Boise State University

ScholarWorks

Electrical and Computer Engineering Faculty Department of Electrical and Computer

Publications and Presentations Engineering

1-2022

Effect of Room Air Exposure on the Field Emission Performance

of UV Light Irradiated Si-Gated Field Emitter Arrays
Ranajoy Bhattacharya
Boise State University

Mason Cannon
Boise State University

Rushmita Bhattacharjee
Boise State University

Jim Browning
Boise State University

Publication Information
Bhattacharya, Ranajoy; Cannon, Mason; Bhattacharjee, Rushmita; Rughoobur, Girish; Karaulac, Nedeljko;
Chern, Winston; . . . and Browning, Jim. (2022). "Effect of Room Air Exposure on the Field Emission
Performance of UV Light Irradiated Si-Gated Field Emitter Arrays". Journal of Vacuum Science and
Technology B, 40(1), 010601. https://doi.org/10.1116/6.0001593

This article may be downloaded for personal use only. Any other use requires prior permission of the author and
AIP Publishing. This article appeared in
Bhattacharya, R., Cannon, M., Bhattacharjee, R., Rughoobur, G., Karaulac, N., Chern, W., . . . & Browning, J. (2022).
Effect of Room Air Exposure on the Field Emission Performance of UV Light Irradiated Si-Gated Field Emitter
Arrays. Journal of Vacuum Science and Technology B, 40(1), 010601.
and may be found at https://doi.org/10.1116/6.0001593.
 LETTER avs.scitation.org/journal/jvb

Effect of room air exposure on the field emission

performance of UV light irradiated Si-gated field
emitter arrays
Cite as: J. Vac. Sci. Technol. B 40, 010601 (2022); doi: 10.1116/6.0001593
Submitted: 2 November 2021 · Accepted: 16 November 2021 · View Online Export Citation CrossMark
Published Online: 20 December 2021

Ranajoy Bhattacharya,1 Mason Cannon,1 Rushmita Bhattacharjee,1 Girish Rughoobur,2

2
Nedeljko Karaulac, Winston Chern,2 Akintunde Ibitayo Akinwande,2 and Jim Browning1,a)

AFFILIATIONS
1
Department of Electrical and Computer Engineering, Boise State University, 1375 W University Dr., Boise, Idaho 83725
2
Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, 77 Massachusetts Ave,
Cambridge, Massachusetts 02139

a)
Electronic mail: jimbrowning@boisestate.edu

ABSTRACT
Field emission cathodes are promising candidates in nanoscale vacuum channel transistors and are used in microwave vacuum electron
devices. Prior research has shown that UV light exposure as well as 350 °C vacuum bake can desorb water vapor from Si field emission tips,
resulting in lower work function and improved emission performance. However, after long exposure to room air (greater than 24 h), the
improved performance is lost as water adsorbs on the tips. In this study, experiments were carried on two sets of 1000 × 1000 Si-gated field
emitter arrays to determine the length of time that emitters can be exposed to room air without degradation. First, the samples were
exposed to UV light irradiation in vacuum, and the I–V curves were measured. Then, the samples were exposed to room air with a relative
humidity ranging from 30% to 40% for varying times (5, 6, 8, 12, 24, and 48 h) and then tested again under high vacuum. It was found that
the emission current did not degrade after room air exposure of 5 h. However, at 6 h of exposure, degradation started to occur, and after 24
h, the emission current went back to the original, pre-UV exposure case. In a separate experiment, UV irradiated samples were stored in
nitrogen for 72 h, with a 10% degradation in current. These results demonstrate that field emission devices with improved performance
resulting from water desorption can be handled in air up to 5 h, depending upon humidity and stored in nitrogen for 72 h while maintain-
ing improved performance.

Published under an exclusive license by the AVS. https://doi.org/10.1116/6.0001593

I. INTRODUCTION
Field emitters (FEs) are being studied as an electron source for
vacuum transistors1 and microwave vacuum electron devices
(VEDs)2,3 and are an attractive choice for high frequency and
harsh-environment electronics due to their radiation resistance,
high temperature capability,4,5 and collisionless electron trans-
port.1,6,7 Silicon based tips are some of the most well-developed8–10
field emitters among different materials. Sharp tips (<10 nm) and
reduced gate-emitter distance have decreased the operating voltage
in Si FEs8,11 to <20 V.6,12 Recently, efforts to combine nanoscale
vacuum channels with the present-day semiconductor fabrication
method has begun to realize vacuum nanoscale devices with high
frequency and fast switching. In a recent study, Han et al.6 demon-
strated an NVCT with a construction comparable to that of the
standard field effect transistor (FET) where the electron emission
takes place due to the tunneling mechanism. Also, the nanoscale
vacuum channel effectively reduces scattering during electron trans-
port, demonstrating its inherent precedence of lower power con-
sumption9,13 and fast transport. More importantly, the Si-GFEAs
has been demonstrated to be insusceptible to high temperatures5
(400 °C) and radiation environment.14 Moreover, with improved
electrode material such as metal or other low dimensional15 materi-
als, e.g., graphene,16,17 NVCTs are an excellent candidate for
on-chip,4,18,19 integrated20 vacuum electron devices. Recent develop-
ments demonstrated that vacuum nanoelectronics can obtain the
device dimensions21 of present-day solid state devices.
To implement NVCTs, it is necessary to understand the phys-
ical processes that control their performance and lifetime including

J. Vac. Sci. Technol. B 40(1) Jan/Feb 2022; doi: 10.1116/6.0001593 40, 010601-1
Published under an exclusive license by the AVS
 LETTER avs.scitation.org/journal/jvb

field emission current and surface leakage between electrodes. In

our prior work, we have studied the effects of water adsorbed on
gated field emission arrays (GFEAs).5,22 In that work, we studied
the field emission performance of 1000 × 1000 Si-GFEAs that were
heated to 400 °C or exposed to ultraviolet (UV) irradiation. Those
studies demonstrated that water vapor was desorbed from the
arrays at ∼350 °C or by the UV exposure time of greater than 80
min. The resulting water desorption increased emission current by
greater than 10 times because of a lower tip work function and the
resulting increased number of emitting tips at the same voltage.
The desorbed water also resulted in greater than 10 times reduction
of the emitter to gate surface leakage, demonstrating that the
adsorbed water was primarily responsible for the leakage path.
Once the GFEAs were returned to room air for greater than 24 h,
the devices returned to their original operating conditions. While
these results are interesting for in situ experiments and for under-
standing the fundamental device physics, it is important to under-
stand the ability to use such techniques in other processes or
applications. FEs in complicated structures such as microwave
vacuum electron devices or in low temperature NVCT packaging
processes may not be able to be UV irradiated or heated to 350 °C
because of the geometric limitations or device compatibility.
Hence, this research looks at the time limits allowed for GFEAs to
be removed from vacuum after UV exposure before water adsorp-
tion causes a decline in performance. In this work, the GFEAs were
exposed to UV light22 for 100 min and then exposed to air for
several fixed time periods, and then the devices were characterized
electrically again in high vacuum. In a separate set of tests, these
GFEAs were exposed to UV light then stored in N2 and again
tested under high vacuum. The experimental setup and procedures
are described in Sec. II, and the results are discussed in Sec. III.

II. EXPERIMENTAL SETUP

The field emitter arrays, built on 150 mm diameter, n-type
doped, single crystal, silicon wafers, have array sizes ranging from
single emitters to 1000 × 1000 emitter arrays. Figure 1(a) shows the FIG. 1. (a) 3D rendering of the device structure. For clarity, layers have been
3D schematic and Fig. 1(b) shows the SEM image of the emitter.8 omitted in different regions of the rendering to show detail. In the front, the bare
The field emitter tips have a 1 μm spacing (pitch). These arrays silicon nanowires (200 nm diameter and 10 μm height) with sharp tips are
shown. (b) SEM image of the completed device at 45° tilt (Ref. 8).
consist of self-aligned gated silicon emitters with integrated nano-
wires of 150 nm diameter and an 8 μm height. Detailed fabrication
methods and descriptions can be found elsewhere.9,23 The
1000 × 1000 arrays were chosen for these experiments because they
(SMU, Keysight B2902A) to sweep the drive voltage and
are able to produce a current density greater than 100 A/cm2.9
measure the gate and collector currents. The gate to emitter
Current-voltage (I–V) characterization measurements were carried
voltage, VGE, was swept from 0 to 40 V DC with a collector voltage
out inside a stainless-steel vacuum chamber. The chamber is
of 200 V DC. The following procedure was used for the air expo-
equipped with electrical feedthroughs, thermocouple feedthroughs,
sure experiments:
a three-axis manipulator probe arm, and an Extorr Inc. XT100
residual gas analyzer (RGA). The system includes a UV lamp (RBD
Instrument, Model- MiniZ, 350 μW/cm2, 2.7500 CF flange 1. The GFEAs are tested, and I-–V curves measured for each
mounted) that is used to desorb water vapor from the tips. A device after a pressure of <10−7 Torr is achieved in the test
detailed description of the chamber and the experimental test setup chamber.
can be found in our previous works.5,22 2. The GFEAs are exposed to UV for 100 min. Water vapor pres-
For the experiments described here, two different 1000 × 1000 sure is monitored.
GFEAs were tested on the same wafer section. The same two 3. I–V curves are again measured for each device.
devices were used for all of the results shown here. The I–V mea- 4. The chamber is vented with N2 to atmospheric pressure, and
surements were performed using a source measurement unit the devices are removed.

J. Vac. Sci. Technol. B 40(1) Jan/Feb 2022; doi: 10.1116/6.0001593 40, 010601-2
Published under an exclusive license by the AVS

5. The GFEAs are stored in room air (relative humidity
measured from 30% to40%) under a laminar flow hood for
varying times.
6. The GFEAs are returned to the chamber, and steps 1–5 are
repeated including exposure to UV after each I–V measurement
and then measured again after UV.
In the final experiment described here, the GFEAs were left in
the vacuum test chamber for 72 h after the N2 vent to simulate
storage in an N2 box. While this is not exactly the same as remov-
ing and storing in a purge box, it is considered a close approxima-
tion. For each of air exposure experiments, the tests were repeated
two times.
FIG. 2. I–V characteristics as a function of the air exposure time for (a) die 1 and (c) die 2 and corresponding F–N plots of (b) die 1 and (d) die 2.
J. Vac. Sci. Technol. B 40(1) Jan/Feb 2022; doi: 10.1116/6.0001593
Published under an exclusive license by the AVS
LETTER avs.scitation.org/journal/jvb
III. RESULTS AND DISCUSSION
The experiments were completed on the same two GFEA
structures with air exposure times of 5, 6, 8, 12, 24, and 48 h.
Shown in Fig. 2 are the I–V and Fowler–Nordheim (F–N)24–27
plots of the two devices before and after UV exposure for the first
test sequence and then after air exposure for the remaining mea-
surements. As can be observed for both devices, the I–V curves are
essentially the same for the 5 and 6 h cases compared to the
post-UV cases, but as air exposure time increases, there is clear
degradation in emission and a change in the F–N plots linearity,
particularly, at higher voltage (greater than 30 V). As explained
earlier, the water vapor increases the emitter tip work function and
40, 010601-3
 LETTER avs.scitation.org/journal/jvb

FIG. 3. Degradation of collector current (normalized to after UV exposure) of

die 1 and die 2 vs air exposure time. At 48 h, the current has degraded to the
pre-UV exposure level. Also shown is the before UV reference and on the right
side of the graph, normalized collector current after 72 h of N2 storage of die 1
and 2 are plotted.

reduces the number of tips that emit. However, the cause of the
slope change28 in the F–N plot for the case before UV and 12 h of
air exposure, for both the die, is not understood and needed to be
studied separately.
To see the water adsorption effects more clearly, the normal-
ized collector current at 40 V on the gate electrode is plotted in
Fig. 3 versus air exposure time. The current at each time point is
normalized to the current after UV exposure. Note that after each
air exposure experiment, the devices were again exposed to UV and
the I–V characteristics measured to ensure there was no other deg-
radation occurring. As can be seen, the current begins degrading at
6 h and continues to degrade rapidly after that. The current levels
reach the pre-UV treatment levels around 24 h after UV exposure. FIG. 4. (a) I–V characterization graph of die 1 and die 2 for before and after 72
This level is also shown on the plot. h of storage in N2 gas. (b) Corresponding F–N plot.

TABLE I. Summary of device F–N coefficients.

From the F–N plots, it can also be observed that the intercept
F–N intercept, ln becomes less negative after UV exposure, suggesting that more tips
F–N slope, bFN (V) (aFN) (A/V2) are emitting as a result of the water vapor desorbing from the
surface.20 Also, it is evident that the water desorption does not
Parameters Die 1 Die 2 Die 1 Die 2 damage the emitters as the F–N slopes are very similar between 20
Before UV 308.229 306.33 −8.008 −7.23 and 35 V for all the sweeps. For an example, the values of F–N
After UV 356.1475 369.68 −4.07 −3.03 coefficients25,26 bFN and ln(aFN) were extracted from the slope and
5 h. 347.69 360.37 −3.8 −3.36 intercept of the F–N plot shown in Fig. 2(b). The extracted bFN and
6h 334.67 359.94 −4.74 −3.51 ln (aFN) are shown in Table I. The decrease in bFN with air expo-
8h 327.255 338.20 −4.465 −4.65 sure time indicates either an increase in the field enhancement
12 h. 317.63 334.67 −5.98 −5.22 factor,25 β, or a decrease in the work function. A decrease in the
24 h. 316.145 310.322 −6.49 −6.78 work function is inconsistent with water vapor adsorption, which
48 h. 303.18 310.37 −7.77 −7.01 should cause an increase in the work function. However, a decrease
in the emission current and in the magnitude of the ln(aFN) is also

J. Vac. Sci. Technol. B 40(1) Jan/Feb 2022; doi: 10.1116/6.0001593 40, 010601-4
Published under an exclusive license by the AVS

observed, indicating the proposed increment in the work function as
expected from water adsorption. The increase in β is due to a change
in the distribution of the emitting tips to the sharper emitters.22
The second experiment shows the results of storage for 72 h
of the devices in N2 after venting. After the 72 h of N2 storage, the
chamber was pumped back to <10−7 Torr. In Fig. 4, the I–V curves
and F–N plots of the same two devices are plotted after sitting in
the test chamber in N2 at atmospheric pressure. The primary result
is that there is roughly 10% degradation in the collector current
after 72 h, so clearly the storage allows for much longer storage
times as expected.
The F–N plots show clearly some level of degradation. Note
that the legend indicates that the data are for the third I–V sweep.
These data are shown because in the first 2 I–V sweeps, the emit-
ters showed increasing current. Hence, some contaminant (likely
water) was on the surface, but sweeping the voltage caused this
water to desorb, so the final sweep is shown.
These normalized current data points at 40 V are also shown
in Fig. 3 with the points again normalized to the after UV condi-
tion. This plot clearly demonstrates that the N2 storage reduced
water adsorption with only a 10% degradation of die 1 and 6% for
die 2. These devices were again exposed to UV light for 100 min,
and it was found that the collector current was restored to the before
storage condition where the difference is ∼1.5% for both the devices.
IV. SUMMARY AND CONCLUSIONS
I–V characterization of Si-GFEA with 1000 × 1000 tip arrays
was carried out after UV light exposure and then after several
hours of room air exposure at a relative humidity of 30%–40%. It
was found that the collector current remains unaffected for up to 5
h of air exposure. However, around 6 h of air exposure, the collec-
tor current begins to degrade, and after 24 h of air exposure, the
current degrades to pre-UV condition because of water adsorption.
From this result, it can be concluded that field emission devices
can be treated in vacuum (high temperature or UV exposure) to
clean the surfaces of water vapor, taken out from high vacuum
using an N2 vent, and then be exposed to room air for ≈5 h
depending upon humidity. This air exposure time window allows
these devices to be mounted or configured for other experiments
where UV exposure or high temperature are not practical. Such a
capability provides much greater flexibility for applications of
field emission devices both in VEDs as well as for NVCTs. In
another set of experiments, it was observed that the devices can
be stored in N2 gas for up to 72 h with only 10% degradation.
Again, these results demonstrate that devices can be stored and
utilized as needed after the UV exposure while maintaining per-
formance. While such techniques are not new, these results here
demonstrate the application of this phenomenon to silicon
GFEAs. In future work, different emitter materials, GaN and TiN,
will also be studied as well as the venting results using different
gases (He or Ar).
ACKNOWLEDGMENTS
Material support for this work is provided by the Air Force
Office of Scientific Research under Grant No. FA9550-18-1-0436.
The authors would also like to thank the support from
J. Vac. Sci. Technol. B 40(1) Jan/Feb 2022; doi: 10.1116/6.0001593
Published under an exclusive license by the AVS
LETTER avs.scitation.org/journal/jvb
undergraduate students Patrick Ward, John McClarin, Robby
Davis, Liz Gaffney, Jessica Carlson, David Vogel, and Gerardo
Herrera.
DATA AVAILABILITY
The data that support the findings of this study are available
from the corresponding author upon reasonable request.
REFERENCES
1
J.-W. Han, M.-L. Seol, D.-I. Moon, G. Hunter, and M. Meyyappan, Nat.
Electron. 2, 405 (2019).
2
A. K. Singh, S. K. Shukla, M. Ravi, and R. K. Barik, IEEE Trans. Plasma Sci. 48,
3446 (2020).
3
R. L. Ives, IEEE Trans. Plasma Sci. 32, 1277 (2004).
4
W. P. Kang, J. L. Davidson, Y. M. Wong, and K. Holmes, Diamond Relat.
Mater. 13, 975 (2004).
5
R. Bhattacharya, N. Karaulac, W. Chern, A. I. Akinwande, and J. Browning,
J. Vac. Sci. Technol. B. 39, 023201 (2021).
6
J.-W. Han, D.-I. Moon, and M. Meyyappan, Nano Lett. 17, 2146 (2017).
7
J.-W. Han, J. Sub Oh, and M. Meyyappan, Appl. Phys. Lett. 100, 213505 (2012).
8
S. A. Guerrera and A. I. Akinwande, Nanotechnology 27, 295302 (2016).
9
S. A. Guerrera and A. I. Akinwande, IEEE Electron Device Lett. 37, 96 (2016).
10
S. A. Guerrera, L. F. Velasquez-Garcia, and A. I. Akinwande, IEEE Trans.
Electron Devices 59, 2524 (2012).
11
L. F. Velasquez-Garcia, S. A. Guerrera, Y. Niu, and A. I. Akinwande, IEEE
Trans. Electron Devices 58, 1783 (2011).
12
J. Xu, Z. Gu, W. Yang, Q. Wang, and X. Zhang, Nanoscale Res. Lett. 13, 311 (2018).
13
D. G. Pflug, “Low voltage field emitter arrays through aperture scaling” Ph.D.
dissertation (Massachusetts Institute of Technology, 2000).
14
J. L. Davidson, W. P. Kang, K. Subramanian, A. G. Holmes-Siedle, R. A. Reed,
and K. F. Galloway, IEEE Trans. Nucl. Sci. 56, 2225 (2009).
15
Q. Tang and Z. Zhou, Prog. Mater. Sci. 58, 1244 (2013).
16
I.-K. Baek, R. Bhattacharya, J. S. Lee, S. Kim, D. Hong, M. A. Sattorov,
S.-H. Min, Y. H. Kim, and G.-S. Park, J. Electromagn. Waves Appl. 31, 2064
(2017).
17
R. Bhattacharya, “Highly electrically conductive tungsten doped reduced gra-
phene oxide film as large current density field emitter” Ph.D. dissertation
(National University, Seoul, 2018).
18
Y. Yang, M. Turchetti, P. Vasireddy, W. P. Putnam, O. Karnbach, A. Nardi,
F. X. Kärtner, K. K. Berggren, and P. D. Keathley, Nat. Commun. 11, 3407
(2020).
19
W. P. Kang, J. L Davidson, K. Subramanian, B. K. Choi, and K. F. Galloway,
IEEE Trans. Nucl. Sci. 54, 1061 (2007).
20
R. Greene, H. Gray and G. Campisi, “Vacuum integrated circuits,” 1985
International Electron Devices Meeting, Washington DC (1985).
21
D. Pavlidis, E. Bellotti, S. Graham, S. Mohammadi, S. Rajan, and J. Redwing,
“III-Nitrides for vacuum nanoelectronics,” in ECS Meeting Abstracts (IOP
Publishing, 2021), p. 1068, No. 33.
22
R. Bhattacharya, N. Karaulac, G. Rughoobur, W. Chern, A. I. Akinwande, and
J. Browning, J. Vac. Sci. Technol. B 39, 033201 (2021).
23
A. A. Fomani, S. A. Guerrera, L. F. Velásquez-García, and A. I. Akinwande,
IEEE Trans. Electron Dev. 61, 2538 (2014).
24
R. H. Fowler and L. Nordheim, Proc. R. Soc. London, A 119, 173 (1928).
25
E. L. Murphy and R. H. Good Jr, Phys. Rev. 102, 1464 (1956).
26
R. G. Forbes, “21st century planar field emission theory and its role in vacuum
breakdown science,” 2020 29th International Symposium on Discharges and
Electrical Insulation in Vacuum (ISDEIV), Padova, Italy (2021), pp. 3–10.
27
R. G. Forbes, Mod. Dev. Vac. Electron Sources 135, 387 (2020)
28
N. S. Xu, Y. Chen, S. Z. Deng, J. Chen, X. C. Ma, and E. G. Wang, J. Phys. D:
Appl. Phys. 34, 1597 (2001).
40, 010601-5