Effects of gases on the field emission performance of silicon gated field emitter array

Field emitters are now widely discussed as an electron source for nanovacuum channel transistors.^1–5 In field emitters, emission of electrons takes place by quantum mechanical tunneling from a micro- or nanometer scale tip by applying a positive potential on a gate electrode that surrounds the tip.^6–9 During the operation of these devices, high vacuum is required to reduce the gas-electron collision probability between the electrodes. Field emitters can be used as a number of vacuum microelectronic devices, such as harsh environment electronics.^10–12 One major hurdle to the successful application of silicon-gated field emitter arrays (GFEAs) is the rapid degradation of field emission current during operation.^13–15 It is well known that field emission is quite sensitive to the surface condition^16,17 and contamination of emitters, which can cause a problem when field emitters are vacuum packaged. Residual gases, particularly oxygen and water vapor not removed by the heat treatment, or gas desorbed by the electron bombardment of device components is known to interact with GFEAs and result in reduction in the field emission current. Research work on GFEAs has focused on assimilation of degradation mechanisms caused by oxidation of the tip material from residual gases, primarily oxygen and other gases containing oxygen.^18,19 In contrast, previous works have demonstrated that the presence of hydrogen gas can improve the field-emission current of metal-based field emitter arrays.^18,20 However, effects of gases on Si-GFEAs are reported to result in reduced field emission current.²¹ Only a few systematic study that examines the effects of different inert gases and different pressures, especially for an extended period of time, has been reported previously.¹⁹ Effects of inert (Ar and He) and less reactive (N₂) gasses are important as they can be used as the backfilled gas during the packaging process. Also, most of the published work discussed about the effect of injected gas without considering the effect of water vapor associated with it. In this paper, we present experimental results conducted using silicon-gated field emitter arrays (Si-GFEA) fabricated using a novel method.^22,23 The influence of gases as well as the effect of water vapor associated with the intended gas on the emission characteristics of field emitters was studied by introducing several inert gases and nitrogen gas within a vacuum chamber. The chamber setup and experimental process are explained including the results on emission performance.

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 3D schematic and Fig. 1(b) shows the SEM image of the emitter.²² The field emitter tips have 1 μm spacing (pitch). These arrays consist of self-aligned gated silicon emitters with integrated nanowires of 150 nm diameter and an 8 μm height. Detailed fabrication methods and descriptions can be found elsewhere.^22,23 The 1000 × 1000 arrays were chosen for these experiments because they are able to produce current density greater than 100 A/cm².²³ Current–voltage (I–V) characterization measurements were carried out inside a stainless-steel vacuum chamber using a Keysight B2902A SMU. The chamber is equipped with electrical feedthroughs, thermocouple feedthroughs, 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, 2.75” CF flange mounted) that is used to desorb water vapor from the tips.²⁴ A detailed description of the chamber and the experimental test setup can be found in our previous work.^24,25

I–V sweeps were carried out for different gases and gas pressures. Time taken for each test was ≈10 min. At first, the tip surfaces were cleaned using the UV light,²⁴ and then the backfill gas was injected slowly using a leak valve until the pressure inside the chamber reached the desired value. After achieving the desired pressure, gate voltage sweeps were carried out keeping the collector voltage fixed (200 V). Collector distance was ≈1 mm. Each voltage sweep took ≈2 min. Similar experimental steps were repeated for each gas pressures (5 × 10⁻⁶, 5 × 10⁻⁵, and 5 × 10⁻⁴ Torr) and backfill gas (N₂, Ar, and He). From the experiments (Fig. 2), it was observed that for a gate voltage of 40 V, the collector current decrease was the least for N₂ [Figs. 2(a) and 2(b)] for each partial pressure, where a drop of 41% was observed for the pressure of 5 × 10⁻⁴ Torr; He [Figs. 2(c) and 2(d)] has a decrease of ≈52%, and Ar [Figs. 2(e) and 2(f)] had the greatest decrease (65% drop for pressure of 5 × 10⁻⁴ Torr). Figures 2(a) (inset), 2(c) (inset), and 2(e) (inset) show the Fowler–Nordheim (F–N) plots for N₂, He, and Ar, respectively, to demonstrate the field emission nature of the device. After the exposure, vacuum was restored by pumping down and the I–V sweep was repeated. It was observed that after pump back down, the collector current was restored to preexposure conditions (≈98%), which suggests that most of tips were not physically damaged. Corresponding F–N plots for N₂ [Fig. 2(a) inset], He [Fig. 2(c) inset], and Ar [Fig. 2(d) inset] also became more linear after the pump down and UV exposure, which suggest that degradation was not permanent.

A degradation in emission current and increase in gate current can be observed after the N₂ exposure experiment and at the beginning of the He exposure experiment (at high vacuum condition, before the gas injection). This degradation is permanent and assumed to be the result of tip damage due to arcing. However, this phenomenon is not in the scope of this study. To keep the consistency, we used the same device throughout this study.

Further study was carried out using both industrial grade (I.G.) and high purity (H.P.) Ar (≈99.999%) to find out the effect of the Ar gas versus the water vapor on the collector current and can be seen in Figs. 3(g) and 3(h). At first, the surface was cleaned using UV light, and then the industrial grade Ar was injected into the chamber until the chamber total pressure reached ≈5 × 10⁻⁶ Torr. This pressure was kept for 30 min by adjusting the leak valve. The I–V sweeps were then carried out every 5 min. The gate voltage was swept to 40 V, and the collector voltage was fixed at 200 V. After 30 min, the collector current degraded ≈60%. For the case of high purity Ar also, similar steps were repeated. However, for this case, it took ≈120 min of gas exposure for the current to degrade ≈60%, and the gate sweep was carried out every 10 min. The experimental data are plotted in Fig. 6 in terms of the normalized collector current versus the gas exposure in Langmuir. One Langmuir corresponds to an exposure of 10⁻⁶ Torr during 1 s. The collector currents were normalized to the pregas injection current. Figure 6(a) shows the total volume of I.G. Ar and H.P. Ar injected during the experiment, where Fig. 6(b) shows the total volume of water vapor that was included in the injected Ar gases. The total volume of Ar is much higher than the total volume of water vapor present in the gas exposure effects. From Fig. 6, it can clearly be observed that the water vapor present in the gas is the primary cause of degradation as there is clear correlation between H₂O exposure and current decrease.

I–V characterization of Si-GFEA with 1000 × 1000 tip arrays was carried out after UV light exposure and then after exposure to gases (N₂, Ar, and He) at different vacuum levels. It was found that the water vapor present in these gases was the major cause of the work function increase. At first, it was observed that the collector current was least effected by N₂ and mostly affected by the Ar gas exposure. The degradation of collector current at 5 × 10⁻⁴ Torr pressure for Ar was ≈65% where for the N₂, at the same level of pressure, the degradation was ≈41%. It was assumed that the gases adsorbed into the tips changed the work function, which, in turn, affected the collector current. Upon pump down, the adsorbed gases were removed from the tips, which restored the pre-exposed collector current. Further experiments carried out with high purity Ar (≈99.999%) revealed that the water vapor present in the gas is the primary reason behind collector current degradation. Further experiments with the water vapor trap is required to find out the effect of gases itself.

Material support for this work was provided by the Air Force Office of Scientific Research under Grant No. FA9550-18-1-0436. The authors would also like to thank support from undergraduate students Brady Sainz, Gerardo Herrera, and John McClarin.

The authors have no conflicts to disclose.

Ranajoy Bhattacharya: Data curation (lead); Formal analysis (lead); Investigation (lead); Methodology (lead); Writing – original draft (lead). Mason Cannon: Investigation (supporting). Girish Rughoobur: Resources (equal). Nedeljko Karaulac: Resources (equal). Winston Chern: Resources (supporting). Reza Farsad Asadi: Data curation (equal); Validation (equal). Zheng Tao: Formal analysis (equal); Validation (equal). Bruce E. Gnade: Funding acquisition (equal); Supervision (equal); Validation (equal); Writing – review & editing (equal). Akintunde Ibitayo Akinwande: Funding acquisition (lead); Project administration (equal); Resources (equal); Supervision (equal). Jim Browning: Conceptualization (equal); Funding acquisition (equal); Supervision (equal); Validation (equal); Writing – review & editing (equal).

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Ranajoy Hattacharya received his Ph.D. degree from the Department of Electrical and Computer Engineering, Seoul National University, Seoul, South Korea in 2018. At present, he is working as a research scientist at Boise State University.

Girish Rughoobur received the M.Eng. degree in Electrical and Electronic Engineering from the University College London (UCL), UK, in 2012, and the Ph.D. degree in engineering from the University of Cambridge, UK, in 2017.

Nedeljko Karaulac received the B.S. degree in Electrical Engineering with the Georgia Institute of Technology (Georgia Tech), Atlanta, GA, USA, in 2015 and the M.S. degree in electrical engineering and computer science (EECS) from the Massachusetts Institute of Technology, Cambridge, MA, USA, in 2019. He is currently pursuing the Ph.D. degree in electrical engineering and computer science (EECS) from the Massachusetts Institute of Technology, Cambridge, MA, USA.

Winston Chern received the B.S. degree in Materials Science and Engineering from the University of Illinois at Urbana-Champaign, Champaign, IL, USA, in 2010, and the M.S. degree in Electrical Engineering and Computer Science (EECS) from the Massachusetts Institute of Technology, Cambridge, MA, USA. He received his Ph.D. degree in Electrical Engineering and Computer Science (EECS) from the Massachusetts Institute of Technology, Cambridge, MA, USA, in 2017.

Reza Farsad Asadi is currently pursuing his Ph.D. degree in Electrical and Computer Engineering from the Southern Methodist University, TX, USA.