WO2006084313A1

WO2006084313A1 - Detection of nuclear quadrupole resonance signals in substances

Info

Publication number: WO2006084313A1
Application number: PCT/AU2006/000156
Authority: WO
Inventors: Taras Nikolaevitch Rudakov; Vassili Timofeevitch Mikhaltsevitch; Peter Alaric Hayes; Warrick Paul Chisholm; John Harold Flexman
Original assignee: Qrsciences Pty Ltd
Priority date: 2005-02-08
Filing date: 2006-02-08
Publication date: 2006-08-17

Abstract

A system for detecting nuclear quadrupole resonance (NQR) in a substance includes a DC magnet (10), a DC magnet control unit (20), a probe (30) for irradiating a volume with RF electromagnetic energy and receiving response signals from a sample disposed within the volume, a control and signal processing unit (70), which with a transmitter unit (60), receiver unit (50) and matching circuit (40) applies excitation to the probe (30) in the form of pulse sequences and processes response signals to detect NQR signals. These signals correspond to a substance containing quadrupolar nuclei and nuclei with a magnetic moment being targeted within the sample. The DC magnet control unit (20) operates the DC magnet (10) to alternately permeate the volume with, or adiabatically remove, a DC magnetic field to provide cross-relaxation and/or cross-polarization effects for improving the speed of NQR detection. A method for providing such cross-relaxation and/or cross-polarization effects is also described.

Description

DETECTION OF ISfUCLEAR QUADRUPOLE RESONANCE SIGNALS IN SUBSTANCES

Field of the Invention!

5 This invention relates to the detection of nuclear quadrupole resonance (NQR) signals in substances including specific substances, such as explosives and narcotics.

Throughout the specification, unless the context requires otherwise, the word "comprise" or variations such as "comprises" or "comprising", will be understood to 0 imply the inclusion of a stated integer or group of integers but not the exclusion of any other integer or group of integers.

Background Art

The following discussion of the background art is intended to facilitate an understanding of the present invention only. It should be appreciated that the 5 discussion is not an acknowledgement or admission that any of the material referred to is or was part of the common general knowledge as at the priority date of the application.

Nuclear quadrupole resonance (NQR) is one of many modem research methods in physics used for the analytical detection of chemical substances in solid form. 0 NQR is a radio frequency (RF) spectroscopy, and it is defined as a phenomenon of resonance associated with RF absorption or emission of electromagnetic energy. It is due to the dependence of a portion of the energy of electron-nuclear interactions on the mutual orientations of asymmetrically distributed charges of the atomic nucleus and the atomic shell electrons as well as those charges that 5 are outside the atomic radius. Thus, the differences observed in the quadrupole coupling constants and NQR frequencies between substances are due to electric effects. The nuclear electric quadrupole moment eQ interacts with the electric field . gradient eq , defined by asymmetry parameter η. Therefore the nuclear quadrupole coupling constant e^zQq and the asymmetry parameter η, which contains structural information about a molecule, may be calculated from the experimental data.

The main spectral parameters observed in typical NQR experiments are the transition frequencies of the nuclei and the line widths άf . Besides those parameters, obtaining the spin-lattice relaxation times T₁ , spin-spin relaxation times T₂ and line-shape parameters Tl (inversely proportional to Δf ) are also of great value. These parameters must also be taken into consideration when choosing the experimental technique and equipment for detecting a chemical substance using NQR.

In contrast to nuclear magnetic resonance (NMR) methods, NQR can be performed without a strong external DC magnetic field. This technique is known as "pure NQR", or direct NQR detection, and has many advantages over other techniques for some applications, such as identification of specific compounds and remote NQR detection. For example these methods are successfully used for detecting the presence of specific substances, such as explosives and narcotics, in airport luggage. Explosives can also be detected in landmines.

The probe of a pulsed NQR detection system is a device providing interaction between the radio frequency (RF) field of a resonant RF transmitter and a substance being targetted, as well as the RF field response from the targetted substance and a receiving part of the NQR detector- Strong RF pulses, typically with the power of hundreds of watts are used. In practical NQR devices, when detecting specific substances (for example explosives and narcotics), the RF pulse power can reach several kilowatts.

FIG. 1 illustrates a conventional system for detecting NQR signals. A transmitter unit 60 and a receiver unit 50 are connected to a probe 30 through a duplexer and matching circuit 40, which switches probe 30 between the transmit and the receive mode. The transmitter unit 60 generates RF pulses and applies the pulses to the probe 30 to excite a substance being targetted. The pulses have a frequency corresponding to the resonance frequency of the quadrupolar nuclei of the substance- After the RF pulse is applied, the probe 30 will detect any NQR signal emitted from the substance. This signal is received by the receiver unit 50 and processed by a control and signal processing unit 70, which also generates all control and RF signals for the transmitter unit and the receiver unit.

The NQR method is an ideal method for the detection and identification of such specific substances as explosive materials and narcotics. Most of these substances contain nitrogen-14 nuclei, the spectral lines of which are usually located at low frequencies where NQR signals detected have low intensity. In order to improve sensitivity, a multi-pulse technique is normally used. In fact this technique can be divided into two main groups of multi-pulse sequences.

The first group is based on the effect of the Steady-State Free Precession (SSFP) discovered in 1951. These kinds of sequences have been used to increase sensitivity in NMR and NQR measurements. The simplest version of the SSFP pulse sequence is a long train of phase coherent RF pulses and can be represented as:

(τ-θ^ -τ-),, ,

where θ¹ is the flipping angle of the other pulses in the sequence, and τ is the time interval between the pulses.

It is a necessary condition for an SSFP pulse sequence that the RF pulse repetition time 2τ must be less than or equal to the spin-spin relaxation time T₂. Under this condition a relatively large and continuous steady-state signal is produced and it will recur for as long as the sequence is applied.

The second group of sequences is based on the pulsed spin-locking effect. These sequences cause a refocussing of transverse magnetization for periods much longer than the spin-rspin relaxation time T₂. The basic spin-locking multi-pulse (SLMP) sequence, which was proposed in NMR in 1966, can be represented as:

^• C -(τ-θ^ -τ-)_N ,

where θ°is the flipping angle of a preparatory pulse and θ¹ is the flipping angle of the other pulses in the sequence, phase shifted by 90° in relation to the preparatory pulse.

The first experimental investigation of this type of sequence in pure NQR was made in 1977. The results obtained at that time were very similar to those obtained earlier in the NMR experiments.

Besides SSFP and SLMP sequences, a mufti-pulse technique combining the properties of both of these groups of sequences can also be used. In most cases this technique gives better results than either the SSFP or SLMP sequences alone.

If a multi-pulse technique is used, the highest sensitivity is attained when the spacing of the RF pulses is adjusted to be short in comparison with the spin-spin relaxation times T₂, and under such conditions a steady-state response is established. Unfortunately under these conditions, however, the signal amplitude strongly depends on the frequency offset Δf and the pulse repetition time τ. This effect is well known in NMR as intensity variations, and has been associated with finite macroscopic transverse components of magnetization at the end of the pulse interval.

Intensity variations can also occur in the case of NQR spectroscopy. The effect of strong intensity variations occurs for both SSFP and SLMP sequences. However, intensity variations are not a large problem for NQR detection systems if the NQR frequency of the substance being detected is known.

However, temperature effects can also degrade the performance of NQR detection systems where the exact temperature of the sample is not known, for example, such as scanning of passenger luggage. Due to temperature effects, an actual NQR frequency of a quadrupolar nucleus may be offset from the expected value. The result of this is that the frequencies of the transmitted RF pulses are far from the actual resonant frequency and the intensity variations can degrade the sensitivity of the NQR detection systems.

Fortunately the methods for suppressing intensity variations that are extensively used in magnetic resonance are also applicable in NQR. The problem is generally solved by using one or more additional pulse sequences (or blocks of sequences) after the first pulse sequence (or block of sequences), where some parameters are different between the two. Such pulse parameters that are different could be a phase, frequency or pulse repetition time.

In order to achieve a reasonably good degree of sensitivity when using these techniques, a waiting time is required between repetition of the sequences or blocks. The waiting time required is a minimum of about 0.5 times the spin-lattice relaxation time T₁, and is ideally about 3-5 times T₁. Consequently, for most applications, it is not time effective to use the technique for the detection of substances that have a spin-lattice relaxation time Ti greater than about a few seconds, and this is especially true for substances such as TNT, PETN and

Ammonium Nitrate (AN).

For several substances the effect of temperature is quite large and the frequency of the transmitted RF pulses may be too far from the actual resonant frequency to produce a resonance signal. As a result, a typical NQR detection system will not detect a target substance that is actually present in the object being scanned. In order to avoid this problem, additional scanning with a frequency shift is required. In this case a waiting time of about 3-5 times T₁ between the scans will also be required, which exacerbates the problem when trying to detect substances with a long spin-lattice relaxation time T-i.

Usually detected NQR signals have low intensity and therefore the presence of noise sources also presents a serious problem, particularly for the detection of certain substances. In practical situations, certain objects within a volume being - B - investigated for the presence of a particular substance responsive to the NQR phenomenon, can become sources of coherent noise (or spurious signals) when irradiated with strong radio frequency (RF) pulses. These objects generate strong magneto-acoustic or piezo-βlectric ringing signals. As these signals are coherent with RF pulses, special multi-pulse techniques can considerably reduce them.

One technique that has been suggested for suppressing magneto-acoustic ringing . is to use two Steady-State Free Precession (SSFP) pulse sequences to eliminate coherent noise:

where θ is a flip angle determined by the pulse length and the value of the B₁ field of the RF pulse, and τ is a pulse repetition time.

The first sequence (also known in NQR as a Strong Off Resonant Comb (SORC)) is referred to as a Non-Pulse-A1temated Pulse Sequence (NPAPS), and the second as a Phase Alternated Putse Sequence (PAPS). It is claimed that if the receiver phase in NPAPS is set as constant and equals 180°, and in PAPS it is alternated with each pulse (0° - 1B0°), this combination ensures the cancelling of coherent noise of up to 20 dB.

Others have suggested two NPAPS - PAPS sequences which are supplemented by two more identical sequences with the corresponding phase settings of 90° and 270°. This allows the achievement of a more complete symmetry in the constructed combination of sequences, and as a result, better elimination of coherent noise can be obtained.

There are also other variations of various SSFP pulse sequence combinations, either with phase alternation or with phase cycling, all of which are effective. All of these sequences have good symmetry and achieve a comparable level of coherent noise elimination. It should be noted, however, that none of the above techniques work well if the substance under test has a long spin-lattice relaxation time T₁. The main reason for this is that all of these techniques use more than one pulse sequence, which required a waiting time of about 3-5 times Ti between the sequences. Consequently, in cases where lines of luggage are being scanned for the presence of explosives or narcotics at airports or the like, delays in the order of 20 seconds or more for scanning a single item of luggage, which would arise from using any of the above multiple pulse sequence techniques, would be unacceptable.

Disclosure of the Invention

It is an object of this invention to provide for varying the waiting time between pulse sequences for NQR detection purposes.

It is a preferred object to provide for the reduction of this waiting for the effective detecting of NQR signals in a substance being targeted for detection.

In accordance with one aspect of the present invention, there is provided a method for varying the spin-lattice relaxation time Ti in substances containing quadmpolar nuclei and nuclei with a magnetic moment, comprising:

controlledly applying a DC magnetic field to a sample that may have a substance containing quadrupolar nuclei and nuclei with a magnetic moment; and

adiabatically removing the DC magnetic field.

Preferably, the method includes applying excitation to the sample to excite a • resonance response signal in the substance if present, prior to applying the DC magnetic field.

Preferably, the method includes detecting any response signals from the sample after applying the excitation and before applying the DC magnetic field. - B -

Preferably, the method includes: applying another excitation to the sample to excite the resonance response signal in the substance if present, after adiabatically removing the DC magnetic field; and detecting any response signals from the sample.

Preferably, the method includes processing any received response signals to detect the presence of an NQR signal corresponding to a substance being targeted.

In accordance with another aspect of the present invention, there is provided an NQR detection system comprising:

an NQR detection apparatus, including a probe for irradiating a volume with RF electromagnetic energy and receiving response signals from a sample disposed within the volume, a control means for applying excitation to said probe, and a signal processing means for processing the response signals to detect the presence of an NQR signal corresponding to a substance containing quadrupolar nuclei and nuclei with a magnetic moment being targeted within the sample;

a DC magnet associated with the probe; and

a DC magnet control means associated with said control means for controlling the operation of said DC magnet in conjunction with said NQR detection apparatus;

wherein said DC magnet control means operates said DC magnet to permeate the volume with a DC magnetic field.

Preferably, said control means: (i) applies excitation to said probe of a form comprising a combination or multiple of pulse sequences of the SSFP or SLMP type, or both; and (ii) reduces the waiting time between the sequences of pulses to correspond with the reduction of spin-lattice relaxation time occurring in the substance in response to the cross-relaxation and cross-polarization effects caused therein by the DC magnetic field applied thereto. Essentially, applying the DC magnetic field causes a cross-relaxation (CR) effect and a cross-polarization (CP) effect between the quadmpolar spin-system (quadrupolar nuclei) and magnetic spin-system (normally protons), in a substance responsive to the NQR phenomenon. Consequently, the spin-lattice relaxation time can be reduced, permitting a combination or multiple of pulse sequences to be applied for exciting the sample with a reduced waiting time between the pulse sequences, where the NQR signal for particular substances can be increased and/or distinguished from noise, which was not possible before.

Brief Description of the Drawings

FIG. 1 (prior art) is a block diagram of a conventional NQR apparatus for detecting a resonance signal in a substance being targeted within a sample.

FIG. 2 is a graph describing a connection between magnetic reservoir (protons), quadrupolar reservoir and lattice.

FIG. 3 is a graph of a level-crossing effect between magnetic (protons) and quadrupolar spin-systems.

FIG. 4 is a flowchart illustrating a method for detecting a resonance signal in the sample according to the first embodiment of the present invention.

FIG. 5 shows the effect of reducing of the waiting time between pulse sequences using the first preferred embodiment of the method.

FIG. 6 is a flowchart illustrating a method for detecting a resonance signal in the sample according to the second embodiment of the present invention.

FlG. 7 is a graph showing the dependence of NQR signal intensity in PETN on the applied DC magnetic field and the effect of increasing the signal intensity using the second preferred embodiment of the method.

FIG. 8 a graph showing the dependence of NQR signal intensity in PETN on the proton polarization time for different applied DC magnetic fields and the effect of iπcreasing the signal intensity using the second preferred embodiment of the method.

FlG. 9 is a block diagram illustrating an NQR apparatus for detecting a resonance signal in a substance being targeted within a sample, according to the first and second embodiments of the present invention.

FIG. 10 is a block diagram illustrating an NQR apparatus for detecting a resonance signal in a substance being targeted within a sample, according to the third embodiment of the present invention.

FIG. 11 shows a flow chart illustrating selecting of a particular DC B field level to produce CP and CR effects in accordance with the fourth embodiment.

FIG. 12 shows a flow chart illustrating moving the sample into the field slowly to avoid destroying magnetic media in accordance with the fifth embodiment.

FIG. 13 shows a side view of a scanner which houses a NQR coil and shield and a DC magnet surrounding the infeed conveyor in accordance with the sixth embodiment.

FIG. 14 shows an example of a coil which can induce polarisation enhancement and cross relaxation effects on luggage in accordance with the sixth embodiment.

FIG. 15 shows a side view of a scanner which houses an NQR coil and shield and a DC magnet surrounding the infeed conveyor and a second DC magnet surrounding the NQR coil in accordance with the seventh embodiment.

FIG- 16 shows a two stage shoe scanner design performing polarisation enhancement and cross relaxation effects in accordance with the eighth embodiment.

FIG. 17 shows a polarisation enhanced landmine detection system in accordance with the ninth embodiment. FIG. 18 is a flow chart showing the method associated with the tenth embodiment

FIG. 19 shows a coil design for producing the eleventh embodiment.

Best Mode(s) for Carrying Out the Invention

The best mo.de for carrying out the present invention is directed towards a method and system using an NQR detection apparatus for detecting the presence of target substances within a sample, containing both quadrupolar nuclei and nuclei with a magnetic moment.

In the preceding Background Art discussion, a conventional NQR detection apparatus was described with reference to Figure 1 of the drawings. In the Best Mode for Carrying Out the Invention, the NQR detection system is Improved from the apparatus shown in Figure 1, to provide a system that can perform any of the methods described in the following embodiments.

As shown in Figure 9 of the drawings, the hardware components of the NQR detection system of the best mode include a probe 30, which is connected to a receiver unit 50 and a conventional transmitter unit 60, via a duplexer and matching circuit 40. The probe 30 includes a coil, a tank circuit and a tune circuit (not shown). The tank circuit is tuned with the tune circuit to a frequency of interest for detecting the presence of a particular substance containing quadrupolar nuclei and nuclei with a magnetic moment targeted by the system, within a sample disposed within a volume circumscribed by the coil.

The duplexer and matching circuit 40 is a circuit that switches the tank circuit between a transmit and a receive mode, as well as matches the receiver unit 50 and transmitter unit 60 to the tank circuit, via control line 45. The transmitter unit 60 generates RF pulses and transfers the pulses to the tank circuit. These RF pulses can excite NQR signals in the targeted substance within the sample under investigation, when it is located in the volume circumscribed by the coil of the probe 30. This signal is amplified and detected by the receiver unit 50 and is then . delivered for further mathematical processing into a computer, which is part of control and signal processing means 70.

Importantly, the NQR system includes a DC magnet 10 to create a DC magnetic field that permeates the volume, and hence any targeted substance with the sample therein. The operation of the DC magnet is controlled by a DC magnet control means 20, vyhich is also controlled by the control and signal processing means 70, via control line 25. The DC magnet control means 20 sets the strength of the magnetic field produced by the DC magnet 10, which controls the operation of the DC magnet via control line 15.

The control and signal processing means 70 comprises the computer, an RF signal source in the form of a digital synthesiser unit and a pulse programmer (for producing control signals). The digital synthesiser unit generates an RF signal, which, from a first output 72 of the control and signal processing means 70, is transmitted: (i) to one of the inputs 62 of the transmitter unit 60 for further formation of the RF carrier of the RF pulses transmitted by the transmitter unit for exciting the probe; and (ii) to one of the inputs 52 of the receiver unit 50 as the reference frequency. The pulse programmer generates control signals, which are transferred from a second output 74 of the control and signal processing means 70, to another input 64 of the transmitter unit 60 in accordance with parameters prescribed by the computer for determining the particular RF pulse sequence to be applied to the probe 30. The pulse programmer is also controlled by the computer to generate another control signal that is transferred from a third output 76 of the control and signal processing means 70, to a second input 54 of the receiver unit 50. The computer also generates control signals for tuning the tank circuit from a third output 78, which are transferred to a control input 35 of the probe.

Now having the described the hardware components of the NQR system the general method of operating the same in accordance with the best mode will now be described. In describing the method, the specific manner of generating the pulse sequences and the types of pulse sequences involved for exciting the probe, along with operating the probe and the specific processing performed to detect any NQR signals will not be described further, as this is known in the art. Accordingly, the description of the method will be limited to specifically describing the process of operating the NQR system that pertains to the novel and inventive aspects of the present invention.

According to the best mode, the method of operating the NQR system is based essentially on generating cross-røiaxation (CR) and cross-polarization (CP) effects in a targeted substance within a sample undergoing investigation for enhancing the detecting of NQR signals identifying the targeted substance. The detection algorithm essentially includes the steps of:

applying a DC magnetic field to the sample, and

adlabati^'cally removing DC magnetic field from the sample;

in some combination with: applying a sequence of RF pulses to the sample, and

detecting response signals from the sample.

Thereafter signal processing of the response signals is undertaken to detect an NQR signal indicative of the targeted substance if present.

The order and number of repetitions of each step in the detection algorithm can be different for different tasks, depending on whether CP (also known as polarization enhancement - PE) or CR is used.

In the case of CP/PE, the actual sequence of steps involves:

applying a DC magnetic field to the sample,

adiabatically removing DC magnetic field from the sample, to reach some energy level where the polarization is transferred to the quadrupαlar nuclei,

applying a sequence of RF pulses to the sample, ^■ detectiπg response signals from the sample,

signal processing the response signals to detect an NQR signal indicative of the targeted substance if present.

In the case of CR, the sequence of steps is performed, more or less, in reverse, involving:

applying a sequence of RF pulses to the sample,

detecting response signals from the sample,

adiabatically applying and removing a DC magnetic field to cause energy in the quadrupolar system to be drained into the proton system, allowing a rapid repeat of the next NQR pulse sequence as the spin-lattice relaxation time Ti has been effectively reduced,

applying another sequence of RF pulses to the sample,, and

Using the best mode improves the results that can be achieved from using a multiple or combination of pulse sequences, without detracting from commercially acceptable detection times, than would otherwise be the case. More specifically the best mode produces a significant reduction in the waiting time in between pulse sequences, improved signai-to-noise ratio (SNR) and/or obtaining a location of the substance in a scanned volume and/or checking a suspicious object in a scanned volume.

As previously described, this waiting time is required for the effective detection of NQR signals in substances using multiple or combination sequences of pulses.

The best mode is particularly effective for the detection and identification of substances containing at least two kinds of spin-systems: quadrupolar nuclei (normally nitrogen ¹⁴N) and nuclei with a magnetic moment (normally protons). Examples of these substances include such explosives as PETN, TNT and Ammonium Nitrate (NH₃NO₃). The theory behind the CP/PE and CR will now be described in more detail below.

The quadrupole reservoir Q of abundant quadrupolar nuclei and the proton reservoir P of abundant protons are connected with the lattice and this connection is characterised by the spin-lattice relaxation times T^Q and TIP respectively. The connection between Q and P reservoirs can be established using special experimental techniques and is characterised by the cross-relaxation time TCR. The diagram describing the connection between the two reservoirs and the lattice is shown in Figure 2.

The enhancement of the NQR signals can be achieved by using the cross- polarization (CP) effect. The underlying physical process constituting the CP effect is to bring the nuclei in the P reservoir, occupying NMR energy levels created by a small DC magnetic field, to the same energy difference as that of an NQR transition, so that an exchange. of polarization can occur between the P and Q reservoirs. The separation in NMR levels can be controlled by the strength of an applied DC magnetic field, while the NQR energy levels are mainly determined by the bonding environment. In this process it is possible to increase the polarization of the NQR system through the CP from the NMR levels.

For the CP method, a basic approach is to initially polarize protons, which are more abundant in a sample in a static (DC) magnetic field, so that the proton energy levels have much greater separation than the NQR levels. Given time to equilibrate, these proton levels will have relative occupation numbers determined by the Boltzman distribution. The relative population difference between the two proton levels, hence polarization, will correspondingly be much greater than would be the case with the NQR levels. By reducing the DC magnetic field adiabatically the proton level splitting is reduced such that the proton and quadrupolar energy level separations equalise allowing a transfer of polarisation (see Figure 3). This results in a net polarisation transfer from the protons to the quadrupolar nuclei. This can also be explained through the concept of spin temperatures, where energy flows from the "hot" quadrupolar spin-system to the "cold" proton spin- system to "cool" the quadrupolar spin-system.

By applying conventional pulse detection techniques soon after removing the DC magnetic field, the NQR response can be improved by virtue of the ratio of the proton NMR to NQR frequency, provided the proton reservoir is sufficient to cool the quadrupolar system to the proton spin temperature.

A feature that is noteworthy is that unlike conventional NMR₁ the increase in signal is not critically dependent on the uniformity of the DC field. Removing the uniformity requirement lowers the technology cost considerably and allows diverse applications compared to NMR alone.

It has been discovered, pursuant to the best mode, that for the efficient detection of NQR in a sample, the cross-relaxation (CR) effect can also be used, which takes place between the Q and P reservoirs.

Furthermore, it has been discovered that the PE NQR technique is also applicable in the case of multi-pulse sequence NQR and can be used in combination with the CR technique.

First Preferred Embodiment of the method

The first embodiment of the preferred method involves the use of the CR effect to improve the recovery times after excitation of a quadrupoJe system. An NQR system once excited by an RF pulse recovers to thermal equilibrium with the lattice in a time known as a spin-lattice relaxation time TV A long Ji relaxation time in some substances significantly reduces the efficiency of an applied multi- pulse sequence or sequences. Examples of these explosives include PETN, TNT and Ammonium Nitrate (NH₃NO₃). To take PETN as an example, which has a long Ti relaxation time, the ¹⁴N NQR signal, which is decaying through-out the application of a first pulse sequence, cannot be re-excited by the application of a second pulse sequence until a period greater than Ti has passed to obtain a total signal to noise advantage. In most applications this period is excessive and so detection results are generally based on the single application of an efficient pulse sequence that is optimised towards the NQR signal decay rates. Being able to effectively manipulate the recovery time Ti allows a new approach to the optimisation of multi-pulse sequences, and potentially major improvements in effectiveness of such sequences.

The. use of the CR effect between quadrupolar nuclei and protons can reduce recovery time from the long T₁ of the substance. A method to achieve this is by using contact between the proton and quadrupolar spin-systems, where the energy level separations in the NQR and NMR systems are made the same. During contact the quadrupolar spin-system relaxes with time T_CR (cross- relaxation time) which is normally much shorter than Ti of the substance. For the technique to be effective the proton system should be "cooler" than the RF pulse sequence heated NQR system. In reducing the Ti time, the CR technique can provide for a better Detection Rate (DR) and lower False Alarm Rate (FAR) while keeping the same detection time in comparison to currently used NQR techniques.

By adopting the CR concept with NQR detection, for a substance with a long Ti the optimal waiting time between the pulse sequences or blocks can be reduced up to 100 times. Consequently, this translates into a significant increase in signal to noise.

Figure 4 illustrates a specific method for detecting NQR, according to the first embodiment of the present invention. Referring now to Figure 4, the detection process starts in step S100, where the sample is irradiated by the sequence or block of sequences of RF pulses to excite the first resonance signal which can include the NQR signal. From step S100 the process moves to step S110 where the first resonance signal is detected.

From step S11Q, the process moves to step S120 where the DC magnetic field is applied to the sample. Then the process moves to step S130 where the DC magnetic field is adiabatically reduced to the lower level. Preferably, this lower level is less than the level that creates a proton level splitting equal to the smallest quadmpolar level splitting.

From step S130, the process moves to step S140 where parameters of the sequence or block of sequences of RF pulses are changed. Then process moves to step S150, where the sample is irradiated by the new sequence or block of sequences of RF pulses to excite the second resonance signal which can include NQR signal. From step S150 the process moves to step S160 where the second resonance signal is detected.

From step S160, the process moves to step S170 where the detected first and second resonance signals are analysed to distinguish NQR signal from the noise.

Figure 5 illustrates the effect of reducing the waiting time between pulse sequences using the first preferred embodiment of the method. In Figure 5, it can be seen that the dependence of the intensity of NQR signal obtained from the ¹⁴N nuclei in PETN on the waiting time between two pulse SLMP sequences. The PETN sample was irradiated by the SLMP sequence and after the waiting time the NQR signal was detected by using the same pulse sequence. For comparison, the signal intensities for the regular detection technique (B=OG) (A curve) and using CR effect (B=SOOG) (B curve) are presented. It can easily be observed that when using the first preferred embodiment of the method the effective waiting time can be reduced considerably. For example, in the case of low DC magnetic field (B=500G) the waiting time can be reduced 58 times (from 175s to 3s).

Second Preferred Embodiment of the method

The second embodiment of the best mode is directed towards the use of a combination of CP and CR effects. The difference between this embodiment and the first embodiment is the use of two additional steps. The two steps are the same as S120 and 5130 in Figure 4, but occur before S100. This embodiment is shown in Figure 6. Moreover, before the sample is scanned in an NQR system (S100), it is initially cross polarised (S80). After this step it is scanned in a NQR system (S100) and then the CR method is repetitively applied.

Figure 7 illustrates the effect of increasing NQR signal intensity using the second preferred embodiment of the method. In Figure 7 the dependence of the intensity of NQR signal obtained from the ¹⁴N nuclei in PETN on the applied DC magnetic field is presented. For this experiment, the polarization time of the protons was quite long - 60s. It can be seen that the signal intensity can be increased up to 5 times (B=900G). According to the experimental results obtained, the second preferred embodiment of the method is still effective at much shorter polarization times. This fact is demonstrated in Figure 8 where the dependences of the intensity of NQR signal obtained from the ¹⁴N nuclei in PETN on a polarization time of the protons for different values of the DC magnetic field are presented. The proton polarization time can be reduced to 10s. For that time it can easily be observed that when using the first preferred embodiment of the method the NQR signal intensity can be increased by 2.8 times (B=700G).

Third Preferred Embodiment of the method

The third embodiment of the best mode is different from the first and second embodiments by the use of a local DC magnetic field, this field being directed at a particular volume inside a larger scanned volume.

The present embodiment uses the CR effect to identify the location of the substance and/or determine which substance is located within a particular volume. The flowchart is very similar to that in Figure 4.

Figure 10 shows the hardware components of the NQR detection system of the best mode, the differences of this NQR detection system from that presented in Figure 9 are an additional switch 80 and two DC magnets 10. Each DC magnet creates a DC the magnetic field in a localised area- Although not shown in Figure 10, ideally many separate magnets can be added if required to obtain a better resolution of the detection volume. The operation of the DC magnets 10 are controlled by a DC magnet control means 20, which is also controlled by the control and signal processing means 70, via control line 25. The DC magnet control means 20 sets the strength of the magnetic field produced by the DC magnets 10, and controls the operation of the DC magnet via switch 80 and control lines 15.

The operation of the system is as follows:

- The substance is moved into the coil which is surrounded by the two polarisation magnets. One polarisation magnet is located down one end of the coil and the other polarisation magnet is located at the other end of the coil.

- The first pulse sequence is applied and the NQR signal is detected.

- If the received signal indicates the presence of a specific substance, then the first polarisation magnet is energised cross relaxing any sample at the end of the luggage which lies within its field.

- After waiting the required time to achieve relaxation, a pulse sequence is then applied to excite the QR signal.

- Then the second polarisation magnet is energised cross relaxing any sample at the end of the luggage which lies within its field.

- Similarly, after waiting the required relaxation time then a pulse sequence is applied to detect the QR signal.

This process repeats for as many polarisation magnets as there are present. The entire process enables the determination of approximately where the sample of interest lies within the luggage, which is useful for airport staff guide them in locating the explosive or in the case of X-ray, checking the particular location on the X-ray image to determine if there appears to anything suspicious present. This technique is also useful for the X-ray operator to resolve false alarms. Fourth Preferred Embodiment of the method

The fourth embodiment of the best mode is applicable to other embodiments of the present invention. In this embodiment the optimal values of a DC magnetic field for generating CP and CR effects are used. Pursuant to the present invention it has been found that the optimal DC field to be used for CP/CR effects lies between 200G and 750G:

According to physical principles, the CR and CP effects take place if the level crossing between quadrupolar and nuclear magnetic spin-systems occurs. To achieve these effects the value of the DC magnetic field should be larger than those that created the level splitting of magnetic spin system to be equal to the quadrupolar level splitting. This is the condition for the minimal value of the DC magnetic field.

On the other hand, the effectiveness of the CR and CP effects depends on the DC magnetic field strength and is directly proportional to the strength. This means that the value of DC magnetic field should be as large as possible. However, in the case of practical applications of the proposed method, such as luggage screening, the strong DC magnetic field can affect and destroy magnetic media which can be located in the investigated volume.

It has been discovered pursuant to the present invention that effective value of DC magnetic field, from a polarisation point of view should be about 200 G or larger.

The largest safe DC magnetic field depends on the characteristics of the magnetic material being scanned In the past credit cards, floppy disks and video tapes were destroyed by fields on the order of 300G. However, more modern credit cards, floppy disks and video tapes use much higher 'write' fields to store information on these types of magnetic media, with the lower limit now being about 750G. Hard disk drives typically use DC write fields anywhere from 2000G to 5000G and therefore have limits well above DC polarisation fields. Therefore, there exists a

'window¹ of possible DC fields for polarisation enhancement from 200 to 750G. In the future this upper limit should become larger as many types of magnetic media are becoming obsolete, such as floppy disks and video tape. These types of media are being replaced by CD and DVD technology. Indeed some computers are now being manufactured with no floppy disk drives and some video rental stores only stock DVD's in the United States.

Hence, as shown in Figure 11, prior to performing CR & CP effects S310, a safe operating DC magnetic field level for magnetic media is applied and this field level is used to produce CP and CR effects S300.

Fifth Preferred Embodiment:

In the fifth preferred embodiment, which follows on from the fourth embodiment, a technique is used which allows the use of a higher DC magnetic field than could otherwise have been used while not destroying magnetic media. In this method the sample to be scanned is moved very slowly into a DC magnetic field, polarised and then slowly removed from the magnetic field. This embodiment will only be applicable to CP, because in CR, the DC magnetic field will most likely surround the NQR coil and will consequently need to be switched off while the NQR scan occurs, otherwise the DC magnetic field will cause NQR line broadening effects which will make detection of the sample of interest impossible. In this case, the CP DC magnetic field is left on permanently and is not switched off, unless the entire machine is switched off.

It has been found pursuant to the present invention, that if the sample is moved/removed slowly into/from the DC magnetic field, then a much higher magnetic field can be used allowing a greater polarisation and greater signal enhancement. Typically modern floppy disks have a write field of about 700G₁ however it has been discovered that if the floppy disk is moved more slowly into a high magnetic field (MOOOG), then the magnetic information stored upon the floppy is not destroyed. If however, the floppy is moved quickly into the magnetic field or is allow to moved back and forth quickly whilst in the DC field, then the magnetic information is destroyed.

Therefore in the fifth embodiment the sample is moved in such a way into the DC field without destroying any magnetic media, such that the sample is polarised and the DC field used is above the normal write field strength used to imprint information upon the magnetic media.

in Figure 12, the sample is moved slowly into the magnetic field S400 where it is cross polarised. The sample is being cross polarised in a coil which lies some distance away from the main NQR coil, and therefore after polarisation is complete the sample is moved slowly S410 from the DC magnetic field and into the NQR coil, where it is scanned using a pulse sequence S415.

Sixth Preferred Embodiment:

In the sixth embodiment a particular QR scanner design is used which allows the system to polarise the next sample whilst the current sample is being analysed within the QR coil, i.e. pipelining. This creates higher efficiency in the QR scanning process allows increased throughput of the QR scanner device.

As shown in Figure 13, and differing from Figure 10, the DC polarisation coils 140, 170 are located on the infeed conveyor 160 of the QR scanner. Typically the QR scanning process can take anywhere from 8-20 seconds to complete. Hence, the bag 150 that follows the one 120 being scanned, lies stationery on the infeed conveyor 160. To take advantage of this period during which the bag 150 is stationery, the bag 150 is polarised whilst it is on the infeed conveyor by a surrounding DC magnetic field supplied by the DC coils 140, 170. This process is repeated for as long as there are bags that need to be scanned.

Hence, the step by step procedure is:

(i) A first bag is polarised in polarisation coils located within the infeed conveyor.

(ii) This first bag is then moved into the QR coil where it is scanned for QR. Simultaneously, a second bag is moved into the polarisation colls and is polarised. (lii) The first bag is then exited from the system, the second bag is moved into the QR coil to be scanned and a third bag is moved into the polarisation coils to be polarised.

(iv) The process continues.

DC coils can interfere with pacemakers of aircraft passengers and other medical equipment. To avoid this problem the shield 175 surrounding the section where the polarisation takes place is made of a material which prevents the escape of DC magnetic fields such as "mu metal".

The DC coils 140, 170 are in fact one coil arrangement designed to produce a reasonably linear magnetic field across the piece of luggage. The preferred option is to use a spiral saddle coil combination to generate a highly uniform magnetic field while still having an opening to allow luggage to pass through the device.

Such a coil design was described in our earlier International patent application

WO2004/042426 for QR scanning of volumes, which is incorporated herein by reference. In the present embodiment, the coil design 11, as shown in Fig. 14, is used for the different purpose of producing a DC magnetic field for polarisation. ,

The top coil 140 is one spiral saddle coil 13a and the bottom coil 170 is the second spiral saddle coil 13b. Both coils are connected in series and powered by a power supply unit (not shown). To limit the power supply requirements, the spiral coil would have at least 50 turns. In alternative embodiments, other coii designs are possible provided they can produce a reasonably uniform magnetic field.

The overall benefit is a similar throughput rate to current QR systems but with the added advantage of increased signal to noise due to polarisation enhancement.

Seventh Preferred Embodiment:

The seventh preferred embodiment is similar to the previous embodiment except that a second polarising DC coil is placed surrounding the QR coil. The arrangement of two DC coils allows both polarisation of the sample at the infeed conveyor and cross relaxation of the sample after the first pulse sequence has been applied. As shown in Figure 15, the bag 150 located on the infeed conveyor 160 is polarised by the first spiral saddle coil 140, 170 for a period of 8-20 seconds. The bag is then moved into the QR coii 110 where it is scanned using a first QR pulse sequence. Then the second spiral saddle coil 180, 190 is switched on to cross relax the sample. Subsequently, a second QR pulse sequence is applied. This process is repeated for as many pulse sequences as are required. Once the pulse sequences are finished being applied the bag exits the system.

Hence, the step by step procedure is:

(ii) This first bag is then moved into the QR coil where it is scanned for QR.

Simultaneously, a second bag is moved into the polarisation coils and is polarised. The first bag is cross relaxed and rescanned with another pulse sequence. This process of cross relaxing and rescanning continues for as long as additional pulse sequences need to be executed.

(Hi) The first bag is then exited from the system, the second bag is moved into the QR coil to be scanned, cross relaxed and re-pulsed, and a third bag is moved into the polarisation coils to be polarised.

(iv) This process continues.

Again to prevent damage to pacemakers and other medical equipment, the shield 175 is composed of a material which prevents the escape of DC magnetic fields. The shield 100 surrounding the QR coil 110 is composed of aluminium or copper metal on the inside to retain the high Q properties of the QR coil and on the outside is composed of a material which prevents the escape of DC magnetic fields. This combination is required because if there was the DC magnetic field limiting shield only, then the QR coil's AC magnetic field would be affected leading to a loss of Q and inductance, which ultimately means a loss of signal to noise in the QR measurement. However, if the shield was left as copper or aluminium then the DC magnetic field could escape to the outside world affecting pacemakers, because DC magnetic field can easily penetrate copper and aluminium sheet metal with out loss.

The spiral saddle coils are 140, 170 are in fact one coil arrangement designed to produce a reasonably linear magnetic field across the piece of luggage. Both coils are connected in series and powered by a power supply unit (not shown).

The benefits of such a system are to enable both cross polarisation and cross relaxation to be performed simultaneously enhancing the signal to noise ratio over what could be achieved by ordinary QR scanners and performing the measurement quickly and efficiently.

Eighth Preferred Embodiment:

In the eighth preferred embodiment, the polarisation coil is added to a QR shoe scanner,

In the prior art technique, a QR shoe scanner scans the shoes of a passenger before they board an aircraft, checking for explosives that might be contained within their shoes. At present, to scan a person's shoes the passenger steps onto a platform, underneath which is a coil for scanning for QR responses from explosives. After the person has stepped onto the shoe scanner, a controlling computer scans sequentially as many explosives as there are required to be scanned. Once the scan cycle has been completed, the result of scanning of the person's shoes is indicated and the person steps off the platform.

In the eighth embodiment a modification to this technique is done by adding a polarising coil as shown in Figure 16. This is achieved by using two platforms 220, 230 instead of one platform. The first platform 220 is for polarising any explosive contained within the passenger's shoes by the use of a DC magnetic field and the second platform is for performing QR detection. In this procedure the person steps onto the first platform for a few seconds where two spiral half-saddle coils either side of their shoes polarise the explosive. Then the person steps off the first platform and stands on the second platform where they are scanned for explosives.

By using two platforms the throughput of passenger's can be increased by parallel processing two passengers simultaneously. One passenger can be scanned for QR responses on one platform, while another can be exposed to the DC magnetic field on the second platform.

Two additional spiral half saddle coils 240, 250 are added to the QR platform to perform cross relaxation measurements. To prevent the coil of the DC coil spirals and the QR coil from interfering with each other, the one coil is switched to be open circuit.while the other coil is in operation

Ninth Preferred Embodiment:

The ninth preferred embodiment is substantially similar to the previous embodiment, except it is directed to application concerning a QR landmine detector

For QR landmine detection, typically a metal detector type coil detection system is used or the detector is attached to the front a vehicle. As shown in Figure 17, in either case the polarising coil 510 is concentric with the QR coil 500. The procedure involves the QR coil circuit initially being switched to open circuit and the polarised coil beings switched to closed circuit. This prevents the QR coil affecting the field strength of the DC coil. The DC power supply is then switched on polarising any explosive contain within a landmine underneath the coil. Then the DC coil is switched to open circuit and the QR coil is switched to closed circuit and the QR detection process begins. The DC coil needs to be switched to open circuit to prevent nulling the QR coil's field- This process can be repeated if cross relaxation is required.

Tenth Preferred Embodiment: In QR detection, the level crossing point for magnetic fields applied to a target sample occurs at around 200G. If a sample is exposed to a DC field below this, point then_, there is no improvement in the signal intensity. However if the sample is exposed to a field greater than the level crossing then contact between the spin systems occurs and there is a signal enhancement. Hence, this fact can be used to advantage in the QR detection of explosives. In particular, this fact can be used to distinguish false alarm signals from real alarm signals.

Hence, in the tenth embodiment as shown in Figure 18 the sample is initially polarised and the NQR signal intensity is determined and compared to a threshold S500. If there is an apparent alarm generated by that detection process, then the sample is not polarised and the sample is remeasured S530. If there is a significant difference in the signal level between the two measurements then it is presumed that the signal is real $540. If however there is no change in the signal level then it presumed that the signal is a false alarm S560.

Eleventh Preferred Embodiment:

The polarisation effect can be used to form two or three dimensional images of quadrupolar nuclei within objects. As shown in Figure 19, a particular section of a piece of luggage is polarised using a very narrow DC coil 300. The narrow DC coil 300 shown in Figure 19 is a single turn coil and is energised by passing a DC current through it. After this section has been polarised for a short period of time the bag is moved into the main QR coil 320 where it is irradiated with a high power pulse sequence. The signal that is generated is stored for future use. The bag is then moved back and a different section of the bag is polarised. After being polarised, the bag is moved back into the main QR coil where it is once again irradiated with a high power pulse sequence and the resultant signal is stored. This process repeats until all sections of the piece of luggage have been polarised and measured. A two dimensional image of the intensity versus the distance along the bag can then be produced allowing the location of the quadrupolar nuclei to be identified. Once located, the object can examined remotely by viewing the bag through an X-ray image or the bag can be manually inspected. It should be appreciated that the present invention is not limited to the specific embodiments described herein and that alternative embodiments may be proposed that fall within the scope of the present invention.

Claims

The Claims Defining the Invention are as Follows;

1. A method for varying the spin-lattice relaxation time Ti in substances containing quadrupolar nuclei and nuclei with a magnetic moment, comprising:

coπtrolledly. applying a DC magnetic field to a sample that may have a substance containing quadrupolar nuclei and nuclei with a magnetic moment; and

adiabatically removing the DC magnetic field.

2. A method as claimed in claim 1 , including applying excitation to the sample to excite a resonance response signal in the substance if present, prior to applying the DC magnetic field.

3. A method as claimed in claim 1 or 2, including detecting any response signals from the sample after applying the excitation and before applying the DC magnetic field.

4. A method as claimed in any one of the preceding claims, including: applying another excitation to the sample to excite the resonance response signal in the substance if present, after adiabatically removing the DC magnetic field; and detecting any response signals from the sample.

5. A method as claimed in any one of the preceding claims including processing any received response signals to detect the presence of an NQR signal corresponding to a substance being targeted.

6- An NQR detection system comprising:

(i) an NQR detection apparatus, including: a probe for irradiating a volume with RF electromagnetic energy and receiving response signals from a sample disposed within the volume,

a control means for applying excitation to said probe, and

a signal processing means for processing the response signals to detect the presence of an NQR signal corresponding to a substance containing quadrupolar nuclei and nuclei with a magnetic, moment being targeted within the sample;

(ii) a DC magnet associated with the probe; and

(iii) a DC magnet control means associated with said control means for controlling the operation of said DC magnet in conjunction with said NQR detection apparatus;

7. An NQR detection system as claimed in claim 6, wherein said control means:

(i) applies excitation to said probe of a form comprising a combination or multiple of pulse sequences of the SSFP or SLMP type, or both; and

(ii) reduces the waiting time between, the sequences of pulses to correspond with the reduction of spin-lattice relaxation time occurring in the substance in response to the cross-relaxation and cross-polarization effects caused therein by the DC magnetic field applied thereto.

8. A method for varying the spin-lattice relaxation time Ti in substances containing quadrupolar nuclei and nuclei with a magnetic moment substaπtially as described herein with reference to the accompanying drawings as appropriate.

9. An NQR detection system substantially as described herein with reference to the accompanying drawings as appropriate.