearly visible in these figures.
(1.9 g/cm3) PyC layer (O–PyC). The average coating
thickness of low dense PyC (buffer), I–PyC, SiC and
O–PyC on ZrO2 microspheres has been determined to be
99, 53, 69 and 95 mm, respectively. These kinds of multiple
coating are important for development of the high-
temperature reactor fuel particles (Nozawa et al., 2007).
Hence, X-ray-based phase-contrast technique offers attrac-
tive possibilities for non-destructive characterization of
these advanced reactor fuel particles.
4.5. Phase images of soft (biological) materials
We have also done studies on some biological samples. A
mouse having tumorous growth just beneath the skin was
imaged using this technique. The picture shows the
presence of tumor very clearly (Fig. 13(a) ). We have also
tried to see other details of the mouse body (Fig. 13(b)).
Some other soft materials such as images of dead bee, dead
cockroach and leaves are also shown in Figs. 14–16. It can
be seen that the details seen in these images are far better
than those obtained from radiography experiments. Hence,
X-ray phase-imaging techniques can become a valuable
tool for clinical application such as mammography.
Similarly, this technique can also be used for laboratory-
based plant and animal studies.
5. Conclusion
We have set up a low-cost experimental facility using a
microfocus X-ray source and CCD detector for X-ray
micro-imaging experiments. We used this system for
obtaining phase-contrast images of various samples chosen
from material science and medical science field. These
results show that the X-ray phase-enhancement technique
can be used to investigate a large class of important
materials. Our future efforts will concentrate on the
development of phase microtomography for industrial
and medical applications.
1089
Acknowledgments
We are highly indebted to K. Dasgupta and D.
Sathiyamoorthy of Powder Metallurgy Section of Bhabha
Atomic Research Centre, India, for providing support for
carrying out some of the experiments.
References
Arfelli, F., Assante, M., Bonvicini, V., Bravin, A., Cantatore, G., Castelli,
E., Dalla Palma, L., Di Michiel, M., Longo, R., Olivo, A., Pani, S.,
Pontoni, D., Poropat, P., Prest, M., Rashevsky, A., Tromba, G.,
Vacchi, A., Vallazza, E., Zanconati, 1998. Low-dose phase contrast X-
ray medical imaging. Phys. Med. Biol. 43, 2845.
Chapman, D., Thomlinson, W., Johnston, R.E., Washburn, D., Pisano,
E., Gmur, N., Zhong, Z., Menk, R., Arfelli, F., 1997. Diffraction-
enhanced X-ray imaging. Phys. Med. Biol. 42, 2015.
Cloetens, P., Pateyron-Salomé, M., Buffière, J.Y., Peix, G., Baruchel, J.,
Peyrin, F., Sclenker, M., 1997. Observation of microstructure and
damage in materials by phase-sensitive radiography and tomography.
J. Appl. Phys. 81, 5878.
Cowley, J., 1975. Diffraction Physics. North-Holland, Amsterdam.
Davis, T.J., Gao, D., Gureyev, T.E., Stevenson, A.W., Wilkins, S.W.,
1995. Phase-contrast imaging of weakly absorbing materials using
hard X-rays. Nature 373, 595.
Donelly, E.F., Lewis, K.G., Wolske, K.M., Pickens, D.R., Price, R.R.,
2006. Characterization of the phase-contrast radiography edge-
enhancement effect in a cabinet X-ray system. Phys. Med. Biol. 51,
21.
Momose, A., Takeda, T., Itai, Y., Hirano, K., 1996. Phase-contrast X-ray
computed tomography for observing biological soft tissues. Nat. Med.
2, 473.
Nozawa, T., Snead, L.L., Katoh, Y., Miller, J.H., 2007. Shear properties
at the PyC/SiC-interface of a TRISO-coating. J. Nucl. Mater. 371, 304.
Olivo, A., Speller, R., 2006. Experimental validation of a simple model
capable of predicting the phase contrast imaging capabilities of any X-
ray imaging system. Phys. Med. Biol. 56, 3015.
Peterzol, A., Olivo, A., Rigon, L., Pani, S., Dreossi, D., 2005. The effects
of the imaging system on the validity limits of the ray-optical approach
to phase contrast imaging. Med. Phys. 32, 3617.
Pogany, A., Gao, D., Wilkins, S.W., 1997. Contrast and resolution in
imaging with a microfocus X-ray source. Rev. Sci. Instrum. 68,
2774.
 ARTICLE IN PRESS
1090 Y.S. Kashyap et al. / Applied Radiation and Isotopes 66 (2008) 1083–1090

Snigirev, A., Snigireva, I., Kohn, V., Kuznetsov, S., Scelokov, I., 1995.
On the possibilities of X-ray phase contrast microimaging by
coherent high-energy synchrotron radiation. Rev. Sci. Instrum. 66,
5486.
Wilkins, S.W., Gureyev, T.E., Goa, D., Pogany, A., Stevenson, A.W.,
1996. Phase-contrast imaging using polychromatic hard X-rays.
Nature 384, 335.
Wu, X., Liu, H., 2003. Clinical implementation of X-ray phase-contrast
imaging: theoretical foundations and design consideration. Med. Phys.
30 (8), 2169.
Yadav, P.S., Kashyap Yogesh, Sarkar, P.S., Sinha Amar, Godwal, B.K.,
2006. Study of phase contrast imaging for carbon fiber, polystyrene
and lung tissue using monochromatic and polychromatic X-ray
sources. Nucl. Instrum. Methods A 564, 496.