PHYSICAL PROPERTIES OF ULTRA THIN GOLD FILMS DEPOSITED  ON AMORPHOUS GERMANIUM SUBSTRATE
Tanina Bradley, Christopher Jessamey and Abebe Kebede,   Brookhaven National Laboratory,  June 6-August 14, 2004
Picture diary


Content
download poster

ABSTRACT

NTRODUCTION

PROCEDURE

EXPERIMENT

  • Evaporation 
  • Surface Morphology
  • Electrical Resistance


A FLOATING ANIMAL

ABSTRACT

In this work we present the results of nanostructural studies and electrical characterization of Au/Ge films. Ultra thin films of gold were grown on amorphous germanium substrates. These films were fabricated under very high vacuum condition using thermal evaporation technique. The nanostructure of the films were determined using the atomic force microscopy method. The thickness of the films range from 2A to 100 A and [the average cluster size ranges from 5nm to 100nm]. It is found that the size of the gold clusters depend on substrate temperature.   In this communication we present the Atomic Force Microscopy data, and the results of the electrical resistance and its dependence on the temperature. Analysis of the data shows that these nano sized films are insulators

INTRODUCTION

In the rich literature of the physics of disordered metals, the structural and electrical properties of ultra thin  (near mono layer) films of metals are shown to exhibit interesting behavior because of their two dimensional (2D) nature. Such materials exhibit high  resistance greater than h/4e2 (~30kW/square). They  display phenomenon related to disorder, such as strong localization effects, Universal conductance fluctuation and suppression of superconductivity. It is also shown that Quantum Size Effects (QSE)" arise when the thickness of the film is comparable to the length scales of the system such as the electron mean free path and the de Broglie wavelength. The recurrent experimental and technical inquiries in the literature about thin films is centered at electrical properties of such materials when the thickness  is comparable to the important quantum mechanical length scales of the system . For  application in molecular electronics, the technical problems include fine tuning material processing of thin films of metals for molecular electronics. One of the experimental problems in this case is the ability to obtain very thin films, free of percolation effects and granularity. In this case such film can be used as electrodes to the molecule under investigation. The man goals this  project include production such film,  investigation of bonding of thiols and biologically active molecules to gold and measurement of electron transport through the adsorbed molecules, using in-situ apparatus. In this communication we describe the procedure we followed to prepare the films as well as some of our results from the measurement of electrical conductivity.

PROCEDURE

SUBSTRATE PREPARATION

Fisher Scientific Premium microscope slides, 1mm thick, are cut with a diamond knife into 1cm x 1cm squares  were cleaned with ethyl alcohol, and masked with aluminum foil, as shown above. Four symmetrically placed silver pads were sputtered using Kurt Lester sputtering system. The Kurt Lesser Sputtering system consists of  a high RF voltage, a pumping system, water cooled electrodes and high purity argon gas.

Four copper wires (40 gauge ) are attached to the silver pads,  using Ted Pella Inc. fast drying silver point (Lot #0378), for in situ measurement of the sheet resistance .
 

The substrate was mounted on a low temperature insert, where the four copper wires are soldered to the sample measurement platform that run to the measuring electronics via a vacuum feed through. 
For most of our measurements the pressure in the chamber was maintained at 2-5 x10-8 torr using nitrogen cooled diffusion pump. A Dycor  mass spectrometer was used to identify molecules that may be outgassed from the sample-platform system as well as the from the chamber itself. In most cases, within the resolution of the spectrometer,  the vacuum was clean from such molecules as water vapor, nitrogen, carbon dioxide and other molecules.

EXPERIMENT
 

Evaporation of Germanium and Gold Films

The samples were prepare using thermal evaporation technique under ultar high vacuum condition ( 2 -5x 10-8  torr) .  The temperature of the substrate is monitored using CY7 series silicon diode sensor. . Two sources were used for evaporating Ge and Au in succession, without breaking the vacuum. A quartz crystal thickness monitor was used to determine the thickness  of  Ge and Au in conjunction with the DC circuit that is used for in situ measurement of the DC sheet resistance.  The electronics is prepared to measure the sheet resistance  using a high impedance Keithley 614 (K614) electrometer. The current through the sample is monitored using Keithley 485 (K485) Pico ammeter, A 6.0 Volt battery was used as a source. The temperature was monitored using a diode temperature sensor using a LakeShore DRC93A temperature controller.

The evaporation rate was tuned to 5A/s for Ge and 0.1 to 0.5 A/s for Au.  The thickness monitor and sample assembly is shown in picture below

After evaporation of Ge, the Inficon apparatus was always reset to measure the thickness of Au.   During the evaporation of Au the current flow through the DC circuit  is monitored. That way we established the formation of monolayers of Au on the on top of the Ge. Once such condition is established the evaporation of Au is halted. 

Surface Morphology
 
The nano-structure of our samples was identified by the Atomic Force Microscope (AFM) in contact mode. In this mode a sharp probe on a cantilever with a spring constant of less than 1 N/m is brought in direct contact with the surface and the repulsive force between the tip and the surface is measured. 
The thickness of the samples mostly correspond to the thickness obtained during the  film growth. The samples contain features of several sizes ranging from 10 nm-1000 nm.  More measurements are needed  to identify the chemical composition of these features

Electrical Resistance

We measured the electrical Resistance(in this case the sheet resistance) of ultra thin gold films deposited on amorphous germanium, as a function of temperature and thickness. Several samples have been measured.  From the room temperature resistance value we estimated the electrical resistivity to be 10-3 W-cm. Such high resistivity led us to speculate that these materials are "semiconducting". For typical semiconductors  such as Si and Ge r(T) =Cexp(-Eg/2kT). Where C is the resistivity when the temperature is very large,  Eg is the energy gap, that is the minimum energy needed to excite electrons from the valence band to the conduction band, and k is the Boltzmann constant. 

In almost all samples we studied the materials showed semiconducting behavior.  We also made a qualitative estimate of the "energy gap" associated with the activation behavior by plotting ln R as a function of 1/T shown in Figures . This gap depends on the thickness of the sample, and ranges from 10 mev for high thickness to 100 mev for low thickness.  This suggests that the system would undergo an insulator to metal transition as the thickness of the gold is increased. 

This analysis provided us with a qualitative measure of the semiconducting behavior.. However, we believe that the conduction mechanism in our sample can not be described by assuming traditional doped semiconductors because the resistivity is too high.