We report on the fabrication, characterization, and field emission behavior of cold electron cathodes that are specifically designed to be used for X-ray sources. The emitter was designed as a two-tier system, with the patterned CNT field emission array and the Pt comb electrode beneath. The micromachining patterning process, which enhanced the edge effect, significantly improved the field emission performance. By using an inorganic binder, the cathode could resist high temperature annealing up to 800°C. Our novel emitter showed relatively good field emission properties such as low turn on field, high current density, and good stability. We ascribed the improved emission properties after the annealing process to the oxygen desorption induced decrease of work function, and defect annealing which could stabilize the emitter. To check the application of the emitter, the fabricated emitter was sealed in a conventional X-ray tube. The blackened shade on the sensitive film confirmed the X-ray generation, which demonstrated that our emitter successfully survived the high temperature processing steps of the X-ray source fabrication.
© 2016 Optical Society of America
Carbon nanotubes (CNTs) exhibit excellent field emission characteristics due to their inherent small tip radius and high aspect ratio combined with robust chemical and mechanical stabilities [1,2]. CNT field emission cold cathodes have been regarded as attractive electron sources in various applications such as field emission displays , lighting devices , x-ray tubes , electron microscopy and microwave power amplifiers. Recently, much effort has been made to develop various methods for fabricating CNT field emitters. Direct growth , electrophoresis , and screen printing  have been recognized as three promising techniques for fabricating field emission cold cathode. Although high performance field emitters with stable emission current of 1 A (current density of 4 A/cm2) have already been reported [9, 10]. It is believed that, as the heart of electronic devices, the carbon nanotube field emitters have not reached their full potential, and there are still lots of challenges in this field .
In particular, screen printing of CNT paste has been proposed as a capable technology for economic fabrication of CNT field emitters. As the technology using CNT paste has many advantages in terms of simple process, uniform emission site, and mass production. However, the screen-printing process has some limitations in obtaining fine-pitched patterns due to the inherent paste rheology. The minimum size of the pattern by the conventional screen-printing method is about 100 um .
Fabrication technology in the engineering of micro-electromechanical systems (MEMS) does well in manufacture devices that have characteristic length of less than 1 mm but more than 1 micron. It is very attractive to fabricate fine field emission arrays using this method, for we are able to design the CNT field emitters in 3 dimension scale, which can bring more options for optimization. As MEMS fabrication technology is always highly compatible with the integrated circuit batch-processing technologies. It is believed that, MEMS fabrication technology will be effective for fabrication of CNT field emitter with low cost.
We describe in this work the fabrication, characterization, field emission and X-ray generation behavior of carbon nanotube cathode produced by micromachining technology. The challenge was to conceive an inorganic paste that had the appropriate rheology for micromachining, which could also resist the high temperature processing steps intervening at later stages of the source fabrication (For example, high temperature brazing above 780°C is needed for vacuum-tight sealing of commercial X-ray sources.) [13, 14]. The inorganic paste also would not break device vacuum during high temperature operation due to volatile components out-gassing that would seriously impact the field emission property of the electron source . Sodium silicate is an environmental water soluble binder. It is widely used for insulation applications, as it exhibits suitably stable mechanical strength under high temperature . By using the inorganic binder, the emitter was designed as a two-tier system, the CNT field emission array and the Pt comb electrode beneath. A vacuum annealing process at 800°C was performed to verify the vacuum-tight sealing process. The fabricated emitters showed excellent field emission properties and stability. To check the application of this novel flexible emitter, an X-ray imaging experiment was performed. A simple diode X-ray tube was achieved by sealing the fabricated emitter into a conventional X-ray tube.
2.1 Materials and processing
Multi-walled carbon nanotubes (MWCNT, 0.5 g) with an average diameter of 30-50 nm and varying lengths of 5-15 um (95% pure) were bought from Timesnano Co., Ltd (Chengdu, China). They were mixed with 10 g sodium silicate (silicon dioxide 24.91% (wt), sodium oxide 7.25% (wt), modulus 3.55, density 1.32 g/cm3). A mechanical ball milling process was carried out resulting in a black nanotube ink (nanotube suspension). To maintain the intrinsic properties of CNTs, no surfactant was added to the ink. Pt comb electrode was designed beneath the emission cells to reduce the film resistance. The field emitter fabrication process is presented as follows (Fig. 1):
- (a) Cr (50 nm) seed layer was sputtered on the Si wafer;
- (b) Photoresist of 7 um was spun on the seed layer, and lithography was performed to develop the pattern area for the comb electrode;
- (c) The Pt comb electrode of 1 um was fabricated by sputtering, and Cr seed layer was removed by wet etching;
- (d) Photoresist of 50 um was spun on the seed layer, and lithography was performed to develop the pattern area for the emission array;
- (e) The emission array was performed by filling the pattern area with the nanotube ink;
- (f) The composite was cured by drying the sample at 300°C for 1 hour under ambient atmosphere, then the field emitter was achieved by etching away a thin sodium silicate layer by hydrofluoric acid. The etching process was carefully controlled to ensure that the CNTs were exposed on the surface and the CNT roots remained in the composite film. At last the remaining photoresist was removed by acetone.
All the above steps were carried out at room temperature. Then a vacuum annealing process was performed in a vacuum furnace at 800°C for 30 mins under 10−5 Torr.
The morphologies of the fabricated emitter were characterized using a field emission scanning electron microscope (FESEM; Zeiss ultra 55, Germany). In order to obtain good dispersion for length distribution measurement, 0.2 ml of each milled sample was distributed in 5 ml distilled water, 0.5 ml polyvinylpyrrolidone (PVP K16-18, Aladdin, China) was added to each sample, then the mixture was sonicated at room temperature for 1 h. The Raman spectrum of the emission cell was obtained using a Raman microscope (Ram, Bruker Opties Senterra R200, US) with 10 × and 100 × objectives at a laser wavelength of 532 nm. Spectrum acquisitions were done with a power of 1mW with integration times of 10-60 s depending on the examined sample.
The field emission characteristics of the emitter were measured in a vacuum chamber with a parallel diode-type configuration at pressure of 1 × 10−6 Torr. The voltage was applied by a high-voltage power supply (HBGY HB-2502-100AC, China) across the cathode and the anode with a distance of 150 μm. The current was measured and saved by a digit multimeter (Agilent 34401A, US). In order to protect high-voltage power supply from high-voltage arcing breakdown, a current-limiting resistor (3 MΩ) was used.
3. Results and discussion
Figure 2 shows a series of SEM images of the emitter. There are 51 cells on the emitter, the diameter of each cell is 180 um, and the height of each cell is 20 um, the gap between every two cells is 20 um. The height of the Pt comb electrode is 1 um. There are also magnitude SEM images of the emitter from the top Fig. 2(c) and 2(d), and the oblique Fig. 2(e) and 2(f) directions. Figure 2(c) and 2(d) shows that the cathode with a flat surface is covered with CNTs. As shown in Fig. 2(e) and 2(f), CNTs are vertically aligned and uniformly distributed in the substrate, and the vertically aligned CNTs can actively contribute to electron emission when an electric field is applied.
Due to their small tip radius and high aspect ratio, CNTs have a highly entangled structure, which needs to be dispersed, preferably up to single nanotube level, for practical applications. Ball milling process was employed to improve the CNT dispersion, and the CNT aggregate (composed of many CNTs) size distribution was investigated by laser particle size analyzer. According to the particle size evolution in Fig. 3(a), the particle size decreased and the length distribution narrowed down with increasing milling time. The aggregate particle size decreased from over 1200 nm to approx 850 nm in the first 1 h milling, then the rate of aggregate size reduction slowed down. Average size of about 770, 670, 430 nm could be achieved after milling for 4, 8, 16 h respectively.
During the ball-milling process, large CNT aggregate died down, and CNTs turned into conglomeration which was closed to granules and sheets because of the friction of rolling between the balls . It is believed that ball-milling process not only decreased the aggregate size, but also changed some CNTs into amorphous carbon. The CNTs structure changes made by the ball milling process were investigated by microprobe Raman technique. As shown in Fig. 3(b), there are two sharp peaks at 1350 cm−1 (D bond) and 1580 cm−1 (G bond) representing typical characteristics of amorphous and graphite carbons, respectively. The D bond at 1350 cm−1 is generally attributed to defects in the curved graphite sheet or other impurities, while the G bond at 1580 cm−1 is corresponding to the opposite direction movement of two neighboring carbon atoms in a graphitic sheet, and it indicates the presence of crystalline graphitic carbon in CNTs .
To study their structure changes in detail, relative value (R-value, ID/IG) are shown in Fig. 3(b). It is generally agreed that the R-value provides a useful index for comparing the crystallite sizes (more specifically, peak area of the Lorentzian functions) of various carbon materials. The R-value of the samples treated for 0, 1, 4, 8, 16 h is 0.736, 0.752, 0.797, 0.851 and 0.948, respectively. It reveals that the R-value of D bond increases against the milling time, which indicates small increase in defect density. From the results above, we can see that, the CNT dispersion was significantly improved by the ball milling process, however shortened CNTs with more structural defects were obtained.
In order to investigate the emission properties affected by the micromachining patterning, field emission characteristics of the rectangle field emitter (RFE) and patterned field emitter (PFE) were tested in a vacuum chamber. The emission area for RFE (SR = 0.272*0.048 = 0.013 cm2) and PFE (SP = 51*π*(Ø /2)2 = 0.013 cm2) are the same. The relationship of the emission current with the applied voltage is shown in Fig. 4. Both of the emission current increased monotonically with the applied field. The turn-on voltages were 401 V and 351 V for RFE and PFE emitter respectively. For RFE, with the applied voltage of 857 V, the field emission current of 2.16 mA (current density of 166 mA/cm2) was achieved. For PFE, with the applied voltage of 781 V, the field emission current of 3.08 mA (current density of 241 mA/cm2) was achieved.
To clarify the origin of the observed efficient field emission from PFE, the electrostatic potential was calculated by using COMSOL. To save the computing resource, the models were simplified and shown in Fig. 5(a) and 5(b) for PFE and RFE. The calculated electric-field distributions of the PFE and RFE at an applied voltage of 600 V (4 V/um) were shown in Fig. 5(c). It is found that the electric field of the PFE is significantly higher at the edge than that at the center, whereas the electric field of the RFE keeps constant all over the emitter surface. The electric field at the edge of the emission cell (15 V/um) was much higher than that at the center (4.1 V/um). These results indicate that the electric field is predominantly concentrated in CNTs at the periphery of the cell, which acts as major emission sites. Therefore, the highly efficient field emission can be attributed to the edge effect that was significantly improved by the micromachining patterning process.
The field emission properties affected by different cell patterns were investigated by using CST particle studio. A simplified 3D model was built (to save the computing resource), the distance between the anode and the cathode is 150 um, the height of the emitter is 20 um, the thickness of the anode and the cathode is 50 um, and the diameter of the anode and the cathode are both 2 mm. Field emission cathode of different patterns (Triangle, Square, Pentagon, Hexagon, Octagon, and Round) were shown in Fig. 6(a), each model composed of 4 cells with an area of 2.54 × 10−4 cm2. And all the cathodes were centered at the origin of the coordinates (0, 0, 0) respectively.
As shown in Fig. 6(b), applied with the same voltage, the hexagon and round arrays emit the highest current, followed by the pentagon, the triangle, the square and the octagon arrays. Take applying voltage of 5,000 V as an example, the emission current densities for the triangle, square, pentagon, hexagon, octagon, and round cell are 5.31, 5.27, 6.85, 7.28, 5.48, and 7.11 A/cm2, respectively. From the results above, we can see that the field emission properties of different arrays differ from each other. Cell patterns optimization plays an important role in improving the field emission properties of the cathode.
The field emission characteristics of the field emitter before and after annealing were tested in a vacuum chamber. The relationship of the emission current with the applied voltage is shown in Fig. 7. Both of the emission currents increased monotonically with the applied field. The turn-on field, that was defined as an electric field required to get an emission current of 10 uA, were 351 V and 249 V for the pristine and the annealed emitter respectively. We simply consider the total area of the emission cells on the emitter as the emission area. The area S = 51*π*(Ø /2)2 = 0.013 cm2. For the annealed emitter, with the applied voltage of 781 V, the field emission current of 3.08 mA (current density of 241 mA/cm2) was achieved. The corresponding Fowler-Nordheim (F-N) plot for the flexible emitter is shown in Fig. 7 inset. All dots on the curve fit a single straight line well, which implies that the field emission process follows the F-N mechanism.
In order to investigate the structural change induced by the annealing process, CNT emission cells at the initial and the final annealing process were measured for comparison using XPS and Raman (Fig. 8). XPS spectra of the emission cell with binding energy ranging from 0 to 1200 eV is shown in Fig. 8(a). Typical O1s peak centered at 532 eV is clearly observed in the XPS spectrum before the annealing process, which is corresponding with oxygen contaminants like C = O  and C-O . The O1s peak becomes weaker in the XPS spectrum after annealing. Figure 8(a) inset shows the fitted results of the C1s peak for the emission cell after annealing. The C1s peak centers at 284.1 eV in the pristine curve while moves to 283.3 eV in the annealed curve, this 0.8 eV left shift (inset of Fig. 8(a)) also confirms the oxygen desorption. In the C-O bond, carbon is positively charged and oxygen is negatively charged. The binding energy in carbon decreased while the binding energy in oxygen increased as a result of the broken C-O bonds. Therefore, it is no wonder that the binding energy of C1s peak decreases after the oxygen desorption. Combining with the emisison properties of the CNT emitter shown in Fig. 7, the decreased binding energy of carbon induced by oxygen desorption is supposed to ascend the Fermi Lever, yielding the decrease of work function, which is beneficial for decreasing the electron tunneling barriers during field emission and improving the emission behavior. Information on structural evolution can also be obtained from the fitted results of the XPS C1s peaks of the emitter (Fig. 8(b)). Three peaks centered at 283.5, 284.8, and 288 eV are corresponding with graphite carbon, diamond-like carbon, and organic contaminated carbon, respectively . The percentage of SP2-hybridized carbon (SP2%) was used to evaluate the crystallinity of the emission cells and find that SP2% increased from 62.5% to 71.6% after annealing, indicating that part of the intrinsic CNT defects have been annealed at the annealing temperature of 800°C.
Further insights of the structural evolution were obtained from the Raman spectra (Fig. 8(c)). The D bond at 1350 cm−1 is generally attributed to defects in the curved graphite sheet or other impurities, while the G bond at 1580 cm−1 is corresponding to the opposite direction movement of two neighboring carbon atoms in a graphitic sheet, and it indicates the presence of crystalline graphitic carbon in CNTs . The slight decrease of the intensity ratio between D peak and G peak (ID/IG) after annealing is another evidence for the defects annealing. Defects, especially vacancy-related defects, increase emission current by providing more effective electron transmission traces, so the annealing of defects is detrimental to FE (which is only related to the J-E behavior) [18, 19]. However, comparing with oxygen desorption induced FE improvement, this defect annealing induced FE degradation is negligible, the J-E plots still shift to the low applied field region (left-shift shown in Fig. 8(a) inset). On the other hand, defect annealing is good for improving the emission stability in a way of reinforcing the defective locations which are more likely to be burned off by Joule heat during FE [20, 21]. The competition between oxygen desorption induced FE improvement and defect annealing induced FE degradation determines the field emission behavior of the CNT emitter.
The short-term stability of the emitter was firstly evaluated by monitoring emission current under pulsed voltage operation for 40 h. As shown in Fig. 9, with applied field of 4.63, 4.08, 3.19, 1.91 V/um, emission current of 1.97, 1.31, 0.66, 0.13 mA could be achieved, and they remained almost constant during the 40 h continuous measurement, and the fluctuation width of the emission current for 40 hours were all in ± 5%. One thing to note here is that a few arcing events occurred when the emission current reached higher than 1 mA, however, the emitter could withstand the arcing and the emission current remained constant with time. We believe that the annealing of defects in a way of reinforcing the defective locations which are more likely to be burned off by Joule heat during the field emission plays an important role in its highly stable emission performance [22, 23].
For comparison, we have included some of the best values obtained from a few past high temperature processable CNT emitters. It is evident from our data that our emitter fabricated with sodium silicate exhibited excellent field emission properties (Table 1). Although the maximum current density for our emitter is relatively low (3.08 mA), it is enough for micro focus x-ray applications. The emitter reported in the present work showed good stability under high emission current. The fluctuation width of the emission current at 1.97 mA was in ± 5%, which was seldom reported in the past high temperature processable emitters.
In this section the experiments are presented that proof the X-ray generation with the electron source described in the previous sections. The fabricated emitter was sealed in a conventional X-ray tube, as shown in Fig. 10(b) and 10(c). The diameter of the glass shell around the anode is 30 mm, while the diameter of the glass shell around the cathode is 20 mm. All of the connection parts of the X-ray tube were tightly vacuum-sealed. The components of the X-ray tube were baked at 550°C for 12 h, and subsequently, these were brazed through a single-step brazing process at 780°C for 30 min in a vacuum furnace.
The distance between cathode and anode was maintained at 10 mm. The proof of the X-ray creation was done by using an X-ray sensitive film. This film was commercially available (Kodak Insight, 31 × 41 mm2) and widely used in dental diagnostics as standard X-ray analogue film plates. The detection was done by placing the film in front of the anode outside the glass shell. The sealed X-ray tube was successfully operated at 20 KV with an extraction current from the cathode of 214 uA. As the emitter was fixed in a “cap”, the electric field between cathode and anode was complex. The X-ray tube can be operated under a relatively low voltage comparing with other works [13, 25, 26]. A round badge with a thickness of 100 um and diameter of 3 mm was put in front of the sensitive plate in order to create a defined pattern when exposed to radiation, the shadow of the aperture thus unequivocally confirms the X-ray emission. The sensitive plate and the badge were exposed to the radiation under emission current of 214 uA for 30 min to compensate the X-ray energy loss. The above exposed plate was then developed in accordance with the photographic processing. The photograph of the badge and the developed copy can be seen in Fig. 10. The blackened shade on this copy confirms the X-ray generation, which demonstrates that our emitter successfully survived the high temperature processing steps of the X-ray source fabrication.
In conclusion, we introduced a novel method by using micromachining for fabricating patterned CNT field emitters using for X-ray generation. Ball milling process was employed to improve the CNT dispersion. The emitter was shown to resist vacuum annealing at 800°C, it is therefore no wonder that it can withstand the necessary brazing steps for vacuum-tight sealing, which is relevant for X-ray source fabrication. Micromachining patterning process, which enhanced the edge effect, significantly improved its field emission performance. Based on the analysis from XPS, we ascribed the improved emission properties after the annealing process to the oxygen desorption induced decrease of work function, and defect annealing which could stabilize the emitter. The emitter shows relatively good field-emission properties such as high current density (237 mA/cm2 at an applied voltage of 781 V), low turn-on field (1.25 V/um), and good stability (40 h for 10% fluctuation around 2 mA). To check the application of the emitter, the fabricated emitter was sealed in a conventional X-ray tube. The blackened shade on the sensitive film confirmed the X-ray generation, which demonstrated that our emitter successfully survived the high temperature processing steps of the X-ray source fabrication. From those results, it is believed that, this new method based on micromachining can be helpful for wide application of CNT based cathode in X-ray tube, and further optimization in device configuration and cathode geometry is still processing.
The authors express their sincere gratitude to the colleagues of National Key Laboratory of Nano/Micro Fabrication Technology, thanks for their support and encouragement. The authors would like to thank supports from the National Natural Science Foundation of China (No. 51305265, No. 51205390) and the Research Fund for the Doctoral Program of Higher Education of China (No.20120073110061).
References and links
1. W. B. Choi, D. S. Chung, J. H. Kang, H. Y. Kim, Y. W. Jin, I. T. Han, Y. H. Lee, J. E. Jung, N. S. Lee, G. S. Park, and J. M. Kim, “Fully sealed, high-brightness carbon-nanotube field-emission display,” Appl. Phys. Lett. 75(20), 3129 (1999). [CrossRef]
2. S. Fan, M. G. Chapline, N. R. Franklin, T. W. Tombler, A. M. Cassell, and H. Dai, “Self-oriented regular arrays of carbon nanotubes and their field emission properties,” Science 283(5401), 512–514 (1999). [CrossRef] [PubMed]
3. M. Deng, G. F. Ding, Y. Wang, H. Q. Wu, Y. J. Yao, and L. D. Zhu, “Fabrication of Ni-matrix carbon nanotube field emitters using composite electroplating and micromachining,” Carbon 47(15), 3466–3471 (2009). [CrossRef]
4. H. C. Wu, M. J. Youh, W. H. Lin, C. L. Tseng, Y. M. Juan, M. H. Chuang, Y. Y. Li, and A. Sakoda, “Fabrication of double-sided field-emission light source using a mixture of carbon nanotubes and phosphor sandwiched between two electrode layers,” Carbon 50(13), 4781–4786 (2012). [CrossRef]
5. J. S. Park, J. P. Kim, Y. R. Noh, K. C. Jo, S. Y. Lee, H. Y. Choi, and J. U. Kim, “X-ray images obtained from cold cathodes using carbon nanotubes coated with gallium-doped zinc oxide thin films,” Thin Solid Films 519(5), 1743–1748 (2010). [CrossRef]
6. S. H. Heo, A. Ihsan, and S. O. Cho, “Transmission-type microfocus x-ray tube using carbon nanotube field emitters,” Appl. Phys. Lett. 90, 183109 (2007).
7. B. Gao, G. Z. Yue, Q. Qiu, Y. Cheng, H. Shimoda, L. Fleming, and O. Zhou, “Fabrication and electron field emission properties of carbon nanotube films by electrophoretic deposition,” Adv. Mater. 13(23), 1770–1773 (2001). [CrossRef]
8. H. Jung, S. Y. An, D. M. Jang, J. M. Kim, J. Y. Park, and D. Kim, “A multi-wall carbon nanotube/polymethyl methacrylate composite for use in field emitters on flexible substrates,” Carbon 50(3), 987–993 (2012). [CrossRef]
9. X. Calderon-Colon, H. Z. Geng, B. Gao, L. An, G. H. Cao, and O Zhou, “A carbon nanotube field emission cathode with high current density and long-term stability,” Nanotechnology 20, 325707 (2009).
10. W. Zhu, C. Bower, O. Zhou, G. Kochanski, and S. Jin, “Large current density from carbon nanotube field emitters,” Appl. Phys. Lett. 75(6), 873–875 (1999). [CrossRef]
11. Y. Saito, Carbon Nanotube and Related Field Emitters: Fundamentals and Applications (John Wiley & Sons, 2010).
12. S. Delpeux, F. Beguin, R. Benoit, R. Erre, N. Manolova, and I. Rashkov, “Fullerene core star-like polymers—1. Preparation from fullerenes and monoazidopolyethers,” Eur. Polym. J. 34, 905–915 (1998). [CrossRef]
14. R. Longtin, H. R. Elsener, J. R. Sanchez-Valencia, D. Cloetta, L. O. Nilsson, C. Leinenbach, O. Groning, and P. Groning, “High-temperature processable carbon-silicate nanocomposite cold electron cathodes for miniature X-ray sources,” J. Mater. Chem. 1(7), 1368–1374 (2013). [CrossRef]
15. Y. Li, X. Cheng, W. Cao, L. Gong, R. Zhang, and H. Zhang, “Development of adiabatic foam using sodium silicate modified by boric acid,” J. Alloys Compd. 666, 513–519 (2016). [CrossRef]
16. J.-H. Deng, Y. Yang, R. Zheng, and G. Cheng, “Temperature dependent field emission performances of carbon nanotube arrays: Speculation on oxygen desorption and defect annealing,” Appl. Surf. Sci. 258(18), 7094–7098 (2012). [CrossRef]
17. S. D. Gardner, C. S. K. Singamsetty, G. L. Booth, G.-R. He, and C. U. Pittman Jr., “Surface characterization of carbon fibers using angle-resolved XPS and ISS,” Carbon 33(5), 587–595 (1995). [CrossRef]
18. G. Kim, B. W. Jeong, and J. Ihm, “Deep levels in the band gap of the carbon nanotube with vacancy-related defects,” Appl. Phys. Lett. 88(19), 193107 (2006). [CrossRef]
19. G. Wei, “Emission property of carbon nanotube with defects,” Appl. Phys. Lett. 89(14), 143111 (2006). [CrossRef]
20. K. A. Dean, T. P. Burgin, and B. R. Chalamala, “Evaporation of carbon nanotubes during electron field emission,” Appl. Phys. Lett. 79(12), 1873–1875 (2001). [CrossRef]
21. N. Zhao, J. Chen, K. Qu, Q. Khan, W. Lei, and X. Zhang, “Stable electron field emission from carbon nanotubes emitter transferred on graphene films,” Physica E 72, 84–88 (2015). [CrossRef]
22. J. H. Park, J. S. Moon, J. H. Han, A. S. Berdinsky, J. B. Yoo, C. Y. Park, J. W. Nam, J. Park, C. G. Lee, and D. H. Choe, “Stable and high emission current from carbon nanotube paste with spin on glass,” J. Vac. Sci. Technol. B 23(2), 702–706 (2005). [CrossRef]
24. J.-W. Kim, J.-W. Jeong, J.-T. Kang, S. Choi, S. Park, M.-S. Shin, S. Ahn, and Y.-H. Song, “Great improvement in adhesion and uniformity of carbon nanotube field emitters through reactive nanometer-scale SiC fillers,” Carbon 82, 245–253 (2015). [CrossRef]
25. J.-W. Jeong, J.-T. Kang, S. Choi, J.-W. Kim, S. Ahn, and Y.-H. Song, “A digital miniature x-ray tube with a high-density triode carbon nanotube field emitter,” Appl. Phys. Lett. 102(2), 023504 (2013). [CrossRef]
26. J. W. Jeong, J. W. Kim, J. T. Kang, S. Choi, S. Ahn, and Y. H. Song, “A vacuum-sealed compact x-ray tube based on focused carbon nanotube field-emission electrons,” Nanotechnology 24(8), 085201 (2013). [CrossRef] [PubMed]
27. R. Longtin, H.-R. Elsener, J. R. Sanchez-Valencia, D. Cloetta, L.-O. Nilsson, C. Leinenbach, O. Gröning, and P. Gröning, “High-temperature processable carbon–silicate nanocomposite cold electron cathodes for miniature X-ray sources,” J. Mater. Chem. C 1(7), 1368 (2013). [CrossRef]