Process for ion implantation

Abstract

The present invention relates to a substrate comprising an ion-implanted layer, for example a cation, wherein the ion implanted layer has a substantially uniform distribution of the implanted ions at a significantly greater depth than previously possible, to a well-defined and sharp boundary within the substrate. The invention further comprises said sub-strate wherein the substrate is a silicon based substrate, such as glass. The invention also comprises the use of said material as a waveguide and the use of said material in measurement devices.

Claims

1. A process for fabricating an ion-implanted layer in a substrate comprising silicate glass, the process comprising: providing a target layer comprising a tellurite glass in a vacuum chamber at a reduced pressure; providing said substrate in the vacuum chamber which includes a surface of the substrate in proximity to said target layer wherein the surface of the substrate is spaced from the target layer by a distance in the range 50 mm to 70 mm; heating the substrate to a temperature in the range from 0.55 T.sub.g to 0.75 T.sub.g where T.sub.g is the glass transition temperature of the substrate; and, directing pulses of incident radiation from a laser at the target layer thereby ablating material of the target layer with said incident radiation to produce therefrom a plume of the ablated target material capable of implanting into the heated substrate with the target layer and the substrate in oxygen gas at said reduced pressure within the range 80 mTorr to 90 mTorr, whereby said ablated material of said plume is implanted into the heated substrate by entering the heated substrate via the surface of the heated substrate such that a film of said ablated target material is not formed on the surface of the heated substrate, thereby forming a layer of implanted ions of said ablated target material within the heated substrate which extends from the surface of the heated substrate to a depth of at least 50 nm and which has a density of said implanted ions of said ablated target material of at least 10.sup.21 ions cm.sup.3.

2. A process according to claim 1 wherein the laser is a Femtosecond laser.

3. A process according to claim 1 wherein said implanted ions of said layer of implanted ablated target material are arranged in a spatial distribution which is substantially uniform along a direction extending into the heated substrate from the surface of the heated substrate.

4. A process according to claim 1 wherein said layer of implanted ablated target material has a density of said implanted ions of at least 10.sup.23 ions cm.sup.3.

5. A process according to claim 1 in which the depth of the implanted ions of said layer of implanted ablated target material is at least 200 nm.

6. A process according to claim 1 where the depth of the implanted ions of said layer of implanted ablated target material is at least 500 nm.

7. A process according to claim 1 wherein said implanted ions of said layer of implanted ablated target material are arranged in a spatial distribution which is substantially uniform along a direction extending into the heated substrate.

8. A process according to claim 1 wherein the layer of implanted ablated target material either: (i) encompasses substantially the whole area of said surface of the heated substrate; or (ii) comprises one or more zones of said surface of the heated substrate.

9. A process according to claim 1 wherein the layer of implanted ablated target material comprises one or more zones of said surface of the heated substrate wherein the zones comprise the same or different ions.

10. A process according to claim 1 wherein the implanted ions are cations.

Description

BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWING(S)

(1) FIG. 1 shows schematically the ablation, plasma production and the multi-ion implantation process.

(2) FIG. 2 shows the SEM and TEM images of the substrate cross sections with a highly defined and substantially uniformly implanted region in silica at two different target ablation energies of 47 J (Sample A) and 63 J (Sample B) respectively.

(3) FIG. 3 shows a 4001600 nm HAADF slices of individual elements of Sample B with line intensity profiles of the elements.

(4) FIG. 4 represent the Raman spectrum of ion diffused glass compared with bare silica and tellurite bulk glass.

(5) FIG. 5A represents a schematic diagram of a biosensor, such as a glucose sensor of the invention.

(6) FIG. 5B represents a schematic diagram of the biosensor of FIG. 5A.

(7) FIG. 6 shows Molar absorptivity spectra of glucose (solid), alanine (dashdot-dot), ascorbate (medium dash), lactate (short dash), urea (dotted), and triacetin (dash-dot) at 37.00.1 C. over the first overtone.

(8) FIG. 7 shows the variation in photoluminescence lifetime measured at three different wavelengths (1543 nm, 1550 nm, 1580 nm) for human blood sample with varying concentrations of glucose.

(9) FIG. 8 shows the non-invasive measurement of glucose carried out on porcine skin.

(10) FIG. 9 shows the refractive index profile of sample B.

(11) FIG. 10 shows the photoluminescence lifetime at a wavelength of 1535 nm, of sample B compared with the target glass used to prepare sample B (i.e. the ablated target glass).

(12) FIG. 11 represents a schematic diagram of a biosensor, such as a glucose sensor of the invention.

(13) FIG. 12 shows measurement results for detection of healthy and unhealthy aortic arch tissue.

(14) FIG. 13 shows the variation in implanted ion layer depth/thickness (measured from the surface of a substrate e.g. waveguide containing the implanted layer) as a function of the parameters of the pulsed laser light employed to ablate a target glass to produce an ion cloud for implantation.

ABBREVIATIONS USED

(15) HAADF high angle angular dark field elemental mapping NIR near infra red SEM Scanning electron microscopy TEM Transmission electron microscopy

DETAILED DESCRIPTION OF THE INVENTION

ExampleImplantation into Silica Glass

(16) Multi-ion implantation into silica glass 4 was produced via femtosecond laser ablation of an erbium doped tellurite glass target containing zinc and sodium. A Ti-sapphire femtosecond laser 1 operating at a wavelength of 800 nm with 100 fs pulse width and a maximum repetition rate of 1 kHz (Coherent Inc, Santa Clara, Calif., U.S.A.) was used to ablate the glass target 2 generating an expanding plasma plume 3 consisting of multiple metal ions (multi-ion). A tellurite glass target with a molar composition of 79.5TeO.sub.2: 10ZnO:10Na.sub.2O:0.5Er.sub.2O.sub.3 produces multiple ions of Te, Zn, Na and Er, which diffuse into the silica glass substrate 4 under certain process conditions. The silica glass substrate was coupled to a heater chamber arranged to heat the substrate to a desired temperature. The ablation, plasma production and the multi-ion implantation process are schematically shown in FIG. 1.

(17) Experiments were carried by varying the laser energy, repetition rate, target to substrate distance and finally the deposition target temperature. The deposition target was not translated for the simplicity of the experiment and for a better understanding of parameter variation along the sample surface. There was a variation in implantation depth and refractive index profile along the surface when radially moving outwards from the centre, therefore all the characteristic properties of the modification provided were measured from the centre of the sample unless otherwise stated.

(18) Optimum results were obtained for laser energies between 40 J-75 J when operated at 500 Hz and 1 kHz. The ablation target to substrate distance was set at 70 mm and the substrate temperature was set at 973K. FIG. 2 represents the SEM and TEM images of the substrate cross sections with a highly defined and substantially uniformly implanted region in silica at two different target ablation energies of 47 J (Sample A) and 63 J (Sample B) respectively. Diffusion depths of the ions increased from 350 nm to 850 nm with laser energy while the deposition time was 6 hours and repetition rate was 500 Hz for both cases. A well-defined boundary of the diffused and pristine region is clearly visible in the FIG. 2 and the modified region does not show any major clustering of ions or particle inhomogeneities.

(19) Further analysis of the diffusion characteristics of each ions in silica was carried out using high angle angular dark field (HAADF) elemental mapping of sample B. FIG. 3 depicts a 4001600 nm HAADF slices of individual elements. A line intensity profile shows the relative concentration profile of each diffused elements with a well-defined and sharp boundary within the silica. The oxygen concentration 6 in silica remained unchanged across the boundary while silicon 7 showed a complementary concentration profile with respect to the diffused elements. This indicates the formation of a complex glass of silica with implanted ions increasing refractive index from 1.457 of that of silica to 1.626. Considering only the cations, the atomic concentration of silicon in the diffused region is determined to be 57% while Te (curve 8), Zn (curve 9), Na (curve 10) and Er (curve 11) constitute the rest in Sample A. This confirms a single step multi-ion implantation process in the silica glass substrate.

(20) The implanted layer is highly substantially uniform and homogenous along the transverse and horizontal sections of the silica substrate. This could vary for silicon or other substrates and formations of crystallites and nano-crystallites are possible.

(21) The process for making the ion-implanted substrate of the invention generally comprises ablating a target layer with incident radiation from a laser in the presence of a substrate whereby to implant a quantity of the target layer in the substrate. The target layer when exposed to incident laser radiation produces a plasma comprising ions capable of implanting into the substrate. The substrate is heated to facilitate the implantation of ions into the substrate. The substrate is spaced apart from the target layer e.g. at a distance of about 70 mm.

(22) One or more masks or stencils may cover the surface of the substrate being implanted with the ions thereby to facilitate implantation of ions in specific zones of the substrate. These zones may form photonlc circuits such as optical waveguide structures and pathways in the substrate, or other optical components/structures.

(23) Structural Properties of Implanted Region:

(24) Silica and tellurtte are completely immiscible and will not form a stable glass under conventional batch melting and quenching process. However in the results presented above it is demonstrated that diffusion of metal cations including Te.sup.4+ ions in to the silica glass network is possible with network modification. The properties of the implanted silica glass were measured. No signals of any kind of crystallization were observed in electron diffraction and XRD characterization proving a complete amorphous phase of silica-tellurite glass. Raman spectroscopy was used to analyse the glass network in the diffused region. FIG. 4 represent the Raman spectrum of ion diffused glass compared with bare silica and tellurite bulk glass which is devoid of any characteristic signals corresponding to TeO.sub.2 glass phase within the network.

(25) FIGS. 5A and 5B represents a schematic diagram of a biosensor 8 of the invention for use as a glucose sensor. The biosensor comprises a pump laser 9, an ion-implanted substrate of glass 10, and a light detector 11 for detecting optical signals 12 emanating from the ion-implanted substrate. The pump laser is arranged to emit laser light 13 having a suitable wavelength for causing implanted ions 14 within the substrate to fluoresce. The light detector is a fast photodiode, such as a microsecond photodiode, or a nanosecond photodiode and is arranged to detect the intensity of fluorescence of the implanted ions within the substrate. That is to say, the photodiode preferably has a response time whereby an electrical detection signal is generated within microseconds, or more preferably nanoseconds in response to the presence of light at the detector, and that such electrical signal stops within the same or similar rapid time period after light is no longer present at the detector. This speed is preferable in order to allow the apparatus to have a desired degree of temporal resolution permitting separate, brief fluorescence detections in rapid succession. A similarly rapid photomultiplier device may be used as an alternative.

(26) The substrate has a substantially uniform thickness which is in the range of about 0.1 mm to about 10 mm, such as in the range of 0.5 mm to about 3 mm. The ion implanted substrate defines at its implanted surface region 14 a functional surface for the optical detection of the presence of glucose via photons of fluorescent light 15 generated in the implanted region of the substrate. It may be referred to as a photonic chip. Indeed, a substrate of the invention may be formed directly upon the surface of a laser such as a solid-state laser whereby the functional surface of the substrate is pumped by the solid-state laser in use. This offers a very compact photonic chip design.

(27) The biosensor includes not only the light detector but also a data processor unit 16 connected to the light detector for receiving light detection signals from the light detector and for generating data representing those detections. The blosensor further includes a control unit 17 comprising software, or firmware or electronic circuitry, arranged for controlling the operation of the pump laser and/or the light detector and/or the data processor unit as desired. The combination of the light detector and the data processor unit may be referred to as a photonic chip integrator, in its relationship to the photonic chip. The control unit may output detected signals, received from the signal processor unit, to display at a display of human interface of the control unit 17. Optical waveguides (18) guide pump light from the pump laser 9 to the substrate 10, and guide fluorescent light from the substrate to the photodiode via a diffraction grating 11A arranged to separate the received fluorescent light into desired spectral wavelength bands for output to the optical detector on a dedicated spectral channel 11B (e.g. a waveguide). This allows the photodiode to detect selected spectral channels independently; e.g. by blocking receipt of light of all but the channel(s) it is desired to detect.

(28) The ion-implanted layer of the substrate is formed to extend into the body of the substrate from a planar surface/face of the substrate to a substantially uniform depth of at least 50 nm (e.g. at least 200 nm) into the glass and has a substantially uniform distribution of implanted ions throughout that depth. The implanted layer also extends across the planar surface to provide a functional surface area of sensitivity for glucose detection. The surface are may be several square mm, or square cm, in size. The distribution of implanted ions throughout that area is also substantially uniform. The pump laser and light detector are both optically coupled to a rear surface area of the substrate which is the reverse surface to functional surface area. This means that the functional surface area may be applied or located at a desired object or target of study (e.g. the skin or an animal, 19), while the pumping of the functional area and the subsequent detection of light emanating from the functional surface area may take place away from the target without requiring the functional surface to be moved relative to the target. This permits compactness in the biosensor, and ease of use.

(29) This use of ion-implanted substrates of the invention permits non-invasive detection of metabolites, such as glucose, in animals. The detection method is a novel method which measures a photoluminescence lifetime to detect metabolites such as glucose, rather than simply detecting from an isolated light intensity measurement.

(30) The photoluminescence spectral band of the dopant(s) within the ion-implanted substrate overlaps with the characteristic absorption bands (e.g. in the range 1530-2200 nm) of glucose molecules in the near-infrared (NIR) wavelengths. An example of these absorption bands is illustrated in FIG. 6 for Glucose and for other molecules (see Airat K. Amerov et al., Appl. Spectrosc. 58 (10), 1195 (2004)).

(31) In particular, FIG. 6 shows the Molar absorptivity spectra of glucose (solid line), alanine (dashdot-dot line), ascorbate (medium dash line), lactate (short dash line), urea (dotted line), and triacetin (dash-dot line) at 37.00.1 C. over the first overtone. Thus, while the present embodiment is described in terms of glucose detection, it will be readily appreciated that it may be applied to detection of other molecules such as those described above.

(32) By measuring/detecting fluorescence lifetimes at selected wavelengths of light falling within regions of the absorption spectrum of a metabolite in which significant spectral variations occur, such as sharp, or sustained changes in absorption levels, one may identify the presence of that metabolite in terms of the corresponding changes in the measured fluorescence lifetimes at different optical wavelengths within that regions of the absorption spectrum of a metabolite. For example, Glucose has a strong spectral slope in the wavelength region of about 1525 nm to about 1650 nm as shown in FIG. 6, whereas the other metabolites represented in FIG. 6 have far less spectral variation there. This means that detected spectral variations in detected fluorescence lifetimes within that spectral region and can be used to infer or confirm the presence of Glucose. This applies similarly to other metabolites at other spectral regions.

(33) FIG. 7 and FIG. 8 show detected changes in fluorescence lifetimes when the biosensor is applied (in the manner shown in FIG. 5 or FIG. 11) to a blood sample containing Glucose (FIG. 7) or a sample of porcine skin (FIG. 8). It can be seen that at longer wavelengths (1580 nm) the change in fluorescence lifetime is the greatest and can be more sustained in its correlation with Glucose concentrations (see FIG. 8). In this regard, the biosensor may be arranged to operate at one particular wavelength (e.g. 1580 nm) selected from amongst a plurality of wavelengths, which provides such a favourable response or sustained correlation.

(34) The measured photoluminescence lifetime of the rare earth implanted ions becomes modified within the glass thin film contained in a photonic sensor due to the amplification by random scattering and localization of photoluminescent photons. When a target medium containing glucose (e.g. skin) interacts with fluorescent photons from the ion-implanted film, via the ion-implanted functional surface, the apparent photoluminescence lifetime becomes modified as a function of glucose concentration within the target medium due to its specific absorption as well as molecular scattering properties.

(35) Thus by an accurate measurement of the photoluminescence lifetime of the rare earth implanted ions in the functional surface, the concentration of glucose in the medium can be deduced. Since the absorption and scattering properties of the photons at different wavelengths within the emission band varies as a function of glucose concentration, the apparent change in the photoluminescence lifetime of the rare earth implanted ions at different wavelengths can be used as an additional feature to enhance the signal due to glucose as compared to that from other interferences (e.g. metabolites) as described above with reference to FIG. 6. This new measurement concept may be referred to as the Spectrally Resolved Photoluminescence Lifetime (SRPL) technique and is the novel principle of detection that may be applied in the biosensor.

(36) The method for the non-invasive measurement of a metabolite using the biosensor includes: applying the biosensor on or near the subject being sensed, such as applying the functional surface of the biosensor to the skin of an animal; irradiating the functional surface with pump light from the pump laser for a pumping period of time to excite fluorescence in the functional surface, and such that a portion of the fluorescent light escapes into the animal; measuring the photoluminescence lifetime of fluorescent light from the functional surface, after the pumping period has ended, using the photodiode and the signal processor unit.

(37) The signal processor unit is arranged to determine the time period (the recovery lifetime) required for the measured intensity of fluorescence light to fall in value by a factor of 1/e. This period preferably begins immediately from the ending of the pumping period such that no pumping of the functional surface takes place during the lifetime measurement period. The measured recovery lifetime of detected fluorescent light from the functional surface is correlated with the level of the metabolite within the subject being sensed as described above. The signal processor is arranged to calculate a value of the metabotite (e.g. Glucose) concentration level using such pre-determined correlations. These pre-determined correlations for a given metabolite such as Glucose (i.e. concentration levels corresponding to measured recovery lifetime) may be pre-stored in a look-up table within the signal processor unit (or in a separate memory accessible by the signal processor unit) from which the signal processor unit determines a concentration level from a given measured recovery lifetime and outputs the result either for display to a user or for storage as desired.

(38) The control unit 17 is arranged to control the pump laser to resume pumping the functional surface with pump light after a defined period of time exceeding the recovery lifetime. This controlled on/off switching of the pump laser may occur at a repetition rate to suit the user. The on period permits the pump laser to pump the implanted ions of the functional surface, while a successive off period permits a recovery lifetime to be measured, and a metabolite concentration to be determined. FIG. 10 illustrates the decay in detected fluorescence light levels in the functional surface when isolated from any measurement subject. The exponential decay falls by a factor of 1/e after 11.1 ms which corresponds to the recovery lifetime of the functional surface in isolation. This recovery lifetime is changed in the presence of a metabolite such as Glucose, and the amount of change is correlated to the amount of glucose. FIG. 7 shows an example correlation when the subject being measured is a free blood sample containing Glucose (in vitro) as shown in FIG. 11, whereas the correlation shown in FIG. 8 is obtained from a porcine skin sample as shown in FIG. 5. FIG. 10 also shows a decay curve for the target glass used to generate the plasma cloud (FIG. 1) during ion implantation of the substrate.

(39) FIG. 12 illustrates the ability of the biosensor to distinguish between normal aortic tissue and abnormal aortic tissue, such as atherosclerotic tissue. This illustrates the use of the biosensor for sensing cardio-vascular disease and cancers. The change in fluorescence lifetime is shown in FIG. 12 for five different circumstances. The leftmost data point 20 in the figure corresponds to a circumstance where the biosensor is not in contact with any subject tissue. The change in fluorescence lifetime is zero. The next data point 21 corresponds to healthy aortic tissue and shows a positive change in fluorescence lifetime, whereas a negative change 22 in fluorescence lifetime is seen for atherosclerotic tissue. This illustrates a measurement method for determining the presence of healthy or unhealthy tissues as follows.

(40) First, the fluorescence lifetime of the functional surface of the biosensor is determined in the absence of tissue; next, the tissue is applied to the functional surface of the biosensor and a new fluorescence lifetime is determined in the presence of the tissue (e.g. in vitro as in FIG. 11). If the latter fluorescence lifetime is greater than the former, then the tissue is healthy, else it is unhealthy if the latter fluorescence lifetime is less than the former. The signal processor of the biosensor may be arranged to make this determination.

(41) In a further enhancement of this method, Erbium (e.g. in a salt solution: e.g. Erbium Chloride) may be added to the tissue, and a second fluorescence lifetime measured. If the second fluorescence lifetime 23 is less than the first-measured fluorescence lifetime (for the Erbium-free tissue) then this is indicative of healthy tissue, otherwise if the second fluorescence lifetime 24 is greater than (i.e. less negative) the first-measured fluorescence lifetime (for the Erbium-free tissue) then this is indicative of unhealthy tissue. The signal processor of the biosensor may be arranged to make this determination.

(42) The substrate of the biosensor, and in other aspects of the invention, is provided with an ion-implanted layer in which the ion implanted layer extends substantially from the outermost surface of the substrate. This is shown in FIG. 3 which shows the concentration of dopant ions, and of substrate glass as a function of depth into the substrate from the outermost surface of the glass through which implantation took place. The left-hand part of FIG. 3 shows a HAADF slices as described above indicating concentrations of dopants and other atoms of the substrate. It is to be noted that the line trace graphs of the HAADF slices shown to the right of FIG. 3 begins just above the outermost surface of the glass, which surface is indicated by the initial sharp rise in HAADF signal intensities at a distance on the graph of a few nm. The ion-implanted layer has a substantially uniform distribution of the implanted ions extending into the substrate from its outermost surface.

(43) This ion distribution produces a region of substantially uniformly increased refractive index in the substrate as is indicated by the approximate (simulated) effective refractive index profile 25 shown in FIG. 9. This profile illustrates an approximate representation of the change in refractive index of the substrate with increasing depth from the surface of the ion-implanted layer. The continuous curve represents a simulation (approximate) of the effective refractive index distribution based on the measured optical properties (e.g. refractive indices) of three different optical propagation modes (26, 27, 28) measured within the substrate, and calculated according to standard techniques for determining effective refractive indices using optical propagation modes as would be readily apparent to the skilled person. The result is a substantially uniform refractive index of about 1.625 (measured at a wavelength of 633 nm, using a He Ne laser) across the implanted layer, falling as rapidly to the refractive index of the substrate glass at a depth of about 1.4 microns where the implanted layer ends. The refractive index distribution is a step index distribution and results from the uniformity of the ion distribution within the substrate. A more accurate simulation would show a sharper step change in refractive index at a depth of 1.4 microns. Note that the refractive index of the substrate material was about 1.457 whereas the refractive index of the implanted layer is 1.625a very significant increase.

(44) FIG. 13 shows the variation in implanted ion layer depth/thickness measured from the surface of a substrate (e.g. waveguide) containing the implanted layer, as a function of the parameters of the pulsed laser light employed to ablate a target glass to produce and ion cloud for implantation. It can be seen that the depth (layer thickness) as well as the refractive index of the layer, increase with pulse repetition rate and laser pulse energy in the ablating laser. Increasing the layer deposition time also increases the layer thickness/depth.

(45) The left graph of FIG. 13 shows the implanted layer depth and refractive index (the four-digit numbers adjacent to data points) as a function of increasing pulse repetition rate. The deposition time was 30 minutes at 51 microjoules per pulse of ablating laser light. The middle graph of FIG. 13 shows the implanted layer depth and refractive index (the four-digit numbers adjacent to data points) as a function of increasing pulse energy. The deposition time was 6 hours at a pulse repetition rate of 500 Hz. The right graph of FIG. 13 shows the implanted layer as a function of increasing deposition time. The pulse repetition rate was 500 Hz at 51 microjoules per pulse of ablating laser light.