US20040253431A1

US20040253431A1 - Supporting substrate used for the deposition, automated recognition and spectroscopic identification of particles

Info

Publication number: US20040253431A1
Application number: US10/478,588
Authority: US
Inventors: Lothar Holz; Oliver Valet; Markus Lankers
Original assignee: rap ID Particle Systems GmbH
Current assignee: rap ID Particle Systems GmbH
Priority date: 2001-05-31
Filing date: 2002-05-24
Publication date: 2004-12-16
Also published as: DE50213993D1; EP1397664A1; WO2002097400A1; DE10127537C1; US7016034B2; EP1397664B1; ATE448476T1; US20060017919A1

Abstract

A supporting substrate for the deposition, automated recognition and spectroscopic identification of particulate impurities in liquid or gaseous media, comprising a filter membrane of polymer materials of a defined pore width, wherein the surface of the filter membrane is coated with metal which in the selected wavelength range for spectroscopic identification has no spectral features and at the selected excitation wavelength absorbs no or only little of the laser energy which is radiated in, and has a very smooth structure.

Description

The present invention concerns a supporting substrate for the deposition, automated recognition and spectroscopic identification of particles, in particular particulate impurities and in particular for use with Raman spectroscopy.

BACKGROUND OF THE ART

Different methods are known for the deposition of solid particulate impurities from air or liquids. The simplest methods are based on the deposition of the particles on filter membranes both in gaseous and also liquid media with subsequent analysis by means of suitable methods such as light microscopy, scanning electron microscopy or gravimetric analysis (see for example Millipore Particle Monitoring Guide, Millipore Corporation, 1998).

The filter membranes generally comprise polymeric materials, such as for example nitrocellulose, nylon, FTFE or PVC, of an exactly defined pore size, wherein the particles of larger diameter than the pore width of the filter accumulate on the latter and can subsequently be analysed. In recent times, for many applications in the area of microelectronics, ascertaining and analysing particularly small particles, so-called micro-particles, in the range of sizes of about 10 μm or less, is of particular interest, the analysis of which is problematical with the hitherto known methods, by virtue of the size relationships of the particles to be analysed.

Metal filters are also known, such as for example metal filters consisting of silver, for filtration purposes, from Millipore, which however because of their method of manufacture have a surface which, by virtue of its roughness, is not suitable for the recognition or identification of individual particles of <5 μm. The corporation alto tec GmbH of Hamburg offers gold-plated filters for determining asbestos concentrations, which are also not optimised for the described use.

Procedures exist for the quantitative contamination analysis of smooth surfaces, such procedures using a laser beam and a laser scanner for scanning surfaces and detecting deviations from a plane by means of the scattered light which is collected with a photodetector. Such a method is set forth in U.S. Pat. No. 5,479,252. It is however not possible to implement chemical characterisation of the particles with that method.

Other methods, such as that set forth, for example, in U.S. Pat. No. 6,178,383, investigate video images in digital form with image recognition programs and, besides the recognition of particulate irnpurities, can also provide information about the shape and/or size thereof. The equipment for methods of that kind however is very costly in comparison with the laser technology, and also identification of the particles with that method is not possible. The resolution of those methods is admittedly theoretically only diffraction-limited but it is difficult to determine the size of particles which are smaller than 1.5 μm.

Methods of Raman spectroscopy are known for the qualitative and quantitative analysis of the composition of a sample, in particular of microparticles (M Lankers, J Popp, G Rössling and W Kiefer, Chem Phys Let 277 (1997) 331-334) and have proven to be advantageous. In that case, a sample is irradiated with intensive electromagnetic monochromatic radiation, for example laser light. For that purpose, electromagnetic radiation from the visible or ultraviolet spectral range is usually employed. Upon measurement of the scattered light with a spectrometer and a suitable detector, that is to say when determining the beam intensity of the scattered light as a function of wavelength, the result obtained is a spectrum which comprises a strong line, the so-called exciter line, and very many weaker lines, the so-called Raman lines. The exciter line has the same wave number as the incident radiation. The Raman lines respectively correspond to specific rotational or vibrational states of the substance to be investigated. The Raman lines are arranged on a wave number scale symmetrically with respect to the exciter line. In addition the Raman lines are of an intensity which is between 10 ⁻³and 10⁻⁴times less, with the intensity of the Raman lines on the low-frequency side usually being substantially greater at ambient temperature than those on the higher-frequency side.

The Raman spectrum, that is to say the sequences of Raman lines, is characteristic in respect of each substance. A compound can be identified by comparison of its spectrum with the spectra of known compounds.

It will be noted however that the low level of efficiency of Raman spectroscopy is found to be problematical when using that procedure. It is necessary to use very high laser powers for investigating small amounts of substances, as is the case when investigating microparticles. In that respect it is undesirable that the focus of the laser beam is generally markedly larger than the diameter of the particle. Thus there is the unwanted consequence that the signal of the supporting substrate is recorded at the same time and in that situation the spectrum of the particle is slightly overlapped. That becomes clear from the area relationships. If a focus of about 10 μm in diameter is used in order to investigate a particle of a diameter of 1 μm, the supporting substrate/particle signal relationship is about 10:1. In most cases that makes it impossible to characterise the particle. In some cases it is possible to resolve the problem by focusing the laser beam to 1 μm. In that situation however the energy density rises severely and results in damage or a modification as a consequence of burning or photochemical reactions on the part of sensitive substances.

Therefore the object of the present invention is to provide supporting substrates for the spectroscopic analysis of particles, preferably Raman spectroscopy, which reduce the above-indicated disadvantages, in particular in the analysis of microparticles, to such a degree that reliable analysis results are obtained and which in addition are suitable for the filtration of both liquid and also gaseous media.

SUMMARY OF THE INVENTION

In accordance with the invention that object is attained by a supporting substrate for the deposition, automated recognition and spectroscopic identification of particulate impurities in liquid or gaseous media, comprising a filter membrane of polymer materials of a defined pore width, wherein the surface of the filter membrane is coated with metal which in the selected wavelength range for stereoscopic identification has no spectral features and at the selected excitation wavelength absorbs no or only little of the laser energy which is radiated in, and has a very smooth structure.

In that way it is now possible to reliably identify even particulate impurities in the micro range, in particular of between about 1 and 10 μm, by means of spectroscopic analysis, in particular Raman spectroscopy, and to obtain virtually unfalsified analysis results. Focusing of the laser beam on to the particle diameter and therewith a great increase in energy density with the unwanted consequences of damage to or a change in the supporting substrate and/or the particle to be analysed is no longer required, whereby the range of application of Raman spectroscopy is enlarged and reliable analysis procedures are achieved. The coating with a thin metal layer permits the inexpensive manufacture of supporting substrates, with the advantageous properties of the metals for the selected wavelength ranges still being retained. The metal-coated membranes permit various pore sizes or widths and are suitable for the deposition of the particles both from gaseous and also liquid media.

The fact that the metal layer, in the selected excitation wavelength range, absorbs no or only little of the radiated-in laser energy, avoids destruction of the coating and/or the particle to be analysed. For example gold-coated filters cannot be used for the investigation of particles in the near-infrared range (700-1070 nm). Power densities of about 80 kW/cm ²are required for investigating the particles. In that range however the gold layers only have a load-carrying capacity of less than 1 kW/cm². It is however possible to use gold-coated filters in the rest of the spectral range. A further example is represented by a silver coating. Here, no investigation procedures are possible in the range of 350-500 nm with 80 kW/cm². Aluminum cannot be used in a spectral range of 240-280 nm.

Preferably filter membranes with a very smooth surface (roughness RMS<1 μm) are used, for example polycarbonate, polytetrafluoroethylene, such as that sold commercially by duPont under the trademark TEFLON, or cellulose acetate membranes, with defined pores of for example 0.2; 0.8 or 1.2 μm.

The metal layer comprises, for example, nickel, aluminum, palladium, platinum, tungsten, iron, tantalum, rhodium, cadmium, copper, gold, silver, indium, cobalt, tin, silicon, germanium, tellurium, selenium or an alloy of those metals. The thickness of the coating is preferably between 50 and 200 nm.

The supporting substrates according to the invention are particularly suitable for Raman spectroscopy, but also for other spectral analysis methods.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention is described in greater detail hereinafter by means of embodiments with reference to the accompanying drawings in which: [0017]
FIG. 1 shows the Raman spectrum of polystyrene beads on an uncoated polycarbonate filter, and [0018]
FIG. 2 shows the Raman spectrum of polystyrene beads on a polycarbonate filter coated in accordance with the invention.[0019]

DETAILED DESCRIPTION OF A PREFERRED EMBODIMENT

For the experiment laser of a wavelength of 785 nm is used, preferably a Tuioptik Laser. The laser light is coupled into a microscope by means of a mirror and focussed on to the particles with an objective lens, preferably a Nikon ULWD [0020] 40 lens. The backscattered light is collected by the same objective lens, coupled into a fiber and the exciter wavelength is filtered out by means of a notch filter, preferably a supernotch filter (Kaiser Optical). Finally the inelastically scattered light is spectrally divided up in a spectrometer, preferably an Acton spectrometer, and the Raman lines are recorded with a detector, for example a Backthinned CCD camera.
FIGS. 1A and 1D respectively illustrate the pure spectra of polystyrene, the particle to be analysed, and polycarbonate, the filter membrane. FIG. 1B shows the spectrum of a polystyrene bead of a diameter of 3 μm which is disposed on an uncoated polycarbonate membrane. The focus of the laser beam with which this investigation was implemented is of a diameter of about 10 μm. In practice only spectral features of the filter material are to be recognised, identification of the material to be analysed is practically impossible (see [0021] 1A, 1D). FIG. 1C shows the spectrum of the same particle upon irradiation with a narrow laser focus of about 1 μm. The spectral components of the polystyrene beads can only be weakly recognised. The overall spectrum however is still strongly dominated by the features of the filter material.
The following comparative example involves the use of a polycarbonate filter of 25 mm diameter and a pore diameter of 0.8 μm which in accordance with the invention is vapor-deposited with aluminum in a layer thickness of 100 nm. [0022]
In comparison with the above-illustrated spectra FIG. 2D shows the spectrum of pure polystyrene and the aluminum coating. The spectrum of the coated filter does not have any structure whatsoever. The spectrum of the polystyrene beads of a diameter of 3 μm, illuminated with a 10 μm laser focus but on an aluminum vapor-deposited filter is to be seen from FIG. 2B. No falsifying bands on the part of the polycarbonate are to be recognised. The essential spectral features of the particle to be analysed are to be seen so that unambiguous identification is now possible. The same effect can also be observed for analysis of the particle with a focus of 1 μm, see FIG. 2C. Most spectral features of the comparative spectrum can be found again here. [0023]
The example makes it clear that identification of small particles by Raman spectroscopy on commercially available filter membranes with laser foci in the range of 2-10 μm is not possible. A marked improvement in identification is achieved with the filter membranes coated in accordance with the invention. [0024]

Claims

What is claimed is:

1. A supporting substrate for the deposition, automated recognition and spectroscopic identification of particulate impurities in liquid or gaseous media, comprising:

a filter membrane of polymer materials of a defined pore width,

wherein the filter membrane has a surface coated with a metal layer that has no spectral features in a wavelength range selected for spectroscopic identification and that absorbs no or only little of the laser energy which is radiated in at a selected excitation wavelength, the surface coating having a thickness of between about 50 and 200 nm, and having a very smooth structure with a roughness of <1 μm.

2. The substrate of claim 1, wherein:

the metal layer is selected from the group of metals consisting of: nickel, aluminum, palladium, platinum, tungsten, iron, tantalum, rhodium, cadmium, copper, gold, silver, indium, cobalt, tin, silicon, germanium, tellurium, selenium and alloys thereof.

3. The substrate of claim 2, wherein:

the metal layer is vapor-deposited.

4. The substrate of claim 3, wherein:

the filter membrane comprises polycarbonate, polytetrafluoroethylene or cellulose acetate.

5. The substrate of claim 4, wherein:

the defined pore width is between about 0.2 and 1.2 μm.

6. The substrate of claim 5, wherein:

spectroscopic identification is effected by Raman spectroscopy.

7. The substrate of claim 1, wherein:

the metal layer is vapor-deposited.

8. The substrate of claim 3, wherein:

9. The substrate of claim 7, wherein:

10. The substrate of claim 1, wherein:

the defined pore width is between about 0.2 and 1.2 μm.

11. The substrate of claim 8, wherein:

the defined pore width is between about 0.2 and 1.2 μm.

12. The substrate of claim 9, wherein:

the defined pore width is between about 0.2 and 1.2 μm.

13. The substrate of claim 5, wherein:

spectroscopic identification is effected by Raman spectroscopy.

14. The substrate of claim 1, wherein:

spectroscopic identification is effected by Raman spectroscopy.

15. The substrate of claim 10, wherein:

spectroscopic identification is effected by Raman spectroscopy.

16. The substrate of claim 11, wherein:

spectroscopic identification is effected by Raman spectroscopy.

17. The substrate of claim 12, wherein:

spectroscopic identification is effected by Raman spectroscopy.