Thickness Measurement with X Ray Polycapillary Optics

Ray Optics; Confocal Pachymetry; Capillary X-ray Lenses; X-ray Fluorescence

Summary

In order to realize the micro-regional non-destructive analysis of the thickness of thin films and coating materials, a confocal microbeam X-ray fluorescence thickness gauge with a common laboratory X-ray light source was designed and built by using a polycapillary X-ray converging lens and a polycapillary X-ray parallel beam lens. , the performance of the confocal thickness gauge was systematically characterized. The thickness of the Ni independent film sample with a thickness of about 25um and the thickness of the Ni film sample with a thickness of about 15um pressed on the silicon substrate were measured by the thickness gauge, and the relative measurement errors were 3.7% and 6.7%, respectively.

In addition, the thickness uniformity of Ni thin film samples with a thickness of about 10 μm was also measured. The confocal thickness gauge can perform micro-area depth analysis of the sample, and has the element resolution capability, so that the spectrometer can measure the film thickness of different layers of the multilayer film sample, and has potential applications in the field of thin film and coating thickness characterization.

Introduction

The performance characterization technology of solid material thin films and surface coatings has always attracted the attention of materials researchers. It has been widely used in metallurgical industry, magneto-optical recording materials, semiconductor materials, optoelectronic functional ceramic materials, electronic components, and aerospace equipment wear-resistant and high-temperature resistant materials. There are urgent needs. Important properties such as chemical composition, uniformity and thickness of film and coating materials directly affect the use of materials.

For example, the thickness of copper-nickel electroplating layer on soft magnetic material has a very important influence on its protective performance, magnetic performance, electrical performance and spot welding performance. At present, the commonly used methods for measuring the thickness of thin films and coatings mainly include: chemical acid leaching method, optical interferometer method, cross-sectional microscopic scanning, etc.

Each of these methods has its own advantages, but also has its own shortcomings. For example, none of these methods has high or no elemental resolution capability; in addition, some of these methods have special requirements for sample carriers, and some need to be performed under the condition of destroying the sample. In order to develop non-destructive thickness measurement technology, X-ray is applied to thickness measurement analysis.

The traditional X-ray thickness measurement equipment mainly includes X-ray penetration thickness gauge, X-ray fluorescence thickness gauge, X-ray scattering thickness gauge and X-ray photoelectron thickness gauge. The penetration thickness gauge is mainly based on the attenuation law in the interaction process of X-rays and matter, that is, the intensity of primary X-rays decays exponentially with the thickness of the penetrated material, and the intensity of initial X-rays and transmitted X-rays are measured.

The intensity of the rays, so that the film thickness can be calculated. But this method does nothing for thin films attached to thick substrates, which absorb the original X-rays, resulting in transmitted X-rays that are too weak to measure. In order to overcome the above shortcomings, X-ray scattering thickness gauges, X-ray photoelectron thickness gauges and X-ray fluorescence thickness gauges have been developed.

These three thickness gauges can be used in the measurement of film thickness on thick substrates. Among them, X-ray thickness gauges Ray scattering thickness gauges include X-ray elastic scattering thickness gauges and X-ray inelastic scattering thickness gauges. Although the elastic scattering thickness gauge can not be limited to the thickness measurement of crystal samples, this scattering thickness measurement method is still insufficient in element resolution.

The X-ray photoelectron thickness gauge can only work in a vacuum environment and can only measure the thickness of ultra-thin films. X-ray fluorescence thickness gauges include conventional X-ray fluorescence thickness gauges, total reflection X-ray fluorescence thickness gauges and grazing emission X-ray fluorescence thickness gauges. X-ray fluorescence thickness gauges have elemental resolution capabilities.

The existing X-ray fluorescence thickness gauge is to determine the thickness of the film through the theoretical relationship between the fluorescence intensity of the characteristic X-ray corresponding to a certain element in the film and the film thickness, and use the measured X-ray fluorescence intensity to invert the film thickness.

Due to the error limit of the theoretical calculation, the accuracy is limited to a certain extent. In addition, the commonly used X-ray fluorescence thickness gauge cannot perform micro-area thickness measurement. To overcome the above shortcomings, this work designs a confocal microbeam based on a polycapillary focusing-ray lens (PCFXRL) and a polycapillary paraIlel x-ray lens (PCPXRL). X-ray fluorescence thickness gauge. The confocal micro-element of the confocal thickness gauge is used to scan the film sample to be tested, and the thickness of the film is obtained according to the full width at half maximum of the scanning curve, which is beneficial to improve the measurement accuracy.

In addition, the confocal thickness gauge can perform multi-point micro-area thickness measurement analysis on the sample, so as to analyze the uniformity of the sample thickness and measure the film thickness of different layers of the multilayer film sample.

1. Experimental part

1.1 Instrument Design

Figure 1 is a schematic structural diagram of the designed and built confocal microbeam X-ray fluorescence thickness gauge for PCFXRL and PCPXRL.

Fig.1 Scheme of confocal microbeam X-ray fluorescence thickness gauge

In order to give full play to the transmission performance of PCFXRL, a small focal spot molybdenum target X-ray light source is used, the focal spot diameter is about 50um, and the weight is about 0.9 kg. The X-ray silicon detector used has a resolution of about 192 eV at 5.9 keV energy and weighs about 0.5 kg. PCFXRL, and PCPXRL are compact integral polycapillary X-ray lenses, the specific parameters are shown in Table 1.

Table 1 Parameter of X-ray optics lens

wdt_ID Lens Length / mm Input focal distance / mm Output focal distance / mm Gain at 17.0 keV
1 PCFXRL 38.5 45.6 16.7 1 517
2 PCPXRL 12.1 11.2 - 6

Use the five-dimensional debugging frame to adjust the relative position between PCFXRL and the light source, so that the light source is just at the entrance focal spot of PCFXRL and fixed. In addition, the solid PCPXRL and silicon detectors are co-located in a lightweight 5D debug stand. With the aid of CCD, the inlet focal spot of PCPXRL and the outlet focal spot of PCFXRL are adjusted to be in the same position, that is, confocal.

At this time, the detector can only detect the X-ray fluorescence signal from the confocal micro-element, so the three-dimensional information of the sample can be obtained non-destructively with the relative movement of the confocal micro-element and the sample. The size of the confocal micro-element depends on the performance of PcFXRL and PCPXRL.

From the geometric relationship between the exit focal spot of PCFXRL and the entrance focal spot of PCPXRL in the micro-element space, the dimensions of the micro-element along the Y-axis and Z-axis in Figure 2 are are equal, and the dimension in the X-axis direction takes the smaller of the diameter of the exit focal spot of PCFXRL ∅1 and the diameter of the entrance focal spot of PCPXRL ∅2. Using dx, dy and dz to represent the dimensions of the micro-element along the X, Y and Z axes, respectively, there is the following approximate theoretical calculation formula:

The sample stage is placed on a three-dimensional electric debugging frame, and the automatic control program of the electric debugging frame is integrated with the control program of the detector, so as to realize the automatic scanning and detection recording of the sample. In order to process the detected data quickly, data automatic reading and processing software has been programmed. The above-mentioned light source, detector and polycapillary X-ray lens are the main components of the thickness gauge. These main components are small in size and light in weight, thereby ensuring the confocal thickness gauge is compact and flexible.

1.2 Measurement of confocal cell size

Figure 2 is a schematic diagram of a confocal micro-element of a microbeam X-ray fluorescence thickness gauge.

Fig. 2 Scheme of cofocal micro-volume

Scan the micro-element with a metal filament along a certain coordinate direction in the figure, and a series of fluorescence spectra of the metal filament can be obtained (Fig. 3).

Fig. 3 X-ray fluorescence spectrum of Ni-Cr wire

Since the initial X-ray is focused by the PC, its intensity follows a Gaussian distribution in the micro-element space. , therefore, the intensity of the element fluorescence peaks in the above series of fluorescence spectra forms a Gaussian distribution along the scanning direction.

The Y-axis and Z-axis coordinates of the fixed sample stage are fixed, and the working voltage and current are 25 keV and 145 uμA, respectively. Use 25 μm NK paste filament to scan along the X axis to find Ni. The maximum strength of the K line. Then, fix the X-axis coordinate to the above-mentioned maximum fluorescence intensity, keep the y-axis still, and scan the Z-axis in the same way to obtain the Gaussian distribution curve of the NH(I) line intensity as shown in Figure 4.

Fig. 4 Fluorescence intensity distribution of Ni-Ka of Ni-Cr wire along Z axis (Fig.2)

Finally, fix the Z-axis coordinate to the maximum fluorescence intensity shown in Figure 4, keep the y-axis unchanged, scan the X-axis again, and obtain the NH(I) line intensity Gaussian distribution curve as shown in Figure 5.

Fig. 5 Fluorescence intensity distribution of Ni-Ka of Ni-Cr wire along X axis (Fig.2)

The confocal cell sizes such as xi and dx are related to the full width at half maximum FWHMz and FWHMx of the Gaussian distribution in Fig. 4 and Fig. 5, And the Ni-Cr filament diameter D has the following relationship:

In the formula,
D=25.0μm,
FWHMz=76.2μm,
FWHMx=44.6μm,
dz and dx are calculated to be 74.1 and 52.2μm, respectively.

1.3 Measurement of film thickness and thickness uniformity

The Ni independent thin film sample S1 with a thickness of about 25μm was measured, and the Gaussian distribution curve of its fluorescence intensity along the Z-axis coordinate is shown in Figure 6.

Fig. 6 Fluorescence intensity distribution of Ni-Ka of 25 μm Ni thin film along Z axis(Fig.2)

In order to simulate the measurement of the surface coating thickness, the same scanning method was used to measure the Ni film S2 with a thickness of about 15μm pressed on the surface of the silicon substrate (Fig. 7).

Fig. 7 Fluorescence intensity distribution of Ni-Ka of 15 μm Ni thin film along Z axis(Fig.2)

Since the fluorescence intensity of the Ni thin film sample scans along the Z axis, the full width at half maximum FWHMs of the Gaussian distribution curve, the size of the confocal microelement dz and the thickness t of the Ni thin film have the following relationship:

As has been determined, the available film thicknesses were calculated from FWHMs1 and FWHMs2 in Figures 6 and 7 (Table 2).

Table 2 Results of thickness measurements

wdt_ID Sample Real thickness / um Measured thickness / um Relative error / %
1 S1 24.5 23.6 3.7
2 S2 14.9 15.9 6.7

In order to measure the uniformity of the thickness of the Ni film of about 10 μm pressed against the surface of the silicon substrate, a micro-thickness analysis was performed on an area of 550 μm × 750 μm (Fig. 8). The scan steps in the X and Y directions were 55 and 75 μm, respectively.

Fig. 8 Thickness of 10 μm Ni thin film

2 Results and Discussion

Experiments show that the relative error of the used confocal microbeam X-ray fluorescence thickness gauge is between 1% and 10% when analyzing and measuring the thickness of thin films and coatings in the range of 10 to 30 μm. The main source of error is that in the process of thickness measurement, the angle between the sample surface and the Z-axis of the confocal micro-element (Fig. 2) deviates by 90º, which causes the measurement result to be too large.

Since PCFXRL also has a gain factor of the order of 103, PC HO also increases the power density of the x-ray beam entering the detector (Table 1), making many experiments that must be performed with synchrotron radiation sources available in the laboratory Ordinary X-ray light source implementation. The confocal thickness gauge is compact and flexible, which is convenient for on-site thickness measurement and analysis. Both the exit focal spot of PCFXRL and the entrance focal spot of PCPXRL have small diameters, therefore, micro-analysis of the thickness uniformity of the sample can be performed. Since different elements correspond to different characteristic X-ray fluorescences, the confocal microbeam X-ray fluorescence thickness gauge has element resolution capability.

The physical parameters of PCFXI and PCPXI used in this experiment determine the size of the confocal micro-element between 60 and 80 μm. In order to improve the spatial resolution of the confocal microbeam X-ray fluorescence thickness gauge, the focal spot diameter can be designed. Smaller PCFXRL and PCPXRL, or other X-ray converging elements, to enable confocal microbeam X-ray fluorescence thickness gauges to measure nano- to sub-micron film and coating thicknesses.

3 Conclusions

The micro-element size using PCFXRL and PCPXRL confocal microbeam X-ray fluorescence thickness gauges is 52.2 μm×74.1 μm×74.1 μm. When using capillary X-ray lens confocal microbeam thickness measurement technology to measure the thickness of thin films and coatings in the range of 10-30 μm, the relative error is less than 10%.

Due to the high power density gain and small focal spot capillary X-ray lens, this confocal microbeam thickness measurement technique can use a portable laboratory low-power X-ray light source. The confocal microbeam thickness measurement technology can perform multi-point thickness measurement analysis on the sample, so that the thickness uniformity of the sample can be analyzed. In addition, the thickness measurement technology can also perform thickness measurement analysis on the layered sample.

Capillary X-ray lens confocal microbeam thickness measurement technology has opened up the application field of capillary X-ray lens confocal technology, and enriched the X-ray non-destructive side thickness technology, which has broad application prospects.