Tunable quantum dots from atomically precise graphene nanoribbons using a multi-gate architecture

2025-05-06 0 0 9.1MB 25 页 10玖币

侵权投诉

Tunable quantum dots from atomically precise

graphene nanoribbons using a multi-gate

architecture

Jian Zhang,∗,†,‡‡ Oliver Braun,†,‡,‡‡ Gabriela Borin Barin,¶Sara Sangtarash,§Jan Overbeck,†,k

Rimah Darawish,¶,⊥Michael Stiefel,†Roman Furrer,†Antonis Olziersky,#Klaus M¨ullen,@Ivan

Shorubalko,†Abdalghani H.S. Daaoub,§Pascal Ruﬃeux,¶Roman Fasel,¶,4Hatef Sadeghi,∗,§

Mickael L. Perrin,∗,†,∇and Michel Calame∗,†,k,††

†Transport at Nanoscale Interfaces Laboratory, Empa, Swiss Federal Laboratories for Materials Science and

Technology, 8600 D¨ubendorf, Switzerland

‡Department of Physics, University of Basel, 4056 Basel, Switzerland

¶nanotech@surfaces Laboratory, Empa, Swiss Federal Laboratories for Materials Science and Technology,

8600 D¨ubendorf, Switzerland

§School of Engineering, University of Warwick, Coventry CV4 7AL, United Kingdom

kDepartment of Physics, University of Basel, 4056 Basel, Switzerland

⊥Department of Chemistry and Biochemistry, University of Bern, 3012 Bern, Switzerland

#IBM Research - Zurich, 8803 R¨uschlikon, Switzerland

@Max Planck Institute for Polymer Research, 55128 Mainz, Germany

4Department of Chemistry, Biochemistry and Pharmaceutical Sciences, University of Bern, 3012 Bern,

Switzerland

∇Department of Information Technology and Electrical Engineering, ETH Zurich, 8092 Zurich, Switzerland

††Swiss Nanoscience Institute, University of Basel, 4056 Basel, Switzerland

‡‡These authors contributed equally to this work.

E-mail: jian.zhang@empa.ch; Hatef.Sadeghi@warwick.ac.uk; mickael.perrin@empa.ch;

michel.calame@empa.ch

arXiv:2210.03366v2 [cond-mat.mes-hall] 27 Oct 2022

Abstract

Atomically precise graphene nanoribbons (GNRs) are increasingly attracting in-

terest due to their largely modiﬁable electronic properties, which can be tailored

by controlling their width and edge structure during chemical synthesis. In re-

cent years, the exploitation of GNR properties for electronic devices has focused

on GNR integration into ﬁeld-eﬀect-transistor (FET) geometries. However, such

FET devices have limited electrostatic tunability due to the presence of a sin-

gle gate. Here, we report on the device integration of 9-atom wide armchair

graphene nanoribbons (9-AGNRs) into a multi-gate FET geometry, consisting of

an ultra-narrow ﬁnger gate and two side gates. We use high-resolution electron-

beam lithography (EBL) for deﬁning ﬁnger gates as narrow as 12 nm and com-

bine them with graphene electrodes for contacting the GNRs. Low-temperature

transport spectroscopy measurements reveal quantum dot (QD) behavior with

rich Coulomb diamond patterns, suggesting that the GNRs form QDs that are

connected both in series and in parallel. Moreover, we show that the additional

gates enable diﬀerential tuning of the QDs in the nanojunction, providing the

ﬁrst step towards multi-gate control of GNR-based multi-dot systems.

Bottom-up synthesized GNRs have attracted considerable interest as possible future elec-

tronic building blocks. This is mainly due to the fact their chemical structure can be con-

trolled with atomic precision, a property that top-down etched GNRs lack.1–3 Bottom-up

synthesized GNRs can, therefore, be regarded as a designer quantum material, where the

material properties can be designed by selecting the appropriate chemical precursors and

synthetic routes4–17. As such, one can largely tune their bandgap,18,19 form pn-junctions

within a single, heterogeneous ribbon,6tailor spin-polarized states7,20 and even topologi-

cally non-trivial phases.8–10,16,21 Exploiting these properties in electronic devices requires

contacting strategies that preserve the integrity of the GNRs, while at the same time allow-

ing for charge carriers to ﬂow through. Moreover, many technological applications require

electrostatic control over the level structure of the GNRs. For example, ﬁeld-eﬀect transis-

tors require the presence of a single gate electrode to tune the channel to conductance, while

multiple gate electrodes are needed for the realization of qubits.

Several prototypical GNR devices have been studied to date10,22–28, exhibiting various

charge-transport characteristics, such as high-performance ﬁeld-eﬀect transistors operating

at room temperature22, gate-tunable QDs at low temperature10,25 and temperature-activated

transport through micron-sized ﬁlms26,28. However, many challenges remain in the device

integration of these materials. On the one hand, the contacts need to be improved further29,

as well as the transfer process from the growth substrate to the devices substrate which can

lead to defects, impurities, and adsorbates at the interface between GNRs and the electrode

material. On the other hand, advanced gating strategies, such as ultrashort transistors30,31 or

multi-gate architectures, are highly desirable for devices that require additional control over

the electrostatic landscape of the device. To date, due the nanoscale size of the GNRs, only

ﬁeld-eﬀect-transistor devices have been realized10,22–28. More advanced device architectures

with multiple gates that are individually addressed require a very high control over the

fabrication of the multiple gates, the electrodes, and the alignment between them.

Here, we report on the integration of GNRs into a multi-gate ﬁeld-eﬀect transistor with

graphene electrodes. Our device design consists of a narrow ﬁnger gate and two additional

side gates. This geometry improves gating capabilities by allowing for the generation of an

asymmetric gate ﬁeld using the side gates. As such, the diﬀerent sides of the nanogaps experi-

ence a diﬀerent gate ﬁeld, providing additional control over the electrostatic landscape of the

junction. The narrow gate is ∼10 nm in length, with an eﬀective channel length of <15 nm,

and is fabricated using CMOS-compatible processing steps. The graphene electrodes are

created using EBL, which has a major advantage of the control of the nanogap position27

and a proper alignment with the underlying gates. This is in contrast to electrodes created

using the electric breakdown procedure that has been commonly used for graphene.10,25,32

Moreover, our fabrication protocol allows for the integration of the GNRs at the very last

stage of device fabrication. Similar approaches with the integration of the sensitive material

in the ﬁnal step have been shown to lead to major improvements in the device performance,

as demonstrated for example for MoS2.33–35 The design of the devices is supported by ﬁnite-

element calculations for optimizing the various geometrical parameters and maximizing the

eﬀective electrostatic potential at the GNRs. Furthermore, low-temperature transport spec-

troscopy measurements reveal quantum dot (QD) behavior with addition energies in the

range of 20-150 meV, and transport characteristics that are tunable using the two side gates.

Our observations are supported by simple model calculations that highlight the importance

of the asymmetric gate ﬁeld in the junction area.

Results

Devices design and fabrication

A schematic of the proposed device architecture is shown in Figure 1a. The ﬁnger-gate

(FG) with nanometer-scale dimensions is ﬁne-patterned under the 9-AGNRs junction, while

two side-gates (SG1 and SG2) are deﬁned under the source and drain graphene electrodes,

respectively. As the electronic coupling between the GNRs and the graphene is weak, we

anticipate the formation of quantum dots (QD) at low temperatures.10,25 The side gates

(SG1 or SG2), as they are located close to the nanojunction, are employed to introduce

an asymmetric electric gate ﬁeld that will couple to the QD. We note that the interplay

between the gate length, gate separation, gate oxide thickness and the applied potentials is

very delicate to optimize to achieve a homogeneous electrostatic potential over the complete

channel length. Thinner oxides lead to higher gate coupling but also lower breakdown

voltages between the various gates. Moreover, reducing the distance between the gates also

increases the screening of the gate potential by the neighboring gates. To investigate this

balance, we performed ﬁnite-element calculations using Comsol Multiphysics. In Section 1

of the Supporting Information, we present the eﬀective potential at the GNRs for various

thicknesses of the Al2O3and the gate separation. Graphene is modeled as a surface charge

density; its value is calculated using the voltage applied on the gate located below the

respective electrode, and the sum of quantum capacitance and geometric capacitance.

We show that thinner oxides down to 12 nm are generally more beneﬁcial. A further

result is that reducing the FG-SG separation is beneﬁcial, but only down to 10 nm. Beyond

that point, the ﬁeld exceeds 1 V/nm, a strength where a breakdown of the oxide is likely to

happen.36

The starting point of the sample fabrication is a highly-doped silicon (Si) carrier chip with

a 285 nm thick silicon dioxide (SiO2), such that the Si substrate acts as a global back gate

(BG). The ﬁnger gates are patterned on top as follows. First an 8 nm platinum (Pt) ﬁlm is

deposited using electron-beam evaporation. Then, a negative resist hydrogen silesquioxane

(HSQ) is spin-coated as a resist to deﬁne the etch mask. This resist turns into SiO2-like

after EBL exposure and development, leading to a highly-resistive etch mask. A subsequent

Ar+-ion milling step transfers the etch mask feature to the metal ﬁlm, separating the ﬁnger

from the side gates. This process leads to very sharp features as it is not limited by grain

sizes or other edges eﬀects which are common when using electron-beam evaporation with a

lift-oﬀ process. Our approach results in a ﬁnger gate with a length of 10-15 nm and a width

of 500 nm. The nanometer-scale dimension of the FG (< d) allows for creating an ultra-short

eﬀective channel length while minimizing parasitic gate to source-drain capacitance. The 9-

AGNRs junction is electrically isolated from the metal gates using a 30 nm thick aluminum

oxide (Al2O3). The graphene electrodes are separated by a nanogap formed with high-

resolution patterning by using EBL, as reported elsewhere.27 Here, the electrode separation

dis set to be ∼15 nm, large enough to eliminate direct tunneling contributions between

the electrodes, but smaller than the average length of the 9-AGNRs.37 In Section 2 of the

Supporting Information, a more detailed description of the fabrication process is given.A

scanning electron micrograph (SEM) of the ﬁnal device before GNR transfer is presented in

文档加载中……请稍候！
如果长时间未打开，您也可以点击刷新试试。

下载文档到电脑，查找使用更方便

10 玖币 0人已下载

立即下载

摘要：

Tunablequantumdotsfromatomicallyprecisegraphenenanoribbonsusingamulti-gatearchitectureJianZhang,,y,zzOliverBraun,y,z,zzGabrielaBorinBarin,{SaraSangtarash,xJanOverbeck,y,kRimahDarawish,{,?MichaelStiefel,yRomanFurrer,yAntonisOlziersky,#KlausMullen,@IvanShorubalko,yAbdalghaniH.S.Daaoub,xPascalRueux,...

展开>> 收起<<

Tunable quantum dots from atomically precise graphene nanoribbons using a multi-gate architecture.pdf

共25页,预览5页

还剩页未读，继续阅读

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

Tunable quantum dots from atomically precise graphene nanoribbons using a multi-gate architecture

相关推荐

开通VIP享超值会员特权

作者详情

相关内容

热门标签

举报选择: