Nanoscale resistive switching devices based on...

Nanoscale resistive switchingdevices based on metal oxides

Jacopo FrascaroliSupervisor Prof. Alberto PulliaCo-Supervisor Dr. Sabina Spiga

Jacopo Frascaroli Milano, October 15th 2013

Outline

Resistive switching:

phenomenology and applications

Device characterization

Nano devices using block copolymers

Pt

TiN

ALD - HfO2

5 - 6 nm

40 μm

40 μm


Resistive switching in MIM structures

AMetal

Metal

Insulator

Capacitor-like structure with electrically alterable resistivity

A large variety of insulating materials shows resistive switching properties:

Transition metal oxides (TMO) metal doped perovskites Polymers ...


Switching behavior

Applying a proper potential difference at the two electrodes, the device resistance changes

Reset

Set


Resistive switching mechanism in TMO

The anion motion leads to a valence change of the metal sublattice (cations) R change

Filamentary type

R. Waser, Adv. Mater. 2009 and ref. therein; A. Sawa, Materials Today 2008

Mobile species:Oxygen ions – oxygen vacancies

Initial state Forming

Reset Set


Very simple structure Highly scalable CMOS process compatibility Low switching time Low power consumption

Crossbar array

Resistive switching applications

Memory

Need for a lithographictechnology to enable cell scaling

Si nanocrystal

NAND-Flash (FG)

Techn.node F [nm]

Cell size[F2]

10

20

30

40

180 130 90 65 5x 3x

NOR-Flash (FG/NROM)

MRAM

FeRAM

PCM

ReRAM ?3D stack

2x4x 1x

niche and niche and

embedded embedded

applicationsapplications

codecode

datadata

evolutionaryevolutionary

disruptivedisruptive

3D-SONOS

Spintronics

NAND-Flash (FG)

Techn.node F [nm]

Cell size[F2]

10

20

30

40

180 130 90 65 5x 3x

NOR-Flash (FG/NROM)

MRAM

FeRAM

PCM

ReRAM ?3D stack

2x4x 1x

niche and niche and

embedded embedded

applicationsapplications

codecode

datadata

evolutionaryevolutionary

disruptivedisruptive

3D-SONOS

Spintronics

MRAM-spin torque


Resistive switching applicationsLogic Artificial neural

networks

Yang, J. Joshua, Dmitri B. Strukov, and Duncan R. Stewart. "Memristive devices for computing" Nature nanotechnology 8.1 (2012)


Experimental results

Up to 2000 reproducible switching cycles demonstrated

Pt

TiN

ALD - HfO2

5 - 6 nm

40 μm

40 μm

Microscopic device characterization

-2 -1 0 1 2

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

C

urr

ent

(mA

)

Voltage (V)

Reset

Set


Possible to tune the low resistance varying the maximum allowed current

during set (current compliance)

0.1 110

2

103

104

RLO

W (

)

Set max current (mA)

Experimental resultsElectrical characterization

-1.5 -1.0 -0.5 0.0

-5

-4

-3

-2

-1

0

Curr

ent

(mA

)

Voltage (V)


1.0 1.5 2.0 2.5 3.0104

105

RH

IGH

(

)

Reset voltage (V)

Possible to tune the high resistance varying the maximum reset voltage

0 1 2 3

0.0

0.2

0.4

0.6

0.8

1.0

Curr

ent

(mA

)

Voltage (V)

Experimental resultsElectrical characterization


• Two different retention behaviors

• Arrhenius dependence of the retention time on temperature

100

101

102

103

104

105

106

101

102

103

104

105

106

Low resistance stateR (

)

t (s)

TiN/HfO2/Pt 150°C

Failure bit

Typical bits

High resistance state

2.2 2.3 2.4 2.5 2.6 2.7 2.8

1000/T (K-1)

Ea

HR = 1.5 +/- 0.1 eV

180 160 140 120 100 80

T (°C)

100

101

102

103

104

105

106

(

s)

Experimental resultsRetention behavior

Retention behavior at 150°C


Slow retention loss• Modification of the residual

filament• The TiOxNy /HfO2 interface

may play a role

Fast retention loss• Very narrow gap.

Diffusion of a few vacancies can reconstruct the conductive channel

100

101

102

103

104

105

106

101

102

103

104

105

106

Low resistance stateR (

)

t (s)

TiN/HfO2/Pt 150°C

Failure bit

Typical bits

High resistance state

Experimental resultsRetention modeling

Pt

TiN

HfO

x

Oxigen speciesOxigen vacancies

Forming

-V

Pt

TiN

HfO

x

Oxigen speciesOxigen vacancies

Reset

+V

Set Reset

J. Frascaroli et al., in preparation


10-5

10-4

10-3

102

103

104

105

RHI

RLOW

R (

)

Area (cm2)

Filamentary conduction

High scalability potential

Experimental results

No area dependence of the resistance

Device area dependence


Previous works

ReRAM nanoscaling

Top-down

B. Govoreanu et al. IEEE Electron DevicesMeeting (2011)

Anodized-Aluminium Oxide templates

Bottom-up

D. Perego et al. Nanotechnology 24

(2013)

Electron beam lithography

• Slow and serial lithographic technique• Limited number of devices • Not compatible with actual

technology


Block Copolymers

Hexagonal cylinders template

Experimental result: block copolymers self assembling of hafnium dioxide


GraphoepitaxyM. Perego et al. Nanotechnology 24, 8 (2013)

Pre-patterned structures can be used to drive the self assembly of block copolymers Courtesy of F. Ferrarese Lupi, MDM IMM-CNR


Top electrode nano patterning

Electrical measurements with c-AFM

Combining directed self-assembling with electron beam lithography

Lift-off process


Substrate: ALD – HfO2

MetalPhysical

depositionThickness

Tungsten (W) sputtering 5 nm

Platinum (Pt)

sputtering 5 nm

ebeam 5 nm

Lift-off process study

Experimental result


200 nm

We demonstrated the possibility to obtain nano scaled top Pt electrodes over

large areas of HfO2

Experimental result


Current activity

Device characterization andstudy of the retention behavior

Implementing the block copolymer self-assembling on HfO2

Study of the lift-off process for top electrodes nano patterning


Future outlook

Electrical characterization of the nano devices using C-AFM to inspect the physical properties at the nano scale

Find alternative strategies for the fabrication of resistive switching devices using block copolymers self-assembly


Initial state

Fresh cell

Resistive switching mechanismH. Akinaga and H. Shima, Proceedings of the IEEE 2010


Soft breakdown

Initial state

GND

SET (forming)

On state(Low resistance)

Fresh cell Forming

V+

Resistive switching mechanism


V-

GND

Off state(High resistance)

RESET

Fresh cell Forming Reset




GND


SET (forming)


Fresh cell Forming Reset Set

V+



V-

GND

GND


SET (forming)


RESET

Fresh cell Forming Reset Set Many cycles

V+



X-ray photoemission spectroscopy• TiO2/TiOxNy in serie with the HfO2 film• Significant percentage of non-lattice oxide

542 540 538 536 534 532 530 528 526 524

1000

2000

3000

4000

5000

6000

7000

8000

9000

10000

Data

sum

peak1 (O1s)

peak2

background

Co

un

ts

BE (eV)

O1s

Non-lattice oxide

475 470 465 460 455 4500

2000

4000

6000

8000

10000

12000

14000

Data

Sum

Peak1

Peak2

Peak3

Background

Counts

BE (eV)

Ti2p

TiO2

TiOxNy

HfO2 filmTiN film

Experimental resultsMetal/Oxide interface


ITRS 2012 UPDATE - Winter Presentations


Conduction in metal oxides

J. Joshua Yang et al., "Memristive devices for computing" Nature nanotechnology 8.1 (2012)

Nanoscale resistive switching devices based on...

Documents

Transcript of Nanoscale resistive switching devices based on...