Method for Radiation Detection and Measurement

When solid-state crystals are subjected to ionizing radiation, several absorption bands make their appearance at increasingly higher levels of radiation damage.sup.(1). In the case of the alkali halides, the F-band is the first of the radiation damage centers to produce a detectable absorption band. Although the F-center provides the greatest concentration of damage centers for a given radiation exposure, its physical properties preclude its use in quantifying ionizing radiation exposure using luminescence techniques. The luminescence of the F-center is thermally quenched below room temperature.

With increasing radiation exposure, a second damage center builds up in the absorption spectrum. The second damage center is known as the M-center and is generally thought to consist of two adjacent F-centers.sup.(2,3). Unlike the F-center, however, excitation into the longer wavelength M-center absorption band at room temperature produces luminescence.sup.(4). M-center luminescence involves a Stokes' shift, allowing the luminescence to be observed at a significantly different wavelength from the exciting wavelength.

Highly purified lithium fluoride (LiF) crystals have long been used as optical windows. Optical grade LiF is known for its excellent transmission mission from the deep ultraviolet through the infra-red. Radiation dosimetry applications have so far been restricted to high-level Mega-Rad gamma dosimetry using radiation-induced absorption peaks measured with a spectrophotometer. However, absorption measurements are a very insensitive way to measure these radiation damage centers.

A more sensitive method of measuring radiation damage than absorption measurement can be accomplished by measuring the M-center luminescence. Excitation of an LiF crystal with a 442 nm He-Cd laser has been found to significantly amplify the luminescence yield of the M-center of LiF. An He-Cd laser was the preferred excitation laser because absorption measurements in LiF indicate the peak of the M-center absorption occurs at 443 nm. Laser stimulation produces an excited state of the M-center, which undergoes a very strong Stokes' shift. The peak of the M-center emission spectrum occurs at 665 nm with a half-width of 0.36 ev. Since the excitation wavelength differs significantly from the emission wavelength, measurement of the deep red emission can be done simultaneously with the excitation. Therefore, optical filtration provides a means of measuring the M-center luminescence simultaneously with laser excitation into the M-center absorption band. The population of M-centers grows with increasing radiation damage, and therefore M-center luminescence provides a basis for radiation dosimetry. Readout of the LiF can be accomplished in a fraction of a second.

While all members of the alkali halide family are candidates for M-center luminescence dosimetry, lithium fluoride (LiF) was chosen for a number of reasons. First, the crystal is commercially available in large quantities through its use as optical window material. Secondly, LiF is nearly tissue equivalent in its energy response to photons, making it a suitable material for mixed field dosimetry applications. Finally, LiF can be excited by a low cost, commercially available blue-light laser and its M-center luminescence occurs significantly within the visible spectrum. This feature simplifies light detection since phototubes that are sensitive to the visible light spectrum are readily available. Data describing the dosimetric properties of the M-center and other general physical data concerning the M-center are presented in the detailed description, which also discusses the possibilities for M-center dosimetry in LiF.

M-center luminescence exists in almost every solid-state crystal which can be artificially grown at the present time and is not restricted to the alkali halide group. Each crystal has a characteristic M-center absorption band which can be measured following a large radiation exposure (around 1 Mega R). Illumination into the M-center absorption band with laser light stimulates M-center luminescence which occurs at a significantly different wavelength from the exciting laser light beam. Easy separation and detection is possible simultaneously with the exciting light by using an optical filter which absorbs the exciting light and transmits the M-center luminescence.

FIG. 1 is a diagram of the M-center luminescence readout system. The He-Cd laser produces 70 mW at 442 nm with a beam diameter of 1 mm. The reader consists of a light tight metal box with a sample drawer for insertion and removal of the LiF crystals. The chamber surfaces are made of a black plastic to reject and shield stray light. An RCA 8852 phototube with a broadband optical interference filter is used to reject the laser light and transmit the M-center luminescence. A personal computer houses a multichannel scaling board for photon counting capable of 50 MHz.

FIG. 2 shows an LiF M-center luminescence optical emission spectrum measured using a Tracor Northern optical multichannel analyzer. The M-center luminescence emission was excited by a 442 nm He-Cd laser. The LiF crystal was given an exposure of 2.6 C/kg using a .sup.60 Co source. The peak of the emission spectrum is 665 nm with a half-width of 0.36 ev.

FIG. 3 shows an optical absorption spectrum for a LiF crystal exposed to 260 C/kg. The optical path length of the LiF crystal was 6 mm. The optical absorption spectrum was taken using a Hewlett-Packard diode array spectrometer. The peak of the absorption spectrum was measured to be 443 nm.

FIG. 4 shows the optical bleaching rate of the M-center luminescence as a function of 442 nm He-Cd laser energy incident upon the crystal. The crystal used in this figure was exposed to 26 mC/kg. The time constant for the optical bleaching was found to be 32 s. The total reduction in M-center luminescent intensity was 20% for a delivered energy of 30J.

FIG. 5 is a schematic diagram of a personnel dosimeter based on an LiF crystal.

Optical grade LiF single crystals (6 mm.times.6 mm .times.6 mm) were selected for the M-center measurements. Since an appreciable population of M-centers are present in commercially available crystals, the centers are erased by heat treatment. A heat treatment in air at 873K for 1 h was found satisfactory in eliminating the M-center background. After heat treatment, some of the crystals were kept as controls, while the remainder were exposed to 0.26, 2.6, 26, and 260 mC/kg from a .sup.60 Co source. After irradiation, the crystals were stored in paper envelopes to prevent unwanted light exposure

As shown in FIG. 1, the readout apparatus consisted of three distinct functional units: a 70 mW He-Cd 442 nm laser 2, a light-tight box 4 that contained the sample chamber 6 and phototube 8, and a photon counting system 10. The He-Cd laser beam was focused to beam diameter of 1 mm at the crystal face and transmitted through the crystal at the centers of two opposite cube faces. To provide the laser excitation light to the crystals, the laser beam path and the sample chamber were made of black plastic which helped shield stray room light and laser scatter. A broadband interference filter 12 centered at 650 nm was chosen to prevent the 442 nm He-Cd laser light from reaching the phototube 8 and to transmit the M-center luminescence. The filter provided a 650 nm peak transmission of 50% and a bandwidth of 70 nm. Transmission of the scattered 442 nm He-Cd laser light was reduced to 0.01% by the filter 12. A 1 mm aperture was placed between the optical filter and the crystal to help reject some of the background light that accompanies the M-center luminescence. The phototube 8 chosen for the experiment was an RCA 8852 with a red-extended multi-alkali photocathode and was mounted at 90.degree. from the axis of the laser beam. This phototube was chosen for its suitability for photon counting and its excellent red response. The photon counting system consisted of a preamplifier, amplifier, discriminator, and a multichannel scaling board housed in a personal computer. The counting system is capable of 50 MHz rates.

Readout consisted of passing the 70 mW laser beam through the sample chamber and into the LiF crystal while simultaneously counting photons with the multichannel scaling board within the computer. A region of interest was chosen which consisted of 400 channels with each channel having a time width of 50 ms. An integration was performed over the 400-channel region of interest which lasted for 20 s.

FIG. 2 is a LiF M-center luminescence optical emission spectrum measured using a Tracor Northern optical multichannel analyzer. The LiF crystal was given an exposure of 2.6 C/kg using a .sup.60 Co gamma source. The peak of the emission spectrum is 665 nm with a half width of 0.36 ev. This result was used to determine the optimal interference filter to use in the M-center luminescence reader. An optical absorption spectrum for a heavily exposed LiF crystal was measured using a Hewlett-Packard diode array spectrometer. This absorption spectrum is shown in FIG. 3. The exposure level for the absorption spectrum was 260 C/kg. To resolve the M-center absorption curve adequately, it was found that 260 C/kg was the minimum exposure that could be used The absorption curve peaks at 443 nm, which matches the 442 nm line of the He-Cd laser.

The following table shows the M-center response as a function of .sup.60 Co gamma exposure of 0.26, 2.6, 26 and 260 mC/kg.

Two laser power levels were used due to counting saturation of the 260 mC/kg exposure using the full 70 mW of laser power. Within experimental uncertainty, the response follows a linear relationship. The crystals exposed to 0.26 mC/kg gave an integral response that was twice as high as a crystal that received no exposure.

FIG. 4 is a plot of the bleaching rate of the M-center luminescence as a function of time. The crystal was exposed to 26 mC/kg and the laser power was 70 mW delivered for 600 s. The time constant for the bleaching process was found to be 32 s and appears to remain constant over time. For the entire time in which the M-center luminescence was monitored, the total reduction due to bleaching was only 20%.

The time constant associated with the decay of the excited M-center was measured in a straightforward manner. The same readout apparatus that was used for the previous measurements was used in the time-constant measurements. A nitrogen-pumped dye laser was used to excite an irradiated LiF crystal within the readout apparatus, the signal was fed into a Hewlett-Packard digitizing fast storage oscilloscope. The oscilloscope was allowed to average the M-center decay signal until a smooth set of data was obtained. The time constant measured was 70 ns. This appears to be consistent with measurements of the other authors .sup.(5,6).

The M-center luminescence in LiF occurs with a time constant on the order of 70 ns. Based on this measurements, each M-center in LiF should be capable of producing around 10.sup.7 photons per second at saturation. A simple calculation based on the oscillator strength of the M-center in LiF predicts the M-center luminescence saturates with a 442 nm laser beam at 40 kw. Since the laser used in the present experiment is 70 mW, large gains in the M-center luminescence signal can be achieved by increasing the excitation power, for example, through use of a high intensity argon laser.

Phototubes that record the M-center luminescence must be extremely sensitive to the red-infra-red emission spectrum if a sensitive dosimeter capable of personnel monitoring is to be realized. Unfortunately, red extended phototubes capable of single photon counting have characteristically large dark counts associated with thermionic emission. A commercially available dye laser provides an elegant solution to the problem of high dark counts and also satisfies the need for large peak powers to achieve saturation of the M-center luminescence. Since dye lasers typically have extremely short pulse lengths, the red extended phototube dark count within the duration of the dye laser pulse becomes negligible. The flashlamp-pumped dye laser can be made to produce very large peak powers with time constants of 10 to 100 .mu.s, orders of magnitude longer than the time constant for the decay of excited M-centers. Therefore, through the use of the flashlamp-pumped dye laser, the M-center luminescence is maximized by producing saturation, while allowing a sampling of the luminescence in a time period that is much longer than the decay constant of the M-center. This condition will provide better statistical results. Fortunately, the M-center luminescence yield at saturation is theoretically optimized with the same laser that produces a negligible dark count.

The most restricting aspect of the M-center luminescence technique in LiF is the large fluorescence background. The unwanted background, stimulated by the direct and scattered laser light, emanates from a number of sources. The first such source is the fluorescence that results within the chamber surfaces from laser light scatter. A careful chamber design both in geometry and in material selection can greatly reduce this source of unwanted light. A second source is the fluorescence that emanates from the crystal surfaces. The surface fluorescence is clearly visible by the human eye under the proper conditions and represents the largest source of unwanted light. The surface fluorescence depends on the nature of the fluorescing species at the surface. The third source of unwanted light is from the bulk of the crystal. Impurities within the crystal that are excited by the laser beam will produce light that interferes with the desired signal. This source of light has not been observed at the present time, due to the large fluorescence signals that mask the bulk crystal effects. The fourth source of background light is from the leakage of the optical filter used both to condition the laser and to block stray laser light from reaching the phototube. Since the phototube is highly sensitive to the exciting laser wavelength, proper filtration is essential. High quality filters and the use of multiple filters may prevent this source of background from causing any serious problems. The final source of background light is perhaps the most troubling.

In the dosimeter application, it is desirable that the dosimeter be reusable, and this implies the need to anneal the crystal so as to remove the M-centers. As noted previously, thermal treatment can be used to eliminate the M-centers. However, at some point of the heat treatment, the M-center will reach an equilibrium level, at which point further reduction in M-center concentration becomes difficult. It remains to be seen whether this concentration level will be suitably low to perform adequate personnel radiation monitoring.

Based on the foregoing, it will be appreciated that M-center luminescence in LiF using a .sup.60 Co gamma source has been demonstrated to be a useful dosimeter in the exposure range of 0.026 to 260 mC/kg. The major difficulty in using LiF for M-center luminescence dosimetry is the large background of fluorescence which accompanies the M-center luminescence signal. Low cost diode lasers emitting in the infra-red are currently available which could be used to excite those crystals which have their M-center absorption band in the infra-red.

Many alkali halides have their M-center luminescence absorption band in the infra-red. An inexpensive infra-red detector can be used to detect the infra-red luminescence allowing a low cost laser and detection system to be technically feasible. In this way, the M-center luminescence can be used as a means of personnel or environmental dosimetry. By coupling a small infra-red diode laser with an infra-red detector and suitable optical filters, the combination becomes a radiation monitor which gives a real-time analysis. FIG. 5 illustrates such a dosimeter, comprising a crystal 20 to which are mounted a diode laser 22, a detector 24 to detect the crystal's M-center luminescence, and an optical filter 26 to block the laser light and transmit the crystal's M-center luminescence.

Optically stimulated luminescence is described in co-pending U.S. Pat. applications Ser. Nos. 07/213,245, now U.S. Pat. No. 4,954,707, and 07/420,293 the contents of which are hereby incorporated by reference herein.

It will be appreciated that the invention is not restricted to the particular embodiment that has been described, and that variations may be made therein without departing from the scope of the invention as defined in the appended claims and equivalents thereof.

1. McLaughlin et al, "Electron and gamma-ray dosimetry using radiation-induced color centers in LiF", Radiat. Phys. Chem. 14, 467-480 (1979).

2. Seitz, "Color centers in alkali halide crystals", Rev. Mod. Physics 18, 348 (1946).

3. Knox, "Inversion symmetry of the M-center", Phys. Rev. Letters 2(3), 87 (1959).

4. Klick, "Luminescence of color centers in alkali halides", Phys. Rev. 79, 894 (1950).

5. Bosi et al, "Lifetime studies on excited (F.sub.2 +)* and M centers in NaF doped with magnesium", Phys. Stat. Sol. (b) 140, 355-360 (1987).

6. Bosi et al, "New results on the decay properties of perturbed and unperturbed M-centers in NaCl:CdCl.sub.2 ", Phys Stat. Sol. (b) 123, 519-524 (1984).

The disclosure of each reference is hereby incorporated by reference herein .