WO2003033993A1 - A kelvin probe instrument - Google Patents

Abstract

The invention gives a method of determining the contact potential of a sample, the volta potential and/or the topography of the sample. The sample is placed on a translatable stage (22) in a chamber (12). The Kelvin probe (16) is brought into close proximity to the sample and an oscillating Kelvin probe signal is obtained according to the well-known principles of Kelvin probes. The signal is processed to obtain the first harmonic (114) of the signal. Using this harmonic signal, the contact potential, topography and/or the volta potential is determined. Use of the first harmonic provides an improved signal to noise ratio.

Description

A KELVIN PROBE INSTRUMENT

FIELD OF THE INVENTION

The present invention relates to an improved Kelvin probe instrument, which is of particular but by no means exclusive, application as a scanning Kelvin probe in the areas of corrosion measurement, solid state physics and electrochemistry.

BACKGROUND OF THE INVENTION

Throughout this document, the term Kelvin probe electrode is used to refer to a needle like metal electrode, typically with a flat end, which is vibrated in close proximity with and normal to a conducting surface, while the term Kelvin probe instrument is used to refer to an instrument system which utilizes a Kelvin probe electrode in conjunction with a signal detection system and electrical feedback circuit and positioning system and optics for viewing the sample.

The Kelvin probe instrument (first described - together with the theory behind its operation - by Lord Kelvin in 1897) essentially comprises a vibrating capacitor, with air as the dielectric between two dissimilar, electrically connected metal plates that are in very close proximity (of the order of 100 μm) . The vibration of one of the plates results in an alternating current. A variable DC voltage source is located in the circuit joining the plates, and varied until the electric field between the plates - and consequently the alternating current - is nulled. The DC voltage signal is then proportional to the contact potential difference (CPD) , the work function difference or the volta potential difference between the two metals. The Kelvin probe instrument can therefore be used to determine variations in a surface by monitoring changes in the CPD, work function or volta potential as a Kelvin probe electrode (of fixed properties) is moved over a sample surface.

Although one of the first Kelvin probe instruments was reported in 1932, it was not until powerful personal computers were being utilized as control units for the instrumentation that the commercial development of scanning probe type of work function measurements become viable on a routine basis. Since the 1960s, work has also been done to reduce the noise of existing Kelvin probe instruments in order to increase their sensitivity. One existing Kelvin probe instrument has been designed to facilitate the measurement of topography simultaneously with volta potential (see The Scanning Kelvin Microscope, Mackel et al., Review of Scientific Instruments 64 (3) (1993) 694-699), while Zhang et al . report a temperature controlled Kelvin microprobe in Sensors and Actuators B 12 (1993) 175-180. Further work has also been reported in Low cost PC Based scanning Kelvin probe (Baikie and Estrup, Review of Scientific Instruments 69(11) (1998) 3902-3907), Noise and the Kelvin method (Baikie et al . ,

Review of Scientific Instruments 62(5) (1991) 1326-1332), Ritty et al . , J. Phys . E: Sci. Instrum. , 15 (1982) 310- 317) and Fujihira and Kawate, Thin Solid Films, 242 (1994) 163-169.

Despite these contributions, however, in many applications greater sensitivity and functionality in a Kelvin probe instrument would still be valuable, especially if this could be provided without undue additional complexity.

In the following, the term "contact potential difference' has been used in preference to "workfunction difference" , though the two are generally regarded as essentially equivalent. Consequently, it should be understood that any reference to "contact potential difference" could be replaced by "workfunction difference", and that the invention should be understood (in those aspects that relate to the determination of contact potential difference) to relate equivalently to the determination of workfunction difference.

SUMMARY OF THE INVENTION

The present invention provides, therefore, a method of determining the contact potential difference between two materials, comprising: arranging said materials in a Kelvin probe instrument; measuring a harmonic of the Kelvin probe signal; and determining said potential difference from said harmonic .

The two materials will generally be a sample and a Kelvin probe electrode.

Preferably the Kelvin probe signal comprises the in-phase current .

Phase-lock devices (such as lockin amplifiers) operate by synchronising the phase of a reference signal and of an AC source waveform so that full rectification is achieved and the DC level is linearly proportional to the r.m.s. value of the source. The in-phase current value is the result of this procedure.

Preferably the harmonic is the first harmonic ,

Thus, as the potential difference between, for example, a sample and the Kelvin probe electrode is proportional to the current, this potential difference can also be determined from a harmonic of that current (being an alternating current signal) . It has been surprisingly found that the signal to noise ratio is greater when measuring the first harmonic of the in-phase current then when measuring the fundamental frequency of that current, and - further - that noise rejection is far better when measuring the first harmonic of the in-phase current then when measuring the fundamental frequency. It has also been found that volta potential measurements can be made more accurately and faster using the first harmonic, and that the initial positioning of the tip of the Kelvin probe can be made accurately using the ratio of the fundamental and first harmonic of the in-phase current.

Preferably said method includes defining the resolution of data from said instrument by defining voltage resolution.

Measuring the harmonic may comprise measuring the entire Kelvin probe signal, and processing said signal to obtain said harmonic.

Preferably said Kelvin probe instrument includes a piezoelectric actuator.

It has been found that signal processing of the first harmonic allows an improved use of piezo-electric actuators.

Preferably said method includes determining the relative topography of two surfaces simultaneously with said determining of said contact potential difference.

Typically one of the surfaces is a Kelvin probe electrode, the other a sample, so the relative topography is a measure of the actual topography of said sample.

Maps can simultaneously be made of the contact potential difference and the topography of surfaces that are conducting, semi-conducting or insulating using the ratio of the fundamental and first harmonic so long as they are not high in capacitance. In one embodiment, the method includes measuring the Volta Potential of a sample surface while tracking the topography of said sample.

This mode is, in some embodiments, referred to as "flyover Volta Potential scanner" mode.

The present invention also provides a method of determining the topography of a sample, comprising: arranging said sample in a Kelvin probe instrument; measuring a harmonic of the Kelvin probe signal; and determining said topography from said harmonic.

Preferably said method includes obtaining a null condition by altering sample potential incrementally until probe current is within a predefined tolerance of zero.

The present invention also provides a method of analyzing a Kelvin probe output signal, comprising: deriving a first harmonic of said signal from said output signal; and determining one or more quantities from said first harmonic selected from the set comprising contact potential difference and topography.

The present invention further provides a Kelvin probe instrument, having: signal processing means for extracting a first harmonic signal from a Kelvin probe output signal; and data processing means for determining from said first harmonic signal one or more quantities selected from the set comprising contact potential difference and topography. BRIEF DESCRIPTION OF THE DRAWINGS In order that the invention may be more clearly ascertained, preferred embodiments will now be described, by way of example, with reference to the accompanying drawings, in which:

Figure 1 is a schematic diagram of a scanning Kelvin probe instrument set-up for small scale x-y scanning according to one embodiment of the present invention; Figure 2 is a schematic diagram of a scanning

Kelvin probe instrument set-up for large scale x-y scanning according to another embodiment of the present invention;

Figure 3A is a top cross-sectional view of the actuator of the Kelvin probe instrument of figure 1;

Figure 3B is a side cross-sectional view of the actuator of the Kelvin probe instrument of figure 1;

Figures 4A to 4D are schematic representations of the various steps in the preparation of the Kelvin probe electrode of the Kelvin probe instrument of figure 1;

Figure 5 is a photograph of the platinum Kelvin probe electrode of the Kelvin probe instrument of figure 1;

Figure 6 is a simplified electronics diagram for the Kelvin probe instrument of figure 1;

Figure 7 is a schematic diagram of the feedback control loop of the Kelvin probe instrument of figure 1, for obtaining the volta potential of a sample;

Figure 8 is a general structural representation of the software interface of the Kelvin probe instrument of figure 1;

Figure 9 is a screen capture from an oscilloscope comparing the actuator driver (top) and the Kelvin probe (bottom) waveforms from the Kelvin probe instrument of figure 1;

Figure 10 is a plot of the in-phase current of the first harmonic (circles) and fundamental frequency (squares) as a function of relative position from the sample surface, for the Kelvin probe instrument of figure

1;

Figure 11 is a plot of fundamental frequency in- phase current versus applied voltage for various values of do, the gradients of which being the fundamental frequency sensitivity in each case, for the Kelvin probe instrument of figure 1;

Figure 12 is a plot of first harmonic in-phase current versus applied voltage for various values of d₀, the gradients of which being the first harmonic sensitivity in each case, for the Kelvin probe instrument of figure 1;

Figure 13 is a plot of the ratio of first harmonic current amplitude to fundamental frequency current amplitude (i.e. iχ/i₀) as a function of m = dι/d₀, obtained from measurement (squares) and from simulation (circles) , for the Kelvin probe instrument of figure 1; Figure 14 is a plot of the ratio ii/in as a function of applied voltage to the sample surface for the Kelvin probe instrument of figure 1;

Figure 15 is a plot of data from simultaneous CPD and topography measurements as x axis tilt is varied, performed with the Kelvin probe instrument of figure 1, in which CPD measurements are plotted as squares, topography measurements as circles;

Figure 16 is a flow chart of the logic of the volta potential program of the Kelvin probe instrument of figure 1; Figure 17 is a cross-sectional view of a metal (M

= Cu, Fe, Zn) cell filled with metal sulfate (MS0₄(_ag)) solution;

Figure 18 is a calibration plot of a platinum

+2

Kelvin probe electrode using M/M (_aq) couples; Figure 19A is a volta potential map obtained using the ratio of the fundamental and first harmonic by means of the Kelvin probe instrument of figure 1; Figure 19B is a topography map obtained simultaneously with the volta potential map of figure 19A using the ratio of the fundamental and first harmonic by means of the Kelvin probe instrument of figure 1; Figure 20 is a 2000 x 2000 μm potential map of a

1 mm copper wire embedded in a zinc substrate obtained with the Kelvin probe instrument of figure 1 fitted with a platinum electrode;

Figure 21 is a 1200 x 1200 μm potential map of a 0.5 mm aluminium wire in a zinc substrate obtained with the Kelvin probe instrument of figure 1 fitted with a platinum electrode;

Figure 22 is a potential distribution map obtained using the first harmonic of gold on aluminium and the Kelvin probe instrument of figure 1;

Figure 23A is a potential distribution map of a filiform track on epoxy coated Al 2024 after 16 hrs at 90% relative humidity obtained with the Kelvin probe instrument of figure 1; Figure 23B is a potential distribution map of the filiform track on epoxy coated Al 2024 referred to in figure 23A after a further 6 hrs at 60% relative humidity, obtained with the Kelvin probe instrument of figure 1;

Figure 24 is three dimensional potential distribution plot over Al 2024 treated with 1 M NaOH for 15 seconds on the (left portion) , for 30 seconds (right portion) and untreated (middle portion) , obtained with the Kelvin probe instrument of figure 1;

Figure 25 is a three dimensional potential distribution plot of corrosion initiation by NaCI micro droplet deposition on zinc metal, acquired at room temperature and 95% relative humidity with the Kelvin probe instrument of figure 1;

Figure 26 is a three dimensional potential distribution plot of a pit and surrounding area on a treated panel of Al 2024 exposed to neutral salt spray for 15 days, acquired at room temperature and humidity using a scanning Kelvin probe instrument according to a preferred embodiment, based on the probe of figure 1;

Figure 27A is a plot of the contact potential of a sample of zinc metal coated with an enamel paint coating obtained using the preferred embodiment scanning Kelvin probe instrument (referred to in figure 26) ;

Figure 27B is an optical image of the sample of figure 27A;

Figure 27C is a topography image of the sample of figure 27A obtained using the preferred embodiment scanning Kelvin probe;

Figure 28 is a plot of the potential transient of a pit in aluminium alloy 2024 as a result of electrochemical treatment, measured by means of the preferred embodiment scanning Kelvin probe instrument;

Figure 29 is a plot of the potential transient from the reaction of magnesium metal substrate with 7,7, 8,8-tetracyanoquinodimethane (TCNQ) at low and high humidity; Figure 30A is a plot of potential transients from the tribocharging of a polymeric film, to illustrate the change in electrostatic charge as a function of time at different positions on the surface of the film relative to the location of tribocharge induction; Figure 30B is a plot of potential transients from the tribocharging of another polymeric film, to illustrate the change in electrostatic charge as a function of time at different positions on the surface of the film relative to the location of tribocharge induction; Figure 31 is an image of the surface potential of a zinc oxide film formed on ITO coated glass in the dark (upper regions in this figure) and under illumination (lower regions) ;

Figure 32 is a plot of potential transient of a zinc oxide film formed on ITO under illumination, hatched regions indicating when the surface is in the dark;

Figure 33A is a contact potential map of a sample comprising gold patterned on aluminium obtained by means of the preferred embodiment scanning Kelvin probe instrument;

Figure 33B is a topography map of the sample of figure 33A obtained by means of the preferred embodiment scanning Kelvin probe instrument; and

Figure 33C is the Volta potential map of the sample of figure 33A obtained by means of the preferred embodiment scanning Kelvin probe instrument in "fly over" mode .

DETAILED DESCRIPTION OF THE INVENTION The principal components of a scanning Kelvin probe instrument for small scale x-y scanning according to one embodiment of the present invention is illustrated generally at 10 in figure 1. The instrument 10 includes an environment chamber 12, a piezo-electric actuator 14, a Kelvin probe electrode 16, x-y translation stages 18, a z- translation stage 20 and a sample stage 22. The chamber 12 includes a chamber door 24, chamber supports 26, electronic connector housing 28, optics ports 30,32 and gas sensor ports 34.

The chamber 12 is an environmental chamber where experiments and Kelvin probe investigations may be performed. The chamber 12 is constructed from a continuous cylinder and circular end-plates of 316 stainless steel. Optionally another highly corrosion resistant metal could be employed. Both top and bottom are pressure sealed with viton o-rings and M4 bolts. The chamber 12 is constructed so that it may be operated at room temperature, though it may optionally be constructed for operation at temperatures in the range 5-40° C by use of a temperature controlled circulation unit. In such a case the chamber enclosure has a cavity for the temperature controlled fluid to be circulated. The chamber 12 may be sealed to prevent the escape of gas or obtain a vacuum not less than 0.1 kPa. It has a large door 24 to allow manual manipulation of the sample stage and sample. The optics ports 30,32 are constructed for long working distance magnification optics, and for irradiation and detection of focused infrared or coherent light. There are also interfaces (not shown) for controlling the translation stages 18,20 and actuator 14. The access door 24 is provided with a 3 mm thick glass window 36. The surface of the window 36 is preferably coated with a transparent conducting film, such as gold or indium tin oxide, to prevent static charging while allowing high visibility.

Referring to figure 2, another embodiment of the invention provides a Kelvin probe instrument for large scale x-y scanning, shown generally at 40. This instrument 40 is identical in most respects with Kelvin probe instrument 10, but has y-translation stages 52 located above chamber 54. The x-translation stage 56 remains below chamber 52, and z-translation stage 58 above chamber 52.

Returning to the embodiment of figure 1, the actuator 14 is a piezo-ceramic vibrator device. It is designed to allow for translation of an electrode post in the z direction only with amplitude 30 to 100 μm at resonance. The actuator housing is constructed of copper, aluminium or stainless steel, and is connected to the lower translation post 38 coupled with an upper-chamber interface (not shown) . Electrical connectors are shielded and the actuator housing is attached to ground.

Figures 3A and 3B are respectively top and side cross- sectional views of the actuator 14. The Kelvin probe actuator 14 is constructed with two piezo sheets 60a, 60b, arranged to minimize the non-reticular motion in directions other that the z plane. The piezo sheets 60a, 60b are mounted on fixed, electrical insulating mounts 62a, 62b at either end of the actuator housing 64. The piezo sheets 60a, 60b are joined by a stainless steel foil membrane 66, mounted on an electrode post 68. The membrane 66 thus mechanically couples the piezo sheets 60a, 60b to the electrode post 68. The electrode post 68 is made from a lightweight, rigid acrylic. A gold electrode mount 70 is located at the lower or Kelvin probe electrode end of the post 68, via which a connection is made and the Kelvin probe electrode attached.

Referring to figures 4A to 4D, the Kelvin probe electrode 16 was prepared from platinum wire, having a starting diameter of 0.5 mm and a gold coated connector. A platinum wire 72 was cut to not less than 10 mm length and soldered into a gold coated connector 74. The Kelvin probe shape and tip diameter was made by mechanically polishing the wire 72 with the aid of a hobby drill, and

1200 and 4000 grit SiC paper. With the tip of the wire at a fine point 76 (less than 20 μm in diameter) , the wire 72 and connector 74 was embedded in resin 78. Subsequently, the electrode tip 76 was polished flat by grinding and polishing the adjacent face 80 of the resin 78. The desired diameter of electrode tip 82 was controlled by the extent of polishing and was evaluated by means of a microscope. The electrode 16 was then recovered from the resin by using tetrahydrofuran (THF) .

Finally, the wire part of the electrode 72 was electrochemically etched (240V, 50Hz) in saturated CaS0₄ (pH adjusted to 2 by H₂S0₄.) Before use, the electrode 16 was rinsed in distilled water and dried with an inert gas. Figure 5 is a photograph of a typical Kelvin probe electrode produced by this method. The sample stage 22 is composed of a two-dimensional tilt and rotation stage, a sample stage base and a sample platform. The two-dimensional tilt-rotation stage is made of anodized aluminium, the sample stage base and the sample platform of stainless steel. The assembly is coupled by a Teflon dovetail coupling which is mechanically fastened to the sample platform. The sample stage base may be coupled to the x-y translation system or the bottom-plate of the chamber. Electrical connections to control the sample potential are made by insulated connectors in the sample stage base and within the dovetail coupling to the sample platform. Digital line connections from the sample stage platform through the dovetail mount to the sample stage base can be made for auto-recognition of the sample platform geometry.

Stepper motors are used to manipulate the translation stages. In the x and y directions the range is 25 mm to 200 mm; in the z direction the range is 50 mm to 100 mm. The translation stage should have a resolution of at least as good as 0.1 μm, and straight line accuracy of less than 5 μm per 100 mm of travel and less than 1.5 μm repeatability.

The translation system may be configured in one of two ways. In the first, for small scale x-y scanning as shown in figure 1, the actuator/reference electrode system is moved only in the z plane, while the sample stage 22 is moved in the x and y planes. In the second configuration, for large scale x-y scanning as shown in figure 2, the actuator/reference electrode system is moved in the y and z planes . The sample is coupled to the x translation stage in both cases.

In the first configuration (see figure 1) , the actuator/reference system is only translated in the z direction, and the sample stage moved in the x and y plane. The z-stage platform 20 is mounted either fixed or freely upon the upper chamber interface, the x-y stepper motor/stages 18 being mounted on the optical table under an elevated chamber 12. The chamber 12 is elevated using three solid cylinders 26 of precision-machined cast iron. Optionally, the cylinders 26 could be instead of some other metal alloy with comparable thermal expansion characteristics. Coupling the x-y stepper motor/stages to the tilt stage platform is a translation post made of stainless steel.

In the second configuration (see figure 2) , the actuator/reference electrode system is moved in the y and z planes and the sample is the x plane. The y stepper motor/stage systems 52 are mounted on a support platform made from cast iron. The chamber 54 and the support platform are positioned so that a platform bracket may be coupled between the y stepper motor/stages 52 and a 90° bracket 84 which is mounted on the z-stage interface 58. The support platform and y-z platform brackets are machined to precision allowing for exact linkage.

In both arrangements, translation of the actuator/Kelvin probe electrode (14,16) is made via a linkage between the stepper motor/stage unit, the upper chamber interface and the z stage platform. This configuration allows for the complete isolation of atmospheres inside and outside the chamber.

The z stage platform is a unit comprising of housing and two concentric sections that slide within each other. However, the sections may not travel past the top edge plane of the housing. Linking the actuator/reference system with the stepper-motor/stage unit are translation posts that are fixed to the central most section of the z- stage interface. The z-stage interface and translation posts are constructed of stainless steel and z translation is facilitated using vacuum grease and o-rings.

The z-stage platform is mounted to the z-stage interface. To the z stage platform is mounted the 90° bracket 84 by which the z stepper motor/stage is supported. The 90° bracket 84 has fine adjustment for exact linkage of the translation post between the stepper motor/stage and the z-stage interface. The z stage interface sits freely on the upper chamber interface. Both the bottom face of z stage interface and the upper surface of the upper chamber interface are coated with Teflon (or similar compound) to enhance frictionless contact.

Schematic diagrams of the measurement electronics and control configuration are shown in figures 6 and 7 respectively. A personal computer, running LabView brand software, is used to control every aspect of the Kelvin probe instrument system.

The measurement electronics 90 include a Lock-in Amplifier (LIA) 92 capable of digital signal processing, and a 200k samples/second 16 Bit AD/DA board (DAQ) with eight digital channels 94, the latter being integrated into the computer system 96. The voltage output on the DAQ 94 is controlled directly by the computer and software. The actuator driver signal 98 is also indicated.

The electronic circuit is completed by arranging the Kelvin probe electrode 16 and the sample surface 100 with a common ground 102.

Other measurement electronics include temperature and humidity sensors (not shown) . For translation system control, the stepper motor controller 104 is connected via the parallel port or dedicated interface card on the computer system 96, and the LIA 92 via an RS232 or GPIB interface card 108 on the computer system 96. Temperature and humidity sensors are connected via the DAQ 94, while the driver voltage 98 to the actuator 14 is supplied by the oscillator of the LIA 92.

Electrical connections for measurements employ the electrode mount 70, voltage feedback circuit to the sample via the DAQ 94 and ground connections. In an electrically unbalanced condition, the current 110 generated between the electrode 16 and the sample 100 is measured by the LIA 92 via the connection at the Kelvin probe electrode 16. This raw signal 110 is processed and DC voltages proportional to the in-phase current of the raw signal for the fundamental frequency 112 and first harmonic 114 are output to the inputs of the DAQ 94. Feedback control by the software sets a voltage at the sample via the analogue output of the DAQ 94.

The stepper motors are also controlled by the LabView brand software, which calls dedicated dynamic link library files (.dll files) located at the operating system level of the computer 96. Utilizing the parallel port or a dedicated interface card for communication, by the LabView brand software sends commands to the controller.

Also depicted in figure 7 are sample stage positioning signals 118 (between the sample stage 22 and the stepper motor controller 104) , nulling voltage 120 (send to the sample stage 22) and GPIB communications signals 122 (from the LIA 92 to the computer 96) .

In use the electrode 16 is vibrated in the z direction by the piezo-ceramic sheets 60a, 60b fixed in the actuator housing 64, as discussed above. A shielded connector from the LIA oscillator is typically set to between 3 and 5 Volts r.m.s., at the resonance frequency of the piezo- ceramic system (typically in the range 550-780 Hz, depending on Kelvin probe electrode mass) . The piezo- ceramic elements 60a, 60b are electrically isolated from the actuator housing 64 and the Kelvin probe electrode. The actuator housing 64 and auxiliary shielding connected to ground prevents the Kelvin probe electrode from picking up noise from the low voltage piezo driver system. The Kelvin probe electrode connection consists of a coaxial cable coupled with a light weight copper braid. The braid has a coiled (diameter 2 mm) length of approximately 4 mm and is soldered to the gold connector 70 (female) fixed to the electrode post 68. A platinum Kelvin probe electrode 16 with connection diameter 0.5 mm is inserted into the gold plated fixture and held in place by friction fit.

To facilitate automatic experimental procedures and obtain electronic data formats, all aspects of the Kelvin Probe instrument are computer controlled using an interface written in LabView brand programming language. The primary interface and sub-programs for performing experiments and controlling the functions of the using the Kelvin probe instrument is shown in a tree diagram shown in figure 8. The primary interface 130 allows the user to perform both stepper motor controller initialization and parameter set-up 132, and call various sub-programs. The sub-programs are divided into three groups: Utility 134, volta potential 136, and Contact Potential Difference (CPD) and Topography 138. Utility sub-programs 134 are grouped into Stepper Motor Utilities 140 and Signal Utilities 142. Shown below are the sub-programs in each group .

Operation of the Kelvin probe instrument 10 involves using a number of sub-programs to perform one or more of the following f nctions :

• Characterization of the Kelvin probe signal;

• Positioning of the Kelvin probe electrode;

• Nulling of the Kelvin probe signal;

• Raster Scanning; and

• Instrument System Diagnostics.

Some of the subprograms only perform one of the above functions, while others - such as Volta Potential Scanner - utilize all.

In the "Fly-over Volta Potential scanner" operational mode, the Volta Potential of a sample surface is measured while tracking the topography of the sample. The topography of the scan area is first obtained by means of the Contact Potential Difference and Topography operation. The coordinates of the topography so collected are then used by the Volta Potential scanner. The resulting data (see the description of figure 33C below) shows that spatial resolution and quality of data are markedly superior in this mode compared with the Contact Potential Difference mode.

The proximity of the Kelvin probe electrode tip to the sample surface approximates a parallel plate capacitor in which the distance between the plates changes with time. The resulting current signal i assuming a homogenous field between the plates can be expressed as:

,_*_ΔF*. 1.1 dt dt

where C is the capacitance and ΔV is the potential difference between the Kelvin probe electrode and sample surface. The capacitance of a parallel plate capacitor is defined by:

c=^≠,

where ε₀ is the dielectric constant, ε is the dielectric of the air, A is the area of the Kelvin probe electrode tip and d is the distance between plates. The Kelvin probe is actuated in a sinusoidal fashion and hence:

1 4 d = d + dι sin(ω ₅

where d₀ is the average distance between the Kelvin probe and the sample surface, d- is the amplitude of vibration and t is time in seconds. The frequency at which the probe is actuated is ω expressed in radians. Substituting equation 3 and 4 into equation 2 , the current for a homogenous field can be expressed as

If d₀ >> d then equation 1.5 can be approximated as:

< ωcos(ω i.6 i = -εεoAAV Hence, the magnitude of i is directly related to ΔV, A and the ratio of d to d₀. Hence, it is convenient to use the modulation factor m to understand the relationship that values of di and d₀ have on the current amplitude, and is expressed as:

m =— d_± . _ „ d₀ ¹ ' ^η

Shown in figure 9 is a screen capture of the waveform of the Kelvin probe signal captured on an oscilloscope, with both actuator driver waveform (top) and Kelvin probe waveform (bottom) shown. As predicted by equation 1.5 and demonstrated in figure 9, the resulting waveform of the Kelvin probe signal has contributions from several harmonics of the fundamental frequency. Illustrated in figure 10 is the change in current magnitude of the fundamental (squares) and first harmonic (circles) as a function of average probe tip distance, d₀. As the average distance approaches d_lΛ the current amplitude increases significantly, hence improving the sensitivity of the measurement. The gradient of the relationship between the applied voltage and the in-phase current is defined as the "sensitivity", as illustrated in figures 11 and 12 for the fundamental and first harmonic respectively. These curves display measurements at different values of o for a constant value of dj.. The intersection of these plots in an ideal case occurs at zero current, and this value corresponds to the volta potential difference between the probe tip and the surface.

The first harmonic of the Kelvin probe actuator frequency is employed rather than the fundamental frequency, as this provides a much improved signal to noise ratio. Signal to noise is calculated by determining the above-mentioned sensitivity, which - as shown in figures 11 and 12 - becomes greater with decreasing distance between the probe tip and the sample surface, while noise is defined as the variation in the out of phase current signal. Although the sensitivity of the fundamental measurement is much greater than the first harmonic, the noise on the first harmonic is far less in proportion to the fundamental frequency.

Thus, from figure 11 the sensitivity of the fundamental frequency signal is 15.83 pA/V, and the noise of the fundamental frequency signal is 1.360 pA. Consequently, the Signal to Noise ratio of the fundamental frequency signal (i.e. the ratio of sensitivity to noise) is 11.63 V.

However, although the sensitivity of the first harmonic signal is lesser (viz. 3.50 pA/V) , the noise in that signal is much less (viz. 0.033 pA) and as a result the Signal to Noise ratio for the first harmonic signal is 106.0 V, an order of magnitude better than for the fundamental frequency signal .

Indeed, the noise rejection is also far better when measuring the first harmonic of the in-phase current in comparison to measuring the current of the fundamental frequency. The null current condition is achieved by adjusting the samples potential so that there is zero current. There will always be some noise associated with the equipment used for signal processing, so an approximation is made for the zero current condition.

Variance on the fundamental channel is 0.0028 pA and on the first harmonic channel 0.00068 pA (n = 50); the ratio of these is about 4 in favour of the first harmonic. Cross-talk from the actuator is a form of noise which is the same frequency as the in-phase current and hence is not rejected by the signal processing performed by the Lockin amplifier. This noise contributes to an offset of the null condition, which is apparent in figure 11. The offset values for the fundamental frequency is 1.45 pA and for the first harmonic 0.019 pA, a ratio of about 76 in favour of the first harmonic.

The accuracy of measurements made by the first harmonic are more accurate than those made by the second. The effect of the offset noise in determining the null condition is also illustrated graphically by figure 11. The most accurate measurement of the sample volta potential is when the in-phase current equals zero. As mentioned above, an approximation is made for zero current, and this is usually a value which is 3-4 times the variance on the measurement channel. At 100 μm, the fundamental frequency has a sensitivity of 10.62pA/V, the first harmonic 1.34pA/V. For the fundamental frequency, accuracy could be no better than 0.26 mV and for the first harmonic 0.51 mV. The fundamental frequency is favoured in this respect because of greater sensitivity, but in addition to the variance, the offset noise value results in a difference between the measured null condition potential and the true null condition potential . In the case of the fundamental frequency this difference is 132 mV, and for the first harmonic 20 mV. The sum of these errors for the fundamental frequency is 132.3 mV and for the first harmonic 20.5 mV. In addition, this offset leads to a volta potential measurement (zero current) which is has a high degree of dependence on do.

Computer controlled positioning of the Kelvin probe tip is achieved by using a feedback loop which is based upon the ratio of the first harmonic to the fundamental frequency, i /io- Illustrated in figure 13 is a plot of the measured ratio ii/io as a function of m (squares) compared with a calculated ratio (circles) based on equations 1.8 and 1.9.

The calculation of the fundamental current amplitude and subsequently the ratio i±/io includes an offset value representative of the value measured from the intercept of the curves in figure 11. For a constant probe tip amplitude d±, accurate and reproducible positioning of the electrode tip based on the ratio can be performed irrespective of electrode area. Figure 14 is a plot of the ratio ii/io as a function of applied voltage to the sample surface, and illustrates that when an applied voltage is more than 500 mV positive or negative of the sample volta potential, the ratio iχ/io can be used to position the electrode tip irrespective of sample type. Deviation near the volta Potential is the result of differences in the rate of signal change between the two signals as they approach zero. By calibrating the Kelvin probe electrode, the contact potential difference (CPD) may be determined over samples which may have a tilted surface or have significant topographical features.

EXAMPLES

Firstly, below is an application and analysis matrix for the instrumentation of the scanning Kelvin probe instrument, in which a number of possible applications of Kelvin probe instruments of this invention are listed, together with examples of each application:

In the examples described below, the frequency of actuator 14 was between 470 and 450 Hz, the time-constant of LIA 92 was 50 ms and measurements were preformed using a platinum electrode.

EXAMPLE 1

Figure 15, for example, is a plot of the position of the z-translation stage and therefore topography (plotted as circles, calibrated on the left vertical axis) and of CPD (plotted as squares, calibrated on the right vertical axis) , from a line-scan over a tilted zinc substrate. In preparation the zinc sample had been polished with 800 grit SiC and then 14, 9, 3 μm diamond paste using non- aqueous lubricants, and rinsed with ethanol and dried under a stream of nitrogen. The measurements were performed with a platinum electrode of 100 μm diameter, an ii/io ratio of 0.2, and a step size of 25 μm, in an ambient atmosphere with 40% relative humidity.

The nulling of the Kelvin probe signal is performed by means of a feedback procedure. Feedback control is achieved using a sub-program which is illustrated in flowchart form in figure 16. The steps are as follows.

After setting the DAQ configuration and Kelvin probe calibration (150) , the feedback control parameters are set (152) . The DAQ reads the DC voltage output by the LIA (154) . This voltage is proportional to the in-phase component of the Kelvin probe signal. Next, the DC voltage is scaled (156) to provide a measure of the in- phase current i and the sign of the value is determined (158) . The value of i is checked to see if it is zero to within a predefined tolerance T (160) . This tolerance is discussed further below. If \ i \ ≤ T, the potential of the sample is not changed and the voltage is recorded as the volta potential (162) . If the current is outside the tolerance (viz. \ i \ > T) , the potential of the sample is incrementally changed in a positive (166) or negative (168) direction depending on the sign of the current signal. The voltage is adjusted by increments of the "sweep voltage resolution" (set by the user) in the positive direction if the in-phase signal is positive, and in the negative direction if the in-phase signal is negative, according to figure 11 and/or figure 12 (where the sign of the slope dictates the sign of the incremental adjustments) .

The tolerance T is determined from the product of the "voltage resolution" and the "sensitivity slope". Thus, T = voltage resolution x sensitivity slope.

For example, if 10 mV resolution of the volta potential was sought, and the sensitivity slope was 3 fA/mV, then the tolerance would be equal to T = 10 x 3 = 30 fA.

The rate at which the voltage is changed to approach the tolerance, depends on the "sweep voltage resolution", the "sensitivity slope" and the "gain". The sensitivity slope is determined from the relationship between the applied voltage and in-phase current, as illustrated in figures 11 and 12. The voltage resolution is multiplied by the quotient of the scaled signal magnitude and the sensitivity slope and sample voltage is adjusted accordingly. A gain factor between 1 and 5 may be used to further increase the sweep voltage resolution value. A threshold is used to define a level that, if exceeded, results in an accelerated voltage scanning rate being applied. This prevents oscillation or "ringing" of the feedback system. This sequence of operations continues until the scaled signal is less than the tolerance, with a delay in the loop structure that is equal to the time constant of the actuator frequency.

When the user starts the sub-program to determine the volta potential (136 in figure 8) , the parameters mentioned above are initialised. These parameters are: LIA sensitivity, sensitivity slope, voltage resolution, sweep voltage resolution and gain. After the null condition is achieved, the user may change these parameters to optimize data acquisition.

By using a calibrated Kelvin probe electrode, the volta potential differences may be compared with the electrochemical potential measured by conventional electrodes such as the saturated calomel electrode (SCE) or normal hydrogen electrode (NHE) . The calibration procedure involves measuring both the volta potential and the electrochemical potential of a reference system. The calibration is obtained by simultaneously measuring the potential of a M/M +2Sθ₄(_aq) couple with a conventional reference electrode and the volta potential using the

Kelvin Probe device.

To calibrate the Kelvin probe against a standardized electrode, a copper-copper sulfate couple is used. The Kelvin reference electrode is brought very close to the solution-air boundary of the copper-copper sulfate couple and the volta potential is measured.

Figure 17 is a cross-sectional view of the couple 170. Copper, iron and zinc cells 172 are filled with, respectively, CuS0₄, FeS0₄ and ZnS0₄ solution 174.

The copper-copper sulfate reference (CSR) consists of a copper well 172 containing 0.5 M copper sulfate solution 174. The copper 172 is approximately 4 mm thick and 10 mm in diameter. In using this reference cell 170, consideration has been made for reducing the surface tension on the solution in the well by ensuring that the centre of the well is close to the height of the edges. This avoids a large meniscus being formed and assists in the stability of the liquid-air boundary, which is important when positioning a vibrating electrode within 100 μm of the surface.

In preparing the CSR for calibration of the Kelvin probe electrode, copper oxide solids normally present after atmospheric exposure must be removed. This is easily done by using 1200 grit silicon carbide paper and 1 M nitric or sulfuric acid. Just before calibration, the CSR is rinsed in distilled water and then is rinsed once with copper sulfate solution and then filled to a level equal to that of the edge of the well sides.

Owing to the hygroscopic nature of copper sulfate, the CSR is ideal for measurements in high relative humidity environments. However, if measurements made in dry atmospheric conditions are problematic due to high evaporation rates and crystallisation of the copper sulfate onto the CSR.

During the measurement of the volta potential with the Kelvin probe electrode above the copper sulfate solution, the Kelvin probe electrode may become contaminated. If a large inverted meniscus is formed in the calibration cell, then when the Kelvin probe approaches the surface of the solution, the solution may "jump" to the Kelvin probe electrode. This results in contamination of the Kelvin probe electrode. If the Kelvin probe becomes contaminated, soaking in distilled water and then rinsing in ethanol may be used to restore it to the non- contaminated state. Drying in dry nitrogen or argon can follow this cleaning procedure. Again, the CSR must then be prepared again as described, without the presence of crystallisation of copper sulfate, nor blackening of the copper due to oxide formation.

EXAMPLE 2

Measurements were then performed with an electrode of 100 μm diameter, and in an argon atmosphere of 75% relative humidity. The results are plotted in figure 18. The plotted points may be fitted by a line with slope equal to 1.0 within experimental error. Once the calibration has been performed, a periodic calibration check may be performed prior or during an experiment using the Cu/CuSθ₄ cell.

By means of the instrument 10, the volta potential (null condition) can be established very quickly as consequence of the reduced noise when using the first harmonic in comparison to the fundamental frequency. Variance in the measurement channel results in the feedback control loop to take a greater number of iterations in incrementally adjusting the sample potential to achieve the null condition.

EXAMPLE 3

Further, simultaneous maps of potential and topography can be made of a sample surface that has a large degree of variation in volta potential such as that shown in figures 19A and 19B. Figure 19A is a volta potential map obtained using the ratio of the fundamental and first harmonic by means of the Kelvin probe instrument of figure 1, of a copper wire embedded in a zinc plate. The sample had been polished with 800 grit SiC and then 14, 9, 3 μm diamond paste using non-aqueous lubricants, and rinsed with ethanol and dried under a stream of nitrogen. The measurements were performed with a platinum electrode of 100 μm diameter, an iχ/io ratio of 0.4, and a step size of 50 μm, in air with 40% relative humidity.

Figure 19B is a topography map obtained simultaneously with the volta potential map of figure 26A, again using the ratio of the fundamental and first harmonic. For the analysis of this sample, the surface was purposely tilted to demonstrate the function of the automatic probe tip positioning.

Another important advantage concerns the methodology used to perform raster scanning. For both the determination of the null condition and height control, the tolerance or the ratio respectively is achieved before a new position will be measured. This ensures that the accuracy of the data is defined by the voltage resolution (entered by the user) for each point in the two dimensional space.

Use of the first harmonic in the measurement of in-phase current for obtaining the null condition and hence the volta potential of the sample surface may be used to determine, for example, differences in work functions of dissimilar metals. This is demonstrated in figures 19A and 19B (described above), and in figures 20, 21 and 22.

EXAMPLE 4

Figure 20 is a potential map covering an area of 2000 x

2000 μm, of a 1 mm copper wire embedded in a zinc substrate obtained with the Kelvin probe instrument 10 of figure 1 (fitted with a platinum Kelvin probe electrode) . In preparation the sample had been polished with 800 grit SiC and then 14, 9, 3 μm diamond paste using non-aqueous lubricants, and rinsed with ethanol and dried under a stream of nitrogen. The measurements were performed with a platinum electrode of 100 μm diameter, with a step size of 50 μm, in air with 40% relative humidity.

EXAMPLE 5 Figure 21 is a potential map covering an area of 1200 x 1200 μm, of a 0.5 mm aluminium wire in a zinc substrate obtained with the Kelvin probe instrument 10 of figure 1 fitted with a platinum Kelvin probe electrode. In preparation the sample had been polished with 800 grit SiC and then 14, 9, 3 μm diamond paste using non-aqueous lubricants, and rinsed with ethanol and dried under a stream of nitrogen. The measurements were performed with a platinum electrode of 100 μm diameter, with a step size of 50 μm, in air with 40% relative humidity.

EXAMPLE 6 Figure 22 is a potential distribution map of gold on aluminium obtained using the first harmonic and the Kelvin probe instrument of figure 1. The sample had been prepared by lithographic techniques, to produce a grid of squares 400 x 400 μm in size set 400 μm apart. The measurements were performed with a platinum electrode of

60 μm diameter, with a step size of 25 μm, in air with 40% relative humidity.

EXAMPLE 7 Defects in polymeric coatings on metals may also be inspected, as shown in figure 23A and 23B. A sample comprising epoxy (in the form of urethane) coated A12024- T3 alloy was prepared and scribed, then exposed to fuming HCI for 30 seconds followed by air at high humidity for 4 weeks in a separate chamber. A region of the surface of the coated alloy that contained a filiform track was then analysed using the Kelvin probe instrument of figure 1. The measurements were performed with a platinum electrode of 60 μm diameter, with a step size of 25 μm, in air with 90% relative humidity.

Thus, figure 23A is a potential distribution map of the filiform track after 16 hours at 90% relative humidity obtained with the Kelvin probe instrument of figure 1. Figure 23B is a potential distribution map of the filiform track referred to in figure 23A after a further 6 hours at 60% relative humidity, obtained with the Kelvin probe instrument of figure 1.

EXAMPLE 8

The instrument 10 can be used to view the effect of oxide formation on metals, as shown in figure 24. A sample of Al 2024-T3 alloy was prepared: the sample was dipped in nitric acid (pH 2) for 10 seconds, rinsed with distilled water and then dried under nitrogen. Next, one end of the sample was immersed in 1 M NaOH for 15 seconds; the other end was immersed for 30 seconds in the same solution.

After each immersion, the sample was rinsed with distilled water and dried under nitrogen.

The measurements were performed with a platinum electrode of 100 μm diameter, with a step size of 25 μm, in air with 40% relative humidity. Figure 24 is a three dimensional potential distribution plot over the Al 2024 sample; the end treated with 1 M NaOH for 15 seconds is at the left in this figure, while the end treated for 30 seconds is at the right in this figure. The middle portion was untreated.

EXAMPLE 9

The effects of the active corrosion of metals can be monitored, as illustrated in figure 25. Figure 25 is a three dimensional potential distribution plot of corrosion initiation by NaCI micro droplet deposition on zinc metal, acquired at room temperature in air with 95% relative humidity with the Kelvin probe instrument 10 of figure 1. The sample had been first polished with 800 grit SiC and then 14, 9, 3 μm diamond paste using non-aqueous lubricants, and rinsed with ethanol and dried under a stream of nitrogen, and the measurements were performed with a platinum electrode of 100 μ diameter, and with a step size of 50 μm. EXAMPLE 10

The location of corrosion products resulting from corrosion processes can be determined, as illustrated in figure 26. An Al 2024-T3 alloy sample was prepared as follows : the sample was first exposed to continuous neutral salt spray for 15 days. Upon removal from the salt spray chamber, the sample was rinsed with distilled water and dried under nitrogen. A region surrounding a pit was identified with the aid of a microscope, and subsequently this area was analysed using the Kelvin probe instrument of figure 1. Figure 26 is a three dimensional potential distribution plot of the pit and surrounding area of the sample, acquired in air at room temperature and humidity (of 40% relative humidity) using the Kelvin probe instrument. The measurements were performed with a platinum electrode of 60 μm diameter, and with a step size of 25 μm.

Other applications of the Kelvin probe instrument 10 include the determination or measurement of surface properties of painted metals, surface charging effects on semiconductors and surface charging effects on freestanding polymeric films.

Samples that have been analysed by the preferred embodiment scanning Kelvin probe instrument 10 principally comprise semi-conducting oxide material and polymeric material that have conducting surfactant coatings. The results from the analysis of these types of materials illustrate the capability of the instrument.

EXAMPLE 11

Thus, a sample of pure zinc sheet was spray painted with an enamel paint and then, on half the sample, the paint film removed using a scalpel blade. Measurements were made with a platinum electrode of 60 μm diameter, an iχ/io ratio of 0.3, and a step size of 50 μm, in air with 40% relative humidity. Figure 27A is a plot of the contact potential of the sample obtained using the scanning Kelvin probe instrument of this embodiment. Figure 27B is an optical image of the same sample, and figure 27C is a topography image of this sample.

EXAMPLE 12

A sample of Al 2024-T3 was prepared as follows: the sample was first polarised anodically 100 mV from the open circuit potential for 30 seconds in 0.1 M NaCI electrolyte saturated with oxygen, and then remained at open circuit potential in the same electrolyte for 20 minutes. The surface of the area exposed to the electrolyte was rinsed with distilled water and dried under nitrogen. A region surrounding a pit on the surface of the alloy was identified using the aid of a microscope, and subsequently this area analysed using the Kelvin probe instrument. Using the potential map obtained from this analysis, the spatial coordinates of the pit was determined and the Kelvin probe was positioned at that location. The Volta potential (null current) was maintained at a period equivalent to that of the measurement time constant.

Measurements were made with a platinum electrode of 100 μm diameter and an acquisition rate of 1 point per second, in air with 90% relative humidity. Figure 28 is a plot of the potential transient U (mV/SHE) of the pit (2200, -1000) versus time t in hours, after 6 hours in an ambient atmosphere, measured by means of the scanning Kelvin probe instrument.

EXAMPLE 13

A sample comprising magnesium metal was polished with 800 grit SiC and then 14 μm diamond paste using non-aqueous lubricants. The surface was rinsed with ethanol and dried under a stream of nitrogen. The sample was placed in the Kelvin probe chamber under an atmosphere of dry argon. The potential of the magnesium surface was measured, and then 7, 7, 8, 8-tetracyanoquinodimethane (TCNQ) dissolved in acetonitrile was drop-coated onto the surface. The Kelvin probe was then repositioned to measure the potential over the region were the thin film of Mg-TCNQ was formed and the potential measured. The Volta potential (null current) was maintained at a period equivalent to that of the measurement time constant.

Measurements were made with a platinum electrode of 100 μm diameter and an acquisition rate of 1 point per second, in an argon atmosphere with a relative humidity that varied over the course of the measurements between 20 and 83%. Figure 29 is a plot of the potential transient from the reaction of magnesium metal substrate with TCNQ at low and high humidity against time t in minutes. Arrow 200 indicates the dropping of the coat of TCNQ, and arrow 210 indicates the introduction of wet argon. Curve 220 is the measured relative humidity (Ar RH (%) ) of the argon, calibrated on the right axis; curve 230 is the measured potential U ( V/SHE) , calibrated on the left axis.

EXAMPLE 14

Sections of commercially available 35 mm polymeric film were mounted on the sample stage of the Kelvin probe instrument, and the chamber humidity was then adjusted to 15% relative humidity in air. Using a probe with a small piece of the same film attached, tribocharging of one side of the sample was achieved by friction of the two film surfaces. Immediately after this procedure, line scans were initiated in the direction toward the region of tribocharging from a distance of typically 8 to 10 mm. Linescans were initiated every 3 to 5 minutes tracking the same lateral coordinates for a time up to 4 hours from the initial tribocharging.

Measurements were made with a platinum electrode of 100 μm diameter and a step size of 50 μm.

Selected data from specific positions on the linear scan over a defined time period is plotted in figure 30A, a plot of potential ϋ (mV/SHE) from the tribocharging of the polymeric film versus time t in minutes. The data show the change in electrostatic charge as a function of time at different positions on the surface of the film relative to the location of tribocharge induction (see figure 3OB for the key to the curves, expressed as distance from the tribocharged region in μm) .

EXAMPLE 15

Figure 3OB is a plot of potential transients from the tribocharging of another commercially available polymeric film, again illustrating the change in electrostatic charge as a function of time at different positions on the surface of the film relative to the location of tribocharge induction. Each curve corresponds to measurements made at a different distance from the tribocharged region; the key to the curves at the right of the plot indicates the distance from the tribocharged region in each case in μm.

EXAMPLE 16

A zinc oxide film was prepared on ITO coated glass according to the method described by Izaki (Izaki, M., "Preparation of Transparent and Conductive Zinc Oxide Films by Optimization of the Two Step Electrolysis Technique", Journal of the Electrochemical Society,

146(12) (1999) 4517-4521). After electrolytic preparation of the zinc oxide film, the sample's surface was rinsed with distilled water and dried under nitrogen. A potential map of the sample was acquired by means of the scanning Kelvin probe instrument in the dark, until a point when a spot light from a conventional microscope illuminator focused at the surface of the sample was turned on. Measurements were made with a platinum electrode of 100 μm diameter, a step size of 50 μm, and in air with relative humidity of 40%. Figure 31 is an image of the surface potential of the zinc oxide film formed on the ITO coated glass in the dark (upper regions in this figure) and under illumination (lower regions) .

A transient was also obtained from this sample by alternating illuminated and dark conditions. Illumination was performed using a conventional light microscope light source. The measurements were made with a platinum electrode of 100 μm diameter, an acquisition rate of 1 point per second, and in air with relative humidity of 40%. The results are plotted in figure 32 as potential 17 (mV/SHE) versus time t in seconds. Hatched regions of the graph indicate when the surface is in the dark.

EXAMPLE 17

A sample comprising gold patterned on aluminium was prepared by lithographic techniques, producing a grid of squares 200 x 200 μm in size set 200 μm apart. To obtain a Volta potential map, a topography map was first obtained using the controlled height "fly-over" technique described above (with an i±/io ratio of 0.4). The topography data was then utilised to accurately position the probe during the Volta potential measurement. Measurements were made with the preferred embodiment scanning Kelvin probe instrument with a platinum electrode of 60 μm diameter and a step size of 10 μm, in air with relative humidity of 40%.

Figure 33A is a contact potential map of the gold on aluminium sample. Figure 33B is a topography map of this sample, and figure 33C is the Volta potential map of the sample obtained with the scanning Kelvin probe instrument in "fly over" mode.

Modifications within the scope of the invention may be readily effected by those skilled in the art. It is to be understood, therefore, that this invention is not limited to the particular embodiments described by way of example hereinabove .

Any reference herein to prior art is not intended to suggest that such prior art is common general knowledge.

Claims

THE CLAIMS DEFINING THE INVENTION ARE AS FOLLOWS:

1. A method of determining the contact potential difference between two materials, comprising: arranging said materials in a Kelvin probe instrument; measuring a harmonic of the Kelvin probe signal; and determining said potential difference from said harmonic.

2. A method as claimed in claim 1, wherein said two materials comprise respectively a sample and a Kelvin probe electrode.

3. A method as claimed in either claim 1 or 2, wherein said Kelvin probe signal comprises the in-phase current.

4. A method as claimed in any one of the preceding claims, wherein said harmonic is the first harmonic.

5. A method as claimed in any one of the preceding claims, including defining the resolution of data from said instrument by defining voltage resolution.

6. A method as claimed in any one of the preceding claims, wherein measuring said harmonic comprises measuring the entire Kelvin probe signal and processing said signal to obtain said harmonic.

7. A method as claimed in any one of the preceding claims, wherein said Kelvin probe instrument includes a piezo-electric actuator.

8. A method as claimed in any one of the preceding claims, including determining the relative topography of two surfaces simultaneously with said determining of said contact potential difference.

9. A method as claimed in any one of the preceding claims, including measuring the Volta Potential of a sample surface while tracking the topography of said sample.

10. A method of determining the topography of a sample, comprising: arranging said sample in a Kelvin probe instrument; measuring a harmonic of the Kelvin probe signal; and determining said topography from said harmonic.

11. A method as claimed in claim 10, including obtaining a null condition by altering sample potential incrementally until probe current is within a predefined tolerance of zero.

12. A method of analyzing a Kelvin probe output signal, comprising: deriving a first harmonic of said signal from said output signal; and determining one or more quantities from said first harmonic selected from the set comprising contact potential difference and topography.

13. A Kelvin probe instrument, having: signal processing means for extracting a first harmonic signal from a Kelvin probe output signal; and data processing means for determining from said first harmonic signal one or more quantities selected from the set comprising contact potential difference and topography.

14. A Kelvin probe instrument as claimed in claim 13, wherein said instrument is operable to measure the Volta

Potential of a sample surface while tracking the topography of said sample.