Monochromatic computed tomography using laboratory-scale setup proof-of-concept Ari-Pekka Honkanen12and Simo Huotari2

2025-05-02 0 0 4.25MB 6 页 10玖币

侵权投诉

Monochromatic computed tomography using laboratory-scale setup:

proof-of-concept

Ari-Pekka Honkanen∗1,2and Simo Huotari2

1Comprehensive Cancer Center, Helsinki University Hospital, P.O. BOX 180, FI-00029 HUS, Finland

2Department of Physics, University of Helsinki, P.O. Box 64, FI-00014 Helsinki, Finland

Abstract— In this article, we demonstrate the viability

of highly monochromatic full-ﬁeld X-ray absorption near

edge structure based tomography using a laboratory-scale

Johann-type X-ray absorption spectrometer based on a

conventional X-ray tube source. In this proof-of-concept, by

using a phantom embedded with elemental Se, Na2SeO3, and

Na2SeO4, we show that the three-dimensional distributions of

Se in different oxidation states can be mapped and distinguished

from the phantom matrix and each other with absorption

edge contrast tomography. The presented method allows for

volumetric analyses of chemical speciation in mm-scale samples

using low-brilliance X-ray sources, and represents a new

analytic tool for materials engineering and research in many

ﬁelds including biology and chemistry.

I. INTRODUCTION

Computed tomography (CT) is a widely used

non-destructive method to investigate the three dimensional

structure of matter. The clinical CT instruments and a

major fraction of laboratory-scale setups are based on

polychromatic broad-bandwidth beam produced with

conventional X-ray tubes. While this produces a sufﬁciently

high ﬂux of photons for imaging purposes, polychromaticity

of the beam has its own drawbacks such as beam-hardening

artifacts and insensitivity to the chemical composition of the

imaged object. Some amount of chemical contrast can be

achieved by dual-energy imaging but the information can be

used to separate elements at best into two or three groups

based on their atomic number [1]. The lack of elemental

sensitivity is a signiﬁcant shortcoming from the viewpoint

of materials research as the properties of material rely not

only on its elemental composition and distribution but also

the chemical speciation of the elements.

These limitations can be overcome with highly

monochromatic and tunable X-ray beams such as ones

produced with synchrotron and X-ray free electron laser

lightsources. One such approach is K-edge subtraction

imaging, which has been utilized for example to map the

ventilation of airways in lungs during an asthma attack

using the xenon gas K-edge absorption imaging [2,3].

By adjusting the photon energy of an X-ray beam with

.eV resolution one can even separate the X-ray signals

of different chemical species which in turn can be utilized

to map the distribution of the species in the sample. This

method, known as x-ray absorption near-edge spectroscopy

∗ari-pekka.honkanen@hus.fi

(XANES), offers a non-destructive tool for the analysis

of the chemistry of a given element, most importantly

its oxidation state and local atomic coordination [4]. It

has shown success in being utilized as a contrast method

for full-ﬁeld tomography in numerous materials research

applications such as investigating nano and mesoscale

chemical compositions and phase transitions in battery

materials [5]–[7], degradation and inactivation of catalyst

materials [8,9], and heterogeneity of defect-engineered

metal-organic framework crystals [10]. It has also been

demonstrated that a similar idea can be applied to inelastic

X-ray scattering (X-ray Raman spectroscopy) to obtain

tomographic data on the chemical state of low-Z elements

to e.g. spatially distinguish sp2and sp3bonds in carbon

materials [11].

The aforementioned techniques require a highly brilliant,

energy-tunable X-ray light source, such as a synchrotron

light source, which limits their applicability in the laboratory

scale. However, due to high demand and scarcity of

beamtime at large scale synchrotron and X-ray free

electron laser lightsources, the laboratory-scale X-ray

spectrometry has experienced a renaissance in the recent

years. Despite their orders of magnitude lower photon output,

laboratory-scale instruments have proven to be a viable

alternative to large-scale facilities in many applications

[12]–[17].

In our previous work [18], we demonstrated chemically

sensitive 2D-imaging using a Johann-type X-ray absorption

(2D-XANES) spectrometer based on a conventional X-ray

tube as presented in Fig. 1. The polychromatic beam of

the primary source is directed at a spherically bent crystal

analyser which monochromatises and refocuses the beam at

on the Rowland circle which acts as a secondary source.

The sample and the imaging detector are set downstream

from the secondary focus. Chemical sensitivity is obtained by

adjusting the energy of the diffracted photons and recording

the spatially resolved changes in the attenuation coefﬁcients.

In this work we develop the imaging capabilities of such

a laboratory setup further by demonstrating 3D imaging

with chemical contrast (3D-XANES) using the low-brilliance

x-ray source. We prepared a PMMA phantom (Fig. 2) which

was embedded with Se in different chemical states and show

that mapping the 3D spatial distribution of different chemical

species is viable using the setup described.

arXiv:2210.00804v1 [physics.ins-det] 3 Oct 2022

Fig. 1: The schematic drawing of the XAS-CT setup. Polychromatic

X-rays produced by the X-ray tube are monochromatised with

the spherically bent crystal analyser. The sample to be imaged

is illuminated by the monochromatised beam by moving it away

from the Rowland circle so that the defocused beam covers it

completely. The beam transmitted through the sample is recorded

with a position-sensitive detector.

Fig. 2: Schematic drawing of the phantom. The cuboid PMMA

(dark grey) was drilled with three holes each of which was ﬁlled

with a mixture of a selenium compound (elemental Se (red),

Na2SeO3(green) and Na2SeO4(blue)) and starch. The holes were

capped with tissue paper (light grey).

II. RESULTS

Fig. 1shows the schematic design of the experiment. The

polychromatic beam is produced by a conventional 1.5 kW

X-ray tube with an Ag anode. The beam is monochromatised

and refocused on the focal spot using a spherically bent

strip-bent Si(953) crystal analyser with a bending radius of

0.5 m [19]. The focal point of the crystal analyzer produces

a secondary source for the cone-beam imaging, with sample

located downstream, and the image of which is captured

by a position-sensitive photon-counting MiniPIX detector

by Advacam, which is based on a TimePIX [20] direct

conversion Si detector with 256 ×256 square pixels with

the side length of 55 µm. To obtain a spectrum, the crystal

analyzer Bragg angle is scanned across the relevant range

(in this case, 77.47◦to 73.01◦, corresponding to a range of

12.54 to 12.80 keV). The sample and the detector follow

the moving monochromatic cone with high precision using

motorized stages.

The contrast of Se-K edge XANES is illustrated in

Fig. 3. It shows the relative increase of the attenuation

coefﬁcient at the photon energy range that corresponds

to the Se K near edge region. The background of the

attenuation owing to other electrons than the Se 1shave

been subtracted for photon energies below the K edge, and

the background-subtracted spectra have been normalized to

an equal area in the energy region of the plot. The CT

scans were obtained at the photon energies labelled A-D.

As will be shown below, sampling the spectra at these four

distinct energies is sufﬁcient to distinguish the three different

oxidation states of Se that are used in this experiment.

A B CD

Fig. 3: Background-subtracted and normalized K-edge absorption

spectra of elemental Se, Na2SeO3[Se(IV)] and Na2SeO4[Se(VI)].

The vertical lines indicate the acquisition photon energies of 12.645

(A), 12.658 (B), 12.662 (C), and 12.685 (D) keV.

For each monochromated photon energy, a transmission

(projection) image is acquired for various sample orientations

at 1.8◦intervals, with a 100 s exposure time.

The ﬁrst step in the data analysis is to calibrate the

instrument via, e.g., the ﬂat ﬁeld image. An example of a

single ﬂat ﬁeld image is presented in Fig. 4. The image is

not completely uniform but it exhibits a horizontal slope in

the intensity. Since the electron beam inside the X-ray tube

propagates horizontally as well, the observed slope could

be due to the intensity variation in the source beam due to

the anode heel effect. Non-uniform stray scattering inside

the chamber may also contribute to the observed gradient.

With better collimation, shielding of the equipment and use

of vacuum or He-ﬁlled chamber to reduce the scatter, this

non-constant background probably can be partly reduced in

the future. Slightly darker vertical lines at x= 60 and

x= 180 px are due to the gaps between the strips of

the crystal wafer analyser [19]. The mean count rate of

photons per pixel was 2.6 ph s−1px−1. Almost all photon

count values fall between the range 2.2–3.0 ph s−1px−1or

approximately within ±15 % of the mean value.

In Fig. 5two projections are presented, taken at energy

A and D, respectively. As expected, there is virtually no

difference in the integrated attenuation coefﬁcient of the

phantom except at the locations which contain Se. The

resolution of the image was calculated from the projection

at energy A by differentiating the edge of the phantom

row-by-row at the lower left corner of the projection, ﬁtting

Gaussian functions to the resulting line spread functions,

and taking the mean of the FWHM (full width at half

maximum) and its standard error of the ﬁtted curves. The

文档加载中……请稍候！
如果长时间未打开，您也可以点击刷新试试。

下载文档到电脑，查找使用更方便

10 玖币 0人已下载

立即下载

摘要：

Monochromaticcomputedtomographyusinglaboratory-scalesetup:proof-of-conceptAri-PekkaHonkanen1;2andSimoHuotari21ComprehensiveCancerCenter,HelsinkiUniversityHospital,P.O.BOX180,FI-00029HUS,Finland2DepartmentofPhysics,UniversityofHelsinki,P.O.Box64,FI-00014Helsinki,FinlandAbstractInthisarticle,wedemon...

展开>> 收起<<

Monochromatic computed tomography using laboratory-scale setup proof-of-concept Ari-Pekka Honkanen12and Simo Huotari2.pdf

共6页,预览2页

还剩页未读，继续阅读

声明：本站为文档C2C交易模式，即用户上传的文档直接被用户下载，本站只是中间服务平台，本站所有文档下载所得的收益归上传人(含作者)所有。玖贝云文库仅提供信息存储空间，仅对用户上传内容的表现方式做保护处理，对上载内容本身不做任何修改或编辑。若文档所含内容侵犯了您的版权或隐私，请立即通知玖贝云文库，我们立即给予删除！

Monochromatic computed tomography using laboratory-scale setup proof-of-concept Ari-Pekka Honkanen12and Simo Huotari2

相关推荐

开通VIP享超值会员特权

作者详情

相关内容

热门标签

举报选择: