Applied Mathematical Modelling 39 (2015) 58965906

Contents lists available at ScienceDirect

Applied Mathematical Modelling

journal homepage: www.elsevier.com/locate/apm

Simulation and modeling of alignment-free eld emission X-ray

tubes q
Meng-Jey Youh a, Yi-Ping Chou b, Yih-Ming Liu c,1, Ming-Der Ger c, Kung-Hsu Hou d,
Nen-Wen Pu e,
a

Department of Information Technology, Hsing Wu University, New Taipei City 244, Taiwan, ROC
School of Defense Science, Chung Cheng Institute of Technology, National Defense University, Dasi, Taoyuan 335, Taiwan, ROC
c
Department of Chemical and Materials Engineering, Chung Cheng Institute of Technology, National Defense University, Dasi, Taoyuan 335, Taiwan, ROC
d
Department of Power Vehicle and Systems Engineering, Chung Cheng Institute of Technology, National Defense University, Dasi, Taoyuan 335, Taiwan, ROC
e
Department of Photonics Engineering, Yuan Ze University, Chung-Li, Taoyuan 320, Taiwan, ROC
b

a r t i c l e

i n f o

Article history:
Received 29 July 2014
Received in revised form 26 March 2015
Accepted 31 March 2015
Available online 6 May 2015
Keywords:
X-ray tube
Field emission
Alignment-free
Carbon nanocoil

a b s t r a c t
A new hemispherical-geometry cathode for eld emission (FE) X-ray tubes was designed
and simulated in this study. The electric eld strength distribution on the
hemispherical-geometry cathode was calculated and used to predict the emission current.
Because of their unique FE properties, carbon nanocoils were used as the electron
emitters for simulation and modeling. The results show that a large tolerance is permitted,
and the performance attained with 20 cathode misalignment was nearly identical to
that without misalignment. The maximum emission current can be achieved using a
hemispherical-geometry cathode with a radius of 1 mm. Additional advantages of FE
X-ray tubes with hemispherical-geometry cathodes are higher FE currents, shorter
response times, and smaller X-ray spot sizes. In addition, lower power consumption, longer
lifetimes, and higher resolutions are expected with this simple and low-cost design.
2015 Elsevier Inc. All rights reserved.

1. Introduction
Conventional X-ray tubes used in industrial applications and medical diagnoses generally employ the thermionic emission (TE) design. This design generally involves a metal lament (cathode) that emits electrons when heated over 1000 C
and a metal target (anode) that emits X-rays when bombarded by the accelerated electrons [1]. The X-ray intensity is proportional to the electron current and the square of the acceleration voltage [2]. Because the lament within the X-ray tube
must be heated by the control voltage, thermionic cathodes tend to have long response times and high energy consumption.
Moreover, the cathode is operated at high temperatures and can easily react with gaseous molecules, mainly H2O and oxygen, to form metal oxides. Such contamination of the metal lament reduces the lifetime of the X-ray tubes. The average
lifetime of a conventional TE X-ray tube is less than a year, and a majority of X-ray tube failures arise from

q
This article belongs to the Special Issue: ASPEC 2013 2013 International Applied Science and Precision Engineering Conference, October 2013 NanTou,
Taiwan.
Corresponding author. Tel.: +886 3 4638800; fax: +886 3 4514281.
E-mail addresses: 095063@mail.hwu.edu.tw (M.-J. Youh), koleon2001@yahoo.com.tw (Y.-P. Chou), liuym@ndu.edu.tw (Y.-M. Liu), mingderger@gmail.
com (M.-D. Ger), khou@ndu.edu.tw (K.-H. Hou), nwpuccit@gmail.com (N.-W. Pu).
1
Tel.: +886 3 3891716.

http://dx.doi.org/10.1016/j.apm.2015.03.033
0307-904X/ 2015 Elsevier Inc. All rights reserved.

M.-J. Youh et al. / Applied Mathematical Modelling 39 (2015) 58965906

5897

lament-related damages. In addition, the electrons from the cathode are emitted in random directions. This non-Gaussian
emission of electrons reduces X-ray image resolution; however, the emission can be focused using an external electric eld.
By contrast, eld emission (FE) electrons are generated through electron tunneling near the Fermi level when a high electric eld is applied; therefore, electron emission through FE is at room temperature (cold emission) and the output current is
voltage controllable. The FE cathode operates at room temperature, thus extending the cathode lifetime and lowering energy
consumption. FE X-ray tubes are the next generation of X-ray tubes and will replace TE X-ray tubes. FE theory is elaborated
upon in Section 2. Moreover, the electron emission from a cold cathode is always along the vector of the applied electric eld,
and the Gaussian electron emission distribution is advantageous in X-ray imaging. Furthermore, FE cathodes have found
application in various vacuum electronic devices, such as FE displays and transmission electron microscopes.
The concept of cold cathode X-ray tubes has been explored using several materials, including diamond-like carbon and
carbon nanotubes (CNTs). In recent years, researchers have found that FE from carbon nanocoils (CNCs) is more advantageous than those from CNTs in terms of long-term current emission stability, current density, and operation lifetime.
Paired with the benet of CNCs, FE is a more effective method of extracting electrons than TE because of room temperature
electron emission and voltage-controllable output power. Instead of a at-plane cathode, a novel hemispherical-geometry
cathode for FE X-ray tubes was designed and simulated in this study.
2. Theory and methodology
2.1. FE theory
The theory of FE was proposed by Fowler and Nordheim in 1928 [3]. Compared with TE, FE is advantageous because electrons are emitted at room temperature (i.e., cold cathode) and FE current is voltage controllable; in addition, the low operating temperature in FE extends the cathode lifetime. FE cathodes are used in several applications such as display panels
[4,5], at panel displays for producing backlight [6,7], FE lamps [812], and X-ray tubes [1318].
FE results from the tunneling of electrons through the potential barrier near the Fermi level when a high electric eld is
applied. The image force is a Coulombic force of attraction pulling the emitted electron toward the surface because of the
induced surface charge on the metal. The net potential energy of the electron is the sum of the image force potential and
the potential created by the applied external electric eld at the surface. The potential barrier is innitely thick in the
absence of an electric eld, but it becomes triangular with a considerably narrowed width when a large electric eld is
applied [1923].
The slope of the potential and consequently the width of the potential barrier depend on the amplitude of the local electric eld on the surface. This local electric eld is considerably enhanced if the electron emitter is a sharply pointed wire with
a high aspect ratio.
The FE mechanism can be summarized by the FowlerNordheim (FN) equation [3], which is shown in (1).

I aAb2 E2 =u expbu3=2 =bE;

where I, A, E, u, and b are the emission current, effective FE area, macroscopic electric eld (in the absence of the local
structure that causes enhancement), work function, and eld enhancement factor, respectively; b is a constant. The eld
enhancement factor b is the most crucial factor determining the performance of FE and is often introduced in the FN
equation to represent the geometrical effects at the cathode surface. The value of b can be extracted from FN plots, as shown
in Fig. 1.

Fig. 1. Conventional FN plot used to extract the b value of FE emitters.

5898

M.-J. Youh et al. / Applied Mathematical Modelling 39 (2015) 58965906

The extremely high aspect ratios of CNTs and CNCs produce a large b [2426], which makes them excellent eld emitters
with low turn-on elds. The FE current is a function of E, and the potential distribution V(x, y, z) can generally be obtained by
numerically solving Poissons equation by using the nite element method (FEM). The electric eld distribution is calculated
by differentiation as shown in (2).

~
~V
E r

The electric eld strength can easily be obtained in at-plane geometries, but it is difcult to calculate in nonplanar
geometries. Therefore, computer simulation and the FEM were applied to determine the electric eld strength in the
hemispherical-geometry cathode used in this study.
2.2. Choices of electron emitter
The unique properties of carbon nanomaterials, such as their high thermal conductivity, low work function, and excellent
electron migration rate, make them promising FE emitters. CNTs with nanoscale dimension have received much attention
since their discovery by Iijima [2733]. An advantage of CNTs in FE is that they are natural eld-enhancing objects because
of their sharp geometry and high aspect ratio.
CNCs [3443] are nanocarbon materials with a coiled structure and are composed of nanometer-scale bers. Compared
with CNTs, CNCs have considerably more electron emission points, not only at their tips but also in their coils. CNCs have
several other advantages over CNTs, including a higher long-term current emission stability, current density, and FE efciency, and longer lifetime [44,45].
CNCs are suitable FE emitters primarily because of their unique geometry and their ability to emit electrons from not only
their tips but also their bodies; therefore, CNCs are relatively more efcient than CNTs. In addition, CNCs are more robust
than CNTs because the former is typically composed of several layers of graphene sheets. Thus, we used CNCs as the emitters
in our X-ray tube.
CNCs also exhibit novel mechanical and electromagnetic properties because of their unique twist morphology. Therefore,
CNCs nd application in several devices such as electromagnetic wave absorbers [46], vibration dampers [47], and FE
emitters.
CNCs are generally synthesized using catalytic chemical vapor deposition (CVD) or arc discharge. Because CVD yields
low-cost and large-scale commercial synthesis of CNCs, it was used to synthesize CNCs in this research.
2.3. FEM analysis
In this study, the FEM was employed in the simulation to calculate the electric eld strength. The FEM is a numerical technique for determining approximate solutions to boundary value problems for differential equations. First, the problem
domain is divided into nite elements, as shown in Fig. 2, with each element represented by a set of element equations
to the original problem. For the nal calculation, all element equations are systematically recombined into a global system
of equations, which have known solution techniques that can be calculated from the initial values of the original problem, to
obtain a numerical answer.

Fig. 2. Top view of the hemispherical-geometry cathode; red points indicate the data extracted for predicting FE current. (For interpretation of the
references to colour in this gure legend, the reader is referred to the web version of this article.)

M.-J. Youh et al. / Applied Mathematical Modelling 39 (2015) 58965906

5899

The FEM is the preferred analysis method in structural mechanics. In our study, the FEM is also a suitable tool for obtaining the numerical solution of the electric eld strength on the hemispherical-geometry cathode.
2.4. Hemispherical-geometry cathode
FE cathodes are mostly planar in order to increase the emission area and emission current. The cathode and anode must
align precisely to avoid poor eld uniformity across the system. According to the FN equation, the FE current is strongly
affected by the electric eld strength applied on the emitter surface. If a at-plane cathode is not aligned precisely, for example, tilted at a small angle, the electric eld lines converge in one corner, resulting in nonuniform FE current density and a
reduction in the cathode lifetime.
To solve this critical problem, a hemispherical-geometry cathode is proposed in this study. It is an alignment-free cathode
design without degradation in emission area or emission current. According to our simulation, hemispherical-geometry
cathodes have the advantage of approximately 20 misalignment tolerance because of the shielding effect. This effect
allows the lower hemisphere of the hemispherical-geometry cathode to have almost zero potential distribution and emission of electrons, enabling the connecting rod to tolerate misalignment.
3. Experiment details and results
3.1. Modeling of the planar cathode
We used SolidWorks to build a two-dimensional (2D) X-ray tube model and subsequently used COMSOL Multiphysics 4.2
to solve the electric eld strength on the cathode surface of various radii.
The simulation results of the potential distribution on the tradition planar cathodes at an applied voltage of 7000 V are
shown in Fig. 3. The X-ray tube was a 4 mm  4 mm rectangle with 3 mm separation between the anode and the cathode. On
precise parallel alignment of the planar cathode and anode, uniform potential distribution was obtained on the cathode surface, as shown in Fig. 3(a). With a 10 misalignment, the potential distribution became nonuniform and the electric eld lines
converged toward to one side, as shown in Fig. 3(b). The emitting currents focused on the left corner, thus reducing the cathode lifetime.
Fig. 4 shows the electric eld distribution on the planar cathode surface. In the misaligned cathode, the maximum electric
eld occurred toward its left; by contrast, the well-aligned cathode had a uniform electric eld strength distribution on its
surface.
Evidently, parallel tolerance is a key issue of the planar cathode system. Even with a small misalignment between the
cathode and anode, because electrons travel through the shortest path, dense emission bombardment on certain regions
damages cathodes.
3.2. Modeling of the hemispherical-geometry cathode
Conventional FE type X-ray tubes are composed of a planar cathode and anode, as mentioned. If a small misalignment
occurs, such as from the spacer being unequal, the electric eld strength concentrates on a certain region, causing the FE
current to focus on certain sites; in such cases, the emitters rapidly burn out.

Fig. 3. Simulation results of the potential distribution on a conventional planar cathode at an applied voltage of 7000 V: (a) well-aligned cathode; (b)
cathode misaligned at approximately 10.

5900

M.-J. Youh et al. / Applied Mathematical Modelling 39 (2015) 58965906

Fig. 4. Electric eld distribution on a conventional planar cathode surface. Black: well-aligned cathode; red: cathode tilted at 10. (For interpretation of the
references to colour in this gure legend, the reader is referred to the web version of this article.)

Therefore, a 2D X-ray model of a hemispherical-geometry cathode with CNCs and 4 mm  4 mm planar anode was
designed and the potential energy distribution was calculated, as shown in Fig. 5. The cathode, a 1 mm diameter ball,
was placed underneath the center of the anode; the distance between the top of the spherical cathode and the anode was
2 mm. The accelerating voltage was set at 7 kV to calculate the potential distribution in this space. Because of the shielding
by the metal cathode, the potential energy distribution in the lower hemisphere of the cathode was nearly uniform. The calculation of the electric eld on the lower hemisphere was omitted in the simulation.
Fig. 6 illustrates the electric eld distribution in the upper hemisphere of the cathode. Compared with that of the misaligned planar cathode, the electric eld strength had little variation above the equator, and the maximum eld strength was
on top of the cathode, demonstrating that a large tolerance misalignment of up to 20 is permitted in this hemispherical
system.
Potential distribution simulation was performed at an applied voltage of 8 kV, and the results were plotted in a
three-dimensional diagram (Fig. 7). As shown in the gure, the potential distribution was uniform below the equator of
the spherical cathode; likewise, the electrons in this area were shielded by the metal cathode. Because the ball cathode must
be connected to a conductive metal electrode for conducting current and for mechanical support, having a shielded space
behind the ball cathode becomes essential.

Fig. 5. Potential energy distribution in the 2D hemispherical-geometry cathode at an applied voltage of 7000 V.

M.-J. Youh et al. / Applied Mathematical Modelling 39 (2015) 58965906

5901

Fig. 6. Electric eld distribution in the upper hemisphere of the hemispherical-geometry cathode at an applied voltage of 7000 V.

Fig. 7. Potential energy distribution on the tube. The hemisphere has a shielding area for 20 from the cathode center.

As shown in Fig. 8, the metal electrode can have up to a 20 misalignment when the cathode is placed in the X-ray tube,
and Fig. 7 depicts a nearly uniform potential gradient with 20 misalignment. That is, the alignment of the
hemispherical-geometry cathode is noncritical.
3.3. Synthesis of CNCs
CNCs were directly grown on the hemispherical-geometry cathode by using a catalytic CVD method. In recent years, Pd
was found to be a promising catalyst for CNC growth; in addition, the structure and characteristics of CNCs can be controlled
by modifying the catalytic composition and the reaction parameters. Therefore, Pd was used as the catalyst for CNC growth,
and 304 stainless steel balls were used as the cathode substrate.

5902

M.-J. Youh et al. / Applied Mathematical Modelling 39 (2015) 58965906

Fig. 8. Because of the shielding effect of the cathode on the metal rod, approximately 20 cathode misalignment is permitted.

Stainless steel balls of various diameters were rinsed and ultrasonically treated in acetone for 10 min and subsequently
washed in deionized water. Next, the samples were etched with 1037% HCl for 310 min to expose fresh metal surfaces and
washed again in deionized water. The Pd catalysts were deposited on the cathode through chemical displacement in a 100
800 ppm PdCl2 solution of pH 1.61.8 at 80 C for 10 min. After the reaction was complete, the cathodes were rinsed with
deionized water and dried with N2 gas. The prepared cathodes were placed in a thermal CVD furnace and the pressure
was pumped down to 102 torr. The temperature was gradually increased to 600 C at a rate of 20 C/min under a H2 and
Ar atmosphere. The ow rates for H2 and Ar were 10 and 100 sccm, respectively. When the temperature reached 600 C,
the Ar gas was evacuated and the H2 ow rate was increased to 20 sccm to pretreat the Pd catalysts for 5 min. Next, to grow
the CNCs, C2H2 gas was fed at a ow rate of 5 sccm for 20 min; the furnace temperature was maintained at 600 C. A scanning
electron microscopy (SEM) micrograph of CNCs grown on the cathode is shown in Fig. 9 (the secondary electron image). The
average pitch and radius of the CNCs are both approximately 1 lm.
3.4. Extraction of b
The eld enhancement factor (b) depends on the microstructural properties, such as length, diameter, density, and morphology, of CNCs. It is independent of the geometries of the macrostructure, such as the cathode and anode geometry.
To evaluate b, a simple test using a cylindrical anode and an emitter cathode lament was designed, as shown in Fig. 10. A
lament with a radius r1 = 0.5 mm was placed on the symmetry axis of a cylindrical anode with an inner radius r2 = 7.5 mm.
The anode was a glass tube with a conductive coating and a phosphor layer on the inner surface. The macroscopic electric
eld E is given by the analytical expression shown in (3), which can be derived from Gausss law.

Er V=r lnr 1 =r 2 

Fig. 9. SEM micrograph of CNCs grown on the cathode surface by using catalytic CVD with Pd as the catalyst.

5903

M.-J. Youh et al. / Applied Mathematical Modelling 39 (2015) 58965906

Fig. 10. Cylindrical geometry for measuring b of carbon nanocoils.

IV curves were plotted for cathodes 10, 20, 30, and 40 mm in length. Subsequently, the FN plots were charted; as shown
in Fig. 11, all curves have similar slopes. The average b = 1075 was used as the eld enhancement factor in the simulations.
3.5. Calculation of FE current
According to the FN equation, the FE current is described by (4). The constants used during the calculation are listed in
Table 1.

ab2 E2 =u expbu3=2 =bEdA:

The calculation results with four cathode radii, 0.5, 1, 1.5 and 2 mm, are presented in Fig. 12. The electric eld strength
increased with a decreasing cathode radius. The hemispherical-geometry cathode with a radius of 0.5 mm had the highest
electric eld strength, and that with a 2 mm radius had the weakest. However, the effective emission area decreased with the
cathode radius; therefore, the integration of the cathode surface area and electric eld strength, as shown in (4), must be
considered in predicting FE currents.
At an 8 kV applied voltage, the predicted FE currents were 5.296, 5.379, and 4.900 mA for cathodes with radii of 0.5, 1, and
1.5 mm, respectively (Fig. 13), indicating that the optimal radius was approximately 1 mm. In the voltage range 840 kV, the
maximum FE current could be obtained using a 1 mm radius cathode. Thus, our design yields optimal emission currents.

Fig. 11. FN plot; cathode diameter = 1 mm; cathode length: 10, 20, 30 and 40 mm. Average b = 1075.

Table 1
Constants used in calculating FE current.

1.54  106
[eV/V2]

1075

6.83
[eV3/2V/nm]

5
[eV]

5904

M.-J. Youh et al. / Applied Mathematical Modelling 39 (2015) 58965906

Fig. 12. Electric eld distribution for cathodes with radii of 0.5, 1, 1.5, and 2 mm.

Fig. 13. Electric eld strength distribution on the hemispherical-geometry cathodes with various radii at accelerated voltages of 8, 16, 24, 32, and 40 kV.

Fig. 14. X-ray intensity for hemispherical-geometry cathodes with radii of 0.5, 1.0, and 1.5 mm at accelerated voltages of 8, 16, 24, 32, and 40 kV.

M.-J. Youh et al. / Applied Mathematical Modelling 39 (2015) 58965906

5905

3.6. Optimal FE currents and X-ray intensity

Because the X-ray intensity is proportional to the electron current and the square of the acceleration voltage, the intensity
of our X-ray tube was predicted as shown in Fig. 14.
In this study, a computer simulation program was utilized to solve the complex geometry of the cathode and the electric
eld strength. Cathodes with radii of 0.5, 1.0, 1.5, and 2 mm were simulated. The cathode with a radius of 0.5 mm had the
highest electric eld strength but a relatively smaller emitting area, whereas the cathode with a radius of 2 mm had the
weakest electric eld strength but a relatively larger emitting area. Cathodes with small radii are advantageous because they
provide a small X-ray spot size. Spot size is a crucial factor in X-ray resolution; the smaller the spot size is, the higher the
resolution.
Calculating the electric eld strength on the cathode surface by using the constants and calculated values revealed that
the optimal FE current occurred on the hemispherical-geometry cathode with a radius of 1 mm. In addition, the cathode with
a 1 mm radius had a small spot size. Higher X-ray brightness and resolution can be achieved using this method.
4. Conclusion
We designed a new alignment-free FE X-ray tube and constructed a numerical simulation model for the electrostatic calculation of its hemispherical geometry. We used CNCs as the emitters because of their higher long-term stability, longer lifetime, higher FE currents, and greater number of emission sites compared with CNTs. A hemispherical-geometry stainless
steel cathode grown with CNCs and an 8 mm  8 mm planar copper anode were used in our design. The primary advantage
of the hemispherical-geometry cathode is its insensitivity to misalignment up to approximately 20. The electrical eld distribution within the tube was computed. Using b extracted from our previous experiment, we calculated the total FE current
by integrating the current density over the cathode surface. The simulation showed that eld strength decreased with the
cathode diameter and cathode-to-anode separation. An additional advantage of the hemispherical-geometry cathode grown
with carbon CNCs is its large current. The maximum FE current of 5.379 mA was predicted to occur for a 1-mm-radius cathode at an applied voltage of 8 kV. In addition, because the electric eld strength is nearly uniform in the upper hemisphere,
electrons are evenly emitted, thus increasing the cathode lifetime. These simulation results are useful in designing an
alignment-free FE X-ray tube in the future.
Acknowledgment
This work was funded by the Ministry of Science and Technology of Taiwan under Grants MOST 103-2221-E-606-003 and
MOST 104-2623-E-155-007-ET.
References
[1]
[2]
[3]
[4]
[5]
[6]

[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]

S.C. Bushong, Radiologic Science for Technologist Mosby, St. Louis, 1997.
W.O. Richardson, Proc. Cambridge Philos. Soc. 11 (1901) 286.
R.H. Fowler, L. Nordheim, Electron emission in intense electric elds, Proc. R. Soc. London A 119 (1928) 173181.
F.G. Zeng, C.C. Zhu, W. Liu, X. Liu, The fabrication and operation of fully printed carbon nanotube eld emission displays, Microelectron. J. 37 (2006)
495499.
A.A. Kuznetzov, S. Lee, B.M. Zhang, R.H. Baughman, A.A. Zakhidov, Electron eld emission from transparent multiwalled carbon nanotube sheets for
inverted eld emission displays, Carbon 48 (2010) 4146.
Y.C. Choi, J.W. Lee, S.K. Lee, M.S. Kang, C.S. Lee, K.W. Jung, J.H. Lim, J.W. Moon, M.I. Hwang, I.H. Kim, Y.H. Kim, B.G. Lee, H.R. Seon, Sang Jin Lee, J.H. Park,
Y.C. Kim, H.S. Kim, The high contrast ratio and fast response time of a liquid crystal display lit by a carbon nanotube eld emission backlight unit,
Nanotechnology 19 (2008) 235306.
Y. Zhang, S.Z. Deng, N.S. Xu, J. Chen, Fully sealed carbon nanotube at-panel light source and its application as thin lm transistorliquid-crystal
display backlight, J. Vac. Sci. Technol. B 26 (2008) 10331037.
N.Y. Bae, W.M. Bae, A.N. Ha, M. Nakamoto, J. Jang, K.C. Park, Low-voltage driven carbon nanotube eld emission lamp, Curr. Appl. Phys. 11 (2011) S86
S89.
H.S. Yoo, W.Y. Sung, S.J. Yoon, Y.H. Kim, S.K. Joo, Novel triode-type eld emission arrays and appropriate driving method for at lamp using carbon
nanobers grown by plasma enhanced chemical vapor deposition, Jpn. J. Appl. Phys. 46 (2007). 4381-4356.
L. Wei, X. Zhang, B. Wang, C. Lou, Z. Zhu, Z. Zhao, L. Chi, H. Kai, A stable eld-emission light source with ZnO nanoemitters, IEEE Electron Device Lett. 29
(2008) 452455.
X.X. Zhang, C.C. Zhu, Field-emission lighting tube with CNT lm cathode, Microelectron. J. 37 (2006) 13581360.
Yi-Ping Chou, Nen-Wen Pu, Ming-Der Ger, Kun-Ju Chung, Kung-Hsu Hou, Yi-Ming Liu, Bulb-shaped eld emission lamps using carbon nano-coil
cathodes, J. Nanosci. Nanotechnol 12 (2012) 83168322.
S. Wang, X. Calderon, R. Peng, E.C. Schreiber, O. Zhou, S. Chang, A carbon nanotube eld emission multipixel X-ray array source for microradiotherapy.
application, Appl. Phys. Lett. 98 (2011) 213701.
R.R. Whitlock, M.I. Bell, D.V. Kerns, S. Kerns, J.L. Davidson, W.P. Kang, Transmission cathode for X-ray production, US Patent No. 6333968, 2001.
R. Baptist, Electron eld emission from carbon nanotubes, US Patent No. 6259765, 2001.
P. Rangstein, C. Ribbing, C. Strandman, R. Hok, L. Smith, Field-emitting structures intended for a miniature X-ray source, Sens. Actuators 82 (2000) 24
29.
H. Sugie, M. Tanemura, V. Filip, K. Iwata, K. Takahashi, F. Okuyama, Carbon nanotubes as electron source in an X-ray tube, Appl. Phys. Lett. 78 (2001)
25782580.
J. Zhang, Y. Cheng, Y.Z. Lee, B. Gao, Q. Qiu, W.L. Lin, D. Lalush, J.P. Lu, O. Zhou, A nanotube-based eld emission X-ray source for microcomputed
tomography, Rev. Sci. Instrum. 76 (2005). 094301-4.
Y. Cheng, O. Zhou, Electron eld emission from carbon nanotubes, C.R. Phys. 4 (2003) 10211033.

5906
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41]
[42]
[43]
[44]
[45]
[46]
[47]

M.-J. Youh et al. / Applied Mathematical Modelling 39 (2015) 58965906

E.L. Murphy, R.H. Good Jr., Thermionic emission, eld emission, and the transition region, Phys. Rev. 102 (1956) 1464.
W.P. Dyke, W.W. Dolan, Field emission, Adv. Electron. Electron. Phys. 8 (1956) 89183.
J.W. Gadzuk, E.W. Plummer, Field emission energy distribution (FEED), Rev. Mod. Phys. 45 (1973) 487.
R. Gomer, Field emission, eld ionization and eld desorption, Surface Sci. 299300 (1994) 129.
L.J. Pan, T.C. Hayashida, M. Zhang, Y. Nakayama, Field emission properties of carbon tubule nanocoils, Jpn. J. Appl. Phys. 40 (2001) L235.
W.Y. Sung, J.G. Ok, W.J. Kim, S.M. Lee, S.C. Yeon, H.Y. Lee, Y.H. Kim, Synthesis and eld emission characteristics of carbon nanocoils with a high aspect
ratio supported by copper micro-tips, Nanotechnology 18 (2007) 245603.
H. Liu, Y. Saito, Inuence of surface roughness on eld emission of electrons from carbon nanotube lms, J. Nanosci. Nanotechnol. 10 (2010) 3983.
W.A. de Heer, D. Ugarte, Carbon onions produced by heat treatment of carbon soot and their relation to the 217.5 nm interstellar absorption feature,
Chem. Phys. Lett. 207 (1993) 480.
S. Fan, M.G. Chapline, N.R. Franklin, T.W. Tombler, A.M. Cassell, H. Dai, Self-oriented regular arrays of carbon nanotubes and their eld emission
properties, Science 283 (1999) 512.
A.G. Rinzler et al, Synthesis of large arrays of well-aligned carbon nanotubes on glass, Science 269 (1995) 1550.
W.A. de Heer, A. Chatelain, D. Ugarte, Electrostatic deections and electromechanical resonances of carbon nanotubes, Science 270 (1995) 1179.
Y. Saito, S. Uemura, Field emission from carbon nanotubes and its application to electron sources, Carbon 38 (2000) 169.
J.M. Bonard, M. Croci, C. Klinke, R. Kurt, O. Noury, N. Weiss, Carbon nanotube lms as electron eld emitters, Carbon 40 (2002) 1715.
J.X. Huang, Jun Chen, S.Z. Deng, J.C. She, N.S. Xu, Field-emission uorescent lamp using carbon nanotubes on a wire-type cold cathode and a reecting
anode, J. Vac. Sci. Technol. B 26 (2008) 1700.
W.B. Downs, R.T.K. Baker, Novel carbon ber-carbon lament structures, Carbon 29 (1991) 1173.
S. Motojima, M. Kawaguchi, K. Nozaki, H. Iwanaga, Preparation of coiled carbon bers by catalytic pyrolysis of acetylene, and its morphology and
extension characteristics, Carbon 29 (1991) 379.
M. Kawaguchi, K. Nozaki, S. Motojima, H. Iwanaga, A growth mechanism of regularly coiled carbon bers through acetylene pyrolysis, J. Cryst. Growth
118 (1992) 309.
X. Chen, S. Motojima, Morphologies of carbon micro-coils grown by chemical vapor deposition, J. Mater. Sci. 34 (1999) 5519.
S. Motojima, X. Chen, Three-dimensional vapor growth mechanism of carbon microcoils, J. Appl. Phys. 85 (1999) 19.
X. Chen, S. Motojima, The growth patterns and morphologies of carbon micro-coils produced by chemical vapor deposition, Carbon 37 (1999) 1817.
K. Hernadi, A. Faonseca, J.B. Nagy, . Fudala, D. Bernaerts, I. Kiricsi, Catalytic production of carbon nanobers over iron carbide doped with Sn2+, Appl.
Catal. A Gen. 228 (2002) 103113.
M. Zhang, Y. Nakayama, L. Pan, Synthesis of carbon tubule nanocoils in high yield using iron-coated indium tin oxide as catalyst, Jpn. J. Appl. Phys. 39
(2000) 1242.
V.V. Chesnokov, V.I. Zaikovskii, R.A. Buyanov, Symmetric twisted carbon laments formed from butadiene-1, 3 on NiCu/MgO-catalyst: growth
regularities and mechanism, J. Mol. Catal. A Chem. 158 (2000) 267270.
X. Chen, S. Yang, S. Motojima, Morphology and growth models of circular and at carbon coils obtained by the catalytic pyrolysis of acetylene, Mater.
Lett. 57 (2002) 48.
W.Y. Sung, J.G. Ok, W.J. Kim, S.M. Lee, S.C. Yeon, H.Y. Lee, Y.H. Kim, Synthesis and eld emission characteristics of carbon nanocoils with a high aspect
ratio supported by copper micro-tips, Nanotechnology 18 (2007) 245603.
Z.J. Zhang, P.G. He, Z. Sun, T. Feng, Y.W. Chen, H.L. Li, B.K. Tay, Growth and eld emission property of coiled carbon nanostructure using copper as
catalyst, Appl. Surf. Sci. 256 (2010) 4417.
G. Wang, Z. Gao, S. Tang, C. Chen, F. Duan, S. Zhao, S. Lin, Y. Feng, L. Zhou, Y. Qin, Microwave absorption properties of carbon nanocoils coated with
highly controlled magnetic materials by atomic layer deposition, ACS Nano 6 (2012) 1100911017.
T. Okumura, T. Nosaka, Y. Kita, S. Akita, Y. Nakayama. Mechanical properties of carbon nanocoil/polycarbonate composite resin. Abstract, 15th Filler
symposium, 2007 (in Japanese).